BUILDING A

BETTER
TEACHER

How Teaching Works

(and How to Teach It to Everyone)

Elizabeth Green

W. W. NORTON & COMPANY

New York • London
For my family: parents, brothers, and Dave.
 Contents

Prologue: HOW TO BE A TEACHER (Part One)

1 FOUNDING FATHERS
2 A TEACHER IS BORN
3 SPARTAN TRAGEDY
4 KNEAD AND RISE
5 AN EDUCATIONAL START-UP
6 LEMOV’S TAXONOMY
7 THE DISCIPLINE OF DISCIPLINE
8 THE POWER OF AN INSIDE JOKE
9 THE HOLY GRAIL
10 A PROFESSION OF HOPE
Epilogue: HOW TO BE A TEACHER (PART
TWO)

Acknowledgments
Notes

Index
BUILDING A
BETTER
TEACHER
 Prologue

HOW TO BE A TEACHER
(Part One)

Open the door and walk in. Remain standing. Or

maybe you should sit down?
This crowded rectangular room is yours. Right
now it has twenty-six chairs with attached desks,
a chalkboard, and early-afternoon sunlight pouring
through windows onto the tabletops. In a moment,
the room will also have twenty-six fifth-graders
whose names are printed on the attendance ledger:
Richard, Catherine, Anthony, Eddie, Varouna,
Giyoo, Awad, Donna Ruth, Tyrone, Ellie, Enoyat,
Leticia, Charlotte, Karim, Shanota, Messima,
Saundra, Dorota, Ivan, Connie, Illeana, Yasu,
Reba, Jumanah, Candice, and Shahroukh.
Your job, according to the state where you
happen to live and the school district that pays
your salary, is to make sure that, sixty minutes
from now, the students have grasped the concept
of “rate.” Specifically, if a car is going 55 miles
per hour, how far will it have traveled after 15
minutes? How about after 2 hours? By the end of
the year, your students should also have mastered
fractions, negative numbers, linear functions, long
division, ratio and proportion, and exponents.
You’re also supposed to teach them to become
good citizens, subtly knitting into your lesson (yes,
this math lesson) the principles of democracy. In
whatever time is left, remember to help the
children vault over any hurdles life has thrown
them—racial, economic, parental, intellectual.
You must bend reality closer to the dream of the
American meritocracy.
Ready?
The door bursts open. With the residual energy
of recess, they surge through the coat room,
rearranging their clothes and jostling for sips from
the water fountain. Here comes Varouna. She is
from Kenya, lithe and dark skinned. Giyoo is from
Japan. He is 4 feet tall and barely speaks.
Catherine is studious and has her hair in braids.
Eddie, freckle faced and hyperactive, takes his seat
in the back. Tyrone just moved from South
Carolina and prefers not to pay attention. He sits
closer to you, in the front.
Don’t just stand there. Teach something!
Richard sits near the front, next to Tyrone.
They’re both new to the school this year. On the
first day, Richard introduced himself and
volunteered that math was his “worse subject.”
Half an hour later, the students are all askew,
murmuring and chatting with each other. They’ve
been working on a math problem you wrote on the
chalkboard while they were out at recess.

Condition: A car is going 55 mph. Make a

diagram to show where it will be
A. after an hour
B. after 2 hours
C. after half an hour
 D. after 15 minutes

Consider how to get everyone to quiet down.

Next to you, on a table, is a small bell. Do you
ring it? Perhaps you should raise one hand and put
the other hand over your mouth. Or what about
that old line? When my hand goes up, your
mouths go shut. You go for the bell. Thankfully, it
works, and you launch a discussion.
Soon, fifteen minutes have passed, and class is
almost over. So far, the students have worked on
the problem in small groups of four to six. You
have circulated around, peering over shoulders at
their varying degrees of success, deciding when to
talk and when to nod and when to hold in a laugh,
letting it shake inside your chest when a student
does something hilarious and adorable. And all of
you, together, have reasoned your way through A,
B, and C.
On the chalkboard, you’ve drawn a straight
horizontal line, with distance represented on top
and time underneath. On the far right is a
crosshatch for 110 miles and 2 hours (B); halfway
in the middle there is another for 55 miles and 1
hour (A); then there’s one more, smaller,
crosshatch halfway between 0 and 55: 27.5 miles
and ½ hour (C).
It looks like this:

Point to the board. Ask: Can anyone show

where the solution to part D should go on the
diagram?
Hands shoot up. Then, right in front of you,
Richard adds his. You know enough about the
others to have an idea of how they understand
“rate,” or at least an idea of what they will be able
to do with the problem. Richard, though, is
something of a mystery. After the “worse subject”
speech, you collected his math notebook at the end
of each week along with the other students’. But
he wrote very little in it and only rarely raised his
hand. Now he’s volunteering to answer the most
difficult part of the question—and you have no
idea what he’ll say.
What do you do?
Look at the clock; only 10 minutes left. Do you
have time to risk a wrong answer? What about
Richard? What if he isn’t even close? If he’s
wrong, will he, an African American boy in a
racially diverse classroom, shut down and hesitate
to participate again? On the other hand, what
message does it send to the others not to call on
him?
“Richard,” you say. He stands up, turning his
notebook so he can see it from the board, and
walks slowly to the front. Everyone waits, silent.
D: Show where the car going 55 mph will be
 after 15 minutes.
Reaching for the miles section, on top, he rests the
chalk halfway between 0 and 27.5. “15 minutes,”
he writes. Below, between 0 minutes and ½ hour,
he writes, “18.” The board looks like this:

“Ummm,” he says. “Eighteen.”

Huh? Not only has he put time (15 minutes)
where distance should go, but he has also
proposed another number, 18, that makes no
sense. A car going 55 miles an hour could not
travel 18 miles in 15 minutes. And what
reasonable computation would get you to 18?
Not dividing 27.5 by 2, or 110 by 4, certainly, and
not anything else related to the numbers on the
board either.
What do you do?
You could quickly correct his time-distance
reversal, not drawing too much attention to the
mistake, on the assumption that it was a careless
error. But what if it wasn’t? You decide to assume
nothing. “Eighteen miles,” you venture, “or
eighteen minutes?”
Clarify: “You wrote 18 next to minutes. Did
you mean 18 miles and 15 minutes?” Richard
nods, erases, and rewrites. Now the numbers are
flipped: 18 miles, 15 minutes. But there’s still that
mystifying 18.
What do you do? Should you say, simply and
directly, That’s wrong? What does Richard mean,
anyway?
Look at the class. Ask: Can anybody explain
what Richard was thinking?
Another jolt of hands. Try to memorize who is
asking to speak, and who is making a fan out of
his pencils. Remember, you aren’t just teaching
Richard; the other twenty-five need to be educated
too. What are they thinking? Are they learning?
Check the time. Just a few minutes left, but this
could take much longer. Maybe better to give up;
there’s always tomorrow. But look at Richard,
who still believes 18 makes sense, who doesn’t
know what he doesn’t know.
Call on studious Catherine. “Ummmmm,” she
says. “I disagree with that.” She pauses. Then,
“Ummm . . .”
Think. She wants to give the correct answer, yet
you said, can anybody explain what Richard was
thinking, not can anybody talk about her own idea.
Catherine seems to know she’s out of order. That
“ummm . . .”—she’s eyeing you, looking for
permission to disobey.
Do you grant it? Maybe you should. Nod, and
the right answer will come—clear and concise,
knowing Catherine, and just in time for the end of
class. But look at Richard. If quick Catherine, a
white girl, jumps in with the save, what effect will
that have on him? On the other hand, if you don’t
let Catherine continue, how will that affect the rest
of the class? In either case, what will the class
learn about race, gender, and—oh yeah, math?
On Monday, November 20, 1989, Magdalene
Lampert, a stoic, watchful woman with straight
blonde hair clipped to the back of her head and
more than a decade of teaching experience, made a
snap decision. She pointed to the 18. “Does
anyone agree with this answer?” she asked.

The common view of great teachers is that they are

born that way. Like Michelle Pfeiffer’s ex-marine
in Dangerous Minds, Edward James Olmos’s
Jaime Escalante in Stand and Deliver, and Robin
Williams’s “carpe diem”–intoning whistler in
Dead Poets Society, legendary teachers transform
thugs into scholars, illiterates into geniuses, and
slackers into bards through brute charisma.
Teaching is their calling—not a matter of craft and
training, but alchemical inspiration.
Bad teachers, conversely, are portrayed as
deliberately sadistic (as with the Sue Sylvester
character on Glee), congenitally boring (Ben
Stein’s nasal droner in Ferris Bueller’s Day Off),
or ludicrously dim-witted (Mr. Garrison from
South Park). These are the tropes of a common
narrative, a story I’ve come to call the “Myth of
the Natural-Born Teacher.”
Even in the rare cases where fictional teachers
appear to improve—as happens in Goodbye, Mr.
Chips, the novel-turned-film, in which a bland
schoolteacher named Mr. Chips comes to
“sparkle”—the change is an ugly duckling–style
unmasking of hidden pizzazz rather than the
acquisition of new skill. Others think Mr. Chips
has become a “new man,” but in fact, we are told,
he has only peeled back a “creeping dry rot of
pedagogy” to reveal the “sense of humor” that “he
had always had.”
 The idea of the natural-born teacher is
embedded in thousands of studies conducted over
dozens of years. Again and again, researchers
have sought to explain great teaching through
personality and character traits. The most effective
teachers, researchers have guessed, must be more
extroverted, agreeable, conscientious, open to new
experiences, empathetic, socially adjusted,
emotionally sensitive, persevering, humorous, or
all of the above. For decades, though, these studies
have proved inconclusive. Great teachers can be
extroverts or introverts, humorous or serious,
flexible or rigid.
Even those charged with training teachers—the
ones who, by definition, should believe teaching
can be taught—believe the natural-born-teacher
narrative. “I think that there is an innate drive or
innate ability for teaching,” the dean of the
College of Education at Chicago State University,
Sylvia Gist, told me when I met with her in 2009.
The consensus seems to be, you either have it or
you don’t.
Before I met Magdalene Lampert, I ascribed to
this view as well. My teacher friends seemed born
for the blackboard. I could see it in their
personalities and in how much they cared—one’s
earnest, unabashed sensitivity; another’s
confident, playful devotion. Gregarious, charming,
and theatrical, they commanded attention
wherever they went. No wonder they decided to
teach, while I—shamefully serious, allergic to
goofiness, prone to skepticism—became a
journalist. They had the magical quality of
“teacherness”—what Jane Hannaway, the director
of the National Center for Analysis of
Longitudinal Data in Education Research and a
former teacher, described to me as “voodoo.”
When I first met Magdalene, her talent was
obvious, and it did, at first, look like voodoo. It
was the winter of 2009, twenty years after she
taught fifth-grade math to Catherine and Richard,
and she was now a professor at the University of
Michigan’s School of Education. We sat in her
sun-soaked office, at the far end of a long table,
looking at the work of a fifth-grader named
Brandon.
In the course of solving a problem about the
price of party ribbons, Brandon had mistakenly
declared that = 1.5. What, Magdalene asked me,
could have made him think that?
This was probably the first time Magdalene
read my mind, which is what she does after asking
a question. She lowers her eyelids slightly, purses
her lips, and peers into your soul. I had no idea
how Brandon could have come up with 1.5, and
she knew it.
But instead of giving me the answer, she
wanted me to think about what might make sense
(just as, back in 1989, she had wanted Richard to
think about his answer, 18). She drew a long-
division sign, that “house” that I remembered
from fifth grade. She placed the numbers in the
wrong spots: 12 under the house and 7 outside of
it, to the left, as if we were asking how many
times 7 went into 12 rather than how many times
12 went into 7. Putting 7 into 12, a student would
find that it went in once, with a remainder of 5 (12
– 7). “1 R 5,” he would have written, in the
language of fifth grade.
When we looked at Brandon’s paper, that is
exactly what we saw: a house over 12, with 7 on
the outside, and then “1 r 5” written next to it in
green marker. Brandon, Magdalene explained,
must have mistakenly translated his “1 r 5” into
1.5. (The answer is actually 1 and .)
It seemed like a magic trick—how quickly
Magdalene moved from noticing a problem to
diagnosing its source. Instead of just looking at
the final wrong answer, she had translated
Brandon’s notes—almost nonsensical to me—into
a logical (if flawed) path, skipped backward
through his thinking, and located the original
point of misfire. It took her no more than a minute.
And what about all the other errors Brandon
could have made as he struggled to find the price
of those ribbons? What about the mistakes
scattered through his classmates’ papers, not to
mention all the ones that weren’t there, but could
have been? This, after all, was just the work of one
class, taken from one day out of the year, in one
grade and one subject, by one student. I watched,
captivated, as Magdalene worked through more
papers, reading backward through the minds of
the children, each prone to his own unique
mistakes.
But the more I learned about Magdalene and
her teaching, the more I saw that what looked like
mind reading was in fact the result of
extraordinary skill, not inborn talent. Her success
did not depend on her personality, which—
inward, pensive, and measured—was in many
ways the opposite of Hollywood’s mythic teachers.
Instead, Magdalene’s success relied on a body of
knowledge and skill that she had spent years
acquiring. Teaching, as she practiced it, was a
complex craft.
Magdalene showed me that the illusion of the
natural-born teacher is at best a polite version of
the old adage attributed to George Bernard Shaw:
“He who can, does. He who cannot, teaches.” By
imagining teaching as a “voodoo” mixture of
personal charisma and passion, we are saying,
essentially, He who has intelligence, does. He
who has charm, teaches. I have come to think that
this is a dangerous notion. By misunderstanding
how teaching works, we misunderstand what it
will take to make it better—ensuring that, far too
often, teaching doesn’t work at all.

“Aha!” Magdalene Lampert’s decision not to

correct Richard had paid off—partly, anyway.
His nonsensical answer, that the car traveling at
a speed of 55 miles per hour would go 18 miles in
15 minutes, remained on the board. But after
Magdalene asked the class whether anyone agreed
with his answer, enduring an uncomfortable pause
when nobody said anything, Richard finally broke
the silence.
“Can I change my mind?” he asked her. Instead
of 18, he wanted to “put thirteen and a half or
thirteen point five.”
Better! The calculation he should have made is
that, since 55 miles corresponds to 60 minutes,
and half of 55 miles, 27.5, corresponds to 30
minutes, then a quarter of 60 minutes would be
half again: 13.75. He was close.
But Magdalene still didn’t understand why he’d
first said 18. She needed to know exactly what had
gone wrong inside his head. Pointing to the place
on the chalkboard where Richard had originally
written “18,” she asked him why he’d changed his
mind.
He was back in his seat. “Because,” he said,
“eighteen plus eighteen isn’t twenty-seven.”
“Aha!” she said, permitting herself a minor
celebration.
 He had it—at least, most of it. Keeping her
hand on the board so that it covered up the old
wrong answer, 18, Magdalene pivoted so that her
body faced Richard and the rest of the class. She
wanted everyone to hear what she had to say.
Richard had gone from stumbling to coming up
with the beginnings of a proof—a mathematical
argument for why 18 couldn’t be the answer—and
she wanted to draw everyone else’s attention to his
work.
This is what the board showed:

Whatever goes in this spot, she said, has to be a

number that, when doubled, comes close to 27.
In the back of the room, students started
murmuring. “Not close!” one shouted out. Another
student threw up his hand.
Magdalene took note but did not make a move
just yet. She thought about 27. The board, of
course, said the correct number of miles that is
half of 55—27.5, not 27. If he had been shooting
for precision, Richard would have tried to find a
number that, when doubled, equaled 27.5, not 27.
But if they were talking about a real car, making a
real trip, would it matter if he calculated a
distance of 13.5 miles, rather than 13.75? It might,
and it also might not. Still, learning to make
approximations was an important skill, and
Magdalene was happy with Richard’s
performance. He was estimating—proving, even
—thinking mathematically.
She did not want to make Richard think he’d
made a mistake, but she also wanted to help him
and the rest of the class reach the exact answer.
After all, if she hadn’t wanted the students to deal
with the tricky matter of having to divide 27.5 into
two pieces, she could have picked a rounder
speed, like 60 miles an hour. Then the math would
have been nice and clean. But one of her
objectives for the year was to have students learn
to convert between decimals and fractions, and to
divide each of them in their heads. She had picked
55 because she wanted the class to struggle with
exactly this problem, in exactly this way.
How to acknowledge Richard’s good work but
also, at the same time, correct it? She surveyed the
growing field of raised hands. Anthony, a small
boy whom Magdalene knew loved to talk, was
waving his hand in the air. Awad, a quiet boy
with neat, curlicue handwriting, had his hand up
too. Who would keep up her ambiguous tone:
accepting Richard’s answer, but expanding on it?
She chose Awad.

Paradoxically, the institution most susceptible to

the fallacy of the natural-born teacher is our
country’s public school system. And that’s despite
the fact that alarm—always high—over the
disappointing level of our national teaching
quality has recently reached a fever pitch.
“From the moment our children step into a
classroom,” Barack Obama said in 2007, “the
single most important factor determining their
achievement is not the color of their skin or where
they come from; it’s not who their parents are or
how much money they have. It’s who their teacher
is.” Obama was then a presidential candidate; in
office, his position only strengthened. Today,
thanks to policies that his administration has
advanced, school districts across the country are
undergoing ambitious efforts to reinvigorate their
teaching force. The debate about these reforms is
fierce; many people, including many teachers,
oppose Obama’s efforts. But their objection is not
usually with his premise. They agree that teachers
matter and that the quality of their work should be
improved. What they dispute is how to enact the
change.
One argument—Obama’s—prescribes
improvement by way of accountability. The
problem with American education, this line of
thinking goes, is that we have for too long treated
all teachers the same: they get the same pay raises,
the same evaluations, and the same job protections
whether they inspire their students like Robin
Williams or stultify them like Ben Stein. But the
fact is that some teachers are good and some are
bad. Some help children learn while others set
them back.
“They have 300,000 teachers in California,”
Obama explained in a speech in 2009. “The top
10 percent are 30,000 of the best that are out there.
The bottom 10 percent are 30,000 of the worst out
there. The problem is, we have no way to tell
which is which.” This, he went on, “is where data
comes in.” By measuring which teachers are
successful and which aren’t, we can reward the
phenoms and discard the duds, thereby improving
the overall quality of the teaching force. Following
Obama’s prescription, revamped teacher
evaluation systems are now being rolled out across
the country, along with rewards and punishments
that will affect teachers’ careers.
The other argument—call it the autonomy
thesis—prescribes exactly the opposite. Where
accountability proponents call for extensive
student testing and frequent on-the-job
evaluations, autonomy supporters say that teachers
are professionals and should be treated
accordingly. Like lawyers or doctors, they will
improve only if they are given the trust, respect,
and freedom they need to do their jobs well.
Lately, proponents of this argument have been
drawing comparisons to Finland. There, a recent
report by the Chicago Teachers Union described,
“teaching is a respected, top career choice;
teachers have autonomy in their classrooms, work
collectively to develop the school curriculum, and
participate in shared governance of the school.” In
Finland, the report concludes, teachers “are not
rated; they are trusted.”
As descriptions, both arguments—
accountability and autonomy—contain a measure
of truth. Teachers do lack some of the freedom
they need to teach well, and they also lack
adequate feedback. But as prescriptions, actual
suggestions for how to improve teaching, the
arguments fail. Neither change, on its own, will
produce better teachers. Basic math makes the
problem with accountability clear: Discard the
bottom 10 percent and, as Obama said, that’s
thirty thousand teachers who will need to be
replaced. And that’s just in California. Nationally,
the number is more than ten times that.
Autonomy, meanwhile, is an experiment that
many schools have tried for years, and still seen
teachers struggle.
Neither accountability nor autonomy is enough,
in other words, because both arguments subscribe
to the myth of the natural-born teacher. In both
cases, the assumption is that good teachers know
what to do to help their students learn. These good
teachers should either be allowed to do their jobs
or be held accountable for not doing them, and
they will perform better. Both arguments, finally,
rest on a feeble bet: that the average teacher will
figure out how to become an expert teacher—
alone.
This bet is especially audacious, considering the
large number of people involved. More people
teach in this country than work at McDonald’s,
Wal-Mart, and the U.S. Post Office combined. In
New York City, where I live, a corps of teachers
seventy-five thousand strong makes up a
workforce roughly the same size as Apple’s global
employee base. As Amy McIntosh, the former
chief talent officer of New York City’s
Department of Education, pointed out, in all the
five boroughs there is no building where all
seventy-five thousand teachers could gather at a
single time. Not even Yankee Stadium (capacity
50,287).
Of the fields to which teaching is commonly
compared—those that require a college degree and
are considered of reasonably high social value—
none come close to matching the number of
employees that teaching has. Consider a bar graph
displaying the number of Americans in different
professions. The shortest bar represents architects:
180,000. Farther over, slightly higher, come
psychologists (185,000) and then lawyers
(952,000), followed by engineers (1.3 million) and
waiters (1.8 million). At the top stand the big
three: janitors, maids, and household cleaners (3.3
million); secretaries (3.6 million); and, finally,
teachers (3.7 million). An ongoing swell of baby
boomer retirements is expected to force school
systems to hire more than three million new
teachers between 2014 and 2020. As the departing
teachers wave goodbye to their students, they will
take all their experience and skill out the door with
them. These new hires will have to replace them.
 One December night in 2009, I watched as
hundreds of the people hoping to become teachers
packed an auditorium at Chicago’s Cultural
Center, home of the world’s largest stained-glass
Tiffany dome, to hear from the city school
system’s director of recruitment. There were no
seats available, and the sea of humanity was as
diverse as it was vast. There was a cross-eyed
woman with white hair and a disheveled look.
There was a dreadlocked recent college graduate
with hair dangling below his belt. There were
many dozens of young midwestern ladies with
their mothers, taking careful notes. There was a
small woman in a Christmas sweater with
ornaments sewn into quadrants, including a
Velcro nameplate stuck on her left breast:
RACHEL.
But even if everyone in the auditorium had
signed up to teach—the mothers along with their
daughters—the crowd still would not have filled
all the available teaching slots. Each year, the city
of Chicago hires two thousand new teachers. That
year, the economic downturn had lowered the
number below its average. But the district still
needed six hundred new teachers. Nationwide,
nearly four hundred thousand new teachers start
work at public and private schools every year.
When all these people take their place in front
of classrooms across the country—from the
overcrowded trailers in Queens, New York, to the
humid, ranch-style spaces serving Alabama
Native American reservations, to the breezy, open-
air classrooms of Cerritos, California—what will
they do? What should they do? And how can we
make sure all of them do the best possible job?
The cold truth is that accountability and
autonomy, the two dominant philosophies for
teacher improvement, have left us with no real
plan. Autonomy lets teachers succeed or fail on
their own terms, with little guidance.
Accountability tells them only whether they have
succeeded, not what to do to improve. Instead of
helping, both prescriptions preserve a long-
standing culture of abandonment. Steven Farr of
Teach For America described this culture by
telling me about the first time his assigned mentor
came to observe his class. The mentor was just
doing her job, but when she walked in, she
apologized, as if for some voyeuristic intrusion.
Teaching, she told him, is “the second-most
private act.” She’d rather not be caught watching
someone else do it.
The sociologist Dan Lortie, in his classic work
Schoolteacher, describes the teaching profession
in the language of Victorian-era sex: a private
“ordeal.” Lortie traces the fundamental loneliness
to the days of the one-room schoolhouse, when
teachers worked in isolation because the other
adults (and some of the children) were busy
farming. These days, there are more personnel and
more students associated with each classroom, but
each teacher still faces a room full of pupils alone.
What do teachers do? They do what any of us
would do. They make it up.

That day in November, Magdalene Lampert’s

gamble to call on Awad—carefully calculated, in
her case—paid off. Awad played exactly the role
she had hoped, correcting Richard’s imprecision
about 13.5 without trampling over his
accomplishment in getting there.
“Ummm,” Awad had said with his typical
deliberation. “I think it’s thirteen point seventy-
five.” Richard kept his composure, and in the
minutes that followed, Magdalene untangled a
series of teaching problems. She called on
Anthony, who had been waving his hand in the
air, but didn’t let him go on too long and even
distilled a clear, concise idea from his confusing,
if enthusiastic, speech.
She then gave the floor to a girl, Ellie,
balancing the gender of speakers and thereby
minimizing the idea that only boys can do math,
which paved the way for an astonishing
performance by another girl, Yasu, who
constructed a sophisticated logical proof that
recalled Richard’s original insight about the
relationship between doubling and halving. All
this had happened in just a few minutes. But now,
it was beyond time for class to end. The teacher
who was to take over the room after math ended
stood at the back of the classroom, giving
Magdalene a look.
“You know what I think?” she said to the class,
nodding at the teacher in the back. “I think that we
are going to schedule a little time on remainders
and division. ’Cause I think we are getting a little
mix—We are mixing up a lot of ideas here and we
don’t have time to go into them.”
She paused again. She wanted to give anyone
who might be deeply confused one last chance to
ask a question. The students sat before her, their
math notebooks still open in front of them:
Richard in the front, Awad in the back, Catherine
to her right. All of them would be there tomorrow
too, and the next day and the next and the next,
until summer.
“Okay?” Magdalene asked, turning the
statement into a slight question—a door just on its
way to being closed. No one said anything.
Okay.

Both sides of the “teacher quality” debate tend to

depict the challenge as a transfer problem—how
to help unsuccessful, often low-income students
(like the ones I cover as a reporter in New York
City) to access the experiences enjoyed by their
more affluent peers (like the ones I had attending
public school in the manicured Washington, DC,
suburb of Montgomery County, Maryland).
The accountability argument holds that
suburban schools have the best teachers because,
with rich coffers and newer, prettier buildings,
they are able to lure top talent. To rebalance this
unequal distribution, Obama has supported
measures to tempt high-quality teachers back to
school districts serving poorer populations.
Proponents of the autonomy argument,
meanwhile, contend that teachers working with
the poor have paradoxically received the least
freedom and the most restrictive working
environments. Make their schools look more like
those enjoyed by the children of the wealthy, and
they will be able to prosper.
Again, neither description is wrong, but as
prescriptions, both are incomplete. Teachers at
affluent public schools do enjoy, on average, better
working conditions and more flexibility. But they
are also victims of the natural-born-teacher
hypothesis. Indeed, the more I learned about
successful teaching, the more I realized how rare it
is, even in the schools with the most resources.
Not long ago, exploring the closet of my
childhood bedroom in Maryland, I discovered a
pink, cardboard filing cabinet that held my
elementary school papers. In the best classrooms I
visited as a reporter, children were reading and
writing by kindergarten. My pink filing cabinet
did not have a kindergarten file. What would I
have put in it? That year, I did not know how to
read.
The first-grade papers, meanwhile, bore little
resemblance to the careful work I saw in
classrooms run by excellent reading teachers. The
file contained words copied from worksheets and
not much original writing. By January, I had
reached my peak level for the year: I was able to
fill a collection of construction paper sheets,
stapled together and labeled—in an adult’s
handwriting—“Writing Journal,” with lifeless
one-sentence entries:

Water is fun.
I ha t cold weather.
I like toe s.
I like Sarah.
 I like sissors.
It was fun at the show.
I haet work. It is to eyse.

No wonder I didn’t read until first grade. The

work was too easy, and as a result I didn’t learn.
Even later, in the special “gifted” programs I
attended from fourth grade on, how much had my
teachers really taught me? Some changed my life
forever, helping me fall in love with journalism,
calculus, and even quantum physics. But what
about the others? Besides that fluke physics year,
my memories from science classes were mainly of
lethargic fruit flies. And only in my last year of
high school did I figure out that history had to do
with evidence and arguments, as well as
memorizing state capitals and the dates of
irrelevant wars.
I attended some of the fanciest public schools in
the country (my school district, in Montgomery
County, Maryland, has an average household
income in the country’s top ten), yet the teaching I
received was just as inconsistent as at the schools
I later visited in Newark, New Jersey, the Bronx,
and San Francisco.
Yet, while I have come to see that the scope of
America’s education challenge is much larger
than I ever imagined, I have also begun to see a
path through which the challenge might be
tackled. For every case I have found of the natural-
born-teacher fallacy hampering progress—and I
have found a lot, stretching far into the past—I
have found another case of a person who thought
differently.
Take Colonel Francis Parker. Born in 1837 in
New Hampshire, the son and grandson of
teachers, Parker believed that teaching well
required intense study. Teaching was, he said,
“the greatest art in all the world”; learning to do it
well could take a lifetime. But it didn’t take Parker
long to learn that this was an unpopular view.
After serving honorably in the Civil War, he was
offered several prestigious jobs that would take
him away from the classroom. “When I said that I
was going to be a school teacher [instead]”—
during the war, he had spent nights before the
campfire, planning future lessons in his head, and
he did not intend to give them up—“my friends
were very much disgusted with me.” Even another
teacher called him a fool.
His fellow teachers, Parker was finding,
mirrored the general public. Many of them didn’t
think about their work as a craft they needed to
study. Later, when Parker took over a struggling
school for teachers in Chicago, most of the city
(including some teachers) wondered whether the
school should exist at all. “The fact of the matter
is, the conviction that young men and women
should be trained for their work in order to teach
little children existed only here and there,” one of
his colleagues said. “The general public was
against it.”
Parker died before seeing his dream of
resuscitating the school and its reputation fully
realized. He was, said the rabbi Emil G. Hirsch in
his eulogy, “another Moses,” destined to behold
his promised land only from afar. The same fate
befell Parker’s successor at the University of
Chicago, the philosopher John Dewey. Expanding
on Parker’s vision, Dewey had written eloquently
about the “science of education” he hoped to
develop—how it would help prevent the
immeasurable “waste” that comes from letting
great teachers’ secrets live and die with them.
“The only way by which we can prevent such
waste in the future,” he wrote, “is by methods
which enable us to make an analysis of what the
gifted teacher does intuitively, so that something
accruing from his work can be communicated to
others.” The science wasn’t to be—at least, not
yet. For half a century after he made it, Dewey’s
prescription lay in hibernation, the victim of the
same forces that Parker had pressed against.
But though Parker and Dewey both died before
seeing “educational Palestine,” as Emil Hirsch
called it, their vision did not. Today, the natural-
born-teacher illusion lives on, but thanks to
Magdalene Lampert and a growing group of
educators like her, so does Parker and Dewey’s
dream.
The educators include some people like
Magdalene, longtime teachers who later became a
unique breed of researcher, studying their own
craft while they worked to pass it on to others.
They include, as well, people who echoed
Magdalene’s conclusions without ever meeting
her—sometimes deliberately (as happened in the
1980s, on an island six thousand miles away from
her fifth-grade classroom), other times not (as in
the case of the movement of entrepreneurial
educators that emerged a decade later). Together,
these educators still constitute a minority. But for
a variety of reasons, their chances of building
Francis Parker’s educational Palestine are better
than any other time in history.
 This book is their story. It is also the story of
teaching, that hilarious and heartbreaking theater
that unfolds between children and teachers every
day. The work that, when done well, with trained
skill, can induce in a student a near-magical
feeling: the trembling sensation of beholding a
new idea where nothing existed before. It begins
with one of the first pioneers—a shy, industrious
man named Nathaniel Gage.
 1
FOUNDING FATHERS

By 1948, when he landed his first academic job at

the University of Illinois, Nate Gage had already
helped the army select and train radar observers
during World War II; worked with the College
Board to develop a new tool—the Scholastic
Aptitude Test; and coauthored a definitive
textbook: A Practical Introduction to Evaluation
and Measurement. The second son of Jewish
immigrants from Poland, he’d made his way from
hanging wallpaper with his father to the top of his
chosen field, educational psychology. At Illinois,
he joined the prestigious new Bureau of
Educational Research. But the breakthrough that
became Nate’s most important finding happened
in the classroom.
Nate was serious, but also passionate and
sweet. At conferences, he would transfix his
colleagues with barroom storytelling late into the
night. And yet, in the classroom, that chemistry
somehow failed to materialize. He simply could
not keep the students’ attention. It was not
unusual for one or more of them to fall asleep in
the middle of his lectures. “He just didn’t have
that certain something,” says one of his students,
David Berliner. For all his success—the multiple
publications in prestigious journals; the glittering
title, professor of education—the data all pointed
to one disturbing conclusion: Nate was a terrible
teacher.
Distraught, Nate turned to the academic
literature. Surely some of his colleagues in
educational psychology had cracked the mysteries
of teaching. That was when he made his second
discovery: the research on teaching didn’t exist. At
least, the findings didn’t. Instead of conclusions,
researchers had developed a bundle of
idiosyncratic hypotheses, focused mostly on
teachers’ personality traits. Were good teachers
warmer? more enthusiastic? more organized?
more interested in their subject? Maybe better
teachers had similar degrees of bohemianism,
emotional sensitivity, and sociability. Perhaps
subpar teachers displayed radicalism, or even
“worrying suspiciousness.” Other studies cast
their searches even more broadly, investigating
traits from age and experience to eye color,
clothing style, and strength of grip.
None of the studies found anything conclusive.
A researcher would publish a discovery, only to
have another produce exactly opposite findings.
The few conclusions that could be squeezed out of
the research tended to be vague and unhelpful.
One set of studies suggested that good teachers
should be “friendly, cheerful, sympathetic, and
morally virtuous rather than cruel, depressed,
unsympathetic, and morally depraved.” Another
study concluded that the best teachers had a
characteristic called, unhelpfully, “teaching skill.”
 Summarizing the research in 1953, Nate wrote:

The simple fact of the matter is that,

after 40 years of research on teacher
effectiveness during which a vast
number of studies have been carried out,
one can point to few outcomes that a
superintendent of schools can safely
employ in hiring a teacher or granting
him tenure, that an agency can employ
in certifying teachers, or that a teacher-
education faculty can employ in
planning or improving teacher-
education programs.

The irony was bruising. The country, at that

point, had dozens of university programs devoted
to recruiting, training, and vouching for America’s
future teachers—education schools, they were
called. Yet somehow all those ed schools’
professors had managed to learn nothing about
teaching. And that was the professors who paid
the topic any attention at all. The most prestigious
among them—the elite education researchers like
Nate—ignored teaching altogether.
You couldn’t help but wonder. How had this
happened? How had an entire field come to
neglect the work at its heart?

One answer was that they did it on purpose. The

tradition began with the first education professors,
who taught the new education courses with
undisguised reluctance. “Educational
psychology?” the philosopher William James was
said to have quipped. “I think there are about six
weeks of it.” James became the grandfather of the
discipline. His student, Edward Thorndike,
another foundational figure, entered the field only
because he had to. After he finished graduate
school in psychology in 1898, the best job offer he
could find was not in psychology but in pedagogy,
at the Women’s College at Western Reserve
University in Cleveland.
“The bane of my life is the practice school they
stuck me with,” he wrote in a letter to a friend
soon after starting the job. Later, when he moved
to Columbia University’s Teachers College, he
spent his first year visiting schools, but he quickly
abandoned the mission, calling the trips a “bore.”
When asked what he would do if faced with a
certain superintendent’s real-world dilemma, he
scoffed. “Do? Why, I’d resign!”
Instead of addressing educational problems,
Thorndike took psychological ones and grafted
them onto schools. He applied to human students
the general laws of learning that he derived from
his experiments with monkeys, dogs, and cats
(“Never will you get a better psychological subject
than a hungry cat,” he wrote). Meanwhile, he
aided the proliferation of new measurement
techniques, assessing everything from intelligence
to memory. But he did not study teachers.
Even John Dewey, who advocated a “science of
education,” wound up retreating to his original
discipline, philosophy. All around him,
educational researchers had followed Thorndike
and abandoned the study of real schools.
Discouraged, Dewey set his work in education
aside.
Nate Gage, too, never intended to study
education. What he really wanted to be was a
psychologist. But after graduating from the
University of Minnesota magna cum laude, a star
student of the young B. F. Skinner, he was
rejected by all ten graduate programs he applied
to. “From the universities’ point of view it would
be pointless to take him into a graduate
programme in psychology and waste resources
training him, since he was Jewish,” explained
Minnesota’s dean of psychology, Richard Elliott.
Graduate programs were judged by their success
at placing professors, and universities did not hire
Jews. The only program that made him an offer
was one he had not applied to—a new program in
educational psychology at Purdue, where the
young director recruited his students by scouring
psych departments’ reject lists.
Another reason early education professors
ignored teaching was that they found it
uninteresting. Learning to teach composition did
not require a method, but rather a “clear head, an
enduring conscience, an elastic enthusiasm, and
uncommon commonsense,” the English professor
LeBaron Russell Briggs insisted. “There is no
such thing as a science of Pedagogy,” Josiah
Royce wrote in the lead article of the inaugural
issue of the journal Educational Review,
published in 1891. “As for a ‘philosophy of
education’ in any other sense,” Royce added, “the
lord deliver us therefrom.”
Yet the subject had to be offered; simple
economics demanded it. In 1890, total enrollment
in US elementary and secondary schools stood at
just under thirteen million. By 1920, the number
was more than twenty million. In the same period,
the ranks of school teachers grew by nearly four
hundred thousand. Another twenty-one thousand
people served as administrators. By the time Nate
arrived at the University of Illinois, in 1948, the
number of teachers alone was nearing one million.
For a university, the calculation was clear: training
teachers made financial sense whether there was
something to teach them or not.

The grim history might have led another man to

surrender. If William James hadn’t been able to
develop a science of teaching, what could honestly
be expected of Nate Gage from New Jersey?
Anyway, as Thorndike had proved, it was
perfectly possible to make a respectable career in
education research without touching the teaching
problem at all. But where others might have seen
a dead end, Nate saw possibility. After all, in
science, the most important discoveries were born
not from answers, but from puzzles. And, studying
the early work on teaching, he had glimpsed a
common and, he suspected, fatal flaw.
None of the traits the first researchers
investigated—eye color? strength of grip?—had
come from the classroom. They had looked into
hundreds of variables but ignored “the primary
data of the teaching process.” That choice, too,
belied the pattern of science’s greatest discoveries.
Johannes Kepler, Dmitri Mendeleev, Gregor
Mendel—all began by scrutinizing phenomena
close up and only then came up with theories to
explain them. Like “Kepler in examining the
orbits of planets, Mendelyeev [sic] in poring over
the properties of the elements, or Mendel in
raising his peas,” Nate decided, education
researchers would only unlock the mysteries of the
American classroom by venturing inside of it.
Nate set out to construct a true science of
teaching. He called his method the “process-
product” paradigm. By comparing the process
(teaching) to its product (learning), researchers
could conclude which teaching acts were effective
and which were not. The ambition was not unlike
John Dewey’s imagined science of “what the
gifted teacher does intuitively.” The only
difference was that, while Dewey favored learning
about teaching in the messy cauldron of a real
school, Nate preferred formal experimentation. A
successful process-product study, in his view,
needed to approximate the natural classroom
habitat while also controlling for extraneous
variables.
In one experiment, Nate focused on
explanation, the slice of teaching that, in his
opinion, formed “the essence of instruction.” He
and his grad students recruited real teachers to
teach real students, but under certain parameters.
One was that the teachers could speak and use the
chalkboard, but they could not invite discussion,
solicit questions, or even ask students to take
notes. (“For some teachers this restriction may
require a difficult departure from their customary
teaching style,” the instructions read. “We hope
that you will bear with us.”) Another restriction
was the content; each lesson corresponded to a
preselected article from the Atlantic Monthly
magazine. The researchers gave students a
comprehension test at the end of the lesson to find
out which teachers had explained it best.
Nate’s students videotaped each lesson and
catalogued the teachers’ behaviors. One graduate
student, Barak Rosenshine, had a list of twenty-
seven qualities to watch for, ranging from the
average length of words spoken (perhaps brevity
was key?) to the frequency of “reference to pupils’
interests” to the number of gestures (“movement
of the arms, head, or trunk”) and paces (walking
from one place to another).
Another group wrote computer programs to
analyze what the teachers had said. One compared
the transcripts against a “vagueness dictionary”
written specially for the occasion (qualifying
words included almost, maybe, generally, and
most). In one lecture that scored as highly vague,
for instance, a teacher began by describing an
author’s name, which he said was “not too
important.” He went on:

I will put his name up on the board

anyway. It is really not very important at
all. MIHAJOV [sic]—that is the way
you pronounce that word, Uh Mihajlov
wrote those articles. And someone, he
has done something that is fine someone
very similar had done and there was
another author whose name, uh, uh, let
us just remember there is another
author. That one has spelling problems
too. Two authors, two authors. One we
know is Mihajlov, the other one wrote
earlier in nineteen sixty-two. Both of
them complained about conditions,
especially in Russia. And this one was
in prison because he wrote a book about
 conversations with Stalin and, I do not
know if you have ever heard of the book
...

The final step compared the recorded teaching

behaviors (process) and students’ comprehension
scores (products). As one might expect, the
students of vagueness offenders had significantly
lower comprehension. Rosenshine’s method
yielded other strong correlations. A high number
of gestures, it turned out, helped improve
comprehension; so did a high level of right-to-left
movement. The research might not have been
quite what Dewey imagined, but it was certainly
unlike anything Nate Gage’s contemporaries had
seen.
Process-product research caught on quickly. In
1957, sharing an elevator with a colleague at the
American Educational Research Association’s
annual conference, Nate joked that if the elevator
crashed, then all of that year’s research on
teaching would go down with it. That year, he and
the colleague were the only two giving papers on
the topic. By the spring of 1963, Nate’s book
collecting the available research on teaching had
converted a new generation of researchers into the
fold. Officially called The Handbook of Research
on Teaching (and unofficially known as “The
Gage Handbook”), the volume sold 30,000 copies.
One chapter, outlining how to design experiments
to study teaching, generated such demand that the
publisher, Rand McNally, released it separately in
pamphlet form in 1966. By 1974, the pamphlet
had sold 130,000 copies.
Perhaps most important, Nate became, if not the
most engaging teacher, certainly a beloved one.
Graduate students devoted themselves to him, and
even the American Federation of Teachers, a
union representing practitioners across the
country, caught on. “They called him the Sage
Gage,” says Lovely Billups, a union official at the
time, who worked with Nate to convert his
findings into usable lessons for teachers.
So when, in 1971, a pair of young staffers at the
new National Institute of Education was charged
with funding the next generation of research on
teaching, they went straight to teaching’s “pooh-
bah,” according to one of them, Garry McDaniels.
Soon, Nate was taking a leave from his university
—by then he was at Stanford—to help them
launch the new round of funding with a
conference suggesting new directions for research.
There was just one twist. Created by the
contrarian new president, Richard Nixon, NIE
was charged not just with supporting existing
research, but with transforming it. “My
assignment,” says McDaniels, “was to change the
field.” Wittingly or not, Nate helped him do it.
The draft conference agenda he circulated for
feedback went to all his colleagues back at
Stanford, including the man who would eventually
inherit Nate’s “pooh-bah” crown—a young
professor visiting from Michigan State named Lee
Shulman.

“Garbage,” Lee Shulman said when Richard

Snow, another Stanford professor, asked him what
he thought of Gage’s draft—the one he was
circulating about the conference planning the
future of research on teaching. “Same old
bullshit.”
Dick Snow was aghast. “Why?”
“It’s nothing but a kind of testimony to the
past,” Lee said. “Doesn’t Nate realize that
behaviorism is on life support?”
It was true. Nate’s process-product approach
depended on a school of psychology that was
falling increasingly out of fashion. Nate was a
behaviorist by default and also by generation; B.
F. Skinner, his old professor, had been
behaviorism’s seminal figure. The founder of
educational psychology, Thorndike, was another
lifelong adherent. Nate’s rise correlated with
behaviorism’s most prominent period.
The behaviorists held that the only scientific
way to study humans was to study their directly
observable features—their behaviors and the
actions (“stimuli”) that triggered them. But the
new generation of psychologists began to point out
that by focusing on stimuli and their responses,
behaviorists were ignoring the mind.
In Thorndike’s model, the human mind was just
an extension of the animal one. Learning meant
responding to repeated rewards or punishments. If
rewarded for one behavior enough times, the
subject learned to keep doing it. If punished, he or
she learned to stop.
But while this pattern might describe some
forms of human learning, critics argued that
behaviorism could never explain them all—
especially not the kind of learning that went
beyond simple actions (will I get food when I
press my cat paw on this pedal?) to more
complicated concepts (when is it useful to
calculate an indefinite integral?). To explain how
people learned higher-level concepts, the critics
held, psychology had to reckon with cognition.
Lee, who’d begun not as a psychologist but as a
philosopher, had never liked behaviorism. It
rejected as unscientific the questions that he found
most fascinating—questions about the mind. Early
on, that opinion was unpopular. But by the time of
Lee’s year at Stanford, in 1973, critics—known as
“cognitivists”—had broken the behaviorist
stranglehold on their field. The cognitive
revolution spread from one area of psychology to
the next, turning attention from behavior to the
working of the mind.
Lee figured the shift should apply to research on
teaching too. The whole point of process-product
research, Nate Gage’s great contribution, was to
study teaching by studying teachers’ behaviors.
But what about their minds?
“Why don’t you write to Nate?” Snow told Lee.
“Come on, Nate personifies process-product
research!” Lee said.
But Snow was insistent. Nate was a serious
scholar. He’d listen. “So I wrote him a two-page
memo,” Lee says. “Probably wrote it on a
typewriter, Selectric typewriter, and I made—I
politely critiqued what he was doing and said,
‘You don’t even have one group looking at the
relevance of cognitive work for the study of
teaching, and my guess is that’s the future of
research on teaching.’”
Lee was mostly just riffing. “I mean, I wasn’t
really in the field at that point. I was teaching
future teachers . . . But research on teaching
wasn’t my area.” So when the phone rang a few
days later and Nate asked him to use the memo as
the basis for leading one of the ten panels at his
conference in Washington, Lee was unprepared.
He didn’t think. He just said yes.

Lee Shulman’s area of expertise was doctors. He’d

begun studying them in 1968, at Michigan State,
as an outgrowth of an idea that first struck him in
graduate school.
Besides education, what Lee had always found
fascinating was thinking. The technical term was
epistemology, the occupation of thinking about
thinking. Like his idol John Dewey, Lee focused
on higher kinds of thoughts, the mental operations
that take place when a person moves from
impression to question to understanding. “The
pedestrian,” wrote Dewey, “feels the cold; he
thinks of clouds and a coming shower.”
The psychology of thinking wasn’t just
fascinating; it also seemed painfully relevant to
education. By understanding complex thought—
the process of making knowledge—researchers
would not just study schools; they would help
improve them. And Lee had an idea for how to
study thinking in a way that could make a real
difference. Other early cognitive psychologists
presented subjects with problems to solve, puzzles
to answer, but Lee knew that, in real life,
problems didn’t come prepackaged. “A problem
well put is half solved,” John Dewey wrote.
“Without a problem, there is blind groping in the
dark.” To get a true grasp on how knowledge was
made, Lee intended to study the blind groping in
action. He only had to find the right research
subjects—people for whom problem solving was
part of the natural habitat.
The idea of studying doctors arrived a few years
into Lee’s time at Michigan State, when a man
walked into his office and introduced himself as
the dean of the university’s new medical school. “I
understand you study complex problem solving,”
he said to Lee. “Well,” he continued, “I think
that’s what medicine’s all about, and we
physicians don’t begin to understand how that
really works. Would you be willing to take 50
percent of your appointment and join the medical
school faculty and do research on medical problem
solving?”
 Doctors. Of course! “It was such an epiphany,”
Lee says. Doctors solved problems all day. It was
the heart of their work. Joseph Bell, Sir Arthur
Conan Doyle’s medical school professor and a
surgeon with legendary capacities of deduction,
had inspired Sherlock Holmes, the greatest
professional problem solver in (fictional) history.
Lee said yes, and it was a perfect fit. Observing
doctors at work with his colleague and childhood
friend Arthur Elstein, he overturned the
conventional wisdom about medical problem
solving—and, ultimately, helped improve medical
education in the process. Lee and Arthur designed
simulations to approximate the circumstances of
daily diagnosis and asked doctors to discuss their
thought processes. Students played the patients. A
lab room became the doctor’s office, staged like a
regular exam room except for the two huge video
cameras mounted on the ceiling. Three real cases
provided the basis for the actors’ improvisation,
and Lee and Arthur concocted a “data bank” with
all the blood levels and X-ray results a physician
might possibly request. As the doctors worked,
researchers stood behind a one-way mirror
watching them “think aloud,” sharing the mental
considerations that usually remain private.
On the first day, Lee, Arthur, and their
colleagues got a preview of what they would find.
Watching their first physician, a chief of medicine,
the researchers expected events to proceed as all
the medical textbooks recommended. First the
doctor would interview the “patient.” Then he
would start ordering tests. Only later, after
reviewing the results, would the doctor start
outlining possible diagnoses.
But the work-up had barely begun when the
chief of medicine turned to the researchers to
announce his first diagnosis. What was going on?
At first the team figured the chief of medicine
must be a maverick, an outlier who followed his
instincts. But as more doctors came into the lab,
each one proceeded in a similar manner,
suggesting two, three, even four possible
diagnoses before even taking the patient’s blood
pressure. The maverick wasn’t a maverick at all.
The majority of doctors worked this way, exactly
the opposite of the meticulous decision tree that
textbooks advised.
But the method seemed to work. When Lee,
Arthur, and their team ran their data, they found
that doctors who made their first diagnostic guess
earlier in the appointment got the answer right just
as often as those who waited. If anything, it
looked like the more guesses were made early on,
the more likely the physician was to reach an
accurate diagnosis. So much for moving “from
symptom to sign to syndrome to disease,” as one
textbook prescribed. With one modest study, Lee
and his team had discovered that medical decision
making was far more complex than the textbooks
portrayed.
Lee thought he could take the research even
further. At Stanford, that was what he planned to
do—extend the problem-solving findings, fleshing
out their implications for education. And that,
ultimately, is what he did. He just didn’t realize
quite what form the transformation would take.

After the NIE conference, writing up a report

based on his panel discussion, Lee’s first move
was to borrow from his own work, crossing out
the word physician and writing teacher instead.
The clinical act of medical diagnosis became the
clinical act of teaching; the questions about which
lab tests to run became questions about how to
group the students, arrange the classroom, and
select a textbook. Where Nate had thought of
teachers as collections of behaviors, Lee borrowed
from the medical project and called them
“information processors.”
Lee had no expectations for his foray into the
study of teaching. Cognitive scientists had started
out by studying doctors, chess masters, and
investors because thinking was an obvious
prerequisite of their job. How much information
was processed by people who spent their days
telling small children, “One, two, three, eyes on
me!”?
But studying teachers by studying their thinking
turned out to be surprisingly generative. The
process-product findings that Nate Gage
championed might have been statistically
significant, but they often seemed to contradict
each other. It was important that every child stay
“on task,” but calling on students at random—the
best way to keep them focused—was not always
the best path to getting a good discussion going.
Similarly, after asking a question, the most
successful teachers waited a few extra seconds
before accepting an answer. But successful
teachers also tended to be the most brisk,
spending the smallest number of minutes between
topics. Pulling a single, clear answer out of the
process-product research was like trying to distill
laws from the Bible. One passage offered perfect
clarity, but the next said the complete opposite.
Lee, who had spent his grade-school years at a
yeshiva, met the task as perhaps no other
psychologist could. “Think about the tradition of
commentaries on the Talmud—this enormously
long historical tradition of interpretation in which
you never get to a settled conclusion,” says Gary
Sykes, who worked with Lee at Stanford. “It’s
brilliant intellectual work with a text. Lee took as
the text intellectual life in classrooms. And from
there, all was commentary and interpretation.”
Take the problem of timing. How could it
possibly be beneficial both to be fast, moving
quickly from task to task, and also to be slow,
pausing beyond the bounds of comfort before
calling on a student to answer your question? Lee
explored this teaching problem by examining what
he called “the anatomy of a turn.” The process-
product researchers had described the visible
elements of turns: teacher asks a question, time
elapses, student answers. But to really understand
the turn, you had to look at it from the teacher’s
perspective.
Building on others’ research, especially Mary
Budd Rowe’s study of “wait time”—the pause
between posing a question and selecting an
answer—Lee pointed out the logic in the apparent
paradox. For a teacher, each second spent waiting
for an answer held both promise and danger. On
one hand, the longer she waited, the more time the
students would have to think. This was good. On
the other hand, the sooner she broke the silence
with the correct answer, the lower was her risk of
exposing the class to a useless diversion. This was
also good. The wisdom and peril of pausing were
both true, and if you thought that didn’t make
sense, well, that was true too. Wait times, Lee
concluded, were “blessings dipped in acid.”
The question for teachers, as for doctors, was
not, What is the best behavior? It was, How do I
decide which of many behaviors to deploy for the
case at hand? It was a problem of diagnosis.
Teachers had to locate their pupils’ pathologies,
determine a best intervention, and act.
With doctors, diagnosis and treatment had clear
beginnings, middles, and ends. With teachers, the
questions kept coming. Since the pathologies—
that is, everything the child didn’t know—were
not physical but mental, how could teachers
diagnose them? How could they understand what
a child had failed to learn? And if they did manage
to teach successfully, how could they confirm it?
There was also the problem of scale. “The
teacher,” Lee realized, “is confronted not with a
single patient, but with a classroom filled with 25
to 35 youngsters.” Even if a teacher could locate
pathologies and somehow do it for all her
students, how did she manage to deploy the
correct interventions, all at once, to the entire
group? “The only time a physician could possibly
encounter a situation of comparable complexity,”
Lee concluded, “would be in the emergency room
of a hospital during or after a natural disaster.”
Studying teachers, he realized, was just as
important as thinking about doctors; in fact, “it is
far more germane.”

The National Institute of Education conference

came and went quickly. Lee moderated his panel;
submitted his summary report, advocating the
usefulness of studying teachers’ decisions; and
soon he was back at Michigan State, working with
doctors. He might have forgotten about the trip
altogether, had NIE not sent him a call for
proposals to build a new research and
development center to study teacher thinking and
decision making.
Lee knew that his proposal would be a long
shot. The likely list of applicants included
Stanford University and his colleague Nate Gage.
And since writing his famous handbook, Nate had
made Stanford into the country’s leading source of
research on teaching. MSU, by comparison, was a
“cow college,” better known for training teachers
than for studying them.
But Michigan State won. Among the losers
were several of the universities that had been
pulling government grants for behaviorist
education research for years, Gage among them.
“Nate lost his grant too,” says Garry McDaniels.
“In the old days they always gave him the grant.
But the work that he had done had been going on
for so long that I was convinced that it had
reached its end.”
Lee is fond of quoting a line from the
psychologist Jerome Bruner about narrative. One
of the cognitive revolution’s leaders and an early
scholar of teaching, Bruner wrote that narrative is
fundamentally composed of “the vicissitudes of
intention.” A protagonist sets out to do one thing,
but along the way something unpredictable
happens, and he decides to do another thing
instead.
 Lee set out to study thinking. By understanding
minds, he thought, he could help improve
education, the work of shaping them. The thing
that happened along his way—the call from Nate
Gage—led him to change not his intention, but his
method. Doctors had provided a neat keyhole into
the mind, but it turned out that another group of
professionals offered a bay window. Teachers not
only had to think; they had to think about other
people’s thinking. They were an army of everyday
epistemologists, forced to consider what it meant
to know something and then reproduce that
transformation in their students. Teaching was
more than story time on the rug. It was the highest
form of knowing.
At a university, traditionally the highest degree
holders are called master or doctor. “Both
words,” Lee discovered, “have the same
definition; they mean ‘teacher.’ ” What was the
best way to show you really understood a subject,
if not to teach it? And what was the best way to
use research to improve education, if not to study
teaching?
Without realizing what he was doing, Lee had
stumbled on Dewey’s lost project. Teaching was
indeed the science of all sciences, the art of all
arts, as Dewey’s predecessor Francis Parker had
put it. And now, thanks to Nate Gage’s nudge,
Nixon’s investment, and his own lifelong
obsession, Lee was going to pick up on the work
Dewey and Parker had never finished.
Lee had written in the NIE panel’s concluding
report that “gifted practitioners are capable of
performances which our best theories are not yet
capable of explaining, much less generating or
predicting.” Future research on teaching, then,
should explore the talents of the best teachers—
the “wisdom of practice,” he called it. All he
needed to do was find the great practitioners.

Lee Shulman was no teacher, but he became one

of two seminal figures in modern thinking about
policies for improving the quality of teaching. The
other figure was not a teacher either. And whereas
Lee focused on education after spending time
inside of it (or at least inside of a school of
education), Eric Hanushek came to it wholly from
the outside. He would go on to have as much of an
influence on education as Lee did, if not more. But
he never worked at an ed school.
Hanushek became fascinated with schools in
the summer of 1966. He was nearing the end of
graduate school in economics at MIT, still lacking
a dissertation topic, when he stumbled on a
remarkable story in the newspaper:

Washington, D.C.—The Johnson

administration Thursday was accused of
ignoring results of a Federal
investigation of inequality in city school
systems because of political
implications . . .
 After a survey of 600,000 school
children and 60,000 teachers, the report
concluded that pupils from poor families
left school “with greater deficiencies”
than when they entered . . .

“This report means that all our

education plans—increasing spending
per pupil, more and better libraries and
books, education devices—won’t solve
the crisis in our schools,” said
[Connecticut congressman Abraham]
Ribicoff.

Given the redistributive goals of Lyndon

Johnson’s “Great Society” programs, it made
sense that the administration would want to cover
up the report. If the study, by a Johns Hopkins
sociologist named James Coleman, was right, then
one of the most expensive educational
interventions in history had failed. According to
Coleman, giving schools additional services—
including more per-pupil spending, the supposed
antidote to underachievement—did not help poor
and African-American students overcome the
challenges of their environments.
Hanushek couldn’t quite believe it. “If in fact
schools don’t make much difference,” he thought,
“why are we continually pumping more and more
money into schools to try and improve them?”
There had to be something else going on, a lurking
variable masking the money’s impact. But what
could it be? Running through Coleman’s data, a
massive set drawing from 645,000 students and
more than three thousand schools across ninety-
three different variables, Hanushek found no
mistakes of consequence. Nor did a working
group convened at Harvard to vet the research.
Coleman’s conclusions largely held up.
But what about other data? Hanushek pulled
together a data set from a school district in
California—much smaller in scope than
Coleman’s national sample, but with two
advantages. First, instead of capturing a snapshot
of one year in students’ lives, the California data
followed students longitudinally. Second, the data
broke down students not just by school, but by the
teachers they’d had. The extra detail enabled
Hanushek to get more specific than Coleman; he
could go beyond whether schools made a
difference and determine whether individual
teachers had an effect as well.
The effect of teachers was no simple thing to
measure, even with the better data. Countless
factors undoubtedly influenced students’
performance in schools, from genetic disposition
to the size of their parents’ vocabularies. How
could Hanushek discern the teacher’s influence
amid all these other variables?
The education literature offered no advice, but
another area of economics did: the study of
industrial production. Like teachers, factories
receive certain raw products (steel, coal, plastics)
and then put their own unique spin on the
business of transforming them into, say, a Chevy
Camaro. To measure the productivity of the
manufacturing process, economists had to extract
the value offered by raw products from the value
provided by the plant assembling them. They did
this by looking for patterns. What was the value of
the raw products before they came to the factory,
and how much did that value rise or fall after the
manufacturer had its way with them?
Applying the idea to education, Hanushek could
control for the effects of nonteacher variables,
from home background to past performance, by
searching for deviations. “If you follow an
individual kid and you see him on some learning
path, and then one year, all of a sudden, he learns
a lot more than in another year, or all of a sudden
he learns a lot less,” Hanushek explains, “that
gives you a hint that maybe it’s something specific
about the teachers, or something specific in that
year. Then, if you see that all of the kids in the one
class have this jump or this fall in performance,
then you start to believe that it’s something in that
class.”
The California data was full of such jumps. If
Hanushek’s method was right, teachers did make
a difference, and the difference was big. Later he
managed to put a number to the effect. Students
assigned to the best teachers, he calculated,
progressed by the equivalent of a whole grade
level more than students assigned to the worst, as
measured by test scores. By Hanushek’s method,
teachers could do what the Coleman Report
suggested schools could not: they could offset the
disadvantage of poverty.
Perhaps Hanushek’s most influential finding
stemmed from his comparison of teachers’
“effectiveness” (the educational equivalent of
productivity) to other characteristics, especially
salary. The amounts paid to teachers were based
on how many years of experience they had and on
how many degrees they had earned; a master’s
won a salary bump, and every additional graduate
class won another. Yet the salary inputs seemed to
have no bearing on the output: teacher
effectiveness. Experience did matter, but only up
to a certain point. In productivity terms, there was
no difference between a teacher who’d been
teaching for three years and a teacher who’d been
teaching for thirteen.
Writing his final dissertation, eventually
published as a book called Education and Race,
Hanushek turned the observation into a suggestion
—one that would reverberate for years into the
future. If school districts stopped rewarding
graduate study and experience, then they could
redirect their investments into something more
efficient: “Teacher Accountability,” he called it.
Accountability would draw on the statistical
method that Hanushek had adapted from studies
of factories. “This procedure,” he explained,
“allows the ranking of teachers on the basis of
teaching ability.” Ranking and then rewarding
teachers according to their effectiveness might
create some problems—including, he suggested,
“problems arising from attempts to ‘teach the
tests.’ ” (Held accountable for how their students
performed on tests, teachers might emphasize the
exam material to the exclusion of other, equally
important lessons.) But, Hanushek wrote, “while
we may not be at a point now where we trust
standardized tests to hold up under concerted
attempts to foil them, conceptually the problem
appears soluble.” A decade later, Hanushek gave
his method a name: “value-added.”*
In the same book, Education and Race,
Hanushek made one more intriguing point. Like
Coleman, he had only examined educational
investments and their effects—comparing, in
economic parlance, education’s “inputs” to its
“outputs.” He had not studied education’s vast
middle. “The black box of the production
process,” he called it. That is, classroom teaching
and learning. He had, in other words, followed the
trail that Nate Gage and Lee Shulman were
blazing, looking not just at schools but at teachers,
yet he had followed them only so far. He looked at
teachers’ effects, but not at their work—at
teachers, but not at teaching.
Hanushek made the observation as an aside, but
the decision to overlook teaching’s “black box”
would prove just as influential as his “value-
added” innovation. By studying teaching, Lee
Shulman and his colleagues were about to explode
many common ideas about how it worked,
including the myth of the natural-born teacher.
Hanushek, meanwhile, ignored teaching and, as a
result, ignored how teaching worked. He could
read his value-added research and draw the
simplest conclusion, the one that matched what
everyone already believed: some teachers were
bad, most were fine, and a few were wonderful—
as if they’d been born that way.
 * The other researcher with a claim to having
invented value-added calculations of teacher
effectiveness is the statistician William Sanders,
who developed the Tennessee Value-Added
Assessment System (TVASS).
 2
A TEACHER IS BORN

Deborah Loewenberg Ball arrived at the Spartan

Village school in East Lansing, Michigan, in the
fall of 1975. Technically still a college student,
Deborah had never taught her own class before.
Barely a decade separated her from the fifth-
graders. But it did not take long for the other
teachers to size her up. Mindy Emerson, who
taught across the hall, could tell almost
immediately.
As Mindy saw it, there were two types of
teachers: those who chose to teach and those who
were born to it. For the latter, teaching was not a
job; it was a calling. (Mindy, for her part, decided
she would be a teacher on her first day of
kindergarten. “I walked in, I smelled the chalk,
and I knew I was home,” she says.) From
Deborah’s first day, Mindy could see that she, too,
had been called. Deborah had finished high school
early, and by the time she turned twenty years old,
she spoke four languages. “She could have been
whatever she wanted to be,” Mindy said. But she
was visibly giddy at the chance to teach, and her
gift was undeniable.
Deborah had joined Spartan Village as a
member of Michigan State’s Elementary Intern
Program, a teacher-training course that culminated
in a one-year, immersive classroom experience at
a local elementary school. Other new teachers
went through a predictable litany of challenges.
They couldn’t get the students to listen, they tried
too hard to be the students’ friends, they doubted
whether all the children could really learn, they
struggled to feel comfortable in the new role. In
one memorable case, one of Mindy’s old
classmates at MSU came back from her first
classroom experience in shock: “The children are
touching me!”
 Mindy couldn’t blame them. Teaching, after all,
wasn’t like other jobs, where new hires take on
new responsibilities only after they’ve mastered
simpler ones. “With us, when the kids walk in the
door, you’re on,” Mindy said. “It doesn’t matter if
you’ve taught one year, ten years, or thirty years.
You’re on.” And you are alone. “You feel like the
lone ranger.”
But if Deborah felt any of the typical jitters, she
did not show it. She had a calm, gentle way with
the children, connecting even with the ones who
came unable to speak English (and there were
many of them at Spartan Village, a public school
literally on the wrong side of East Lansing’s
tracks, with more than the usual share of
immigrants). Her discipline struck that rare
balance, leaving the children both happy and well
behaved. Mindy had never seen better
penmanship: neatly lined print and cursive that
made the discriminating heart flutter. And her
lesson plans! Organized and comprehensive, they
left nothing to chance. All Mindy could think was,
Wow. “A natural-born teacher,” she said
confidently, thirty-six years later.
Even the new principal, a young woman named
Jessie Fry, had to admit that there was something
extraordinary about Deborah. And if Jessie had
seen any weakness, she would have pointed it out.
Jessie was strict. The teachers made no secret of
their frustration with her mandates. Lesson plans
turned in a full week in advance; goals and
objectives outlined for every single child; visits to
every family’s home by Thanksgiving;
unannounced classroom observations, and always
followed by a pointed note “FROM THE DESK
OF JESSIE J. FRY.”
Deborah was different. “Great!” and “Thank
you!” and “You are coming along nicely,” Jessie
wrote, because when she watched Deborah teach,
she couldn’t think what to improve. Her only
advice was to slow down. “She was animated, and
she talked very fast, very fast,” Jessie says. “I said,
‘Okay, Deborah, you got to slow this down!’ ” By
the end of her first year, Jessie had violated an
unwritten law of the Elementary Intern Program.
Tradition held that interns left after graduation to
find a permanent job; that way, another intern
could have a chance to teach the next year. But
Jessie asked Deborah to stay permanently. Within
two years, Deborah was petitioning her to create
an unheard-of first-, second-, and third-grade
combination class with another teacher. And, call
Jessie naïve, she heard herself saying yes.
The staff slipped the custodian a little
something extra, and he tore down the wall
separating the two classrooms. Deborah,
meanwhile, dragged in a working refrigerator and
stove off of the street. For cooking lessons, she
explained. (In high school, she had worked in a
bakery; that was where she had met her husband,
Richard, whose mother decorated the cakes.) She
used baking as a teaching opportunity and also to
help raise the money needed to take her students
on a train trip to visit the Kellogg Cereal factory in
Battle Creek. Together, she and her second-
graders opened a “dessert restaurant”—staffed, of
course, by eight-year-olds.
Some of the older teachers, already annoyed
with Jessie, began to grumble. Why didn’t every
classroom have a stove? And how could they be
sure that this special combination class wouldn’t
skim off all the best students? If Deborah is so
creative, they suggested, Jessie should give her the
troublemakers. But as the years passed, the
grumblers became acolytes. They rearranged
schedules so that their students could use
Deborah’s stove. They requested subs so that they
could leave their own classrooms and watch
Deborah teach. And sometimes, they invited her to
join them in theirs.
By 1980, only one person in all of Spartan
Village doubted Deborah’s gift. That was
Deborah.
Mindy, Jessie, and the others had one thing right.
Teaching had become Deborah’s passion. But she
did not find it easy. To the contrary, her favorite
part about teaching was how hard it was. Before
deciding to become a teacher, she’d majored in
French, taking classes in phonology and culture
and philosophy. She seemed to enjoy courses in
direct proportion to their difficulty. She also
needed to study something that would lead quickly
to a decent job. (Rich, the baker’s son, was her
high school sweetheart. Once they married, at
nineteen, Deborah no longer had financial support
from her parents.) The Elementary Intern Program
offered both an intellectual challenge and a job.
She became particularly fascinated by the
puzzle of how to teach children to read—a mystery
for any literate adult, but especially so for
Deborah, who had learned to read at age four and
could not remember a time when words weren’t
synonymous with sounds and meanings. The job
of a teacher was to explain to others what you
already knew by heart. To teach children to read,
Deborah had to understand all the things that
make reading hard. Adults know intuitively all the
different ways that “ea” can be pronounced,
depending on the context. But teachers needed to
be able to explain that break, beak, read (present
tense), and read (past tense) all make different
sounds. Same with the “r-controlled vowel.” Few
adults could explain it, but children certainly
noticed the confusing way that “r” can transform a
vowel from its usual sound to something entirely
different—so that the “e” in her is nothing like the
“e” in hen. Those children’s teachers had to know
how to help them figure that out, selecting specific
words for each student to work on: milk, store, lot,
her; say, ran, down, right; laundry, laundromat,
iron, fall, bag, them, from, girls. Each group of
words had a purpose, carefully selected to match
exactly what the children needed to learn.
The next challenge was science, the other
subject Deborah taught in that first year at Spartan
Village. (The school was organized into
departments, like a high school, with different
teachers handling different subjects.) The
Elementary Intern Program had no science
equivalent to its reading course, and Deborah’s
own knowledge of science was limited to what she
had learned in high school, where her teacher had
staged a debate on the topic of whether evolution
should be taught in school. (The teacher’s
position: no.) At Spartan Village, Deborah had to
learn the content and pedagogy simultaneously.
How did electric circuits work, again? What was
the difference between a closed and an open
system, and how did this apply to the systems that
her students were supposed to study (weather,
electricity, and heat)? What should she expect out
of a third-grade science fair project? And why did
the textbook think it was a good idea for third-
graders to grow cotyledons and brine shrimp?
For every teaching problem Deborah conquered
in those early years, a new one seemed to emerge.
The greatest arose in 1980, when, teaching fifth
grade, she discovered that her students were
struggling with math. And not just a few of them.
All of them. One day, twenty-five notebooks
would fill with perfect little number houses as she
walked them through the steps of long division.
The next day, left to reconstruct the steps alone,
the children would flounder. Weekends were her
worst enemy. “My students would go home on
Friday able to solve a long division problem and
on Monday no longer seem to remember how to
begin,” she wrote, reflecting on the experience in
an essay.
Worst of all, the students’ misunderstandings
included even the simplest ideas. Trying to
subtract a number like 55 from a number like 72,
the students would subtract up, taking 2 from 5
instead of 5 from 2, and decide the answer was 23.
Deborah handed out bundles of sticks to represent
tens and ones and repeatedly explained the idea of
“borrowing”—take ten from the 7, carry it over to
the 2—but her students would still make
mistakes. Reading a word problem like “So-and-
so has 12 and so-and-so has 20. How many more
does so-and-so have than so-and-so?” the students
would see the word more—and decide to add. But
the question was asking how many more: the
difference. A child who added because of seeing
the word more either didn’t understand the
meaning of addition or hadn’t read the words.
It did not escape Deborah’s notice that a sizable
number of these clueless students came from the
very same first-, second-, and third-grade math
classes she had taught with such apparent success
a few years before. What had she been doing
when she thought she was teaching? What had
they been doing when she assumed they were
learning?

Deborah’s first move was to ask other teachers for

help. But almost everything they suggested, she
had already tried. She followed the textbook,
divided the children into “skills groups” (the
Triangles, the Diamonds, the Circles, they were
called), constructed clear explanations, and used
paper cutouts to illustrate key ideas. Yet when her
tests came back, they told the same story. The
children had not understood.
Stumped, she took her problem to a professor
back at Michigan State’s College of Education.
The professor suggested that she try out a new
experimental curriculum for elementary school
math.
“Experimental” put it mildly. For first grade
alone (Deborah had moved again after teaching
fifth), the teacher’s guide filled more than seven
hundred pages. The curriculum was strange,
suggesting that lessons move forward according to
scripted discussions about, in one case, an
elephant named Eli. (“There is an elephant named
Eli who lives in the jungle and is always very
hungry. What do you think is Eli’s favorite food?”
the guide suggested teachers ask, as a way to
begin a lesson on negative numbers.) But the
baffling methodology began to make sense when
Deborah steered the students through the
“dialogues” that punctuated each lesson.
On its face, the idea of a math lesson as an
extended conversation, rather than a set of ideas
and related practice problems (subtraction with
regrouping, say, or counting), was as unusual as
using an elephant story to teach math. Deborah
had held discussions in reading, where students
could talk about stories—discussing the characters
and what the students thought might happen next
—and in science, where they could guess the
results of an experiment. But she had never led a
discussion about math, and despite the assurances
in the teacher’s guide, she wondered whether her
students would really have much to say. 2 + 2
always equaled 4. What was there really to
discuss?
Yet when she tried it, the curriculum began to
show its wisdom. Eli the elephant, for example,
turned out to love peanuts—and his peanuts took
two forms: regular peanuts and “magical” ones,
which were like negative numbers. When a
regular peanut and a “magical” one met, it was
like adding 1 and –1: both disappeared. Following
the story, Deborah’s first-graders quickly picked
up the idea of negative numbers—a concept that
often befuddled much older students.
In discussions, meanwhile, the challenge
wasn’t in getting the students to talk, but in
making sense of all they had to say. The deeper
into the math the students got, the more questions
they presented to their teacher. For instance, are
there “afinidy” possible ways to use a 24-story
building’s elevator to get to the second floor—or
are there 25?* What exactly did it mean to add a
negative number? What about to subtract one (for
example, 3 minus −5)? Even rudimentary ideas
now raised complicated questions that Deborah
felt unprepared to answer. Talking about math
was surprisingly interesting, but for the talking to
lead to learning, it seemed the teacher needed to
know something more.
Once again, Deborah turned to Michigan State,
this time not to the College of Education, but to
the Department of Mathematics. As an undergrad,
she had tested out of math completely. Now she
started from the beginning, venturing through
elementary algebra, geometry, and calculus while
her students studied addition, subtraction, and
fractions. The courses produced some exciting
aha! moments, as when the details of limits and
integrals helped her steer students through a
problem about area.
The most important revelation arrived in her
final course: Number Theory. Taught by the chair
of the math department, a professor named Joseph
Adney, the class addressed a mathematical subject
Deborah hadn’t studied before (one deeply
relevant to an elementary school teacher). And,
even more important, Adney taught it in a new
way. Instead of marching methodically through a
list of concepts, he invited his students to discover
the ideas for themselves. He’d write something on
the board and then ask, with a straight face, “Is
that always true?”
Deborah found herself thinking about these
problems for hours. Sometimes the writing on the
board took the form of a statement: a “conjecture,”
Adney said. A conjecture was like the
mathematical version of a hypothesis, a question
without a question mark. The sum of two odd
integers will always be even, for instance, or the
sum of an even and odd integer will always be
odd. Is that always true? When the students came
up with arguments—proofs—he would present
everybody’s attempt without prejudice. Then
they’d have to defend their reasoning before the
class. If anyone could find a counterexample, then
poof! The conjecture would explode.
Deborah had seen proofs before. But she had
never been asked to make one from scratch, and
she’d never realized how many different proofs
could support a single statement. Adney took
special interest in oddball proofs (Deborah, with
her abbreviated math background, was a
connoisseur of these). The more diverse ideas they
could pull together, the richer was their
exploration of the math.
Discussions in Adney’s class were not just fun
ways to pass the time. They were vital to the work
of doing math. By talking about math—puzzling
over problems, making conjectures—they
practiced it. In the process, Deborah—the classic
humanities type, who never took a single left-
brained class in college—fell in love with the
subject. The math she’d learned in school was
dull, rote, blah—“uninspiring at best, mentally
and emotionally crushing at worst,” she wrote not
long after taking the class. The procession of rules
and procedures flattened any latent pleasure in the
neat finality of the right answer. (“How many
more” = subtraction, she’d reminded her students
at Spartan Village just a few years before.
Subtract the ones first. And always subtract
down.) Sometimes the procedures made sense.
More often, they were just a predetermined path to
the right answer. Adney presented a different
subject altogether. In his class, math was
powerful, rich, even awe inspiring. For days, a
problem boggled. But then someone would offer
another way of looking at it, and suddenly it
would make sense.
What would it look like to teach elementary
school children math in the way she was learning
it? Adney, who taught undergraduates, could take
her only so far. Deborah needed another resource.

A few years after her classes with Adney, Deborah

decided to teach a summer school section outside
her usual repertoire. She’d just taken a class on
research methods, and the material had struck her
as potentially powerful for eight- and nine-year-
olds. In particular, she wanted to teach inferential
statistics, a kind of math in which students use
tools like curves and intervals to draw conclusions
about data. But, finding no research or curriculum
on how to teach the subject to young children,
she’d had to create the course from scratch. This
proved more challenging than she anticipated, so
she decided to recruit help. Not a coteacher—the
class had only eighteen students, a perfectly
manageable number for one person. What
Deborah needed was another brain. Better yet, a
dozen of them.
Recruiting teachers was simple; by
participating, they could cross off a required
professional development session. Every day that
summer, before the children arrived, the group
walked through the lesson Deborah had drafted,
trying out problems, imagining how students
might react, and discussing what Deborah might
say in response. When the lesson began, the other
teachers served as extra eyes and ears, studying
each child and noting what they did and did not
understand. At the end of each day, Deborah had
the students leave their notebooks behind so the
teachers could study those too. Then they all sat
together and talked about what had just happened.
What did everyone think about what this or that
student had said? What ideas did the class still not
seem to grasp? What should Deborah do
tomorrow?
In a way, this method was no different from her
normal practice. At Spartan Village, she
frequently pulled other teachers into her class to
help her solve problems. But at Spartan Village,
moments like these were merely friendly favors
offered by busy colleagues. At the summer
program, the group’s focus was sustained, the
tone serious; it was as if they were not in an
elementary school, but in a laboratory. Or maybe,
Deborah thought, a surgical theater.
Technically, only Deborah taught the children.
But really she was the group’s surrogate—a kind
of “pedagogical daredevil,” she decided, trying out
ideas on everyone’s behalf. “Whatever we decided
to do,” she wrote later, “I was the one who had to
try to make it fly.” The group, meanwhile, formed
her safety net, making sure the students didn’t
become casualties of the experiment.
The students learned, and, just as importantly,
so did Deborah. Looking back, she says it is
impossible to recall any one moment of epiphany.
Her teaching was evolving quickly, and she hadn’t
yet begun to make records capturing each lesson
and the discussions that followed. But similar
public lessons, held years later, at the annual
program known as the Elementary Math Lab, shed
light on what she and those first co-conspirators
must have seen that first summer in 1984.
During a lesson in July 2012, a group of
observers took notes from bleacher-style seats as
Deborah asked a class of rising sixth-graders to
consider a rectangle. The rectangle looked like
this:
 What fraction of the rectangle, Deborah asked
the students, is shaded? The first student she
called on, a girl named Anya, gave the correct
answer, ¼, explaining how she had drawn an
additional line to help her solve the problem:

But when Deborah asked for comments on

Anya’s answer, a boy named Shamar, with puffy
cheeks and long dreadlocks, said something
curious. “I think the answer was one-half and a
one on the side of it,” he said. The mysteries
multiplied when Deborah brought him up to the
board to explain. “1 ½,” he wrote. But he kept
saying the number backward, as if reading from
right to left: one-half first, then one, which he
called the “remainder.”
What was he thinking? Under what
assumptions might 1½ make sense? Scrutinizing
Shamar’s responses during the debriefing
Deborah held after the lesson, once the students
had left, one group of observers pieced together a
hypothesis. Perhaps he had flipped the question.
Instead of looking at the fraction of the rectangle
that was shaded, he focused on the fraction that
was empty. He might have even seen the image as
its inverse:

Others focused on Shamar’s description of 1½

as “one-half and a one on the side of it”—more
like ½ 1 than 1½. Maybe he had transcribed the
inverse image into the numbers that it resembled:
½ on the left, 1 on the right. If you saw math as a
set of rules and procedures, as so many children
were taught, then you did not think about fractions
as holding meaning. They were simply numbers
with lines through their middles.
Whatever his exact thought process was,
Shamar had clearly become confused about an
idea that stood at the heart of fractions—one that,
over the years, Deborah and those who joined her
at the lab had come to see as a typical stumbling
block for children (and many adults): the idea of
the whole. To answer any fractions problem, you
had to define the thing that you wanted to know a
fraction of.
In this case, the whole was the largest
rectangle, the one that also happened to be a
square. Shamar’s answer suggested that he had
defined a different rectangle as the whole—the
one that, with its longer sides, looked more like a
child’s idea of a rectangle. (Children often do not
understand that all squares are rectangles too, just
with equal-length sides.) If you defined that
slimmer rectangle as the whole, and you accepted
Shamar’s inversion of the shaded and empty
space, then 1½ made sense.
The misunderstanding offered an opportunity.
Encountering the math through the students’ eyes,
the group could figure out what needed to be
clarified. Then together, these observers could
figure out what Deborah might say and do to get
Shamar to understand the importance of defining
the whole.
There were many possible paths into the
material: questions to ask, explanations to give,
problems to assign. Over the course of many
teaching labs, the most productive methods and
problems made themselves clear. One good
approach was to have students with different ideas
present them to the class. Some of them
undoubtedly shared Shamar’s misunderstanding,
in one form or another. (Even adults in the room
could forget sometimes that a fraction was
meaningless without its unit.) Listening to their
peers could help the confused students sort out
their ideas. When a boy named Eduardo jumped
in to clarify, Shamar seemed to understand his
own idea better too. Then, when Eduardo
explained why he agreed with Anya anyway about
¼, Shamar decided to change his answer.
The lab group also studied turns—which
students Deborah called on, in what order, and
what she asked each of them to do. Her decisions
had come to hold more significance over time, as
she learned the many different types of turns, each
with varying dimensions of both academic
difficulty (offering a math fact versus offering an
interpretation) and social risk (giving an answer
even though you hadn’t raised your hand was
moderately risky; coming up to the board and
offering a detailed description of your incorrect
answer, much more so).
Other considerations mattered too. To make
sure everyone participated, it was advisable to call
on the three students who had not spoken yet, but
this might not be a good strategy if all three had
the same answer. Order also made a difference. In
certain cases, there was wisdom to calling on, say,
Anya before Shamar. Shamar’s answer had
assumed an idea that Anya’s, by drawing in the
previously invisible line, made explicit—that
fractions made sense only if they formed equal
parts of the whole. The discussion would go better
if Deborah could get Anya’s idea on the table
before tackling Shamar’s confusion.
Over time, more conventions emerged. It was
crucial, for instance, to make sure that students
did not talk just to Deborah, but to the entire class.
Everyone had to learn everyone else’s name. Then,
instead of saying “that weird idea about one-half
and one beside it,” they could simply say
“Shamar’s idea” or, if Shamar posited an
argument, “Shamar’s conjecture.”
Deborah came to see these named conjectures
as “fence posts” for a productive conversation.
The students could peer backward over the
landscape of their evolving understanding and
name the key turning points. And when another
part-whole misunderstanding inevitably arose,
they could undo it more quickly by thinking back
to Shamar’s idea and the reasons it didn’t hold up.
The precise wording of questions also mattered,
and the lab group spent hours debating Deborah’s
constructions. That same year, hoping to introduce
students to the concept of infinity—one of those
dazzling ideas that could spin in a student’s mind
for days—she had presented a problem with
endless answers. Then, asking the students to
guess how many solutions they could come up
with, she added an extra question, apparently as
an afterthought. “After you write down your
answer,” she asked, “can you write how long it
will take [to come up with all the solutions]?”
The lab group devoted several minutes to
considering the value of that extra question. By
asking the students to write down how long
writing the solutions would take, hadn’t Deborah
suggested that writing them all down was actually
possible? And so, argued a teacher from Chicago,
hadn’t the question inadvertently tilted the
students away from the correct answer? But the
question had done exactly the opposite, another
group of teachers argued. “We could be doing this
forever!” the students might realize, thereby
jumping closer to the key idea.
The group dissected the problems Deborah
selected too. On the day of Shamar’s
misunderstanding, another confusion had arisen—
this one not about the whole, but about the parts.
Counting the shaded part and then counting the
total number of parts, some students had called
the fraction ⅓. They had missed what Anya saw
about drawing a line to make the parts equal. The
class discussed why dividing a shape into equal
parts was important, but some lab observers
wondered whether all the students really grasped
this idea. One person offered a proposal. In the
next class, why not present a problem that forced
the students to draw even more lines? Something
like this:

The next day, Deborah added the problem to the

warm-up.
Back at Spartan Village, the lessons from the
early summer labs—which began in 1984 and
continued for years after—were combining with
the new curriculum to create a kind of magic.
Now that the students conjectured, reasoned,
argued, and proved, they were building one idea
on top of another. They sometimes forgot what
they’d learned, like all students do. But now,
when they stumbled, they could pick themselves
up. Deborah saw it happen one day a few weeks
into the fractions unit, when two third-graders
were puzzling over a problem about cookies that
involved the number .
“How can we have this?” Betsy asked Jeannie,
pointing to the confusing fraction.
“I don’t know,” Jeannie said.
“Four twoths?” Betsy asked.
“We take something and divide it into two parts
. . . and take four of those parts?” Jeannie asked.
“I’m confused,” Betsy said.
“Me too,” Jeannie said.
Just then, Sheena walked up. “Four halves,
isn’t it?”
“Yeah!” Betsy exclaimed. “Four halves! Halves
are two parts. So . . .”
“So we need two cookies and cut them each in
half, then we have four halves,” Jeannie said.
“One, two, three, four. Twoths. I mean halves.”

While Deborah worked on the puzzle of how to be

an effective teacher, another question pulsed in the
back of her mind: Why hadn’t she learned any of
this before? As a double major in French and
elementary education, she’d taken a methods
class, supposedly about how to teach math. Later,
of course, she’d taken nearly the entire strand of
university-level math classes. But none of these
classes had prepared her to help children learn
math. That class did not exist.
The trouble, she suspected, lay in the kind of
knowledge one needed to teach well. It fit in
neither the category of general education nor that
of pure math, though both kinds of knowledge
were helpful. In addition to the math itself, she
reasoned, math teachers needed to know the kinds
of activities and tasks that turned a student’s
slippery intuition into solid understanding. Not
only did they have to master procedures, concepts,
and the special cycle of conjecture to argument to
proof, but they also had to know the students: how
much they were capable of; the iterative, circling
way in which they learned; and the kinds of
representations—the particular configurations of
pictures, numbers, and blocks—that best helped
them to understand.
No wonder the class did not exist. It would
have had to teach a subject with no name. Even
Deborah—who was now both a teacher at Spartan
Village and the special “math helping teacher” for
all East Lansing elementary schools, not to
mention a doctoral student at Michigan State’s
College of Education—could not articulate the
parameters of this knowledge. But that began to
change in the mid-1980s, when she decided to
study teachers’ mathematical knowledge as part of
her dissertation.
Her hunch was that Michigan State was still not
equipping future teachers with the knowledge and
techniques they would need. But to be sure, she
devised a test, a short set of teaching problems
that she thought math teachers should be able to
answer, and gave it to education majors about to
graduate.
One question described a group of eighth-grade
teachers who “noticed that several of their
students were making the same mistake.” When
multiplying large numbers, like 123 × 645, their
students “seemed to be forgetting to ‘move the
numbers.’ ” Their work looked like this:

when it should have looked like this:

“While these teachers agreed this was a

problem,” Deborah’s question went on, “they did
not agree about what to do about it.” She turned
the question to the future teachers. “What would
you do if you were teaching eighth grade and you
noticed that several of your students were doing
this?”
To answer the question well, Deborah decided,
teachers would need to identify the ideas the
students lacked. Two were particularly important.
One was the concept of place value, the
convention that gives integers different values
depending on where they sit, so that the second 3
in 79,335 actually means 30, whereas the first one
represents 300. The second missing idea was the
distributive property, which explains why the
common procedure depicted in her second picture
worked—why, in order to find 123 × 645, you
could add up the results of three multiplication
problems (123 × 5, 123 × 40, and 123 × 600). By
not moving the numbers over, eighth-graders
showed they had followed a procedure blindly,
and then fallen over the inevitable cliff. To help
them understand the steps that did make sense, a
teacher would have to acquaint them with the
reasons why the algorithm worked.
As it turned out, of the nineteen future teachers
Deborah interviewed, only five mentioned either
idea. Most described how they would remind
students of the right steps, especially what a
teacher named Teri called “shift[ing] things over.”
Some referenced the idea of place value, but
obliquely, without remembering its name,
meaning, or why it was important. “Since you are
working with such a large sum,” explained a
teacher named Rachel, “you have to know how to
work in the thousands, you know, to keep your
numbers that way.”
They had taken classes in both education and
math, but the Michigan State students didn’t have
another kind of knowledge required for teaching
—“pedagogical content knowledge,” Lee
Shulman called it. Not just teaching methods or
the intricacies of the subject, but the perfect mix of
the two.
Even future high school teachers with a joint
major in math struggled to produce clear
explanations. More often, like a teacher named
Barb, they remembered the reason for the
procedure only in the course of trying to explain it,
and then stumbled through it. “ ’Cause you’re
going to take 5 times that, and you take 40, and
then 600, and you can see where those zeroes
come from,” Barb told Deborah. Deborah could
see, but would Barb’s student?
Other questions were as puzzling to the math
majors as to everyone else. One query asked the
future teachers to come up with a way to represent
a common part of the curriculum, division by
fractions. Deborah picked a specific problem: 1¾
÷ ½. Math teachers, she reminded her
interviewees, often try to explain problems by
relating them to real-world situations or “models
that make clear what something means.” Could
the interviewee think of a situation or model for
1¾ ÷ ½?
A good answer would help the students
visualize what it means to figure out how many
½’s go into 1¾. In her dissertation, Deborah
described one possibility: “A recipe calls for ½ a
cup of butter. How many batches can one make if
one has 1¾ cups of butter?” The answer was 3½
batches, because 1¾ cups of butter contains 3½
half cups. The story not only represented the
problem; it clarified the concept, offering the
students one way to imagine division (as creating
groups of a certain size) and cutting through the
confusion of defining the whole by making the
unit clear. 3½ whats? 3½ halves.
Once again, of the nineteen interviewees, only
five came up with representations that Deborah
could call mathematically correct. And of those
five, only one made up a representation that was
decipherable, though it lacked the crispness of the
butter example. (The teacher said she would use a
number line to mark off 1¾ and ½, and then ask
the students how many ½’s went into 1¾.)
The other four teachers-in-training offered
examples that, while correct, strained even
Deborah’s imagination, like one by a young man
named Terrell. He said he would have students
imagine getting three pizzas: one whole pizza, ¾
of another pizza, and ½ of a third. Then he would
ask them to imagine placing the ½ pizza on top of
the first and then the second pie, each time taking
away that amount of pizza. How many times
would they perform this strange ritual before there
was no pizza left?
Deborah asked Terrell to explain what the
answer, 3½, would mean in this story. Terrell
stumbled. “The answer,” he said, “would be how
many times you got a whole half (if you want to
say that). Of the . . . whatever’s left over, what
part of it is of the half, I guess you could say.” He
obviously knew what fractions were, but when it
came to explaining the idea to another person, he
was at sea.
Another five future teachers came up with
stories or diagrams that did not actually represent
the problem. Several made up problems that
divided 1¾ “in half”—that is, by 2 instead of by
½. The remaining eight came up with nothing.
Deborah didn’t extrapolate the finding in her
dissertation, but the reader had to wonder: Besides
Deborah herself, how many people at the College
of Education could have answered even one of
those questions correctly? How many teachers at
Spartan Village?
In 1984, MSU announced the arrival of a new
professor with a unique joint appointment.
Magdalene Lampert (or Maggie, as everyone
called her then) was to serve as both assistant
professor of education at Michigan State and,
simultaneously, math teacher for grades four and
five at Spartan Village. She was both a researcher
of education and a practicing math teacher. And,
as Deborah soon learned, not only could she have
breezed through all the questions on Deborah’s
quiz; she could have written a better one.
Magdalene came from Cambridge,
Massachusetts, but she might as well have sprung
from Deborah’s imagination. Here, in silk blouses
and pulled-back blonde hair, was Deborah’s
special teaching knowledge personified.
Magdalene spoke about problems the way a potter
talked about clay, turning them over to see just
what they could do and then saving them for the
perfect opportunity. The best problems, the ones
that really pushed students into just the right
mathematical territory, Magdalene deemed “rich,”
“open,” “productive.” In class, she infused lessons
with the ideas and also the habits of math,
teaching her fifth-graders to “confer,”
“conjecture,” and “prove.” Confronted with a
student’s wrong idea, she often spied the hidden
misunderstanding faster than anyone else.
And just like a good math problem, she gave
nothing away. With lips pursed, her face was
perfectly opaque. “Nobody knew what the right
answer was for Maggie,” said Thom Dye, whose
fifth-grade classroom at Spartan Village became
her home (and the place where she eventually
taught Awad, Ellie, and Richard about “rate,”
helping them see how far a car going at 55 miles
per hour would travel in 15 minutes). “She’s very
stoic. And so the students had to look to
themselves for the correct answer . . . They
couldn’t just say, ‘Oh, it’s the right answer
because the teacher said so.’” Magdalene gave
them no other choice. They had to think.
The fact that Deborah hadn’t met Magdalene
until she came to MSU made the resonance
between their work that much more incredible.
What did it mean that two people, living hundreds
of miles apart, had stumbled on the same
approach—Deborah’s still nascent, Magdalene’s
far more developed, but in spirit the same? In
time, as they began collaborating—Deborah
learning from Magdalene—the work they were
doing, the specific kind of teaching, began to
demand a name, an easy tag for referencing in
discussion, like one of Deborah’s students’ fence
posts. Unable to come up with something
adequately distinctive, the Michigan State faculty
settled on a compromise: “This Kind of
Teaching”—TKOT (pronounced tee-kot)—or,
sometimes, “teaching for understanding,” though
no one really liked that term. (“What other kind of
teaching is there?” someone would inevitably
ask.) More specific labels simply didn’t fit.
Progressive, for instance, was a political
movement, not a pedagogical approach;
constructivist, meanwhile, had to do with a theory
of learning, not teaching. So they stayed
purposefully, playfully vague: “This Kind of
Teaching” would do fine.
Deborah protested the impulse to name; the
teaching she and Magdalene did, she insisted, was
simply teaching, not a special subset or approach.
They taught so that children learned. Wasn’t that
the whole point? For her part, Magdalene
acknowledged that her teaching was a kind of
“existence proof”: living evidence that it was
possible to teach math in the way Deborah aspired
to teach it.
A graduate of one of the country’s most
prestigious ed schools—the Graduate School of
Education at Harvard—Magdalene Lampert had
eschewed a conventional academic path. After
getting her doctorate, she’d taken a job teaching
elementary school math at Buckingham Browne &
Nichols, a private school in Cambridge,
Massachusetts, that also doubled as a teacher-
training program, working with students from
nearby Lesley College. There, her classroom, a loft
space shaded by trees, became her laboratory. Her
daily journal entries chronicled the room’s
happenings—her own ideas as well as her
students’. At Harvard, she’d read the formal
research on teaching, but none of it harmonized
with her own experience. Between lessons at
BB&N, she took careful and thorough notes,
reassembling the day’s events into an account of
what teaching really entailed.
 She was happy at BB&N and had no desire to
leave. But with the grant he’d gotten from the
National Institute for Education, Lee Shulman had
built Michigan State a new research group that
suited Magdalene perfectly: the Institute for
Research on Teaching (IRT). Charged with
exploring what Lee called the “wisdom of
practice,” IRT professors were expected not just to
do research and not just to teach future teachers,
but also to teach school. In other words, at
Michigan State, Magdalene could do everything
she’d been doing at BB&N, except with the
support of a full research university. She was sold,
and soon she had conscripted Deborah into her
cause—the transformation of Spartan Village into
a full-fledged teaching laboratory, a project that
came to transform both women’s careers.
Magdalene first got the idea from a former
colleague, who made a career of teaching school,
teaching teachers, and writing about his teaching,
all at once. To help unlock her own hidden
teaching expertise (the pedagogical content
knowledge behind her TKOT), Magdalene had
turned her classroom into a petri dish open for
study. At BB&N, back in Cambridge, her
observers had been the cohort of teachers she
worked with. At Spartan Village, a small army of
teachers-in-training, grad students, and fellow
professors followed her turns. On any given day,
two dozen or more ed school students left
Erickson Hall and drove past the football stadium
and over the train tracks to the Spartan Village
school. There they crowded into the back of
Magdalene’s classes (and, soon, in Deborah’s
too), taking notes.
But Magdalene quickly found that her observers
failed to see the subtlety of her methods. It was as
if their microscopes were smudged. They saw only
the least important details. After all, Magdalene’s
most important work, her moment-to-moment
decisions about what to do, lived only in her own
head. If she interrupted her teaching to make them
visible—to think aloud—she stopped teaching.
Deborah had used the metaphor of a surgical
theater. But unlike surgery, the act of teaching
took an entire year. In that way, it was more like a
novel. Skipping one chapter meant missing
everything.
An MSU graduate student offered a suggestion.
Why not videotape an entire year in her class,
starting with the first day and going all the way to
the last? It was a “wild idea,” Magdalene reflected
later. But it was the mid-1980s by then, and
technology was improving fast. (The grad student
had come to MSU by way of a new company
called Apple Computer.) Magdalene and Deborah
both already kept diligent journals, but video
would expand their data dramatically—and help
convey it too. Video, after all, could be paused as
well as rewound. To review a particularly
productive or confounding turn, all an observer
would have to do was click a button.
Working with Deborah, Magdalene drafted a
grant application, and by the 1989–90 school year,
the two of them had recruited two teams of
graduate students—the Lampert team and the Ball
team, they called themselves—to do the filming.
Careful schedules outlined which days the grad
students would man the cameras and which days
they would take notes. Each student was on a
quarter-time appointment, which paid enough for
just ten days of work per month. But after a while,
the students began giving up their days off
because they didn’t want to miss anything. This
Kind of Teaching, TKOT, might have a
ridiculously vague name, but it was riveting.
“They couldn’t keep us away,” laughs Kara
Suzuka, who was on the Ball team. “It was just
hard not to be there. You know, the story
continues! I mean, class just ends, and they still
don’t have a resolution. Kids are still confused. Or
they just came up with an incredible conjecture,
and you don’t know what’s going to happen with
that. You know, are they going to use it? What’s
the next thing? . . . It was just a very compelling
story.” Just how compelling, they had no idea.

Hyman Bass first watched the videos in 1996,

after a package of VHS cassettes arrived in his
campus mailbox at Columbia, where he was then
a tenured mathematics professor.
Sixty-four years old, having worked in math for
decades already, Hy was known professionally for
expanding a new field of algebra. But he had also
taken a late-career interest in the way children
learned his subject in schools. As far as he was
concerned, math was not just beautiful and
fascinating, but vital. “One of the noblest
expressions of humanity,” he said. Yet, instead of
encountering the beauty of the discipline, children
slogged away at a mere facsimile.
This was hardly a new worry. Math education’s
woes had always drawn extra attention in the
United States. Partially, this had to do with the
country’s belief in the economic power of the so-
called STEM fields (science, technology,
engineering, and math). Another factor was poor
test scores. US students regularly ranked behind
Canada, Germany, and Japan, reflecting a math
aversion that plagued many of their parents too.
As a country, it seemed, Americans simply were
not “math people.”
Hoping to make a difference, Hy joined policy
groups, math education boards, and advisory
panels. Despite the progress being made by Lee
Shulman and his colleagues at Michigan State, the
most prominent education reforms in the 1980s
stemmed from economist Eric Hanushek’s ideas
about accountability—an attention to outputs
rather than inputs, production rather than process.
“The conventional wisdom about public schools is
that they face serious problems in terms of
performance and that improving schools requires
additional money,” Hanushek explained in a 1981
article. “However, the available evidence suggests
that there is no relationship between expenditures
and the achievement of students.” Instead of
investing in traditional remedies like lower class
sizes or better teacher training, he wrote, “more
attention should be given to developing direct
performance incentives.”
The attention to incentives manifested as a
movement to craft more demanding educational
standards. The tragic flaw of the public school
system, standards advocates argued, was that it
had neither attended to the outputs of its students
nor defined what those learning goals should be.
Of course the schools wasted money; the system
literally had no standards!
Following the trend, Hy’s early forays in
education focused on writing better goals. He was
serving on a board devoted to just this task when
he met Deborah Ball. Deborah was unlike anyone
else Hy had met in math education. While he
believed in the power of standards, the efforts to
write them felt disconnected. Everyone seemed to
have an idea of what better math learning could
look like, but no one could describe it, and they
had certainly never seen it. Deborah was the first
person he met who actually seemed to know
something about the school side of the equation.
She was also the first person to imagine a way that
Hy himself, with his extensive math background
but limited knowledge of classroom teaching,
might be helpful. So when she asked him to
review the videotapes that she and a colleague had
made, Hy said sure.
Now, popping the tape into his VCR, he knew
his instinct had been right. The video opened on a
classroom that looked, at first glance, ordinary. At
the front was a long, green chalkboard with little
posters pasted on either side; in the middle stood
the standard beige desks with smooth, laminate
tops and built-in storage below. And of course,
there were children—nineteen of them. As the
video opened, one child stared listlessly at the
floor, her head slumped onto the back of her chair;
another perched his chin on his hand, pensive; a
third pulled her desk open, super quick, and
grabbed a pencil. The scene couldn’t have been
more mundane. Yet the class was unlike anything
Hy had ever seen.
The first voice to break the silence was
Deborah’s. “More comments from the meeting?”
she asked from the side of the room, out of the
camera’s sight. A transcript explained the context.
The day before, the third-graders had held a
meeting with a group of fourth-graders who’d
taken Deborah’s class the previous year. The
“Conference on the Number Zero,” the fourth-
graders had called it, lining up their desks in an
authoritative row to present their findings on a
question the third-graders had only just begun to
puzzle over. Was zero even, odd, or, as some
children argued, neither one?
Now it was a day later, and Deborah was
giving the third-graders a chance to debrief. The
discussion, she figured, would last only a few
minutes. They’d talk quickly about what they had
learned (zero is even),* and then they’d move on
to the real plan for the day, an activity Deborah
had been anticipating for weeks. A few days
earlier, working on a problem she’d designed
specifically for this purpose, the students had
come up with conjectures about the properties of
even and odd numbers. An odd number plus an
odd number, they’d noticed, always seemed to
equal an even, while two evens always made an
even. Now, she wanted to see if they could do
what no third-graders she knew had ever done
before. She wanted them to prove the statements
true—not just for the numbers they’d tried so far,
but for all numbers.
It wasn’t to be—not that day, anyway. Before
she could get through the discussion, Deborah
found her plan hijacked by a tall boy named Sean.
She wouldn’t regain control for another several
days.
“Sean?” she’d said, noticing his hand.
 “I don’t have anything about the meeting
yesterday,” he said, “but I was just thinking about
six. I was thinking that it can be an odd number
too, ’cause there can be two, four, six, and two—
three twos—that’d make six.”
“Uh-huh . . .” Deborah said.
“And two threes. It could be an odd and an
even number. Both! Three things to make it and
there could be two things to make it.”
Deborah jumped in. “And the two things that
you put together to make it were odd, right? Three
and three are each odd?”
“Uh huh,” Sean replied, “and the other, the
twos were even.”
Maybe, she thought, Sean was responding to
the earlier comment about how even numbers can
be made up of two even numbers. Maybe he
wanted to point out that some even numbers, like
six, were actually made up of two odd numbers
instead.
Knowing they’d get to all that in just a few
minutes, Deborah decided to let the idea rest.
“Other people’s comments?” she asked, returning
to the debriefing.
But Cassandra, a tall girl with a yellow hair
clip, who raised her hand next, was stuck on 6. “I
disagree with Sean when he says that six can be
an odd number,” she declared, rocking her chair
back on its legs. “Because—”
Hy watched Cassandra stand up and walk to the
board, where she picked up a long pointer.
“Look,” she said, directing the pointer at the
number line high above the chalkboard and
landing it on zero. “Six can’t be an odd number,
because this is, um”—she pointed to zero
—“even.” She walked through the rest of the
numbers, “Odd, even, odd, even, odd,” until she
landed on six. “Even.” She turned back to Sean.
“How can it be an odd number?” But Sean
persisted. “Because,” he said, “because six—
because there can be three of something to make
six, and three of something is, like, odd.”
 Next came Keith, who threw up his hand. “That
doesn’t necessarily mean that six is odd,” he said,
to a chorus of agreement from the class. “Just
because two odd numbers add up to an even
number doesn’t mean it has to be odd.”
Hy marveled as the video continued. These
third-graders—not a gifted class, but average,
public school third-graders from, Deborah said, a
wide range of backgrounds and ability levels—
were having a real mathematical debate. One of
them had made a claim, and then the others were
trying to prove him wrong. Cassandra’s proof
followed a classic structure. First, she had invoked
one definition of even and odd—the fact that
integers alternate between the two types on a
number line—to show that six could only be even.
Then she had drawn out a counterargument. To be
odd and still fit the alternating definition, she’d
shown, zero would have to be odd too. But, she’d
concluded with a flourish, they had just decided
the other day that zero was even. QED: Sean’s
conjecture was impossible.
Deborah had asked Hy to watch the video for
important mathematics. Well, here it was: a third-
grader doing a fairly sophisticated mathematical
proof.
More proofs followed, none of which helped
Sean articulate his idea. Jeannie reminded her
classmates of their working definition of an even
number—one “that you can split up evenly
without having to split one in half.” Sean had
agreed that, yes, 6 fit that definition. Later, from
Ofala, a girl from Nigeria, came a derivation of a
definition the class apparently had not made
before, for odd numbers—“my conjecture,” Ofala
called it. If even numbers were those that could be
split evenly in twos, without any left, then odd
numbers were those that also “have two in them,
except they have one left.” Drawing out six lines,
she showed, there were none left. “I already have
all the twos circled,” she said.
But the most intriguing proof belonged to a
little girl in a purple headband, the one who’d
started the day staring at the floor: Mei. “Oh!”
she’d exclaimed, out of nowhere, not long after
Deborah had declared herself confused. “I think I
know what he’s saying! . . . What he is saying is
that—you have three groups of two, and three is
an odd number. So six can be an odd number and
an even number.” That is, it could be even because
6 is broken into groups of two, but odd because
the number of groups of two is odd. “Is that what
you’re saying, Sean?” Deborah asked. Finally, it
was!
Mei was not done. She had clarified Sean’s
argument, and now she intended to destroy it. “I
disagree with that,” she said, when Deborah
asked her opinion. “Here.” She was already
halfway out of her seat. “Can I show it on the
board?” Before her teacher could say yes, Mei was
pushing in her chair and marching to the board.
Sean, still standing at the board, took a step to the
side to make room. “It’s not according to, like,
how many groups it is,” Mei said, her long, black
hair wagging from side to side behind her as she
reached the board and grabbed a piece of chalk.
Her head barely reached Sean’s elbow.
She explained. “Let’s see if I can find . . .” she
said, pointing her chalk at the green board and
staring ahead, deep in thought. Her voice was
high, even for a nine-year-old. “Let’s say ten.” She
began drawing a line of circles. “One, two,” she
counted. She drew ten in a row. “And here are ten
circles,” she said.
Sean stood with his back against the board,
watching Mei press down her chalk again. “And
then you would split them,” she was saying.
“Let’s say I want to split them by twos. Go one,
two . . .” She drew vertical lines between every
other circle. The board looked like this:

“Well, look!” she said, gaining speed as she

tapped each pair with her chalk—“one, two, three,
four, five!”—and turned to face Sean, who now
had his entire body facing hers.
“Then why do you not call ten, like—a—,” Mei
stopped for a moment, and Sean said something,
but she didn’t hear it. She had turned to face the
rest of the class, and she was throwing her hands
out to either side of her, summing up her case like
a trial lawyer reaching the climax of a closing
statement. Just as 6 divided by 2 produced an odd
number (3), 10 divided by 2 was 5—another odd
number. Why didn’t Sean call 10 “an odd number
and an even number?” Mei asked. She dropped
her hands to her sides and stepped back,
scratching her nose. Case closed.
Mei had missed the key sentence, but it had not
escaped Hy. Right in the middle of her crescendo
conclusion, Sean had mumbled the following four
words: “I disagree with myself.” To Hy, all this
was stunning—an extraordinary episode of
mathematical reasoning, enacted entirely by nine-
year-olds. First, Mei had pulled off something that
is often challenging, even to mathematicians. She
had listened to Sean’s confusing argument, and
she had translated it into an impeccably clear
explanation of his own thinking. Mei’s analysis
helped the class, and it also allowed Mei to
articulate to the whole group why she disagreed.
What she did at the board was even more
amazing. Until that point, all the arguments
against Sean had followed the same pattern: Sean
made his claim, and then the students attacked the
conclusion, offering up different proofs of why 6
was actually even. Mei took a much more
sophisticated stance. Instead of challenging his
conclusion, she challenged his reasoning. And in
the process, she took a leap Sean had not yet made
or even seen. Six, she showed him, wasn’t the
only number that met his odd-groups-of-two
criteria; 10 did too, and possibly others.
“What about other numbers?!” Mei had said.
“Like, if you keep on going on like that, and you
say that other numbers are odd and even, maybe
we’ll end it up with all numbers are odd and even.
Then it won’t make sense that all numbers should
be odd and even, because if all numbers were odd
and even, we wouldn’t be even having this
discussion!”
Sean seemed to have no choice but to fold.
Except, that’s not what happened. After a pause—
Mei staring at Sean, Sean staring at Mei, Mei
scratching her nose, Sean rocking back and forth
—Sean extended his gratitude. “I didn’t think of it
that way,” he said, smiling. “Thank you for
bringing it up. So, I say it’s—ten can be an odd
and an even.”
Instead of quieting Sean, Mei had unleashed
him. Not only that, but soon other students were
joining his cause, deriving more numbers that fit
his criteria (not just 6 and 10, but also 14 and even
2!). When Deborah tried to steer the conversation
to a close—one more idea, she allowed, “but then
I think maybe we’re going to have to stop with
this”—it was no use. The children, Hy thought
with delight, had been ignited.
Not only were they constructing proofs. They
had invented an entirely new category of numbers.
“Sean numbers,” Deborah christened them a few
days later, deciding to turn the diversion into an
opportunity to enhance a point she’d been trying to
teach, about how to make mathematical
definitions. (A point, she had to admit, that the
Sean detour had offered multiple chances to
underscore, what with Ofala’s new definition of
odd numbers and Jeannie’s restatement of the
definition of an even number.)
When the video came to an end, Hy considered
the question Deborah had asked him. What math
could he see in the videos? He’d seen math in the
kids, of course, but also in Deborah. Upon
reflection, it was the teacher, not Sean or Mei or
Ofala, who had provided the kids with two critical
turning points. Deborah had directed them back to
their “working definition” of an even number,
laying a foundation for discussion. Sean, in calling
6 “even and odd, both,” was actually positing an
entirely different definition. And then there was
the moment, right after Mei suggested that 10 fit
his criteria too, when Deborah suggested that the
children consider 14—the launch point that had
sparked a girl named Riba to derive another
mathematical definition. In fact, Riba had shown
that, if you followed Sean’s logic, every fourth
number on the number line could be called “odd
and even.”
Another mathematician might look at “Sean
numbers” as a mistake. After all, numbers that
met the boy’s criteria were, ultimately, even. But
Hy knew that math was all about definitions—
coming up with the specific rules and restrictions
that made one number positive and another
negative, or one prime and another composite. The
usefulness of these definitions determined which
ones stuck and which ones mathematicians
discarded. The concept of Sean numbers would
ultimately end up in the dustbin, but by inventing
the numbers in the first place, the students had
learned something fundamental about how to
think about math, something they certainly
wouldn’t have gotten just from learning the
difference between odd and even.
Deborah amazed Hy. He had viewed only what
she had done on this particular day. But what
about everything she had to know to get to this
point? The problems she’d had to pick, the habits
she’d had to teach, the decisions about when to let
a detour happen and when to shut it down? Every
amazing student epiphany in a TKOT classroom
reflected an equal capacity on the part of the
teacher orchestrating it. And Deborah’s capacity
was unlike anything Hy had ever seen. “Watching
Deborah teach,” he said, “is like listening to
chamber music.”
He liked it so much that, not long after
watching that first tape, he handed in his
resignation at Columbia and moved to the
University of Michigan, where he took a joint
appointment in the math department and the
education school. Standards, curriculum, and
assessments were important. But math education,
he had realized, could not change unless the
teachers could turn those tools into everyday
lessons. He and Deborah began a formal inquiry
into the kind of knowledge required to teach math
well. Soon, they had a definition of their own—
and a name: Mathematical Knowledge for
Teaching, or MKT, “the mathematical knowledge,
skills, habits of mind, and sensibilities that are
entailed by the actual work of teaching.” The math
version of Lee Shulman’s pedagogical content
knowledge. The wisdom of expert practice.
Some parts of MKT overlapped with knowledge
held by any educated adult, but other parts, like
knowing how to analyze incorrect or nonstandard
solutions, identifying the student thinking that
might have produced an incorrect answer,
anticipating likely student errors, and
understanding what kinds of representations offer
the best explanations, did not. Even Hy, a
professional with decades of experience, did not
possess these parts of MKT. And later,
administering a test of MKT, he and Deborah saw
that neither (to the subjects’ horror) did other
mathematicians.
In time, the Sean episode went viral, playing at
conferences from California to Korea. Magdalene
Lampert’s class, meanwhile, was featured in Life
magazine and eventually came to inspire an entire
television show on PBS: a math program for
young children called Square One TV in which
noir detectives George Frankly and Kate Monday
of “MathNet” worked to solve a new case each
week. (In addition to the cases cracked by
MathNet, recurring sketches offered the two-
minute television equivalent of Magdalene’s
problem of the day. In the “Bureau of Missing
Numbers,” distressed citizens reported absentee
numbers to an FBI-style investigator, who hunted
them down by interviewing witnesses about their
characteristics; “Prime Club” depicted a nightclub
that admitted only prime numbers; a musical
number parodied “Climb Every Mountain” in a
song about counting to the highest number.)
The two teachers—Deborah Ball and
Magdalene Lampert—were a form of proof
themselves, evidence that a different kind of
teaching was possible. If two women in Michigan
could teach this way, people began to wonder,
why couldn’t everyone?

* The correct answer depends on the

problem’s assumption, which wasn’t given. If the
passenger can stop only once, then there are only
25 ways to get to the second floor. If trips can
include multiple stops, then the number of
possible trips is infinite.
 * Like all even numbers, zero can be divided
evenly by 2, is surrounded on either side by odd
numbers, and when it is subtracted from an even
number, produces an even result.
 3
SPARTAN TRAGEDY

It was one thing to prove that excellent teaching

was possible, quite another to teach it to people
without the extraordinary skill of Magdalene
Lampert and Deborah Ball. In 1982, after Lee
Shulman left Michigan State University for
Stanford, the woman who took responsibility for
that task was Lee’s original IRT partner in crime,
a Michigan native and former teacher named
Judith Lanier.
Judy modeled her reform efforts on two of the
schools where she’d been trained herself—lab
schools, they were called. Even back then (she
started teaching in the late 1950s), the schools had
been the last of a dying breed. Lab schools, in
turn, were the offshoots of another antiquated
institution, the “normal” school, a college
alternative that thrived in the early twentieth
century before universities took over the job of
training teachers.
Aimed both at training future teachers and
inventing better ways of teaching, normal schools
served two kinds of students: kindergarteners
through twelfth-graders and, simultaneously, the
college students who wanted to learn how to teach
them. Each normal school was really two schools:
the “normal” part, for teacher training, and the
lab, or “practice,” school, where K–12 kids
learned while the future teachers watched and,
eventually, stepped in to try out teaching
themselves.
John Dewey’s lab school at the University of
Chicago, which he inherited from his mentor
Francis Parker, fashioned itself on this model.
Judy’s own experiences—first at a lab school in
Paw Paw, Michigan, and later at a school in
Milwaukee, Wisconsin—showed her the model’s
power. In Paw Paw, she and her fellow teachers
learned to keep detailed logs of their daily
practice. Later they met with researchers to go
over what they’d done, receiving lectures in a
building adjacent to the school. The lab school in
Milwaukee even had an upstairs viewing area, a
glassed-in catwalk raised around the classroom’s
perimeter, from which education students could sit
and observe, opening the glass window to
eavesdrop on the lesson without interrupting its
flow. Afterward, the students and their professor
would go over teaching problems together.
Working at the lab school changed Judy’s view
of teaching. Originally, she’d taken up the work
by default. (When she finished high school, in the
early 1950s, women seemed to have only three
choices: nursing, secretarial work, and teaching.
Since Judy disliked blood and found office work
boring, teaching was it.) But after the experiences
in Paw Paw and Milwaukee, teaching was no
longer a job she had settled for. It was, she saw, a
craft—one that a person could spend a lifetime
mastering.
As the years passed, however, the lab schools,
and the view of teaching they supported, became
increasingly obsolete. The main trigger was
universities, which began to add the lucrative
teacher-training business to their repertoires,
putting normal schools out of work. But in taking
over teacher training, universities marginalized it.
Instead of training, their professors—men more
like the psychologists William James and Edward
Thorndike than like the former schoolteacher
Francis Parker—focused on research that offered
“hardly a nod toward the public schools,” Judy
later wrote. At the university, “schoolteachers and
young learners, who should be the focus,” became
a “sideshow to the performance in the center
ring.”
An Oxford professor named Harry Judge,
touring American ed schools at the request of the
Ford Foundation, described the university
approach as “the doctrine of Anything-But.” That
is, ed schools were “anything but schools of
pedagogy,” an ed school professor told Judge.
The doctrine of Anything-But began with
professors. In his final report, Judge described
how faculty recruitment happened at a fictional
university he named Waterend—a composite
representing the elite ed schools he’d toured:

The dominant tactic was to make a

foray into the disciplines, to track down
a scholar of achieved distinction or of
sparkling promise, and to carry him
triumphantly through the gates of
Waterend. Thereafter, the professor
would be careful to explain that this was
the first appointment he had ever held in
a school of education, that he was
unsullied by contact with the lower
worlds of educational practice, that he
was first and foremost a Waterend
Professor—with at least a courtesy
 appointment in another department as
well.

At land-grant universities like Michigan State,

which Judge satirized as a fictional place he called
HSU, the same practices ruled, but on a larger
scale. Instead of a handful of unsullied
psychologists, he observed, HSU’s ed school hired
sixty.
At both Waterend and HSU, education
professors tended to feel more loyal to their
discipline of origin than to the study of education.
Among the subjects of interest, Judge recounted, a
professor of education might study “the history of
the family, the role of the media in the formation
of public opinion, the structure of higher
education, the changing shape of macro-
economics, or the evolution of organisational
theory” before ever visiting a classroom. The
neglect was sometimes benign (for example, Lee
Shulman studied doctors before his encounter with
Nate Gage), but often it was outright hostile. One
professor told Judge how happy he was not to
have to work with “dumb-assed teachers.”
Fueling the doctrine of Anything-But were the
perverse pressures of tenure. Even those with good
intentions learned that the work that led to that
ultimate academic accolade did not also lead to
good schoolteaching or teacher training. A young
professor might be a masterful trainer of
undergraduate teachers, but her CV needed to list
publications like “Sex Stereotypes of Secondary
School Teaching Subjects” if she wanted a job
after grad school.
Judge concluded that the tradition, while
“indefensible,” was nevertheless unchangeable.
The universities depended on ed schools for
tuition and thus suppressed reform. The ed
department became “our dumping ground,” one
professor told him.
When Judy Lanier arrived there in 1964, as a
graduate student, the MSU ed school epitomized
the doctrine of Anything-But. It was a “good old
boys’ paradise,” Judy told one historian. The
reigning clique of big-thinker types met regularly
in the lounge on the top floor of Erickson Hall,
where they spent hours smoking their pipes,
sipping coffee, and generally infuriating the rest of
the faculty, who noted with aggravation that they
rarely strolled in before 10:00 a.m.
These “good old boys” had little interest in
research on teaching. When Judy first described
her dissertation idea—a study of the features that
separated excellent teaching from mediocre cases
—several senior faculty told her that the project
was impossible. They were unimpressed by her
proposed methodology: surveying principals,
teachers, and parents and videotaping the teachers
with the best reputations. This was the mid-1960s,
pre–Nate Gage, and most studies had failed to
identify any common ingredients of good teaching.
The faculty doubted there was any way for Judy to
identify the best teachers, much less to discern
what made them succeed.
MSU’s “teacher educators,” meanwhile, formed
a distinct, marginalized group. While the “good
old boys” enjoyed light teaching loads and ample
time for research, the teacher educators endured
monumental class sizes, heavy loads, and slim to
nonexistent research budgets.
As for teacher preparation itself, rigorous tracks
like the Elementary Intern Program that fed
Deborah Ball into Spartan Village were rare. A
typical American undergraduate, Judge’s report
had observed, could pass through the courses
necessary to become a teacher even if she suffered
“a prolonged fit of absentmindedness.”
Rather than closely guided classroom
experiences, the average MSU education student
followed a three-part curriculum. First, there was
the overview of the academic subjects, basic
survey courses, taught by what Judy considered
the department’s lightweights. Then came the
“foundations” courses in the psychology, history,
and philosophy of education, taught by the most
junior of the “good old boy” types (including,
when he first arrived, the young educational
psychologist Lee Shulman). Finally, there were the
methods classes, taught by the teacher educators.
In theory, these focused on the craft of teaching,
the “how” rather than the “what.” But more often
they reflected what Judy called the “boots and
galoshes” vision of teaching—“the idea,”
according to Francesca Forzani’s history of the
period, “that all teachers needed to learn was how
to help children dress for recess.”
Next, future teachers embarked on the student
teaching experience, ten weeks in a classroom
buoyed only by whatever limited guidance their
host teacher could provide. Sometimes, Judy
observed, schools assigned student teachers to
their weakest staff members—the ones who
struggled to keep the children in order and could
use the help. During her own student teaching, at
Western Michigan University, she’d wound up
doing more of the teaching than the host teacher
did.
In 1980, when she became dean of the ed
school, Judy upended the doctrine of Anything-
But. Instead of dallying in other disciplines,
professors would spend their time mining the
secret wisdom of teachers, modeling for the entire
school the approach that she and Lee Shulman had
perfected at the IRT. They would transform MSU
into a modern version of her lab school in Paw
Paw. Michigan State would set an example for
universities across the country, raising the level of
teaching nationwide.
Judge’s scathing report to the Ford Foundation
in 1981 might as well have been Judy’s blueprint.
First came housecleaning. Judy cut ed school
spending by 40 percent. In return, she extracted a
promise from the provost: a 25 percent increase
for future initiatives focused on her mission. Then
she went in search of new hires. Just as she and
Lee had done at the IRT, Judy recruited professors
whom other ed schools might have ignored,
faculty who were expected to conduct research
about teaching and to train teachers. The goal was
not just to do teacher education, but to transform
it. Judy even convinced Harry Judge to join the
cause. For five years, he served a joint
appointment, working at both Oxford and MSU.
She capped off her hiring spree in 1984 with
Magdalene Lampert. Magdalene had sworn off ed
schools long before then, banishing herself instead
to the classroom at Buckingham, Browne &
Nichols. Her experience at the Harvard Graduate
School of Education had thoroughly disenchanted
her. Of all the listings in Harvard’s course
catalogue, only one had the word “teaching” in its
title—and she ended up marrying the professor,
David Cohen. But even David, a historian by
training, had never taught school himself. He had
only recently moved from studying education
policy and history to observing
classrooms. “Crouching,” he called it.
Not only that, but Michigan State was a cow
college. When Magdalene and David first visited
MSU, one of the first departments they passed had
a sign that announced, in bold MSU green and
white, “DEPARTMENT OF SHEEP TEACHING
AND RESEARCH.” Another: “SWINE
TEACHING AND RESEARCH.” Erickson Hall,
home of the education school (human teaching
and research), sat on a street called Farm Lane,
less than a minute’s drive from a long stretch of
cornfields. Finally, they passed a sign reminiscent
of home: “AI School,” it said. “Well, in
Cambridge, that’s MIT, and that’s artificial
intelligence,” David says. At Michigan State,
“AI” meant artificial insemination.
But Judy Lanier took what was most foreign
about the Midwest—that staggering flatness, the
swine—and spun it as an advantage. As the
country’s first land-grant university, Michigan
State prided itself on its commitment to producing
knowledge for the field, literally. In the nineteenth
century, it was a professor at MSU who first
devised a procedure for hybridizing corn, helping
to modernize agriculture.
The education school, Judy explained to
Magdalene and David, could do the same for
teaching. With such a huge student body, their
experiments might change the lives of thousands
of student teachers in classrooms throughout
Michigan each year. Equip them with the right
ideas and skills, and they could change education.
How could Magdalene say no?

Judy Lanier’s ambitions were not novel. Efforts to

transform teaching stretched back to the early
nineteenth century. In a speech to a gathering of
school leaders in 1830, an educator named Warren
Coburn announced that he wanted to extinguish
what he called “the old system.” In that approach,
“the learner was presented with a rule, which told
him how to perform certain operations on figures .
. . But no reason was given for a single step,”
Coburn wrote. “As he began in the dark, so he
continued; and the results of his calculation
seemed to be obtained by some magical operation
rather than by the inductions of reason.”
But the “old system” was still current in 1911,
when the mathematician Alfred North Whitehead
described the “road to pedantry” offered by most
school math. Poorly taught, with a focus only on
brute memorization and not any of the subject’s
more intricate concepts, he said, the great science
became like the ghost of Hamlet’s father: “ ’Tis
here, ’tis there, ’tis gone.” Nor had much changed
by 1957, when the competitive panic wrought by
the Soviets’ Sputnik launch inspired a fresh
curriculum called the “New Math”—a program of
study that, Suzanne Wilson summarized in her
history of American math reforms, would help
“any normal human being [to] appreciate some of
the beauty and power of mathematics.” The state
of math teaching as Deborah Ball and Hy Bass
encountered it in the 1990s showed how well that
had gone.
Math got the most attention (not to mention
more research dollars), but it was not the only
school subject to inspire, and then resist, calls for
change. Studying classrooms in Portland, Oregon,
in 1913, a survey team found pedantry
everywhere. In geography, “the questions, almost
without exception, called for unreasoning
memorization of the statements of the book.” In
grammar, much of the work “had little meaning
for most of the children.” In history, “there was
not the slightest evidence of active interest in the
subject; the one purpose seemed to be to acquire,
by sheer force of memory, the statements of the
assigned text.” Yet in 1970, the journalist Charles
Silberman was still diagnosing “mindlessness”
across the board. This was the story and, perhaps,
the destiny of American schoolteaching: always
admonished, never changed.
But Judy Lanier benefited from good timing.
Like-minded comrades might have come and gone
before, but none had arrived at a moment as
auspicious for teaching reform as the 1980s. One
advantage was the emergence of academic
research that, for the first time, bolstered (rather
than ignored) educators’ notions that learning was
more complicated than Nate Gage’s behaviorism
suggested. Studying the inner workings of the
mind, the new breed of cognitive scientists had
found that learning did not respond to common
teaching techniques.
One study examined the math capabilities of
child street vendors in Brazil. Selling coconuts
and watermelons on the street, the children
tabulated prices and counted out change with
impressive facility. But when the researchers
transferred the problems the children had
encountered on the street to paper, the children
floundered. Over and over again, the researchers
watched them miscalculate in the same way: by
incorrectly following procedures they had learned
—and completely misunderstood—in school.
They were more than capable of complex
computation. School just seemed to conspire
against their ability to do it.
The pattern repeated itself again and again. On
the street, a child would improvise mental math to
figure out his customer’s price. Then, on paper, he
would switch off that part of his mind—the part
where multiplication and division represented real
transformations—and instead do his best imitation
(often incorrect) of the steps he’d been taught to
memorize in school. It was as if the two problems
were completely separate: one an actual
manipulation of real numbers, the other a series of
steps performed to please a teacher.
A twelve-year-old boy who had just fluently
calculated the price of 4 coconuts at 35 cruzeiros
a coconut, 140 cruzeiros, was flummoxed when
researchers presented him with the exact same
problem on paper. 35 × 4? Instead of following
the same calculation he’d done a minute before in
his head (“Three will be 105, plus 30, that’s 135 .
. . one coconut is 35 . . . that’s 140!”), he tried to
walk through the multiplication procedure he’d
learned in school, stacking one number on top of
the other:

He got the main pieces right, correctly

multiplying the 4 by 5 to get 20 and then carrying
the 2. But instead of waiting to multiply 4 by 3
before adding the carried 2, he added the 3 and 2
and then multiplied by 4. He produced his answer
for the researchers—200—apparently without
wondering about the difference from his first
calculation. School, the study suggested, not only
failed to help students learn; it actually seemed to
confound them.
In addition to the cognitivists, an even more
powerful group had begun to influence schools in
the 1980s—the business and political elite.
Alarmed by new international tests showing that
US students were falling behind their counterparts
around the world, these CEOs, elected officials,
philanthropists, and advocates worried about what
the apparent downturn might mean for the
national interest. After all, the economy was
shifting from moving and making physical objects
(cars, food, coal) to constructing what economists
called “information products” (software, video
games, cell phone calls). Floundering American
students did not seem poised to participate in this
new economy.
One CEO, Alfred Taubman, a billionaire
businessman whose empire included the A&W
restaurant chain, became alarmed about schools
after a major product flop. Hoping to challenge the
famous McDonald’s Quarter Pounder, he’d
released the A&W one-third-pound burger—at the
same price. But though the A&W burger beat the
Quarter Pounder in taste tests and value, the one-
third-pounder did not sell. Only after hiring a
market research firm to mount customer focus
groups did Taubman understand why. Half the
participants in the focus groups believed that
A&W had overcharged. “Why should we pay the
same amount for a third of a pound of meat as we
do for a quarter-pound of meat at McDonald’s?”
they asked. Some customers actually thought that
a third, having to do with the number 3, was less
than a fourth, having to do with 4.
The math abilities of customers augured poorly
for those of workers. “When companies have to
spend billions of dollars providing remedial
instruction in reading, simple math, and problem
solving,” Taubman concluded, “that’s a double
tax.” They paid once for the official education
system, through government taxes, and then, when
the schools failed, they paid again for their own.
Economic competitiveness was on the mind of
President Reagan’s first secretary of education,
Terrel Bell, when he commissioned a study of
American schools in 1981. Titled A Nation at
Risk, the report described the deteriorating
condition of the American education system. Of
special concern was the fact that even students
with basic competency failed at the “higher order
intellectual skills” that would be vital in the
postindustrial economy. The Nation at Risk report
launched dozens more. One history concluded,
“Within a few years, it was no exaggeration to
speak of a ‘movement’ for school reform.”
Reformers waged their fight on many fronts.
Over time, the most prominent was the push for
standards. After A Nation at Risk, governors
began meeting to plot new learning standards—a
development that sowed the seeds for the No
Child Left Behind law two decades later. Judy
Lanier, meanwhile, used the budding concern to
build an impressive coalition around her reform
agenda: transforming the study and training of
teachers. When Terrel Bell announced the Nation
at Risk report, Judy traveled to Washington, DC,
for the occasion. Later, Bell visited Michigan
State, where he gave an award to Judy and to the
Institute for Research on Teaching. Judy’s
advisers also included Alfred Taubman, Jim
Blanchard (the Michigan governor), and the
leaders of several national philanthropies.
Rallying the reform movement to her cause, Judy
didn’t have to persuade the establishment about
the importance of training better teachers. For the
moment anyway, she was the establishment.

But, Magdalene Lampert would later ask, exactly

what did teachers need to learn, and how were
they going to learn it? Not long after Magdalene
agreed to come to MSU (she signed on for a
provisional two-year stay, short enough for David
to keep the option of returning to Harvard without
losing his tenure, and long enough for her to give
MSU a real shot), a teacher named Ruth
Heaton came to embody the challenge of teacher
education.
A first-year graduate student, Ruth first met
Magdalene midway through the school year in a
state of distress. She’d come to grad school to
become a teacher educator, but despite nine years
of experience in elementary school classrooms, her
confidence was suffering at MSU. All around
Erickson Hall, she heard people diagnosing the
pitfalls of the traditional math classroom. And,
with horror, she realized that for nine years she
had been perpetuating those same mistakes.
Magdalene took Ruth on as her new student.
She put her on the Lampert team, the group of
grad students who came to watch her teach during
the year of the videotapes. The following
September, Magdalene installed Ruth in a fourth-
grade class right next door to her own. Ruth
became the math teacher for that class, as well as
Magdalene’s unofficial apprentice. Twice a week,
Magdalene sat in Ruth’s classroom, watching and
composing comments. In the years that followed,
both women taught each other; Magdalene taught
Ruth how to teach math, and in turn, Ruth taught
Magdalene how to teach teaching.
One early lesson started off simple and then got
more complicated. Watching Magdalene and
Deborah teach, Ruth had grasped the importance
of getting her students to talk. In her math lessons,
she dutifully plied the fourth-graders with
questions, often imitating Magdalene and
Deborah word for word. “How do you know
that?” “What do other people think about that?”
But although she asked the right questions, she
wasn’t sure what to do with the students’ answers.
As a result, class discussions felt less like
explorations and more like a series of dead ends.
Each comment fell with a thud—the sound of no
one thinking.
In one typical sequence, Ruth introduced a
lesson on functions. She was using the same
experimental curriculum that Deborah had tried
with her first-graders. Following the teacher’s
guide, Ruth had the class make up a list of
numbers that fit a simple function: f(x) = x + 10 +
2. Plug in any number for x, and what would you
get? The students came up with a list easily: 99
and 111, 8000 and 8012, 250 and 262, 4988 and
5000, and so on.
Next, Ruth was supposed to ask them what
patterns they saw in the numbers. The teacher’s
guide described the rich dialogue that would
ensue. The students would make sharp
observations—if the number on the left is even,
then so is the number on the right; the number on
the right is always bigger than the number on the
left—and then, as in one of Magdalene’s or
Deborah’s lessons, they would move from
noticing to verifying (“Is this always true?”) and
from verifying to a deeper understanding. Who
knew, maybe they’d even invent a new class of
numbers!
Instead, it was another dreary parade of dull
ideas marching nowhere. There are two “80s,” one
student offered, pointing to 8000 and 8012.
Another pointed to 8000, 111, and 5000. “Here is
three zeroes in a row and three ones in a row, and
then three zeroes in a row,” the student, a boy
named Richard, said. What was the pattern? Ruth
asked hopefully, and he said it again: “000, 111,
000.”
Ruth was despondent. “I felt like I was
floundering today,” she told a colleague later that
afternoon. But where Ruth saw failure, Magdalene
saw room for improvement. Ruth knew her
students needed to talk about math. She just didn’t
know how to turn the talking into learning. That
was what Magdalene would have to explain.
To teach math to a child, the best strategy was
to design a productive problem. To teach TKOT,
Magdalene needed a parallel opportunity, a
teaching problem to help Ruth see the difference
between her solution (repeating the question from
the textbook) and other, better possibilities (a
more fertile way of responding to students’ ideas).
 A productive teaching problem arose one day in
late September, when Ruth assigned the fourth-
graders a problem of the day:

What whole numbers could be put in the

boxes?
26 – =

Instead of plunging into the problem, like

Magdalene’s students always seemed to do,
Ruth’s diverted. “What’s a whole number?” one
girl asked her. Thrown off guard, Ruth tried to
answer the question quickly and move on. She
directed the girl to the number line on the wall,
which displayed a list of whole numbers. But
instead of returning to the problem, the same
student piped up again. “I don’t understand,” she
said. “What’s not a whole number?” Frustrated by
this waste of time—they needed to be coming up
with solutions, not debating the directions!—but
also trying to listen to the students, Ruth paused
one more time, helping the students list more
examples of whole numbers.
Watching from the sidelines, Magdalene saw a
classic teaching problem. Ruth seemed to have
missed what the girl was asking for. She didn’t
want examples of whole numbers; she wanted a
definition. That was why she asked the second
question, “What’s not a whole number?” There
was no easy solution to this problem (or to any
teaching problem, for that matter), but Magdalene
could help Ruth think more carefully about her
response. She could, for instance, help Ruth see
the value in listening to what students were asking
her for. Sometimes, a teacher needed to steer
students away from questions that threatened to
take them off on tangents because, sometimes, the
tangents were a waste of time. But in this
particular case, clarifying the students’ confusion
about whole numbers was core to helping them
work on the task. They needed to know what a
whole number was before they could think of
whole numbers to fit in the boxes.
 Magdalene could also help Ruth devise a better
response by helping her understand the math she
was dealing with. Her decision to point to the
number line had its merits, as it gave the students
a ready and visible list of possible numbers to try.
But Ruth had confused the whole-number
numerals written on the line with the line itself,
which represented not only whole numbers but
also all the fractions in between. Magdalene
understood the confusion. “Within mathematics
the importance of the number line is that it
represents continuity,” Magdalene wrote in a note
to Ruth. “That is, it represents the idea that there
are always more numbers in between the other
numbers.” The challenge for teachers was to walk
the line between these two equally important uses
of the number line—a stock of discrete numbers to
draw on, and also an expression of continuity—
without confusing the students or being incorrect.
Reading over Magdalene’s note later, Ruth felt
relief. There was no magic bullet that took
students’ ideas and created a rich conversational
environment. But there were better and worse
ways of making sense of their comments—and
better and worse ways of responding to them. For
instance, had Ruth done a better job of listening to
what it was the students actually wanted—a
definition—she could have focused her efforts on
helping them generate one. And to get them there,
she could have elaborated on a move that
Magdalene pointed out approvingly, when Ruth
suggested that a student discuss the meaning of
whole number with other students at her table.
Instead of just letting them talk, she could have
steered their conversation toward the right answer.
The key to moving a discussion forward was to
listen to students’ questions, figure out what they
needed to understand, and construct a response to
pull them there.
When Ruth finally managed to pull this off, she
didn’t even notice she’d done it until Magdalene
pointed it out to her. Magdalene always scribbled
comments right after watching a lesson and then
gave them to Ruth to read before they discussed
them in person. This time, Magdalene’s note
underlined a certain moment, when Ruth had told
the students, “I want to show you something.”
“Did you get some kind of ‘bright idea’ about how
to pull all this together when you said [that]? Or
were you following the script?” Magdalene wrote
in her observation note that day. “It seemed to me
as if you were more engaged here, more thinking
about the kids and the subject matter and the
representation rather than reading the manual.”
At first, reading the note inside her car in the
Spartan Village parking lot, Ruth had no idea
what Magdalene meant. “Did I get some bright
idea? What moment was she referring to?” she
thought. At home that night, she opened the
audiotape she had made of the lesson, found the “I
want to show you something” moment, and hit
play. The lesson had begun like so many others,
with Ruth throwing out questions and the students
handing back duds. They plodded on like this—
nothing interesting, no grist for exploration—until,
more than half an hour into class, a boy named
Arif stepped to the front. The problem was 2 × (3
× x), and they were trying it out with different
numbers standing in for x. In this case, x equaled
35, and another student had offered the solution,
210, counting out the calculation with checkers.
Arif volunteered to do the problem another way.
But now he stood at the board, stuck. Ruth
listened again to the awkward silence that
followed. “I am confused,” Arif said. “Why?”
Ruth asked. “Because over here we added three of
them [35] and we got 105,” Arif said, “and I
thought over here we were supposed to add two
more of them.” He meant two more of 35, but the
other student had added two more 105s, not two
more 35s.
Ruth could remember the moment now, and
Magdalene was right—she had been struck by a
“bright idea.” While preparing for class, she
herself had misread 2 × (3 × x) as 3 × x + 2 × x.
She had soon realized her error; the problem
actually called for multiplying, not adding—for
tripling x and then doubling the product. This was
a common misunderstanding of multistep
multiplication. When Arif said he thought they
should add two more 35s, Ruth could tell he’d
made the same mistake. “Does anybody have
thoughts for Arif about this?” she asked the class.
A student named Bob jumped in, but Ruth
heard herself cut him off. She remembered why.
Knowing exactly where she wanted the discussion
to go—a point about order of operations and the
properties of multiplication—she had seen almost
immediately that Bob’s answer wouldn’t take
them there. He hadn’t understood Arif’s
confusion, so his comment wasn’t going to move
anything forward.
That was the moment when she said those
words—“I would like to show you something
here.” She walked the class back through their
steps, starting with 35 × 3 = 105. “Now,” she
heard herself saying, “and this is what some
people were having problems with, I want to
double this.” “This” meant the 105—not, she was
making clear, the 35. She drew an arrow from the
105 and wrote “2x” over it, the class’s symbol for
multiplying a number by 2.
A chorus of “Ohhhhhhhh” filled the room.
Listening to the tape, Ruth experienced her own
aha! moment—not about math, but about
teaching. “I did see what they needed,” she wrote
in her own journal. “The point was to see the
connection between addition and multiplication . .
. They were missing the point and I could see
that.”
In her notebook that night, Ruth paraphrased
the lesson she took from Magdalene’s comment.
“Things came together in that moment because I
was thinking about the subject, listening to the
students and trying to make sense of what they
were saying,” she wrote, “and then I acted.”
Discussions wouldn’t work if she simply let the
students talk on their own. The best exchanges
actually happened when she figured out what the
students needed to understand and guided their
conversation to a place where she could teach it to
them.
By the end of the year, the challenge was not
how to get a discussion going but how to end it.
(Class often went on straight through the bell,
stopped only by the cries of other children making
their way to lunch.) Ruth still had a lot to learn,
but the success was undeniable. Now her only
question was, “How do I keep it up?”

At Spartan Village, Ruth wasn’t the only one

mastering the techniques of TKOT. Down the hall,
a veteran teacher named Sylvia Rundquist was
studying with Deborah Ball the same way Ruth
watched Magdalene—and changing her own
teaching in response.
 Sylvia was nineteen years older than Deborah,
and her teaching experience dated back to the
1960s. She taught third grade but gave her class
over to Deborah for math. Sitting in the back of
her own classroom during Deborah’s hour each
day, she became a student herself, full of
questions. Was zero odd or was it even? Why
wasn’t –7 prime if 7 was? Did 7 × 4 really mean
something different from 4 × 7? One morning
each week, she and Deborah met outside of
school, talking over the answers. (Zero is even;
negative numbers aren’t prime because they’re all
divisible by –1 as well as 1; and yes, though they
both have the same answer, the difference between
7 × 4 and 4 × 7 is an important concept for math
teachers to know.)*
Watching Deborah teach made Sylvia question
her own teaching. Midway through the year, she
began to refer to her old habits as BDB, “Before
Deborah Ball.” One day, attending a required
professional development training on math, Sylvia
found herself following along easily—and even, in
some instances, catching errors. “[The leader]
simply announced that 3 × 4 and 4 × 3 were the
SAME!” she wrote to Deborah in an e-mail later
that night. “Val and I strongly disagreed with her,
and she agreed that they were different, but the
end result was the same . . . Amen and thank you!
(It’s working.)”
One major challenge for Sylvia was managing
her fear. Deborah always looked so calm and
serious. When Sylvia opened up a discussion for
children’s thoughts and ideas, her heart raced, her
stomach got tight, and her face grew warm. What
if a student asked a question she didn’t know how
to answer? TKOT obviously worked, but it was
also scary. To do it well, Sylvia didn’t just have to
learn about math and how children understood it.
She had to muster a kind of confidence she hadn’t
previously thought to find.
Sylvia found herself changing the way she got
ready for class. “Whereas when she used to go in
on weekends it was to clean up and to correct
papers,” Sylvia and Deborah wrote in a summary
of the experience, “now she looked for materials,
read, and organized areas of the room. She tried to
imagine various paths the students might want to
take in investigating what was going on with their
plants, or their bread mold cultures, or their
magnets. She looked for books. She gathered
magazines and newspapers. In short, she realized
that she was preparing, rather than planning, for
teaching.”
Like Ruth, Sylvia became more comfortable
over time. She never taught a full math lesson, but
the effects of what she saw trickled over into the
way she taught science, reading, and English. She
stopped using a basal reader, one of those
textbooks with pre-prepared passages, and started
having the class read complete works of children’s
literature. “What do the rest of you think?” she’d
ask, moving from the usual ask-tell ping-pong to
something looser. “Facilitating,” she called it.
 Sylvia and Ruth weren’t the only ones learning
from Magdalene and Deborah. Every day,
Magdalene invited MSU undergrads to sit in the
back of her fifth-grade classroom and observe.
Every week, she met with a small group of
Spartan Village teachers to work on math
problems and talk about teaching. Deborah,
meanwhile, continued playing pedagogical
daredevil, teaching mini-lessons in other teachers’
classrooms.
At first, the visiting undergrads’ reactions
traced the same superficial territory that had
inspired Magdalene to make the tapes. “The
teacher doesn’t say much—she doesn’t do
anything to reinforce the kids who are getting it
right,” one student said. “I wonder if this is a
gifted class,” said another. “Most third graders
can’t talk like this.” And, noting that the students
worked on just one problem a day, another
wondered, “Don’t they need to get through
everything they are supposed to learn in third
grade? How can they do that if they work so long
on one simple problem at a time?”
By the end of the year, after watching videos
and also practicing the material on their own
(writing and solving fraction problems, for
example), their opinions had changed. They
noticed the ways that Deborah and Magdalene did
give feedback, if not by simply decreeing each
answer right or wrong. They saw that in the space
of one problem, the class often touched multiple
parts of the curriculum. And they watched as
misunderstandings that the teacher seemed to
ignore one day were taken up and obliterated the
next. They even shed their ideas about their own
math abilities. “I’m just not a math person,”
they’d said at the beginning. By the end, one such
student wrote, “This course has enlightened me to
a whole world.”

As the years went by, Judy Lanier’s ambitions

grew. She gave the model that Magdalene and
Deborah were creating at Spartan Village an
official name, the “professional development
school,” a modern-day lab school. And with a
grant from the Michigan state legislature, she
began creating more of them. By the mid-1990s,
more than a dozen MSU professors had made
home bases of local schools and embarked on new
teaching research.
These professors were just the first wave of the
recruits Judy and Lee Shulman brought in. In
1986, two years into their hesitant tryout,
Magdalene and her husband, David Cohen, had
decided to stay, making David perhaps the first
tenured Harvard professor in history to leave
Cambridge for East Lansing. With David on
board, Judy began persuading professors from all
across the country to come to Michigan State.
Judy needed the extra faculty members because
she was eyeing an even bigger expansion. The
same year Magdalene and David decided to
commit to MSU, Judy launched a national
campaign with fellow disaffected ed school deans.
They called themselves the Holmes Group after a
maverick Harvard dean who was committed to
training teachers. Their first report urged ed
schools to better prepare teachers or “surrender
their franchise.” Among the group’s
recommendations: create more professional
development schools. By the end of the year, to
everyone’s surprise, membership in the group
included more than a hundred deans from colleges
and universities all across the country.
It was hard not to feel like MSU was the center
of a new universe, ground zero for a new national
reform movement. The sense had been underlined
in 1985, when the California Department of
Education announced its intention to adopt
“teaching for understanding” in math classes
throughout the state. A year later, David and a
group of young MSU researchers boarded a plane
for California. The group included Deborah and
Ruth, though not Magdalene, who preferred to
study teaching rather than policy. They were
excited. What would it look like if an entire state
committed to teaching math with more than just
rote exercises and memorization? They were about
to see for themselves.
Over the next several years, the group observed
classrooms across the state, watching elementary
school teachers teach math. They saw some
promising changes. One teacher who viewed the
reforms with suspicion nonetheless used the new
method of teaching fractions and reported being
“amazed” by what the students achieved. “He
never had imagined that his fifth graders could
think and reason in such advanced ways,” the
team wrote in one report. Another teacher, a
woman David called Mrs. Oublier (a pseudonym),
proudly declared that her classroom had
undergone a “revolution” as a result of the new
ideas.
But on closer inspection, the MSU team
wondered how extensive California’s changes
really were. Mrs. Oublier’s “revolution”—which
David observed from the back of the room,
crouched among the second-graders—seemed to
have real limits. Following the state’s decree that
math lessons “involve concrete experiences” with
numbers, Mrs. Oublier had replaced her pen-and-
paper worksheets with “manipulatives”—little
dried beans and drinking straws. Instead of
seating the children in rows, she arranged them in
clusters of four or five, in line with the state’s new
emphasis on “cooperative learning groups.” And
she zealously incorporated new topics that the
state said were important, like estimation.
Yet each adjustment did little to achieve the
state’s goals. “Concrete experiences,” for instance,
were supposed to help children “develop a sense
of what numbers mean and how they are related,”
according to the state’s new math framework. Yet
Mrs. Oublier paid more attention to the activities
themselves than to the math they were supposed to
teach. In one activity that David observed, the
children used beans and cups to model place
value. But Mrs. Oublier focused most of her time
on whether the children were holding the beans
correctly, sometimes physically moving their arms
to make sure they made the motions as she’d
instructed. When she got to the activity’s key
mathematical point—the moment when the
students had to enact subtraction with regrouping,
effectively taking a larger number from a smaller
one—she flew right past it. They did the exchange
and they moved on—no emphasis or discussion.
The supposedly cooperative learning groups,
meanwhile, were intended to give Mrs. Oublier’s
students opportunities for “speculating,
questioning, and explaining concepts in order to
clarify their own thinking.” But David never once
saw students speak to each other about math.
“Indeed,” he wrote, “Mrs. O specifically
discouraged students from speaking with each
other, in her efforts to keep class orderly and
quiet.” As far as David could tell, she used the
groups only as a means to call on individual
children to come up to the board (for example, so
that they could note their response to a yes-or-no
question for use in a graphing activity), to pass out
or collect papers or materials, and to dismiss the
class for lunch and recess. “She would let the
quietest and tidiest group go first,” David
observed.
Another lesson David sat in on had to do with
estimation. Like the place value lesson, the
activity had potential. Mrs. Oublier asked the
students to guess how many paper clips it would
take to line an entire edge of her desk, and then,
after they wrote down their ideas, she collected
each student’s guess on the chalkboard, asking
each time if the class found the guess
“reasonable.” But instead of discussing what
makes a guess reasonable, or helping the students
to discriminate between more or less reasonable
estimates, Mrs. O treated all the guesses equally
—even some that were obviously far off.
Mrs. O wasn’t the only one whose revolution
fell short. The team visited the classrooms of
nearly three dozen teachers and, wrote one
researcher, “To a one, we never saw radical
change.” The teachers sat children in groups and
even assigned each group member a “cooperative
learning” role, but the roles didn’t translate into
conversations about math. They emphasized the
importance of knowing the “why” of a procedure,
but only accepted one kind of why as correct, even
when more existed. And when students presented
explanations that teachers didn’t understand,
instead of digging into the ideas, the teachers
steered away.
Deborah watched that particular drama play out
in a lesson focused on one of her own old teaching
challenges: subtraction with regrouping. After
explaining “Mrs. Turner’s law of math”—“Never
subtract the top number from the bottom
number”—the teacher harped so much on the
importance of regrouping, or “borrowing,” that
one little boy borrowed on every subtraction
problem, even when it wasn’t necessary. But
instead of unpacking the child’s
misunderstanding, Mrs. Turner (a pseudonym)
seemed to brush it aside. After asking him a short
series of stacked questions and getting the desired
response (“You have 4 cookies. Can you eat 3
cookies? . . . So there’s no reason to borrow
there.”), she moved on.
What explained these poor choices? Some
might argue the problem was an active resistance
to change. And there might have been some
teachers in California who did resist the ideas in
the new California math framework. One MSU
researcher reported visiting a teacher who
swapped the problem-solving pages of a new
textbook his district had adopted for old
worksheets. Another scoffed at the term teaching
for understanding. “What do they think we’ve
been doing—teaching for misunderstanding?” the
teacher asked the MSU team.
But many teachers, like Mrs. Oublier, plainly
embraced the changes. What stopped them from
implementing the ideas more effectively wasn’t a
lack of will, but a lack of clarity about what to do.
Like Ruth and Sylvia, the California teachers were
struggling to understand students’ ideas, figure
out what the students needed to know, and then
use that information to respond. They thought that
simply giving students a chance to talk was
enough. But without the mathematical training to
respond to students’ comments, they weren’t able
to translate confusion into understanding.
Change was also difficult without good models.
In the absence of proper coaching, many like Mrs.
Oublier believed they had undertaken a revolution.
And with many visible changes in their
classrooms—more children talking or playing
with blocks—they had reason to believe the
revolution was real. Yet when David asked Mrs.
Oublier if she had actually read the state’s
manifesto outlining the changes, she couldn’t
remember. That response was repeated over and
over again. Outside of a small group of math
specialists, who had their own worn copies of the
math framework, teachers told the MSU
researchers that they’d either never read the
document or didn’t even know it existed. The
teachers did receive new textbooks, but the books
had not actually made the changes that California
education officials hoped for. Despite hard
lobbying by the state’s Department of Education,
publishers’ revisions were minimal. According to
Suzanne Wilson’s history of the period, California
Dreaming, state officials “estimated that 90
percent of the texts had remained essentially the
same.”
Professional development sessions, meanwhile,
made matters worse. Wilson watched one session
in which an instructor, explaining the new focus
on open-ended rather than multiple-choice
problems, emphasized that students must
communicate their ideas clearly but failed to
mention that teachers also need to make sure the
students’ answers are correct.
For her part, Mrs. Oublier relied on a book
written before the framework came out, a teacher’s
guide called Math Their Way. Reading it, David
found that the book centered on a strange idea.
Young children can’t actually understand abstract
numbers, it argued. But if they work enough with
physical representations of numbers—beans, say,
or straws—then when they are old enough,
numbers themselves will come quite “naturally.”
The process, the book said, would be “effortless.”
To MSU researchers, Mrs. Oublier’s decision to
use Math Their Way was hard to understand. The
book was not part of the state’s reforms. Indeed,
its magical thinking directly contradicted
psychologists’ findings about how much abstract
math young children are capable of doing.
But Mrs. Oublier’s decision had its own logic.
As Ruth and Sylvia had found, changing the way
you taught was a major undertaking. A teacher
had to revise everything from the kinds of
questions she asked to her very understanding of
the subject she was teaching. Implementing the
activities in Math Their Way, meanwhile, was
more like what Mrs. Oublier did with the desks: a
redesign, but not an overhaul. The same old wine
in new bottles, David said. She could carry out the
activities without rebuilding her core beliefs.
More than that, nobody had challenged Mrs.
Oublier or any of the California teachers on their
fidelity to the reforms. Mrs. Oublier’s principal
admired the changes in her classroom and even
called it a model for others. Instead of really
teaching Mrs. Oublier, giving her opportunities to
learn, and noting what she did and did not
understand, the state simply said, here’s the
framework; good luck.

Back at MSU, Judy Lanier’s plans were faltering

too. Support for overhauling the ed school had
never been universal. But in the beginning, Judy’s
supporters had usually drowned out the skeptics.
Now the balance began to flip. Hiring education
researchers who also did teacher education and
taught in a school classroom was a lovely ideal,
but difficult to carry out. Magdalene’s and
Deborah’s positions at Spartan Village had grown
organically. Building new relationships with new
schools took time. According to a history of the
period by Francesca Forzani, one faculty member
had to spend a year “hanging out” in a school
before a teacher finally agreed to collaborate with
her.
Time spent in a professional development
school, meanwhile, meant time away from the
usual tasks of being a professor, like doing
original research and joining professional groups.
According to Forzani, several young professors
scaled back their involvement in professional
development schools to focus on boosting their
academic résumés.
Harry Judge, the observer from Oxford, had
predicted this: the American practice of tenure,
historically determined not by the number of days
spent working in elementary schools, but by the
number of publications in peer-reviewed journals,
would undermine ed school reform efforts. Judy
Lanier swore that those who followed her into the
classroom would not be punished, but, according
to one professor, young faculty “saw the writing
on the wall in terms of the productivity expected
for tenure.”
Instead of relaxing her goals, though, Judy sped
up. Even with more professional development
schools under way, a majority of MSU undergrads
still did their student teaching at schools selected
essentially at random. There simply weren’t
enough professional development schools to
accommodate every trainee.
Judy’s ambitions were also influenced by the
people whose support she spent more and more of
her time courting—potential donors who might
provide the money needed to expand her
operations. Especially influential was Alfred
Taubman, the billionaire A&W proprietor, who
began to brainstorm with Judy about how to take
the professional development school idea
statewide—a “scaling” project that he modeled on
his own experience growing supermarkets and
chain restaurants. The spiraling plan called for
building fifty or sixty professional development
schools all across the state. “M.S.U. has only 140
faculty members and the numbers doing teacher
education are even smaller,” a faculty member
told Forzani. “That’s just not enough people to
make it work.” For her part, Judy (known today as
Judith Gallagher) points out that she pursued the
expansion only at the insistence of some of the
same colleagues who later questioned it. Whatever
the source of the plans’ ambitions was, though,
the gap between what Judy and other faculty
members believed was needed in both Michigan
and the rest of the country and what the resources
at hand made possible was undeniable.
According to a faculty member interviewed by
Forzani, Judy came to think of her job as
analogous to the queen Scheherazade from One
Thousand and One Nights. Just as Scheherazade
had to tell the king a new captivating story every
night to stay alive, Judy felt that she constantly
had to spin better and better plans before potential
funders. “You had to propose a grand vision that
is [in fact] cockamamie,” the faculty member told
Forzani. “I mean, people in this place would read
Judy’s plans and say, ‘What is she thinking?!’ But
that’s what it took in the corporate community;
that’s the story you’ve got to tell.”
Fed up, other faculty members aired their
frustrations publicly, publishing a newsletter filled
with “enraged and sometimes satirical essays
about Judy as well as cartoons that depicted, for
example, the dean smashing hammers and other
instruments over the heads of her colleagues.” By
the early 1990s, a group of particularly frustrated
faculty members—many of them members of the
old boys’ club Judy was trying to change—began
holding regular meetings to discuss how to resist
her plans.*

Even if Judy and company were unable to reform

the university system and the ed school institution,
Spartan Village might still have offered an
example of what professional development
schools could do on their own. It’s true that
Deborah’s, Magdalene’s, and Ruth’s relationships
with the school had been forged through MSU,
but much of their work could have continued even
without an ed school to support it.
But the work at Spartan Village proved
unsustainable too. Though she never mentioned it
to Deborah, Principal Jessie Fry caught a lot of
complaints from teachers who didn’t want to visit
each other’s classrooms. At first, Deborah’s
unique powers of persuasion and Jessie’s own
iron will kept the changes intact. But as they
sought more ambitious reforms, Jessie began
running into roadblocks more formidable than
veteran teachers’ skepticism.
The evolution of the staff meeting told the
whole tale. Traditionally, the meeting existed for
the purpose of exchanging business unrelated to
teaching: the state of the school budget,
applications for supplies, news from the district,
building concerns, fire drills, tornado drills,
parent-teacher conferences, schedules, upcoming
events. But as the school began to change, the
meeting did too. “I would say, ‘So-and-so, I was
sitting in your room and I saw what you were
doing,’ ” Jessie described. “Would you kind of
share what you were doing with so-and-so little
kid?” Teachers started out shy, but over time,
more and more of them shared, until eventually,
the staff meetings had so many non-“business”
items that Jessie ran out of space. The meeting
simply wasn’t long enough to deal with both
school business and teaching practices.
Finding another time to meet was not easy.
According to all the official district policies, the
teacher and teacher’s aide contracts, and the
school calendar, the school week was full. The
Spartan Village school also had no physical space
to meet. The school already used a tiny hallway
alcove as a library, and the room where they met
for staff meetings doubled as a classroom. So
Jessie had to negotiate. Through meetings with
both relevant unions, as well as the school district
and the school board, she won permission to
change the school’s calendar, eventually building
in extra time for “professional development.”
The agreement solved the meeting-time
problem, but not its corollary. Teachers wanted to
observe their colleagues but had no one to step in
and watch their own classes. Determined to give
them more chances to observe, Jessie negotiated a
separate arrangement to bring more substitutes to
Spartan Village—and, because her teachers
wouldn’t leave their students with just anyone,
Jessie had to negotiate something even trickier:
permission for her teachers to screen their subs,
something the district had not previously allowed.
Grants, meanwhile, paid for additions to the
school building. She built a new library and a new
room just for teachers to meet. Improving
teaching, it turned out, required not only new job
descriptions for the teachers, but also a new floor
plan.
After all of these acrobatic feats, Jessie still
faced another hurdle. Like the staff meetings,
Jessie’s own official schedule allowed for only the
“business” part of her job. She alone was
responsible for writing the school’s budget,
ordering supplies, managing the maintenance
staff, and dealing with parents. Working with
teachers on their teaching—the schedule simply
didn’t allow for that. Jessie began working even
more overtime than usual. During the school day
she moved from one classroom to another,
watching teachers work, leaving notes with
feedback, and thinking of which teachers might
benefit from talking together. At night, she played
official principal, filling out the endless paperwork
that kept the school humming. Most nights, she
didn’t leave school until 8:00 or 9:00 p.m.
The arrangements worked for a while, but over
time, strains began to appear. Each time a new
superintendent arrived—Jessie was principal
through at least four—she had to defend the
Spartan Village exceptions. Every time budgets
grew tight, the school board always seemed to
turn to Spartan Village. Did that training school
across the tracks really need to exist? The long
hours and rising stress strained Jessie’s private
life. Her marriage ended, and her health declined.
Teachers, meanwhile, came to resent the
growing demands on their own schedules as more
MSU professors sat in on their new professional
development meetings. Some refused to attend the
meetings at all. Others joked about being
“bugged”; to study the school’s transformation,
MSU researchers had begun videotaping the
school’s meetings. A few teachers even refused to
let Jessie watch them teach. “There were a couple
of people that, they didn’t want anything to do
with the professional development school,” Jessie
says. “They didn’t want to meet extra hours or
anything. They wanted to just stay in their
classrooms, do their teaching. They would do
everything else I required, the lesson plans and all
this. ‘Nope! Don’t want you to come into my
room.’ ”
When Judy Lanier asked Spartan Village to
become not just a professional development
school, but a demonstration school, teaching not
just future teachers but other schools, Jessie and
the staff said no. It would be too disruptive.
“You’ve got to remember,” Jessie says, “that this
is the university.” She held out one hand. “Here’s
us out here.” She held out another, way over to the
other side. “The two don’t meet. So we had to
learn to work together and to share our knowledge
and our own expertise. We are on one side, we’re
a team, and they’re on the other side, and it’s like
us and them, us and them. We weren’t all
together. We grew together.”
Until they didn’t.
Soon, David and Magdalene had announced
their plans to leave MSU for the University of
Michigan, beginning a wave of departures. By
1992, the self-described faculty “mutiny” against
Judy Lanier had expanded to include even
Taubman, who told Judy she needed to become
more realistic about what she could accomplish,
according to Forzani’s account.
That October, Judy resigned as dean. A few
years later, she moved to Flint, Michigan, to work
with the distressed city’s public schools. Seven
years later, when those reforms crumbled under
opposition too, she left education altogether and
moved permanently to Beaver Island in Lake
Michigan, the most remote inhabited island in the
Great Lakes. She has lived there ever since.

* Though it’s true that the two expressions are

equivalent (both equal 28) math teachers need to
see that 4 × 7 and 7 × 4 represent different ideas:
one means four groups of seven; the other, seven
groups of four. (Imagine, for instance, seven cars
with four wheels each versus four cars with seven
wheels each; both have a total of twenty-eight
wheels, but through very different means.)
The distinction grows even more important in
division. The corollary of the idea that 7 × 4 is
different from 4 × 7 is the fact that there are two
ways to understand the meaning of 28 ÷ 4. One,
called partitive division, asks questions like, If we
have 28 wheels and 4 cars, how many wheels can
we give to each car? Another, called quotative
division, asks, If we have 28 wheels, and we
know we need 4 wheels per car, how many cars
can we fit with new wheels? In both cases, the
answer is 7, but again, the configurations look
very different.
Armed with this understanding, Sylvia no doubt
would have been able to answer the division-by-½
problem that stumped so many MSU undergrads.
Dividing by ½ makes no sense from the
perspective of the most common conception of
division (how can you make ½ number of
groups?) but it makes perfect sense conceived
quotatively (you can easily make groups of size
½).

* Judy, for her part, disputes the account that

building resentment against the reform work at
Michigan State was targeted at her personally. She
remembers support for the reforms staying strong
until after she took time off as dean in 1989.
(Judith Gallagher, interview by Jessica Campbell,
November 2013.)
 4
KNEAD AND RISE

Creating a country full of teachers like Magdalene

Lampert and Deborah Ball might have failed in
the United States, but that didn’t necessarily mean
it was impossible. Magdalene learned this lesson
one day in 1985, after giving a talk at the
University of Chicago. She’d opened her remarks
with a warning. The videos the audience was
about to see, taken in her classroom at Spartan
Village during a series of lessons on
multiplication, would depict teaching that
deviated markedly from traditional math class
culture. As far as she knew, no other teacher in the
world taught in quite the same way.
That last comment stuck in the mind of one
member of the audience as he watched the videos.
James Stigler, then a young psychology professor
at the University of Chicago, knew that
Magdalene was only partially right. Yes, her
teaching did look different from that in most
American schools. But, as Stigler told Magdalene
later, she was not, in fact, the only teacher in the
world who taught that way. Indeed, a whole group
of teachers taught almost exactly like she did.
They just happened to live in Japan.
Stigler knew because he’d seen them do it,
starting when he was in grad school at the
University of Michigan. He’d gone to Japan with
Harold Stevenson, a psychology professor who
studied children in Japan and China who
struggled with reading. By comparing how
children learned to read in different languages,
Stevenson and Stigler hoped to get a better sense
of the process in general. But the pair’s focus
shifted when they ran a test of students’ math
achievement.
Comparing reading abilities, they had found
some discrepancies between countries. But, says
Stigler, “the reading differences were minor
compared to the math differences.” Japan stood
out most of all. Comparing children from
Minneapolis, Taipei, and Sendai, they found that
73 percent of Japanese six-year-olds scored higher
than the average American child. The advantage
grew even larger as children got older. Among
ten-year-olds, the percentage of Japanese students
scoring higher than the average American was 92.
Even the Japanese ten-year-olds with the lowest
average math scores in Sendai scored better than
those with the highest scores in Minneapolis.
Stigler and Stevenson’s finding echoed a
growing set of international comparisons that put
statistics behind deepening anxiety about
America’s educational standing, especially in
matters of science and math. One study funded by
the US government compared achievement across
twelve different countries (not just the United
States and Japan, but also Israel, Sweden,
England, and others) and found that the average
Japanese student scored as well on a math test as
the top 1 percent of students around the world.
Another, comparing high school students in
Illinois and Japan, found that the average
Japanese student performed better than roughly 98
percent of Americans. A third study,
commissioned by the Dallas Times Herald, found
that out of eight countries, Japan ranked number
one in math achievement, while the United States
ranked number eight. “There is no doubt that the
Japanese . . . have built up their educational
system in a manner comparable to the heralded
‘economic miracle,’ ” a New York Times reporter
visiting Japan concluded, just after the release of
the Nation at Risk report.
With concern mounting about American
schools’ performance, explanations for the gap
proliferated. Some commentators pointed to
cultural factors, noting the Japanese emphasis on
effort above ability. Stevenson and Stigler
themselves argued that home life had to play a
role; in Japan, they found, 98 percent of fifth-
graders had a desk at home, while the percentage
among their American counterparts was 63.
Others speculated that Japanese children had
inherently higher IQs, though Stevenson and
Stigler could find no significant differences when
they gave children a test of cognitive ability.
Reviewing the math results, Stigler thought the
extracurricular factors must be important. But he
doubted that they could completely explain the
difference. “It’s not like your parents sit you down
and teach you algebra,” he says. “You go to
school, and your teachers are teaching you these
things.” What the Japanese teachers did in the
classroom had to matter too. Each time he visited
a Japanese school, he began to ask the local hosts
a favor. Would they mind taking him into a
classroom to watch a lesson or two?
On early visits, Stigler had noticed superficial
differences. Instead of the one-floor buildings
common in America, in Japan elementary schools
were all three-story concrete palisades, with the
hallways wrapping around the circumference like
a multilevel motel. (Indeed, every elementary
school had a full swimming pool, although not
usually on the ground floor.) Children snapped
between opposite poles of activity, shrieking and
running chaotically one minute—the boys on stilts
twice their size, the girls on unicycles—and
silently studying the next. At the front door,
everyone exchanged their shoes for slippers (as is
standard everywhere in Japan). Schools used heat
sparingly, creating a constant chill. And when the
principal received visitors, he always served them
hot tea.
But it was only when he started visiting
classrooms—not just poking a head in, but sitting
through an entire lesson—that Stigler noticed the
deeper differences. Japanese math teachers led
class with a different pace, structure, and tone
than did other countries’ teachers. Instead of a
series of problems, the teachers used just one, and
instead of leading students through procedures,
they let students do much more talking and
thinking.
Watching Magdalene’s videos in Chicago
transported Stigler back to those classrooms in
Sendai. She had the same slow, methodical way of
studying students’ work, asking a question, and
channeling their replies toward the desired
conclusion. How had such uncannily similar
pedagogy evolved in teachers an ocean apart? And
how had the Japanese managed to do what eluded
Americans, training what appeared to be the entire
profession to use TKOT?
In the years after meeting Magdalene, Stigler
could only guess. The number of Japanese
classrooms that he’d visited was tiny compared to
the total number of schools in the country. His
knowledge of what happened inside American
classrooms, meanwhile, was also imperfect. Many
people thought they knew how most American
teachers taught math, but no one had ever
mounted a large-scale scientific study to confirm
it.
Stigler’s opportunity arrived in the early 1990s,
when the group behind the new international tests
was preparing its third and largest comparison
study yet: the Third International Mathematics and
Science Study, or TIMSS. This time the
participating countries (a group that had now
swelled from twelve to more than forty) agreed to
examine not only scores on achievement tests, but
also other measures that might shed light on
international differences—including a variable to
account for classroom teaching.
Applying the sampling methods that had been
used to compare achievement, TIMSS could build
the first-ever study of international classroom
teaching, using video cameras to capture teachers
and students at work. Because video technology
was still relatively expensive, they narrowed the
recording to just three countries. In addition to the
United States, TIMSS organizers picked Germany
(a major economic competitor) and Japan (the
reigning king of the international tests).
They selected James Stigler to lead the study.

Stigler guessed that he would find differences

across countries. But he didn’t anticipate just how
similar teachers would turn out to be within the
countries. Common lore, of course, held that a
wide gap separated the best American teachers
from the worst. But compared to their German and
Japanese counterparts, even the two most
disparate Americans looked identical.
The consistencies stood out most when Stigler
got people from different countries in one room to
watch the videos together. One day early on, a
Japanese researcher abruptly stopped a video of an
American classroom right in the middle of the
lesson. “What was that?” he asked. The teacher in
the video had been demonstrating a procedure at
the chalkboard when an invisible voice interrupted
him. “May I have your attention, please,” the
voice said. “All students riding in bus thirty-one,
you will meet your bus in the rear of the school
today, not in the front of the school. Teachers
please take note of this and remind your students.”
The Americans had barely noticed the public
address interruption. “Oh, nothing,” they told the
Japanese researcher, pressing the button to start
the video again. But the Japanese researcher
persisted. “What do you mean, nothing?” he said.
Stigler wrote:

As we patiently tried to explain that it

was just a P.A. announcement, he
became more and more incredulous.
Were we implying that it was normal to
interrupt a lesson? How could that ever
happen? Such interruptions would never
happen in Japan, he said, because they
would ruin the flow of the lesson. As he
went on, we began to wonder whether
 this interruption was more significant
than we had thought.

Later, going over all the videos, they found that

the researcher was right. Thirty-one percent of the
American lessons contained some kind of an
interruption, either a PA announcement or a
visitor walking in to deal with administrative
business (like collecting the lunch count). Zero of
the Japanese lessons did. But they never would
have thought to count interruptions, had the
observer not singled out that moment. Sometimes
the most distinctive features of a country’s
teaching were also the most difficult for natives to
notice.
One striking example was the way teachers
structured their lessons. American teachers rarely
talked about lesson structure—the way class
proceeds from a beginning to a middle to an end
—and yet, watching each individual teacher at
work, Stigler felt as though they’d all read the
same recipe. “A cultural script,” he called it. The
American and Japanese scripts were the most
different from each other—a limerick versus a
sonnet. Some American teachers called their
pattern “I, We, You”: After checking homework,
teachers announced the day’s topic, demonstrating
a new procedure—Today, we’re going to talk
about dividing a two-digit number by a one-digit
number (I). Then they led the class in trying out a
sample problem together—Let’s try out the steps
for 24 ÷ 6 (We). Finally, they let students work
through similar problems on their own, usually by
silently making their way through a worksheet
—Keep your eyes on your own paper. If you have
a question, raise your hand! (You).
The Japanese teachers, meanwhile, turned “I,
We, You” inside out. You might call their version
“You, Y’all, We.” They began not with an
introduction, but a single problem that students
spent ten or twenty minutes working through
alone—24 chocolates to be shared with x number
of people (no leftovers); come up with as many
solutions as you can (You). While the students
worked, the teacher wove through the students’
desks, studying what they came up with and
taking notes to remember who had which idea.
Sometimes the teacher then deployed the students
to discuss the problem in small groups (Y’all).
Next, the teacher brought them back to the whole
group, asking students to present their different
ideas for how to solve the problem on the
chalkboard. Give the answer and the reason for
your answer. Finally, the teacher led a discussion,
guiding students to a shared conclusion—What
did you learn from today’s problem, or what new
questions do you have, if any? (We).
The patterns didn’t dictate everything each
teacher did, of course, and the researchers found
some cases of departures. But even departures
happened inside the spirit of the scripts, which
encouraged some moves more than others. Take
the kinds of questions each country’s teachers
asked. Americans asked a lot of simple questions
and sought quick answers. 1 − 4: What does it
equal? Japanese teachers, working at the slower
pace provided by a single focused problem, used
questions not simply to understand whether the
child had the right answer, but to peek into her
mind, discerning what she understood and what
she didn’t: Who had the same thinking? Anything
to add to this way of thinking? Did anybody else
use another way?
In a ministudy of four lessons, two American
and two Japanese, Stigler counted the types of
questions that arose in each one. In the Japanese
lessons, the most common question took the form
of what he called “explain how or why”: How did
you find the area of this triangle? for instance, or
Why is the area here 17? Problems, meanwhile,
seemed to be designed with great care: they were
generative enough to fill one or two forty-five-
minute lessons each, and carefully selected to lead
students not just through interesting math but to
an important new idea. They tied lessons together
like daisy chains, with the fruits of one day’s
problem leading directly to the task of the next.
On its own, the task of deriving a formula to find
the area of any triangle would be a lot to ask of a
fifth-grader, but coming right on the heels of a
lesson on parallelograms, children could use the
formulas they’d derived just days earlier (often
conveniently pasted to a wall for easy review) to
come up with that day’s answer.
Stigler called the second most common
question in the Japanese lessons “check status”:
Who agrees? Japanese teachers often asked,
tallying up whether other students had been
persuaded by a classmate’s idea. Or, checking
whether students were following the progression
of thoughts: Is anyone confused? In the American
lessons, meanwhile, the most common question
was what Stigler called “name/identify”: What
kind of triangles have we studied so far? the
American teacher might say in her version of
review, or What is the length of this shape? The
second most common question was “calculate”:
What is 90 divided by 2? Neither of the two
Japanese teachers asked a “calculate” question,
and neither of the Americans asked a “check
status” question.
The different sorts of questions led to different
forms of participation. The Japanese students
spoke more often and said different things. For
instance, they were much more likely than the
Americans or the Germans to initiate the method
for solving a problem. Whereas students initiated
the solution method in just 9 percent of American
lessons, in Japanese lessons that number was 40
percent, Stigler’s team found. Students in different
countries also did different kinds of work. The
researchers found that 95 percent of American
students’ work fell into the category of “practice,”
while Japanese students spent only 41 percent of
their time practicing. The majority of work fell
into a category the researchers termed
“invent/think.” A solid 53 percent of Japanese
lessons included formal mathematical proofs. In
all the American lessons collected, the researchers
found zero mathematical proofs.
The TIMSS study revealed vastly different
approaches to teaching exactly the same material.
Lessons on the difficult problem of adding
fractions with unlike denominators (for example,
½ + ) exemplified the gulf. American teachers
were encouraged to build up to the challenge step
by step, starting with like denominators ( + ),
and then moving on to the simplest unlike ones (½
+ ¼). They did this only after warning students of
the importance of not adding the denominators
and demonstrating exactly what to do instead.
Japanese teachers, meanwhile, gave students
unlike fractions without commentary. When
students inevitably made mistakes (adding
denominators together, for instance), the teachers
embraced the error as a chance to see why
converting to like denominators makes more
sense.
Even the architecture of the classrooms reflected
the national predilections. To supplement their
lessons visually, for instance, US teachers usually
used overhead projectors, but in Japan, every
observed teacher used the chalkboard. At first it
seemed like a trivial difference. But on closer
inspection, the researchers could see that each
device created a specific mood. In the American
classrooms, where teachers seemed to value
attention more than any other form of participation
(“Eyes on me!”), the overhead forced light onto
everything the teacher wrote. A strategically
placed sheet of paper, meanwhile, covered up
everything but the latest problem. Guiding
students step by step, teachers brought all eyes to
the immediate idea—and prevented any reflection
on what came before. In Japan, where teachers
cared more about the attention students paid to the
ideas as they unfolded, a chalkboard that could
hold the full trajectory of forty-five minutes’ worth
of insights served teachers better.
Taken together, the findings confirmed Stigler’s
hunch. American teachers reported in large
numbers that they knew about the new math
reform ideas that David Cohen and Deborah Ball
had tracked in California. Like Mrs. Oublier,
many reported that they were adopting the ideas in
their own classrooms. But the videos disputed
their accounts. In some cases, the reforms actually
made matters worse. One eighth-grade teacher,
following directions to use calculators for
problems where practicing computation wasn’t the
point, guided her students to use the machines to
find the answer to 1 − 4. (“Take out your
calculators,” she said. “Now, follow along with
me. Push the one. Push the minus sign. Push the
four. Now push the equals sign. What do you
get?”) No wonder parents and some
mathematicians had begun protesting that the
reforms constituted “fuzzy math.” In the warped
way teachers interpreted them, they were fuzzy.
 One surprise finding did not appear in the
videos or even in Stigler’s final report, but in
informal interviews with Japanese teachers and
education leaders. Asked when and how they had
learned to teach this way, they all responded the
same way. The changes, they said, began in the
1980s. Before that, math classes were more like
what Stigler saw in the United States: rote,
mechanical, dull. After the reforms, Japanese
teachers took inspiration from three main sources:
John Dewey, the American philosopher; George
Polya, a Stanford mathematician whose writing
about problem solving had influenced Magdalene
Lampert; and “NCTM,” the acronym for the
standards produced by the National Council of
Teachers of Mathematics—the ones inspired by
Magdalene Lampert and written in part by
Deborah Ball. How did Japanese educators learn
to teach this way? You, they’d sometimes tell
Americans more simply. We learned from you!
The exchanges reminded Stigler of a confusing
conversation repeated often during his first visits
to Japan. Asked to explain the success of the
country’s thriving companies, Japanese would
pronounce with reverence the name of a
management expert: De-Ming. “Gee,” Stigler
thought. “I wonder if this guy’s work has ever
been translated into English!” Only later did he
learn that “De-Ming” was William Edwards
Deming, an Iowa-born statistician and
management consultant who had begun his career
in the United States after World War II, but whose
ideas had gained traction only in Japan. (Later,
after word spread of Deming’s fame in Japan,
American companies paid hefty fees to seek his
advice about how to compete with their Asian
counterparts.)
Like Deming’s work, the NCTM standards had
a more loyal following in Japan than in the
country that birthed them. Not only had the
Japanese discovered the American math
standards; they’d accomplished what California
never could. They’d taken a population of earnest
but ordinary teachers and produced a country full
of Magdalene Lamperts.

How had they done it? While Stigler pondered that

question, Akihiko Takahashi found himself
obsessed with another. The son of a Tokyo police
officer, Akihiko himself had stumbled into
teaching. Then, in the fall of 1991, he and his wife
found themselves in Chicago, halfway across the
world.
Officially, Akihiko had moved under orders
from Japan’s Ministry of Education, which sent
teachers abroad to work at the Japanese schools
serving the children of traveling businessmen—in
Akihiko’s case, the Futabakai School in Chicago.
But in fact, he’d asked for the assignment. During
twelve years of teaching in Japan, he’d become a
careful student of American educators, especially
George Polya, John Dewey, and the NCTM. Now
he wanted to see the schools they’d built. With the
ministry order, not only did he get to go to the
United States; he got to go to Chicago, the home
of Dewey’s original lab school at the University of
Chicago.
At the Futabakai School, Akihiko had to teach
classes, of course, but he also received permission
from the principal to spend part of his time
visiting American schools. It didn’t take long for
the other teachers to grumble. Why is Takahashi
always traveling? The truth was he was searching
for the classrooms he’d read about in books.
Mecca did not reveal itself. At an elementary
school outside the city, the teacher kept saying
“Shh!” There must have been one hundred
“Shh!”s. “I thought, well, that’s only this class,”
Akihiko says. “I came to the wrong class.” But
“Shh!” turned out to be the rule. Math classes, his
specialty, were nothing like what the NCTM had
described. They were rote, tedious, and full of
mistakes. A member of the Japanese Ministry of
Education, traveling in America a few years later,
watched a math teacher calculate that 2 + 3 × 4
equaled 20, first adding 2 and 3 to get 5, and then
multiplying by 4. Astonished, the ministry official
wondered for a minute whether perhaps
Americans followed a different order of
operations, in which addition preceded
multiplication. (He quickly confirmed that math is
indeed the same all around the world.)
Even the Chicago lab school, the one that
Akihiko had read so much about, betrayed no
trace of its founder. “I was shocked. Like, I read
John Dewey!” Akihiko says. “But they don’t do
anything like that.” The Americans produced
wonderful intellectual work on what teaching
could look like, but they had failed to implement
any of it. He’d told the principal he needed time
off to do research on the American classroom. But,
he says, “rather than findings, there were a lot of
puzzles. Good documents and good research and
good materials . . . but somehow it disappears in
the classroom. So how does this happen?”
Not long after the lab school visit, Akihiko took
a trip to the University of Illinois campus in
Champaign. A colleague in Japan had introduced
him to a professor there and his wife, Jack and
Elizabeth Easley. Elizabeth was Japanese
American, and ten years earlier the two had spent
four months embedded in a Tokyo elementary
school, with Elizabeth translating while Jack
observed the Japanese teaching style. The visits to
Japan had inspired Jack to improve math and
science teaching in the United States, but he had
not been able to import the ideas into actual
schools. The problem, he thought, was a lack of
communication between the two worlds:
researchers and teachers. So, with a few
colleagues, he’d created a new group to bridge the
gap. Dialogues in Mathematics Education, he
called it, or DIME—a regular workshop for
professors, teachers, even principals. When he
connected with Akihiko, he invited him to join
DIME for a meeting.
The meeting fascinated Akihiko. People from
all around the Midwest presented work they were
doing in classrooms, generating rich
conversations. But one thing stood out: the DIME
meeting seemed to be each participant’s only
chance to discuss her work, and as far as Akihiko
could tell, the group met only twice a year. Could
this really be the United States’ best mechanism
for translating ideas into practice? Most stunning
of all was the fact, confirmed to him by members
of the group, that the conversations were just that
—talk. The teachers described lessons they gave
and things students said, but they did not see the
practices. When it came to observing actual
lessons—watching each other teach—they simply
had no opportunity. Indeed, the researchers and
teachers viewed it as a triumph that they were
meeting together at all. To Akihiko, the
unusualness of the affair spoke volumes. What
happened in Japan as a matter of everyday
business (meetings between professors and
teachers) was, in the United States, a
revolutionary act.
The realization helped explain something else
that had been puzzling him. Almost every time he
tried to visit an American classroom, he would get
the same frustrating response. Instead of letting
him watch quietly while they taught, teachers
would halt the lesson to welcome their guest from
Japan. Dozens of minutes would disappear as he
introduced himself and fielded questions.
Afterward, his conversations with the teachers
imitated the distraction. Any question he asked
about the actual content of the lesson got batted
away in favor of something completely unrelated
to education. It was as if, instead of colleagues in
the same profession, they were strangers meeting
at a dinner party. What do you think of the United
States? Where do you live? Social questions, not
professional talk. When a conversation did veer in
an interesting direction—How do you teach this
topic in Japan? a teacher might ask, beginning the
discussion Akihiko longed to have—it always
ended too quickly.
The experience would have been unbelievable if
it had not repeated itself so many times. After
visiting more than a handful of math classes, he
understood. The teachers didn’t let him stand in
the back of their classrooms quietly to watch
because nobody ever stood in the back of their
classrooms and watched. The same went for
conversation. They didn’t talk about their teaching
with him because they didn’t discuss their
teaching with anyone.
They had, he realized, no jugyokenkyu.
Translated literally as “lesson study,” jugyokenkyu
is a bucket of practices that Japanese teachers use
to hone their craft, from observing each other at
work to discussing the lesson afterward to
studying curriculum materials with colleagues.
The practice is so pervasive in Japanese schools
that it is like the PA interruption to Americans:
effectively invisible. For a Japanese observer like
Akihiko, asking if schools had jugyokenkyu in
America would be like asking if they had
students.
And here lay the answer to his puzzle. Of
course the American teachers’ work fell short of
the model set by their best thinkers—Polya,
Dewey, and the NCTM. Without jugyokenkyu, his
own classes would have been equally drab.
Without jugyokenkyu, how could you even teach?

Akihiko was not a natural-born teacher. He had

become a teacher mainly because the university
that had accepted him, Tokyo Gakugei,
specialized in education. Even in his final years of
college, he was indifferent to teaching. But during
the second semester of his junior year, he stepped
into the classroom of Takeshi Matsuyama at the
Setagaya Elementary School, and everything
changed.
 The school stood at the end of a curving
cobblestone driveway in Tokyo’s affluent
Setagaya Ward, a residential neighborhood
famous even in inscrutable Tokyo for its mazelike
streets. The grounds were unusually large for a
center-city school, but the building was ordinary:
three drab concrete stories, practical wood floors,
drafty, and cold. Setagaya, however, was different
from any other Japanese elementary school. A
fuzoku school, meaning “attached”—as in, part of
the university, or, in American parlance, a lab
school—Setagaya conducted its hiring with the
thoroughness of a corporate recruitment office,
combing the country for the best teachers. They
had to be true masters because, as fuzoku teachers,
they were responsible for educating both children
and future teachers. For three weeks each autumn,
these college students trooped into the Setagaya
school, breaking into groups of five per teacher.
They lined the back of the classroom, notebooks in
hand, unsure of what to expect.
 Among the masters, Matsuyama stood out. His
public lessons attracted so many teachers that to
make room for them all, he had to hold class in the
cafeteria. This made the job of his student teachers
especially daunting. In order to graduate,
education majors not only had to watch their
assigned master teacher work, they had to
effectively replace him, installing themselves in
his classroom first as observers and then, by the
third week, as a wobbly five-person
approximation of the teacher himself. It worked
like a kind of teaching relay. Each trainee took a
subject, planning five days’ worth of lessons in
language or math or science or history. Then each
took a day. To pass the baton, you had to teach a
day’s lesson in every single subject: the one you
planned and the four you did not. You had to do
this whether or not the teacher before you made it
through the full material the day before. And you
had to do it right under your master teacher’s
nose. Afterward, everyone—the teacher, the
college students, and sometimes even another
outside observer—would sit around a formal table
to talk about what they saw.
During the observation week, the trainees
stayed in Matsuyama’s class until the students left
at 3:00 p.m., and they didn’t leave the school until
they’d finished discussing the day’s events,
usually around eight o’clock. They talked about
what Matsuyama had said and done, but they
spent more time poring over how the students had
responded: what they wrote in their notes; the
ideas they came up with, right and wrong; the
architecture of the group discussion. The rest of
the night was devoted to planning; some days,
these teachers-in-training didn’t go home until
10:00 p.m. It was intense, exhausting, terrifying—
and thrilling. Watching Matsuyama teach, with all
the intellectual rigor that entailed, was inspiring.
The trainees began to come up with their own
ideas. Akihiko saw for the first time what it meant
to be a teacher. He was hooked.
 For his focus, Akihiko selected math, which
was Matsuyama’s specialty too. This was 1977,
and Matsuyama, a student of John Dewey and
George Polya, was an early proponent of the
changes just arriving in Japanese math teaching.
His technique tantalized Akihiko. In grade school,
Akihiko had loved math but hated math class,
where teachers always acted as if there were only
one correct way to solve a problem. Matsuyama
offered the exact opposite. He not only rewarded
students who came up with their own solution
methods; he depended on them.
Admiring Matsuyama’s teaching style and
carrying it out, of course, were two different
things. Akihiko’s first lesson, which he would still
remember thirty years later, began easily enough.
That week, sixth-grade classrooms across Japan
were introducing the concept of proportional
relationships—as in, if 5 cookies cost 300 yen,
then how much do 2 cookies cost? A traditional
lesson might have introduced the topic and then
demonstrated how to calculate the unit rate (60
yen per cookie) and use that to find the answer
(120 yen). But under Matsuyama’s guidance,
Akihiko had devised a problem that guided the
children to map a set of relationships that would
turn out to be proportional. Then he led the
students in a discussion of the line they had
drawn, showing the relationship between the
number of cookies and how many yen they cost:

Everything was going well enough, when a

student raised his hand with a question. Why, he
asked, couldn’t they connect the line all the way
down to zero? Another student asked a different
question: What does the line connecting the dots
mean?

Akihiko was stuck. He knew that, in general, a

graph representing proportional relationships
should connect zero to all the other data points.
However, the quantities in the cookie problem
were discrete; the case of buying less than one
cookie or even zero cookies for zero yen simply
did not exist. But how could he explain that to the
sixth-grader? “I still remember the feeling,”
Akihiko said. “Even though I knew, I could not
explain it. I felt like my back is to a cliff, and I
cannot go back anymore.” He had no answer, but
having no answer was not an option. Like a
student called on to talk about a book he hadn’t
read, he bluffed—and felt that everyone could tell.
In that moment, he knew. Teaching wouldn’t just
be his college major, something to study before
finding his real profession. He was going to make
it his life’s work.
Over the next six years, as he graduated and
went to work at another nearby elementary school,
Akihiko experimented with more Matsuyama-
style lessons. He knew enough not to try the
approach every day. Veering from the textbook’s
suggested lesson plan required more preparation
than he had time to do, and there was no
guarantee of success. Instead, he deployed it in
select cases, always remembering to spend as
much time imagining how students might respond
to a problem as he spent inventing it. Still, no
matter how late he stayed at school, planning, the
lesson would begin with a leap into the unknown.
The students loved to surprise him. He took to
keeping a journal; he wanted to remember every
lesson. When parents complained about the young
new teacher experimenting on their children, he
turned his notes into a newsletter, sent to parents
first once a month and then, by his third year,
every day. If they were going to support their
children, and support Akihiko, the parents needed
to know the math as well.
Year by year, the lessons got better, and the
parents’ confidence grew. By his fifth year, he was
teaching almost every lesson in the single-problem
style. By his sixth, he received unexpected news.
Back at the Setagaya fuzoku school, Akihiko’s old
mentor Matsuyama had just received a promotion
from the school district. The new job meant that
his position as master was now empty, and the
district officials had asked him to suggest a
replacement. He had named Akihiko.

To an American, Akihiko Takahashi seemed like

another diamond teacher in the rough, the
precocious counterpart to the madogiwa (“window
gazers”), as weak employees are called in Japan.
By the time he arrived in Chicago, he’d become as
famous as Matsuyama, giving public lessons that
attracted hundreds, and, in one case, an audience
of a thousand. He had a seemingly magical effect
on children.
But Akihiko knew he was no virtuoso. “It is not
only me,” he always said in English. “Many
people.” After all, it was his mentor, Matsuyama,
who had taught him the new approach to teaching
math. And Matsuyama had crafted the approach
along with other math teachers in Setagaya Ward
and beyond. Together, the group met regularly to
discuss their plans for teaching differently, a
Japanese version of TKOT; at the end of a
discussion, they’d usually invite each other to their
classrooms to study the results. In retrospect, this
was the most important lesson Matsuyama taught
Akihiko: not how to give a lesson, but how to
study teaching, using the cycle of jugyokenkyu to
put his work under a microscope and improve it.
Those three weeks of student teaching at the
Setagaya fuzoku school had been a jugyokenkyu
cycle in miniature: early planning based on the
curriculum and potential student response; the
observation of another teacher (first Matsuyama,
then each student teacher); teaching a public
lesson; and finally, a discussion of observed
events. Each public lesson posed a hypothesis, a
new idea about how to help children learn. And
each discussion offered a chance to determine
whether it had worked.
The typical postlesson discussion began at
school, around tables arranged in a U, a cup of tea
at each seat, and continued over beers at the local
izakaya. In addition to his own notes, each
observer had a copy of the teacher’s lesson plan.
The plan explained what the teacher intended to
do and why, describing the advantages of 12 – 7
instead of 13 – 6 to introduce subtraction with
regrouping. It also offered context. So far, all
students have mastered subtraction without
regrouping except Sayaka, the plan might say.
Beneath that detail would be a list of techniques
all the students had mastered—for example,
counting on fingers, using manipulatives, or
breaking up the numbers mentally.
In the discussion, the best comments were
microscopic, minute-by-minute recollections of
what had occurred, plus commentary. These
ranged from pragmatic tweaks—since the students
were struggling to represent their calculations
visually, why not arrange the tile blocks in groups
of ten rather than individual blocks?—to insights
that spilled outside the bounds of the lesson.
 After a lesson on finding the area of a rhombus,
for instance, an observer recalled a powerful
moment when one student had asked another why
she calculated the way she did. Usually, teachers
struggled to persuade students to talk not just to
the teacher, but to each other. They often had to
force the issue by having students stand at the
board and present their ideas to the class or by
asking one student to respond to another’s
confusion. But this kind of spontaneous discussion
—one student asking another about her thinking
—proved much harder to engender. What was it
about this class’s culture, the observer asked, that
had taught the children to communicate so well?
The group reviewed the evidence, searching for
occasions when the teacher said or did something
to encourage this kind of dialogue. In transitioning
the students from working on the problem
independently (“You”) to conferring with a
neighbor (“Y’all”), for instance, the teacher had
been very deliberate in her language. Instead of
describing the step of finding a partner or talking
to a friend—a common pattern, and one that could
feel forced—she’d focused on the exchange itself,
telling students to “look at each other’s papers.”
“Maybe,” she’d said, “you will find an idea you
never knew!”
Someone else pointed to the way the teacher
had begun the discussion part of the lesson
(“We”). Like other teachers, she had her students
record their ideas on pieces of paper that they then
tacked on the chalkboard for the whole class to
see. And like other teachers, she divided the
different students’ ideas into groups: triangle
method here, square method there, parallelogram
method here. But she added one unusual twist:
instead of grouping the ideas herself, she
delegated the task to the students, thereby forcing
the thinking work onto them too. In the rhombus
lesson, a spirited discussion had ensued as one
little boy picked up a piece of yellow chalk,
unprompted, and began dividing the board into
four different sections for each type of idea,
leading the others to scramble as they figured out
the section to which each idea belonged. By
creating a simple, predictable routine—after we
have our ideas, we arrange them on the board in
groups—the teacher had inculcated a spirit of
ownership.
Another routine cemented the culture. Each
time, after sharing an idea with the class, a student
asked the same question: “Who thinks the way I
am thinking?”
On its own, each of the routines might have felt
forced. Indeed, teachers observing the lesson
recalled other classrooms where similar exchanges
fell flat. But together, the routines formed a
powerful combination, getting students to ask
each other earnest questions—without having to
be told.
Other postlesson discussions focused more on
the subject matter itself, noting, for instance,
which part of the material the students
misunderstood and whether all children struggled
with that same difficulty. In a lesson about angles,
for instance, an observer commented about the
inherent challenge in seeing angles as not just
shapes but quantities—a more difficult stretch
than making the same mental step for area.
Problems, too, could come under scrutiny. The
same lesson on angles stemmed from a question
asking students to come up with as many angle
combinations as they could, given two triangles—
without using a protractor. But instead of leaving
it at that, a basic math problem, the teachers who
designed the lesson embroidered the question into
a story about an imaginary king who loved to wear
hats of all different angle sizes. (“One day,” the
story began, “there was a country without a
protractor.”) The story made for some fun
moments in class, as when one teacher playfully
anointed a boy king of the class, directing
questions to him. (“King, is this okay?” the
teacher asked, prompting another boy to ask, of no
one in particular, “How did he become the king?”)
But in the postlesson discussion, the observers
noted that mostly the story seemed to leave
students confused. Angles just don’t look like
hats. By the end of the lesson, the teacher had
spent so much time clarifying the bounds of the
question that the students hadn’t gotten to dig into
much math.
The power of jugyokenkyu, from the planning
process to the discussion afterward, lay in the fact
that no teacher worked alone. To solve the puzzles
that teaching posed, teachers needed the push and
pull of other people’s opinions.

Jugyokenkyu pervaded Japanese elementary

schools. But how directly each teacher participated
in it was up to the individual. After graduating
from Tokyo Gakugei University, Akihiko took two
steps to become a power user: he made a vow to
perform (not just help plan) one public lesson a
year, and he joined a volunteer study group of
math teachers in the area. The math group
operated under the sway of American documents
that kept coming out in Japanese translation.
Though he’d first read John Dewey and George
Polya in college, it was with this group that
Akihiko really delved into those Americans’
writings for the first time—and discovered just
how complicated it was to apply their ideas every
day in the classroom.
Take Polya, the mathematician whose problem-
solving manual How to Solve It became like a
bible to the group. Polya argued that the process of
solving a problem had four key steps. At first, the
Japanese teachers followed Polya’s recipe
faithfully, guiding their students to begin by first
understanding a problem, then planning a
solution, and only then attempting to implement
the plan. (The fourth step advised students to look
back on their work, checking for mistakes and
thinking about the solution’s implications.) But
when the teachers in Akihiko’s study group tried
using the steps with their students, nobody ever
followed step two: “devise a plan.” They always
jumped straight to step three: “carry out a
solution.” Again and again, the teachers tried to
get the students to follow the steps, without
success, until it emerged in their discussions that
perhaps they didn’t need step two at all. Three
steps were enough.
Akihiko’s teaching group formed at about the
same time that the National Council of Teachers
of Mathematics released its standards. As with
Polya, Japanese teachers took the NCTM
standards not as a recipe book, but as a guideline.
Instead of following the ideas step by step, they
thought carefully about the goals, tried out ways to
achieve them in their classrooms, and compared
notes on what worked.
Nobody expected progress to happen
immediately. Akihiko followed other teachers’
advice when choosing subjects to approach with
the single-problem format. Area lessons, for
instance, were a natural fit: students could play
around with shapes, experimenting with different
methods for finding the area of a parallelogram or
a triangle, and then, by comparing methods, they
could derive a formula that would work for any
such shape.
As Akihiko worked on the lessons, the rest of
the group looked at his plans and offered
feedback, guiding him through changes. He
especially treasured public lessons, held on special
days in the Japanese school calendar when
teachers were released from their regular teaching
loads to travel and only the students of teachers
giving a lesson stayed in school. With public
lessons, Akihiko learned as much from the ones
he taught as the ones he attended. Visiting a
school in Nagano one year, he watched a teacher
do something intriguing. One component of
teaching that Japanese teachers often discussed
was bansho, or “board writing”—the art of
writing on the chalkboard in a way that helps
students learn. Each teacher had her own style, but
over time, intricate conventions evolved. Usually,
a title went in the upper left-hand corner; the
problem of the day, right underneath. The writing
on the board then proceeded in columns: selected
students’ solution methods, then thoughts about
how to connect them, followed by a concluding
statement (a final formula, definition, or
observation). The key was to make the space a
visible representation of the lesson’s unfolding
ideas.
Carrying this out, of course, posed all kinds of
problems. Not running out of space was a big one.
If the teacher paced herself just right, she recorded
all the important ideas, right up to the conclusion,
without having to erase any that had come before.
But one too many notes could throw off the whole
balance. When students suggested lots of great
ideas, teachers needed strategies to keep them
from getting lost in the dense progression.
 The Nagano teacher had found a novel solution
to this bansho challenge. On the right-hand side of
the chalkboard, the class kept a collection of
magnets, each inscribed with a different child’s
name. When a new idea emerged, the teacher
wrote it out—and attached its author’s name
magnet above it.
The innovation served multiple purposes. On an
aesthetic level, it helped set off the students’
proposed solution methods from the other parts of
the board. It also made discussion smoother.
Talking about the area of triangles, it was easier to
refer to “Nori’s hypothesis” than it was to
constantly summarize its crux. Teachers already
used students’ names to mark ideas this way, like
Deborah’s fence posts, but labeling them with a
magnet made the process even more efficient.
Finally, appending an idea with a name magnet
rewarded students for sharing thoughts, equipping
teachers with a new weapon in their continual war
on shyness.
 Akihiko took up the idea, and soon he was
seeing other teachers do the same. By the time he
returned to Japan after his stay in Chicago, you
could hardly go to a classroom without seeing a
collection of name magnets dotting the right side
of the chalkboard.

The one-problem approach to teaching math took

hold just like the magnet idea. Akihiko wasn’t the
only teacher to observe and emulate Matsuyama.
As the number of teachers experimenting with the
approach expanded, so did the number who saw it
during a public lesson or just while walking down
the hall of their own school. Often, all it took was
one lesson to be persuaded that the approach was
worth trying. Many teachers could still remember
the exact lesson that had opened their eyes.
Not every attempt succeeded, of course. The
difference was that, in Japan, a teacher who
treated group work as merely an end in itself, or
tried so hard to engage the children with a fun
story problem that she distracted from the content,
was likely to hear about it from her colleagues.
And then, after learning what she needed to work
on, she didn’t have to come up with a solution all
by herself. She could observe other teachers’
classrooms with that problem in mind and learn
something.
Take a second-grade teacher’s lesson on bar
graphs. To get the students engaged, the teacher,
Mr. Hirayama, didn’t come up with a goofy fairy
tale. He simply designed a version of the
textbook’s suggested lesson—survey students
about a preference, then have them plot the results
on the chalkboard—around a topic he knew the
students were interested in. They’d been talking
about growing plants in the classroom, and the
teacher decided to use the bar graph lesson as step
one in their planning process.
Hirayama began not by announcing the lesson’s
mathematical purpose (today we will learn about
bar graphs), but by telling the children that today
they would decide which plants to grow. First, he
solicited ideas. What might they want to try?
When almost everyone had shouted out their
preference, Hirayama—a young, tall man with a
smiling, calm demeanor—wrote the final list
along the lower part of the chalkboard, left to
right: potatoes, carrots, okra, sweet potatoes,
tomatoes, cucumbers, green peppers, cosmos.
Then he announced the next step. The students
would vote, placing their magnetic name cards
next to the plant they wanted to grow. The
opportunity caused a minor ecstasy, with some
children making a choice quickly and defending it
and others mulling until the last minute,
engendering a small mutiny from cucumbers to
sweet potatoes right before the deadline.
Between shrieks, Hirayama paused to ask
questions. Which plant has the most votes? Which
has the least? How do you know? When all the
children had entered their final votes, he asked the
question that steered them closer to their secret
purpose. “Did you realize,” he said thoughtfully,
putting his hand on his chin, “you piled up your
names on top of each other?” Indeed, above each
plant name stood a little makeshift bar line made
of name magnets. Inspired by the discussion about
which plant had the most votes, they’d started to
sort them out. With the help of Hirayama—
conveniently more than twice their height—the
name magnets made columns stretching toward
the top of the board. Without being told the idea of
a bar graph, the second-graders had come up with
it intuitively.
Now Hirayama just needed them to notice what
they’d done and think about why they’d done it.
“You didn’t have to do that,” he said. “Why did
you do it that way?” The students threw out ideas,
and the teacher listened calmly, waiting for the
one that would take them closer to grasping the
purpose of arranging data in a bar graph. Finally,
a boy named Ano gave him the opportunity he
was looking for, not in the form of an answer, but
as a question. Why, Ano asked, did Hirayama
make some categories line up in a single column
when others had two columns per plant?
The question turned into a debate. “I don’t
really see the reason of making it two,” another
boy retorted. “Maybe in your head, Ano, you’re
thinking 2, 4, 6, 8?” He mimicked counting by
twos.
More comments followed. Using two columns
might help the second-graders reach the name
cards more easily, and it also helped them count
by twos. But what about the plants that received
an odd number of votes? Running out of time,
Hirayama abandoned his original plan. He’d
expected to have the students replace the name
cards with circles, solidifying the transition from
data collection to chart making. Instead, he let
them pursue their discussion about columns. He
didn’t regret the decision. Even if he had insisted
on moving to the circle technique, they probably
wouldn’t have grasped it; their attention was
somewhere else. He could always make the
transition in the next day’s math lesson.
For the moment, Hirayama concluded the Ano
debate with a poll: How many students thought
that arranging the votes in single columns made
the most sense, and how many agreed with Ano
about two? A consensus emerged. One column,
they decided, was the best way to help them see
which plant had received the most votes. A quick
experiment confirmed the hunch; arranging the
votes in single columns, they all sang out the
winner in unison. “Potatoes!”
Just as the second-graders learned more by
sharing their thoughts with each other, the fluid
exchange of ideas accelerated progress among
teachers too. The beauty of watching multiple
teachers at work was that you could see the many
different facets of a single practice. Sometimes the
different solutions built on each other.
Take the challenge of ending each lesson with a
neatly summarized main point, or matome—and
getting all the students to really consider it. One
matome innovation involved adding a new
minisegment to the very end of the lesson, in
which the teacher asked the students to scribble
down what they had learned that day. “Today
unlike the other days we talked about plants and
we compared the heights and I’m happy,” one girl
wrote in her notebook at the end of Hirayama’s
lesson. “Today we did graphs for the first time. I
didn’t know about it so it was fun,” a boy wrote.
Another group of teachers expanded on the
idea, not only asking students to write a summary
for themselves, but then asking certain students to
share with the group. Sharing had multiple
benefits. Students with excellent summaries got
recognition, and they also served as models,
giving others a chance to discreetly revise their
notes. And everyone got a few minutes to revisit,
record, and (the teacher hoped) remember the key
thing they’d just learned.
 A third group of teachers took a slightly
different approach, replacing summaries with a
competition to give the lesson a title. Usually, the
title to a lesson was written in the upper left-hand
corner of the chalkboard as soon as the students
sat down. In this approach, the teacher left that
part of the board blank. Then, at the end of the
class, she asked students for nominations about
what to fill in. Like summarizing, title writing
helped tie the lesson together. “Times 2 and
divided by 2 are brothers!” one little boy
suggested after a lesson on division, in which the
class had noted a pattern connecting products and
dividends. The teacher encouraged the idea, but it
was the more descriptive suggestion—“The
relationship between the answer and the number
to be divided”—that won a space at the top of the
blackboard (and in every child’s math notebook).
Even teachers who didn’t observe these
methods in classroom visits found the new ideas
entering their classrooms. Leading teachers in
Japan not only attracted crowds to their lessons;
they also took jobs with textbook publishers,
helping to write the texts they had to teach. And so
the math textbooks, too, started to take up the new
ideas, gradually centering each lesson around a
single problem. They refined the problems over
time as they tried them out in lesson studies,
finding out which ones were most productive.
Akihiko was the author of several textbooks; so
was Hirayama’s mentor, a colleague of Akihiko at
Tokyo Gakugei named Toshiakira Fujii.
Take subtraction with regrouping. The numbers
1 through 19 produce thirty-six different problems
that introduce the idea, from 11 – 2 to 18 – 9. But
over time, five of the six textbook companies in
Japan converged on the same problem: 13 – 9.
Other problems were likely to get students
discovering only one solution method. For
example, taking on a problem like 12 – 3, the
natural approach for most students was to take
away 2 and then 1 (the subtraction-subtraction
method). Very few would take 3 from 10 and then
add back 2 (the subtraction-addition method). But
Japanese teachers knew that students were best
served by understanding both methods. Knowing
two methods would come in handy when students
encountered new problems that worked better with
one or the other. And in general, seeing two paths
to a solution helped students understand just how
subtraction worked.
That was why 13 – 9 almost always came out
ahead. When tackling that problem, teachers
knew, students were equally likely to devise
subtraction-addition (break 13 into 10 and 3, and
then take 9 from 10 and add the remaining 1 and 3
to get 4) as they were to devise subtraction-
subtraction (take away 3 to get 10, and then
subtract the remaining 6 to get 4). Because both
approaches were likely to be tried, they could
count on the class to come up with the most
productive path to understanding.
The layout of the textbooks evolved too.
Traditionally, a new unit would begin with the
point of the lesson in bold letters at the top—how
to make proportional relationships, say. Units also
usually began on the left side of a two-page
spread, so that the next page, with the formula or
concept spelled out, was immediately visible to a
reader. But this format, of course, gave away what
should have been a delightful and important
discovery. So, the textbooks began opening units
on the right-hand page, leading off with broader
topics and a single problem, so as not to spoil the
ending—or, more important, to make sure
children understood it for themselves.
One could also get ideas from the essays
teachers published after a lesson study, describing
their plan, what had actually happened, and then
discussing what they had learned from it.
Magazines full of these essays lined the shelves of
local bookstores, offering everything from ways to
introduce a particular concept to transcripts of the
lectures on school culture given by elementary
school principals each month.
The education system was no utopia, of course.
Japanese teachers rolled their eyes at Ministry of
Education bureaucrats as much as teachers
anywhere else do. And just like in the United
States, official ministry-run professional
development sessions, usually held outside of
classrooms, could feel disconnected and pointless
—wastes of time to teachers for whom time was a
scarce resource. Indeed, one idealistic official
hoping to breathe new life into the sessions
traveled all the way to America, only to be told
about a Japanese practice called “lesson study.”
And while teaching evolved impressively inside
elementary schools, high school teaching changed
more slowly because the teachers were bogged
down by the pressure of preparing students for the
cutthroat college admissions contest, not to
mention the requirement of spending their after-
school hours running extracurricular activities.
Meanwhile, lower-income students received
more struggling teachers and fell behind their
peers in achievement, just like in the United
States. And citizens worried about falling behind
in the international achievement race, as other
countries inched ahead on global tests—some of
them, like Singapore, by deliberately adopting
Japan’s approach to jugyokenkyu; others, like
China, by using their own native jugyokenkyu-
style traditions (zuanyan jiaocai, or “studying
teaching materials intensively,” Chinese teachers
call it) or, like Finland, by creating them (“field
schools,” lab schools are called in Finland).
But Japanese education officials also found
ways to support the learning that teachers found
most valuable, writing research lessons into
school districts’ schedules and even inviting
leading teachers to help revise the national
curriculum every ten years. As a result, just like
the textbooks, the curriculum began to incorporate
the new ideas.
The whole process was not unlike a great
lesson. By trying out new ideas in real teaching
experiments, noting what happened, and refining
their craft in response—and (crucially) by doing
all of this together, and frequently, with colleagues
who were both fellow teachers and visiting
subject-matter experts—they learned much faster
than if they had tried to learn on their own.
Working alone, a teacher might excel or innovate,
or might not; the outcome depended mostly on the
individual. Working together increased every
single person’s odds of improving. Through
jugyokenkyu, teachers taught themselves how to
teach.

To James Stigler, the American running the

TIMSS study, jugyokenkyu jibed with what he’d
learned about the country’s culture generally.
Japanese companies had gleaned a similar concept
from William Edwards Deming: the idea of
continuous improvement. Organizations, Deming
argued, could improve only if they constantly
studied their practices, always looking for little
things they could do better. A dud in America,
Deming became a sensation in Japan. The idea
captured a commitment to craftsmanship that was
already at the heart of the country’s most prized
traditions, from the careful lifelong study of the
sushi chef, who spent decades mastering the
particular flip of the hand required to make a
perfect rice pillow, to the slow and steady
apprenticeships in kabuki theater, where students
spent decades mastering the special poses.
Akihiko’s colleague Toshiakira Fujii used the
analogy of his own pastime, kendo. A person
could spend a lifetime slowly advancing through
the kendo ranks. That “do” at the end of kendo,
Fujii pointed out, could be appended as well to
other careful crafts: sado, for the tea ceremony;
shodo, for Japanese calligraphy; karate-do, for the
martial art. Translated literally, it meant “the way”
or “road”—a long way to go. Think of it, Fujii
said, as “lifelong learning.” Considering this
tradition, Japanese teachers had done nothing
particularly innovative when they created
jugyokenkyu—except, perhaps, not calling it
jugyo-do (jugyo means lesson).
Stigler saw the same attitude when he first
pitched the Japanese government on the video
study, one of dozens of optional components of the
international test. As Stigler walked into the
conference room to make his case, a colleague
grabbed him with a warning. “He said, ‘Oh, this
is horrible. The Japanese have turned down every
option.’” But when Stigler made his pitch, the
official “looked up and said, ‘Yes.’ Just like that.
And everybody’s jaw dropped.” Later, when
Stigler asked the official why he’d said yes after
turning so many other components down, he gave
an answer no American official ever mentioned.
“He said, ‘because we want to watch the videos to
see if we can get any ideas to see how to teach
better’ . . . No American had ever brought that up
as a reason why they would do this study.”
Americans wanted answers, not improvement—a
report filled with bar graphs and tables, not new
teaching cases to study.
In 1999, when Stigler and the researcher James
Hiebert published their findings in a book called
The Teaching Gap, they made lesson study their
triumphant conclusion. American teaching, their
study had shown, was failing American children,
denying them deep opportunities to learn. But the
American approach to the problem largely missed
the point.
At first, the business and political elites who led
the US education reform movement had embraced
many different approaches to improving schools.
But by 1999, they had increasingly settled on just
one: standards. It was the kind of argument that fit
nicely into a PowerPoint checklist, a committee
hearing, or a new bill. Documents soon laid out
what students should know and be able to do in
each grade (like “identify a main idea” and
“distinguish between fact and opinion,” in fourth-
grade English; and “understand the concept of
rate” and “add and subtract fractions with unlike
denominators,” in sixth-grade math). In just a
year, the number of states with approved learning
goals on the books would grow to forty-eight, and
two years later, a diverse coalition including
business and labor groups would support the No
Child Left Behind law to make standards and
accompanying annual tests a requirement—along
with consequences when they weren’t met. The
country was moving toward the accountability
idea that Eric Hanushek had articulated thirty
years earlier.
Stigler and Hiebert supported this movement
too. “Without clear goals, we cannot succeed, for
we cannot know in which direction to move,” they
wrote in The Teaching Gap. Yet they suspected
that simply setting standards and consequences for
failure would not ensure that they were met. “It is
equally important to recognize that standards and
assessments, though necessary, are not enough.”
Reform’s “next frontier,” they wrote, was teaching
—the way students and teachers worked together
in school. “Standards set the course, and
assessments provide the benchmarks, but it is
teaching that must be improved to push us along
the path to success.”
The movement toward accountability ignored
this vast middle piece of education, the part
Hanushek had once called “the black box of the
production process.” This blind spot was the real
reason efforts to scale TKOT had failed. Instead of
incorporating the laboratory-style training process
that Magdalene Lampert and Deborah Ball had
devised for their students at MSU, the California
and NCTM reformers described the changes they
wanted without offering a plan to implement
them.
“We have this idea that if you discover
something quantitatively in a research study, and
then you tell everybody about it, that’ll improve
teaching,” Stigler says. “The truth is, with
teaching, 10 percent of it is the technology or the
idea or the innovation. Ninety percent of it is
figuring out how to actually make it work to
achieve our goals for students.”
American ideas might have taken the Japanese
10 percent of the way there, but Japanese
jugyokenkyu had done the rest. To change
teaching, Americans needed to learn as much from
the Japanese as the Japanese had learned from
them.

One year after The Teaching Gap came out,

Hyman Bass found himself standing at the back of
a classroom at the Setagaya Elementary School
with a group of American math teachers, watching
a lesson unfold. With its carefully plotted
beginning, middle, and end, the class reminded
Hy of great theater. Then there was the way the
teacher used manipulative tools, unlike anything
he’d seen in the United States—so deliberately,
with incredible precision. But he still felt some
distance from the Japanese teachers. Everyone else
in the delegation did too—except one person. “The
only American researcher who really connected
deeply with the Japanese was Deborah,” Hy says.
“She noticed things and asked questions of them
that were unlike what anybody else did.”
What Deborah Ball noticed most of all was
language. Translating what the teachers said after
the lesson, Deborah’s interpreter kept stumbling.
The Japanese teachers would say something, and
it wasn’t that the interpreter couldn’t hear them or
that they weren’t making sense. The problem was
that the words the Japanese teachers kept using
had no English equivalent; the language simply
didn’t exist.
It’s okay, Deborah told him, entranced. Just
translate literally. The neriage section of a lesson,
in which many different ideas yielded to a
consensus and a new academic concept, might not
make sense to the interpreter—“knead and rise”—
but it resonated with Deborah. There was a word,
bansho, to describe the art of writing clearly on
the chalkboard; another, kikanjunshi, to describe
the part of the lesson in which the teacher walks
between students’ desks, looking at their work to
determine which student should share and in what
order. There was a word to describe the process of
effectively using students’ ideas to achieve a
lesson’s goal and another for the category of
mistakes that, when shared with the whole class,
offer the richest opportunities to learn (neriage
and tsumazuki, respectively). There were key
questions for posing the problem of the day (shu-
hatsumon) and the practice of observing students
(mitori) and the lesson opener (donyu). The words
were another product of jugyokenkyu. To talk
about teaching and all its components, teachers
had invented new words to describe them.
There, in the middle of residential Tokyo, a city
she’d never visited and probably would never see
again, Deborah had the peculiar feeling of coming
home. “I was like in heaven,” she says. “It would
be as if, I don’t know, you really like good food,
and you’re always eating McDonald’s, and then
suddenly, you’re in this good restaurant.” Or, she
thought, as if after years painting alone, she had
finally found the artist’s colony.
On top of everything else, what Japan had was
language. Of course Americans struggled to
improve their teaching. When they tried to talk or
even think about it, they suffered a fundamental
handicap: they had no words.
 5
AN EDUCATIONAL START-UP

The man who invented an American language of

teaching never visited Japan, never attended ed
school, and, until recently, had never met Deborah
Ball.
Instead of a traditional ed school like Michigan
State, Doug Lemov came from the world of
educational entrepreneurs. A new class of
educators, the entrepreneurs emerged in the 1990s
just as the reforms at Michigan State and
California were winding down. Unlike Deborah
and her cohort, Doug and his friends were just as
likely to have degrees in business as in education.
Instead of epistemology, child psychology, and
philosophy, their obsessions were data-based
decision making, start-ups, and “disruption.” They
were more likely to know the name of Eric
Hanushek, the economist who invented the value-
added teacher evaluation model, than Judy Lanier.
They thought of themselves less as educators than
as activists, members of a movement: the
movement, some of them said. And they kept their
distance from Deborah’s world not only out of
ignorance; the separation resulted from conscious
—even righteous—design.
Their movement was born out of moral outrage.
Doug Lemov’s involvement was ignited in 1994,
during grad school at Indiana University. The son
of a lawyer and a journalist from Bethesda,
Maryland, the upscale suburb of Washington, DC,
he had supplemented his studies (in English) with
a side job tutoring members of the Indiana football
team. Growing up, Doug was small, painfully shy,
and a mediocre athlete. But toward the end of high
school, the growth spurt he’d prayed for finally
arrived, and in college, his six-foot-two frame won
him a spot on the varsity soccer team. College
athletes did not faze him, and he found that he
was good at helping them study. Then, one day,
the coaches presented him with a new pupil—a
nose tackle named Alphonso, who, they told
Doug, “needs more help than just study table.”
Doug met with Alphonso and suggested that he
start off by writing a brief autobiography
introducing himself. Alphonso was not unlike
Doug: sweet, gentlemanly, eager to please. But
when he sat down to write, he struggled. His
paragraph was virtually incomprehensible. Doug
couldn’t find a complete sentence in it. “More
help” had been an understatement. Alphonso was
practically illiterate—and he didn’t even know it.
First, Doug was indignant at the university.
Admiring the young man’s considerable athletic
talents, Indiana U, it seemed, had led him to think
he had the skills to get through college. Doug felt
this deceit was cruel and confronted the study-
table official who’d given him the assignment.
“I’m flattered that you think I’m the solution to
Alphonso,” he said. “But let me tell you, meeting
with me three or four times a week isn’t going to
solve the issue here. He writes on the fourth-grade
level.”
“Actually,” the staff member replied, “we tested
him, and he writes on the third-grade level.” Doug
bristled. Shouldn’t the coaches have considered
that before persuading the university to take him?
“That’s the interesting thing,” she said. “He’s not
a sponsored case.”
The true deception, she explained, was
perpetrated neither by the university nor the
football team, but by his high school in the Bronx,
which had promoted him year after year without
complaint. By the time he graduated, he had good
grades, flattering teacher recommendations, and
not a clue as to how much he did not know.
“Because he was not a troublemaker at a bad
school, no one wanted to shit on his dream,” Doug
says. “Nobody wanted to tell him, I’m not going
to pass you. They thought they were helping him
by passing him along every year. And they killed
his dream.”
In 1971, writing about the birth of modern
feminism in New York Magazine, the writer Jane
O’Reilly described experiences like Doug’s
realization about Alphonso as “clicks”—the
moments when an abstract social ill intersects
with the daily minutiae of life and becomes
personal:

In Houston, Texas, a friend of mine

stood and watched her husband step
over a pile of toys on the stairs, put there
to be carried up. “Why can’t you get this
stuff put away?” he mumbled. Click!
“You have two hands,” she said,
turning away.
Last summer I got a letter, from a
man who wrote: “I do not agree with
your last article, and I am canceling my
wife’s subscription.” The next day I got
a letter from his wife saying, “I am not
 cancelling my subscription.” Click!
...
In New York last fall, my neighbors
—named Jones—had a couple named
Smith over for dinner. Mr. Smith kept
telling his wife to get up and help Mrs.
Jones. Click! Click! Two women
radicalized at once.

They were clicks “of recognition,” O’Reilly wrote,

“the moment that brings a gleam to our eyes and
means the revolution has begun.”
Alphonso provided Doug’s click moment. After
encountering unequal educational outcomes
firsthand, he couldn’t get the injustice out of his
head. He began to think less and less about
graduate school and more and more about
Alphonso. The last time Doug saw him, Alphonso
was sitting in the computer lab, trying to write a
paper. “He couldn’t figure out how to scroll down
on the screen, and the letters in his title weren’t
capitalized,” Doug says. “I thought, he’s going to
fail out, he’s going to go back to the Bronx, and
he’s going to have no idea what he did wrong.”
What kind of country let children turn into
young men without teaching them how to read?
What kind of dysfunction led a public high school
to pass a student who could not write a complete
sentence? What had the civil rights movement and
its Brown v. Board of Education decision
accomplished if boys from the Bronx and their
future tutors from Bethesda still got completely
different educations?
Just as with feminism, Doug’s click didn’t
happen in a vacuum; all over the country, other
people were becoming, in O’Reilly’s words,
“clicking-things-into-place angry” after their own
encounters with the injustice of the American
public school system. There was Jay Altman, who
realized that his rural California high school had
shortchanged him, leaving him years behind most
of his classmates at Williams College. There was
John King, of East Flatbush, Brooklyn, who,
comparing himself to the friends he grew up with,
realized that he might never have graduated high
school, much less gone on to Harvard, had an
extraordinary New York City public school
teacher not set him on a different path. And there
was Wendy Kopp, who watched her Princeton
roommate from the South Bronx struggle to keep
up with her peers.
After meeting Alphonso, Doug had to call only
a few friends before he found one boiling with the
same frustration—a college classmate named
Stacey Boyd. She was living in Boston, where she
planned to foment an educational revolution. In
just a few weeks, Doug had decided to move to
Boston and join her. Together they would overturn
educational inequity. They just needed to figure
out how.

The approach they came up with reflected the

spirit of the decade—and, in particular, two
emerging theories that were then hardening into
conventional wisdom. The first was Erik
Hanushek’s accountability idea. The economist’s
articles about the problem of throwing money at
schools were eccentric, even radical, when they
first appeared in the 1970s. But by the 1990s,
Hanushek’s ideas had become a bipartisan truism
that stretched beyond education to all social
programs. The roots of ignorance, unemployment,
and other ills did not lie in a lack of government
support, but an overabundance of it. By throwing
money at poverty, the government had exacerbated
it, giving the poor new reasons to be complacent
rather than empowering them to change their
stations. “The problem they were trying to solve,”
Irving Kristol put it, “was the problem they were
creating.”
In 1994, the year Doug met Alphonso, the
Democratic president, Bill Clinton, vowed to “end
welfare as we know it.” Future antipoverty efforts,
even Democrats agreed, would have to make
support contingent on results. In other words,
programs would have to attend not only to
“inputs” like how much money the programs got
and how many people they served, but also to
results, or “outputs”—whether they showed
quantifiable improvements.
In education, the assumption was that schools’
dysfunction stemmed from an absence of
accountability. Education had been a cornerstone
of Lyndon Johnson’s war on poverty, receiving
billions of dollars in support for disadvantaged
students like Alphonso by way of a new fund
called Title I. (“Poverty has many roots,” Johnson
had explained, “but the taproot is ignorance.”) In
turn, between 1961 and 1991, real annual per-
pupil spending nearly tripled, rising from $2,835
to $7,933 in constant dollars. But during roughly
the same period, reading achievement as measured
by the National Assessment of Educational
Progress remained essentially flat. Just as welfare
programs had allegedly created a permanent
underclass by supporting the poor with no strings
attached, federal spending had fostered a school
system in which districts got increasingly large
grants whether or not they successfully educated
their students. The arrangement, according to the
new conventional wisdom, conditioned schools to
fail. They were like any monopoly; unchecked by
the free market, they underperformed. Why did
Alphonso’s high school let him fail? According to
the accountability advocates, the answer was
simple. The school let Alphonso fail because
failure carried no penalty.
The second influence on Doug and his friends
was a new national obsession with the power of
quantitative data. The metrics revolution elevated
a budding class of nerdy “quants” (people like the
statisticians hired by Oakland A’s manager Billy
Beane, as described in Michael Lewis’s influential
book Moneyball) above an old guard who
followed intuition and business as usual, never
guessing that within the piles of available
performance data lurked a more effective
approach. In education, the quants were people
like Hanushek, outsiders who analyzed the new
data pouring out of districts across the country and
discovered that schools had no idea how to
effectively spend their money.
These two influences—accountability and
quantitative metrics—led to sweeping reforms in
education, including the proliferation of standards,
stakes, and tests that James Stigler observed. They
came to take their purest manifestation in the
charter school, a new category of school financed
by the district, but under different management.
Stacey Boyd, Doug’s friend from college, wanted
to start one in Boston. Targeted at poor students
like Alphonso, living in heavily black and
Hispanic neighborhoods, charter schools followed
different rules. At traditional public schools,
funding was tied only to attendance; schools
stayed open as long as they had local children to
teach. At charter schools, support was to flow only
as long as student achievement met certain
benchmarks. At the Academy of the Pacific Rim,
or APR, the charter school founded by Stacey and
Doug, students had to meet the targets or the
school would shut down. Stacey took the promise
a step further, vowing that if any student failed to
pass the required tenth-grade state test, she would
send the funds APR would have spent on the
student to whatever school that student chose to
attend instead—essentially a money-back
guarantee.
Traditional public schools all reported to a
single bureaucratic school district that ruled their
operations with Byzantine obtuseness. Charter
schools, on the other hand, could be like lean,
efficient FedExes to the district’s Postal Service—
or, in the language of the decade, spirited dot-
coms battling the mighty Microsoft.* They
received government support, but they controlled
their own affairs.
 The best charters also obsessed over metrics. By
studying everything they did and analyzing the
results, they hoped to figure out what was really
working and what wasn’t, and then change their
habits accordingly. In a newly competitive
educational marketplace, they assumed that the
“winner” would be the school that pursued
effective innovation, even if the result was vastly
different from the way things had always been.

At APR, the first educational convention that

Stacey and Doug discarded was architecture. The
school occupied the second floor of the Most
Precious Blood parochial school, a building
constructed around a courtyard, like an inner-city
DoubleTree. The setup did not include an athletic
field; Most Precious Blood’s only outdoor space
was a parking lot. So athletics, including APR’s
daily Tai Chi lessons, took place on a landing.
Staff meetings, in keeping with the latest
management trends, were held standing up.
APR’s founders also rejected almost everything
associated with ed schools, including their ideas
about teaching. Many of them, Doug included,
hadn’t gone to ed school. But those who had gave
the rest an idea of what education professors
advocated, and as far as they could tell, it was
exactly the opposite of what their students needed.
The broad label for the ed school approach to
teaching, as they understood it, was progressive.
The goals of this progressive pedagogy were
laudable. In the hands of progressive teachers,
classrooms were supposed to be little
democracies, with children working with the
teachers to create the rules; structures were
supposed to be relatively loose, giving students a
chance to express themselves and pursue their
own interests; and instead of focusing on rote
memorization, teachers were supposed to plan
careful lessons guiding students through big
concepts and ideas.*
 But there was theory and then there was reality.
The teachers who had worked at traditional public
schools before coming to work with Stacey and
Doug told horror stories about the attempted
implementation of progressive pedagogy. Instead
of inspiring creative learning and self-expression,
the progressive ideas made chaotic urban schools
more disorderly and struggling students more
confused. Test scores were lower than ever. Scott
McCue, an early APR teacher who came to the
school after teaching at an alternative high school
in New York City, described spending twenty
minutes in a fifty-minute class period just getting
students to pay attention. Progressive pedagogy
sounded nice, but experience didn’t support it.
Anyway, the entrepreneurs had more pressing
concerns. Their students didn’t need democracy;
they needed the basics. At APR many of the
students were poor, black, and years behind their
more affluent peers. They needed to learn to read
and write and add and subtract. None of the loftier
goals—critical thinking, imagination, and
creativity—could happen without a grasp of these
fundamentals. How could a child study chemistry
without knowing how to multiply? Or create a
historical argument without being able to read?
APR students needed to learn something else
too—the skill that undergirded all academic study,
even simple number and letter fluency: discipline,
the art of paying attention, obeying instructions,
and following through. Disorder ruled in the
schools that Stacey and Doug sought to replace—
the failing inner-city public schools, where fights
broke out in the hallways, homework assignments
were roundly ignored, and noise levels tested the
limits of the human eardrum.
At APR, Doug and Stacey threw away the ideas
about democracy and open-ended projects in favor
of a pathological (some said authoritarian) focus
on behavior. A supporter of the school, Linda
Brown, noticed the germ of the habit when she
telephoned Stacey Boyd on the school’s inaugural
first day, a sweltering July morning. (Hoping to
pull struggling students ahead, the school
launched with a summer boot camp.) “I called in
the morning, and I said, ‘Stacey, this is the day
you’ve been waiting for!’” Brown says. “‘The
students have come, right?’ And she said, ‘Yes,
yes, they have, but we had to send a few home.’ I
said, ‘What are you talking about?’ She said,
‘They weren’t in uniform.’ ” The offending
children, Brown learned, had worn every piece of
the required wardrobe (khaki pants, standard-
issue polo shirts) except one: a belt. She pictured
students trudging miles home in 100-degree heat.
What kind of school was this, exactly?
But Stacey explained that what seemed
punishing, even cruel, in fact represented a radical
act of kindness. By being scrupulous about order,
starting with the tiniest symbols, they could build
a school where students obeyed more important
codes of conduct, paving the way for the
emotionally safe, academically challenging
learning experiences that would be truly
progressive. She cited the broken-windows theory
—the argument, devised a decade earlier by two
social scientists, James Q. Wilson and George L.
Kelling, that catastrophic urban chaos, cascading
from harmless drunks wandering the streets to
violent crime, could be undermined by eliminating
the tiniest signs of disorder. When a broken
window is left unrepaired, “all the rest of the
windows will soon be broken,” Wilson and
Kelling argued.

This is as true in nice neighborhoods

as in run-down ones. Window-breaking
does not necessarily occur on a large
scale because some areas are inhabited
by determined window-breakers
whereas others are populated by
window-lovers; rather, one unrepaired
broken window is a signal that no one
cares, and so breaking more windows
 costs nothing. (It has always been fun.)

Stacey had to make sure that students knew

from day one that, at APR, breaking windows—or
its school equivalent, not wearing a belt—cost
something. Brown was convinced. “I realized . . .
if you don’t get that culture right at the beginning,
you don’t get another chance,” she says.
The idea persuaded others too, less because of
the philosophy than because of the response it
engendered from the parents. Spencer Blasdale,
another early APR teacher, and one of the few
teachers there who had attended ed school, at
Harvard, recalls hearing Stacey announce the zero-
tolerance approach for the first time at a series of
informational meetings for parents the year before
APR opened. Every time, parents would push
back on the idea. And every time, Stacey would
hold her ground. “There’s one case that I
remember viscerally,” Spencer says. “A mom
said, you know, ‘My daughter sort of swears
sometimes. She’s 13 years old, she’s a teenager. If
she swore under breath or something, you’re really
saying that you would call me at home to come to
the school to get her?’” Stacey’s reply: “Yup,” she
said, as Spencer recalls. “I will call you, you may
not like it, and I think that’s a service that we are
doing.”
“And every other parent in the room—they
stood up and started clapping,” Spencer says. Of
all the pieces of APR’s pitch to parents—from the
plans to teach international finance in eleventh
grade, to the Mandarin classes for every student
(not to mention Tai Chi), to the glittering résumés
of the founders (Stacey had just finished Harvard
Business School; Doug came with his master’s in
English from Indiana; Spencer came from
Princeton)—it was discipline that captivated
parents most: a promise to keep their children
safe. Once again, the hard data of experience
trumped pretty dreams. And teachers saw
immediate payoff in their results. Teaching at
discipline-obsessed APR, Scott McCue now spent
fifty-four of the fifty-five class minutes on content.
The staff at APR knew their zealous approach
made a number of teachers uncomfortable. A
student forced to apologize to the entire class for
his misbehavior—wasn’t that one step away from
a dunce cap? But this discomfort with
consequences was the conventional wisdom the
entrepreneurs wanted to overturn—another
romantic notion that experience did not support.
Doug called it the “Hug ’Em to Harvard” principle
—the idea that what underprivileged students
needed most was warmth and kindness. The
Hug-’Em-to-Harvards were “like, ‘Oh, when
those kids meet us, and they see how much we
love them, then there won’t be any behavior
problems!’ And you’re like, how dare you think
that when they meet you they won’t show all the
manifestations of the organizational dysfunction
and scars of poverty,” Doug says. “We were not
going to be those fools.” More than hugs, the
faculty felt, the APR students needed limits.
Other charter schools were coming to the same
conclusion. At Roxbury Prep, another Boston
charter school, John King—the Harvard graduate
from East Flatbush—debated with his cofounder
Evan Rudall about how far to take their own
radical orderliness. Like APR they had a strict
dress code, but what about passing periods? John
thought the kids should be allowed to talk as they
moved from class to class. Evan disagreed. “Evan
said, ‘I’m telling you, if we don’t do silent
hallways, transitions are going to take forever, and
it’s going to be a total disaster for classes,’” John
says.
To settle the debate, they decided to call Evan’s
wife—John’s idea, because he knew she would
sympathize with his softer approach. She did, and
students were allowed to talk between classes. It
took only two days of school for John to abandon
ship. As Evan had predicted, moving between
classes not only took up more time than they had
allotted; it also created opportunities for
misbehavior that spilled into classrooms, wasting
valuable instructional time. By the time Jay
Altman—who was in the process of opening his
own school in New Orleans—visited Roxbury
Prep a few years later, John had become the silent
hallway czar. “He was radical about it,” Jay says.
“Every time, about two minutes before the bell
would go off before classes, all year long, he or
someone would get on the speakers and say,
‘Teachers, the bell is about to ring. Please get in
the hallway.’ ” According to John, the PA
message came on only as the bell was ringing,
and the request was for teachers to “join us in the
hallway,” but the result was the same. Without
missing a beat, the teachers without a class that
period appeared in the halls, ready to keep watch.
Soon enough, the entrepreneurs’ approach
—“countercultural,” McCue called it—had
solidified into a philosophy. It was the deliberate
inverse of “Hug ’Em to Harvard”: no excuses.*
The second piece of the educational orthodoxy that
Doug and the other no-excuses entrepreneurs
rejected was isolation.
American teachers might work in school
buildings with dozens of colleagues. But, as
education scholars have noted, they operated
essentially solo, like “lone rangers,” as Deborah
Ball’s Spartan Village colleague Mindy Emerson
put it. Sociologist Dan Lortie, who argued that
American classrooms still operated like a one-
room schoolhouse, described the approach as the
“single cell of instruction” model. Despite
growing faculties with greater numbers of
specialists, individual teachers rarely interacted.
Some education professors used the metaphor of
the “egg-crate school,” which carefully separated
teachers, as if to keep them from touching.
The entrepreneurs hadn’t read much of the
academic theory, but they were zealous about their
work nonetheless. Inexperience seemed to
motivate them. Most of the teachers had just a few
years of teaching under their belts (Doug, for his
part, had taught English for three years at a private
day school in Princeton, New Jersey, before grad
school), yet they were responsible for running the
school. Every day they faced a stream of difficult
questions. “You know, the Tai Chi instructor
didn’t show up,” says Spencer Blasdale. Or, “like,
okay, what’s the first report card going to look
like? How are we going to do our first school
dance? What are we going to do because this Tai
Chi teacher’s been absent three times?”
For Doug, the questions multiplied at the end of
the first year, when Stacey decided to leave. “She
was like, ‘Guess what? You’re principal!’” Doug
says. “‘It’s your school. Go and get ’em!’” He was
twenty-seven years old. One day, not long after
that, a group of auditors approached him about an
accounting problem he hadn’t even known
existed, which, if unaddressed, would jeopardize
the school’s future. “Every day there was
something like that,” Doug says. “We were so
incredibly vulnerable. In part, because [we] were a
tiny little organization with no infrastructure. In
part, because there was no experience and no track
record and no one knew anything to go back on.
And none of us had any experience. And, in part,
because none of the rules were written. It was
terrifying.” Running a charter school, he thought,
was the ultimate crucible. Every mistake you
made, you made in public, for all your staff and
students to see.
Another group of people might have turned
inward. But in their response, Doug and the rest of
the APR teachers were distinctly more
“Japanese.” With no experience to fall back on,
they tackled the problems together. Instead of
regressing to the egg-crate tradition, the faculty at
APR opened their doors—and learned from one
another.
Indeed, the Academy of the Pacific Rim was
“pacific” by definition. Its two founding board
members had read James Stigler and Harold
Stevenson’s early research on Asian schools, and
they had attended Asian schools themselves. They
handed the charter over to Stacey, who had taught
in Japan, with a mandate to blend the best of East
and West.
As a result, Doug found himself leading study
sessions on joint readings of Teaching and
Learning in Japan and essays on Japanese
management techniques. The TIMSS video study
that introduced Stigler to lesson study had not yet
been published, but the cultural approach
underlying jugyokenkyu was embodied by the
concept of kaizen—then popular among the
Harvard Business School crowd—which
described the “continuous improvement” of
Toyota assembly lines. Imagining the educational
equivalent of an efficient and responsive assembly
line, Doug and his colleagues did not quite
reinvent lesson study. But they came close,
holding standing meetings on the minute details of
the homework system and devising schedules to
enable teachers to make regular visits to each
other’s classrooms. (In APR’s case, it helped that
so many of the teachers were young, unmarried,
and childless; days started at 6:00 in the morning
and didn’t end until 7:00, 8:00, sometimes 9:00 at
night.)
The school’s culture of constant collaborative
learning also stemmed from Doug himself.
Though he outgrew his childhood smallness,
Doug never shed the shyness. He was
compulsively humble, painfully self-deprecating,
and prolific on the subject of his own faults. In
college, after a soccer game, he would spend the
next two weeks fixating on the mistakes he’d
made, even if his team won. When he did stumble
on an idea that he knew was good, he always
deferred responsibility to a colleague. “You can
assume, because it was really smart, that it was
not my idea,” he’d say. Whenever he gave advice,
he delivered it with a hint of a question mark. He
was so self-effacing, so obsessed with his own
shortcomings, that being around him could
sometimes be physically uncomfortable.
Colleagues couldn’t talk to Doug for long without
his casually declaring a failure at some deeply
personal task, like raising his children or
supporting his wife.
Doug thought there were two kinds of people in
the world: the virtuosos, who could run or jump or
write or act without thinking much about it; and
the strivers, who studied the naturals’ every move,
perfecting their skills in an attempt to reach the
higher level. Doug was sure he was a striver, a
person for whom brilliance required imitation.
Doug responded to failure at APR by doing
what he had always done: he looked for people
who excelled at their work and studied what they
did. His staff was his first and best source of
ideas. Walking through the classrooms at APR, he
learned how to become a better teacher.
Sometimes the lessons were obvious, as with
Molly Wood, whose peppy persona made her
teaching techniques—an intriguing mix of
extreme structure and warm emotional connection
—easy to identify. Other times they were more
mysterious, as with quiet, understated Kate
Glendenning, who put her students into a kind of
trance as they considered Shakespeare, never
rushing into answering a student’s question (the
way Doug always did), but instead pausing a few
seconds before crafting the most effective reply.
Doug studied them all, even bringing a video
camera to Kate’s class one day to enable further
review. He also looked beyond APR, to the new
movement of no-excuses charter school leaders in
Boston and around the country. Together, the
teachers and principals talked on the phone,
huddled over coffee or at a party, and visited each
other’s schools, discussing everything from how to
deal with troublesome parents to how to manage a
school budget. “Any difficult decision, I would
pick up the phone and call John [King] or Mike
[Goldstein, of the Match charter schools] or Evan
[Rudall] or Brett [Peiser] and say, what would you
do in this situation, or what have you done in this
situation? And we always would send teachers,
groups of teachers—go spend a day at Roxbury
Prep or [later] Boston Prep,” says Spencer
Blasdale, who became Doug’s assistant principal
when Stacey left.
Little comments often made the biggest
difference. One day, watching Doug teach, a
colleague gave him a simple suggestion. Trying to
save time, Doug had given directions while he
moved around the room, passing out papers. But
the students weren’t paying full attention. “When
you want them to follow your directions,” the
colleague suggested, “stand still. If you’re
walking around passing out papers, it looks like
the directions are no more important than all of the
other things you’re doing. Show that your
directions matter. Stand still. They’ll respond.”
Doug tried it, and he was floored. “I could see the
difference right away.”
The comments multiplied, and over time,
everyone’s confidence grew. Teaching presented a
million problems, but by working together,
watching each other and being watched, he and
his colleagues could craft solutions. “The great
thing about being in the crucible is that if you see
something successful, you know right away,”
Doug says. “It’s not working, and it’s going to be
a long hour! Or, ‘Oh my God. It’s working!’ It
was just a ton of very visible—like a laboratory—
a ton of very visible feedback coming back to you.
You’re getting this, or you suck.”
Not every member of the no-excuses movement
was a congenital learner like Doug. But the ones
who were—John King and Evan Rudall at
Roxbury Prep; Jay Altman in New Orleans; Brett
Peiser, a former teacher in Brooklyn who had
started another Boston charter school; David
Levin and Mike Feinberg, early Teach For
America corps members who went on to found the
KIPP charter schools in Houston and New York
City; and (the rare early female founder) Dacia
Toll, a Yale Law School graduate who had created
Amistad Academy in New Haven, Connecticut,
which later grew into the Achievement First
network—became a community. Together they
ensured that each school not only learned to solve
its own problems, but also benefited from
solutions invented by its compatriots.
After visiting the fifth-grade classroom of a
teacher named Julie Jackson at the North Star
Academy in Newark, New Jersey, Jay Altman
decided to devote an entire professional
development session back in New Orleans to
reproducing her lesson. “Ms. Jackson’s Mythical
Math Class,” he called the workshop. “I had
everyone go through the experience of her class
and used all these techniques I’d learned from the
visit. One of them was oral warm-up in math, like
language class.” He brought his staff to their feet
and peppered the room with questions, just as
Julie Jackson had done with her fifth-graders.
“Like, ‘Who can point out a pair of parallel lines
in this room? Okay, three hundred times thirty
plus ten times five equals what?’ And then, if a
kid answered, say, ‘Okay, who agrees with that
answer? Who disagrees? Why or why not?’” In
one short exercise, she’d managed to get her
students thinking, talking, practicing fluency,
using math vocabulary, and giving herself lots of
insights into what they did and didn’t understand.
Altman wanted all that for his students too.
Another trip, organized by Linda Brown, the
Boston charter school advocate who had been at
first perplexed and then impressed by APR’s belt
policy, offered a tour of innovative schools in New
York City. “We rented a bus that held 44 people,
and we took the Massachusetts charter schools
that had opened,” Brown says. It was 1996, the
second year of charter schools in the state; only
twenty-two schools had opened so far. “We filled
the spaces in about 25 minutes.” The trips
eventually became a formal training program for
no-excuses principals: the Building Excellent
Schools fellowship.
Reading Po Bronson’s account of life in Silicon
Valley during the dot-com boom (The Nudist on
the Late Shift), one APR teacher, Chi Tschang,
saw himself and his colleagues in the descriptions
of the scrappy Hotmail programmers who slept
under their desks. Later, after Chi introduced
himself to Bronson by e-mail, the author visited
and wrote about APR. An “educational ‘start-
up,’ ” Bronson called it.

As no-excuses charter schools proliferated, their

students prospered. All the students in APR’s first
graduating class passed the state exams, just as
Stacey had promised (or their money back!), and
all went on to four-year colleges. At Roxbury
Prep, a middle school, students consistently led
the state, fancy Boston suburbs included, in math
and science proficiency. Jay Altman’s New
Orleans Charter Middle School became the
highest-performing nonselective middle school in
the city. North Star, KIPP, and Amistad followed
the same pattern.
Observers interpreted the success in the same
accountability framework that had generated
charter schools in the first place. The no-excuses
schools must have done better because they were
more accountable to results—and therefore, in the
language of the day, “results-driven.” Free of
bureaucratic red tape (such as the rules about
teaching hours and substitutes that eventually
doomed Jessie Fry’s efforts at Spartan Village),
and finally given the incentive to succeed, they
produced the outputs demanded of them. “Had we
tried to invent North Star at the federal
government,” joked then-governor George W.
Bush on a visit to the school in 2000, “they’d still
be in committee hearings.”
 Doug Lemov believed in accountability too.
Indeed, he believed in it so much that, three years
after APR opened, he decided to leave for business
school at Harvard, where he hoped to learn skills
to improve school accountability. While many
charter schools served children well, others, Doug
knew, stayed in business even though they posted
less sparkling results—the opposite of charter
laws’ intent. At business school, he thought, he
could figure out how to make accountability work
better for all schools, not just successful charters.
Eventually, Doug put the idea into practice at a
new dream job, managing the accountability
systems for charter schools across New York
State. Later, he went off to start a company of his
own, building diagnostic tests to help schools
meet their goals. At Harvard he’d become
especially bullish about data. Treated carefully, he
learned, data could both paint an accurate picture
of which companies (or schools) were performing
and which weren’t—and give them the tools to get
better. The diagnostic tests were an example.
Schools that Doug worked with could use them to
track their students’ progress toward state
standards throughout the year, recording each
child’s advancement in order to spot lapses: Why
does Kayla understand two-dimensional figures
and time, but not money? Why is Destiny doing
well with congruence but not the calendar?
Scrutinizing data could help teachers make
decisions. Ahzheona, with her 50 percent scores in
a certain category, probably needed a day of
tutoring; Jasmine, who scored below 50 percent
across the board, needed tutoring every day; and
Kendra and Amirah, with the class’s highest
marks across all the standards, were ready for a
new challenge.
Doug’s work at both jobs earned him praise.
New York’s charter school standards were
heralded as the best in the country, and the
diagnostic testing company, called School
Performance, won him the acclaim of the US
Department of Education, which asked him to
speak about the approach at an education summit
in 2006.
By the time he gave the presentation, though, he
was already beginning to see the limits of
accountability. The year before, he’d worked with
a school in Syracuse that presented a conundrum.
Walking through the hallways and meeting with
teachers and administrators, Doug found a school
drenched in just the kind of great expectations he
had practiced at APR. The school used Doug’s
diagnostic tests, and the principal was so obsessed
with goal setting that she had hung a banner in the
entrance listing her top three goals for everyone to
see:

1. Increase parental involvement to 100%

2. Intentional practice for goal-setting
3. Increase achievement in reading, math

Doug could see that everyone in the building

shared the principal’s goals. The teachers were not
lazy monopolists, enjoying tenure, raking in
attendance checks from the district, and promoting
kids who hadn’t earned it. Instead, they sat down
on the floor, holding books right in front of the
children’s faces, practically begging them to
succeed.
But as he spent more time in its classrooms,
Doug saw that the school’s high hopes, however
staunchly held, failed to materialize in practice.
This was not a Hollywood-style urban jungle, the
kind of brutal failure Doug had seen where
students called their teachers motherfuckers and
instructors made basic factual mistakes. (At one
school, a teacher had solemnly lectured about the
“farmers” of the Constitution.) Here, the failure
was much more insidious and also, Doug
suspected, much more common.
Students obeyed basic instructions, and
teachers’ lesson plans had beginnings, middles,
and ends. But it was as if the two sides had
reached a truce: the students would create minimal
chaos, provided the teacher wouldn’t demand too
much of their concentration. It was a lose-lose
compromise. Class discussions dragged
grudgingly forward. The same three students
always raised their hands. And the same ones
launched subtle protests against serious learning.
In one class that Doug observed, the teacher spent
several minutes debating a student about why he
didn’t have a pencil. Another divided her students
into two groups to practice multiplication together,
only to watch them turn to the more interesting
work of chatting. A single quiet student soldiered
on with the problems, alone. The teacher looked
the other way, and Doug couldn’t watch the rest.
He walked out the door.
Teaching did not have to feel that way, like
suffocating slowly. And high expectations had to
be more than a poster hanging from the ceiling or
a law signed in Washington, DC. To guarantee
that real learning happened, something had to
change in the actual classroom.
The problem did not seem to be a lack of
models. Following Doug’s old strategy of
watching better teachers at work, the school had
already taken a group field trip to one of the crown
jewels of the no-excuses world: KIPP Academy in
New York City. Created by David Levin and Mike
Feinberg, two early Teach For America corps
members, KIPP was a perfect model of both the
zero-tolerance discipline approach and the
sermonizing school-as-pep-talk culture.
Yet when Doug asked the Syracuse teachers
about the trip, he found that the visit had not
proved instructive. The teachers had seen plenty of
things—the arrangement of the reading rugs, the
colors of the uniforms. But like the visitors who
watched Magdalene Lampert and Deborah Ball
teach at Spartan Village, they had not seen the
things they needed to learn. “I just remember
thinking, ‘Holy shit. That’s what you took away?’
The things they took away were so random, and if
you ranked the most important things about a
high-performing school from 1 to 100, they had
seen number 63, number 84, and number 47. As
opposed to numbers 1, 2, and 3.”
But what were 1, 2, and 3? Doug always told
client schools that his data reports were only the
beginning. “This tool is only as good as what
people do with it,” he told the audience at the
national summit. “Empower them to take real
action.” But what actions should teachers take?
Doug understood the Syracuse teachers’
struggle, going to KIPP and not knowing what to
focus on. He could think of plenty of great
teachers he’d seen since he started working in
urban schools. But describing the things that made
them great was like trying to describe a dream. He
could explain how their teaching made him feel:
good or bad, pained or giddy. But he could not
explain exactly what happened or why or how to
make the bad moments better.
He thought about soccer, the sport he’d played
through college and beyond. If his teammates
wanted him to do better, they didn’t just say,
“Improve.” or “Be more like Beckham!” They
broke that “it” factor down, telling him to “mark
tighter” or “close the space.” Maybe the reason he
struggled to talk and even to think about teaching
was that the right words didn’t exist—or at least,
they hadn’t been invented. Not yet anyway.

* Some of the charter schools lived up to this

promise, but many did not. Multiple studies of
charter school performance have shown that the
schools often perform just as poorly as the district-
run schools they seek to outdo. And across the
country, charter schools have been the victim of
the same inefficiency and corruption challenges
that plague neighborhood public schools.

* When similar characterizations were used to

describe Deborah Ball and Magdalene Lampert’s
teaching—what other Michigan State educators
named TKOT—the two women resisted them as
false dichotomies. In good teaching, they
maintained, structure was as important as
freedom, fluency as important as concepts. But the
teachers at APR had not heard of Deborah or
Magdalene. Deborah and Magdalene also both
rightly questioned what “progressive pedagogy”
actually referred to. While some teachers took on
the label proudly to describe a “child-centered”
approach, there was no coherent school of
pedagogical thought that could be called
“progressive.” Historically, progressive education
described a political movement advocating certain
educational goals, not a pedagogical approach to
achieving them.

* As with TKOT (the phrase describing the

kind of teaching practiced by Deborah Ball and
Magdalene Lampert), I borrow the “no excuses”
label from some members of the group of people
I’ve called the entrepreneurial education
movement to describe their pedagogical approach.
“No excuses” was popularized by Samuel Casey
Carter in his book, No Excuses: Lessons from 21
High-Performing High-Poverty Schools
(Washington, DC: Heritage Foundation, 2000).
As with TKOT, the “no excuses” descriptor is not
embraced by all the entrepreneurial educators I
write about. Notably, as I explain further in
Chapter 7, Doug Lemov dislikes the term and
does not use it to describe his own work.
 6
LEMOV’S TAXONOMY

Driving home from Syracuse with his colleague

Karen Cichon, a former Catholic-school teacher
who now worked with him at School Performance,
Doug Lemov couldn’t stop thinking about the
soccer metaphor. As a teacher, he had always been
an evangelist for the power of clear language to
help students understand exactly what you wanted
them to do. He often crusaded against what he
called “the fundamental ambiguity of ‘shh,’ ” one
of the most widely used teacher phrases in the
American school and also, tellingly, one of the
least specific. “Are you asking the kids not to talk,
or are you asking the kids to talk more quietly?”
he would ask teachers. The same lesson applied to
teachers. Learning how to teach, they needed
specific instructions as much as their students did.
 But where could the words come from? Doug
thought of a recent visit he’d made to a classroom
at Roxbury Prep with one of the school’s
administrators, Josh Phillips. At a certain point,
Josh pulled Doug into the hall. “Did you see
that?” he said. “The teacher told the kids to put
their hands down.” It had been a counterintuitive
move. Presented with eager hands, most teachers
would naturally reward their enthusiasm by letting
the children speak. But this teacher decided the
questions had become redundant, and she moved
on. “That’s how we do it here,” Josh told Doug.
A decade of work had come down to “how we
do it here,” a way of doing school so detailed and
intricate it even included an opinion on when to
stop taking students’ questions during a lesson.
Learning together, the no-excuses leaders had
taken a blank canvas and turned it into a system of
thought and practice. What Doug needed was a
way to communicate that vision, to foreground all
the little things that made up the no-excuses way
so that any visitor would be sure to see them. He
needed better words.
As Karen drove them down the highway back
to Albany, Doug explained his idea from the
passenger seat. They needed to create a “common
vocabulary” to describe the elements of good
teaching. He wanted the Syracuse principal to be
able to sit in the back of a classroom, take notes,
and, just as the bell rang, pass the teacher a Post-
it: “Nice job closing the space. Next time, mark
tighter.” Or whatever.
Whatever.
Sitting at the wheel of her Jeep, Karen thought
it over, rewinding through her memory of
everything they’d just seen at the school.
Suddenly, she had one. It was a small thing she’d
seen a few teachers do that always drove her crazy
—maybe because, when she taught, she’d often
succumbed to the same temptation. Frustrated
with bad behavior, teachers often fixated on it,
complaining so much about what students were
doing wrong that they forgot to explain how to
behave right.
Suppose a boy named Daniel tossed a pencil up
and down until, whoops! the pencil parachuted
over to the other side of the room. The sensible
teacher would want Daniel to quit playing with
that stupid pencil. Then she would want him to
put it in the little groove on his desk and not touch
it again. Then she would want the class to return
immediately to the matter at hand. Given these
goals, the sensible teacher might say something
calm like, “Daniel, put that pencil in the pencil
holder and look at me.”
But the frustrated teacher is not usually
sensible. Watching that pencil sail across the
room, and maybe imagining it spiking
Lawrenesha in the eyeball or skewering Dante the
goldfish, the words flying out of the frustrated
teacher’s mouth, more likely, would be something
like, “Daniel! I’m going to glue that pencil to your
fingers!” Not only would these words fail to
describe the correct behavior; they would draw
unnecessary attention to the misbehavior. Any
student who hadn’t noticed that flying pencil
would know all about it now, and the power in the
room would shift from the diligent students to the
dunce.
More examples emerged. “David,” Doug said,
taking on a teacher voice, “I asked you to sit up,
and I still don’t see several of you sitting up.
Pause.” He went back to his regular voice. “Okay,
so what [I] just did was make it explicit that the
kids didn’t do what [I] asked them to do, and there
wasn’t a consequence for it. You almost couldn’t
do anything less productive in your classroom.”
If a principal caught a teacher making this
mistake, Doug and Karen thought, he should be
able to send her a one-phrase reminder. Maybe
they could call it “What to Do,” to represent the
goal of responding to misbehavior by pointing out
exactly what to do instead, rather than giving
attention to the failure. “John,” the teacher might
say, “that’s the third time you’re out of your seat.”
Okay, the principal could write, but What to Do?
Or, “Andrea, why do you have to go to the
bathroom so much? Why didn’t you go at
lunchtime?” Another What to Do.
Karen and Doug thought about problems
beyond behavior too. At the time, everyone had
“high expectations.” There wasn’t a teacher to be
found who would admit to having low
expectations. Yet in Syracuse, and in other schools
across upstate New York, they’d seen a thousand
small ways that teachers unwittingly showcased
that soft bigotry. Like, when a teacher asked a
question and all hands in the room went up but
three, and the teacher was happy because that was
pretty good. But she never followed up with those
three, and the ones who got the question right—
she never asked them to take a step further and try
to solve a slightly harder question.
Or what about when a student answered a
question with almost the right answer, but not
quite, and that was good enough? Or when a
teacher asked a specific student—say, Benjamin
—for an answer, and the student shrugged? “I-o-
no,” Benjamin would say, looking somewhere
else. And because it seemed impossible to pull
anything more out of him, especially under the
time constraints and considering the fact that he
was one of only thirty students in the packed,
maybe also hot room, the teacher would move on.
And Benjamin would get away without saying
another word.
Karen and Doug came up with more categories
to go along with What to Do. “100 Percent”
would remind the teacher to make sure that every
single student in the room was engaged, following
along, and understood. “Right Is Right” would
encourage a teacher to insist on getting the precise
correct answer from the student, not a close-
enough one. “Stretch It” would demand that
teachers press students who easily provide the
correct answer, challenging them to take the
problem a step further.
Doug, who’d been writing their ideas furiously
in his notebook, looked up at the dashboard.
MILES TO EMPTY, the light read: 0. They’d
been so engrossed in creating their new language,
they hadn’t realized they’d run out of gas.

A week later, Doug and Karen returned to the

road. This time their destination was an
elementary school in Brooklyn. Taking notes in
the backs of classrooms, they once again tried to
think of constructive feedback, concrete “real
action” that they could tell the teachers to help
them improve. Except this time, they actually had
some words. It was like finally visiting the
optometrist after a lifetime of nearsightedness:
Fuzzy lines suddenly clicked into focus revealing
clear, discernible letters. The trees had leaves.
In the first classroom a teacher was running an
exercise in proofreading. To practice symbols like
delete, insert, and new paragraph, students walked
one by one up to an overhead projector and filled
in the correct signs. One student came up and
announced that a word needed to be removed. But
she could not remember the name of the symbol,
that loopy thing. “It’s okay,” the teacher said.
“You can say loopy thing. I’ll know what you
mean.”
“Right Is Right,” Karen wrote.
Later the teacher moved on to a vocabulary
lesson. She gave the students a word from their
list and then asked them to use the word in a
sentence. The first word was enjoy. A student
said, “I enjoy my weekend.”
“Can you add on that?” the teacher said.
“I enjoy my weekend by going to my cousin’s
house,” the student said.
“Well,” the teacher urged, “can you describe
that boy [his cousin]?”
Karen and Doug took notes. The teacher was
clearly trying to draw the student out, to push him
to describe the cousin in a way that would
demonstrate that the student understood the
meaning of enjoy. “Stretch It,” Karen wrote in her
notebook; this time, she’d found a good example.
The next word was poison. “One day when my
uncle came,” a student offered, “I made a poison.”
The teacher nodded. “Okay,” she said. “Now, next
word.”
Karen wrote this down in her notebook too—as
a bad example of “Stretch It.” Without any context
clues explaining why the student had made poison
when his uncle came, he hadn’t demonstrated that
he understood the word’s meaning. Not only that;
in the process, he’d gotten the whole class
thinking about his uncle, rather than the meaning
of the word poison—and the teacher had done
nothing to steer back their focus.
Later, Karen took account of the lesson in her
notebook: “Think of all of the things going on
there,” she wrote. “What are the important words
to learn? In the sentence they’re using, is that the
best way to decide: is that the right word to use or
not? If a sentence comes up that takes them off
task—well, now there’s 18 kids that really want to
hear about the uncle making poison, as opposed to
deciding what is the word, what does it mean, on
we go. There’s a whole lot that you could analyze
just based on what was a quick homework
review.”
The simple vocabulary that Doug and Karen
were developing—only a handful of terms so far,
representing a handful of techniques—helped
them say so much more.

By 2010, Doug had traveled to dozens more

schools, this time not to solve problems, but to
find solutions, more techniques to name. He
always began a school visit by asking the
principal for a report on the teachers who helped
students the most. Then he went into their
classrooms, bringing a videographer with him—
the first videographer was basically a wedding
photographer, whom Doug had met through a
friend—so that afterward, he could study the tape
more carefully, rewinding the key parts like an
NFL coach reviewing the opposing team’s plays.
Soon, he’d brought the wedding guy on full-time.
For Doug, too, the project had grown from a side
interest to a nearly full-time job.
A few years earlier, Doug had left his data
business to start two new charter schools—one in
Rochester and another one in Troy, New York,
just outside Albany. The schools were part of the
new Uncommon Schools charter network, a
collaboration that brought Doug together with
some of his old friends: John King and Evan
Rudall of Roxbury Prep in Boston; Norman
Atkins of North Star in Newark, New Jersey; Brett
Peiser of Boston Collegiate, who was now
opening schools under the Uncommon banner in
Brooklyn. Uncommon had institutionalized the
group’s informal community. Now, rather than
just calling each other up when they had a
question, Doug, John, Evan, Norman, and Brett
held regular “managing director” conference calls
to chart strategy. And instead of sharing ideas and
curricula and worksheets and tests haphazardly,
they shared them by design.
The taxonomy project emerged in the same way
—a weird side project of Doug’s that gradually
became something bigger. One reason Doug and
his cohorts built Uncommon was to help all their
schools expand. After achieving isolated
successes, they were under pressure from funders,
parents, and themselves to create more high-
performing charter schools. But growth had
introduced a million new problems; chief among
them was talent. They had all been in the practice
of finding great teachers the same way corporate
head hunters find skilled employees: they scouted
out the best talent at other schools and recruited
those teachers away by offering a better work
environment and, when necessary, a nicer salary.
 Now, as the group opened more schools, and
other charters did the same (Achievement First,
KIPP, and other no-excuses schools were
becoming franchised networks at exactly the same
time), the competition for great teachers was
growing fierce. Uncommon decided to change its
approach. Rather than buy talent, the network
would try to build it. The “build it/buy it”
epiphany gave Doug the same illuminating sense
he had felt in Syracuse when he realized the limits
of data and accountability. Attracting the best
teachers through incentives got the charters only
so far. To scale up, they needed a way of helping
any average teacher get better. They needed a kind
of playbook, an understanding of what made the
best teachers great so that they could help the
merely ordinary get even better. They needed
Doug’s taxonomy.
That was what teachers had started calling his
vocabulary project: the Doug Lemov taxonomy, an
organized breakdown of all the little details that
helped great teachers excel. Soon, the taxonomy
had become part of Doug’s official job. Instead of
expanding his upstate New York branch of
Uncommon at the same pace as the others, he
would focus half of his time on building his
taxonomy and recruiting a small team of video
“analysts” to help him do it. Charter school
supporters on the outside might conclude that the
key driver was accountability, but inside
Uncommon, Doug and the others knew that
training was just as important.
As Doug built the taxonomy, he took advantage
of the growing Uncommon network, which gave
him a wide new pool of teachers to study. Soon
his list of techniques had expanded to forty-nine.
Some, like “Right Is Right,” covered academic
standards. But the techniques that got the most
attention covered discipline and attention, the vital
core of no-excuses culture.
At its heart, the no-excuses idea actually
represented an end, not the means. Doug and
Karen had called one of the first techniques “100
Percent.” It described a goal—a classroom where
100 percent of students meet 100 percent of the
expectations 100 percent of the time—rather than
the means of achieving it. How did the best
teachers get all those eyes on them, or all those
brains thinking through the problem?
To the untrained eye, the key to obedience
seemed to be sheer personality. Observers often
spoke of how great teachers worked “magic,” as if
they turned a hat into a rabbit. In a way, they had.
Where previously had sat defiant, misbehaving
children, voilà!: eager, attentive, curious scholars.
These virtuoso teachers had “it,” whatever “it”
was. But when Doug studied these teachers more
closely, he saw that the students had not actually
transformed—not completely anyway. Presented
with the most charismatic, engaging teachers,
some students would still deviate—refusing to pay
attention, for instance—or they would forget to
follow a rule.
 The difference lay in what happened just after a
student strayed. What did the teacher do then? The
more Doug studied his videotape, the more
intricate the answer seemed to be. Just as students
seemed to find a million different ways to
misbehave or make a mistake, there was also no
one way for a teacher to respond, no secret
formula. But there did seem to be a series of
principles that all the “it” teachers followed,
consciously or not, and the principles seemed to
lead toward certain important moves, a sliding
scale of responses to select and deploy in each
particular moment.
One of the first teachers to impart this lesson to
Doug was Colleen Driggs, a Teach For America
alumna who taught at his school in Rochester.
Colleen had “it,” that enchanting quality that
transformed children into students. But one day,
watching a videotape of her teaching a vocabulary
lesson, Doug noticed something he hadn’t seen
before. Just beneath the surface of her calm, cool
teaching—she was especially good at getting
students to talk not only to her, but to each other—
lay a series of practically invisible hand gestures.
Midsentence, she would point two fingers at her
eyes, bat down an imaginary fly with two quick
swipes, or, with no explanation, briefly clasp her
hands before her, as if in prayer. During one five-
minute video clip, a discussion of the meaning of
the word scarce, Doug counted fifteen such
gestures, one every twenty seconds.
Watching again, he could see that the gestures
clearly meant something, both to Colleen and to
her students. Two fingers to her eyes—that meant
“track the speaker,” code for paying attention to
the person talking, usually another student. The fly
swat, applied to a raised hand, meant “I’m not
taking questions right now.” And the prayer sign
reminded students to get into the attentive position
that no-excuses schools called SLANT or STAR, a
back-straight pose tied off with primly clasped
hands.*
 Colleen had created the gestures, she explained,
so that she could subtly correct students’
misbehavior without interrupting the flow of her
lesson. At the beginning of each year, she taught
the three gestures explicitly. For the first few
weeks, every time she used one, she would say its
name too. But pretty soon, she could pull off
performances like the one Doug had witnessed.
When she said, “The food in Mrs. Driggs’
refrigerator is scarce because the inconsiderate
guests came over and ate almost all of it,” the only
person who noticed that she’d also gripped her
hands in prayer just between “scarce” and
“because” was the student who had lapsed from
SLANT. The practically invisible corrections
explained her “magical” command of the
classroom. By nipping interruptions in the bud,
she kept everyone in the room on task.
The practice became one of the core principles
of the “100 Percent” technique. What did “it”
teachers do when confronted by student
misbehavior? They used “the least invasive form
of intervention,” Doug wrote in the taxonomy.
He found another example in a video of another
Rochester Prep teacher, named Patrick Pastore.
Asking a class of sixth-graders for attention,
Patrick counted down how many pairs of eyes he
needed, until the number was just one. The last
student was a boy named Dwayne. But Patrick
had never said the boy’s name. Instead, he’d just
said, “We need one more set of eyes.” The move
echoed Colleen Driggs’s technique. Presented
with a child refusing to follow directions, Patrick
corrected him—but in a way that was almost
invisible. The alternative, of course, would have
been to call his name. Dwayne, Patrick could have
said, hands on his hips. We’re all waiting. But
imagine, Doug thought, what would have
happened then.
“Everyone’s looking at Dwayne, and then
Dwayne has a choice,” Doug explained to a group
of teachers later, giving a workshop on the
taxonomy in Boston. “Am I going to have
everyone see me bend and do what Patrick asked
me to do or am I going to [act] out, even at great
cost to myself, to save my honor?” A power
struggle would ensue, and, in all likelihood, the
situation would escalate, with Dwayne resisting
harder, forcing Patrick to push back equally hard.
They might spend two minutes just resolving the
fight. Or they might never resolve it.
In the second it took to notice that Dwayne
wasn’t paying attention, Patrick seemed to have
made all these calculations and decided to try his
hardest to avoid a showdown. Using the least
invasive form of intervention lightened a teacher’s
job, but it also lightened the student’s.
Compliance came at a much lower cost.
Too often, Doug knew, teachers made another
choice, and the consequences multiplied. “The
death spiral,” he called it. “Let’s say I’m teaching,
and Anne is slouching,” Doug said at the
workshop in Boston, motioning to a teacher
named Anne. “I could stop and say, ‘Just a
minute, class. Anne, I really need you to sit up.’ ”
He switched perspectives. When he made that
move with Anne, what was likely to go through
the minds of the rest of the class? “The kids who
were least engaged are least likely to get back on
task with me,” he said, answering his own
question. “I stop my lesson to correct one student,
and when I correct that student, three more kids
get off task, and then I have to run over here, and I
have to correct another student, and then three
more kids over here get off task. And then I never
catch up. I never win. It’s the death spiral.”
But what about the cases when it simply wasn’t
possible to correct a student without naming him?
Colleen Driggs could point to her eyes because the
student who had failed to watch the speaker was
watching her. Same with the student who lapsed
from SLANT and the ones with their hands up at
the wrong time. But what if the child was staring
into space? What if the hand gesture or the
anonymous hint (“We need one more set of eyes”)
didn’t work?
Doug grouped possible responses into six
accelerating options, each one slightly more
invasive than the last. Just one step above
“nonverbal intervention,” a nearly invisible hand
gesture like the ones Colleen Driggs used, was
“positive group correction.” By positive Doug
meant constructive—describing the desired
behavior, rather than the problem. “We’re
following along in our books,” a teacher could
say, posing the statement like self-evident
narration, even if it also contained a hint of
aspiration, serving to remind the boy in the back
that he shouldn’t be looking out the window. To
make the point slightly more clear, a teacher could
shift to second person: “You should be following
along.” Level three, “anonymous individual
correction,” was Patrick’s move. “We need two
people,” a teacher could say. Often, “it” teachers
even fudged numbers, optimistically declaring
they needed two when really the number was four
or five, a purposeful misstatement designed to
create the illusion that the order under request was
in fact happening now. Act as if it is already so,
and before you know it, it will be.
Level four approximated anonymity as much as
possible without actually preserving it. “Private
individual correction,” Doug called it, after
watching several “it” teachers casually place
themselves next to a misbehaving child (oh, did I
end up here? what a coincidence!), kneel down,
and whisper a reminder. The best executions
failed to distract anyone except the student who
was not paying attention in the first place. The
teacher did this by somehow making an action
that was both purposeful and antagonistic appear
exactly the opposite, a friendly accident.
One ingenious Rochester Prep teacher, Jaimie
Brillante, began with a diversion: she gave the
whole class a quick, silent task: “Copy it down,
please.” Then, with the students’ faces pointed to
their papers, she walked casually in the direction
of a girl on the right who hadn’t been paying
attention. But instead of marching straight over to
her, she meandered, picking up two tissues from
the classroom Kleenex box, dropping them
warmly on the desk of a girl in her path, and
peering curiously over the shoulder of another girl.
When she took a few more steps over to the one
who was her real destination all along, nobody
noticed, and they didn’t eavesdrop on her stage-
whispered correction either.
The whole move required strategic finesse,
especially considering that it was laced within a
lesson that Jaimie never stopped teaching. It also
depended on the ordinariness of Jaimie’s
happening to weave from one corner of the
classroom to another. She could walk around like
that without attracting attention because she
walked around all the time, not just when
someone misbehaved.
Doug labeled the next intervention, level five,
“lightning-quick public correction.” This, too, was
best conducted surgically. “Andrew, I need you
with me, just like Jeremy and Anne and David.
Now we’re looking sharp!” Doug modeled at the
workshop. “So I corrected Andrew publicly,” he
explained, “but I did a couple of things. One, I
instantly diverted the gaze from him to someone
else or something else, and, when possible, that
something else is much more positive. So if I said,
‘Andrew, I need you with me,’ then you’re all
going to divert your gazes to Andrew, and we’re
in that situation where I have to win, it’s all
public, and then I can’t afford to lose.” Instead, he
let Andrew take the stage for half a second before
quickly moving on to Jeremy, Anne, and David
and the idea that “we” (read: everyone, even
Andrew) were now “looking sharp.”
The sixth and final level was “consequence,”
the most visible response. But even a consequence
did not need to look the way one might picture it,
a tense exercise of power that grabbed everyone’s
attention. (How many times had a teacher handed
out a detention only to find the room taken over by
a chorus of “ooooh!” solving one problem but
creating another?) He referred the Boston teachers
back to a video he’d shown earlier, a cheerful
sequence in which a kindergarten teacher, George
Davis of the Leadership Prep school in Brooklyn,
declared that “one of our friends is not ready yet”
and asked his students to try composing
themselves in the SLANT position one more time.
Making the students get to SLANT again, Doug
pointed out, was a consequence—physical work
doled out because of a small behavioral lapse. “It”
teachers did sometimes resort to detention or that
easiest of Hail Marys—“Dean’s office, now!”—
but they usually reserved these for the most severe
cases, and even then, they kept their cool, working
to keep interruptions to a minimum.
The six levels of behavioral intervention
represented just one subprinciple of one technique
among Doug’s ultimate list of forty-nine, the
“Taxonomy of Effective Teaching Practices.” In
addition, there were Technique No. 43, “Positive
Framing” (“Narrate the world you want your
students to see even while you are relentlessly
improving it”), with its six rules (including one,
“Assume the Best,” describing how the best
teachers acted as if students misbehaved out of
ignorance, rather than willful defiance, unless
proven otherwise); Technique No. 26, “Everybody
Writes” (in which a teacher follows up a question
by asking every student in the class to write down
her thoughts on paper); and Technique No. 44,
“Precise Praise” (reflecting Doug’s conclusion
that successful teachers distinguish praise, which
includes a positive judgment, from
acknowledgment, which merely describes an
expectation met, and offering two other “rules of
thumb” for giving good praise, including the
suggestion to “praise loud; fix soft” and to ensure
that all praise is genuine—not a means to another
end, like pointing out one student’s success in
order to chide another on her failure).

But did the taxonomy capture the magic of

teaching as well as the science? Doug’s
experience with Bob Zimmerli suggested it could.
Bob was a teacher Doug had first met in 2005,
when he interviewed for a position on the
founding team at Rochester Prep by teaching a
sample lesson on place value. “It was like seeing
the truth,” Doug says. “Bob just walked in with a
pencil, and it was like a bolt of lightning came
down.” Doug knew that the students, on loan from
a school he worked with, were not the most
enthusiastic bunch. But in Bob’s hands, they
turned into new people, individuals who took an
interest not just in school but in math. Doug
immediately asked him to teach a second sample.
He had just started building the taxonomy, and he
needed to capture Bob on videotape. “I knew that
if you could bottle what he’d done, you would
have something incredible.”
Five years later, the tape of the second sample
lesson had become a classic in the taxonomy
canon. The video opens at the start of class, with
Bob standing in front of another group of students
he has never met. As class begins, the children—
fifth-graders, all of them black, mostly boys—are
looking everywhere but at Bob, who stands at the
board. One is caught playing with a pair of
headphones; another pages slowly through a giant
three-ring binder. “Okay, guys,” Bob says, “before
I get started today, here’s what I need from you.”
He’s dressed in a neat suit and tie, with the thin,
athletic build of a cyclist. “I need that piece of
paper turned over and a pencil out.” Almost no
one is following the directions, but Bob persists.
“So if there’s anything else on your desk right
now, please put that inside your desk.” He
motions a quick underhand pitch, demonstrating
what he wants the students to do. “Just like you’re
doing, thank you very much,” he says, pointing to
a student in the front who has put her papers
away. Another desk emerges neat; Bob zeroes in.
“Thank you, sir.” “I appreciate it,” he says,
pointing to another student. By the time he points
to one last student—“Nice . . . nice”—the
headphones are gone, the binder has clicked shut,
and everyone is paying attention.
Bob Zimmerli might have been the perfect case
of the unexplainable “it” teacher. A preacher on
the side, he possessed a charm that seemed
downright ethereal. But watching the video in the
context of the taxonomy research, Doug saw the
lesson with new eyes. “Imagine,” Doug told the
teachers in Boston, after showing the clip, “if his
first direction had been ‘please get your things out
for class.’ ” Instead, he’d deployed a perfect
version of one of the first techniques Doug and
Karen Cichon had named: “What to Do,” which
had now become Technique No. 37, an extension
of the principle behind “Assume the Best.” More
often than defiance, Doug had noticed that
misunderstanding lay behind students’ failure to
follow directions. “It” teachers got students to do
something by being brutally specific about exactly
what they wanted. Bob had also deployed rule
number four of the “Positive Framing” technique:
“Build momentum, and narrate the positive.”
Instead of focusing on the binder or the
headphones, he’d pointed out the students who did
listen. “It’s this positive wave,” Doug said. “You
can almost see it going across the classroom from
right to left.”
He played the clip one more time, directing the
group’s attention to the boy with the binder. When
the video starts, his head is down as he pages
slowly through his papers. Ten seconds in, he
looks to his left, where another boy has the paper
and pencil and is staring at the teacher. For the
first time, he looks up at Bob and stops paging.
“He’s like, ‘OK, what’s this?’ ” Doug narrated.
“ ‘I guess I’m going to go with it.’” Thirty
seconds later, his binder is closed, and he’s
pushing it inside his desk, underhand—just like
Bob showed him.
Not only did the taxonomy explain Bob’s
success; Bob himself said he learned something
from it. His testimonial was one of dozens. Doug
had been sharing his ideas since the beginning,
typing the taxonomy into increasingly dense Word
documents and e-mailing them to colleagues
across the country with video files to match. Every
time he shared the list with a new person, he
learned something new, and he built the new ideas
back into the document.
The effects accelerated as he started collecting
more video. When he and the new team of data
analysts hired to help sift through the growing
volume of classroom tape went through the new
videos, they not only found new ideas; they also
found evidence that the original techniques were
spreading. Having watched Bob Zimmerli
“narrate the positive,” other teachers tried the
same thing themselves. And often they improved
the techniques in the process. Sometimes the new
footage even outshined the originals.
An early star, a teacher at the all-boys Brighter
Choices charter school in Albany named Darryl
Williams, had shown Doug the first case he’d ever
seen of Technique No. 39, “Do It Again,” in
which teachers deploy the most minor form of a
consequence, teaching students to perform simple
routines by asking them to repeat them until they
can do them well. But as the DVD clip of Darryl
made its way around the country, more teachers
picked up on the idea, and soon their
interpretations took the idea to new heights.
The new elaborations made Darryl’s execution
look commonplace, even weak—his delivery too
slow, the routine too insipid, his students
needlessly morose. The new videos showed how
to keep the power of the exercise without
sacrificing time or pep. “Saying, ‘Oooh, let’s line
up again and prove why we’re the best reading
group in the school,’ ” Doug wrote in the
taxonomy, “is often better [than] saying ‘Class,
that was very sloppy. We’re going to do it again
until we get it exactly right,’ even if the purpose is
to Do It Again until you get it exactly right.” “Do
It Again” also seemed to work just as well when
teachers called for a video game–style redo right
in the middle of a routine, rather than waiting until
the end. Indeed, interrupting at the first sign of a
slip was less dismal and took up less time. The
taxonomy became a kind of recursive process.
Doug and the analysts sent a technique out into
the world, and then the world sent it back: the
same idea, refined.
Perhaps the best evidence of the taxonomy’s
success was Doug himself. He’d considered
himself a weak teacher since his first job, teaching
English in Princeton, New Jersey, where his
lovingly crafted lessons always seemed to fizzle,
leaving him to count the remaining minutes. Now,
teaching fellow teachers, it was like he had left his
old persona back in Princeton and wrapped
himself in new skin.
He called on his students cold (“Cold Call,”
Technique No. 22), without making anyone feel
pressured, creating the feeling of a natural,
spontaneous conversation while still carefully
keeping track of time and steering the learning in
one purposeful direction. When he posed a
particularly difficult question (“Number one, what
does [this teacher] add to our understanding of
nonverbal intervention? And then two, can you
find evidence of the other principles of 100
Percent?”), he made sure to give each participant
time to write down her thoughts on a carefully
prepared worksheet. And he built in jokes, getting
the whole room to laugh just when the proximity
of lunch might have stolen attention from the
collective enterprise. The result was the truest
measure of the master teacher. Sitting in a Lemov
taxonomy workshop, a participant experienced
that most satisfying buzz, the revelatory aha!
feeling that comes from thinking deeply—the
unmistakable pleasure of learning.

The taxonomy was not jugyokenkyu, exactly, but it

was a twenty-first-century American hybrid.
Instead of devising a language organically, as an
accidental by-product of postlesson discussions,
Doug and his team built the new words
deliberately, during structured “cutting log”
meetings where analysts sat around a conference
table, watching videos together and making notes.
And instead of building time into the school
calendar for city- or even statewide lesson
observations, they sent around video files and an
accompanying Word document (and later, DVDs
and a book).
Yet, in more important ways, the taxonomy and
the crucible that wrought it were just Japanese
lesson study in a slightly different form. They
produced the same effects. For one, American
teachers began to develop followings. Bob
Zimmerli, Colleen Driggs, Darryl Williams—they
became the American equivalents of Takeshi
Matsuyama, Akihiko Takahashi, and Toshiakira
Fujii. If the collaborative development of the
taxonomy was the entrepreneurs’ version of lesson
study, then the no-excuses schools were their
laboratories, the equivalent of the fuzoku schools
where Akihiko had learned and taught. And their
ranks were growing. According to Linda Brown,
the Boston charter school supporter who went on
to train charter school principals, by the time
Doug wrote the taxonomy, the number of no-
excuses schools had grown from about fifteen to
more than a hundred.*
The biggest difference between the American
and Japanese teaching laboratories was not how
they studied teaching, but what kind of teaching
they described. Both Doug and Akihiko
disavowed the usefulness of “shh,” but for
different reasons. Whereas Akihiko thought that
children needed structured opportunities to talk in
order to learn, for Doug, learning first required the
foundational ability to be quiet and listen. He just
didn’t think “shh”—with its fundamental
ambiguity (should students stop talking or just
talk more quietly?)—was a good way to get them
there. And so, whereas in Japan school was
interrupted by a chaotic burst of screams as
children took their hourly break, at no-excuses
schools students walked between lessons in a
determined (and not necessarily unhappy) silence.
Indeed, had James Stigler and his TIMSS
colleagues taken their video cameras into the no-
excuses movement’s classrooms, they would have
found teaching that looked not Japanese, but
classically American. Doug and his colleagues
had advanced far beyond the chaos of the average
urban school, creating schools where studying was
possible and achievement was valued—no small
feat. They had even eliminated the tyranny of the
unannounced PA interruption. But in many other
respects, their classrooms looked no different from
any other in the United States. Like their fellow
Americans, the no-excuses teachers used the “I,
We, You” structure (I explain, we try an example,
and then you practice) for most if not all of their
lessons, asked questions designed mainly to
generate simple answers rather than to “explain
how or why,” and devoted most of their students’
work time to practice, rather than the equally
common Japanese activity, “invent/think.” It was
no wonder that they valued quiet, “eyes on me”–
style attention so highly; for them, as for so many
American teachers, attention held the keys to
learning.
Ironically, this orthodoxy followed, in part, from
a move that had originally seemed countercultural:
the entrepreneurs’ rejection of ed schools. They
had spurned the schools for understandable—
indeed, data-based—reasons. Many ed school
professors’ ideas about teaching—abstract
advocacy for classrooms where order was less
important than creativity and students’ voices
could always be heard—really did not work in
practice. The sad truth was underlined in study
after study showing that ed schools failed to help
their teachers teach well.
Yet by disregarding ed schools, Doug and his
colleagues also unwittingly distanced themselves
from the ed school professors who were producing
innovative ideas—people like Magdalene Lampert
and Deborah Ball and their colleagues doing
similar work in history, English, and science.
(People who, incidentally, had made the same
observation about the tendency of lovely teaching
ideas, like the new California math framework, to
wilt in practice.) As a result, the American
jugyokenkyu that was the Lemov taxonomy had
all the features of its Japanese counterpart except
one: American ideas.
While Akihiko and his colleagues combed
through the standards of the National Council of
Teachers of Mathematics, devoured descriptions
of Magdalene Lampert’s lessons, and closely
studied videos of Deborah Ball’s teaching—all the
way from Tokyo—Doug and the no-excuses
educators unknowingly ignored their own
country’s best teaching innovations. As a result,
when Doug built the vocabulary that Deborah
longed for, he built it to describe a sort of teaching
that she didn’t do. Inspired by Japan, Doug
created the structure that the United States had
never had—a system focused on helping teachers
learn—but with none of the ideas (originally
American) that the Japanese filled it with.
For any other group, that strange state of affairs
might have continued for years. But if anything
matched the strength of the entrepreneurs’ allergy
to the educational status quo, it was their itch to
improve. If Doug’s generation had built the start-
ups to disrupt the old monopoly, they had also
built a machine designed to take new ideas and
improve them—even if that meant attacking the
core of the no-excuses approach.
 * Although the wording varies slightly at each
school, in general, SLANT stands for “Sit up,
Listen, Ask questions, Nod, and Track the speaker
with your eyes.” STAR stands for “Sit up, Track
the speaker, Ask and answer questions, and
Respect those around you.”

* That was a tiny fraction of the number of

charter schools in the United States—just over six
thousand by 2013. (National Association of Public
Charter Schools,
http://dashboard.publiccharters.org/dashboard/scho
 7
THE DISCIPLINE OF DISCIPLINE

When Rousseau Mieze joined the education

reform movement, he was a true believer. The son
of Haitian immigrants—his mother cleaned
houses, his father drove a cab—he knew firsthand
how the right school could change your life. With
his parents always working, it was his teachers
who first fed him sushi, his teachers who showed
him how to fly through two hours talking about
history and philosophy and humans’ place in the
universe, and his teachers who helped him return
a proper handshake without laughing. Teachers
acquainted Rousseau with the best tool he had: his
intellect. And then his intellect took him to
Williams College, the most academically
stimulating place he’d ever been.
He wanted to give other kids the same gift, and
he counted himself lucky that when he started
college, in 2004, an ecosystem of schools already
existed to do just that. Instead of interning with
professors or catching up on sleep or whatever it
was that other people did during breaks at
Williams, Rousseau spent his time apprenticed to
the movement. One summer he worked in the
dean’s office at a Boston charter school. The next
he went to Rochester Prep and shadowed some of
the Lemov taxonomy’s biggest stars—Patrick
Pastore, Colleen Driggs, Bob Zimmerli. To
Rousseau they were heroes in the flesh.
Rousseau would defend charter schools to
anyone, and by that time, he sometimes had to. By
the mid-2000s, the no-excuses movement’s
growth had generated a considerable amount of
attention and then, in turn, a backlash. Preparing
for his first job, at a charter school in Rhode
Island, Rousseau found himself confronted by
another first-year teacher who wondered out loud
whether the “no excuses” style amounted to
“brainwashing.” All those single-file lines and
mandatory chants seemed like prison, some said.
The teacher wondered if they squashed
independent thought. “Nah, man,” Rousseau told
him, running through the key arguments, which
boiled down to: Look at everything the students
learn!
And yet, by his second year teaching, Rousseau
found himself wondering what, exactly, the
students were learning. That year, he was working
at a charter school in Harlem where, on the
surface, the teachers seemed to be doing
everything right. The Lemov taxonomy was
omnipresent—teachers quoted from it like the
Bible. But something about the school made
Rousseau feel uncomfortable. It had to do with the
way teachers talked to students. They commanded
them to follow rules, but they didn’t offer them
any rationale. Sometimes they even seemed to
mock them. The kids, in turn, were great—funny,
smart, sweet, like all the kids he’d ever met. But
too often, they marched into class either defiant or
sullen. Lifeless. Not happy. “We weren’t treating
kids like they were people,” he says.
This was not what the taxonomy advised.
“While you should expect students to do
something when you ask them to, it’s not really
about you in the end,” Doug Lemov wrote in a
part of the taxonomy he called “purpose, not
power.” “Command obedience not because you
can or because it feels good but because it serves
your students.” Yet somehow, the teachers seemed
to be divining exactly the opposite lesson. If the
point of no excuses was to teach discipline—the
habits required to live a happy, productive life—
then why did these children keep acting out? The
teachers had only the best intentions. So did Doug
Lemov. But what were all these taxonomy
techniques actually teaching?
Sometime during that year, another educator
sent Rousseau a chilling report about a KIPP
charter school in Fresno, California. In a sixty-
three-page investigation, the Fresno Unified
School District described dozens of cases of what
it called “inappropriate” student discipline, most
of it meted out by the school’s principal, a young
man in his thirties named Chi Tschang. The
accusations were extreme—stories of punishing
disobedient children by putting them outside in
the cold, locking an entire class in a two-stall
bathroom, and putting a trash can on a student’s
head—and Chi and his teachers (as well as some
of their students) argued that many were also false
or misleading, given the school’s ongoing feud
with the school district. Even so, the overall
picture was shocking.
Many dismissed the school and its principal as
bad apples—an isolated case of good intentions
gone wrong. That was the position that KIPP
implied, declining to make a public statement
backing Chi after he resigned. But others hinted at
a broader takeaway message. One educator who
was friendly with Chi and Rousseau—a KIPP
principal in Newark, New Jersey, named Drew
Martin—wrote out his thoughts in a weekly staff
memo. The memo urged Drew’s teachers to read
the Fresno report and reflect on a single question,
highlighted in a gray box: What does good
discipline look like?
Chi Tschang may have been extreme, but he
had drawn on a specific tradition, and the tradition
deserved scrutiny too. The question Drew Martin
wanted teachers to consider was not whether Chi
had taken the idea of no excuses too far, but what
the no-excuses approach taught in the first place.
For all the infelicities described in the report,
Drew knew that Chi was more typical than most
people were comfortable acknowledging.
Rousseau knew this perhaps better than anyone.
As one of the first graduates of the Academy of
the Pacific Rim charter school, in 2004, discipline
had turned his life around. Chi Tschang had been
his history teacher at APR, and the person who
had had the biggest influence on both of them was
another young principal, named Doug Lemov.

Back then, students at the Academy of the Pacific

Rim could agree about one thing: they absolutely
hated school. “Nothing about school was fun,”
Rousseau says. Instead, APR was a parade of
rules that dragged everyone down—“like a
whipping and ball and chain,” one student told a
researcher at the time. Every chair had to be
pushed perfectly in, every shirt tucked, every
instruction followed, every tooth unsucked and
eyeball unrolled. Either that, or you’d find yourself
sitting in the dean’s office with Mr. Lemov,
explaining how you planned to act differently next
time. When students did come to love APR—and
many of them did, fiercely—their affection was
always bracketed: “love-hate,” said one of the
school’s star students and strong admirers,
Millisent Fury Hopkins.
Rousseau’s first memory of APR was from the
second day of school, which was also the first time
he got suspended. The class was working on an
activity called Math Minutes, in which students
competed to see how many math problems they
could solve in a minute. “So I got through them, I
finished, I did it, and I celebrated,” Rousseau
says. “I was like, ‘Yesssss!! I got 100 on my Math
Minute!’ ” The next thing he knew, he was being
sent to the office. From there, he was sent home
early for disrupting class.
“I was terrified,” he says. Rousseau was a
talker, yes, and a class clown, but he was not a
bad kid. “Please don’t do this!” he pleaded. But
he had called out, disrupting his class, so they sent
him home. The pattern was set. Practically every
day, a teacher would send Rousseau out of class.
Sometimes it happened three or four times. Most
of his offenses were like that first Math Minutes
celebration. No matter how hard he tried, he
always seemed to find himself talking when he
was supposed to be quiet, making a comment
when he hadn’t been called on, joking around
when he should have been working. Suspensions
piled up. Sometimes he got suspended three times
in one week.
The flip side of Stacey Boyd’s “broken
windows” theory of discipline—squashing the
littlest signs of disorder before they exploded into
chaos—was that students spent a lot of time inside
the dean’s office. According to the school’s annual
report, in the 2002–03 school year, when
Rousseau was a junior, 38 percent of students
received at least one out-of-school suspension.
That was down from 58 percent the year before.
The combined suspension and expulsion rate in
the traditional Boston Public Schools, meanwhile,
was estimated at 6 percent for all students and 8
percent for black students.
Punishments at APR could be embarrassing.
An untucked shirt yielded “shirt tuck-in
exercises”—a calisthenics routine performed in
view of the whole school. “You know, we touch
our toes, we reach for the sky, we jump, and then
we have to tuck our shirt back in,” says Kevin
Thai, a member of the school’s first graduating
class. With every jump, the offending shirt
untucked itself again, requiring another round of
calisthenics. The smallest infractions could
produce the most extreme reactions. Once, in sixth
grade, Chimel Idiokitas, a friend of Rousseau’s,
spent an entire day in the office just for dropping
his pencil. Another student was joking, accusing
Chimel of hitting her with his pencil, and “at that
exact time, I dropped my pencil. And my teacher
just assumed I did that on purpose.” One time,
Kevin even got in trouble for the color of his gym
pants. They were supposed to be navy blue, but
Kevin’s were king’s blue.
The students weren’t always clear on the
reasons for such draconian discipline. “They want
you to come in like a robot. Like, just follow and
follow and follow,” another student told a
researcher at the time. “It’s really fake,” said
another. “It doesn’t really mean anything in the
real world . . . I think some of it is just a lot of
pointless rules.” (On the other hand, if it weren’t
for APR, Kevin Thai would never have known
that “king’s blue” is a color, not to mention have
been awarded a full scholarship to college.)
All of Rousseau’s friends complained. That was
how APR kids bonded. They followed instructions
in the manner that researchers Jere Brophy and
Mary McCaslin term “grudging compliance.” But
while the school could be strict, the teachers also
worked hard to build in time for fun. Chimel
Idiokitas couldn’t understand why he spent that
one day in the office instead of in class, but he
loved the plays APR teachers helped students put
on, the field trips they took them on (including one
to China), the raps they let him and Rousseau and
their friends perform in the lunchroom. Even the
shirt-tucking exercises, a creation of Doug
Lemov’s, were led in the spirit of laughter—an
ironic send-up of the school’s strictness, in which
offending students were the “shirt tuck-in team,”
Doug played the role of overzealous coach, and a
longtime repeat untucker was appointed team
captain. Kevin Thai felt embarrassed, but many
other students laughed as their principal jumped
around wildly, making a deliberate fool of himself.
Rousseau defended the school, even back then.
He might have been terrible at following the rules,
but he believed in them. His teachers were trying
to get him to college, and the rules were there to
make that happen, so he did his best to find the
value in them. The teachers were tough, but they
cared deeply about the students. During a bad
period when Kevin Thai was fighting with his
parents, Chi Tschang mediated a session at school
that helped them reconcile. Another time, Chi
stayed at school until 11:00 at night just to make
sure he replied to an e-mail Kevin had written
him. Later, when Kevin became a teacher, he often
thought of an APR teacher, Alexander Phillips,
and tried to model his own devotion to his
students on Mr. Phillips’s example.
But for all that APR gave them, there were
some legacies Rousseau would have preferred not
to carry. Just as Kevin Thai had looked up to
Alexander Phillips, Rousseau looked up to Chi
Tschang as a model later, when he became a
teacher himself. “If I’m a good history teacher, I
should be as prepared as he was,” he would think.
But Chi also “became the embodiment of my low
self-esteem,” Rousseau says. In high school, Chi
had always been the one to remind the students of
what they needed to do to succeed—and what
would happen if they didn’t: the colleges they
wouldn’t get into, the dreams they wouldn’t fulfill.
So in college, whenever Rousseau questioned
himself, he found himself thinking of Chi. “I
would always hear Chi’s voice, even while I was a
freshman at Williams, telling me that I didn’t
belong there and that I didn’t work hard, that I
didn’t deserve to be there.”
As a teacher, Rousseau thought of that legacy
too. Just like Chi, he often stood in front of groups
of students, knowing the grim odds against them
and feeling compelled to let them know too: “This
is where you’re going to end up if you don’t get it
together.” But he couldn’t help wondering what
effect his words would have on his own students,
years later.
Fear represented a fundamental irony of no
excuses. The point of all those rules and
consequences was to teach the students the
discipline they needed to succeed. But in practice,
the same consequences that seemed necessary to
help students succeed could also make them more
anxious, even angry. Researchers studying school
discipline found that punishment often produced
“resentment, retaliation, and/or emotions that are
counterproductive to learning.” At APR,
sometimes the punishments could even push
students out the door. Many of Rousseau’s friends
came to appreciate APR later, and even love it (the
most notorious complainer, Jonathan Correia,
wound up joining the school’s board). But
Rousseau knew that for all the students APR
helped, there were others the school lost.
The first year, Doug Lemov and Stacey Boyd
had started out with a class of fifty-five or so
seventh-graders. But by the time that class made it
to senior year, only eleven students remained. And
three of them had only joined later on, in ninth
grade. In Rousseau’s year, the winnowing went
from about a hundred kids who had started
together in sixth grade to thirty-four in ninth.
About thirty graduated together as seniors. (By
comparison, the percentage of ninth-graders who
entered Boston’s public high schools in 2000 but
didn’t make it to graduation was 21.6.)
Much of the attrition was innocent. Even in the
beginning, APR never asked students to leave (a
badge of honor that some other charter schools—
and many noncharter schools, for that matter—
could not claim). Some of the students who didn’t
make it to graduation, meanwhile, were like
Rousseau’s friend Chimel Idiokitas, who left after
middle school when he got into the prestigious
Boston Latin Academy, an exam school. And over
time, as APR found ways to support more of its
students, the leakages declined.
Yet even with the best intentions, APR—like
many schools serving high-risk communities—
still lost some of the students the school most
wanted to help. Some left because, fed up with
either the discipline or the long school day or both,
they sought another option. According to their old
classmates, the students who left because they
grew tired of APR’s discipline tended to be the
ones with the least support at home. “They were
making their own decisions,” Chimel says. “There
wasn’t a parent telling them that they had to stay
because it would pay off.” Without that push,
“they couldn’t cut it,” Kevin says. So they left.*
What wasn’t clear was whether the paradoxical
effects of no excuses—life-changing for some,
crushing for others, and sometimes, as with
Rousseau and Chi Tschang, both at the very same
time—needed to be that way. Did the teacher who
inspired Rousseau also have to gut his self-
esteem? Or was there another way?

For Drew Martin, there were many moments of

truth—“stop-everything moments,” one of his
teachers, Ranjana Reddy, called them. The Chi
Tschang report, which Drew had written up in his
weekly staff memo on February 22, 2009, was one
of the big ones. Another came just two weeks
later, when Drew fielded reports of bad behavior
on the No Limits bus.
His school, Rise Academy, a middle school that
is part of the KIPP network in Newark, New
Jersey, had four bus routes, all named after the
school’s values: No Limits, No Shortcuts, No
Excuses, and Opportunity. The buses all had strict
rules. Since the bus was a place where disruptive
behavior could put children in real danger, Rise
banned every behavior that might distract the
driver. Moving around was prohibited, and so was
talking. Transgressors, meanwhile, faced steep
consequences. Talk once, and your parents got
called; talk twice, suspension. Get up from your
seat? Don’t even think about it. Automatic
suspension.
For two years, the rules had paid off. Students
got to and from school safely and obediently. But
now, halfway through the school’s third year,
something had evidently shifted. Students were
not only getting up out of their seats; they were
pushing each other midtransit. Alarmed, Drew
and the other school leaders temporarily
suspended the entire No Limits route. Then, on
Friday afternoon, just as they were preparing for
an emergency parent meeting the next week to
figure out what to do with No Limits, the
Opportunity bus surprised them by returning to
school midway through its route, still half full of
children. The driver provided another unsettling
report. “It turns out,” Drew wrote in his weekly
staff memo that weekend, “there were 3 fights on
the bus that afternoon and, according to the kids,
this has been going on for quite some time.”
Drew’s response was a bit overblown. What he
called “fights” were really more like arguments,
with some pushing. And, outside of the bus, a
healthy majority of students at Rise were still
happy, hardworking, and successful. Yet Drew
could still remember Rise’s first bus rides—
hushed, punctilious affairs, with picture-perfect,
adorable students in brightly colored backpacks
riding silently to school. It pained him to watch as
those same students—a few years older, sure, and
unavoidably unrulier as they entered adolescence,
but the same children!—now openly flaunted
everything they’d been taught. “Two years ago,”
Drew wrote, “if I had seen a paragraph like the
one written above I would have been apoplectic
and completely inconsolable.” The students not
only talked; they fought. Official student monitors
not only failed to report incidents; they actively
lied about what went on.
Like the rest of the no-excuses schools, Rise
was built on the proposition that behavior—or
culture, as some of them put it—came first.
Alumni of Teach For America, many of the no-
excuses teachers got their starts in urban schools
where learning was subordinated to mayhem.
Their first pressing order of business was to
change what school felt like. Plus, as Drew and
the others saw it, interpersonal skills were as
important as any academic subject, maybe even
constituting their own subject: the discipline of
discipline. Just as in math, reading, science, and
history, in discipline the measure of success was
simple. Had the students learned?
The bus episodes suggested that the answer was
no. Like Deborah Ball’s fifth-graders, who
mastered long division one day only to have to
relearn it from scratch the next, the students at
Rise appeared to learn—and then, somehow, by
seventh grade, they forgot. They took all that order
their teachers had given them, and they turned it
back into chaos.
The failure reminded Drew of a book he’d read
a few years earlier: A Short History of Progress.
The author, Ronald Wright, describes a
phenomenon he calls the progress trap, in which
societies, pursuing what they think is progress,
instead create the machinery of their own demise.
Weapons, for instance, initially solve a short-term
problem—easier hunting and readier access to
food. But they also create a long-term menace,
threatening human survival. Nuclear power begets
nuclear war, and the expansion of energy yields to
environmental devastation. “A seductive trail of
successes,” says Wright, “may end in a trap.”
Perhaps the logic of no excuses was a progress
trap too. Silent hallways were the best example.
“When you go and you see that,” Drew says,
“you’re impressed. Maybe that’s because you’ve
been to a traditional public school, where the kids
are off the hook, and therefore, ‘Wow. This shows
that we’re in control.’ ” Plus, the hallways solved
an obvious short-term need: getting students
between classes quickly, calmly, and without
disruption, so that teachers could maximize their
time on academics.
The approach built on some of the early ideas
that defined the movement—like broken windows
theory. Eliminate short-term disorder, the thinking
went, and teachers could create long-term gains.
“The way the thinking goes, if you give an
instruction, and it’s not followed,” Ranjana Reddy
says, “every child in that room learns the lifelong
lesson that instructions don’t have to be
followed.” The entrepreneurs had to respond to
even the smallest infractions; it was the most
important thing they could do to help.
But there was evidence that the thinking,
however logical in theory, was flawed in practice.
The most frightening data came from noncharter
public schools, where, during the 1990s (the same
time period during which the no-excuses
movement arose), a parallel approach to discipline
called “zero tolerance” gained increasing
popularity. Like the broken windows theory, zero
tolerance made powerful intuitive sense. Yet, a
decade and a half later, a growing body of
evidence suggested that, in practice, the policies
did not pan out. One seemingly obvious idea had
to do with the calculation—supported by the
experience of educators in classrooms all across
the country—that removing a few very disruptive
students from a school would create a better space
for the many remaining students to learn. In fact,
as the American Psychological Association
reported in a 2008 summary of the research, “data
on a number of indicators of school climate have
shown the opposite effect.” Schools with higher
suspension and expulsion rates actually had worse
school climates and spent more time on discipline.
The policies disproportionately targeted
nonwhite students, even though studies suggested
they were not disproportionately disruptive. And
increasingly severe consequences for misbehavior
led schools to refer more students to the juvenile
justice system—a phenomenon known as the
“school-to-prison pipeline.” There was also
evidence, the APA report said, that “zero tolerance
policies may create, enhance, or accelerate
negative mental health outcomes for youth by
creating increases in student alienation, anxiety,
rejection, and breaking of healthy adult bonds.”
At the best charter schools, educators poured
their energies into building strong relationships
with students as zealously as they meted out
consequences for misbehavior. From the
beginning, they aimed at getting students not just
to comply with rules, but to decide to act morally
all on their own. At the first KIPP school in
Newark, teachers talked with students explicitly
about these levels of moral development, as the
psychologist Lawrence Kohlberg called them,
even writing them on the steps of the school’s
staircase to underscore the point. Yet at those
same schools, no excuses could take on a life of its
own. If, as Ranjana Reddy understood early on,
the best way to serve children was to respond to
even the littlest sign of disorder, then shows of
extreme, even macho, toughness should be
celebrated. Indeed, for better or worse, stories
about educators’ most outrageous efforts—like the
tale, chronicled in Jay Mathews’s history of KIPP
(Work Hard. Be Nice.) of how KIPP cofounder
Mike Feinberg once traveled to a student’s house
to uninstall her television—became legends.
The stories arguably did some good. In
Mathews’s book, Feinberg says that while he
wasn’t sure if taking out the TV set was the right
thing to do, he did think it “sent the message to
students that he would go to crazy lengths to make
sure they had the time and opportunity to get a
good education.” But the stories could also create
a perverse pressure to prove one’s dedication
through crazy acts. One educator, after being told
that she was described as a loving, sensitive, and
understanding teacher, felt compelled to share a
tale that seemed purposefully shocking—of a time
she’d compared her students’ failure to pick up a
single cereal piece to the September 11 tragedy.
The extreme efforts impressed observers and
produced visibly obedient children. And yet, Drew
says, “some of the things that I think sometimes
look the greatest are not always necessarily the
greatest practices.” What mattered wasn’t what a
visitor saw; it was what all that control did to
children in the long run. In Rise’s case, the kids
who grew up with silent hallways were now
unable to sit through a bus ride. “If that’s
progress,” Drew says, “we’re all completely
screwed. We’re running ourselves off the cliff.” To
produce a different outcome for the students at
Rise, they were going to have to find a different
way to teach discipline. They just had to figure out
what that way was.
As it turned out, Rise’s teachers already had
another tool. They’d been building it for years
without quite knowing they were doing it. The
approach evolved from their frustration, early on,
with the limits of a longtime KIPP tradition, the
Bench.
The Bench was like a glorified time out,
punishing misbehavior with social isolation. A
student on the Bench wore a different colored T-
shirt and was barred from talking to peers. The
KIPP twist was that students on the Bench did not
actually miss class; instead, they sat either at the
edge of the room or in a normal seat, separated
only by the color of their shirt and the prohibition
on talking (although talking for the purpose of
class was often permitted).
The Bench was effective as a deterrent. From a
fifth-grader’s point of view, it was “the worst
thing you could ever get on,” according to Malik,
a Rise seventh-grader. “I see fifth-graders that be
walking through this hall like they’re scared,” he
said. “They’re scared to even look at me.” They
didn’t want to do anything that would land them
on the Bench. (Interestingly, from an adult’s point
of view, the fifth-graders could not have looked
happier: walking through the hallways, they
grinned and held each other’s hands; later, in
class, they threw their hands into the air to answer
a teacher’s question with gleeful gusto.) But the
Bench also had a perverse side effect. The students
who had the hardest time interacting with each
other were also, naturally, the ones most likely to
get on the Bench. Some of them would stay on
“for weeks and weeks and weeks,” says Ranjana
Reddy, the former Rise teacher. As a result, “our
biggest behavioral problem kids would have no
practice interacting with other kids socially.”
The irony wasn’t lost on the teachers. So as the
students got older, Ranjana and the other teachers
decided to modify the Bench. Soon, seventh- and
eighth-graders—the “upper school” at Rise—had
a new Bench alternative. Choices, the teachers
called it. Choices traded the Bench’s immediate
social isolation for the relatively less disruptive
consequences of detention and silent lunch.
(Instead of eating in the cafeteria, silent-lunch
participants ate in a classroom with other
misbehavers.) Choices added a new requirement
too. To get off of it, students had to deliver a
public apology to the class in which their original
offense occurred—and the class had to accept.
The goal was to switch the way the school
taught its disciplinary lessons. The Bench
presumed that students would learn from the
consequence of their actions; the punishment of
social isolation would be enough to teach them
never to misbehave in the same way again. And
maybe that worked with fifth-graders. But for
seventh- and eighth-graders, the teachers
suspected, a punishment wouldn’t be enough to
teach better behavior. Some kind of reflection was
also necessary. With Choices, Ranjana says, “the
theory here is that the learning happens at the end
of that piece—that there is a conversation when
the person gives their apology later, and people are
allowed to ask them questions when they give
their apology.” The resulting dialogues forced
students to make sense of their mistakes, to think
them through.
As Rise’s teachers facilitated more and more of
these public apologies, they also started to lead
spontaneous conversations about behavior.
“Culture conversations,” they called them. Like
Choices, culture conversations had their origins in
more choreographed routines, in which teachers
would introduce the reasons behind the school’s
rules by reading a book or showing a clip from a
movie that illustrated the idea they wanted
students to learn. But over time, the conversations
about behavior started to happen right in the
middle of class. Some conversations were modest,
small-group affairs. Watching two students
deliberately leave out another as they picked
partners for a science lab, for instance, a teacher
named Shannon Grande turned the moment into a
chance to discuss the pitfalls of cliques. Later, in
the same lesson, Shannon responded to a
meltdown of a student named Destiny—“You
need to be in LINES and QUIET!” she had
screamed before storming out of the room—by
asking the whole class to pause and discuss how
to respond when Destiny returned.
Shannon didn’t know about Deborah Ball and
Magdalene Lampert’s TKOT, but the culture
conversation approach bore similarities to it. Just
as with TKOT, they didn’t forsake lectures, which
were still crucial for clarifying important concepts
and introducing new ideas; they simply
supplemented them with responses to students’ in-
the-moment misunderstandings. And just as in
TKOT, the culture conversations didn’t just
happen over the course of a single lesson. They
sometimes stretched over weeks and months as
students helped each other think through their
behavior decisions—the source of which they
came to think of not as “discipline,” but as social
and emotional growth.
In Shannon’s classroom, for example, a
seventh-grader named Jamal often failed to control
his anger. Even the littlest comment could set him
off. Your shirt’s not tucked in, another student
would tell him, and he would scream, slam his
books on the floor, and shut down. “Ohh, why do
you always have to say this to me?” he’d say.
And, “I hate this place!” Sometimes, he would
break out and cry during the middle of class with
no explanation. At one point, he even got into a
fight with one of the grade’s best students, usually
a paragon of perfect behavior. Frustrated, the girl
lashed back, telling him that he was dirty, that he
smelled. Shannon couldn’t believe it. “I pulled the
girl, and she was like, ‘Well, he brings it on! He
gets angry and then what am I supposed to say?’ ”
Perhaps, Shannon suggested, don’t tell him he
smells. “And she’s like, ‘Well, he does!’ ”
That was true. Shannon knew that Jamal’s
hygiene was an issue. His home was a chaotic
place. Just the week before, the power had gone
off. The week before that, the water was shut off.
At another point, part of the roof caved in. Often,
with their mother gone, Jamal and his brothers
would be on their own to cook dinner. He and one
of his brothers didn’t have regular access to
showers until the staff at Rise arranged a private
bathroom they could use before school.
Faced with the girl’s legitimate statement—yes,
Jamal did sometimes smell—Shannon decided to
give the girl more context to shape the idea into
something more constructive. One day, in between
classes, she took that girl and three other high
performers in Jamal’s advisory (Rise’s version of
homeroom) aside. “I said, I’m going to have a
very mature conversation with you right now,” she
says. Jamal, Shannon explained, “struggles just to
get here everyday. I was like, ‘you wake up in the
morning and you’ve got somebody that’s getting
you out of the door, giving you breakfast, making
sure your clothes are clean, making sure you’re
getting here on time. He doesn’t have that.’ ” His
anger wasn’t acceptable, but it was
understandable. And what he needed was not
more reasons to feel attacked, but support—and
examples of ways to react differently. Shannon
concluded with a plea. “I was like, ‘So I need you
to keep this quiet, but I also need you to use your
power and influence in the advisory to start
making a change.’ ”
The girls were taken aback. “I didn’t know
that,” Jamal’s original combatant said. Then she
asked her a question that Shannon hadn’t
anticipated. “She’s like, ‘Well, can we talk to
him?’ ” At first, Shannon balked. She’d hoped the
girls could just lead by example, responding to
Jamal’s outbursts with more equanimity. She
certainly didn’t want to create a situation that
would make him even more vulnerable. But after
thinking for a minute, she decided to let them try.
The risk in drawing more attention to a behavior
problem was obvious. As with Destiny, the
original trigger for Jamal’s meltdowns was often
other students. Giving the students an opportunity
to share their feelings could exacerbate the
teasing, making Jamal feel worse. But Shannon
also knew that middle school was a cauldron of
social relationships, whether teachers paid
attention to them or not. When Destiny stormed
out of science class, she was gone for the moment,
but soon enough she had to return, and when she
did, the students had to figure out how to treat her.
Similarly, Jamal and these girls saw each other
every day. They went to class, lunch, and recess
together. They were all on Facebook. If they didn’t
talk now, in the safety of a conversation that
Shannon could moderate, they would talk
someplace else.
Plus, just as Deborah Ball and Magdalene
Lampert had found with TKOT, the upside of a
conflict—the behavioral equivalent of a
mathematical mistake—was a chance to learn.
Opening up Jamal’s behavior problem for
collective examination might help him think about
it differently. The group might do it better than
Shannon could on her own. “Social problem
solving,” researchers called similar approaches.
Still, when the conversation began, Shannon
was nervous. And the girls launched it with a
bomb. “When you get angry, we get angry back,”
they said. “What do you expect us to say?” But to
Shannon’s amazement, Jamal responded not by
lashing out, but by talking calmly. Soon enough,
all of them—Jamal and the girls—were asking
Shannon to let them finish the conversation in
private. She decided to let them go for it. All of the
students were in her advisory, and by that point,
she was not only holding culture conversations
with them, but letting students lead them.
Whatever they said, over the next month their
advisory transformed. When Jamal had an
outburst, instead of lashing back at him the other
students would lead him in a conversation on
what he was doing to change: journaling or deep
breathing or taking a minute in the hall. The most
powerful strategy of all seemed to be not any
single technique, but the conversations
themselves. “Even when he has a rough day, we
talk about it in advisory, and he gets it out, and he
moves on,” Shannon says.
For his part, Jamal began to see school
differently. Early on, with all its crazy rules, Rise
had felt like the rest of his life: a series of personal
attacks all devised to break him down. But as
Shannon Grande and his classmates worked with
him, Rise started to feel different, even safe. The
new feeling transformed him. “Rise was my
turning point,” he said that year. The change
wasn’t academic, but much more abstract. What
Rise gave him, Jamal said, was “a different view
on life.” Before coming to the school, he was an
expert at negative thinking. He felt alone, and he
thought a lot about death. At Rise, he learned to
imagine success as clearly as he could imagine
failure. One morning, he even woke up
remembering a dream he’d never had before. “I
was living by myself, but with my family, in this
big house, and it was peaceful. No yelling, no
arguing. Nobody saying somebody stole this.
Nobody dying. Just one peaceful area.” Before, he
said, “I was expecting to die.” But “at Rise, it
gave me a new future. It gave me hope.”
Educators often talk about discipline as a choice
between rules and autonomy, systems and
freedom, “tight” and “loose”—as if they have only
two choices: either build elaborate behavior
systems like the Bench or let students roam free.
At Rise, changing the way the school exerted
control by no means meant letting go of it. In fact,
it was the opposite. Just as Ruth Heaton and
Sylvia Rundquist had found at the Spartan Village
school, giving students more independence meant
the teachers had to do more work, building more
intricate systems with more deliberate supports—
the stuff teachers call “scaffolding.” They weren’t
less strict. They just changed the point of
intervention. Instead of building a million rules to
prevent misbehavior from happening in the first
place, they loosened up at the start—only to erect
ever more elaborate responses when, inevitably,
students crossed a line.
In addition to culture conversations, teachers
increasingly sought outside help. As Paul Tough
explains in How Children Succeed, a combination
of research in economics, psychology, and
neuroscience has shown that the students targeted
by charter schools—the children of families
suffering from multigenerational poverty and a
history of racism—face more challenges than their
more affluent white peers face. Even the best
teachers often needed the help of psychologists
and other professionals for students who
continued to struggle.
Sometimes teachers used more specialized
approaches. “Does it make sense for one fifth-
grader to bring up a concern about another fifth-
grader in front of thirty kids?” says Mariel
Elguero, one of the school’s founding teachers,
who became dean of instruction. “Maybe in some
contexts, yes, and maybe in some contexts, no.
Maybe what would actually make sense is if we
got those two kids together and they sat in the
corner by themselves, and we gave them tools, and
they figured it out.”
The more complex approach did not solve all
behavior problems, not even close. Although Rise
was religious in its refusal never to ask a student
to leave (a policy the school did not share with all
other charter schools, or all district schools for that
matter), every teacher could still rattle off a list of
the students who made the decision themselves,
resisting Rise’s considerable efforts to change
their minds: Aisha, who twice had to be taken
from the school in an ambulance and once
threatened to kill both of her parents in the
school’s main office; Deon, who missed two
months of school one year and failed all of his
classes; Abdul, the best basketball player ever to
attend Rise, but a struggling student whose
mother refused to let him repeat a grade. Some
days, even after Jamal’s turning point, it was hard
not to worry that he would fall down the same
path. “We have a lot, a lot of kids that it doesn’t
work for,” Drew says. “We don’t have an answer.
We just hold on as long as we possibly can.”

At the Academy of the Pacific Rim, culture

conversation was not a term teachers used, and
neither was TKOT. But the moments that
Rousseau remembered best from school bore
striking similarities to those conversations at Rise.
Each punishment came with the requirement that
the offending student reflect on the mistake made.
As a result, students spent almost as much time
talking about their behavior mistakes as they did
complaining about all the school’s rules.
Sometimes the reflection took the form of a
public apology like the ones Rise students had to
make to get off of Choices. For more serious
offenses, APR required students to write out their
apologies in an essay. The process was “what I
think suspension should be,” says Kevin Thai.
“Every time, they sat me down, they talked to me,
they figured out what was wrong, and tried to
[help me] learn from that experience.” They didn’t
“just throw [me] into a box and say go home.”
Rousseau went through a similar routine with
Doug Lemov every time he was sent to the dean’s
office. Looking back, these were the moments he
appreciated the most, the minutes between the
delinquency and the consequence, when the two of
them sat down and talked—not just about what
had happened, but about why. Usually, Doug
began by listening. What was Rousseau
struggling with? What had made him call out?
Why hadn’t he turned in his homework? Then
they talked about his response—why calling out
was unproductive for the whole group, or why
disrespecting the teacher prevented Rousseau from
working well with him or her.
Doug himself had gone through an evolution
parallel to Drew Martin’s at Rise, enhancing these
more purposeful moments over time. That was the
beauty of the crucible. “We were rebuilding that
machine weekly, monthly, yearly,” he says. The
school Rousseau and his friends arrived at, in
APR’s first year of existence, was not the same
school they graduated from. Just like Drew, Doug
questioned almost everything he did. And one of
the practices he thought about most—changed the
most—was discipline.
So did many no-excuses educators. The
evolution at Rise tracked similar changes at
dozens of other schools. At the first KIPP
replication school, in the South Bronx, the
network’s cofounder, David Levin, never used the
Bench at all. The original KIPP school, in
Houston, replaced its rigid rules with a more
invisible (and more intricate) structure that gave
students the space to make more mistakes. And
across the network, officials added research-based
substance to their intuitive early ideas about
culture, creating a character curriculum that
helped give teachers like Shannon Grande words
and ideas to flesh out their culture conversations.
The simple way to understand the change over
time was that educators collectively decided to
stop crossing certain lines. And that was partly
true. Indeed, every no-excuses educator seemed to
have a story for the moment he decided to act
differently. For Drew, it was when he found
himself hoping a girl’s mother wouldn’t catch him
disciplining her—and realized he never wanted to
make that wish again. For Doug, it was the time
he yanked a wandering dog out of APR, only to
have a student remind him, sagely, “Mr. Lemov!
He’s just a dog!”
Chi Tschang had a story too. It happened
midway through his time at KIPP Fresno, when he
took some of his students on a field trip to visit the
classroom of Rafe Esquith, the California educator
and author who influenced many KIPP leaders.
After Rafe’s students performed a Shakespeare
play, all of the students went out together to share
a meal. “One of the things that we realized,” Chi
says, “is that Rafe’s kids were so much nicer than
our kids. So much more refined, more thoughtful
and articulate—and more considerate.” He and a
colleague puzzled over the difference on the bus
ride home. “And what we realized is, ‘They’re 10
times nicer because Rafe is 10 times nicer. Rafe
never yells.’” Chi changed—and so did his
colleagues.
But reality was more complicated than just
deciding to act more humanely. The most
thoughtful schools—the ones that Doug ran, the
ones like Rise—had wanted to teach their students
something more than just compliance from the
beginning. And to get there, they did more than
simply swing back and forth between two sides of
a pendulum: loose and strict. They reflected on the
same question Rousseau asked himself, about
whether the good outweighed the bad, or whether
the bad was even necessary—and they decided
that, yes, it was possible to push students without
crushing them. What was required was not just
the right motivation, but the right skill. Discipline
was not a black and white choice—tight or loose,
structured or joyful. It was, says David Levin,
“unstructured structure.” And it was much harder.
It required skill. Ultimately, that was what the
entrepreneurs gained over time—not new
ambitions, but the expertise to make their fullest
vision come true.

What, then, explained the charter school in

Harlem? The one where Rousseau found a
building full of teachers, religiously versed in the
Lemov taxonomy and yet defying it?
The irony was that Doug’s other reason for
writing the taxonomy—just as important as his
commitment to developing a shared professional
language—was to eliminate the possibility of such
a school. Like Chi and Drew, Doug had learned
some of the most important teaching lessons the
hard way, and he’d done so, he knew all too well,
on the backs of some children. (Indeed, he not
only rejected many of his own early practices; he
rejected the “no excuses” label altogether.) Why
should other educators have to do the same?
“Culture,” as he called the discipline of discipline,
was one of the teaching problems he’d worked on
the most. And he knew it was also one of the
easiest to wreck.
Teachers, especially teachers who had taught in
dysfunctional schools, had a tendency to glob onto
discipline tools “like catnip.” “Like rats in the lab,
you keep pressing the lever, even when they
become bloated and distended,” Doug says. By
being meticulous about the details, by
emphasizing lessons like “purpose not power” and
“The J Factor,” for joy (Technique No. 46), he
hoped to help more schools achieve what a few
schools were figuring out how to do—how to get
students not just compliant, but invested; not just
obedient, but happy.
In some cases, the taxonomy had done just that.
Eight or ten years ago, Doug says, “the number of
schools that had outstanding culture were very few
and far between.” Now “it’s common. It’s
common as day to have a charter school that’s
figured out how to manage behavior and culture.”
If the taxonomy hadn’t worked for all teachers,
perhaps it was because—on its own—a book
could go only so far.
Rousseau called the difference between the
schools where discipline worked and the schools
where it just didn’t “that intangible piece.” What
was that extra piece, exactly? Some charter school
teachers pointed to relationships. The teachers
who really got kids to change, they said, were the
ones who built strong personal bonds with them.
Mariel Elguero, the dean of instruction at Rise,
talked about the importance of expectations. Early
on, she said, “We were like, fifth-graders can’t do
talking during guided practice” (the part of the
class when students work on the lesson with
teacher oversight).” “They can’t get along! That
was the perception.” Now, “it’s like, of course
they can. I think it’s just expecting more.
Expecting what you didn’t think was possible.”
Doug, for his part, said the difference boiled down
to something even simpler: love. “In your heart,
you have to think, ‘I love these kids, and I want
the best for them,’ ” he says.
When he became a teacher, Rousseau tried to
do all of those things—build strong relationships,
expect what didn’t seem possible, teach with love.
But he found that trying these techniques lacked
“that intangible piece.” As he saw it, this
intangible piece was like the preparation he did to
teach history. Getting ready for a lesson, he would
force himself to imagine the mental steps the
students would need to take to accomplish the
goal of the lesson. By walking through an
assignment himself first, he could do a “think
aloud” when it came time to work with the
students, bringing to the surface the invisible leaps
that came naturally to him but that the students
still needed to learn.
With discipline, Rousseau did something
similar, digging into his memory to recall his own
lowest moments. “How did I deal with it as a
kid?” he’d ask himself. “How do kids deal with it
now? How am I dealing with it now? What has
helped me to break through now? How can I help
a kid break through now with that same thing?”
Teaching discipline required the same amount of
mental work as teaching history, plus an extra
dose of courage.
But soul searching was not the same as
teaching. Once Rousseau could see the idea his
students were missing, he still had to figure out a
way to give it to them. He had to marshal a kind of
knowledge that didn’t seem to have a name—a
knowledge that was neither about teaching in
general nor about the subject he was trying to
teach, but a combination of the two. Lee Shulman
would have called it pedagogical content
knowledge, the skills needed to teach a subject.
Rousseau needed a special knowledge for teaching
discipline.

Deborah Ball’s and Magdalene Lampert’s

research on pedagogical content knowledge in
math teaching had parallels in other subjects, from
history and science to English and even physical
education. But there was no parallel in the
discipline of discipline.
Researchers have begun to uncover some of the
dynamics of discipline. When working with
difficult students, researchers have found, many
teachers lean heavily on rules and punishments.
And this “operant conditioning” seems to work in
the short term, especially when teachers follow the
guidelines outlined in the Lemov taxonomy. If the
least invasive, most positive, and most private
correction is used, students are more likely to
comply not just “grudgingly,” but willingly. But
psychologists have increasingly found that, on its
own, this “conditioning” is not enough to help
students in the long term. Teachers also have to
teach their students what the researchers call
“social problem-solving”—mental and emotional
approaches to dealing with interpersonal
challenges.
But examples of this more complex approach—
of a pedagogical content knowledge for discipline
—were harder to find. The researcher who came
closest, a teacher and education professor at
Northwestern named Carol Lee, never intended to
study the subject. Inspired in part by Magdalene’s
and Deborah’s work, she decided to videotape her
own classroom in an inner-city public high school
every day for three years, hoping to gain insight on
teaching literature. But Carol found that helping
her students become better readers required a
focus beyond just academics. “The framework,”
she wrote in a book reflecting on the project,
Culture, Literacy, and Learning, “had to address
the developmental as well as the cognitive needs
of students.” By “developmental,” she meant “the
process through which over time humans learn
how to address the challenges of cultivating and
sustaining a sense of well-being and of
competence, of nurturing interpersonal
relationships first within families and later across
wider social networks, and of navigating
obstacles.” Over the course of the three years,
Carol thus worked on two kinds of teaching at
once. The first had to do with literature; the
second focused on discipline and development.
One episode showcasing the latter kind of
teaching occurred in 1997, at the start of her
students’ senior year, when Carol noticed that one
student, a girl named Taquisha, was reading that
morning’s copy of the Chicago Sun-Times. A few
minutes before, Carol had shown the class a five-
minute film, one of the everyday cultural artifacts
that she used to prepare students to analyze texts
like the Toni Morrison novel Beloved. She was in
the middle of asking them to comment on what
they’d seen when she noticed Taquisha’s
unconventional choice of reading material.
The moment posed a challenge. As Carol
recounts in her book documenting the year,
“Taquisha is actually reading the newspaper while
I am conducting the lesson.” And she had spent
enough time in the classroom with Taquisha to
know that she was “a very strong-willed young
lady . . . the sort of person with whom you do not
pick a fight unless you’re willing to go to the mat
and be ready to strike with definitive force to
win.”
Carol knew that she had several options for
how to respond. “I could have embarrassed her,
punished her with additional work, lowered her
grade, or sent her to the office.” But while she
believed that some punishments could teach, she
also knew that, in Taquisha’s case, a punishment
like sending her to the office would give the
student exactly what she wanted—“to be removed
from any responsibility for active participation.”
And that was the opposite of Carol’s goal.
So instead of jabbing, she struck a playful tone.
“Other questions. Other questions,” Carol said
calmly before making her move. “Taquisha, you
have a question inside that paper there?”
“Yup,” Taquisha replied emphatically,
retaining her fighting pose. Yes, I am indeed
reading the Chicago Sun-Times during your
lesson, Taquisha was effectively saying. The move
was a taunt, a second invitation to spar. But
instead of taking the bait, Carol deployed a
response that she later called “a kind of Tai Chi
move.” Instead of responding to the “yup,” Carol
simply tilted the subject in her desired direction.
She “deflect[ed] her motion toward me by simply
getting out of the way.”
 “What’s your question?” she asked.
Taquisha punched again, saying she wanted to
know what the short film Sax, Cantor, Riff, by
Julie Dash, had to do with the book they were
preparing to read, Beloved. “Essentially,” Carol
translates, “Taquisha is publicly assaulting the
design of my lesson.”
Faced with this third attack, another teacher
might have finally given in. But Carol, determined
to force Taquisha into the discussion, deployed
another Tai Chi deflection. “Well that might be a
question for me,” she said. Taquisha had, after all,
posed a reasonable question—why were they
doing what they were doing; what was the
purpose?—and Carol wanted to acknowledge it as
such.
“Well,” Taquisha persisted, her permission to
challenge granted, “what does the book have to do
with the girl and the man singing?” (Sax, Cantor,
Riff depicted a sequence that included a woman
singing an African American spiritual and a
Jewish man singing in Hebrew.) “Say that again,”
Carol said. “That’s a good question.” She couldn’t
quite hear Taquisha’s response—other students,
apparently curious about the same question, spoke
over her. “What does the girl have to do with
what?” Carol asked again. “What does the girl
[sic] have to do with the girl and the man
singing?” Taquisha repeated.
Later, watching the exchange on videotape,
Carol marveled at the evolution. In just a few
minutes, Taquisha had shifted from open defiance
to challenging the lesson plan to asking a
sophisticated question that was exactly what Carol
had hoped for when she first launched the
discussion.
The two sequences in the film—the young
African American woman singing, on the one
hand, and the Jewish man singing, on the other—
were disconnected scenes. As far as the viewer
knew, the two hadn’t even seen each other. But
Taquisha made two important assumptions: one,
that the author (in this case, the director) had
created the work with some kind of intent; and
two, that her role as a reader (or in this case,
viewer) was to “impose some form of coherence”
on the apparently disparate scenes.
“That is a beautiful and sound question,” Carol
said, after rephrasing it. “Found it in the Sun-
Times too, didn’t you?”
“Yup!” Taquisha said. Except this time, the yup
had a different sound. She was no longer resisting.

In part, Carol Lee’s pedagogical content

knowledge was a perfect analogue of Magdalene’s
and Deborah’s. To teach English, Carol found that
she needed to know how people learned it—what
distinguished a novice reader from an expert? She
also needed to know the mistakes and
misconceptions that might cause a young person
to stumble—for instance, the fact that incorrectly
pluralizing a word like “child” as “childs” is
actually a step forward for beginning speakers of
English, rather than a step back (the mistake
shows that the student grasps the concept that
plural words usually end in “s”). And she needed
to understand the nuances of the discipline that
might escape even an excellent reader or
professional writer—like precisely how irony,
satire, and unreliable narration differ, and how the
three can and cannot intersect.
But Carol added another dimension, one that
matched her developmental work. She called it
“the ethical and moral” part of teaching. A
teacher, she said, “must come to know each
student and the life circumstances that student
brings with him when he enters a classroom.”
That didn’t mean every single detail—a hopeless
task, especially for a high school teacher working
with a hundred or more students a year. But, like
reading or math mistakes, students’
developmental challenges came in relatively
predictable patterns, so teachers could get to know
the broad strokes. (For her part, Carol drew on a
mix of her own experience and the research of the
University of Chicago psychologist Margaret
Beale Spencer, who describes young people’s acts
of resistance as “maladaptive coping strategies.”)
That way, meeting a student like Taquisha, a
teacher could quickly predict how she was likely
to think and act, without having to take a detailed
inventory with each new student.
Whereas charter schools handled discipline
separately from academics, Carol blended the two,
not only in the same class, but in the same
discussion. She responded to Taquisha’s
resistance not with a conversation, but with a
subtle invitation to participate in class more
productively—and a careful, gentle modeling of
how to do that, all in the context of the close
reading of a film.
Similarly, although Magdalene Lampert did not
use the same “developmental” frame, she did talk
about the work of classroom management. She
called it “Teaching Students to Be People Who
Study in School,” or, more ambitiously, “academic
character education.” The values she taught were
versions of discipline, just fused with the habits
that students needed to do well in math class (and,
she thought, in life)—an “inductive attitude,”
“curiosity,” and a broad sense of self as “a person
who could have ideas.” And just like Carol Lee,
Magdalene found she could teach these attitudes
best by connecting them with math or other
academic activities.
The difference in approach had something to do
with each woman’s identity as a teacher of an
academic subject; that was their primary focus, so
that was how they looked at everything they
taught, even behavior. But it also probably
stemmed from their different approaches to
teaching academic subjects. A key difference was
what their classrooms demanded of students. As
James Stigler found in his TIMSS video study, a
principal characteristic of American classrooms
was the expectation of attention (also known as
the ability to lock eyes with an overhead projector,
a teacher, or a worksheet). Magdalene Lampert
and Carol Lee excelled because they stretched that
expectation.
Like teachers in Japan, Carol required more
elaborate activities in her classroom. In order to
participate well, Taquisha didn’t just have to
throw away her newspaper and stare at the board.
She had to formulate an insightful question.
Similarly, in Magdalene’s class, Richard, one of
the students who struggled with what Magdalene
called academic character, needed to do more than
just stop fooling around with his friends. He also
needed to learn how to get started on the problem
of the day without asking Magdalene what to do.
He needed to think about math not just as an
assignment forced on him by his teacher, but as a
challenge that he could be curious about—and a
puzzle that made sense. And he needed to figure
out how to disagree respectfully, whether about a
mathematical idea or something more personal.
In Carol and Magdalene’s view, getting
Taquisha and Richard to behave was not a step
toward getting them to learn—first, pay attention;
then, learn. Behavior was, instead, synonymous
with learning. The two things happened at the
same time. That was why Carol responded to
Taquisha’s resistance not by mounting a
discussion about it, but with a series of pointed
questions designed to show her how to respond
more productively. And it was why Magdalene
gave Richard not just a lecture about the need to
start on problems by himself, but also a new seat
next to two students she knew were “helpful to
their peers but not in a way that would be
embarrassing.” That way, he could work on his
problem of the day without the teacher’s help by
learning to turn to a fellow student first.
Perhaps the entrepreneurs were right to separate
discipline into its own subject. By making a
separate space for culture conversations, they gave
them more credence. And the separate space
certainly didn’t preclude blending
“developmental” and academic work. They could
do both.
But it was also true that if they wanted a more
integrated pedagogy—one that could
simultaneously teach a student multiplication and
curiosity—they would have to change the way
they taught, and not just the way they taught
discipline. Compliance, the end point of a simpler
approach to teaching behavior, didn’t just make
way for more attentive learners. For the
entrepreneurs as for so many American teachers,
compliance was also often what their math,
reading, and science lessons required.
Magdalene Lampert sometimes took the
argument even further. Certain kinds of teaching,
she argued, were simply more likely to produce
resistance than others. Classroom management
challenges, she contended, could be dealt with in
part by redesigning the activities teachers asked
students to do. Carol Lee found something similar
in her English class. As the newspaper example
made clear, asking demanding questions of
Taquisha and her peers did at first make some of
them resist. But over time, the approach opened
them up. By asking more of students, she also
gave them more of a voice. And in this way, Carol
could make them feel safer. The classroom
belonged to them as much as it belonged to her.
The idea that classroom management
challenges could be solved by well-designed
lesson plans was exactly the kind of argument that
caused some entrepreneurs to reject ed schools in
the first place. Yet by 2010, many of the
entrepreneurs had begun to wonder whether the
academics had a point. In the taxonomy—which
Doug published that year in the form of a best-
selling book, Teach like a Champion—Doug
listed five principles for building a positive
classroom culture. Predictably, the first four were
discipline, management, control, and influence.
But the fifth was what Doug called “engagement,”
the process of getting students to do “productive,
positive work.”
Rousseau’s first teaching assessments
illustrated this shift. Observing him, his principal,
Stacy Birdsell O’Toole, always said the same
thing: he was great at getting the kids to behave,
but to what end? “Rousseau is an incredibly
charismatic teacher, and his management is
incredible. He invests kids wholeheartedly in what
he wants kids to do,” she says. “But what we
really had to work on is, ‘What are you asking
kids to actually do?’ ”
Later, when Stacy became director of training at
the Match Teacher Residency, a program in
Boston designed to prepare teachers to work in no-
excuses charter schools (including the school
associated with the residency, also called Match),
she found that Rousseau wasn’t the only one
working on this problem. All the charter schools
she visited were talking about the same thing.
Stacy had diagnosed herself with the problem too.
“The absence of misbehavior,” she had realized,
“doesn’t mean the presence of high levels of
learning.” They’d cracked the code of how to get
kids to behave. But they were missing a vital
academic ingredient. “Rigor,” they called it.

* There is some evidence to support concerns

that charter schools serve selective student
populations, even though they are supposed to be
open to all children. My own reporting has found
evidence that some charter schools take actions
that tailor the populations they serve, including
discouraging applicants and “counseling out”
students who are already enrolled. But I have also
found cases of schools, including APR and Rise
Academy (described later in this chapter), that
actively oppose these selective policies.
 8
THE POWER OF AN INSIDE JOKE

Seneca Rosenberg entered Teach For America in

2001, eleven years after Wendy Kopp founded the
program and half a decade before Doug Lemov
began his taxonomy. Placed at Cassell Elementary
School in East San Jose two weeks before the start
of school, she spent her first weeks as a teacher
lurching through a series of trials. First came the
dirt, a thick film of which covered every surface of
her classroom. Although the school did employ a
janitor, his responsibilities apparently did not
include scrubbing down desks. Next came the
musical chairs. On the second day of the year,
after Seneca had dutifully memorized all her
fourth-graders’ names, the administration
assigned her a brand-new class. A month and
dozens of new names later, the same thing
happened again. “And I didn’t get the first class
back, of course. I got a new class of fourth-
graders,” she says.
What struck Seneca most of all, however, was
not the chaos of her own assignments, but the
radically disparate experiences of her fellow corps
members. She noticed this most acutely in her
third year, when TFA made her a “learning team
leader,” charged with coordinating a group of
corps members working across the South Bay.
The teachers had almost everything in common.
They all came from the same training program
(TFA followed by a crash course at San José State
University); they had students from essentially the
same demographic (fourth- and fifth-graders
living in a poor, urban area, many of whom spoke
different languages at home); they taught under
the same state standards (California’s) and in the
same district (Alum Rock Union Elementary
School District); and they had roughly the same
number of years of teaching experience (between
one and three). But as Seneca traveled from school
to school visiting the teachers’ classrooms, or
investigated the products of their students’ work at
monthly meetings, she saw that their teaching
differed wildly. From the way they organized their
classrooms to the kinds of projects they assigned
and the work that hung on the walls, almost
nothing about the teachers’ approaches was
consistent.
The randomness was most visible in students’
writing samples. “Length, the strength and clarity
of an argument, complexity and variety of
sentence structure, the type of vocabulary, the
accuracy of conventions”—every aspect varied
from teacher to teacher, Seneca remembers.
“Despite all that we had in common with respect
to training and standards and beliefs, it actually
seemed we might be learning quite different
things, with pretty deep implications for our
students’ learning.”
Seneca had originally thought that talent alone
separated one teacher from another. Some teachers
simply figured out how to help their students learn
and others didn’t. But it didn’t take her long to
realize that this explanation didn’t capture the
problem. What mattered most, for her colleagues
at least, was not will or natural skill, but luck.
What workshop had the teacher happened to
stumble into? Whose classroom did she happen to
get placed next to? Which mentor happened to
take an interest in her work? Thinking back,
Seneca realized that her own trajectory had
depended on a handful of accidents. If she hadn’t
taken that specific workshop, or hadn’t met her
teacher friend, Laura, or hadn’t been assigned that
particular supervisor, her teaching would have
evolved differently. Probably for worse, but who
knew? Maybe for better.
The whole arrangement seemed “absolutely
insane,” she says. How had something so
important been left to chance? Searching for
explanations, Seneca took to the web. Surely she
was not the first to marvel at this randomness;
some education researcher must have studied the
problem, maybe even suggested a remedy. But
instead of answering her question, the policy
papers she found skirted the issue.
One typical study, hosted on the website of a
leading education school, examined parents—
specifically, the kinds of teachers parents
requested for their children when enrolling them
in a new grade. The study found that preferences
differed depending on families’ demographic
background. Low-income parents of color cared
more about teachers’ academic records than their
popularity with students. Wealthier parents,
meanwhile, tended to make the opposite choice.
Across the board, many parents—30 percent—
requested a particular teacher when given the
chance. The implication, the researchers pointed
out, was potentially profound. The national trend
toward greater school choice might end up
exacerbating segregation as parents with different
race and class backgrounds looked for different
strengths in teachers.
Another study took a novel approach to
investigating the challenge of teacher turnover.
Instead of looking at data on teachers who left the
classroom, the researchers studied teachers who
returned to it. By figuring out why those teachers
came back, they surmised, they might find a way
to prevent the teachers from leaving in the first
place. Indeed, the data suggested that the pool of
returning teachers was large. More than 40
percent of those in the study who left the
classroom later came back. The data also pointed
to an intriguing pattern in those who left and then
returned. Many of them, especially women, were
the parents of young children. When they first
taught, they had no children; when they came
back, their children were old enough to go to
school. The pattern suggested that districts could
avoid the costs associated with retraining
returning teachers by investing in convenient child
care options so that they wouldn’t leave in the first
place.
The studies asked important questions, but
though parent preferences and teacher turnover
were related to how people learned to teach, they
were ultimately different subjects. Then one day,
Seneca stumbled onto a professor’s web page at
the University of Michigan. A photo showed an
older man with downcast eyes, smiling in front of
a bed of flowers. In contrast to the other studies,
this professor’s research asked the questions
Seneca had been asking herself, only in more
formal language. Where she thought about the
“randomness” of what her fellow teachers’
learned, he described the “inconsistency” of
“instructional guidance.” Where she diagnosed
teachers’ varied learning experiences as a “mess,”
he described “a blizzard of different and often
conflicting ideas” that added up to an overall
“variability” or, more plainly, “incoherence.”
Different words, but the same conundrum.
 Even more important, he had an explanation for
the source of the trouble. Seneca later learned that
his students had come to name the quandary after
their professor: the “David Cohen coherence
problem.”

Since leaving Michigan State with his wife,

Magdalene Lampert, in 1993, David had been
doing more and more thinking about what had
gone wrong. And not only in East Lansing, but
also in California and all across the country. Why
did every American effort to improve teaching
seem to fail? From the slow dissolution at Spartan
Village to the false revolutions in California, did
the disappointments have a common cause?
Over time, David found his answer in his
original discipline, history. Specifically, he
realized that he could trace the American
resistance to reform back to the founding fathers
and their disagreement about centralized power.
Should the federal government hold the highest
power, or should the states? The debate
foreshadowed much that was to come in
education. Nearly two and a half centuries after
the constitutional convention, the question of who
should control the schools remained unresolved.
American education was like the story that
David’s old friend Lee Shulman told about a rabbi
mediating a dispute between two men over the
ownership of a chicken. After the first man
explained why the chicken was his, Lee said in a
talk recounting the story, “the rabbi nodded sagely
and stated, ‘You are right. The chicken is yours.’ ”
When the second man gave his testimony, the
rabbi nodded again. “You are correct. The chicken
must be yours,” he said. Confused, the rabbi’s
wife spoke up. “My dear, it is impossible for this
one to be right, and that one too,” she said.
“That’s correct,” the rabbi replied a third time.
“You’re also absolutely right!”
Similarly, in American schools, the federal
government was sovereign and the states were
sovereign. Both. And if you thought that situation
couldn’t possibly hold, you were correct about that
too.
Instead of guidance, American schools endured
mass confusion. Principals received mandates
from the feds and from the state and from the
district, sometimes matching and sometimes not
(and only sometimes funded). Teachers got advice
and orders just as contradictory as the directives
their bosses received. Their local curriculum said
one thing; their education school another. And the
textbook, when there was a textbook, said
something else altogether. With fifty states, more
than fourteen thousand school districts, and nearly
a hundred thousand schools, the law of the
educational land was incoherence.
Incoherence sabotaged quality. If teaching was
a skill or trade to be mastered, then teachers were
also apprentices, students of another kind. Like
students, they either had to figure out the material
on their own, or they had to be taught. The law of
incoherence meant that instead of training with
one good teacher, they received the equivalent of
seventeen bad ones, each one saying something
completely different.
The lucky ones made progress against the odds.
Like diligent students in chaotic schools, these
resourceful apprentices took clear guidance when
it emerged and ignored the rest. The majority,
meanwhile, took pragmatic steps to inure
themselves to chaos. Faced with yet another
conflicting order, they responded with what the
educator Lovely Billups once described as the
American teacher’s creed: “This too shall pass.”
Nod politely, thank the state/district/professional
developer/professor for the suggestion, and then,
as Billups described in a speech, “close the door
and go back to what you believe in.” Often, those
still open to change ended up like Mrs. Oublier of
California, whose teaching “revolution” bore only
superficial resemblance to the state’s intended
reforms. These teachers took the new ideas and
incorporated them into the very different kind of
teaching they’d learned somewhere else.
Impressed by David Cohen and his work,
Seneca decided to go to grad school and learn
from the professor himself. Over the next eight
years, she found herself especially struck by the
flip side of coherence—all the stuff that couldn’t
exist without it, the institutions and shared
knowledge base that incoherence had prevented
American schools from building. David called this
“infrastructure.” The dictionary defined
infrastructure as “the basic facilities, services,
and installations needed for the functioning of a
community or society”—building blocks like
roads, bridges, and power lines. Educational
infrastructure, as David defined it, was a school
system’s intangible equivalent, the foundation for
all teaching and learning. It comprised three main
categories: a common curriculum suggesting what
students should study, common examinations to
test how much of that curriculum the students had
grasped, and teacher education to help the faculty
learn to teach exactly what students were
supposed to learn.
Acting in an environment of incoherence, the
U.S. government had never built educational
infrastructure. Without infrastructure, meanwhile,
schools also failed to develop other crucial
resources. “Chief among these,” David explained,
in a book he cowrote with the political scientist
Susan Moffitt, “is a common language concerning
teaching, learning and academic content.” Doctors
had their Physicians’ Desk Reference, with its
technical terminology and its evolving
descriptions of common problems and treatments.
Electricians, plumbers, and pilots had
continuously updated “standard operating
procedures” outlining best practices. Teachers had
—well, they had a bunch of question marks. No
agreement on what their work should aim to teach,
no common vocabulary to describe how to do it,
and no standard measures to know whether they
had succeeded.
No wonder Seneca and her fellow Teach For
America corps members entered a single school
district and encountered a dozen different
experiences! With no infrastructure, they were like
acrobats walking across a moving tightrope
blindfolded—no spotter, no safety net, and no
map. It was no surprise that so many of them
tumbled to the ground.

Seneca arrived in grad school in 2004, two years

into the life of the landmark No Child Left Behind
education law, which took the standards
movement and nationalized it, requiring every
state to set learning goals and judge schools
according to whether they met them. In many
ways, the accountability law would seem to be the
perfect solution to the David Cohen coherence
problem. Marshaled by a historic coalition that
included labor and business leaders, Republicans
and Democrats, and representatives of multiple
branches of government (not only congresspeople
and senators, but also a group representing state
school superintendents), NCLB proved that
Americans could take common action on schools.
And by requiring that states write standards for
what students should learn, it took a step toward
coherence.
But watching the effects of the law unfold
across the country, Seneca saw that No Child Left
Behind was not creating the infrastructure that
coherence was supposed to bring. In place of
common curricula, tests, and teacher education,
NCLB created standards, tests, and accountability
measures. The only overlap—tests—were hardly
the kind imagined in David Cohen’s vision. For
one thing, the new state tests were rarely tied to a
clear curriculum outlining what students should
study. That wasn’t surprising, given that most
states didn’t have a common curriculum. But
standards didn’t clarify the matter either. They
offered learning goals, whereas curricula provided,
in the words of one writer, a “day-to-day, week-to-
week, year-to-year road map for reaching those
goals.” Drawing this road map was still left up to
individual school districts or, in many cases, the
teachers.
Accountability measures might have acted as a
form of teacher education, levying consequences
to schools that failed to meet standards and
thereby at least suggesting when teachers needed
to improve. So might tests, which could give
teachers a report on what their students had and
hadn’t learned. But as David knew from his
research in California (and as Seneca had learned
through her experience in San Jose), standards and
matching tests—while a good start—were still no
guarantee that teachers would learn to teach in a
better, or even uniform, way.
If what education needed was infrastructure, No
Child Left Behind was “best understood as a sort
of exoskeleton,” David wrote. It outlined goals
and offered consequences for failing to meet them,
but it only skimmed the surface of schools’ core
work. Teaching—that daily crucible on which
change depended—was left untouched.
Accountability provided a benchmark, but no
guidance for how to get there. As far as David
could tell, American schools still operated more or
less the same way they had for years, just with
more tests.
As David’s student, Seneca had no reason to
think otherwise. And so, on the day in 2005 that
Seneca discovered a case of burgeoning
infrastructure in American education, the
realization struck her like a “lightning bolt.” She
was sitting in an auditorium in Washington, DC,
with hundreds of other Teach For America
alumni, there to celebrate the organization’s
fifteenth anniversary (at that point, TFA had
trained more than ten thousand corps members),
when the speaker on stage cracked a joke—some
inside line about TFA—and the whole audience
laughed. A perfectly banal moment. But to
Seneca, it was a revelation: the way a group
representing fifteen years of alums laughed with
one voice.
“I had just been reading about hallmarks of a
profession in other professions,” she said, “and
thinking about this lack of common language
thing, and our inability to communicate in any
substance or depth.” But here was a room full of
strangers—education professionals of varying
levels of experience—all laughing at the same
joke. That was the lightning bolt. “Ohhhh!” she
thought. “Teach For America!”
Teach For America was the organization that
had introduced her to the coherence problem, but
since then, it had built an impressive community
for corps members, with shared jokes and an
expanding curriculum. And this was nothing
compared to what its alumni were building—a
new parallel system of public education,
composed mostly of charter schools, and
increasingly speaking the common language
spelled out in Doug Lemov’s taxonomy.
Seneca’s friends had told her all about this new
entrepreneurial world. The school names sounded
like Boy Scout badges: the Knowledge Is Power
Program, Uncommon Schools, Achievement First,
Aspire Public Schools. But until that moment, she
had marveled at their work—the strangely
deliberate chants, the almost militaristic “no
excuses” strictness, the superhuman, TFA-
powered workforce of mainly childless young
people—without thinking much about what it all
meant.
Now she beheld the entrepreneurs in a new
light. Seneca had been studying the infrastructure
for teaching by examining its absence. Was it
possible her old friends were actually building a
real-life version of it?
David loved working with Seneca. Like many of
his best collaborators, she played Pooh to his
Eeyore, balancing out his perpetual cynicism with
an unblemished faith in the possibility of change.
But when she came to him with her idea about
TFA and the no-excuses entrepreneurs, he was
incredulous. “I would be shocked—shocked,
shocked—if any of these organizations were doing
something different,” Seneca remembers him
saying.
There were many reasons to doubt her
hypothesis—and not just the fact that David had
spent his entire career documenting the factors
that made a spontaneous emergence of American
educational infrastructure unlikely. For one, TFA
and the no-excuses schools seemed to be
distinguished not by how they worked, but with
whom. TFA famously recruited high-achieving
college students, many of them Ivy Leaguers, the
idea being that they were recruiting not just
teachers, but a new generation of leaders to solve
educational inequity. Critics, however, argued that
this merely layered smart people on top of a
broken system rather than creating fundamental
change.
Moreover, the program seemed to eschew craft,
sending bright young people into classrooms with
just five weeks’ summer training. Brand-name
diplomas and high hopes, these critics said, could
hardly make up for a lack of training and
experience. Indeed, even the most optimistic
researchers found that corps members had an
uneven impact on students’ learning. Studies
found positive effects on students’ math
achievement—the equivalent of about a month of
extra instruction, by one estimate; more than two
and a half, by another. But when it came to
reading, the best you could say was that corps
members did no harm.
There was also the fact that, after years of
searching, the only contemporary cases of a
strong, coherent educational infrastructure that
David had ever seen were outside the United
States. In countries like Japan, France, and
Singapore, national education ministries were
strong enough to write clear curricula, and the rest
of the educational machinery—education schools,
textbooks, test makers—aligned itself accordingly.
As for his own country, the historic ambivalence
about federal power made consensus impossible.
“The dispute has deep roots,” David wrote in a
paper with Jim Spillane, “it would be astonishing
if it were settled easily or soon.”
David had found only two American programs
that came close to infrastructure—a pair of whole-
school reform projects called America’s Choice
and Success for All. Although they did offer real
curricula, plus matching tests and teacher training,
the projects had to implement their programs on
top of the existing school system, which snarled
their efforts.
But America’s Choice and Success For All
were promising. And if Seneca’s hypothesis was
correct, then TFA and the charter networks might
represent the next evolution of their model.
Operating outside the traditional district system,
but not depending on it, charter schools could
build infrastructure from scratch. The
presumptuousness of the entrepreneurs nauseated
some of David’s colleagues—the way these
arrogant, self-righteous “movement” types
marched into public education like they were the
first to discover its dysfunction, and especially the
way they dismissed their predecessors as part of
the problem. Nevertheless, the entrepreneurs had
figured out a way to build from scratch, and that
was undoubtedly an advantage.
Another problem was that the movement was
tiny, especially compared to the vast U.S. school
system. In 2005, the number of TFA corps
members and alumni working in schools totaled
only about seven thousand, less than 1 percent of
the 3.6 million teachers then working in all U.S.
schools. And that year, a million students were
enrolled in charters, compared to almost forty-
eight million in traditional public schools. And
only a fraction of the charters subscribed to the
learning culture that Doug Lemov and his
colleagues were shaping. Even so, David wanted
to believe that infrastructure was possible. His
advisee was likely to be disappointed by her
research. But why not try and see?
Seneca began with a “really motley crew” of
interviews with no-excuses teachers from her
personal network. By the end of her research,
she’d conducted formal interviews with forty-one
of them. She also analyzed the new projects that
Teach For America had undertaken since she left
it. “I kept this as an empirical question: Were they
actually doing something different?” she says. The
deeper she looked, the more she felt that they
were.
One project was not unlike Doug Lemov’s
taxonomy. After two decades of observing,
interviewing, and surveying the organization’s
most effective teachers, Teach For America had
distilled their common attributes into a framework
it called Teaching as Leadership. The framework
had become a touchstone—the organization’s
“intellectual centerpiece,” in official TFA-speak. It
informed everything from TFA’s recruitment of
new corps members to its expanding efforts to
train teachers, including reliable access to the
mentorship opportunities that had arisen so
haphazardly for Seneca.
Seneca found herself especially drawn to the
work of the charter networks. Selecting one
network for closer study, the Achievement First
charter schools in New York and Connecticut,
Seneca found further evidence of a developing
infrastructure. Achievement First employed a
standard “Cycle of Highly Effective Teaching” to
structure teachers’ work, from setting goals to
planning units and lessons and then revising to fit
students’ evolving needs; “scope and sequence”
documents for each major subject that outlined
periodic learning goals; regular low-stakes tests,
called interim assessments, designed to help
teachers diagnose how their students were
progressing; an “Essentials of Effective
Instruction” document naming the twenty-four
elements required to teach well, in the
organization’s view; and a series of trainings
directly connected to the rest of the infrastructure,
including weekly jugyokenkyu-style sessions for
teachers.
The program offered Achievement First
teachers not only standards, but infrastructure; not
only support, but coherent support. The cycle of
highly effective teaching mapped onto the scope
and sequence documents, the scope and sequence
documents mapped onto the interim assessments,
and the regular professional development sessions
were designed to support the entire structure, from
“data days” for studying interim assessment
results to a formal coaching system pairing every
teacher and principal with someone chosen
specifically to help them improve.
Infrastructure meant that, where others had four
walls and a locked door, Achievement First
teachers had a herd. At her old school, an
Achievement First teacher told Seneca, the
attitude was “do what you will.” When she didn’t
know how to teach a certain topic—she
particularly struggled with math—she resorted to
the only available recourse. “Guys,” she would tell
her students, “let’s just do multiplication again.”
Like most American teachers—like Seneca herself
back in San Jose—she was on her own. At
Achievement First, meanwhile, the teacher had a
framework around which to build each day’s
lesson; prewritten interim assessments to get a
regular peek inside her students’ minds; special
designated data days to dig into the results; and an
army of colleagues to help her think through the
problems that teaching presented every day. An
elementary school teacher, she did not have to
plan lessons for all four of the core subjects on her
own. Instead, mimicking a common practice in
Japanese elementary schools, Achievement First
teachers shared the planning work among “grade-
level teams,” with a more experienced teacher
taking responsibility for the toughest lessons,
while novices handled the easier plans. When she
really struggled with a particular topic, the teacher
told Seneca, her coach would literally stand in
front of her classroom with her. “Look, this is how
you’re supposed to do it,” the coach would say.
With each new dispatch from Seneca’s field
notes, David dialed back his skepticism.
“Achievement First was building a new education
system,” he says. “And that was very exciting . . .
Here was this organization that was building a
version of what America has never—or hardly
ever—had.”
Seneca had discovered a system through which
teaching might be improved. And not just for one
or two gifted teachers, but for a whole school
district.
When Seneca published her dissertation, in 2012,
her final assessment of Achievement First
(abbreviated AF in her paper) was
overwhelmingly positive. “AF provides a rich and
generative new model for thinking about what
organizing for quality teaching and learning might
look like in the US context,” she wrote. But she
also described challenges that could impede the
organization’s success. One had to do with the
model’s ability to be replicated. The network’s
infrastructure could easily grow to serve more
students. But another key factor—Achievement
First’s unusually talented and hardworking staff—
would be harder to extend. Another challenge was
the problem of rigor.
Like Rousseau Mieze’s old principal, Stacy
Birdsell O’Toole, some leaders at Achievement
First worried that their students’ learning was too
superficial. This was especially true in math.
Students did well on state standardized tests—
even better on the math tests, in fact, than on the
English ones. But did they really understand?
Would they continue to learn and eventually be
prepared for college?
“I’m watching them count objects to add them
together,” a third-grade teacher told Seneca in one
of her interviews. Given a set of five, the students
had to count, “one, two, three, . . .” when they
should have just been able to look and know, five.
Given a set of ten, with five objects on one row
and five identical objects lined up right beneath,
they couldn’t just count the first row and double it.
They counted the second one too: one, two, three, .
. . nine, ten.” “I think all of those should be
mastered and solid by the end of second grade,”
the teacher told Seneca, “and they’re not.”
The teacher felt just as culpable as the second-
grade teachers who sent students to her
unprepared. In third grade, students were
supposed to master fractions, but she never found
enough time to get them to really understand.
They moved on to fourth grade able to master the
state test, but with very little idea of what kind of
number a fraction actually represented.
Seneca thought that the superficiality might
have to do with the network’s emphasis on
posting strong results on state tests. The tests set
relatively low bars for learning, emphasizing a
wide but shallow set of skills instead of a steady
progression of deeper understandings. Many
teachers felt that the tests didn’t measure the
higher-order skills children needed to reach and
succeed in college. One academic dean told
Seneca that while she and her teachers preferred a
curriculum that emphasized deeper understanding
—a TKOT-like math textbook called
Investigations—the curriculum clashed with what
students needed to master in order to do well on
the state math test, and, therefore, with the “scope
and sequence” Achievement First had written to
prepare them for it.
 And for a network of charter schools, test scores
mattered. Scores determined not only whether the
schools stayed open, but also whether the
infrastructure continued to receive the support of
the private donors who looked to the tests as
indicators of a school’s success. The same forces
that enabled Achievement First to build a strong
infrastructure for teaching had also conspired to
make the teaching within that infrastructure utterly
conventional.
Seneca observed a striking contrast between
how Achievement First taught its teachers and
how it taught its students. Working with their
designated coaches, teachers focused their studies
on only one or two learning goals at a time. But
with students, even when teachers wanted to
spend more time on a single, complex goal—like
understanding fractions—the state test thwarted
them. Taking more time for fractions, after all,
meant taking less time for another unit, like
money. And when it came time for the test, one
educator told Seneca, “you know that you’re going
to have three questions on money and if you never
got to the money unit . . . you’re not sure how your
kids are going to do.”
Having some infrastructure was certainly better
than having none. But the big question remained.
Could an American infrastructure support high-
level teaching at scale? Thinking over that puzzle,
Seneca found herself turning not to David but to
David’s wife, Magdalene Lampert.

Both Magdalene and David had been despondent

in the years after fleeing Michigan State. For
David, the disappointment was an academic
problem, another educational failure to analyze.
But Magdalene had given her entire self to the
work. At Michigan State, she had made herself
the template for both the new math teacher and the
new ed school professor. The national reform
groups had held her up (along with her star
protégé, Deborah Ball) as the American exemplar,
the one who demonstrated what was possible.
When first the Michigan State experiment and
then the math reforms failed, one after the other,
Magdalene felt that she had failed too.
A moment in 1991 had exemplified the pain.
Asked to present to a commission advising
President George H. W. Bush on new education
standards to take effect by the year 2000, she had
gone to DC with what was by then her usual spiel.
To highlight the changes that both teachers and
students would have to undergo, she described a
day in her fifth-grade classroom at Spartan
Village. She did not sugarcoat the difficulties.
“Unfortunately,” she said, “very few Americans—
and remember that American teachers are only a
subset of Americans—have any idea what a
mathematical community is or what a conjecture
is or what it would look like to do mathematical
reasoning. Most of us have never done that.” She
went on, “The goal that all students by some year,
whether it’s 2000 or 2061 or whatever year you
want to pick, are going to be able to do this thing
that most Americans have no sense of right now,
let alone many teachers, seems like a rather
ambitious goal.”
She was not saying that the country shouldn’t
try. She was simply drawing the obvious
conclusion. Change—real change—would require
a lot of learning, a lot of support, and a lot of time.
But instead of embracing these challenges, the
commission attacked her. “Let me use an
analogy,” said the chairman, Roy Romer, the
tough-talking, square-jawed governor of Colorado.
“In Desert Storm, when the president wanted to
move, he called in the generals from Saudi Arabia.
You know, flew them in. He said, give me a
plan.” But the plan they developed called for
spending way too much time on the ground—a
whole year. So the president demanded another
plan, and what did the generals do? “They came
up with another plan that did work.” The
implication was clear. If the work as Magdalene
imagined it would take too much time, then she
needed to imagine something else.
Later, when she published the research that had
informed her testimony to Romer—a detailed,
five-hundred-page book documenting a year inside
her classroom at Spartan Village called Teaching
with Problems and the Problems of Teaching—
the public’s reaction echoed the governor’s. “The
response initially was, this is way too
complicated. If it takes [five] hundred pages, you
know, like—this is amazing what you’re doing
here, but any novice teacher is gonna read this and
say, you know, I’ll never be able to do this,”
Magdalene says. “And any experienced teacher is
going to say, I don’t have time.”
She had dedicated her career to making her
work accessible to other teachers, but still readers
rejected the work as too hard, too complex,
impossible to scale. When her sabbatical year
arrived, she picked a destination as far away as
she could imagine: Rome, where her plan was to
spend three months learning Italian. “What I
needed was just some time off from being in the
center of this controversy,” she says.
But after just a few weeks at Italiaidea, her new
language school, Magdalene found herself
thinking about teaching. She couldn’t help it. Her
Italian class felt eerily familiar. First, there was
the way the class was structured, always starting
with some kind of problem (how do you order off
a menu? or how do you make a complaint
politely?); then moving onto hypotheses made by
students, which the teacher wrote up on the board;
and finally, ending in a discussion. “In our very,
very halting Italian—because you didn’t speak a
word of English in that school from day one—we
were meant to consider each hypothesis and talk
about why it made sense or didn’t make sense,”
she says. “And I’m thinking, wait a minute. This
is how I teach math! And even though I didn’t
want to be there as an educational researcher, I
started thinking, where did they learn to do this?”
The question struck her again one day early in
the course, when, coming back from a break, she
noticed that the teacher had rerouted the group.
Instead of the task they’d been working on, they
now found on their desks photocopies of a
newspaper article along with an assignment to
underline all the personal pronouns. The change
was clearly strategic. Before the break, the teacher
had been walking between their desks, observing
the students’ work. He’d seen that many of them
were struggling with the same challenge, the
placement of personal pronouns in relation to
verbs. The new assignment helped them correct
the misunderstanding.
In between Italian classes, Magdalene had been
sitting in Roman cafés, editing the page proofs of
her book, which devoted considerable space to
defining this precise activity—“the work of
teaching while students work independently,” she
called it. “And again I thought, that’s really pretty
complicated. Where did these tasks come from? Is
it only my teacher who’s doing this? Did he read
my book?”
At first, she asked no questions. After all, she
was there to learn Italian, not to do research. And,
given Italiaidea’s no-English rule, she couldn’t
have posed the questions even if she’d wanted to.
“First of all, I didn’t know how to ask,” she says.
“And secondly, I don’t know how I would have
understood the answer!”
But as the weeks went by, she noticed patterns.
Students at Italiaidea transitioned to a new teacher
every month. Magdalene took note of each
instructor’s style. The routine stayed the same—
the same lesson structure, the same teaching while
students worked independently, the same habit of
regularly revising the plan midclass. Her first
teacher was not an exception; at Italiaidea, he was
the rule.
The implications for her old work were
profound. The students at Italiaidea had all signed
up for the class voluntarily, but they had a variety
of academic and class backgrounds. So did the
teachers, a mix of full- and part-time, experienced
and novice teachers who were not particularly well
paid, hardworking, or reform-minded. If a modest
Italian language school in Rome achieved
routinely high-level teaching, why couldn’t
American schools?
After her sabbatical ended, Magdalene returned
to Rome, studying the school deliberately this
time, as a researcher. Italiaidea’s success, she
learned, depended not just on an impressive bank
of resources—the newspaper article on personal
pronouns was just one of hundreds. It also
depended on a teacher education program that
carefully trained teachers in the school’s special
method.
Before Magdalene went to Italy, she knew that
what David called “coherence” was technically
possible—that countries like Japan had it, and that
it enabled them to teach at a high level in large
numbers. But until she saw Italiaidea herself, she
didn’t really understand what coherence could
mean for her and for American schools. John
Dewey’s fantasy about preventing “waste and
loss” because “the successes of [great teachers]
tend to be born and to die with them” was within
reach. Great teaching did not have to perish with
the teacher; the right system could teach more than
just one person to do it, without sacrificing any
complexity in the process.

When Seneca told Magdalene Lampert about the

work going on at Achievement First, Magdalene
couldn’t help but notice the parallels to Italiaidea.
Like the language school, Achievement First had
created a detailed set of resources that teachers
could share, plus an organization designed to help
teachers as much as students. Both organizations
had created infrastructure. The only difference was
the level of teaching. Achievement First had all
the supports she’d seen at Italiaidea, but less of
the TKOT-like rigor.
Magdalene discovered Seneca’s research in
2009. Around the same time, her colleague
Anthony Bryk, a sociologist, invited Magdalene to
speak at a conference hosted by the NewSchools
Venture Fund, a philanthropy whose donations
had helped build the entrepreneurial education
world. (Out of $248 million that NewSchools
invested between 2012 and its founding in 1998,
Achievement First received over $6 million;
Uncommon Schools, more than $7 million; and
KIPP, more than $6 million.) The conference was
meant to target one of the charter networks’ latest
challenges—a “pain point,” in NewSchools-
speak.
In working groups, conversations, and board
meetings, the entrepreneurs all described the same
problem. Some called it “human capital.” Doug
Lemov called it the “build it/buy it” challenge.
Either way, the point was the same. In their early
days, the charters had hired the best teachers they
could find. But as they grew, they could no longer
rely on recruiting the cream of the crop (“buy it”).
They had to “build it”—to teach their teachers
how to teach. The conference was one of a series
of events designed to launch a new “portfolio” of
funding for projects tackling the training problem.
From Magdalene’s perspective, speaking at the
NewSchools conference was a risk. She knew the
entrepreneurs might lump her together with other
ed school professors—as part of the problem. But
after talking with Seneca about Achievement
First, she was intrigued to meet them. Plus, as
Tony Bryk pointed out, not only had she spent
most of her career thinking about the best way to
train teachers; after her research in Italy she was
building a new model at the University of
Michigan that had her more optimistic than ever.
A summer program for teachers, it built on the
ideas she’d learned at Italiaidea’s teacher training
school. She told Bryk yes.
 At the conference, some of the entrepreneurs
displayed the bluster that Magdalene feared,
making their disinterest in her work clear. But
others were friendly. One stunned her by walking
up and citing her research. “I read your book,” he
said, meaning the five-hundred-page tome
describing a year inside her classroom at Spartan
Village. “It was awesome.”
The conference-goer’s name was Jesse
Solomon. Jesse, Magdalene learned, had taught
math in the Boston public schools before teaching
at one of the city’s original charter schools, City
on a Hill. There, after running into his own
version of Doug Lemov’s build it/buy it problem,
he’d started a program called the Teachers
Institute to help prepare new teachers. The
institute had since grown to serve not just City on
a Hill, but the entire Boston public school system.
Its new name was BTR, for the Boston Teacher
Residency.
Eight years into running BTR, Jesse’s
challenge, like that facing Achievement First (and,
it turned out, most of the other entrepreneurs at the
conference), was academic rigor. For Jesse, the
problem was especially pressing. A professor at
Harvard had just completed a multiyear study of
the Boston Teacher Residency, and the results
were disturbing. Although BTR graduates were
more likely to continue teaching than were their
counterparts from other programs, they were no
better at raising students’ test scores in English
than was the average Boston teacher with the
same amount of experience. And in math, they
were worse. Something had gone wrong, and
Jesse needed to make a change.
After watching Magdalene’s presentation on
Italiaidea, he made the mental leap Tony Bryk had
been hoping for. He realized that BTR needed
Magdalene Lampert.
 9
THE HOLY GRAIL

Jesse Solomon was especially taken with an idea

that was at the core of Magdalene Lampert’s
findings about Italiaidea: the concept of
instructional activities, or IAs.
The hard part about getting teachers to teach at
a high level, Magdalene explained, was not to
sacrifice complexity for the sake of accessibility.
This was not so different from the challenge of
helping students who knew very little about math
to nevertheless grasp the bigness of it—getting
them not just to memorize, but also to reason,
conjecture, prove, and understand. It wasn’t easy,
but with a well-chosen problem a teacher could
make the subject’s big ideas come alive, even to
little children. The same held for teaching: focus
on only the simpler parts of instruction, and
teachers would learn only superficial techniques.
The trick was to get new teachers teaching
rigorously right from the start.
At Italiaidea’s training school, a place called
Dilit, teacher-educators divided Italian teaching
into fourteen core instructional activities. Each IA
was like a rich math problem. Even a brand-new
teacher could try it out, it could be adapted across
any grade or competency level, and it was both
accessible and rigorous.
Take one IA, called “Conversation Rebuilding.”
In the classroom, the routine resembled a game of
communication Pictionary. The teacher began by
pantomiming a conversation, using only gestures
and drawings. She spoke no words. Then, letting
the students speak only Italian, she invited them to
imagine what had been said. What would a person
ordering a meal in a restaurant say to get the
waiter to bring over a wine list? Once the list
arrived, how might the person respond, if she still
wasn’t sure what she wanted to drink? As
students proposed hypotheses, the teacher helped
steer them toward an understanding of how the
conversation could have actually proceeded.
(When a hypothesis made sense, the teacher
signaled that by having the whole group repeat it;
when it didn’t, she said “excuse me,” mimed the
act again, and got them to start over.) Along the
way, the students learned not just new vocabulary
words and grammar, but how to feel their way into
the language, to communicate.
Steering the conversation demanded a
complicated set of maneuvers for a beginning
teacher, but Dilit made it easier by spelling out the
steps and having the teachers try them out, first
with the professor and other students acting in the
role of students—the teacher-educator throwing
out common student errors—and then with real
students. During rehearsals, Dilit teacher-
educators gave live suggestions in the middle of
an activity. They reminded new teachers not to
forget key pieces of the IA, like always making a
student repeat a correct hypothesis. Later, as the
trainee teachers became more advanced, they
learned to analyze students’ thinking, ignore likely
diversions, and guide students toward increasingly
accurate responses. By the end of a session, the
whole group began to chime in, giving each other
reminders and suggestions about how to proceed.
Before she began working in Boston,
Magdalene had been trying to do the same thing at
the Summer Learning Institute she had built in
Ann Arbor. IAs, she said, were like “containers.”
They let new teachers learn what they needed to
know. One piece of that learning consisted of
practices, the actions required to help children
learn. In the “Choral Counting” IA, for example,
teachers learned how to lead a group of students in
counting aloud by a particular number (tens, say,
or twos); how to write the sequence on the board
(for instance, using columns, so that, counting by
twos, 2, 12, 22, and 32 would sit side by side,
helping the children see a pattern); how to stop the
count at a deliberately chosen number to ask a
question, like “What’s the next number?”; how to
help the students look for patterns; and, finally,
how to facilitate a discussion leading to the key
mathematical idea.
The IAs also let new teachers work on the core
math knowledge they needed to teach—the stuff
Deborah Ball and Hyman Bass had named
“Mathematical Knowledge for Teaching,” or
MKT. (In Choral Counting, this included an
understanding of mathematical patterns, the
common ways students come to understand
numbers, and representations that teachers could
use to advance the students’ understanding.)
Following the Dilit model, the Summer Learning
Institute had groups work together between
rehearsals to give each other feedback and
contribute ideas and techniques to improve
students’ number sense. Finally, they tried the IAs
with local children enrolled in the summer
program.
 Jesse Solomon wanted to incorporate IAs into
the Boston Teacher Residency. Within a year, the
rest of the BTR leaders had traveled twice to
Michigan to watch the Summer Learning Institute
in action. By 2011, they had asked Magdalene to
come work with them full time to redesign the
entire BTR program.
The Boston program posed greater challenges
than the summer program in Ann Arbor.
Residents were placed in some of the city’s most
difficult schools, including several “turnaround”
schools, representing Boston’s portion of the
country’s five thousand worst-performing middle
and high schools. While the students Magdalene
had worked with in Michigan were racially and
culturally diverse, the students in Boston were
more likely to be impoverished, more likely to be
new speakers of English, still learning the
language, and more likely to struggle with
learning and emotional difficulties (20 percent of
students in the Boston public schools are
classified as needing special education).
And yet, when touring the Boston schools,
sitting in classrooms that were often chaotic and
unruly, Magdalene always had the same thought.
The problems she saw, all the challenges—they
were difficult, but they were also solvable. The
BTR teachers, she knew, could get their students
to learn. They only needed to be trained. So, when
Jesse asked her to leave Michigan and work at
BTR full-time, Magdalene said yes.

Two years later, Magdalene Lampert found herself

standing in the first-grade classroom of a BTR
resident named Sabine Ferdinand, holding up an
iPad to record a lesson that would help determine
whether BTR would award Sabine the
certification she needed to teach in Massachusetts.
Technically, the classroom belonged to Ilene
Carver, a fifteen-year veteran teacher. But over the
course of the year, Sabine had taken more and
more responsibility, becoming just as much the
teacher as Ilene, at least in the eyes of the
students. That day, she was leading an activity
called “Quick Images,” an IA adapted from the
math curriculum used throughout Boston
elementary schools.
The lesson began with Sabine counting down
—“eight, seven, six, . . .”—as the students
arranged themselves on the classroom’s
rectangular rug, sitting on masking tape X’s with
their names written in marker. “Three, two, one,”
Sabine said, pausing patiently between each
number. “The expectation is that you’re in your
rug spot with your pencil.”
Sabine sat in the usual teacher spot: the corner
closest to the door, near the place where the
class’s cloth calendar hung on the wall.
“MARCH,” it said, counting out the days in
bright red and white.
“You’re sitting on your bottoms, crisscross
applesauce,” she said, warmly. “I don’t want to
remind you again. Thank you.”
Then she shifted into the work of the moment.
“Who here remembers early on, when we used to
do Quick Images, with the dots?” Hands shot up.
One squealing student gulped and jumped from
her bottom to her knees in glee.
They reviewed the rules: The teacher flashes
each image only twice, and only very briefly each
time. After each flash of the image, the students
write down what they’ve seen. But while the
image is up, they can only think—no pencils.
On this day they were working on geometry, so
the images Sabine showed wouldn’t be dots to
count, but shapes to draw. “Ready?” she asked.
“One, two, three.” The students took in deep
breaths as Sabine flashed the drawing, rotating it
for everyone to see, Vanna White–style. A perfect
square on a white sheet of paper.
After the second viewing, she pinned the
drawing to an easel on her left. “All right,” she
said. “So my question is, how can we describe this
shape?” There was another show of hands, but
Sabine waited patiently, giving more students an
opportunity to think. “How can we describe this
shape?” she repeated. “Rafael?”
“Uhh,” Rafael said thoughtfully. He was a
heavyset child with a big, nervous smile. “It’s a
square.” Sabine could have left it there—yes, a
square—and moved on to the next shape. But she
continued to probe as Magdalene watched.
“Who agrees with Rafael?” she asked, holding
up a model thumb to her chest to suggest how
students could signal their answer, up or down.
“Stephanie,” she asked. “Do you have anything
you want to add on? What else—how can you
describe your shape?”
Stephanie, an energetic girl who sat in a special
chair on the side of the rug, presumably to keep
her focused, mulled that over. “Um,” she said, and
then announced: “It is not long!”
“Can you say more about that?” Sabine asked.
“What’s not long?”
 “The sides are not long,” Stephanie offered.
Malcolm had his hand up too. “Malcolm, what
do you want to say?” Sabine asked. “It has”—he
paused to count—“four sides!”
They were getting more specific, but Sabine
decided they could say even more. “Now,” she
said, “what do we notice about these sides?”
“They’re medium-sized,” offered Oscar, from
the back.
“True,” she said. “Danica,” she continued,
turning to someone else, “what do you notice
about the sides?”
“I want to add on to what Oscar said,” Danica
said thoughtfully. Magdalene took note of her
phrasing—here was a first-grader engaging in a
mathematical discussion, in the classroom of a
first-year teacher. It was remarkable.
Danica went on. “They are large,” she said,
“but on the top of it—they are large—on the sides,
they’re even longer.” Danica’s description wasn’t
accurate, of course; the shape was a square, and so
the sides were actually equal. But she was the first
in the class to compare the lengths at all, and that
pulled the discussion in an important direction.
Sabine repeated the observation, pointing to the
drawing on the board: the sides, Danica was
arguing, were even longer on the left and right
than they were on the top and bottom.
“Interesting,” Sabine said.
The students were treading into significant
mathematical territory, and at just the right
developmental moment. By the spring of first
grade, these students clearly grasped the difference
between broad geometric categories, like triangles
versus squares. But grappling with finer
distinctions—the difference, for instance, between
a right triangle and an isosceles triangle, or in this
case, between a rectangle and a square—proved
more of a challenge. The children’s ideas
suggested they had some understanding of the
difference (the relationship between the sides), but
they were struggling to describe it. Were the left
and right sides longer than the top and bottom
ones, or were they the same?
The discussion now moved into more important
terrain: the shift from what the psychologist Jean
Piaget called “animism”—the idea that objects
have their own consciousnesses, like people, and
so can move and change, rather than staying
stable and constant—to a more abstract
understanding of a square as a category describing
shapes with four equal sides of unchanging
lengths. For children still thinking animistically, a
square was not a solid and permanent fact, but an
object that could decide to expand in any direction
if it chose to do so.
Not long after Danica’s misguided comment
about the square’s longer vertical sides, a girl
named Luisa made the observation that tugged the
class in the correct direction. The shape, she said,
actually had four equal sides.
“Luisa,” Sabine had asked, seizing the
opportunity to underscore an important point,
“why were you sure to say equal sides? What does
that mean?” Clarifying for the rest of the class’s
benefit, Luisa had replied that equal meant “the
same as.”
Yet some students still seemed confused.
Building on Danica’s comment, Oscar, the eager
boy in the back, had propped himself up on his
knees to share an idea. “If you put it a little up,
and a little up,” he said, motioning to show how
he could move the top and bottom sides farther
apart, then, he explained, “it would be longer,
because you’re putting it a little upper.” But while
the right and left sides then would have to get
longer, the top and bottom would become shorter,
he said.
“So are you saying if we were to squish this
shape?” Sabine asked, to confirm. “Yeah,” he
said, nodding.
One purpose of studying math in school,
Magdalene knew, was to help children wrestle
with just the ideas Oscar was working on. Could
squares really squish themselves? Or were they
more stable than that?
“But,” Sabine asked Oscar, “what did Luisa tell
us about this shape? Malcolm?”
“She said it’s four equal parts,” Malcolm said.
“Equal parts,” Sabine repeated. “Equal sides.”
As a more experienced teacher, Magdalene
knew that there were other ways Sabine could
have helped the students work on these ideas. To
get them closer to grasping the distinction
between a rectangle and a square, for instance, she
could have used Danica’s incorrect observation
about the sides as an opportunity. “I could have
pulled out a picture of a rectangle and said, ‘Are
these the same or are they different? In what ways
are they the same and in what ways are they
different?’” Magdalene says.
But that kind of response takes longer to
cultivate (starting with learning to recognize what
all this strange squishing was about—no simple
thing), and she didn’t expect a first-year resident
to figure it out on the fly. What Sabine was doing
was exactly what Magdalene hoped for. Using a
routine she’d rehearsed many times before, she
was able to keep the students focused—and, at the
same time, to listen to their math. Her preparation
had helped her learn not only what to do (the steps
of choral counting) but how to make sense of the
math the students were working on. As a result,
she had gotten the kids thinking about
fundamental concepts in geometry, and when
wrong ideas arose, she didn’t just swat them
away; she put them on the table for the class to
probe. Ultimately, she had managed to elicit the
pivotal idea about four equal sides from Luisa. So
when Oscar brought up his animistic idea that the
shape could “squish,” Sabine could hold it out
against Luisa’s more sophisticated conjecture,
helping Oscar to reconsider what it meant to be a
square.
Even if the students didn’t end that particular
discussion fully grasping the difference between a
rectangle and a square, or the difference between
an abstract shape and a living, elastic object, they
had made important progress. “They’re struggling
with a fundamental concept, and they should be,”
Magdalene said. It was more than could be said of
many first-grade classrooms. It was exciting.
Later, after they finished the Quick Images IA,
Sabine introduced a problem about addition. She
showed the students a puzzle and then gave them
a chart that tracked the blocks she’d used to
complete it—“two hexagons, zero trapezoids, one
blue rhombus, zero squares, three tan rhombuses,
and seven green triangles,” she read out loud.
Now their challenge was to find the total number
of blocks.
Magdalene knew Sabine had selected the
numbers deliberately to add up to a sum greater
than 10 (13). They’d been working on more
complicated methods of adding numbers, like
“counting up” from the largest number or
breaking an unfamiliar problem into familiar parts
or using a number line to skip from one number to
the next. A problem with a solution larger than 10
would nudge the students to try out the new
methods for themselves, rather than using the one
that many of them still preferred: counting on their
fingers.
Indeed, just as Sabine had hoped, the students
walked through all kinds of novel combinations of
the methods they’d been working on. Magdalene
watched as one girl, lying on her stomach, wrote
out two number sentences—3 + 3 = 6 and 6 + 7 =
—and then filled in the empty box by starting
with 7 and then drawing 6 lines:

She caught another student writing out different

sets of number sentences:
Another student, Faith, took a similar approach.
“That is very cool,” Magdalene told her, as Faith
showed off what she’d done:

Later, as the lesson moved from individual

work time to group discussion, Sabine invited a
quiet boy named Kevin to share his strategy.
“I did three plus seven equals ten and then—”
he began, before Sabine interrupted.
“How did you know that three plus seven
equals ten?” she asked.
“Because I knew my combinations of ten,”
Kevin replied.
“So three and seven’s just another combination
of ten,” Sabine repeated, turning to the rest of the
students to make sure they’d understood Kevin’s
strategy. “So you can use what you already know
to help you figure out this problem. So he knew
three plus seven equals ten.” She turned back to
Kevin. “Go on.”
Kevin described his next step. He was left with
2 and 1, and he knew that 2 + 1 = 3, so he added
up the two final sums, 10 and 3. “And you got?”
Sabine asked. “Thirteen,” he said. Sabine added
that to the board, where she was chronicling each
student’s steps:

But before she could summarize the important

point—once again, Kevin had used a combination
he already knew to find the final sum—another
boy blurted out an idea from the carpet. Earlier,
he’d been squirming and talking out of turn. But
now he was interrupting with an observation
about Kevin’s strategy. “If you take away the
zero,” he said, pointing to 10, “and put the three,
it’s thirteen!”
 Sabine looked at Magdalene. A common course
for young children was to see “3 + 10” and
mistakenly add the three and one together, since
the one is closest to the three, getting 4. Then, not
knowing what to do with the zero, and not fully
grasping the difference between the tens and the
ones places, they would write a zero next to the
four: 3 + 10 = 40. This boy, still just a first-grader,
had leapt headfirst into the correct idea: three ones
and one ten meant that the three could effectively
replace the zero: 13. He had intuited place value.
Sabine and Magdalene smiled at each other. “I
see what you’re saying,” Sabine told the boy. And
the class moved on.

Jesse Solomon wasn’t the only entrepreneur to

seek help from the academics. Heather
Kirkpatrick, a leader at the Aspire charter network
in California, came to the Learning to Teach
summit focused on the same problem as the
others. “We looked so good on paper; we were
kind of killing it,” Heather says. “But we all felt
like, jeez, when we walk into the classroom, we’re
not where we want to be.”
What they wanted was rigor—more
specifically, something they called “academic
discourse.” To them, discourse meant four things.
First, adults couldn’t do all the talking (and
therefore all the thinking). Second, the students
had to talk about the academic idea at hand and,
third, they had to talk using academic vocabulary.
Finally, they had to do what Aspire called
“bringing evidence to bear”—quoting the text in
English class, citing a primary source in history,
reasoning through a proof in math, pointing to
experimental evidence in science.
“It was those four things. And we said we
should be able to see that in math, science,
English language arts, history. That is the holy
grail,” Heather says. “Then we said, okay, how do
we get there?”
 Heather was struggling with this question when
someone suggested she talk to Pam Grossman, a
professor at Stanford’s ed school and one of Lee
Shulman’s first students after Lee left Michigan
State for Stanford. Pam had come to grad school
after nearly a decade of teaching high school
English. During her time at Stanford, Lee
dispatched a group of his students to study the
teaching of individual academic subjects; Pam
selected English. Over time, she came to think of
Lee’s students in family parlance. With Lee as
their shared mentor, Deborah Ball and the others
were like Pam Grossman’s academic “cousins,”
working on the same questions, just at different
universities and in different subjects. There was
one other difference for Pam: instead of
experimenting with her own teaching, she studied
other teachers’ work.
One episode, caught on videotape, showed
seven students at a struggling urban high school
in San Lorenzo, California, discussing “The
Yellow Wallpaper,” the short story by Charlotte
Perkins Gilman. Their teacher was not visible in
the video, but the students carried on as if he was
right there, paging through the Xeroxed story in
front of them and even calling on each other to
speak.
“What do you have to say, Jim?” a blonde girl
named Amy was asking the boy across from her
when the clip opened. “My interpretation of this,”
Jim said, “goes back to what Ms. McWilliams”—
his student teacher—“said before we even read the
story, about how it gave her chills.” Jim wore
glasses and had a knack for the theatrical pause.
“And actually, my interpretation of this is that she
was dead from the very end of page thirty.”
The other students looked up from their papers.
The story, written as a series of diary entries by a
woman suffering from anxiety, used the word
dead only once, and that was to describe the paper
on which the woman wrote her diary. But while
the diary did chronicle the woman’s worsening
condition after her husband, a physician, ordered
that she isolate herself from work and society, it
never mentioned that the woman was not alive.
Now Jim was saying she’d been dead since page
thirty. “What?” one student asked.
“That whole conversation” on page thirty, Jim
continued, “the very last line says, ‘I am securely
fastened now by my well-hidden rope—you don’t
get ME out in the road there!’ ” The line described
how the narrator, after feeling trapped behind the
wallpaper of her isolated room, had used a rope to
escape it. Jim continued, “I think at that point,
she’s dead. This is her talking to John”—the
narrator’s husband—“as a ghost.”
“Ah!” exclaimed Jade, a girl in denim who had
been listening quietly, with one eye on the page
and another on Jim. She bolted up, putting a hand
over her mouth. The girl next to her, Sariah, had
her mouth wide open too.
“She’s free in the house,” Jim went on, “but she
is never, like, free—”
 “—OUTSIDE OF THE HOUSE!” Jade and
Sariah shouted in unison, as the rest of the group
talked over each other in an excited rush.
But Amy, the one who had called on Jim in the
first place, wasn’t buying Jim’s idea. “So then,
wait, wait,” she said. She pointed to another
passage that didn’t seem to fit Jim’s story. The
line about the “securely fastened” rope was
preceded by another, speculating that other
women might have made the same escape as the
narrator, fleeing from behind the room’s
oppressive wallpaper.
“Right there,” Amy said, “it says, ‘I wonder if
they all come out of that wallpaper as I did?’ ”
How did that description jibe with Jim’s
interpretation? “Does that represent the people
who died just before her, or something? Other
people who’ve died?” A few students attempted
an explanation before Jim spoke. “I wonder if it
sort of represents society,” he said. “Because she’s
freeing herself, and she’s wondering, are all the
other women doing this too?”
“So her way of freeing herself was killing
herself,” Amy replied, repeating his point. You
could connect the two passages, Jim was saying,
by interpreting all the women’s escapes from
behind the wallpaper as suicides.
Soon Jade had a question. “But what about this
house?” she asked. “This house! This house! This
house has to represent something too.” Amy took
the opportunity to offer her own interpretation of
the story. “Maybe,” she said, “maybe the house
and the area can represent life, right? There are
parts of life, places in life you want to go, things
you want to do, right? She was talking about that
one room she wanted, but her husband said no.”
Jade nodded. She was persuaded by this
interpretation too. “The different rooms could be
different lifestyles!” she said, jumping in. “Or
different things she can or cannot do,” Amy said.
“Or,” Jim said, quietly, “different parts of her
life.” “Yeah,” Amy said, pointing at him with her
pencil and nodding. They had different
interpretations, but they were on the same page.

The video showed just the kind of conversation

Heather and her team wanted to cultivate at
Aspire—a pristine example of “academic
discourse.” The teacher, Peter Williamson, might
not have been on the screen, but, as Pam
explained, his work was all over the lesson. More
specifically, he had set the students up to have a
productive discussion, first having them write out
two types of questions, literal and interpretive;
then having them go over the questions with each
other, getting feedback on how to improve on
them; and finally, after they’d finished talking,
leading a debriefing of the conversation centered
on how they could have gotten even more out of
talking to each other. (That was one reason Peter,
who later became a professor of teacher education
at the University of San Francisco, had videotaped
the session—so that the students could watch it
and think about what to do better next time.)
And these steps were only what Peter had to do
to prepare for the single “Yellow Wallpaper”
lesson. It had taken him more work to get the
students to that point. Eventually, Pam and her
graduate students broke the practice of English
teaching down into key parts—“core practices,”
Pam called them, an English counterpart to
Magdalene’s instructional activities.
Among the practices Pam outlined was
“modeling.” This was a core part of the best
English teachers’ repertoire, a way of walking
students through the processes they needed to
perform in English class—not just reading and
writing, but their component parts, like annotating
a text to help understand its meaning or using
evidence to construct an explanation. To teach
students each part, a teacher not only had to walk
them through what, for instance, an annotation
looked like (here are my highlights!) or show
them an explanation (this sentence right here!);
she also had to break the activity down into its
invisible mental steps. Pam called this “making
your thinking visible.”
Modeling worked best on texts that resonated
with the students. Even better, the teacher could
use the students’ own work. For instance, a
teacher might take the draft of a student’s
persuasive essay and use it to model, say, the
writing of explanations, walking step by step
through the evidence and narrating how a writer
might think about using that evidence to support a
point. “ ‘So what?’ is the question I need to ask,”
the teacher could tell the students. “And my
answer should tell you . . . oh, that’s why this
evidence is so important!”
Another category in Pam Grossman’s taxonomy
of English teaching, “classroom discourse,”
helped teachers work with students on their ideas.
A classroom discussion shouldn’t operate as a
floating alternate reality, the entertainment before
the real work begins. At their best, discussions
were the first step in the writing process, verbal
editing sessions in which students worked
together to sharpen their ideas. What did the text
literally mean? What did it mean symbolically? If
the discussion went well, then, by the end of the
lesson, a classroom full of bland observations
would transform into thirty well-articulated
interpretations.
When Pam started grad school, scholars of
English teaching had written about the importance
of discussion in a literary class. Studying
American classrooms, they had also discovered
how rare it was. But few had thought about how
to help teachers do more of it and do it better. Pam
and her students “decomposed” discussion,
breaking the practice down into teachable parts.
What they found was that great discussions did
not happen by accident. They required serious,
deliberate preparation. One teacher whose practice
Pam studied—Yvonne Divans Hutchinson, who
taught in the Los Angeles Unified School District
at a high school lodged between the
neighborhoods of Watts and Compton—handed
her students detailed lists of what she called
“stock responses,” possible ways of participating
in a discussion, including half a dozen alternatives
to a shrug:

• You don’t know the answer? Try saying, “I

don’t know, but I will try to find out the
answer and get back to you.”
• You didn’t come prepared to talk? “I regret
to say that I am not prepared.”
• You didn’t understand the question? Just
ask, “Would you please repeat (or restate) the
question?”
• You did the homework and understood the
question, but still couldn’t come up with an
answer? How about, “Please come back to
me; I’m still thinking.”

The stock responses might seem forced, but

without them there was no guarantee students
would talk at all. “You’re not born with a gene
that tells you how to talk about Beloved,” Pam
says. “Actually, that’s something you learn to do
over time, and there are things that teachers can do
to make kids successful.”
Yvonne also wrote out suggestions for ways of
making a contribution. To disagree, first say so: “I
respectfully disagree.” Then give your opposing
idea—“and justify it.” To agree and then extend,
say, “I want to add to what (person’s name) said.”
Other rules added to the class’s discursive
repertoire. To make sure a wide variety of students
spoke, and to increase the likelihood of getting an
answer, Yvonne took advantage of peer pressure
and had students call on each other. Asked for a
response by a peer rather than the teacher, she
found, teenagers were more likely to comply. She
also prepared for the case of a student with
nothing to say. She often reminded her classes to
pause, giving the students Yvonne called
“reticent” more time to put their thoughts together.
And through her modeling, the students learned
how to coach each other too, coaxing contributions
from even the quietest peers.
That was just the beginning. In Yvonne’s class,
every discussion began with an “anticipation
guide,” a list of questions designed to get the
students thinking about subjects covered in a
reading before they began it. Next came a
“reading response prompt” that each student
answered individually, complete with reminders
about the best way to read—“mark up the text in
the way you choose,” the instructions said,
“including the use of highlighters and
metacognitive marking”—and instructions asking
the students to write questions of their own. (In
Yvonne’s class, even questions had a careful
taxonomy, from basic factual “right there”
questions [level one] to “global” questions [level
three] that took a text’s substance and expanded
beyond it; for one prompt, the students were to
write two level-one questions, three level twos,
and one or two level threes.) Finally, in the
minutes before the whole-group discussion, they
held miniconversations in small groups. “If they
come to the work with their own frame of
reference, then they’re much more apt to be
engaged,” Yvonne explained.
The planning got the students to the starting
line, but to pull off a lively and productive
discussion, Yvonne had to teach in the moment
too. Pam and another one of her grad students,
Lisa Barker, used videos from Yvonne’s class and
others to further break down the art of leading a
discussion. Drawing on a term coined by early
scholars of classroom discussions, Pam and Lisa
called one of the practices Yvonne often deployed
“uptake.” A teacher practiced uptake when she
listened to a student’s contribution and then
repeated it in some way, by summarizing the idea
(So her way of freeing herself was killing
herself), elaborating on it, or pushing the student
to do the same.
Working with Pam, Lisa broke down “uptake”
into nine subparts, which teachers used at
different times. “Restatement,” the simplest,
involved summarizing a student’s claim, but this
time adding academic language, such as better
grammar or more precise terminology.
“Revoicing,” a subset of restatement, summarized
a student’s contribution for the even more specific
purpose of aligning it with the particular side of
the discussion it belonged to—Amy is clarifying
Jim’s statement about “freeing herself” in order
to support his interpretation that the woman is a
ghost.
Other moves had the teacher directly pushing
students for better contributions. A “challenge”
move responded to a claim by taking the opposite
stance, just for the sake of argument. “Press”
asked the speaker for more information—evidence
of a claim, maybe, or clarification of meaning.
“Post” held up a student’s claim and solicited
comments on it—Who thinks they can articulate
what Jim is trying to say?
Not only had Peter Williamson, the teacher who
assigned “The Yellow Wallpaper,” mastered
classroom discourse himself; his mastery served
as a model for his students, who used uptake to
discuss the story on their own. Amy knew to ask
for clarification; Jade knew to repeat Jim’s claim
to make sure she understood it and to press him to
elaborate on the idea when she wanted to
challenge it; and at the end, when Amy put
forward her own interpretation, Jim knew how to
use “uptake” to build on the idea, listening as she
described the house’s symbolic meaning and then
helping her burnish her explanation. The rooms
represented not just “different things she can or
cannot do,” but “different parts of her life.”
Heather Kirkpatrick loved Peter’s video. She
talked with Pam, and in no time, Pam and Lisa
were coming to Aspire to teach a session at its
summer retreat.
One way to think about what the academics
offered the entrepreneurs was “content.” Whereas
the entrepreneurs like Doug Lemov looked at
teaching generically, across all kinds of subjects,
Magdalene Lampert looked only at math and Pam
Grossman, only at English.
But just as important as their content
knowledge, and maybe more so, was the
academics’ theory of learning. Ironically, this was
a legacy of the same academic structure that had
once hindered research on teaching: the close
relationship between education research and
psychology. It was true, as Lee Shulman’s
predecessor Nate Gage had discovered, that the
science of teaching was not simply the inverse of
the science of learning. But the corollary was also
true. It wasn’t possible to understand teaching
without understanding learning.
Perhaps unintentionally, the charter school
educators had adopted a linear model of learning.
Learners, they assumed, started with the basic
fluency skills needed to do what they called
“higher-order” work. In math, that meant
memorizing the multiplication tables before
working on problem solving; in English, it meant
mastering simple vocabulary words before
learning to construct an argument. They thought of
learning as if it were architecture: a fantastic
design was nothing without the materials to build
it. Something complex and beautiful could not be
accomplished without first mastering the
mundane.
The idea that facts laid the foundation for
concepts yielded a basically behaviorist theory of
learning. If learning began with facts, and facts
began with memorizing—because memorizing (or
“fluency”) was separate from concepts (“critical
thinking”)—then the best method to teach
children to learn was not so different from what
Edward Thorndike had hoped to accomplish with
his cats. Practice, practice, practice, with regular
punishment and rewards. The “rigor” could come
later.
The resulting teaching style was especially clear
in the handling of mistakes. In a behaviorist
model, every mistake should be greeted with a
quick and firm correction. Otherwise, students
won’t learn that an idea is wrong. The best charter
school teachers took this maxim seriously. One
math teacher, heralded as one of the best in the
KIPP network, decided never to give his students
chances to practice problems at homes that they
hadn’t already been taught how to solve. The
danger, he explained, lay in the likelihood that,
alone at home without the teacher to stop them,
they would practice doing the steps wrong. Absent
a response that corrected or approved the step, the
mistake might be ingrained in the category of
uncorrected, and therefore accurate, truths.
Several of the techniques in Doug Lemov’s
taxonomy (for example, “Do It Again”) rested on
this belief. Teaching behavior, in the world of the
taxonomy, often boiled down to the imperative of
responding to every visible misbehavior. A teacher
was bound to give a swift and clear correction to
every mistake. Doug applied the idea to teaching
academic content too. Writing about how to teach
children to “decode,” the work of deciphering a
string of letters into a pronounceable word, he
emphasized the importance of letting no error go
unnoticed. “Given the bedrock importance of
decoding at every level,” he wrote, “teachers
should strive to correct decoding errors whenever
possible, no matter what subject or grade level
they teach.” “Punch the Error” was the name he
gave the technique of notifying a student swiftly of
her mistake.
But by the time Pam, Deborah, and Magdalene
started their study of teaching in the 1980s,
research had begun to show the limits of this
behaviorist view. Learning among humans,
psychologists were discovering, was more than
just a sum of experienced stimulus-and-response
yes-no pairs, and concepts didn’t wait for facts to
accumulate; the two were enmeshed together.
The best memorizers, for instance, succeeded by
embedding their object of study within a more
abstract map of big ideas. One psychologist,
studying a college student he called “S.F.,” found
that the student could memorize long strings of
numbers only by attaching the digits to others that
held more meaning. A competitive runner, S.F.
translated numbers into race times; 3492, for
instance, became “3 minutes and 49 point 2
seconds, near world-record mile time.” Just 3.492
wouldn’t have been enough; he also had to place
the number in a context that made sense to him.
After a year and a half of using the racing
mnemonic, the number of digits S.F. could
memorize had grown from 7 to 79. The only cases
where he stumbled were numbers that simply
couldn’t be mapped back to a memorable race.
Children, similarly, learned to add and subtract
through strategies that built on their intuitive
sense of numbers, not what their teacher told them
was correct. Like the Brazilian street children
selling fruit, who managed to make multidigit
calculations in their heads, they counted, grouped,
and regrouped until they arrived at a solution that
corresponded to what they knew about how
numbers worked.
Humans appeared to practice this reasoning
—“critical thinking” or “rigor,” the
entrepreneurial educators might call it—
practically from birth. In experiment after
experiment, psychologists studying infants
showed that they looked at the world not via a
system of rewards and punishments, but through a
web of generalizations, rules, and principles
derived from observations.
In one experiment, psychologists pushed a blue
cylinder down a ramp until it hit a toy bug. Their
six-and-a-half-month-old subjects watched as the
blue cylinder propelled the bug forward, so that it
traveled all the way to the middle of a horizontal
track. Then the researchers rolled down two more
cylinders, a larger yellow one and a smaller
orange one. Predictably, the larger yellow cylinder
knocked the bug farther along the track, all the
way to the end. But the orange cylinder, although
smaller than both of the other cylinders,
nevertheless moved the bug to the end of the track
as well. Presenting the same strange events to
adults, researchers found that they reacted with
surprise. But would infants, who had never been
taught the laws of physics, do the same?
They did. Shown the surprising case of the
smaller orange cylinder that knocked the toy bug
farther than made sense, infants stared longer than
a control group of infants shown a sequence of
events that did not violate physics. They’d been on
Earth for under a year, and certainly had never
been given a gold star for knowing that larger
cylinders have greater mass than smaller ones. But
after taking in the world’s data—all the times that
large objects pushed things farther than smaller
ones—they had come up with the abstract mental
model that made sense.
The takeaway message was not that conceptual
understanding is more important than
memorization; it was that the two are inextricably
enmeshed. Any supposed dichotomy between
them was false. Magdalene summarized the
lesson in a single phrase. Children, she said, were
“sense makers.” Like the babies staring at the
cylinders, they took in data and reasoned about it,
working from their own evolving grasp of how the
world worked. Educators who imagined otherwise
—assuming, for instance, that memorization took
place outside the context of concepts and
principles, or that repeated rewards and
punishments were enough to help a person learn
—did so at their own peril. Children would try to
make sense of rules, even rules that made no
sense. Then, when violations inevitably arose, they
would apply the rules anyway, as the California
teacher who overemphasized subtraction with
regrouping found when her student, told of the
importance of borrowing, began doing so on every
problem, no matter what size the bottom number
was.
Magdalene and Pam and Deborah’s kind of
teaching, TKOT, was more academically rigorous
not because their problems were harder, or their
expectations higher, or their grading curve steeper,
but because their vision of learning was more
refined. Not only had they read the general
research on learning. They had also studied the
specific rules of “knowing”—the epistemology—
for the individual subjects they taught. Each field
had its own specialized definition of what it meant
to know something—of the way, in math,
conjectures built to proofs, or, in literature,
evidence became explanation and finally
interpretation.
Because the definitions were not all the same,
neither was the teaching they argued for. In math,
for instance, the “You, Y’all, We” lesson pattern
popularized in Japan (as well as in Magdalene
and Deborah’s math classrooms) made sense for
structuring investigations of big ideas, like the
meaning of fractions or negative numbers. In
English, meanwhile, where students needed to
learn specific reading and writing strategies—how
to figure out the meaning of an unfamiliar word,
for example, or how to build ideas for an essay—
the “I, We, You” pattern of modeling followed by
guided practice was more appropriate. And within
each subject, different topics could call for
different structures.
Drawing on these very specific traditions,
Magdalene, Pam, Deborah, and their colleagues
had an easier time achieving the academic rigor
that the no-excuses teachers also longed for. They
taught by helping students see the world
differently, pushing their intuitive knowledge
closer to the bank of understandings and rules of
operation that mathematicians (and scientists,
historians, literary theorists, and so on) have
arrived at over centuries. Teaching, in this view,
began with listening. “Part of interacting with
kids,” Magdalene said, “is assessing where they
are and thinking about what experiences you can
give them that will challenge their way of seeing
the world.”
The different approach to learning also led
Magdalene, Pam, and Deborah to take a different
view of children’s mistakes. In TKOT, mistakes
were not worrisome ills to stamp out on sight, but
precious opportunities to begin the longer process
of correcting misunderstandings over time. One
purpose of teaching, in their view, was to draw out
mistakes. The best English teachers, Pam saw,
helped children write better by showcasing real
examples of student writing that needed work—
and then, holding weak models up against the
strong, by describing exactly what students could
do to improve. Magdalene and Deborah, similarly,
built their problems of the day around the goal of
eliciting misunderstandings that could move the
class toward more accurate ideas.
Some Japanese teachers took this notion even
further. In Japan, the portion of the lesson that
Magdalene called “teaching while students work
independently” actually had two competing
names: kikan-shido and kikan-junshi. The first,
kikan-shido, described the act of observing
students’ efforts to solve the problem of the day
and, when necessary, intervening to resolve their
confusion by offering a hint or an extra instruction.
But the second, kikan-junshi, adopted by a
contingent of purists, described observing without
comment. When a student made a mistake or
became confused, the teacher simply noted the
error (maybe on a pad of paper or maybe just in
her head), nodded, and walked on by. An English
translation clarified the difference: kikan-shido
meant “between desks instruction,” whereas
kikan-junshi meant “between desks patrolling.”
Students learned better when they saw the error
of their ideas on their own, the kikan-junshi
purists felt. And the teacher made sure they did
see it, in the course of the “We” part of the lesson,
the group discussion.
In this regard, Doug Lemov seemed conflicted.
On one hand, as the “Punch the Error” technique
exemplified, much of his taxonomy was built
around the eradication of mistakes. Yet Doug also
wrote about the importance of making class a safe
space for errors. Indeed, “Normalizing Error” was
Technique No. 49, the last one in the taxonomy. It
described how teachers could get students feeling
comfortable with mistakes. And so, in Teach like
a Champion, on the same page that Doug
emphasized the importance of fixing errors “as
quickly as possible,” he also called them “a
normal and healthy part of the learning process.”
The tension was much less apparent in Doug’s
work teaching adults, his teacher training. In this
regard, the entrepreneurs and the academics took a
strikingly similar approach. At taxonomy
workshops, attendees practiced techniques in
simulations that looked practically identical to
those used at Dilit in Italy. Doug always
emphasized that teachers should use these
evolving techniques only if they made sense. The
job of administrators, meanwhile, was not to
punish bad performers for poor teaching. It was to
give them opportunities to learn. To teach them.
And over time, even without direct intervention
from the academics, the entrepreneurs’ approach
to teaching children was beginning to bear more
resemblance to their approach to teaching adults.
In 2013, Doug began crafting “Taxonomy 2.0,” a
second edition of Teach like a Champion, in
which he revised large portions of his approach to
error. Instead of focusing on ways that teachers
could eliminate mistakes as soon as they arose, the
new document tried to give them tools to use
errors as learning opportunities, naming new
techniques they could use to help students feel
comfortable making mistakes.
 As Doug’s changing ideas made clear, the
biggest question was not whether the
entrepreneurs’ teaching would evolve over time.
Their teaching already was evolving, even without
the academics’ help. The biggest question was
whether the rest of the country’s teaching would
change too. Academics like Pam, Magdalene, and
Deborah still made up only a minority of ed school
professors. The entrepreneurs, meanwhile, were
growing in number, but by 2011, charter schools
still reached only 4 percent of American public
school students. And although the outside world
was paying more attention to the charters, the
lessons that observers drew didn’t necessarily
reflect the reality inside. Instead, as usual, the
observers focused on the idea that had spawned
charter schools in the first place: accountability.
 10

A PROFESSION OF HOPE

In 2004, as Doug Lemov began thinking about his

taxonomy, Deborah Ball found herself talking
about her future with two of her mentors—David
Cohen and the University of Michigan’s provost,
Paul Courant.
After spending eight years at Michigan State,
Deborah had followed David and Magdalene
Lampert to the University of Michigan in Ann
Arbor in 1996. There, her career had blossomed.
Her videotapes of Sean, Mei, and the rest in her
classroom at Spartan Village had gained a
growing following among academics and math
teachers, and so had her research. She had
cowritten several papers with David describing
the infrastructure problem. And, working with the
mathematician Hyman Bass, she had built on the
corpus of Mathematical Knowledge for Teaching,
MKT, expanding the focus from content
knowledge to accompanying practices that could
be taught.
And that year, another prestigious university
had begun recruiting her for its ed school. She was
discussing her options with David when he asked
her a question. “Something like, ‘What do you
actually want to do?’ Like, not where do you want
to live, but what do you actually want to do?’”
Deborah recalls.
The answer spilled out. “I want to completely
change the way teacher education works in this
country.” Very little had changed over the two
decades she’d been studying teaching. The new
entrepreneurial sector was an intriguing exception,
but it served only a tiny fraction of American
students. The vast majority of kids still learned
with teachers who were as unprepared as Deborah
had been when she first came to Spartan Village.
At university ed schools, aspiring educators still
sat in five-hundred-person lectures like the ones
Lee Shulman had taught at MSU in the 1970s.
Even Teach For America, despite its
improvements, struggled to ensure that all its
corps members entered the classroom ready to
help students learn at a high level.
The classroom was no different. Curricula still
varied from state to state, district to district, even
school to school. Tests still confused rather than
complemented each curriculum—or simply
overrode it. Professional development was
haphazard at best. The David Cohen coherence
problem, in other words, was still alive, well, and
widely ignored: Americans still lacked any
discernible agreement on what students were
supposed to learn, and teachers were still left
alone to help them learn it. The whole education
world lived with the consequences of incoherence
every day, yet the number of people who really
understood what educational infrastructure meant
could be counted on a first-grade pattern block set,
and all of them seemed to work in the same
building in Ann Arbor.
What did Deborah really want to do? She
wanted to build the infrastructure to support
“responsible teaching” (the phrase she preferred
over “TKOT”)—and not just for her students at
Michigan, but for teachers and students all across
the country. By David’s definition, infrastructure
had three key elements: a common curriculum
suggesting what students should study; common
examinations to test how much of that curriculum
they learned; and finally, teacher education to help
teachers learn to teach exactly what students are
supposed to learn. Since her expertise was in the
third category, Deborah thought she could start
there.
Deborah started looking ahead. It was currently
2004. For the 2007–08 school year, nearly two
hundred thousand new teachers would enter
classrooms for the first time, up from sixty-five
thousand just twenty years earlier. By 2011, 3.7
million people would work in the profession. And
the modal number of years of experience of the
American teacher—fifteen years in 1987—was
now just one. If these new teachers were going to
come to the classroom ready to teach, somebody
needed to help them prepare.
At Michigan, Deborah could test a model that
would arm teachers with the knowledge and
practices they needed to teach students well. If the
model worked, she could expand it to the rest of
the country.
David and the provost gave Deborah their full
support. If what she wanted to do was transform
teacher education, they said, “then that’s what you
should be doing.” She decided to stay in Ann
Arbor, where she became director of the
university’s teacher education program, and
received a significant grant from the provost to
reshape the program from scratch. The next year,
she became dean of the school of education.
As Deborah’s project took off—the Teacher
Education Initiative, she decided to call it—her
timing began to look remarkably apt.
Of the three elements of a David Cohen–esque
infrastructure, she’d picked teacher education
partly because it seemed especially important and
partly because it was what she knew best. But
soon, signs of the other two elements—a common
curriculum and assessments to match—also began
to emerge. The driving force were the Common
Core standards, a new attempt to write national
education goals for all American students. The
Common Core standards weren’t themselves a
curriculum, and they weren’t assessments either,
but they paved the way for both. At first, the effort
seemed just as quixotic as Deborah’s Teacher
Education Initiative. Every previous effort to write
national standards had imploded. Close reading of
the master document always set off debate; critics
found too many instances of Harriet Tubman and
too few of Robert E. Lee, or a reading goal
matched to third grade instead of first, and poof!,
the coalition backing the standards would
disintegrate.
But the agitators for the Common Core had
learned several lessons from their predecessors.
They assembled the standards in relative solitude,
avoiding a big public campaign. They deliberately
sought input from states, organizing not through a
federal government agency, but through the
National Governors Association, and thereby
preempting cries of federal intrusion. They also
had the advantage of time; after several rounds of
math and reading wars in the 1980s and 1990s,
certain ceasefires had been wrought. Reading
experts now largely agreed that both phonics
instruction and an emphasis on comprehension
were important for teaching children to read, and a
core group of mathematicians had come to accept
that educators might be onto something with their
“fuzzy math.” Inevitable disagreements remained,
especially in the less organized middle and high
school English community. Nevertheless, less
than two months after the standards were released,
in June 2010, twenty-seven states had vowed to
adopt the standards. By the end of 2013, the
number was forty-five (plus the District of
Columbia). With so many states signing up,
common curricula and common assessments
weren’t far off.
The remaining leg of the infrastructure triangle,
teacher education, was perhaps the hardest to
build. After all, reformers had been trying to
reinvigorate teacher education for decades with
little success. But here again, Deborah’s timing
gave her a unique opportunity. By 2004, when the
Teacher Education Initiative began, researchers
knew more than ever about how to teach teachers
to teach. In large part, this was Judith Lanier’s
legacy. Her own reform effort might have failed at
Michigan State, but the faculty she and Lee
Shulman had recruited twenty-five years earlier
had, over two and a half decades, uncovered a
great many practices that successful teachers
employed. They had begun to codify “the wisdom
of teachers.” That meant they could lay out in
detail the things a new teacher needed to learn
how to do. And as Magdalene Lampert and Pam
Grossman showed with their instructional
activities and core practices, they had also begun
to develop ways of passing that wisdom on to new
teachers.
Judy Lanier’s faculty members at Michigan
State had been confined to general goals like
“connect teacher education more closely to the
classroom” or “make the academic preparation of
teachers more intellectually sound.” By 2004,
Deborah and her faculty at the University of
Michigan could be much more specific. Drawing
on the research from Judy Lanier’s Institute for
Research on Teaching, they could draft a
curriculum for new teachers that described a full
course of techniques: “high-leverage practices,”
the Michigan faculty called them.
 The first high-leverage practice, for instance
—“making content explicit through explanation,
modeling, representations, and examples”—drew
on the focus by Pam Grossman and other
researchers on modeling. Both Pam’s and
Magdalene’s work, meanwhile, inspired the
second high-leverage practice (“leading a whole-
class discussion”) as well as “eliciting and
interpreting individual students’ thinking,” which
was high-leverage practice number three. And the
practices related to mathematical knowledge for
teaching, Deborah’s MKT, helped ground high-
leverage practice numbers five and six:
“recognizing particular common patterns of
student thinking in a subject-matter domain” (like
the tendency of elementary school children to
mistake the “R” in their remainder findings for a
decimal point—turning, say, “1 R 5” into “1.5”)
and “identifying and implementing an
instructional response to common patterns of
student thinking” (like deciding on a way to help
fifth-graders notice the importance of defining the
whole of any fraction).
Collaborating with her colleague Francesca
Forzani—a Michigan graduate student who had
begun working with Deborah on the project—
Deborah also drew on the work of the
entrepreneurial education movement (of which
Francesca, a Teach For America alum and former
staffer, was a member). Like the academics, the
entrepreneurs were busy codifying teaching. In
particular, Deborah drew from the curriculum at a
teacher residency program spun off from the
Boston charter school Match—the one where
Rousseau Mieze’s former principal, Stacy Birdsell
O’Toole, worked.
Inspired by the school’s founding principal, a
beloved educator named Charles Sposato, who
died of cancer in 2007, the Match training
program distilled Sposato’s magic into a teachable
science. A key element was the way he had
established exceptionally strong relationships with
students and their families through methodical
habit. Every August, before school started, he
would telephone each family to build what he
called “relationship capital.” He continued calling
throughout the year, making sure to vary the kinds
of conversations so that he didn’t always bear bad
news. The Match training program, in turn,
required that each of its teacher candidates
practice six different types of phone calls, from the
“praise quickie” to the “powwow.” The
curriculum also required several hours of calls to
different parents each week: one hour a night each
weekday and two hours each weekend.
“When he called, you didn’t know what he was
calling for,” says Venecia Mumford, the mother of
two Match graduates, who saved a voicemail from
Sposato, left just before he passed away, for years
after he died. “Hey Venecia,” he said in the
message, “I’m so proud of Ed.” (Venecia’s son Ed
had just made the honor roll during his first year
at Virginia State.) “Would you tell him
congratulations and please keep the hard work up?
And thank you for always thinking of us. I love
you.”
The idea of methodical relationship building
resonated with Deborah, who had made regular
contact with her students’ families at Spartan
Village and found it immensely helpful. So, in
addition to modeling, leading discussions, and
eliciting thinking, the nineteen high-leverage
practices included “engaging in strategic
relationship-building conversations with students”
and “communicating about a student with a parent
or guardian.”
None of the high-leverage practices were easy.
All required diligence, care, thought, and a certain
amount of courage. But they had an outsized
impact; if a teacher was going to spend her time
on something, it would best be spent on one of the
high-leverage practices. And if teachers were
encouraged to make habits of the practices from
their first day in ed school, they might continue
them for the rest of their careers.
Deborah’s final stroke of luck came from an
unexpected source—not the school of research on
teaching launched by her own mentor, Lee
Shulman, but the other one, created by the
economist Eric Hanushek.

When Eric Hanushek first proposed the

accountability idea back in 1972, arguing that
education spending didn’t matter unless it was
paired with expectations, he was received as a
gadfly—“the neighborhood kook,” Hanushek
says. But by the first decade of the twenty-first
century, as first Republicans and then many
Democrats came to agree with him, the idea
became policy in states across the country. With
the passage of the No Child Left Behind law, the
kook became the establishment.
But the second piece of Hanushek’s argument
took longer to catch on. That was his case for what
he called “Teacher Accountability.” Since his
early work studying teachers in California,
Hanushek had continued to find that large
differences in effectiveness existed between the
best and worst teachers. The gaps persisted even
as he refined his “value-added” calculation, the
statistical technique that he borrowed from the
literature on the productivity of industrial
factories. Each finding reinforced the idea that by
using the value-added estimates to reward top
teachers and fire the lowest performers, American
education could be fundamentally transformed.
One Hanushek calculation compared American
students’ educational performance and that of
their Canadian peers, who performed, on average,
half a standard deviation higher on international
tests. (Canadian students were above average,
among the top ten of young people in countries
belonging to the Organisation for Economic Co-
operation and Development, whereas American
students just barely escaped the bottom ten.) The
entire gap could be wiped away, Hanushek
realized, by eliminating the bottom 6–10 percent
of American teachers as judged by value-added
scores, or, at a school with thirty teachers, by
firing two.
Among the early skeptics of Hanushek’s value-
added calculations for teachers was Tom Kane,
another economist studying education. Kane
didn’t believe the value-added numbers. At least,
he didn’t believe anyone should take them too
seriously. In 2002, when the Federal Reserve
Bank of Boston invited Kane to respond to a new
paper of Hanushek’s, he aired his concern.
“Value-added might be a useful concept,” his
thinking went, “but there’s so much noise in the
measure that it’s hard to imagine it ever being a
useful thing.”
Kane had good reason to be skeptical. A year
before, as Congress began considering President
George W. Bush’s proposed education bill—the
one that would become No Child Left Behind—he
and another economist, Douglas Staiger, ran an
analysis of the year-by-year test results that would
determine whether schools received rewards or
penalties. They found widespread variability. A
school on a mostly good trajectory could have one
bad year and thus, in the eyes of the bill, be
deemed a failure. Kane and Staiger summarized
their findings in an op-ed in the New York Times.
“Because the average elementary school has only
68 children in each grade,” they wrote, “a few
bright kids one year or a group of rowdy friends
the next can cause fluctuations in test performance
even if a school is on the right track.”
The variability was so pervasive that if No
Child Left Behind had been enacted in North
Carolina and Texas as the bill was written, only 2
percent of the states’ schools would have met its
standard of continual progress—and that was in a
period when both states showed significant
academic growth. “At the typical school, two
steps forward were often followed by one step
back.” If the legislation remained as written, they
concluded, the law “is likely to end as a fiasco.”
At the Federal Reserve Bank of Boston, Kane’s
remarks about teachers reflected his experience
studying schools. If schools’ year-to-year test
score results fluctuated that much, each teacher’s
had to be even more volatile. After all, the core
cause of variability for schools seemed to be the
small sample size of students they worked with
each year. Teachers had even fewer students per
year than schools had, meaning that more
variability was practically guaranteed. Surely,
value-added data for them would prove even less
trustworthy than the accountability data on
schools.
After Kane and Staiger’s study attracted
publicity, Congress rewrote the bill’s definition of
what was called “adequate yearly progress,”
significantly reducing the impact of the year-to-
year variability. Struck by the influence of his
research compared with the relatively small
amount of work he and Staiger had put into it,
Kane decided to take on the next natural question
—teachers. “We thought, ‘Oh it’s gotta be worse
at the teacher level, because the sample sizes are
smaller,’” Kane says. They set to work on a data
set of their own, from the Los Angeles Unified
School District.
The data took them by surprise. Just as they had
predicted, teachers’ value-added scores fluctuated
from one year to the next, the same way schools’
results did. But the fluctuations were not nearly as
arbitrary as Kane had expected. Indeed, the
teacher scores had the same predictive power as
the school scores, despite the smaller sample size.
“There was,” Kane says, “more signal to detect.”
The effect of the individual teacher was, in other
words, actually stronger than the effect of the
school—so much so that it resonated even through
the statistical haze of assorted bright kids and
rowdy friends. Indeed, the effect of a teacher was
stronger than the effect of any other educational
variable that Kane and Staiger could identify. Put
a student with a top-rated teacher, they found, and
she scored an average of 5 percentile points higher
than a similar student assigned to a middle-rated
teacher. Put her with a bottom-rated teacher, and
her scores fell an equal amount in the opposite
direction.
Kane and Staiger ran a few calculations, and
the results astonished them. The size of the
achievement gap between black and white
students, they knew, was about 34 percentile
points. “Therefore,” they estimated, “if the effects
were to accumulate, having a top-quartile teacher
rather than a bottom-quartile teacher four years in
a row would be enough to close the black-white
test score gap.”
The findings matched almost perfectly what
Hanushek had discovered thirty years earlier.
Kane had to concede that the other economist was
onto something. Noisy signals might be
distracting, he said, but “if the underlying effects
are big enough,” that would outweigh the problem
of variability. Take smoke detectors. They might
give some false alarms, but “we don’t completely
ignore them,” Kane said, “because they could save
our lives.”
And unlike in 1972, when Hanushek first made
his value-added calculations, in 2006, when Kane,
Gordon, and Staiger published their findings, the
rest of the world seemed ready to listen. By that
time, the accountability idea was not only
conventional wisdom but, judging by the well-
publicized successes of charter schools like KIPP,
it seemed to be working. And the key to the
success of schools like KIPP, observers
increasingly concluded, was not just school
expectations, but teacher accountability.
Unfettered by unions, charter schools were able to
do just what Hanushek had suggested—hire and
fire on the basis of performance alone. They
improved education by holding their teachers
accountable in ways no school had ever done
before.
In the wake of Kane and Staiger’s findings,
gathered together and published for popular
consumption by the Brookings Institution’s
Hamilton Project, Hanushek’s second kooky idea
finally began to go mainstream. Indeed, Kane and
Staiger’s Hamilton Project paper—cowritten with
Robert Gordon—produced even more impressive
results than their work on No Child Left Behind.
In 2007, presidential candidate Barack Obama
referenced the Hamilton Project findings in his
major education speech. “From the moment our
children step into a classroom,” he said, “the
single most important factor in determining their
achievement is not the color of their skin or where
they come from; it’s not who their parents are or
how much money they have. It’s who their teacher
is.”
“Perhaps,” wrote New York Times columnist
Nicholas Kristof two years later, citing the
calculations, “we should have fought the ‘war on
poverty’ with schools—or,” he added, “with
teachers.”
For Deborah Ball, the sudden swell of interest
in teachers seemed to offer the final boost her
project needed. The Common Core offered
coherence, the research on teaching and teacher
education offered a starting point for a curriculum,
and the entrepreneurs added passion and a
laboratory for experimentation. Now economists
like Eric Hanushek and Tom Kane provided a
warrant to proceed. If teaching really was the most
important of all the educational interventions, then
the only logical conclusion was that American
educators ought to build a coherent infrastructure
—clear goals, accurate tests, trained instructors—
to teach teaching.

Tom Kane didn’t know Deborah Ball, and he

didn’t know about the David Cohen coherence
problem. He only knew what his data told him.
So, when he wrote up his Hamilton Project report
with Doug Staiger and the policy maker Robert
Gordon, he made a different suggestion.
Looking at value-added scores, Kane and his
colleagues had been surprised to find that the
identity of a student’s teacher not only dwarfed the
power of key “school-level variables,” predicting
success more reliably than the size of their classes
or the funding allotted to each student. They also
outpaced every factor currently used to hire, fire,
and reward teachers.
Whether a teacher was certified, for instance,
bore almost no relationship at all to whether the
teacher’s students performed well on achievement
tests.
Nor did the scores correlate with a teacher’s
level of graduate education, even though most
school districts rewarded advanced degrees with
salary increases. Trying to find something to
contradict this finding, Kane and some colleagues
ran a study they jokingly called the “kitchen sink”
test, looking at everything from SAT scores to
“extraversion” in New York City teachers. No
meaningful exceptions emerged.
Writing about the findings in the New Yorker,
Malcolm Gladwell named the dilemma the
“quarterback problem.” Just as the NFL
Combine’s predraft tests (bench press, forty-yard
dash, and so on) appeared to bear no relationship
to a quarterback’s abilities in the game itself, there
seemed to be no way to predict whether a teacher
would succeed until he or she actually taught.
Professions like football addressed this problem
with ruthless pragmatism. When quarterbacks
failed, coaches pulled them out of the game and,
eventually, cut them from the team. School
systems, meanwhile, did almost the exact
opposite, investing heavily in the factors that
predicted teachers’ success the least. Likewise,
they ignored the one area where the research
suggested teachers could be graded accurately.
Almost no states performed on-the-job evaluations
using value-added scores, the measure the
economists had found to be most predictive of
success. Many states even prohibited the
collection of the data that would allow districts to
calculate these scores. When districts did perform
evaluations, they rarely used value-added metrics
—and rarely found poor teachers. Surveying
fifteen thousand teachers in twelve school districts
across the country, a research group at The New
Teacher Project found that, of all the teachers in
all the districts they polled, less than 1 percent had
ever been deemed unsatisfactory.
In the Hamilton Project paper, Kane, Staiger,
and Gordon laid out the obvious conclusion, the
same one that Hanushek had reached thirty years
earlier: if all the variables currently used to hire,
fire, and reward teachers were useless at
predicting student achievement, then they should
not be used at all. Instead of erecting barriers to
entry, districts should hire at will (or randomly),
steering the best teachers to the neediest students
and then weeding out the worst with evaluations
—evaluations with real teeth. Using value-added
measures to determine whether a teacher kept her
job, for instance, could give students a substantial
academic boost. Kane, Staiger, and Gordon
estimated that, in Los Angeles, letting go of
teachers who scored in the bottom 25 percent
would raise student test scores by about 14
percentile points—a boost equivalent to as much
as $169,000 extra in each student’s career
earnings. (In 2013, working with a more precise
data set, Raj Chetty, John Friedman, and Jonah
Rockoff reached a similar conclusion. Replacing a
teacher in the bottom fifth percentile with an
average teacher, they found, would increase
students’ lifetime earnings by roughly $250,000
per classroom.) All schools needed to do was
think more carefully about how they sorted
teachers after they hired them—which ones they
kept, which ones they rewarded, and which ones
they let go—and they could generate dramatic
change.
Gladwell summarized the suggestion in his
New Yorker article, with only slight exaggeration.
“Teaching,” he wrote, “should be open to anyone
with a pulse and a college degree—and teachers
should be judged after they have started their jobs,
not before.”
The argument was seductive, especially because
it seemed to explain the success of the rising
entrepreneurial education movement. Hadn’t
charter schools like Doug Lemov’s succeeded by
acting like the most cutthroat NFL franchises,
recruiting and keeping only the very best teachers?
Unfettered by the usual union and tenure
protections, they had made gains by discarding
those who couldn’t keep up.
But what sounded to the economists like simple
logic—try a lot of teachers, keep the best, fire the
rest—sounded to Deborah like a recipe for
educational malpractice. Drop an unprepared
quarterback in a game, and the only risk was lost
points. Put an unprepared teacher into a
classroom, and the students would suffer. The
economists’ own value-added research showed
how grave a risk that was. By Kane, Gordon, and
Staiger’s estimates, a year with one weak teacher
added up to a drop of 5 percentile points in
academic performance rank. The study by Chetty,
Friedman, and Rockoff had gone even further,
connecting strong teachers not only to their
students’ future earning levels, but also lower
teenage pregnancy rates and higher college
attendance.
The idea not only discounted and ignored the
needs of the children that “anyone with a pulse”
would be responsible for helping. It also flouted
the previous twenty-five years of research on
teaching, all of which suggested that good
teaching was not an innate quality, a mysterious
idiosyncrasy some people were randomly assigned
at birth. Just as most brilliant future
mathematicians couldn’t reinvent calculus on their
own, even the most talented future teachers had to
be taught. But given rigorous and regular
opportunities to work on the core practices of
teaching, a new teacher could learn to help her
students.
Deborah also knew that most of the so-called
hurdles the economists pilloried—the credentials
and licenses and master’s degrees that determined
who could teach—were, in practice, not much of a
hindrance. Pretty much anyone with a college
degree could become a teacher. In New York, for
instance, the pass rate for the teacher certification
exam in 2009 was 92 percent. By comparison, the
pass rate for the cosmetology certification exam
was 59 percent.
As for successful charter schools, they might
not have labor contracts or district guidelines on
who to hire and fire. But that didn’t mean they
used their freedom to dump the bottom quarter of
their teachers each year. As Seneca had
discovered, in the charter school world,
performance evaluation was just one of many
spokes in the complex infrastructure that helped
teachers achieve their mission. Places like
Uncommon Schools (where Doug Lemov
worked), KIPP (the network where Drew Martin
and Shannon Grande taught at Rise Academy),
and Achievement First (the infrastructure-building
network that Seneca Rosenberg studied) evaluated
their teachers’ performance, and they did let some
unsuccessful teachers go. But they also worked on
recruitment, selection, incentives, material
resources like textbooks and tests, and
professional development. One study of
Achievement First by the group Education
Resource Strategies found that the charter school
network spent less than 1 percent of its operating
budget on teacher evaluation, compared to nearly
10 percent on time for teachers to learn.
Indeed, what the entrepreneurs were clamoring
for was not more evaluation, but more guidance.
How can we get more rigorous? they asked. And
when they saw something promising, they leapt to
try it.
Their efforts to adjust to the new Common Core
standards made that clear. “Thank goodness
someone—not someone, a group of really
thoughtful people, did this,” said Joe Negron, a
middle-school math teacher and the founding
principal at KIPP Infinity in Harlem. As a teacher,
Joe’s reputation crossed state lines. Ryan Hill, the
founder and executive director of the Newark
KIPP schools, called Joe one of the best math
teachers in the entire network. When Drew Martin
at Rise found himself competing to hire a teacher,
KIPP Infinity was one of the only rivals that made
him nervous. But Negron had spent the years
before the Common Core feeling deeply frustrated.
The students practiced strategies and memorized
techniques until their “eyes popped out,” but ask
them to explain what the strategy meant, or to
reason about why their answer made sense, and
they couldn’t even begin. “I would go home and
be like, ‘I’m making robots,’ ” he says. “But I
didn’t know how not to.”
The Common Core changed Joe Negron’s
whole approach. For instance, he’d always taught
the division of fractions through mnemonics.
“Keep it, switch it, flip it” reminded students to
keep the first fraction the same, switch the
division sign to multiplication, and then flip the
second fraction (“THAT’S IT!” he sang in a song
to remind them called “Fractions and We Know
Them,” modeled off of the song “Sexy and I Know
It” by LMFAO). Now, reading over the standards
and other resources he found (he especially
appreciated books by the California math educator
Marilyn Burns, who helped teachers learn a
TKOT-style approach), Negron created tasks that
helped students understand what dividing
fractions really meant. He wanted his students to
understand fractions as their own kind of number,
participants on the number line. He made them not
only write out the equations, but also draw
pictures describing what the fractions represented.
But for all the exciting changes, Negron was
struggling to keep up. That year, he’d passed the
principalship of KIPP Infinity to a colleague in
order to return to teaching full-time. The switch
should have given him more free time. Instead, he
found he was working just as hard, if not harder.
Every night, he stayed up late, reworking his
lesson plans from scratch. What he needed was
guidance. Help. A coach.
The Hamilton Project report looked at high-
performing teachers like Joe Negron (who
presumably achieved a high value-added score)
and saw stars. By the report’s logic, Joe and other
teachers who outperformed their peers would get
rewards but receive no further training. But as
Deborah saw it, leaving the Negrons of the world
to their own devices would be like telling talented
high school musicians they’d made the symphony
—and then asking them to learn the repertoire on
their own. If she and Francesca had their way, the
Common Core would be just the beginning,
engendering materials to help all teachers achieve
the standards and to propel talents like Joe to new
levels of expertise. If the Hamilton Project
argument was sustained, the Common Core would
become one more piecemeal mandate handed to
teachers without any guidance on what to do.
Another incoherent layer in an incoherent system.
“We have a moment when we could do
something different,” Deborah said one day,
sitting in a coffee shop in Ann Arbor. “But if
everybody does it their own way, forget it. It’s
going to be the same thing again.”

For better or worse, the Hamilton Project paper

gained traction. In 2007, Bill Gates read a copy. A
few weeks later, Tom Kane met him in Manhattan
to discuss it. The next year, Gates announced a
major shift in his philanthropy. Instead of
investing hundreds of millions of dollars in small
high schools, his first big idea about how to
improve schools, the Gates Foundation would now
devote its education resources to the teacher
quality problem. The old project, Gates explained
in a TED talk, “had a good effect. But the more
we looked at it, the more we realized that having
great teachers was the very key thing.”
Gates described the differences between top and
bottom teachers as measured by Hanushek’s
value-added statistics. “If the entire U.S., for two
years, had top quartile teachers, the entire
difference between us and Asia would go away.
Within four years we would be blowing everyone
in the world away.” The conclusion, he said, was
“simple. All you need are those top quartile
teachers.” His answers echoed the Hamilton
Project paper: use on-the-job performance data to
keep top performers (and not others), steer them to
the neediest students, and give them raises.
To help districts do that, Gates promised to
invest $45 million toward designing better teacher
evaluations. He hired Kane to lead the project,
called the Measures of Effective Teaching project,
or MET. Four school districts were already on
standby, ready to implement the MET
conclusions.
Evaluation was not the only investment Gates
made in his efforts to improve teaching. From the
beginning of his efforts to grow what he called
“teacher effectiveness,” he also wanted to help
them get better, according to his lead education
adviser, Vicki Phillips, a former superintendent
and teacher. Indeed, Phillips says that from the
beginning, the foundation sought to build
“development and evaluation systems.” But,
especially for the first several years of their work,
it was the evaluation work that came out first, and
the evaluation work that got the most attention.
Then came a new force that pushed the country
even more in the direction of evaluation. The
impetus came from Barack Obama. Now
president, he had hired Robert Gordon, Kane and
Staiger’s coauthor, to help lead his Office of
Management and Budget. And in 2009,
announcing his own education plan, Obama
echoed the Hamilton Project paper once again.
Recommendation five of the paper read:

Provide federal grants to help states

that link student performance with the
effectiveness of individual teachers over
time.

The Obama administration called the grant

program it created “Race to the Top.” A
competitive fund, Race to the Top offered grants
only to those states that were willing to overhaul
their teacher evaluation systems—identifying,
promoting, and rewarding effective teachers and
removing those “who aren’t up to the job.”
Race to the Top didn’t only recommend
evaluation. The legislation also included other
policy suggestions, like urging states to give
teachers “effective support” and improving their
local teacher preparation programs. But of the
total five hundred points available in the
competitive scoring system, the largest portion
came from a category called “great teachers and
leaders.” And within that category, the largest
factor by far, at fifty-eight points, was “improving
teacher and principal effectiveness based on
performance,” meaning evaluation. A special
condition further encouraged evaluations by
disqualifying states that banned assessing teachers
by students’ test scores—a restriction that led
several states to revise their laws.
With Race to the Top nudging them along,
school districts increasingly saw teacher
evaluations as their most important tool for
improving teacher quality. By 2012, all twelve
states that were awarded Race to the Top grants,
and many that weren’t, had overhauled their
teacher evaluation systems, including, in many
states, raising the stakes so that well-rated
teachers stood to receive bonuses, higher salaries,
and tenure protection, while poorly rated teachers
could be denied tenure or fired.
Exactly how school districts expected tougher
evaluations to lead to better teaching depended on
the district. One common theory, stemming from
the Hamilton Project paper, claimed that
achievements would be gained through better
sorting. By steering the best teachers to the
students who needed them the most and removing
the worst, districts could arrange their teaching
force so that they had “the right people standing in
front of the classroom,” as Kane, Staiger, and
Gordon put it. Interviews with thirteen state policy
makers showed that some education leaders had
picked up on the idea. They intended to improve
the teacher pool by weeding out the bad ones.
“We’re talking employment decisions,” one
interviewee told researchers. “Two years of
ineffective teaching means that a teacher shall not
be reemployed.”
Increasingly, however, education leaders were
voicing a third idea, one that didn’t come from the
Hamilton Project paper: the idea that evaluations
could serve not as sorting tools, but as diagnostic
tools. By knowing how they performed, teachers
could figure out what they needed to do to
improve in the next year. “The purpose behind it is
really to help teachers that are struggling to be
better teachers,” another state official told the
group of researchers. “We’re hoping,” said
another, “that the evaluation is designed to give
very specific and actionable feedback to teachers.”
Measure something, the thinking went, and it will
get better. “Simple,” as Bill Gates said.

Could evaluations really help teachers get better?

There was some evidence to support the idea. One
teacher evaluation program in Cincinnati gave
teachers focused feedback on specific teaching
practices, from the level of classroom discourse to
the quality of the questions they asked. A study of
the Cincinnati program found that students
performed better in the years after their teachers
received focused evaluations. But evaluations, as
they had been conceived by most states, were
essentially a two-step process. The teacher was
observed, and then the teacher received a grade.
Asking teachers to learn granular teaching
practices from this system would be like asking
students to figure out trigonometry from their SAT
scores.
Another effort, led by Pam Grossman, took an
observation rubric that had been studied as a
possible evaluation device and adapted it into an
actual professional development tool for use with
a group of urban teachers, with impressive results.
The rubric, called PLATO, for the Protocol for
Language Teaching Observations, was one of
several that Tom Kane and the Gates Foundation
used as part of their MET study. Comparing
twenty-five thousand videotapes of three thousand
teachers at work against the rubrics, the study had
sought to discern whether certain teaching
practices led to gains in students’ learning. Many
of the practices did, including several of the
elements in PLATO, suggesting that the rubrics
could be used to evaluate teachers.
Pam’s professional development project began,
like MET, by scoring teachers’ lessons against the
PLATO rubric. At first, the teachers’ PLATO
scores roughly matched those from the MET
study: on average, teachers’ scores were a
disappointing 2 out of 4; on the crucial “strategy
instruction” component, the score was 1.33.* In
interviews, only one teacher demonstrated an
understanding of what strategy instruction was.
Among the eleven who didn’t, one teacher who
mentioned using a strategy talked about teaching
“organization and self-direction”—not anything to
do with the skills students needed for English
class. But by the end of Pam’s project, just as in
Cincinnati, things had changed.
The teachers noticed the difference as much as
the researchers. After the first PLATO training
session, Lorraine McLeod, a twenty-five-year
veteran who taught sixth-grade English in the San
Francisco Unified School District, said she
immediately “started doing things differently in
my classroom.” She’d always prided herself on
her charisma with the students. Her “Pharaoh
Game” lesson, in which she dressed up as an
Egyptian monarch and theatrically mimicked
knocking down the students’ paper “pyramids”
when they weren’t sturdy enough, got students
hysterical every year. Her writing lessons helped
even her many students for whom English was a
second language craft decent persuasive essays.
But PLATO stretched her ideas in new and helpful
ways.
PLATO’s “strategy instruction” unit, for
example, helped Lorraine break down the drafting
and revising processes into smaller parts. In the
case of a persuasive essay, for example, students
should learn not just to identify evidence, but to
discover ways to collect it (by highlighting and
making deliberate annotations) and explain it
(build it into an argument by describing its
importance). Lorraine designed specific lessons
about each strategy. By modeling what it looked
like to identify and then explain evidence, she
helped her students to take their writing (and their
thinking) to a higher level. As a result, when she
assigned an essay on whether Emperor Qin was
an effective leader, her students didn’t just declare
that the emperor was an effective leader because
he standardized currency; they explained that, by
standardizing currency, he had helped strengthen
the economy, thereby improving the lives of his
subjects. PLATO’s “classroom discourse” unit,
meanwhile, helped her structure paths for her
discussions to assist children in learning how to
talk about ideas.
The PLATO approach turned even Lorraine’s
favorite lessons into gems. One, a unit on poetry
in which students listened to Billy Joel songs, had
always been fun in the past. (She loved getting her
young charges to fall for her old music.) But the
class usually got stuck when she tried to transition
them from singing along with the songs to
thinking about the metaphors in the lyrics. “They
didn’t get it, and I had to explain it over and over
and over again,” she says. Even then, “only some
kids got it.”
Inspired by the PLATO trainings, Lorraine
revamped the unit, injecting it with strategy
instruction and discussion. Instead of just
listening to the song, she had the children listen
while filling out blank words in a handout. Then,
when all the words turned out to be landforms, she
transitioned into a strategy lesson on how to spot a
metaphor—and then another on how to write one.
From there, the students filled out a graphic
organizer, brainstorming four different landforms
and their possible associated emotions and
personifications. Only then did they write poems
modeled after the Billy Joel songs.
 For the first time in her career, not just a few,
but all, of the poems showed metaphorical skill, at
least for sixth-graders. “A mountain of anxiety/full
of fear . . ./A mysterious swamp/you better veer”;
“A plateau as flat as bird wings/passionate but
blind”; “The feeble cliff of old age”; “A volcano
throwing furious fists.” Before PLATO training,
she knew in theory that all kids were capable of
symbolic expression. After PLATO, she saw the
evidence. “It wasn’t just you have to have poetry
in your soul,” she said. Any student, properly
taught, could make words sing.
Lorraine wasn’t the only one. By the spring of
the PLATO study’s first year, teachers across the
project were describing big changes in their
classrooms. PLATO training, one experienced
teacher told Pam and her students, “has taught me
how to teach.” When the researchers rescored the
teachers, the improvements were confirmed. After
just three cycles of professional development, the
average PLATO score had significantly improved.
Pam was quick to point out the differences
between PLATO as used in the MET study and
PLATO as used in their professional development
program. They had the same name and the same
foundational ideas about good English teaching,
but that was about it.
For teachers, the PLATO rubrics used for
evaluation meant that either someone came into
your classroom to watch you teach, or you
installed a video camera for a day or two and
someone you never met watched the footage. The
only opportunity to improve would arrive when
the scores came back, especially if the person
who’d seen you teach or reviewed your tape
helped you think about how you might do better
next time. But at most, the review would consist
of a series of conversations. The evaluation’s main
function, at least as it was used in the Gates MET
study, was identification (who taught well and
who taught weakly?), not improvement.
PLATO as a teaching tool drew on studies of
learning. Like children, adults needed chances to
make connections between new content and what
they already understood. They also needed
resources to help them make the connections, and
they needed to feel ownership of their own
learning. As a result, the PLATO training began
not by giving teachers their individual scores, but
by sharing their aggregate scores—the group
average, broken down by each of the program’s
thirteen elements. Then Pam and her team of grad
student collaborators let the teachers decide which
two of the elements they wanted to improve.
Lorraine’s group picked strategy instruction and
classroom discourse, the two elements on which
their scores had started out the weakest.
In the workshops that followed, Pam and her
grad students dissected the two elements.
Teaching teachers to use the elements forced the
team to develop materials beyond what was in the
rubric. The starting point was description. A core
practice like “uptake,” for instance, includes three
submoves—pressing, revoicing, and connecting to
the speaker’s ideas—all of which required
definitions of their own. Next, Pam and her team
devised real examples of these moves in practice,
drawing on preexisting classroom videotapes or
creating their own. Finally, they came up with
what Pam called “approximations of practice,”
exercises designed to give the teachers a chance to
try it for themselves. Pam assigned different
teachers to different activities according to their
starting points. Those like Lorraine, with more
experience leading a discussion, practiced
“uptake” in a simulated whole group. Those with
less experience practiced in a small group.
As a result, when Lorraine worked on strategy
instruction or classroom discourse with her
students, she had more to turn to than vague
buzzwords or the scores on her evaluation. She
now drew on tangible examples (she especially
liked videos of Pam’s old collaborator from Los
Angeles, Yvonne Divans Hutchinson) and her
experience practicing with fellow teachers. She
even adopted special materials designed by Pam
and her team, like sentence stems to help students
participate in a discussion and anticipation guides
to get them thinking about the reading before
discussing it with their peers. In fact, Lorraine
didn’t realize that PLATO had any evaluation
aspect at all. As far as she knew, it was
professional development—the rare kind that
actually helped her grow rather than wasting her
time. PLATO the development tool was much
more than an evaluation system; it was an
education in how to teach.

Might the states’ evaluation systems offer this

kind of learning experience? The early signs were
not promising. One challenge was that, at the
district level, most of the observation instruments
were generic, meaning they could apply to any
academic subject, from math to English to history
to science. Generic instruments saved money, but
using them meant that any feedback would be
generic as well.
And though the MET study did not produce
clear findings about classroom practice, it did
reveal that teachers needed the most work not on
the generic elements of teaching—classroom
management challenges like keeping the students
focused and engaged—but on academic ones, like
facilitating discussions, speaking precisely about
concepts, and carefully modeling the strategies
that students needed to master.
It would be difficult for evaluators to explore
these specific shortcomings with each individual
teacher, especially given that the designated
observer in most states was not someone with the
same subject-matter specialty, but the school
principal. Principals often had teaching experience
themselves, but rarely did they have experience
across every grade and subject.
Early studies indicated that principals lacked
the kind of pedagogical content knowledge—like
MKT in math, or the equivalents in English,
history, and science—that would be required to
help a teacher, say, come up with a better
representation of dividing fractions. One pair of
researchers asked 430 principals to comment on a
teaching case study—the exact activity they would
do when evaluating their teachers for the district.
In the example, a teacher encouraged students to
discuss whether 5 can be divided by 39. In their
responses, almost half of the principals made no
reference to the math at hand, and another 25
percent made only cursory references.
The ability of evaluations to improve teaching
was also hindered by the individual value-added
scores that made up a substantial portion of
teachers’ ratings. Kane had demonstrated that
teacher effects were less meaningfully volatile
than school-wide results. And in the long term, the
scores seemed to be impressively predictive of a
teacher’s performance. But that didn’t mean the
scores were perfect measures. While a teacher
judged effective one year by value-added
techniques was likely to continue to be judged
effective the next year and the year after that, in
practice many teachers still wound up mislabeled.
Measurement error and statistical realities of large
numbers meant that some teachers who were good
would inevitably be labeled neutral or even bad by
their value-added score, even if the percentage of
the misidentified was relatively low. Other
teachers would receive value-added scores calling
them ineffective one year, then very effective the
next—and vice versa.
It was one thing to use the estimates to study
the teacher population in the aggregate and quite
another to use them to make decisions about
individual teachers’ careers. When the economist
Sean Corcoran looked at value-added scores of
New York City teachers, roughly 12,000 of whom
received ratings in the 2007–08 school year, he
found that 31 percent of English teachers who
ranked in the bottom quintile of teachers in 2007
(that is, those who were less effective than 80
percent of teachers) had jumped to one of the top
two quintiles by 2008 (more effective than 60
percent). In other words, assuming half of the
rated teachers taught English, of the 1,200 “worst”
English teachers in New York, 372 of them
became above average just one year later, at least
according to the value-added rankings. If the
bottom quintile had been fired, the district would
have lost nearly four hundred teachers who were
destined for effectiveness.
Of course, nobody was suggesting that value-
added rankings stand alone. Even the strongest
supporters of the measures advocated using them
as just a portion of a teacher’s rating. But that
portion—50 percent in some cases—still had
power.
In a paper summarizing the challenges of using
evaluation as an improvement tool, Pam
Grossman and Heather Hill wrote, “Changing
practice is slow, steady work.” If policy makers
wanted to help teachers improve, they said, they
had to “engage in the kind of high-demand, high-
support policies that . . . help teachers learn.”

Deborah Ball approached the challenge of

responding to what she called the “evaluation tidal
wave” in the way that felt most familiar: as a
teaching problem. Thanks in large part to the
value-added research publicized by Tom Kane’s
Hamilton Project paper, people all across the
country were beginning to see the importance of
high-quality teaching. What they still didn’t
understand—what Deborah and others needed to
teach them—was the best way to get more
teachers to do it.
Just as she had always done with her students,
Deborah targeted her response at what seemed
like the root of the misunderstanding: the
widespread idea that teaching was a natural gift,
something you either could or couldn’t do.
Following that logic, it made sense to try to
improve teaching through sorting. If some
teachers were simply born to the job and some
destined to fail, the logical course of action was to
throw “anyone with a pulse” into the classroom,
and then, after seeing how they did, dispense with
the duds and save the stars. But in fact, as
Deborah explained to anyone who would listen—a
group that began to include state lawmakers,
philanthropists, and congresspeople—teaching
was anything but natural work. She could prove
that with those same math problems she’d
developed back in grad school, the ones that only
a trained teacher could solve. She could also prove
it with simple logic.
Think, she told people, about all the ways good
teachers need to depart from normal human
protocol. In everyday life, when conflict emerged,
the polite approach was to smooth it over, smiling
away differences of opinion or pretending not to
notice when a friend made a mistake. Teachers, by
contrast, had to deliberately “provoke
disequilibrium,” Deborah and Francesca Forzani
wrote in an essay. Similarly, while everyday life
called for immediately helping people in need,
teachers, in order to help their students really
learn, sometimes first had to let them struggle.
And while everyday norms required people “to
assume commonality with others’ understanding
of ideas and arguments and with others’
experience of events,” teachers could not rest on
the comfortable presumption of common ground.
They had to probe.
The unnaturalness of the profession, combined
with the specialized knowledge and skill it
required, meant that improving teaching simply by
sorting the better and worse among the untrained
would be not only ineffective, but irresponsible.
“We would do that in no other sector,” Deborah
said in a speech in 2012. “In no other sector in
this society would we think the way to supply . . .
skillful work, would be to go find people, hope
they do it well, leave them on their own to figure it
out. We don’t do that with nursing, and we don’t
do that with surgery. We don’t do that with
hairdressing,” she joked, “and I’m pretty happy
about that, to tell you the truth.” Her punch line
comparison was pilots. “Every single time I get on
a plane,” she said, “I’m really glad that the plane
is not being flown by someone who just always
loved planes . . . But that’s what we do in this
country. We take people who are committed to
children, and we say here. You know, it’s
individual, work on it, figure it out.”
By 2013, Deborah thought she was, if not
winning, at least getting much better at
explaining. A combination of private donors and
support from the University of Michigan had
helped Deborah and Francesca take their Teacher
Education Initiative national, forming a new
organization called TeachingWorks. At the
University of Michigan, the School of Education
faculty had incorporated high-leverage practices
and restructured the teacher education program.
Prospective teachers now had chances to learn the
core skills and then practice them in increasingly
authentic settings. And students couldn’t graduate
until they passed a series of assessments
deliberately designed to measure whether they’d
mastered the key practices.
At TeachingWorks, Deborah and Francesca
designed a common curriculum for teacher
education, complete with everything from models
(“exemplar” videos of teachers working on the
high-leverage practices) to instructional activities
(the best way to help a future teacher learn
discussion leading, for instance), to assessments
that could be used by any training program,
whether at the school, district, or state level. The
new TeachingWorks curriculum also included a
revision of the high-leverage practices and added a
new list of “high-leverage content,” the topics in
the K–12 curriculum that all teachers needed to
know, no matter their subject (things like how to
write a thesis statement, but not a more granular
and less essential idea like the intricacies of rhyme
and meter in poetry).
Some researchers remained skeptical about the
possibilities of helping teachers learn to teach.
Pointing to the reams of studies showing no
positive effects of professional development, Eric
Hanushek, for his part, deemed efforts to revamp
it a fool’s errand. But others demurred. That group
included Bill Gates and Tom Kane, who now
emphasized the power of “feedback” and coaching
in addition to evaluation. After leaving the
foundation, Kane returned to Harvard to take on a
major new study examining the effect of targeted
feedback on teaching quality. And the Gates
Foundation, in turn, issued a major grant to help
TeachingWorks expand its infrastructure-building
efforts.
 But while the infrastructure idea was
understandable enough in theory, enacting it was
much more difficult. It would be one thing simply
to inject the TeachingWorks curriculum for
teacher education into the entrepreneurial world
that was already building an American version of
educational infrastructure. But Deborah and
Francesca wanted to reach a larger group of
teachers than that, so they had to work with the
patchwork that did exist—incoherence and all.
Deborah also saw the growing signs of a
backlash developing against evaluations that
swung in the opposite (but equally flawed)
direction. Like the charter school educators,
Magdalene Lampert, David Cohen, Pam
Grossman, Heather Hill, and the teachers in
Japan, Deborah didn’t oppose teacher evaluation.
She just didn’t think that, on its own, evaluation
could improve teaching. But as the evaluation
movement gained momentum, many teachers
understandably turned their frustration with
accountability into an argument for its opposite.
The only way to get better teaching, they argued,
was to leave teachers alone—“liberate” them, one
columnist put it, and “let them be themselves.”
Yet leaving teachers alone was exactly what
American schools had done for years, with no
great success.
The Common Core standards, vital to
establishing coherence, came under a similar
pressure. In June of 2013, Deborah watched with
dismay as lawmakers in her state, Michigan,
voted to block funding that would have supported
implementation of the new standards. Later,
legislators voted the funding through, but at the
end of 2013 they were still debating whether to
fund new Common Core–aligned tests. Other
states were backing away from the standards by
opting out of the tests that were being built to
measure them. The standards were, their critics
said, an unwelcome federal intrusion or even,
according to some, a march toward fascism. (“If
this isn’t Nazism, Communism, Marxism and all
the ‘ism’s,’ I don’t know what is,” one critic said.)
The tide was turning toward autonomy. But
autonomy offered the same prescription as
unsupported accountability: an absence of
infrastructure.
Even if Deborah and Francesca could write a
curriculum for teacher education that would
receive wide support—and they believed they
were doing just that—the two educators still faced
a slew of tactical questions. How quickly should
they expand the TeachingWorks curriculum, and
to how many teacher-training institutions? And
how could they ensure that the new teacher
trainers would teach the TeachingWorks
curriculum effectively? A growing number of ed
school professors and entrepreneurs who taught
teachers were embracing an approach that Pam
Grossman called the “practice-based teacher
education” movement. They appreciated teaching
that aspired to get students really thinking and
learning, striving toward TKOT. And they had
experience doing it themselves. But they were still
a minority.
Even the entrepreneurs’ infrastructure of
practice was full of uncertainty. Doug Lemov
might have been working on a more rigorous
taxonomy 2.0, and schools like Rise Academy
might be evolving their approach to discipline.
But whether the rest of the movement would
follow them was not clear.
Rousseau Mieze’s experience at the Harlem
school presented an alternative path, and not
necessarily a better one. Many charters maintained
high standards for teacher training, but under
pressure to replicate their schools quickly, they
sometimes failed to extend the learning culture to
every new franchise. Some became merely
superficial replicas, enforcing certain techniques
without inculcating an understanding of why they
made sense—or, even more important, revising
them when they didn’t. There was a real tension
between the desire to scale programs and the
imperative to preserve those programs’ quality.
For her part, Deborah’s most optimistic
estimate was that TeachingWorks would take at
least ten years to have an effect. Yet even this
cautious outlook fell short of expectations;
everyone else she dealt with operated on much
tighter timelines. The state and federal education
officials and the national philanthropists they
depended on for support wanted schools better not
a decade from now, but tomorrow. “Everyone gets
impatient, but ten years from now, if we had a
different system, that would be a revolution, not a
modest change,” Deborah says. The question was
whether the pieces could actually fall into place
without their backers moving on to the next big
thing.
Deborah is not the kind of person to let these
challenges deter her. In her office hangs an excerpt
from Lewis Carroll’s Alice’s Adventures in
Wonderland:
 “There is no use in trying,” said
Alice; “one can’t believe impossible
things.”
“I dare say you haven’t had much
practice,” said the Queen. “When I was
your age, I always did it for half an hour
a day. Why, sometimes I’ve believed as
many as six impossible things before
breakfast.”

Teaching, David Cohen once said, is a

profession of hope. It requires, on top of
everything else, a leap of faith. A willingness to
believe that something that does not currently exist
might one day come to life. Deborah was a great
teacher because she had spent a long time learning
how to teach. But she was also a great teacher
because she knew how to hope.
Whoever made the Alice in Wonderland poster
for her had highlighted the last line: sometimes
I’ve believed as many as six impossible things
before breakfast. Deborah hung it on her door,
like a conjecture made by one of her Spartan
Village students. Every time she left her office,
there it was, reminding her.

* The poor performance matched the

disappointing teaching quality that the MET
researchers found nationally. Nearly two-thirds of
teachers scored less than proficient on PLATO’s
measures of “intellectual challenge” and
“classroom discourse,” and more than half of all
lessons were rated unsatisfactory for “explicit
strategy use and instruction” and “modeling.”
 Epilogue

HOW TO BE A TEACHER (Part Two)

One day in 2013, midway through the writing of

this book, I finally let one of the teachers I had
been observing beat me in an argument.
I met Andy Snyder at an event hosted by the
news organization where I work, Chalkbeat (then
GothamSchools). Later, a mutual friend (himself
an excellent teacher) told me that Andy was the
best teacher he’d ever seen. Students, the friend
said, actually post photographs of the whiteboard
in Andy’s classroom on their Facebook profiles.
That’s how much they admire him.
Our mutual friend was right. Andy, who
teaches high school social studies in the New
York City public schools, is an extremely skilled
teacher. Sitting in his classroom, I often felt the
same buzzing sensation that I got watching old
tapes of Magdalene Lampert and Deborah Ball,
sitting in the classrooms of Mariel Elguero and
Shannon Grande, or observing lessons in Tokyo.
The material was for students half my age, but I
could still feel myself learning.
The argument between me and Andy went like
this: He thought that I would be a fraud to write a
book on teaching without ever doing it myself.
Did I really want to join the ranks of those who
pontificate about teaching but have never
attempted it themselves? I responded that if the
only warrant for writing about something was
doing it, then why not also suggest that political
journalists stop covering government until they
themselves hold office? Teaching shouldn’t be
exempted from outsider inquiry just because so
many people underestimate it as personal, natural
work.
I lost the debate. At least, I lost the argument
about whether or not I should try to teach. And so,
on a gray morning in March of 2013, I woke up
exceedingly early, and, with my heart pounding,
rode the subway from my apartment in Brooklyn
to a Manhattan high school called the School of
the Future, where I was to teach Andy’s class.
Two days earlier, planning the lesson, I’d felt
an ethereal mania. The thrill and pleasure of
putting together a plan had taken me by surprise. I
was having fun. But my exuberance was shattered
when I sat down with Andy to make final
preparations. A fellow teacher happened to walk
into the room where we were meeting, and he and
Andy each talked about the first class they had
taught. Andy asked the other teacher whether he’d
cried, explaining that Andy had, his first time—
more than once. Then he looked at me. “You
might cry,” he said. “I just want to warn you.”
The following evening, the night before class, I
had a nightmare. The details evaporated as soon
as I woke up, but the gist was clear: utter,
spectacular failure. The students already knew
everything I wanted to teach them. Presented with
my lesson, they handed me back indifference.
Waking up that night in an exam-level panic, I
tossed and turned and never fell back to sleep.

The lesson went by fast. I had just remembered to

look at the clock to see how much time we had left
—and like that, it was done. Andy and I had one
period to debrief, and then it was on to the next
batch of students. Although he had several more
classes that day, I taught only two.
In one way Andy misjudged me. I did not cry,
though I came close, feeling, in the period
between lessons, an exhausted, washed-out
emptiness after all the wild excitement receded
and my sleep loss set in. But Andy was right in
another, more important way. Trying to teach
showed me things I could never have known from
watching and interviewing teachers. Doing it
myself, I relearned everything that Deborah Ball,
Magdalene Lampert, Pam Grossman, and Doug
Lemov had taught me, but in more profound,
permanent ways.
I had understood that teaching was difficult
intellectual work from the first time Deborah Ball
gave me a Mathematical Knowledge for Teaching
test. But I hadn’t understood how difficult until I
worked with Andy. To plan the lesson, the first in
a unit he was building on biographical writing, I
spent hours thinking about what it meant to do
biography, what made it hard, and where a student
of the craft needed to start. I wondered if it was
more important to focus on the claims biographers
make about people, how they come to make them
—or, alternatively, whether the claims are fair.
Was it even possible to separate those problems
and work on only one at a time? Paging through
books and journalistic profiles for passages that
raised the issues I thought we should focus on, I
thought about what questions and texts I could use
to get students thinking about these issues. Then I
spent hours selecting excerpts we could read and
discuss in a sixty-minute lesson—short enough to
be understandable, yet complex enough to address
key topics I wanted the students to consider.
Yet even after all that, on the day of the class
Andy and I still struggled to articulate the “goal
for today,” which he always wrote out on the same
spot on his class whiteboard. Trying to distill the
purpose of the lesson into a single sentence, I
wrote and rewrote. By the time the lesson started,
crumbles of dry-erase marker already covered my
palm.
I learned (again) what Lee Shulman said about
pedagogical content knowledge: knowing the
content was simply not enough. I also needed to
know the students. Sitting down to plan earlier in
the week, I had proudly shown Andy the stack of
books I’d selected, with bookmarks flagging the
pages containing passages I thought we could use.
He leafed through them quickly and, one by one,
pronounced them not good enough. One, a
brilliant section of Taylor Branch’s book Parting
the Waters that introduced the civil rights activist
Bayard Rustin, was too “boring” to hold eleventh-
and twelfth-graders’ interest. So was another
Branch passage I’d chosen, introducing Roy
Wilkins. Another passage, by Michael Lewis,
which I took from his book The Big Short, might
be okay—the character being described, the hedge
fund manager Vincent Daniel, grew up nearby in
Queens, and there were some saucy details, like
the fact that his father was murdered, “though no
one ever talked about that.” But I needed to make
the selection much shorter. And it was still pretty
boring, according to Andy. He gave me a stack of
biographies the students were currently reading—
a possibility I had not anticipated. Try these, he
suggested.
The assessment stunned me. I’d spent hours
selecting these texts, and I thought the Bayard
Rustin one was particularly good. But though I’d
considered ways that I could use the passages
with the students—questions I could ask and
problems we could think about together
—“boring” was never a factor that occurred to me.
I’d spent hours interviewing Pam Grossman about
English teaching and the importance of picking
not only canonical texts, but ones that students
would actually find interesting. But somehow it
was only when Andy described how students
would react to even the most brilliant biography
that I really understood. What you assigned them
meant nothing if they didn’t read it.
When the lesson began, I learned that modeling
is just as powerful as Pam Grossman says. Andy
and I had planned for this part to happen right
before the whole-group discussion—five minutes
when we would sit in the middle of the students’
circle, reading one of the passages together and
posing the questions we wanted the students to
learn to ask. “What is the author claiming about
his subject?” “Imagine the moment when the
author first decided to write about the subject.
What actions do you think he took to learn about
him? What thinking did he do? What choices did
he make?” We used one of the passages Andy had
thrust at me the day before, from Mitch Albom’s
book Tuesdays with Morrie. Later, after the first
lesson, when Andy gathered a few students to give
me feedback, they pinpointed this moment as one
of the most helpful. One student, Marcus, told me
he’d never thought about the fact that authors had
to learn about their subjects before writing about
them until we broke down how Albom might have
learned about Morrie. As a journalist, the
challenge of overcoming that unknowing was part
of my daily life. But how were high school
students supposed to know that? Of course they
didn’t know. They had to learn. And so, of course,
they had to be taught.
I learned, again, that what Heather Kirkpatrick
had told me about “academic discourse” was
brutally accurate. Discussions are wonderful in
theory and eyeball-yankingly difficult to facilitate
in a live classroom. I had tried to arrange the
lesson in three parts: modeling, to start; then
individual practice working through the questions
with a different text (in the end, I took a risk and
used the Michael Lewis passage; nobody fell
asleep), and finally a group discussion of what the
students had learned that would, I hoped, make
the specific ideas become more abstract, taking
them from what one author did to what they might
do, if they were to write a biography of their own.
But when I began the discussion, in the first
period, I realized quickly that I was in way over
my head. Trying to keep track of the students’
ideas in my notebook as they came, I wrote
gibberish instead. And the note-taking disrupted
my working memory. Someone said something,
and I was so distracted by everything I had to do
—keep everyone focused, watch the clock, think
about where I wanted them to go next, remember
their names, call on this and that student who’d
written down an interesting idea while not
ignoring the students whose ideas were more
conventional—that I forgot to listen. I just nodded
blankly, and called on someone else.
In the second period, Andy and I decided to
give up the discussion entirely. It was simply
better not to try. Not surprisingly, the class went
much more smoothly, and I spent much less time
freaking out. Maybe the lesson simply didn’t call
for discussion. Or maybe, like so many teachers, I
took the path that felt best: easier, but not
necessarily better.
I learned, again, what the early studies had
found while searching for the optimal teacher
personality. Character traits and teaching skill are
not the same thing. They interact, but personality
does not lead to skill, or vice versa. In some ways,
I was a natural. During our debriefing, Andy
pointed out that when I walked into the room, the
students immediately treated me as the teacher—
no questions asked. Without earning it, I had their
full attention, if not yet their respect. Many
teachers aren’t that lucky, Andy said. Something
in their gait or their posture just makes it harder
for them to hold the spotlight.
But while I had some “it” in me—“Strong
Voice,” in Lemov taxonomy parlance—in other
ways my personality betrayed me. My friends and
colleagues know my habit of blurting comments
that come out rude or, as my friends never ceased
telling me in high school, “awkward.” As an
adult, I’ve worked to temper this filterless part of
myself, at least with people I’ve just met. But in
the crucible of the classroom, I reverted. When a
girl with her hair in a skull-tight ponytail said
something I didn’t understand, instead of asking
her politely for clarification, I said something like,
“How could you think that?” as if she must be an
idiot to have such a strange idea. My words
slaughtered her. As soon as I said them, the spell
I’d managed to cast was broken. Suddenly, all
twenty-some students were looking at me and
laughing, and not in a good way. The most awful
part was that I couldn’t have picked a worse
student to offend. The girl in the ponytail was the
most defiant student in the class, the one most
determined not to buy in. I offended her just by
existing, and now she was lost.
I learned, again, what I’d learned reading
Magdalene Lampert’s book about a year in her
classroom. A single lesson is not the important
unit in teaching. My initial metric of success was,
“Did they learn anything?” But as the weeks went
by, and I stayed in touch with Andy as he
continued, day after day, with the unit I’d
launched, I realized how silly that was. Learning
in school happens over weeks and months, not
periods of sixty minutes. By the time Andy and the
students finished the biography unit, they’d made
progress, then hit a dead end, restarted, and
recovered the very territory Andy and I thought we
were sowing the day of my lesson. By the time he
finally got them to produce biographical essays
that they could all get excited about, no
assessment on Earth could discern the
effectiveness of my single lesson.
 The first and last thing I relearned—the one that
stuck with me the most—had to do with that four-
letter word, love. Many times, Doug Lemov had
earnestly explained to me the importance of love
in teaching. Good discipline, he told me, required
that teachers work with the students “from love.”
When Doug said this, I always nodded. But it
wasn’t until I taught Andy’s class that I
understood what he meant.
Just before the lesson started, Andy and I stood
in the Xerox room in the school’s basement,
frantically making copies of the passages I’d
brought and the graphical organizers I’d made. I
had a million questions, and Andy answered them
patiently, one by one. He gave his own advice too,
but I can’t remember what he said. Something
about how to work the Xerox machine, maybe, or
exactly what time class began and ended and what
to say when I walked into the room. Then he
turned around, his shirt disheveled from near
oversleep and his eyes red.
 “Here’s another thing,” he said, “and this might
be the most important point. You have to look at
them with love in your heart. Once they know that
you care about them, then they can relax a lot.” It
was the only thing I managed to remember when I
made that comment to the girl in the ponytail and
nearly turned the entire class against me. Staring
at the girl I’d offended, the one who’d unwittingly
caused me to cede my command—the only thing I
had, really—I forced myself to follow Andy’s
instruction. The girl looked like she wanted to
throw me out the window. Staring back at her, I
thought about how she was a human, a person I
cared about. I decided that I loved her.
I managed to keep going. And later, when I
grabbed her as she walked out the door at the end
of class to tell her, privately, how sorry I was for
putting her on the spot like that, she gave me the
most precious gift. She turned up her lips in the
tiniest approximation of a smile, finally looked me
in the eye, and shrugged. Whatever, her face said.
But it was the first thing she’d communicated to
me all period since our incident. I could have
hugged her.

Teaching that lesson, I relearned one more thing: a

person absolutely can learn to teach. Working
with Andy, I didn’t do a great job; I did okay. But
I know with mortal certainty that if I had tried the
same thing before I began the reporting for this
book, I would have done dramatically worse.
I could tell how far I’d come when I talked to
friends who aren’t teachers about what I was
doing. “What will you teach?” they asked me, a
question that repeatedly took me aback. “What
will you teach them to do?” the question
suggested. In their questions, I heard my own
voice, circa 2009, imagining classroom work as a
presentation of expertise. Possible answers to their
question passed through my mind: I will teach
them to make a sandwich, to write a headline, to
dance, to blog, to juggle—to do anything they had
never encountered. “What is the theme of the
speech?” another friend asked, conflating a lesson
with a lecture (a form of teaching, certainly, but
just one part of it).
My own question was not so much what would
I teach, but how would I manage to do it? I
thought not just about topics, but also about
activities and ideas. What could I help the
students learn, and how could I help them learn it?
I shudder to imagine how I would have
prepared for the lesson if it had happened three
years earlier—had I, like so many new teachers,
gone into the classroom without understanding
how teaching really works. Probably I would have
been like the filmmaker who had come to teach
Andy’s class the week before me. Obsessing over
my lesson—its goals, content, and sequence—I
had forced Andy to exchange multiple e-mails
with me and then sit down for more than two
hours, going over what I’d do, step by step. The
filmmaker, Andy said, had resisted planning
altogether, and then came in and bombed, failing
to keep the students engaged, much less working
with them on anything interesting.
I still needed a ton of help, and no amount of
reading and watching and interviewing could
substitute for real practice working with students.
But at least I had an understanding of what made
teaching work—and that carried me farther than
either Andy or I could have imagined.
 Acknowledgments

“The teaching I write about in this book is not

mine alone,” Magdalene Lampert declares in the
opening of her book, Teaching Problems and the
Problems of Teaching. To crib her one last time,
this book is not mine alone.
Thank you to those who opened your hearts,
minds, and classrooms and became my teachers as
well as my sources. The list is led by but in no
way limited to Deborah Loewenberg Ball, David
Cohen, Pam Grossman, Magdalene Lampert,
Doug Lemov, Drew Martin, Rousseau Mieze, and
Akihiko Takahashi. You gave me the most
incredible gift, and I have tried my best to give it
back to everyone who reads this book.
Thank you to Aaron Pallas for introducing me
to Deborah Ball, and to Norman Atkins and David
Levin for introducing me to Doug Lemov.
Thank you to the many people who provided the
time, space, and resources that this project
required. Paul Tough gave me the magazine
assignment that became the book and then
committed his trademark grit to seeing me through
to the very bitter end, despite many moments
when giving up would have been substantially
easier. Paul also introduced me to Vera Titunik,
another terrific editor whose mark lasts here. My
superstar agent, Alia Hanna Habib, along with
David McCormick, helped me see that what
began as a magazine story could become a book
and then shepherded me into just the right hands.
Those belonged to Tom Mayer, whose expert
knowledge and skill—poured into innumerable
editing sessions—will be examined in our next
collaboration, Building a Better Editor. Thank you
also to Louise Brockett, Erin Lovett, Ryan
Harrington, Stephanie Hiebert, and the rest of the
incredible team at W. W. Norton.
Thank you to LynNell Hancock, Nicholas
Lemann, and the other brilliant minds behind the
Columbia Graduate School of Journalism’s
Spencer Fellowship in education journalism.
Thank you to Mike McPherson and the Spencer
Foundation for supporting this project not once
but twice, not to mention for supporting a healthy
portion of the actual research articles that make
this book possible. Support from the Abe
Fellowship for Journalists at the Social Science
Resource Council was invaluable in sending me to
Japan and making sure I thrived once I got there; I
am especially grateful to Nicole Restrick,
Fernando Rojas, and Takuya Toda-Ozaki. Akihiko
Takahashi and Toshiakira Fujii generously
facilitated many life-changing classroom visits in
Tokyo. Also in Tokyo, Yvonne Chang offered
translation, interpretation, directions, and good
company.
Thank you to team Chalkbeat, especially
Philissa Cramer, Alan Gottlieb, Sue Lehmann,
and Gideon Stein, for ensuring that I didn’t have
to give up one dream for another. I am also
indebted to Jill Barkin, Daarel Burnette II, Geoff
Decker, Scott Elliott, Todd Engdahl, Anna
Phillips, Maura Walz, and the ever-growing ranks
of our fellow nuance crusaders, all of whom also
helped pick up slack when I was away. You have
all taught me so much, and you inspire me every
single day.
Thank you to the generous friends and
colleagues who read this book in early and last-
minute stages and made it so much better with
your comments—namely, Drew Bailey, Jessica
Campbell, David Cohen, Philissa Cramer, Rachel
Dry, Nick Ehrmann, Alia Hanna Habib, Ryan
Hill, Timothy Pittman, Andy Snyder, Emma
Sokoloff-Rubin, Ira Stoll, Paul Tough, and Maura
Walz. I am also in debt to Elana Eisen-
Markowitz, Nitzan Pelman, and Dale Russakoff
for conversations that left a big impression.
Thank you to Jessica Campbell. Your research
assistance and fact-checking improved this book
from start to finish.
 Thank you to my teachers, especially (in order
of appearance) Lesley Wagner, Ralph Bunday,
Nanette Dyas, John Mathwin, and Darra
Mulderry, who added immeasurable value.
Thank you to my friends, who kept me
grounded and understood every time I could not
leave my apartment.
Thank you to the Epstein family for unwavering
support and many quiet writing rooms.
Thank you to my grandparents—including,
among them, three first-generation college
graduates, a normal school alum, a nonprofit
media entrepreneur, and a former math teacher.
Thank you to Andrea Weiss, John Green,
Daniel Green, and Benjamin Green, my first and
best teachers, readers, fact-checkers and friends.
Finally, thank you to David Epstein, who
teaches by example and looks at me with love in
his heart.
 Notes

The reporting for this book included many dozens

of interviews with teachers, administrators, policy
analysts, and researchers over the course of more
than five years. In addition to interviews, I relied
on a large number of research articles, books, and
records of teaching, including classrooms that I
observed directly and others made accessible to
me after the fact through videotapes, transcripts,
lesson plan books, and other records.
For space considerations, I have not included a
comprehensive list of all these interviews, texts,
and classrooms. These notes list sources that
directly informed the words on the page but are
not otherwise obvious from the text.
Except where noted otherwise, pseudonyms to
replace children’s names were provided by the
author.
 Prologue
whose names are printed on the attendance
ledger: The student names in this classroom
scene are pseudonyms invented by
Magdalene Lampert.
“Does anyone agree with this answer?”: The
classroom scenes in this chapter were
described in Magdalene Lampert, Teaching
Problems and the Problems of Teaching
(New Haven, CT: Yale University Press,
2001); in many interviews by the author with
Lampert between 2010 and 2013; and in
videotapes obtained from Lampert.
to reveal the “sense of humor” that “he had
always had”: James Hilton, Goodbye Mr.
Chips, Project Gutenberg Australia,
http://gutenberg.net.au/ebooks05/0500111h.ht
accessed September 2013.
thousands of studies conducted over dozens of
years: Thousands is not an exaggeration.
Even in 1929, the researcher Seneca
 Rosenberg reports, scholars of teaching
described “an unwieldy mass of information
. . . too large for assimilation in a lifetime”;
and by 1974, another pair of researchers
estimated that more than 10,000 teacher-
effectiveness studies had been published.
Rosenberg, “Organizing for Quality in
Education: Individualistic and Systemic
Approaches to Teacher Quality” (PhD
dissertation, University of Michigan, 2012).
extroverts or introverts, humorous or serious,
flexible or rigid: See, for instance, A. S. Barr
et al., “Wisconsin Studies of the
Measurement and Prediction of Teacher
Effectiveness: A Summary of
Investigations,” Journal of Experimental
Education 30, no. 1 (September 1961); and
Jonah E. Rockoff et al., Can You Recognize
an Effective Teacher When You Recruit
One? NBER Working Paper, no. 14485
(Cambridge, MA: National Bureau of
 Economic Research, 2008),
http://www.nber.org/papers/w14485.
described to me as “voodoo”: Jane Hannaway,
interview by the author, January 13, 2010.
“He who can, does. He who cannot, teaches.”:
George Bernard Shaw, Maxims for
Revolutionists, Project Gutenberg, Kindle
edition,
http://www.gutenberg.org/ebooks/26107,
accessed October 2013.
“It’s who their teacher is”: Barack Obama,
“Our Kids, Our Future” (speech, Manchester,
NH, November 20, 2007), American
Presidency Project,
http://www.presidency.ucsb.edu/ws/?
pid=77022.
thereby improving the overall quality of the
teaching force: Barack Obama, “Remarks by
the President on Education” (speech, US
Department of Education, Washington, DC,
July 24, 2009), White House,
 http://www.whitehouse.gov/the_press_office/R
by-the-President-at-the-Department-of-
Education.
teachers “are not rated; they are trusted”:
Chicago Teachers Union, The Schools
Chicago’s Students Deserve: Researched-
Based Proposals to Strengthen Elementary
and Secondary Education in the Chicago
Public Schools (Chicago: CTU, 2012),
http://www.ctunet.com/blog/text/SCSD_Repor
02-16-2012-1.pdf.
roughly the same size as Apple’s global
employee base: “U.S. Jobs Supported by
Apple,” Apple.com,
http://www.apple.com/about/job-creation,
accessed July 27, 2013.
and, finally, teachers (3.7 million): Deborah
Loewenberg Ball, “The Work of Teaching
and the Challenge for Teacher Education”
(lecture, Vanderbilt University, September
11, 2008). Data are based on an analysis by
 Francesca Forzani of the Household Data
Annual Averages from the Bureau of Labor
Statistics.
[number of Americans in different professions]:
This presentation was made in 2008. The
latest Census Bureau data continue to
establish teachers as the largest
occupational group in the United States. See
Richard Ingersoll and Lisa Merrill, Seven
Trends: The Transformation of the Teaching
Force, CPRE Working Paper, no. #WP-01
(Philadelphia: Consortium for Policy
Research in Education, 2012).
more than three million new teachers between
2014 and 2020: William J. Husser and
Tabitha M. Bailey, Projections of Education
Statistics to 2020, 39th ed. (Washington,
DC: National Center for Education Statistics,
2011),
http://nces.ed.gov/pubs2011/2011026.pdf.
See Table 16, p. 53.
But the district still needed six hundred new
teachers: Nancy Slavin (then director of
recruitment, Chicago Public Schools),
interviews by the author, December 4 and 18,
2009.
start work at public and private schools every
year: Husser and Bailey, Projections, Table
16, p. 53.
She’d rather not be caught watching someone
else do it: Steven Farr, interview by the
author, January 18, 2010.
better working conditions and more flexibility:
Susanna Loeb, Linda Darling-Hammond,
and John Luczak, “How Teaching Conditions
Predict Teacher Turnover in California
Schools,” Peabody Journal of Education 80,
no. 3 (2005): 47.
an average household income in the country’s
top ten: “2011 American Community
Survey,” US Census Bureau,
http://www.census.gov/acs, accessed
 November 2011.
“the greatest art in all the world”: Francis W.
Parker, Notes of Talks on Teaching,
reported by Lelia E. Patridge (New York: E.
L. Kellogg, 1891), 21,
http://books.google.com/books?
id=9aLsAAAAMAAJ&printsec=frontcover&s
accessed December 29, 2013.
and he did not intend to give them up: Ibid., xii.
Even another teacher called him a fool: William
Milford Giffin, School Days in the Fifties: A
True Story with Some Untrue Names of
Persons and Places (Chicago: A. Flanagan,
1906), 125, http://books.google.com/books?
id=P449AAAAIAAJ&pg=PA63&dq=school+
HAUvikIdWvsQTK6ICYDg&ved=0CC0Q6A
accessed December 29, 2013.
“The general public was against it”: Orville T.
Bright, Homer Bevans, and John Lancaster
Spalding, “Addresses Delivered at the
Memorial Exercises Given by the Public-
 School Teachers of Chicago and Cook
County, Auditorium, April 19, 1902,”
Elementary School Teacher and Course of
Study 2, no. 10 (June 1902): 728.
destined to behold his promised land only from
afar: William R. Harper et al., “In
Memoriam. Colonel Francis Wayland
Parker, Late Director of the School of
Education, University of Chicago,”
Elementary School Teacher and Course of
Study 2, no. 10 (June 1902): 715.
letting great teachers’ secrets live and die with
them: John Dewey, The Sources of a Science
of Education (New York: Horace Liveright,
1929), Kindle edition,
http://www.archive.org/stream/sourcesofasci
p. 10.
“can be communicated to others”: Ibid., 11.
both died before seeing “educational
Palestine”: Harper et al., “In Memoriam,”
715.
 Chapter One
he joined the prestigious new Bureau of
Educational Research: David Berliner,
“Toiling in Pasteur’s Quadrant: The
Contributions of N. L. Gage to Educational
Psychology,” Teaching and Teacher
Education 20, no. 4 (May 2004): 329–40.
with barroom storytelling late into the night:
The following portrait of Nate Gage as a
teacher and scholar is based on David
Berliner, “Toiling in Pasteur’s Quadrant,”
and on interviews by the author with David
Berliner (February 7, 2012), Barak
Rosenshine (February 19, 2012), Frank
Sobol (February 27, 2012), Lovely Billups
(February 4, 2012), and Garry McDaniels
(February 17, 2012).
to fall asleep in the middle of his lectures:
Berliner, interview.
eye color, clothing style, and strength of grip:
The studies referenced in this paragraph are
 summarized in Thomas L. Good, Bruce J.
Biddle, and Jere E. Brophy, Teachers Make
a Difference (New York: Holt, Rinehart, and
Winston, 1975), 14; and in A. S. Barr et al.,
“Wisconsin Studies of the Measurement and
Prediction of Teacher Effectiveness: A
Summary of Investigations,” Journal of
Experimental Education 30, no. 1
(September 1961), 103.
“cruel, depressed, unsympathetic, and morally
depraved”: Quoted in Egon G. Guba,
“Review of Handbook of Research on
Teaching, by N. L. Gage,” Theory into
Practice 2, no. 2 (April 1963): 114.
called, unhelpfully, “teaching skill”: Donald M.
Medley, “Early History of Research on
Teacher Behavior,” International Review of
Education 18, no. 4 (1972): 431.
or improving teacher-education programs:
Quoted in Good, Biddle, and Brophy,
Teachers Make a Difference, 13.
“I think there are about six weeks of it”:
Geraldine Joncich, The Sane Positivist: A
Biography of Edward L. Thorndike
(Middletown, CT: Wesleyan University
Press, 1968), 156.
Edward Thorndike, another foundational figure:
David Berliner, “The 100-Year Journey of
Educational Psychology,” in Exploring
Applied Psychology: Origins and Critical
Analyses, eds. Thomas K. Fagan and Gary
R. VandenBos. (Washington, DC: American
Psychological Association, 1993), 193.
calling the trips a “bore”: Joncich, Sane
Positivist, 163, 230.
“Do? Why, I’d resign!”: Ibid., 217.
“Never will you get a better psychological
subject than a hungry cat”: Edward Lee
Thorndike, Animal Intelligence (New York:
Macmillan, 1911), 54.
But he did not study teachers: Ellen Condliffe
Lagemann, An Elusive Science: The
 Troubling History of Education Research
(Chicago: University of Chicago Press,
2000), 56–66.
Dewey set his work in education aside: Ibid.,
55–56.
and universities did not hire Jews: The portrait
of Nate Gage in this paragraph draws on an
interview by the author with David Berliner
on February 7, 2012; as well as on David C.
Berliner, “Toiling in Pasteur’s Quadrant:
The Contributions of N. L. Gage to
Educational Psychology,” Teaching and
Teacher Education 20 (2004): 329–40.
by scouring psych departments’ reject lists:
Barak Rosenshine, interview by the author,
February 19, 2012.
“and uncommon commonsense”: Quoted in
Arthur G. Powell, The Uncertain Profession:
Harvard and the Search for Educational
Authority (Cambridge, MA: Harvard
University Press, 1980), 48.
“the lord deliver us therefrom”: Quoted in
Willis Rudy, “Josiah Royce and the Art of
Teaching,” Educational Theory 2, no. 3 (July
1952): 158–69.
the number of teachers alone was nearing one
million: The enrollment and staffing figures
in this paragraph are taken from Thomas D.
Snyder, ed., 120 Years of American
Education: A Statistical Portrait
(Washington, DC: National Center for
Education Statistics, 1993), 34.
“or Mendel in raising his peas”: A. S. Barr et
al., “Report of the Committee on the
Criteria of Teacher Effectiveness,” Review
of Educational Research 22, no. 3 (June
1952): 261.
in the messy cauldron of a real school: See, for
example, John Dewey, The Sources of a
Science of Education (New York: Horace
Liveright, 1929), Kindle edition; and
Lagemann, Elusive Science, 48–51.
to find out which teachers had explained it best:
This paragraph draws on N. L. Gage et al.,
Explorations of the Teacher’s Effectiveness
in Explaining, Technical Report, no. 4
(Stanford, CA: Stanford Center for Research
and Development in Teaching, 1968).
“I do not know if you have ever heard of the
book . . .”: The description of Barak
Rosenshine’s research draws on Gage et al.,
“Explorations,” 48.
so did a high level of right-to-left movement:
Gage et al., “Explorations,” 39–40.
the only two giving papers on the topic: Berliner,
“Toiling in Pasteur’s Quadrant,” 339.
the volume sold 30,000 copies: Ibid., 334.
the pamphlet had sold 130,000 copies: Nancy J.
Hultquist, “A Brief History of AERA’s
Publishing,” Educational Researcher 5, no.
11 (December 1976): 12.
to convert his findings into usable lessons for
teachers: Lovely Billups, interview by the
 author, February 4, 2012.
but with transforming it: Richard Nixon,
“Special Message to the Congress on
Education Reform” (speech, Washington,
DC, March 3, 1970), American Presidency
Project,
http://www.presidency.ucsb.edu/ws/?
pid=2895#ixzz1njSzPyoX.
“was to change the field”: Garry McDaniels,
interview by the author, February 17, 2012.
a young professor visiting from Michigan State
named Lee Shulman: The account of the
NIE’s support for teaching research draws on
Garry McDaniels, interview by the author,
February 17, 2012.
“Doesn’t Nate realize that behaviorism is on
life support?”: Lee Shulman, interview by
the author, November 2010.
psychology had to reckon with cognition: For
more background on the origins of cognitive
science, see Howard Gardner, The Mind’s
 New Science: A History of the Cognitive
Revolution (New York: Basic Books, 1985).
what Lee had always found fascinating was
thinking: This chapter’s portrait of Lee
Shulman is drawn from interviews by the
author in November 2010, January 2011, and
February 2012, as well as records captured
by the Inside the Academy project at Arizona
State University. See
http://insidetheacademy.asu.edu/photo-
gallery-lee-shulman and
http://insidetheacademy.asu.edu/lee-shulman.
“he thinks of clouds and a coming shower”:
John Dewey, How We Think (Boston: Heath,
1910), Kindle edition.
“there is blind groping in the dark”: John
Dewey, Logic: The Theory of Inquiry (Henry
Holt, 1938), Kindle edition.
far more complex than the textbooks portrayed:
The description of Lee Shulman’s research
on medical problem solving is drawn from
 Arthur S. Elstein, Lee S. Shulman, and Sarah
A. Sprafka, Medical Problem Solving: An
Analysis of Clinical Reasoning (Cambridge,
MA: Harvard University Press, 1978); Lee S.
Shulman and Arthur S. Elstein, “Studies of
Problem Solving, Judgment, and Decision
Making,” Review of Research in Education
3 (1975): 3–42; and interviews by the author
with Shulman (November 2010, January
2011, and February 2012) and Elstein
(February 6, 2012).
quite what form the transformation would take:
In addition to the work on teacher thinking
that Lee Shulman’s work on medical
problem-solving eventually inspired
(described later in this chapter), Shulman
also coauthored, with Arthur Elstein, an
article more straightforwardly examining the
early work’s implications for education
research during that year at Stanford: Lee S.
Shulman and Arthur S. Elstein, “Studies of
 Problem Solving, Judgment, and Decision
Making: Implications for Educational
Research,” Review of Research in Education
3 (1975): 3–42.
and called them “information processors”: Lee
S. Shulman et al., “Teaching as Clinical
Information Processing,” ed. N. L. Gage,
Panel 6 Report (Washington, DC: National
Conference on Studies in Teaching, 1975).
“And from there, all was commentary and
interpretation”: Gary Sykes, interview by the
author, February 20, 2012.
especially Mary Budd Rowe’s study of “wait
time”: See, for example, Mary Budd Rowe,
“Wait Time: Slowing Down May Be a Way
of Speeding Up!” Journal of Teacher
Education 37, no. 1 (1986): 43–50,
http://jte.sagepub.com/content/37/1/43.abstrac
“blessings dipped in acid”: Lee S. Shulman, The
Wisdom of Practice: Essays on Teaching,
Learning, and Learning to Teach, ed.
 Suzanne M. Wilson (San Francisco: Jossey-
Bass, 2004), 263.
“it is far more germane”: Ibid., 258.
and he decides to do another thing instead:
Jerome Bruner, Actual Minds, Possible
Words (Cambridge, MA: Harvard University
Press, 1986), 17.
they mean “teacher”: Shulman, Wisdom of
Practice, 197.
“much less generating or predicting”: Shulman
et al., “Teaching as Clinical,” 19.
“said [Connecticut congressman Abraham]
Ribicoff”: United Press International (UPI),
“Was School Racial Report Buried?” August
18, 1966. Versions of the report ran in many
papers, including the Boston Globe.
overcome the challenges of their environments:
James S. Coleman et al., Equality of
Educational Opportunity (Washington, DC:
U.S. Department of Health, Education and
Welfare, 1966).
“and more money into schools to try and
improve them?”: Eric Hanushek, interview
by the author, March 31, 2009.
across ninety-three different variables: Richard
John Murnane, “The Impact of School
Resources on the Learning of Inner-City
Children” (PhD dissertation, Yale
University, 1974), 22–23.
a data set from a school district in California:
Eric A. Hanushek, Education and Race
(Lexington, MA: Heath, 1972), 36.
whether individual teachers had an effect as
well: Richard John Murnane, “Impact of
School Resources,” 26, explains the
limitations of the Coleman Report data.
as measured by test scores: A clear summary of
the findings is in Eric A. Hanushek,
“Throwing Money at Schools,” Journal of
Policy Analysis and Management 1, no. 1
(1981): 29.
“conceptually the problem appears soluble”:
 The description of “Teacher Accountability”
is in Hanushek, Education and Race, 115.
Hanushek gave his method a name: “value-
added”: To my knowledge, the first
appearance of the term value-added came in
a footnote in Eric A. Hanushek, “Education
Policy Research—An Industry Perspective,”
Economics of Education Review 1, no. 2
(1981): 8.
That is, classroom teaching and learning:
Hanushek, Education and Race, 15.

Chapter Two
“and I knew I was home,” she says: The
following account draws on interviews by the
author with Mindy Emerson on March 30,
2012, and July 17, 2012.
“FROM THE DESK OF JESSIE J. FRY”: The
portrait of Deborah Loewenberg Ball’s
teaching career in this chapter draws on
many interviews by the author with Ball from
 April 2009 to November 2013, as well as on
records of the period obtained from Ball and
from Jessie Fry (now known as Jessie Storey-
Fry), including photographs, curriculum
materials, and lesson plan books.
to grow cotyledons and brine shrimp?: Deborah
worked with an experimental science
curriculum developed through the Science
Curriculum Improvement Study, which was
supported by a grant from the National
Science Foundation beginning in the 1960s.
reflecting on the experience in an essay:
Magdalene Lampert and Deborah
Loewenberg Ball, Teaching, Multimedia,
and Mathematics: Investigations of Real
Practice (New York: Teachers College Press,
1998), 14.
when she assumed they were learning?: Ibid.
a new experimental curriculum for elementary
school math: The Michigan State professor
whom Ball consulted was Perry Lanier. The
 curriculum he introduced her to was called
the Comprehensive School Mathematics
Program, or CSMP. CSMP’s creation was
heavily influenced by two Belgian math
educators, Georges Papy and Frédérique
Papy-Lenger.
as a way to begin a lesson on negative numbers:
CSMP Mathematics for the First Grade:
Teacher’s Guide (Aurora, CO: McREL
Institute, 1992),
http://ceure.buffalostate.edu/~csmp/CSMPPro
p. 4-473.
What about to subtract (for example, 3 minus
−5)?: Ball recounts her teaching of negative
numbers through an elevator problem in
Deborah Loewenberg Ball, “With an Eye on
the Mathematical Horizon: Dilemmas of
Teaching Elementary School Mathematics,”
Elementary School Journal 93, no. 4 (1993),
378–81.
“mentally and emotionally crushing at worst”:
 Deborah Loewenberg Ball, “Knowledge and
Reasoning in Mathematical Pedagogy:
Examining What Prospective Teachers Bring
to Teacher Education” (PhD dissertation,
Michigan State University, 1988), 1,
http://www-
personal.umich.edu/~dball/books/DBall_disse
and suddenly it would make sense: The
description of Ball’s experience in Joseph
Adney’s class draws on interviews by the
author with Ball in September 2010, as well
as on Ball’s description in Lampert and Ball,
Teaching, Multimedia, and Mathematics, 16.
What should Deborah do tomorrow?: The
description of Ball’s summer school section
draws on interviews by the author with Ball
(September 2010 and May 26, 2012), as well
as on Lampert and Ball, Teaching,
Multimedia, and Mathematics, 16–18.
didn’t become casualties of the experiment:
Lampert and Ball, Teaching, Multimedia,
 and Mathematics, 17.
asked a class of rising sixth-graders to consider
a rectangle: The description of a day at the
Elementary Math Lab that follows draws on
the author’s personal observations on July
23, 2012.
a girl named Anya: The pseudonyms in this
scene were provided by the Elementary Math
Lab.
Deborah added the problem to the warm-up:
The author’s observations at the Elementary
Math Lab were supported by many
participants. This description draws
especially on the insights of Hyman Bass,
Catherine Ditto, and Brian Cohen.
“How can we have this?” Betsy asked Jeannie:
The pseudonyms in this scene were provided
by Deborah Ball.
“Twoths. I mean halves.”: Deborah Loewenberg
Ball, “Halves, Pieces, and Twoths:
Constructing and Using Representational
 Contexts in Teaching Fractions,” in Rational
Numbers: An Integration of Research, eds.
T. P. Carpenter et al. (Hillsdale, NJ:
Erlbaum, 1993), 192.
and gave it to education majors about to
graduate: The following description draws
on Ball, “Knowledge and Reasoning.”
explained a teacher named Rachel: Ball,
“Knowledge and Reasoning,” 52.
but the perfect mix of the two: Lee S. Shulman,
The Wisdom of Practice: Essays on
Teaching, Learning, and Learning to Teach,
ed. Suzanne M. Wilson (San Francisco:
Jossey-Bass, 2004), 203.
The remaining eight came up with nothing: Ball,
“Knowledge and Reasoning,” 65.
“This Kind of Teaching” would do fine: I first
heard the term “This Kind of Teaching” and
its abbreviation, TKOT, from Sharon
Feiman-Nemser, now the Mandel Professor
of Jewish Education at Brandeis University.
 According to Feiman-Nemser, the term was
developed by her and her then colleagues at
Michigan State’s National Center for
Research on Teacher Education as part of
the university’s Teacher Education and
Learning to Teach project. “This Kind of
Teaching” also appears in Lampert and
Ball, Teaching, Multimedia, and
Mathematics, 31–35. Although neither
Magdalene Lampert nor Deborah Ball uses
the term TKOT today, I use it throughout
this book for the sake of clarity. For a
description of Michigan State educators’
thinking at the time about “teaching for
understanding,” see David K. Cohen,
Milbrey W. McLaughlin, and Joan E.
Talbert, eds., Teaching for Understanding:
Challenges for Policy and Practice (San
Francisco: Jossey-Bass, 1992).
her teaching was a kind of “existence proof”:
Magdalene Lampert, “When the Problem Is
 Not the Question and the Solution Is Not the
Answer: Mathematical Knowing and
Teaching,” American Educational Research
Journal 27, no. 1 (Spring 1990), 36.
and writing about his teaching, all at once: The
colleague on whose unique mix of pursuits
Magdalene Lampert modeled her own career
is Marvin Hoffman, a longtime teacher,
teacher educator, and currently the associate
director of the University of Chicago Urban
Teacher Education Program and founding
director of the UChicago Charter School
North Kenwood/Oakland Charter Campus.
all an observer would have to do was click a
button: Magdalene Lampert, Teaching
Problems and the Problems of Teaching
(New Haven, CT: Yale University Press,
2001), 39.
which days they would take notes: A description
of the project’s methodology is in Lampert
and Ball, Teaching, Multimedia, and
 Mathematics, 38–60.
“It was just a very compelling story”: Kara
Suzuka, interview by the author, July 2012.
“to developing direct performance incentives”:
Eric Hanushek, “Throwing Money at
Schools,” Journal of Policy Analysis and
Management 1, no. 1 (1981): 19.
that looked, at first glance, ordinary: The
following account of a scene in Deborah
Ball’s classroom and Hyman Bass’s
experience watching it is based on interviews
by the author with Bass (July 2012) and Ball
(also July 2012), and on videotape and
transcripts obtained from the University of
Michigan’s Mathematics Teaching and
Learning to Teach Project.
hijacked by a tall boy named Sean: The
pseudonyms in this scene were invented by
Deborah Ball.
“that are entailed by the actual work of
teaching”: Hyman Bass, “Mathematics,
 Mathematicians, and Mathematics
Education,” Bulletin of the American
Mathematical Society 42, no. 4 (2005): 429.
neither (to the subjects’ horror) did other
mathematicians: Deborah Loewenberg Ball,
Heather C. Hill, and Hyman Bass, “Knowing
Mathematics for Teaching,” American
Educator, Fall 2005, 14.
by interviewing witnesses about their
characteristics: An archive of materials
documenting Square One TV is available at
http://www.squareonetv.org, accessed
October 27, 2013.

Chapter Three
a college alternative that thrived in the early
twentieth century: An excellent history of
normal schools is available in Francesca M.
Forzani, “The Work of Reform in Teacher
Education” (PhD dissertation, University of
Michigan, 2011), 16–71.
a “sideshow to the performance in the center
ring”: Judith Taack Lanier et al.,
Tomorrow’s Schools of Education: A Report
of the Holmes Group (East Lansing, MI:
Holmes Group, 1995), 17.
“anything but schools of pedagogy”: Harry
Judge, American Graduate Schools of
Education: A View from Abroad: A Report
to the Ford Foundation (New York: Ford
Foundation, 1982), 42.
“with at least a courtesy appointment in another
department as well”: Ibid., 10.
not to have to work with “dumb-assed
teachers”: Ibid., 31.
The ed department became “our dumping
ground”: Ibid., 21.
they rarely strolled in before 10:00 a.m.:
Forzani, “Work of Reform,” 191.
much less to discern what made them succeed:
Ibid., 179.
“a prolonged fit of absentmindedness”: Judge,
 American Graduate Schools, 21.
“how to help children dress for recess”:
Forzani, “Work of Reform,” 198.
a 25 percent increase for future initiatives
focused on her mission: The percentages are
based on an interview by the author with Lee
Shulman in November 2010. In a separate
interview (September 2013), Judy Lanier—
now Judith Gallagher—could not confirm
whether these precise numbers were correct,
but she did confirm the general strategy of
cutting more in the short term in order to
win a larger budget tailored to her vision in
the future.
working at both Oxford and MSU: David Carroll
et al., eds., Transforming Teacher
Education: Reflections from the Field
(Cambridge, MA: Harvard Education Press,
2007), 12.
“rather than by the inductions of reason”:
Quoted in Suzanne M. Wilson, California
 Dreaming: Reforming Mathematics
Education (New Haven, CT: Yale University
Press, 2003), 9.
“ ’Tis here, ’tis there, ’tis gone’”: Alfred North
Whitehead, An Introduction to Mathematics
(New York: Holt, 1911), 7–8.
“some of the beauty and power of
mathematics”: Wilson, California
Dreaming, 13.
“the statements of the assigned text”: Larry
Cuban, How Teachers Taught: Constancy
and Change in American Classrooms, 1880–
1990, 2nd ed. (New York: Teachers College
Press, 1993), 28–29.
diagnosing “mindlessness” across the board:
Charles Silberman, Crisis in the Classroom:
The Remaking of American Education (New
York: Random House, 1970), 10–11.
without wondering about the difference from his
first calculation: Terezinha Nunes Carraher,
David William Carraher, and Analucia Dias
 Schliemann, “Mathematics in the Streets and
in the Schools,” British Journal of
Developmental Psychology 3 (1985): 6.
(software, video games, cell phone calls): Linda
Darling-Hammond, The Flat World and
Education: How America’s Commitment to
Equity Will Determine Our Future (New
York: Teachers College Press, 2010), 4.
was less than a fourth, having to do with 4: A.
Alfred Taubman, Threshold Resistance: The
Extraordinary Career of a Luxury Retailing
Pioneer (New York: HarperBusiness, 2007).
they paid again for their own: Ibid.
“it was no exaggeration to speak of a
‘movement’ for school reform”: David K.
Cohen and Heather C. Hill, Learning
Policy: When State Education Reform Works
(New Haven, CT: Yale University Press,
2001), 14.
and how were they going to learn it?:
Paraphrase of Magdalene writing about Ruth
 Heaton in Ruth M. Heaton and Magdalene
Lampert, “Learning to Hear Voices:
Inventing a New Pedagogy of Teacher
Education,” in Teaching for Understanding:
Challenges for Policy and Practice, eds.
David K. Cohen, Milbrey W. McLaughlin,
and Joan E. Talbert (San Francisco: Jossey-
Bass, 1992), 53.
the sound of no one thinking: Ruth Mary Heaton,
“Creating and Studying a Practice of
Teaching Elementary Mathematics for
Understanding” (PhD dissertation, Michigan
State University, 1994), 130. See also Ruth
M. Heaton, Teaching Mathematics to the
New Standards: Relearning the Dance (New
York: Teachers College Press, 2000).
the student, a boy named Richard: Pseudonyms
for all of Ruth Heaton’s students were
invented by Ruth and published in her
dissertation (“Creating and Studying a
Practice of Teaching Elementary
 Mathematics for Understanding”).
and he said it again: “000, 111, 000”: Heaton,
“Creating and Studying,” 131–34.
“I felt like I was floundering today”: Ibid., 136.
(a more fertile way of responding to students’
ideas): Heaton and Lampert, “Learning to
Hear Voices,” 55–58.
and construct a response to pull them there: The
description of this teaching episode draws on
interviews by the author with Ruth Heaton in
August 2012, and on Heaton and Lampert,
“Learning to Hear Voices,” 62–70.
counting out the calculation with checkers: Ruth
used a counting tool called a minicomputer
to help students learn mental computation
skills. A set of multicolored cardboard
sheets on which children laid checkers, the
minicomputer allowed students to use a
small number of checkers to convey large
numbers. Placing the checkers on squares of
different colors conveyed different values.
 For example, on one sheet, a checker on a
purple square represented 4, while a
checker on a red square represented 2—a
total of 6. On a second sheet, each square
represented 10 times the value, so one
checker on a purple square became 40, and
one checker on a red square became 20—a
total of 60. Remarkably, working with a
skilled teacher, young children quickly
become fluent in using the minicomputer to
add, subtract, and multiply. The
minicomputer was created by the Belgian
mathematician and math educator Georges
Papy, who, with his wife Frédérique Papy-
Lenger, was a major influence on the
experimental curriculum that Ruth,
Deborah, and Magdalene all used at the
Spartan Village school.
“How do I keep it up?”: The description of this
teaching episode draws on interviews by the
author with Ruth Heaton in August 2012,
 and on Heaton, “Creating and Studying,”
173–224.
“Facilitating,” she called it: The preceding
description of Sylvia Rundquist’s teaching
draws on Deborah L. Ball and Sylvia S.
Rundquist, “Collaboration as a Context for
Joining Teacher Learning with Learning
about Teaching,” in Cohen et al., Teaching
for Understanding, 13–37.
“This course has enlightened me to a whole
world”: Magdalene Lampert and Deborah
Loewenberg Ball, Teaching, Multimedia,
and Mathematics: Investigations of Real
Practice (New York: Teachers College Press,
1998), 35–155.
or “surrender their franchise”: Judith Lanier,
Tomorrow’s Teachers: A Report of the
Holmes Group (East Lansing, MI: Holmes
Group, 1986).
from colleges and universities all across the
country: Frank Murray, interview by the
 author, February 18, 2010.
“could think and reason in such advanced ways”:
David K. Cohen and Deborah Loewenberg
Ball, “Relations between Policy and
Practice: A Commentary,” Educational
Evaluation and Policy Analysis 12, no. 3
(Autumn 1990): 333.
even some that were obviously far off: The
description of Mrs. Oublier’s teaching draws
from David K. Cohen, “A Revolution in One
Classroom: The Case of Mrs. Oublier,”
Educational Evaluation and Policy Analysis
12, no. 3 (Autumn 1990): 311–29.
“To a one, we never saw radical change”:
Wilson, California Dreaming, 207.
she moved on: Deborah Loewenberg Ball,
“Reflections and Deflections of Policy: The
Case of Carol Turner,” Educational
Evaluation and Policy Analysis 12, no. 3
(Autumn 1990): 250–51.
had adopted for old worksheets: Wilson,
 California Dreaming, 207.
“teaching for misunderstanding?”: Cohen and
Ball, “Relations between Policy and
Practice,” 331.
“had remained essentially the same”: Wilson,
California Dreaming, 55.
to make sure the students’ answers are correct:
Ibid., 85–93.
here’s the framework; good luck: Cohen,
“Revolution in One Classroom.”
“in terms of the productivity expected for
tenure”: Quoted in Forzani, “Work of
Reform,” 246.
“That’s just not enough people to make it work”:
Ibid., 253.
at the insistence of the same colleagues who later
questioned it: Judith Gallagher, interview by
Jessica Campbell (fact checker for the
author), November 2013.
“and other instruments over the heads of her
colleagues”: Ibid., 257.
to deal with both school business and teaching
practices: The following account of the
Spartan Village school’s struggle in
sustaining its reforms is based on multiple
interviews by the author with Jessie Storey-
Fry between April and August 2012, a
review of records from the time provided by
Storey-Fry, and interviews by the author with
several former Spartan Village teachers.
Others joked about being “bugged”: A
photograph provided by Jessie Storey-Fry
records an arch note from her staff. “We’re
so glad you’ll be leading us through the
hard times ahead. We’re happy we can
continue to work in a bug-free
environment,” the teachers wrote to her.

Chapter Four
She’d opened her remarks with a warning: For
this chapter the author relied on two trips to
Tokyo, Japan, in November–December 2011
 and April 2012, where interviews and
observations were translated by the reporter
Yvonne Chang.
better than those with the highest scores in
Minneapolis: Reported in Richard Lynn,
“Mathematics Teaching in Japan,” in New
Directions in Mathematics Education, ed.
Brian Greer and Gerry Mulhern (London:
Routledge, 1989).
especially in matters of science and math: For
background on studies of international math
achievement, see Ina V. S. Mullis and
Michael O. Martin, “TIMSS in Perspective:
Lessons Learned from IEA’s Four Decades
of International Mathematics Assessments,”
in Lessons Learned: What International
Assessments Tell Us about Math
Achievement, ed. Tom Loveless
(Washington, DC: Brookings Institution
Press, 2007).
the top 1 percent of students around the world: T.
 Husen, International Study of Achievement
in Mathematics: A Comparison of Twelve
Countries (New York: Wiley, 1967).
better than roughly 98 percent of Americans: The
Illinois-Japan Study of Mathematics,
reported in Richard Lynn, “Mathematics
Teaching in Japan,” in New Directions in
Mathematics Education, ed. Brian Greer
and Gerry Mulhern (London: Routledge,
1989).
while the United States ranked number eight:
Associated Press, “Test Results ‘Embarrass’
U.S.,” December 12, 1983.
“in a manner comparable to the heralded
‘economic miracle’ ”: Edward B. Fiske,
“Japan’s Schools: Intent about the Basics,”
New York Times, July 10, 1983.
when they gave children a test of cognitive
ability: Harold W. Stevenson, Shin-ying Lee,
and James W. Stigler, “Mathematics
Achievement of Chinese, Japanese, and
 American Children,” Science 231, no. 4739
(February 14, 1986): 695, 696.
“and your teachers are teaching you these
things”: James Stigler, interview by the
author, August 30, 2012.
including a variable to account for classroom
teaching: James W. Stigler and James
Hiebert, The Teaching Gap: Best Ideas from
the World’s Teachers for Improving
Education in the Classroom (New York:
Free Press, 1999), Kindle edition.
“was more significant than we had thought”:
Ibid., Kindle locations 720–30.
Zero of the Japanese lessons did: Ibid., Kindle
locations 814–17.
or what new questions do you have, if any? (We):
The description of a lesson is drawn from a
fourth-grade class observed at Koganei
Elementary School on April 17, 2012.
and neither of the Americans asked a “check
status” question: The ministudy of four
 lessons is reported in James W. Stigler, Clea
Fernandez, and Makoto Yoshida,
“Traditions of School Mathematics in
Japanese and American Elementary
Classrooms,” in Theories of Mathematical
Learning, ed. Leslie P. Steffe and Pearla
Nesher (Mahwah, NJ: Erlbaum, 1996), 149–
75.
in Japanese lessons that number was 40 percent:
Stigler and Hiebert, Teaching Gap, Kindle
locations 901–3.
why converting to like denominators makes more
sense: Ibid., Kindle location 1183.
forty-five minutes’ worth of insights served
teachers better: Ibid., Kindle locations 957–
73.
“Now push the equals sign. What do you get?”:
Ibid., Kindle locations 1377–1379.
advice about how to compete with their Asian
counterparts: John Holusha, “W. Edwards
Deming, Expert on Business Management,
 Dies at 93,” New York Times, December 21,
1993.
“I came to the wrong class”: The quotes by
Akihiko Takahashi in this section come from
an interview by the author on December 21,
2011.
math is indeed the same all around the world:
Toshiya Chichibu, interview by the author,
November 27, 2011.
“and I cannot go back anymore”: Akihiko
Takahashi, interview by the author,
November 29, 2011.
as weak teachers are called in Japan: Harold
Stevenson et al., The Educational System in
Japan: Case Study Findings (Washington,
DC: National Institute on Student
Achievement, Curriculum, and Assessment,
1998), 201. A principal interviewed in the
report also cited the term nimotsu, meaning
“baggage.”
or breaking up the numbers mentally: The
 description of a typical postlesson discussion
includes excerpts reported in Clea Fernandez
and Makoto Yoshida, Lesson Study: A
Japanese Approach to Improving
Mathematics Teaching and Learning
(Mahwah, NJ: Erlbaum, 2004), 110–112.
without having to be told: The description of a
postlesson discussion is based on the
author’s personal observation of a research
lesson at Tokyo’s Wakabayashi Elementary
School on December 7, 2011.
than making the same mental step for area: The
postlesson discussion about the lesson on
angles, which the author observed, occurred
at Tokyo’s Hashido Elementary School on
April 25, 2012.
hadn’t gotten to dig into much math: The research
lesson on angles, which the author observed,
occurred at Tokyo’s Hashido Elementary
School on April 25, 2012.
“Potatoes!”: The author observed Mr.
 Hirayama’s lesson on April 18, 2012, at
Takehaya Elementary School, one of four
fuzoku schools affiliated with Tokyo
Gakugei University.
the most productive path to understanding:
Toshiakira Fujii, interview by the author,
April 25, 2012.
just like in the United States: Heidi Knipprath,
“What PISA Tells Us about the Quality and
Inequality of Japanese Education in
Mathematics and Science,” International
Journal of Science and Mathematics 8, no. 3
(June 2010): 389–408.
(“field schools,” lab schools are called in
Finland): Evidence of Singapore’s conscious
efforts to learn from Japanese lesson study is
drawn from interviews with officials at the
Singapore Ministry of Education, April 2012,
and from lectures at the World Association of
Lesson Study conference, 2011. China’s
version of lesson study is described in
 Liping Ma, “Profound Understanding of
Fundamental Mathematics: When And How
Is It Attained,” chap. 6 in Knowing and
Teaching Elementary Mathematics:
Teachers’ Understanding of Fundamental
Mathematics in China and the United States
(Mahwah, NJ: Erlbaum, 1999). Finland’s
“field schools” and their role in the country’s
recent education reforms are described in
Pasi Sahlberg, Finnish Lessons: What Can
the World Learn from Educational Change
in Finland? (New York: Teachers College
Press, 2011), 17.
to make a perfect rice pillow: See Jiro Dreams of
Sushi, directed by David Gelb (2012).
where students spent decades mastering the
special poses: Tokunaga Kyoko, “The
Kabuki Actor Training Center,” Nipponia no.
22 (September 15, 2002), http://web-
japan.org/nipponia/nipponia22/en/feature/featu
“as a reason why they would do this study”:
 James Stigler, interview by the author,
August 30, 2012.
“with unlike denominators,” in sixth-grade math:
The examples of standards are from New
York State’s English Language Arts and
Mathematics standards, published in May
and March 2005, respectively. See
http://www.p12.nysed.gov/ciai/mst/math/stand
would grow to forty-eight: Margaret A. Jorgensen
and Jenny Hoffmann, History of the No
Child Left Behind Act of 2001, Pearson
Education, August 2003 (revision 1,
December 2003),
http://images.pearsonassessments.com/images
WT.mc_id=TMRS_History_of_the_No_Child
p. 5.
“to push us along the path to success”: Stigler
and Hiebert, Teaching Gap, Kindle locations
130–37.
“to achieve our goals for students”: James
Stigler, interview by the author, September
 29, 2011.
teachers had invented new words to describe
them: The vocabulary here draws on
interviews by the author with teachers in
Tokyo, and on Fernandez and Yoshida,
Lesson Study.
“and then suddenly, you’re in this good
restaurant”: Deborah Ball, interview by the
author, May 16, 2012.

Chapter Five
“He writes on the fourth-grade level”: The
quotes by Doug Lemov throughout this
passage are from an interview by the author
on November 10, 2009.
“and means the revolution has begun”: Jane O.
Reilly, “The Housewife’s Moment of Truth,”
New York Magazine, December 20, 1971.
struggle to keep up with her peers: Wendy Kopp,
One Day, All Children . . . : The Unlikely
Triumph of Teach For America and What I
 Learned along the Way (New York:
PublicAffairs, 2003), Kindle edition,
locations 114–15.
“was the problem they were creating”: Irving
Kristol, “The Best of Intentions, the Worst of
Results,” Atlantic Monthly, August 1971.
vowed to “end welfare as we know it”: Jason
DeParle, “President Would Not Limit
Welfare Plan’s Public Jobs,” New York
Times, June 13, 1994.
“but the taproot is ignorance”: Quoted in David
K. Cohen and Susan L. Moffitt, The Ordeal
of Equality: Did Federal Regulation Fix the
Schools? (Cambridge, MA: Harvard
University Press, 2009), 45.
rising from $2,835 to $7,933 in constant dollars:
National Center for Education Statistics,
Digest of Education Statistics, Table 191
(“Total and Current Expenditures per Pupil
in Public Elementary and Secondary Schools:
Selected Years, 1919–20 through 2008–09”),
 http://nces.ed.gov/programs/digest/d11/tables
essentially a money-back guarantee: Charles A.
Radin, “Charter School Offers a Guarantee:
If Student Fails, Parents Get Tuition Free,”
Boston Globe, April 7, 1998.
were held standing up: This account of the
Academy of the Pacific Rim draws on
interviews by the author with half a dozen
staff from the time, as well as on interviews
with former students.
“(It has always been fun)”: George L. Kelling
and James Q Wilson, “Broken Windows:
The Police and Neighborhood Safety,”
Atlantic Monthly, March 1, 1982.
the “single cell of instruction” model: Dan C.
Lortie, Schoolteacher: A Sociological Study
(Chicago: University of Chicago Press,
1975), 15.
“Stand still. They’ll respond.”: Doug Lemov,
Teach like a Champion: 49 Techniques That
Put Students on the Path to College (San
 Francisco: Jossey-Bass, 2010), 3.
“like language class”: Jay Altman, interview by
the author, October 1, 2011.
“the Massachusetts charter schools that had
opened”: Linda Brown, interview by the
author, September 13, 2012.
only twenty-two schools had opened so far: The
Massachusetts Charter School Initiative
(Malden, MA: Massachusetts Department of
Education, 2001),
http://web.archive.org/web/20061019161538/
p. 62.
An “educational ‘start-up,’” Bronson called it:
Po Bronson, What Should I Do with My
Life?: The True Story of People Who
Answered the Ultimate Question (New York:
Random House, 2002), 338–39.
and all went on to four-year colleges: Katherine
Boo, “The Factory,” New Yorker, October
18, 2004.
in math and science proficiency: Sam Allis,
 “Closing the Gap,” Boston Globe, June 27,
2004.
“they’d still be in committee hearings”: Maria
Newman, “Newark School Shows Off
Educational Approach,” New York Times,
March 30, 2000.
Why does Kayla understand: The students’ names
in this description are pseudonyms that come
from a presentation that Doug Lemov gave
about how to make diagnostic testing data
useful using the example of an invented class
of second-grade girls.
at an education summit in 2006: The slides from
the presentation are available at
http://www2.ed.gov/admins/tchrqual/learn/nc
slide002.html.

Chapter Six
Driving home from Syracuse: This chapter is
based on extensive interviews by the author
with Doug Lemov and his current and former
 students and colleagues between 2009 and
2013.
challenging them to take the problem a step
further: The account of the car ride from
Syracuse to Albany draws on interviews by
the author with Doug Lemov (December 16,
2009) and Karen Cichon (January 27, 2010),
as well as notes provided by Karen Cichon.
to write down her thoughts on paper: Doug
Lemov, Teach like a Champion: 49
Techniques That Put Students on the Path to
College (San Francisco: Jossey-Bass, 2010),
140.
in order to chide another on her failure: Ibid.,
213.
brutally specific about exactly what they wanted:
Ibid., 177.
was less dismal and took up less time: Ibid., 194.
from about fifteen to more than a hundred: Linda
Brown, interview by the author, September
13, 2012.
 Chapter Seven
he was a true believer: The following passage is
based on extensive interviews by the author
with Rousseau Mieze between August and
December 2013.
the most academically stimulating place he’d
ever been: This chapter draws on dozens of
interviews by the author with no-excuses
charter school teachers and leaders, as well
as on many school visits and personal
observations.
because it serves your students: Doug Lemov,
Teach like a Champion: 49 Techniques That
Put Students on the Path to College (San
Francisco: Jossey-Bass, 2010), 175–76.
the overall picture was shocking: Fresno Unified
School District, Chartering Authority,
“Notice to Cure and Correct,” sent to Nolan
Highbaugh, General Counsel, KIPP
California, December 11, 2008.
“like a whipping and ball and chain”: These
 comments were published in a private
research report prepared for the Academy of
the Pacific Rim in the 2002–03 school year
and obtained by the author.
their affection was always bracketed: “love-hate”:
Millisent Fury Hopkins, interview by the
author, September 2013.
down from 58 percent the year before: Academy
of the Pacific Rim Charter School, “Annual
Report 2002–03,” 2.
and 8 percent for black students: Rebecca
Gordon, Libero Della Piana, and Terry
Keleher, Facing the Consequences: An
Examination of Racial Discrimination in
U.S. Public Schools (Oakland, CA: Applied
Research Center, 2000),
http://www.arhsparentcenter.org/files/Racial-
Discrimination-in-US-Public-Schools.pdf, p.
29.
“And my teacher just assumed I did that on
purpose”: Chimel Idiokitas, interview by the
 author, September 20, 2013.
but Kevin’s were king’s blue: Kevin Thai,
interview by the author, September 2013.
“just follow and follow and follow”: These
comments were published in a private
research report prepared for the Academy of
the Pacific Rim in the 2002–03 school year
and obtained by the author.
“just a lot of pointless rules”: Ibid.
“grudging compliance”: Jere Brophy and Mary
McCaslin, “Teachers’ Reports of How They
Perceive and Cope with Problem Students,”
Elementary School Journal 93, no. 1
(September 1992): 14.
to model his own devotion to his students on Mr.
Phillips’s example: Kevin Thai, interview by
the author, September 2013.
“emotions that are counterproductive to
learning”: George G. Bear, “School
Discipline in the United States: Prevention,
Correction, and Long-Term Social
 Development,” School Psychology Review
27, no. 1 (1998): 14–33.
had only joined later on, in ninth grade: The
statistics are based on the recollections of two
members of APR’s first graduating class:
Millisent Fury Hopkins and Kevin Thai.
to thirty-four in ninth: Rousseau Mieze, interview
by the author, September 23, 2012.
didn’t make it to graduation was 21.6: “Boston
Public Schools 2007–2008: Student
Dropout,” Office of Research, Assessment,
and Evaluation, February 2009,
http://www.bostonpublicschools.org/files/Drop
08.pdf.
“they had to stay because it would pay off”:
Chimel Idiokitas, interview by the author,
September 2013.
So they left: The empirical research on charter
school attrition is mixed. One study, of
students in Texas, found that students across
all racial and income groups leave charter
 schools at significantly higher rates than
they leave noncharter schools, although the
study did not investigate the reasons for the
departures. Eric Hanushek et al., “Charter
School Quality and Parental Decision
Making with School Choice,” Journal of
Public Economics 91 (2007): 823–48.
However, other studies have found no
significant difference in charter and
noncharter school attrition rates. See, for
example, Scott A. Imberman, “Achievement
and Behavior in Charter Schools: Drawing a
More Complete Picture,” Review of
Economics and Statistics 93, no. 2 (May
2011): 416–35; and Ira Nichols-Barrer et al.,
Student Selection, Attrition, and
Replacement in KIPP Middle Schools,
Mathematica Policy Research Working
Paper, September 2012.
reports of bad behavior on the No Limits bus:
The following description of Rise Academy
 is based on multiple visits to the school by
the author and on author interviews with
Drew Martin, Shannon Grande, Ranjana
Reddy, and more than a dozen other teachers
and students at Rise Academy between
December 2010 and February 2013.
“may end in a trap”: Ronald Wright, A Short
History of Progress (New York: Carroll &
Graff, 2004), 5.
“and breaking of healthy adult bonds”: American
Psychological Association Zero Tolerance
Task Force, “Are Zero Tolerance Policies
Effective in Schools? An Evidentiary Review
and Recommendations,” American
Psychologist 63, no. 9 (December 2008):
852–62.
“time and opportunity to get a good education”:
Jay Mathews, Work Hard. Be Nice.: How
Two Inspired Teachers Created the Most
Promising Schools in America (Chapel Hill,
NC: Algonquin, 2009), Kindle edition,
 location 2745.
to the September 11 tragedy: Interview by the
author with the educator.
“no practice interacting with other kids
socially”: Ranjana Reddy, interview by the
author, November 10, 2012.
“he gets it out, and he moves on”: The
descriptions of Shannon Grande’s teaching
are drawn from visits to her classroom in
June 2011, September 2012, and February
2013, and from an interview with Shannon
by the author, October 2011.
more challenges than their more affluent peers
face: Paul Tough, How Children Succeed:
Grit, Curiosity, and the Hidden Power of
Character (Boston: Houghton Mifflin
Harcourt, 2012).
“we gave them tools, and they figured it out”:
Mariel Elguero, interview by the author,
February 2013.
“just throw [me] into a box and say go home”:
 Kevin Thai, interview by the author,
September 2013.
to flesh out their culture conversations: The KIPP
character curriculum and its basis in research
are described in Paul Tough’s book, How
Children Succeed.
Chi changed—and so did his colleagues: Chi
Tschang, interview by the author, September
28. 2012.
And it was much harder: David Levin, interview
by the author, December 18, 2013.
“Expecting what you didn’t think was possible”:
Mariel Elguero, interview by the author,
April 2012.
approaches to dealing with interpersonal
challenges: Bear, “School Discipline.”
“and of navigating obstacles”: Carol D. Lee,
Culture, Literacy, and Learning: Taking
Bloom in the Midst of the Whirlwind (New
York: Teachers College Press, 2007), 28.
a girl named Taquisha: “Taquisha” is a
 pseudonym created by Carol Lee.
that morning’s copy of the Chicago Sun-Times:
The following account draws on Lee,
Culture, Literacy, and Learning, 132–41.
and how the three can and cannot intersect: Ibid.,
118–23.
“the ethical and moral” part of teaching: Ibid.,
128.
“maladaptive coping strategies”: Margaret Beale
Spencer et al., “Vulnerability to Violence: A
Contextually-Sensitive, Development
Perspective on African American
Adolescents,” Journal of Social Issues 59,
no. 1 (2003): 33–49.
“a person who could have ideas”: Magdalene
Lampert, Teaching Problems and the
Problems of Teaching (New Haven, CT:
Yale University Press, 2001), 265–72.
without asking Magdalene what to do: Ibid., 278.
“but not in a way that would be embarrassing”:
Ibid., 279.
getting students to do “productive, positive
work”: Lemov, Teach like a Champion,
144–49.

Chapter Eight
“I got a new class of fourth-graders,” she says:
The description of Seneca Rosenberg’s
teaching and research career draws on
interviews by the author in January,
February, and March of 2013, and on e-mail
exchanges with the author on July 1, 2013.
looked for different strengths in teachers: Brian
A. Jacob and Lars Lefgren, What Do
Parents Value in Education: An Empirical
Investigation of Parents’ Revealed
Preferences for Teachers, NBER Working
Paper, no. 11494 (Cambridge, MA: National
Bureau of Economic Research, 2005),
http://www.nber.org/papers/w11494.pdf?
new_window=1.
so that they wouldn’t leave in the first place:
 Jason A. Grissom and Michelle Reininger,
“Who Comes Back? A Longitudinal Analysis
of the Reentry Behavior of Exiting
Teachers,” Education Finance Policy 7, no.
4 (Fall 2012): 446.
the “inconsistency” of “instructional guidance”:
D. K. Cohen and J. Spillane, “Policy and
Practice: The Relations between Governance
and Instruction,” Review of Research in
Education 18, no. 1 (January 1992): 17.
“variability” or, more plainly, “incoherence”:
David K. Cohen, “Standards-Based School
Reform: Policy, Practice, and Performance,”
in Holding Schools Accountable:
Performance-Based Reform in Education,
ed. Helen F. Ladd (Washington DC:
Brookings Institution, 1996), 108–9.
“You’re also absolutely right!”: Lee S. Shulman,
The Wisdom of Practice: Essays on
Teaching, Learning, and Learning to Teach,
ed. Suzanne M. Wilson (San Francisco:
 Jossey-Bass, 2004), 102.
more than fourteen thousand school districts:
“School Districts,” U.S. Census Bureau,
http://www.census.gov/did/www/schooldistric
accessed November 2013.
and nearly a hundred thousand schools:
“Educational Institutions,” National Center
for Education Statistics Fast Facts,
http://nces.ed.gov/fastfacts/display.asp?
id=84, accessed November 2013.
“and go back to what you believe in”: Lovely
Billups, interview by the author, February 4,
2012.
like roads, bridges, and power lines: The
American Heritage Dictionary of the
English Language, 5th ed. (Boston:
Houghton Mifflin Harcourt, 2011–13).
exactly what students were supposed to learn:
David K. Cohen and Susan L. Moffitt, The
Ordeal of Equality: Did Federal Regulation
Fix the Schools? (Cambridge, MA: Harvard
 University Press, 2009), 3–4.
“concerning teaching, learning and academic
content”: Ibid., 4.
“standard operating procedures” outlining best
practices: David K. Cohen, Teaching and Its
Predicaments (Cambridge, MA: Harvard
University Press, 2011), 56–57.
“year-to-year road map for reaching those
goals”: Peter Meyer, “The Common Core
Conflation Syndrome: Standards &
Curriculum,” CUNY Institute for Education
Policy at Roosevelt House, June 12, 2013,
http://roosevelthouse.hunter.cuny.edu/ciep/the
conflation-continues-or-bring-on-the-
comfederal-stational-curstandalums.
“best understood as a sort of exoskeleton”:
Cohen and Moffitt, Ordeal of Equality, 10.
more than ten thousand corps members: Greg
Toppo, “Teach For America Turns 15,” USA
Today, October 6, 2005.
about a month of extra instruction, by one
 estimate: Paul T. Decker, Daniel P. Mayer,
and Steven Glazerman, “The Effects of
Teach For America on Students: Findings
from a National Evaluation,” Mathematica
Policy Research, June 9, 2004, 31.
more than two and a half, by another: Melissa A.
Clark et al., The Effectiveness of Secondary
Math Teachers from Teach For America and
the Teaching Fellows Programs
(Washington, DC: Institute for Educational
Studies, National Center for Education
Evaluation and Regional Assistance, 2013).
the corps members did no harm: Steven
Glazerman, Daniel Mayer, and Paul Decker,
“Alternative Routes to Teaching: The
Impacts of Teach For America on Student
Achievement and Other Outcomes,” Journal
of Policy Analysis and Management 25, no.
1 (Winter 2006): 75–96.
“if it were settled easily or soon”: Cohen and
Spillane, “Policy and Practice,” 24.
which snarled their efforts: Cohen and Moffitt,
Ordeal of Equality, 172.
totaled only about seven thousand: Toppo, “Teach
For America Turns 15.”
less than 1 percent of the 3.6 million teachers:
National Center for Education Statistics,
Digest of Education Statistics, Table 69
(“Public and Private Elementary and
Secondary Teachers, Enrollment, and
Pupil/Teacher Ratios: Selected Years, Fall
1955 through Fall 2020”),
http://nces.ed.gov/programs/digest/d11/tables/
almost forty-eight million in traditional public
schools: Ibid., Table 108.
formal interviews with forty-one of them: Seneca
Rosenberg, “Organizing for Quality in
Education: Individualistic and Systemic
Approaches to Teacher Quality” (PhD
dissertation, University of Michigan, 2012),
viii.
that had arisen so haphazardly for Seneca:
 Steven Farr, Teaching as Leadership: The
Highly Effective Teacher’s Guide to Closing
the Achievement Gap (San Francisco:
Jossey-Bass, 2010).
jugyokenkyu-style sessions for teachers: Ibid.,
136–41.
to a formal coaching system: Ibid., 148.
“Look, this is how you’re supposed to do it”:
Ibid., 246–47.
“what America has never—or hardly ever—
had”: David Cohen, interview by the author,
February 26, 2013.
“might look like in the US context”: Rosenberg,
“Organizing for Quality in Education,” 183.
to get them to really understand: Ibid., 170.
“you’re not sure how your kids are going to do”:
Ibid.
“They came up with another plan that did work”:
The described exchange is based on
recollections shared with the author by
Magdalene Lampert in April 2012, July
 2012, February 2013, April 2013, and
August 2013; and on video footage from
“Standards for National Testing and Exams,”
C-SPAN Video Library, July 19, 1991,
http://www.c-
spanvideo.org/program/Exams.
“the work of teaching while students work
independently”: Magdalene Lampert,
Teaching Problems and the Problems of
Teaching (New Haven, CT: Yale University
Press, 2001), 121.
Out of $248 million: NewSchools Venture Fund,
“2012 Annual Report,”
http://issuu.com/nsvf/docs/2012annualreport?
e=7139272/2303874, accessed February
2013.
Achievement First received over $6 million:
“Venture Snapshot: Achievement First,”
NewSchools Venture Fund,
http://www.newschools.org/venture/af,
accessed February 2013.
Uncommon Schools, more than $7 million:
“Venture Snapshot: Uncommon Schools”
NewSchools Venture Fund,
http://www.newschools.org/venture/uncommo
schools; “Venture Snapshot: Roxbury
Preparatory Charter School” NewSchools
Venture Fund,
http://www.newschools.org/venture/north-
star; and “Venture Snapshot: North Star
Academy Charter School of Newark,”
NewSchools Venture Fund,
http://www.newschools.org/venture/roxbury-
preparatory-charter-school, both accessed
February 2013.
and KIPP, more than $6 million: “Venture
Snapshot: KIPP Foundation,” NewSchools
Venture Fund,
http://www.newschools.org/venture/kipp-
foundation; “Venture Snapshot: KIPP D.C.,”
NewSchools Venture Fund,
http://www.newschools.org/venture/kipp-dc;
 “Venture Snapshot: KIPP MA,” NewSchools
Venture Fund,
http://www.newschools.org/venture/kipp-ma;
and “Venture Snapshot: TEAM Charter
Schools,” NewSchools Venture Fund,
http://www.newschools.org/venture/team-
charter-schools, all accessed February 2013.
“It was awesome”: Magdalene Lampert, e-mail
message to the author, February 14, 2013.
And in math, they were worse: Jesse Solomon
letter to friends of Boston Teacher Residency,
December 14, 2011. The study was
conducted by Harvard University’s Center
for Education Policy Research at the request
of the Boston Teacher Residency.

Chapter Nine
if she still wasn’t sure what to drink?: Magdalene
Lampert and Filippo Graziani, “Instructional
Activities as a Tool for Teachers’ and
Teacher Educators’ Learning,” Elementary
 School Journal 109, no. 5 (2009): 497.
and got them to start over: Ibid., 499–500.
reminders and suggestions about how to proceed:
Ibid., 499–500.
a discussion leading to the key mathematical
idea: Magdalene Lampert et al., “Using
Designed Instructional Activities to Enable
Novices to Manage Ambitious Mathematics
Teaching,” in Instructional Explanations in
the Disciplines, eds. M. K. Stein and L.
Kucan (New York: Springer, 2010), 136.
five thousand worst-performing middle and high
schools: Michele McNeil, “Tight Leash
Likely on Turnaround Aid,” Education Week,
September 2, 2009.
a fifteen-year veteran teacher: Ilene Carver,
interview by the author, April 23, 2013.
The lesson began: The description of this lesson is
based on a video provided by Magdalene
Lampert and on interviews by the author with
Magdalene Lampert (April 2013), Ilene
 Carver (April 23, 2013), and Sabine
Ferdinand (April 23, 2013).
“we’re not where we want to be”: The quotes
from Heather Kirkpatrick in this section
come from an interview by the author on
January 23, 2013.
“that’s why this evidence is so important!”: The
description of modeling is drawn from the
author’s observation of a PLATO workshop
for teachers led by Pam Grossman, Michael
Metz, and other colleagues in San Francisco
on March 14, 2013.
between the neighborhoods of Watts and
Compton: Yvonne Divans Hutchinson,
“About My School and My Classroom,”
Inside Teaching, a project of the Center to
Support Excellence in Teaching at Stanford,
http://insideteaching.org/quest/collections/site
hutchinson_yvonne/teachingcontext.html,
accessed November 2013.
“I want to add to what (person’s name) said”:
 Yvonne Divans Hutchinson, “Promoting
Literate Discourse in the Classroom,” Inside
Teaching,
http://insideteaching.org/quest/collections/site
hutchinson_yvonne/promlitdis.html,
accessed November 2013.
“they’re much more apt to be engaged”: Yvonne
Divans Hutchinson, videotaped interview,
http://insideteaching.org/quest/collections/site
hutchinson_yvonne/cleanwlfc.mov, accessed
November 2013.
and solicited comments on it: Lisa Marie Barker,
“Under Discussion: Improvisational Theatre
as a Tool for Improving Classroom
Discourse” (PhD dissertation, Stanford
University, 2012), 16.
notifying a student swiftly of her mistake: Doug
Lemov, Teach like a Champion: 49
Techniques That Put Students on the Path to
College (San Francisco: Jossey-Bass, 2010),
267.
couldn’t be mapped back to a memorable race:
K. Anders Ericsson, William G. Chase, and
Steve Faloon, “Acquisition of a Memory
Skill,” Science 208, no. 4448 (June 1980):
1181–82.
what they knew about how numbers worked:
Thomas P. Carpenter et al., Children’s
Mathematics: Cognitively Guided
Instruction (Portsmouth, NH: Heinemann,
1999).
the abstract mental model that made sense:
Renee Baillargeon, “Physical Reasoning in
Infancy,” in The Cognitive Neurosciences,
ed. M. S. Gazzaniga (Cambridge, MA: MIT
Press, 1995), 190.
“a normal and healthy part of the learning
process”: Lemov, Teach like a Champion,
222.
only 4 percent of American public school
students: National Center for Education
Statistics, Digest of Education Statistics,
 Table 108 (“Number and Enrollment of
Public Elementary and Secondary Schools,
by School Level, Type, and Charter and
Magnet Status: Selected Years, 1990–91
through 2010–11”),
http://nces.ed.gov/programs/digest/d12/tables

Chapter Ten
“but what do you actually want to do?”: Deborah
Ball, interview by the author, June 2013.
exactly what students are supposed to learn:
David K. Cohen and Susan L. Moffitt, The
Ordeal of Equality: Did Federal Regulation
Fix the Schools? (Cambridge, MA: Harvard
University Press, 2009), 3–4.
up from sixty-five thousand just twenty years
earlier: Richard Ingersoll and Lisa Merrill,
Seven Trends: The Transformation of the
Teaching Force, CPRE Report, no. RR-79
(Philadelphia: Consortium for Policy
Research in Education, University of
 Pennsylvania, 2012),
http://www.cpre.org/sites/default/files/working
9.
was now just one: Thomas G. Carroll and
Elizabeth Foster, Who Will Teach?
Experience Matters (Washington, DC:
National Commission on Teaching and
America’s Future, 2010), http://nctaf.org/wp-
content/uploads/2012/01/NCTAF-Who-Will-
Teach-Experience-Matters-2010-Report.pdf,
p. 10.
and too few of Robert E. Lee: Lynne Cheney,
“The End of History,” Wall Street Journal,
October 20, 1994.
a reading goal matched to third grade instead of
first: Phyllis Schlafly, “School-to-Work and
Goals 2000,” Phyllis Schlafly Report 30, no.
9 (April 1997),
http://www.eagleforum.org/psr/1997/apr97/ps
were important for teaching children to read:
National Reading Panel, Teaching Children
 to Read: An Evidence-Based Assessment of
the Scientific Research Literature on
Reading and Its Implications for Reading
Instruction (Washington, DC: National
Reading Panel, 2000),
https://www.nichd.nih.gov/publications/pubs/
Inevitable disagreements remained: For a look at
early push-back to the Common Core
standards, see Stephanie Banchero, “School-
Standards Pushback,” Wall Street Journal,
May 8, 2012.
twenty-seven states had vowed to adopt the
standards: Tamar Lewin, “Many States
Adopt National Standards for Their
Schools,” New York Times, July 21, 2010.
“academic preparation of teachers more
intellectually sound”: Francesca Forzani,
“The Work of Reform in Teacher Education”
(PhD dissertation, University of Michigan,
2011), 206–22.
“And thank you for always thinking of us. I love
 you”: Charles Sposato, voicemail recorded in
2007 by Venecia Mumford and played for the
author during an interview, June 2008.
the kook became the establishment: Eric
Hanushek, interview by the author, October
25, 2012.
or, at a school with thirty teachers, by firing two:
Eric A. Hanushek, “Teacher Deselection,”
in Creating a New Teaching Profession, eds.
Dan Goldhaber and Jane Hannaway
(Washington, DC: Urban Institute Press,
2009).
“it’s hard to imagine it ever being a useful
thing”: The complete discussant paper is
Thomas J. Kane, “Improving Educational
Quality: How Best to Evaluate Our
Schools?” in Education in the 21st Century:
Meeting the Challenges of a Changing
World: Conference Proceedings, Conference
Series (Federal Reserve Bank of Boston), no.
47 (Boston: Federal Reserve Bank of Boston,
 2002). See also
http://www.bostonfed.org/economic/conf/conf4
accessed November 2013.
the law “is likely to end as a fiasco”: Thomas J.
Kane and Douglas O. Staiger, “Rigid Rules
Will Damage Schools,” New York Times,
August 13, 2001.
“enough to close the black-white test score gap”:
Robert Gordon, Thomas J. Kane, and
Douglas O. Staiger, “Identifying Effective
Teachers Using Performance on the Job,”
Hamilton Project Discussion Paper 2006-01
(Washington, DC: Brookings Institution,
2006),
http://www.brookings.edu/views/papers/20060
p. 8.
“because they could save our lives”: Tom Kane,
interview by the author, April 17, 2013.
“It’s who their teacher is”: Barack Obama, “Our
Kids, Our Future” (speech, Manchester, NH,
November 20, 2007), American Presidency
 Project,
http://www.presidency.ucsb.edu/ws/?
pid=77022.
“or,” he added, “with teachers”: Nicholas D.
Kristof, “Our Greatest National Shame,”
New York Times, February 14, 2009.
performed well on achievement tests: Robert
Gordon, Thomas J. Kane, and Douglas O.
Staiger, “Identifying Effective Teachers
Using Performance on the Job,” Hamilton
Project Discussion Paper 2006-01
(Washington, DC: Brookings Institution,
2006),
http://www.brookings.edu/views/papers/20060
p. 7.
No meaningful exceptions emerged: Jonah E.
Rockoff et al., Can You Recognize an
Effective Teacher When You Recruit One?,
NBER Working Paper, no. 14485
(Cambridge, MA: National Bureau of
Economic Research, 2008),
 http://www.nber.org/papers/w14485.
had ever been deemed unsatisfactory: Daniel
Weisberg et al., The Widget Effect: Our
National Failure to Acknowledge and Act on
Differences in Teacher Effectiveness, 2nd
ed. (Brooklyn, NY: The New Teacher
Project, 2009),
http://widgeteffect.org/downloads/TheWidgetE
$169,000 extra in each student’s career earnings:
Gordon, Kane, and Staiger, “Identifying
Effective Teachers,” 14–15.
by roughly $250,000 per classroom: Raj Chetty,
John N. Friedman, and Jonah E. Rockoff,
“Measuring the Impacts of Teachers II:
Teacher Value-Added and Student Outcomes
in Adulthood,” NBER Working Paper, no.
19424 (Cambridge, MA: National Bureau of
Economic Research, 2013),
http://www.nber.org/papers/w19424.
“after they have started their jobs, not before”:
Malcolm Gladwell, “Most Likely to
 Succeed,” New Yorker, December 15, 2008.
a drop of 5 percentile points in academic
performance rank: Gordon, Kane, and
Staiger, “Identifying Effective Teachers,” 8.
a new teacher could learn to help her students:
Donald J. Boyd et al., “Teacher Preparation
and Student Achievement,” Educational
Evaluation and Policy Analysis 31, no. 4
(December 2009): 416–40.
for the teacher certification exam as 92 percent:
John Hildebrand, “New Schools Chief Calls
for Tougher Teacher Standards,” Newsday,
July 27, 2009.
for the cosmetology certification exam was 59
percent: Interview by the author with an
employee of the New York Department of
State, Division of Licensing Services, 2009.
nearly 10 percent on time for teachers to learn: A
New Vision for Teacher Professional
Growth & Support: Six Steps to a More
Powerful School System Strategy
 (Watertown, MA: Education Resource
Strategies, 2013),
http://www.erstrategies.org/cms/files/1800-
gates-pgs-white-paper.pdf, p. 33.
“not someone, a group of really thoughtful
people, did this”: Joe Negron, interview by
the author and Emma Sokoloff-Rubin, April
2013.
called “Fractions and We Know Them”: A
rendition of the song is available in Fractions
and We Know Them, YouTube,
http://www.youtube.com/watch?
v=lUygYN6tgyI, accessed October 2013.
“having great teachers was the very key thing”:
Bill Gates, “Mosquitos, Malaria and
Education” (TED Talk), TED 2009, February
2009,
http://www.ted.com/talks/bill_gates_unplugge
“All you need are those top quartile teachers”:
Ibid.
“development and evaluation systems”: Vicki
 Phillips, interview by the author, October 14,
2013.
“who aren’t up to the job”: Barack Obama,
“Remarks by the President on Education”
(speech, US Department of Education,
Washington, DC, July 24, 2009), White
House,
http://www.whitehouse.gov/the_press_office/R
by-the-President-at-the-Department-of-
Education.
“based on performance,” meaning evaluation:
See for example, New York State’s Race to
the Top, Panel Review by Applicant for New
York, Phase 1,
http://www2.ed.gov/programs/racetothetop/p
applications/score-sheets/new-york.pdf,
accessed September 2013.
banned assessing teachers by students’ test
scores: US Department of Education, “Final
Priorities, Requirements, Definitions, and
Selection Criteria,” Federal Register 74, no.
 221 (November 2009): 59692.
that led several states to revise their laws:
Associated Press, “States Change Laws in
Hopes of Race to Top Edge,” January 20,
2010.
could be denied tenure or fired: Corinne Herlihy
et al., “State and Local Efforts to Investigate
the Validity and Reliability of Scores from
Teacher Evaluation Systems,” Teachers
College Record (forthcoming).
“the right people standing in front of the
classroom”: Gordon, Kane, and Staiger,
“Identifying Effective Teachers,” 5.
“a teacher shall not be reemployed”: Herlihy et
al., “State and Local Efforts,” 17.
“very specific and actionable feedback to
teachers”: Ibid.
after their teachers received focused evaluations:
Eric S. Taylor and John H. Tyler, “Can
Teacher Evaluation Improve Teaching?”
Education Next, Fall 2012.
the skills students needed for English class: The
description of the PLATO group’s research
findings draws on Pam Grossman et al.,
“From Measurement to Improvement:
Leveraging an Observation Protocol for
Instructional Improvement” (paper presented
at the annual meeting of the American
Educational Research Association, April 20,
2013).
could make words sing: The preceding sequence
draws on interviews by the author with
Lorraine McCleod in March 2013, and on
observations in her classroom.
the average PLATO score had significantly
improved: Grossman et al., “From
Measurement to Improvement,” 12–17.
from math to English to history to science:
Heather C. Hill and Pam Grossman,
“Learning from Teacher Observations:
Challenges and Opportunities Posed by New
Teacher Evaluation Systems,” Harvard
 Educational Review 83, no. 2 (Summer
2013): 379.
the strategies that students needed to master:
Elizabeth Green, “Gates Foundation Study
Paints Bleak Picture of Teaching Quality,”
GothamSchools, January 6, 2012,
http://gothamschools.org/2012/01/06/gates-
foundation -study-paints-bleak-picture-of-
teaching-quality.
a better representation of dividing fractions:
Barbara Scott Nelson, Virginia C. Stimpson,
and Will J. Jordan, “Leadership Content
Knowledge for Mathematics of Staff
Engaged in Key School Leadership
Functions” (paper presented at the
University Council of Education
Administration annual meeting, November
2007).
25 percent made only cursory references: Lynn
T. Goldsmith and Kristen E. Reed, “Final
Report: Thinking about Mathematics
 Instruction,” NSF grant EHR 0335384 (in
preparation), cited in Hill and Grossman,
“Learning from Teacher Observations.”
impressively predictive of a teacher’s
performance: Jonah Rockoff et al.,
“Information and Employee Evaluation:
Evidence from a Randomized Intervention in
Public Schools,” American Economic
Review (forthcoming).
who were destined for effectiveness: The
preceding calculations draw from Sean P.
Corcoran, Can Teachers Be Evaluated by
Their Students’ Test Scores? Should They
Be? The Use of Value-Added Measures of
Teacher Effectiveness in Policy and
Practice, Education Policy Action Series
(Providence, RI: Annenberg Institute for
School Reform at Brown University, 2010).
“high-support policies that . . . help teachers
learn”: Hill and Grossman, “Learning from
Teacher Observations,” 382.
They had to probe: Deborah Loewenberg Ball and
Francesca Forzani, “The Work of Teaching
and the Challenge for Teacher Education,”
Journal of Teacher Education 60: 497–511.
“it’s individual, work on it, figure it out”:
Remarks by Deborah Loewenberg Ball at the
launch of the Sposato Graduate School of
Education, Boston, MA, September 21,
2012.
“feedback” and coaching in addition to
evaluation: Bill Gates, “Teachers Need Real
Feedback” (speech, TED Talks Education,
May 2013), TED,
http://www.ted.com/talks/bill_gates_teachers
Colleen Walsh, “Changing How Teachers
Improve: Emphasis on Bettering
Performance Rather Than Simply Rating
Success,” Harvard Gazette, February 3,
2011,
http://news.harvard.edu/gazette/story/2011/02/
how-teachers-improve.
“let them be themselves”: Philip K. Howard,
“Free the Teachers,” New York Daily News,
November 28, 2010.
supported implementation of the new standards:
Brian Smith, “Common Core Standards
Funding Officially Blocked in New Michigan
Budget after Senate Vote,” MLive.com, June
4, 2013.
“I don’t know what is”: Sher Zieve, “Common
Core Forcing Marxism/Nazism on America’s
Children,” Canada Free Press, May 9, 2013.
 Index

Page numbers listed correspond to the print

edition of this book. You can use your device’s
search function to locate particular terms in the
text.

“academic character education,” 226, 227

academic discourse, 264–67, 268, 271–72, 304,
305, 318
see also class discussions
academic research, 20–21, 26, 30–31, 88–89, 135,
147, 272, 276, 278, 279, 284, 291, 293, 294,
332n
of Ball, 61–65, 78–79, 102, 148, 221, 222, 225,
255, 274, 276, 280, 282, 284–85
entrepreneurs seeking perspectives from leaders
in, 252, 253, 255–72
Hamilton Project report in, 290, 291–93, 294,
296, 297, 298, 299–300
 of Hanushek, 41–44, 150, 290, 292, 328n
inconclusive findings in, 24–25, 27, 36–37, 84
lack of communication between teachers and
leaders of, 125–27
of Lampert, 20–21, 65, 67–69, 92, 102, 221,
222, 225, 247–48, 249–50, 251, 253, 272,
274, 276, 284, 320
of Lanier, 83–84, 91, 101, 107
MET project in, 298, 301, 303, 304, 305
natural-born-teacher fallacy in, 6–7, 24, 44, 294
on pedagogical content knowledge, 61–63, 78–
79, 221, 222, 225, 247–48, 255, 272, 280,
284–85, 306
“process-product” experiments in, 28–30, 31,
32, 36–37
teacher decision-making studies in, 35–36, 37,
38, 39
TIMSS study in, 116–23, 144, 145–46, 147,
164, 194, 226
see also educational psychology
academic “rigor,” 229, 244–46, 250, 252, 264,
 273, 274–77
Academy of the Pacific Rim (APR), 156, 157–61,
163–67, 168, 169, 170, 199–202, 204n, 216–
17, 218
discipline and order enforced at, 158–61, 168,
199–202, 203–4, 216–17
high attrition rate at, 203–4
accountability movement, 12, 13, 15, 17–18, 43–
44, 70, 146–47, 237–38, 286–305, 310–12
entrepreneurial education movement influenced
by, 154–55, 156, 169–71, 181, 279, 290, 295
limits to approach of, 170–71, 181
performance incentives in, 70, 169, 181, 287,
291, 293, 295, 296, 297, 298, 299
“value-added” calculation and, 43–44, 287–94,
296, 297
see also evaluations, teacher; standards
movement
Achievement First charter network, 167, 180, 239,
242–46, 250–52, 295
Adney, Joseph, 52–54
Albom, Mitch, 318
Alphonso (college student, pseudonym), 151–52,
153, 154, 155, 156
Altman, Jay, 153, 162, 167–68, 169
Alum Rock Union Elementary School District,
231
American Educational Research Association, 30
American Federation of Teachers, 30
American Psychological Association (APA), 207,
208
America’s Choice, 241
Amistad Academy, 167, 169
anonymous individual correction, 185–86
apologies, public, 210–11, 216
Aspire Public Schools, 239, 264, 267, 272
“Assume the Best” technique, 188, 190
Atkins, Norman, 180
attrition rates, 203–4, 216, 345n
autonomy thesis, 12–13, 15, 17–18, 311

Ball, Deborah Loewenberg, 45–67, 68, 70–71,

 78–79, 80, 87, 93, 98–101, 102, 104, 107,
109, 113, 122, 137, 147, 150, 157n–58n,
162n, 163, 172, 194, 195, 206, 211, 213,
246, 264, 274, 276, 277, 279, 280–82, 293,
294, 310–11, 316, 333n–34n, 335n, 338n
academic background of, 45–46, 48, 52, 60–61,
84, 308
baking used as teaching tool by, 47–48
in Japan, 148–49
pedagogical-content-knowledge research of,
61–63, 78–79, 221, 222, 225, 255, 280, 284–
85
restructuring of teacher education by, 281–82,
284–86, 291, 296–97, 307–10, 311–13
summer teaching labs of, 54–60
TKOT method of, see TKOT approach
videotaped lessons of, 68, 69, 71–78, 79, 92,
100–101, 280, 314, 336n
Ball, Richard, 47, 48
bansho (“board writing”), 122, 136–38, 141, 148
Barker, Lisa, 270–71, 272
basal readers, 100
Bass, Hyman, 69, 70–71, 73–74, 75–76, 77, 78,
79, 87, 147–48, 255, 280, 336n
behaviorism, 31–32, 36, 37, 38, 88, 273–75, 276
Bell, Joseph, 34
Bell, Terrel, 90, 91
Bench, The (disciplinary tool), 209–10, 215, 217
Berliner, David, 23
Billups, Lovely, 30, 235
Blanchard, Jim, 91
Blasdale, Spencer, 160, 161, 163, 166
Boston, Mass., 154, 156, 167, 169, 196, 203, 229,
251, 255–56, 285
Boston Collegiate, 180
Boston Latin Academy, 203
Boston Prep, 166
Boston Public Schools, 200
Boston Teacher Residency (BTR), 252, 255–64
Boyd, Stacey, 153–54, 156, 157, 158, 159–61,
163–64, 166, 168, 200, 203
Brazil, 88–89, 274
Briggs, LeBaron Russell, 26–27
Brighter Choices charter school, 191
Brillante, Jaimie, 186–87
broken-windows theory, 159–60, 200, 207
Bronson, Po, 168
Brookings Institution, 290
Brophy, Jere, 201
Brown, Linda, 159, 160, 168
Bruner, Jerome, 38–39
Bryk, Anthony, 250, 251, 252
Buckingham Browne & Nichols private school,
67, 68, 86
Building Excellent Schools fellowship, 168
Burns, Marilyn, 296
Bush, George W., 169, 288

California, 12, 13, 41, 42, 153, 218, 231, 235,

238, 264, 276, 287, 296
Education Department of, 102, 105–6
TKOT model attempted in, 102–7, 122, 123,
147, 150, 195, 234
California Dreaming (Wilson), 106
Carter, Samuel Casey, 162n
Carver, Ilene, 256
Cassell Elementary School, 230
Chalkbeat, 314
“challenges,” in class discussions, 271
charter schools, 155–73, 180, 196–99, 200, 204–
7, 229, 239, 252, 290, 294–95, 311
academic performance and student success at,
156n, 168–69, 252
academic rigor issue at, 244–46, 250, 252, 264,
274, 277
accountability and standards at, 156, 169–71,
181, 242, 279, 290, 295
attrition rates at, 203–4, 216, 345n
culture of collaborative learning in, 164–68,
172, 180, 243
discipline and order stressed at, see discipline,
at charter schools
educational infrastructure built by, 241, 242–
46, 250, 295, 310, 312
 funding of, 250–51
growth and expansion of, 180–81, 197, 251,
279, 312
jugyokenkyu technique adopted by, 164–65,
193, 195
Lemov’s taxonomy techniques deployed at,
189–92, 193, 197, 273–74
metrics analyzed by, 156–57, 169–70, 181,
343n
principal training programs of, 168
recruiting of teachers for, 180–81, 240, 251,
293, 295
teacher education and training at, 229, 242,
243, 245–46, 278, 285–86, 312
test scores emphasized in, 155, 156, 169, 242,
244, 245, 246
see also entrepreneurial education movement;
specific charter schools
Chetty, Raj, 293, 294
Chicago, Ill., 20, 123, 124, 131, 138
Cultural Center in, 14–15
Chicago, University of, 20, 81, 113, 124, 225,
335n
Chicago State University, College of Education at,
7
Chicago Teachers Union, 13
China, 114, 201, 341n
zuanyan jiaocai tradition in, 144
Choices (disciplinary action), 210–11, 216
“Choral Counting” (IA), 255
Cichon, Karen, 174, 175, 176–79, 181, 190
City on a Hill charter school, 252
class discussions, 36, 51, 92, 100–101, 171, 248,
286, 306
“anticipation guides” in, 270
balancing gender participation in, 16
on behavior, 211–12
in charter schools, 168, 171
conjectures in, 53, 58, 60, 61, 65, 69, 72, 74
diagnosing student mistakes and thought
processes in, 4–5, 9–11, 57, 58, 59, 72–74,
76, 94–95, 96–97, 101, 104, 105, 119, 254,
 284
difficulties and obstacles in facilitating of, 93–
95, 97, 104–5, 132–33, 137, 171, 269, 318
in English lessons, 264–67, 268–72, 302, 305,
318–19
Grossman’s taxonomy of English teaching and,
267–68, 271–72
guiding class to correct answer in, 11, 16–17,
57, 58, 77, 95, 96–97, 116, 119, 192, 254
IA concepts and, 254–61
in Japanese lessons, 119–21, 128, 129–30,
132–35, 138–41, 278
in math lessons, 3–6, 9–11, 16–17, 51–54, 57,
58–59, 60, 65, 71–78, 79, 92–97, 104, 105,
119–21, 129–35, 168, 255, 256–61, 262
“modeling” in, 267–68, 270, 271, 284, 318
pre-lesson teacher prep work for, 99–100, 131,
267, 269, 316–18, 322
“stock responses” for teachers in, 269
student “turns” in, 16, 37, 58
training teachers in facilitating of, 100, 128,
 132–34, 254, 255, 260–61, 269–72, 284,
304, 305, 310
“wait time” paradox in, 37
war on shyness in, 137, 269, 270
see also academic discourse; TKOT approach
Coburn, Warren, 87
cognitive psychology, 32–40, 88–89, 332n
Cohen, David, 86, 92, 101, 102, 103–4, 105, 106,
112, 122, 233–34, 235–41, 243–44, 246,
250, 280–82, 291, 311, 313
coherence problem, 233, 234–38, 239, 240, 242,
250, 281, 291, 297, 310
“Cold Call” technique, 192
Coleman, James, 41, 43, 44
Columbia University, 69, 78
Teachers College at, 25
Common Core standards, 283, 291, 295–97, 311
Comprehensive School Mathematics Program
(CSMP), 334n
Congress, U.S., 288, 289
conjectures, 53, 58, 60, 61, 65, 69, 72, 74, 276
“consequence,” as level of behavioral intervention,
187–88, 191
“Conversation Rebuilding” (IA), 254
cooperative learning groups, 103, 104
Corcoran, Sean, 307
“core practices,” for teaching English classes,
267–68, 271–72, 284
Correia, Jonathan, 203
Courant, Paul, 280
Culture, Literacy, and Learning (Lee), 222–23
culture conversations, 211–16, 218, 227
curricula, 13, 235, 236, 237, 239, 241, 245, 281,
282–83, 291, 310
Ball in drafting of, 284–86, 309–10, 311–13
experimental, 50–51, 52, 60, 92, 334n, 338n
in Japan, 144
in teacher-training programs, 26, 84–85, 285,
309–10, 311–13

Dash, Julie, 223–24

Davis, George, 187
Deming, William Edwards, 123, 144–45
Democrats, 154, 237, 287
Dewey, John, 20, 26, 28, 29–30, 33, 39, 81, 122,
124, 127, 129, 135, 250
Dialogues in Mathematics Education (DIME),
125
Dilit, 253–54, 255, 278
discipline, 158–60, 181, 197–99, 203, 207, 215,
219–29
“broken windows” theory of, 159–60, 200, 207
operant conditioning in, 221–22
pedagogical content knowledge in teaching of,
221–28
“social problem solving” and, 214, 222
“zero-tolerance” approach to, 172, 207–8
see also misbehavior, student
discipline, at charter schools, 158–62, 168, 172,
181, 197, 198, 199–205, 206–7, 208–20,
226, 228–29, 273
APR and, 158–61, 168, 199–202, 203–4, 216–
17
 The Bench as tool of, 209–10, 215, 217
on bus routes, 204–6, 209
Choices as form of, 210–11, 216
culture conversations in facilitating of, 211–16,
218, 227
evolution of, 209–16, 217–19, 220, 227–28,
312
failure of, 219–20, 298
“inappropriate” and extreme examples of, 198–
99, 208–9
Lemov’s taxonomy and, 219–20, 221, 228
as progress trap, 206–7, 209
public apologies and, 210–11, 216
Rise Academy and, 204–6, 209–16, 217
SLANT position and, 183, 185, 187
unintended negative consequences of, 197,
202–4, 206–7, 209, 210
“Do It Again” technique, 191, 273
donyu (“lesson opener”), 148
dress codes, 159, 161, 168, 172
Driggs, Colleen, 182–84, 185, 193, 197
Dye, Thom, 65–66

Easley, Jack and Elizabeth, 125

East Lansing, Mich., 45, 46, 61, 234
see also Spartan Village school
education:
in Japan, see Japanese education
medical, 34
science of, see science of teaching
of teachers, see teacher education and training
see also specific school subjects
education, American, 116, 124–27, 146, 149, 194,
226, 239, 269, 287, 311
architecture of classrooms in, 121–22, 157
charter schools in, see charter schools
coherence problem in, 233, 234–38, 239, 240,
242, 250, 281, 291, 297, 310
fight over federal vs. state control of, 234–35,
237, 240, 283, 311
frequent lesson interruptions in, 117–18, 126,
194
 “gifted” programs in, 18–19, 156
inequity in, 17–18, 40–41, 42, 43, 144, 151–52,
153–55, 158, 240
infrastructure as lacking in, 236, 238, 239, 240,
281, 311
Japanese vs., 118–21, 193, 194
lack of communication between educators in,
125–27, 163
lesson structure in, 118–19, 120–21, 194
poor international test scores and, 70, 89–90,
114–15, 144, 287
reform movements in, see education reform
size of teacher workforce in, 12, 14, 27, 328n
teacher lesson mistakes in, 124, 171
see also teachers, teaching; specific school
subjects
educational infrastructure, 236, 237, 238, 239–46,
280, 281–84, 291, 297
academic rigor issue and, 244–46, 250, 252,
264
American school system as lacking in, 236,
 238, 239, 240, 281, 311
Ball’s efforts in creating of, 281–82, 284–86,
291, 296–97, 307–10, 311–13
of charter network, 241, 242–46, 250, 295, 310,
312
Common Core standards and, 283, 291, 295–
97, 311
definition of, 236
difficulties in creating high-level teaching at
scale in, 181, 244, 246–48, 249–50, 281,
310, 312
flaws in AF’s model of, 244–46, 250
in foreign countries, 240
Rosenberg’s research on, 241–46, 250, 295
TFA in creation of, 238–40, 242
three key elements of, 281–84
educational psychology, 23, 25–33, 150, 272–76
behavioral perspective in, 28–30, 31–32, 36–
37, 38, 88, 273–75, 276
character-trait hypothesis in, 24, 26, 27, 319
cognitive perspective in, 32–33, 35–36, 37–40,
 88–89
disregard and abandonment of teaching in, 25–
27
Gage’s draft conference agenda on, 30–31
inconclusive research findings in, 24–25, 27, 84
natural-born fallacy as hindrance to field of, 24,
26
see also pedagogical content knowledge;
science of teaching
Education and Race (Hanushek), 43, 44
Education Department, U.S., 170
education reform, 80, 83, 89–91, 196–97, 246–48,
284
accountability movement in, 12, 13, 15, 17–18,
70, 146–47, 154–55, 169–71, 181, 237–38,
279, 286–95, 310–12
autonomy thesis in, 12–13, 15, 17–18
business and political elites’ perspective on,
89–90, 146
in California, 102–7, 122, 123, 147, 150, 195,
234, 235
“cooperative learning groups” in, 103, 104
educational inequity issue and, 17–18, 40–41,
43, 144, 151–52, 153–55, 156, 158, 240
entrepreneurial education movement in, see
charter schools; entrepreneurial education
movement
failed experiments in, 87, 102–12, 122, 147,
150, 169, 195, 234, 246, 284, 339n
infrastructure and, see educational
infrastructure
Johnson administration approach to, 40–41,
154–55
in math lessons and teaching, 70–71, 87, 90–
91, 102–107, 122, 123, 136, 146, 195, 240,
243, 246, 283, 296; see also TKOT approach
natural-born teacher fallacy as impediment in,
13, 19
No Child Left Behind and, 91, 146, 237–38,
287, 288, 289, 290
NTCM standards in, 122, 123, 124, 127, 136,
147, 195
 quantitative metrics in, 155, 169–70, 181, 194,
343n
resistance to change in, 105, 108–10, 111–12,
234
standards movement in, 70, 71, 91, 146–47,
155, 170, 237–38, 242, 246, 283, 291, 295–
97, 311, 342n
“teacher quality” issue in, see “teacher quality”
issue
in teacher-training programs, see teaching
reform
testing emphasized in, 13, 146, 155, 156, 169,
237–38, 242, 244–45, 246, 288, 311
TKOT approach in, see TKOT approach
U.S.’s declining academic performance and, 70,
89–91
Education Resource Strategies, 295
education schools, 7, 20, 61, 67, 68, 78, 80–88,
107, 109, 150, 160, 232, 235, 240, 246, 264–
65, 267, 268–69, 280, 281
entrepreneurial movement’s rejection of, 194–
 95, 228
faculty recruitment at, 25, 82, 85, 101, 107
in Japan, 127–30, 131, 132–35, 142
lab schools, 80–81, 85, 101, 124, 125, 127–30,
131, 132–35, 144, 147, 193; see also
professional development sessions/schools
“progressive pedagogy” approach taught at,
157–58, 159
“science of pedagogy” initially ignored in, 25–
27, 61, 86
student teaching programs in, 85, 86, 108, 128,
129–30, 132
see also teacher education and training; specific
schools and universities
Elementary Math Lab, 55–60, 335n
elementary schools, 27, 61, 107, 124, 143, 178,
288
in Japan, 115–16, 125, 127, 130, 131, 135, 136,
137, 341n
see also charter schools; specific elementary
schools
Elguero, Mariel, 215–16, 220, 314
Elliott, Richard, 26
Elstein, Arthur, 34, 35, 332n
Emerson, Mindy, 45, 46, 48, 163
engagement, student, 138–41, 228
England, 114
English education and lessons, 100, 146, 166,
194, 218, 222–27, 228, 244, 252, 254, 264–
72, 276, 277, 283, 301, 302, 305, 307
class discussions in, 264–67, 268–72, 302, 305,
318–19
“core practices” for teaching of, 267–68, 271–
72, 284
lesson planning for, 224, 315, 316–18
pedagogical content knowledge in teaching of,
221, 225–27, 306, 317
PLATO approach in, 302–3, 305
entrepreneurial education movement, 21, 150–51,
154–73, 193, 197, 239–40, 264, 278, 279,
293
academic perspectives sought in, 251, 252, 253,
 255–72
accountability and standards theories as
influence on, 154–55, 156, 169–71, 181,
242, 279, 290, 295
education schools rejected by, 194–95, 228
linear model of learning adopted in, 272–74,
275–76, 295
“no excuses” approach in, 159–162, 162n–
163n, 169, 172, 175, 181, 183, 194, 195,
197, 199, 202–204, 206–207, 208, 217, 219,
239; see also discipline, at charter schools
“progressive pedagogy” approach rejected in,
157–58, 159
progress traps in, 206–7, 209
quantitative metrics used in, 155, 156–57, 169–
70, 181, 194, 343n
quick and firm correction of student mistakes
in, 273–74, 278
teaching-reform efforts in, 157–158, 161, 162–
168, 169–72, 175, 194–195, 238–239, 241–
246, 250–252, 253, 255–264, 272, 273–274,
 278–279, 281, 285, 291, 293, 295, 312; see
also “Taxonomy of Effective Teaching
Practices”
see also charter schools
epistemology, 33, 39, 150, 276
see also cognitive psychology
Esquith, Rafe, 218
evaluations, teacher, 12, 13, 47, 282, 283, 292,
293, 295, 297–308, 310–11, 320
backlash to, 310–11
as diagnostic and development tools, 300–301,
303–5, 307, 310
MET project for redesigning of, 298, 301, 303,
304, 305
PLATO rubric appropriated for use in, 300–
301, 303–5
“Race to the Top” emphasis on, 298–99
as sorting tools, 12, 293, 297, 299–300, 308
“value-added” scores as hindering effectiveness
of, 306–7
“Everybody Writes” technique, 188
expulsion rates, 200, 207

Farr, Steven, 15
Federal Reserve Bank of Boston, 287, 288
Feiman-Nemser, Sharon, 335n
Feinberg, Mike, 167, 172, 208
Ferdinand, Sabine, 256–64
Finland, 13, 144
Ford Foundation, 82, 85
Forzani, Francesca, 84–85, 107, 108, 112, 285,
297, 308, 309–10, 311, 328n
Friedman, John, 293, 294
Fry, Jessie, 46–47, 48, 109, 110–12, 169, 333n–
34n, 339n
Fujii, Toshiakira, 142, 145, 193
Futabakai School (Chicago), 123, 124
fuzoku schools, 127–30, 131, 132–35, 193, 341n

Gage, Nathaniel, 21, 23–25, 27–31, 32–33, 36–

39, 44, 83, 84, 88, 272
Gallagher, Judith, see Lanier, Judith
Gates, Bill, 297–98, 300, 304, 310
Gates Foundation, 297, 298, 300–301, 304, 310
Germany, 70, 117, 120
“gifted” programs, 18–19
Gist, Sylvia, 7
Gladwell, Malcolm, 292, 293
Glendenning, Kate, 166
Goldstein, Mike, 166
Gordon, Robert, 290, 292, 293, 294, 298, 299
grammar, 87, 254, 271
Grande, Shannon, 211, 212–14, 218, 295, 314
Grossman, Pam, 264–65, 267, 270, 272, 274, 276,
277, 279, 300–305, 307, 311, 312, 316, 317
taxonomy of English teaching developed by,
267–68, 271–72, 284

Hamilton Project, 290, 291–93, 294, 296, 297,

298, 299–300, 308
Handbook of Research on Teaching, The (“The
Gage Handbook”), 30
Hannaway, Jane, 7
Hanushek, Eric, 40–44, 70, 146, 147, 154, 155,
 286–87, 291, 310
teacher-effectiveness studies of, 41–44, 150,
290, 292
“value-added” calculations of, 43–44, 150,
287–94, 297, 333n
Harvard Business School, 161, 164, 169
Harvard University, 41, 91, 101, 153, 160, 252,
310
Graduate School of Education at, 67, 86
Heaton, Ruth, 91–98, 100, 102, 105, 106, 109,
215, 338n
Hiebert, James, 146–47
“high-leverage practices,” 284–86, 309–10
high schools, 143, 265, 297, 314, 315
Hill, Heather, 307, 311
Hill, Ryan, 295
Hirayama, Mr., 138–41, 142, 341n
Hirsch, Emil G., 20
history lessons, 87–88, 128, 194, 220, 221, 264,
305, 306
Hoffman, Marvin, 335n
Holmes Group, 101
Hopkins, Millisent Fury, 199
How Children Succeed (Tough), 215
How to Solve It (Polya), 135–36
Hutchinson, Yvonne Divans, 269–71, 305

Idiokitas, Chimel, 200–201, 203, 204

Illinois, University of, 23, 27, 125
incoherence, law of, see coherence problem
Indiana University, 151–52, 161
Institute for Research on Teaching (IRT), 67, 80,
85, 91, 284
instructional activities (IAs), 253–55, 256–61,
267, 284, 310
interim assessments, 242, 243
international tests, 70, 89–90, 114–15, 116–17,
144, 145, 287
Italiaidea, 248–55, 278

Jackson, Julie, 167–68

James, William, 25, 27, 81
Japan, 2, 70, 113–14, 117, 118, 123, 125, 126,
 129, 138, 150, 195
Deborah Loewenberg Ball in, 148–49
education schools and teacher training in, 127–
130, 131, 132–135, 193; see also
jugyokenkyu (“lesson study”)
elementary schools in, 115–16, 125, 127, 130,
131, 135, 136, 137, 341n
kaizen (“continuous improvement”) concept in,
144–45, 164–65
math achievement scores in, 114–15, 144
Ministry of Education in, 123, 124, 143
Japanese education, 113, 115–16, 122–23, 125,
129–46, 147–49, 164, 193, 194, 226, 240,
243, 250, 311, 314
architecture of classrooms in, 121–22
bansho (“board writing”) in, 122, 136–38, 141,
148
charter schools as adopting lesson study
techniques of, 164–65
“check status” question in, 120
class discussions in, 119–21, 128, 129–30,
 132–35, 138–41, 278
lesson structure in, 116, 118, 119, 133, 134,
140–41, 148, 194, 276–78; see also
jugyokenkyu
lower-income vs. affluent students in, 144
NCTM standards as influential in, 123, 136
similar approach to TKOT model in, 113, 116,
119–21, 122–23, 129–30, 131, 132–35, 136,
137, 138–42
textbooks in, 142–43, 144
U.S. vs., 118–21, 193
Johnson, Lyndon, 40, 41, 154–55
Judge, Harry, 82, 84, 85, 107
jugyokenkyu (“lesson study”), 126–27, 132–33,
134–37, 138, 140, 142, 143, 144–46, 147,
148, 192, 193, 195, 242, 341n
charter schools as adopting similar technique
to, 164–65, 193, 195
public lessons and, 127–30, 131, 132, 135, 136,
137, 138, 140–42
kaizen (“continuous improvement”), 144–45,
164–65
Kane, Tom, 287–94, 300, 310
as critical of “value-added” calculation, 287–
89, 306, 308
Hamilton Project report of, 290, 291–93, 294,
297, 299, 308
MET project and, 298, 301
Kelling, George L., 159
kikanjunshi, 148, 277–78
kikan-shido, 277, 278
King, John, 153, 161–62, 166, 167, 180
KIPP charter schools, 167, 169, 180, 198, 208,
209, 217, 237, 251, 273, 290, 295
KIPP Academy, 172
KIPP Fresno, 218
KIPP Infinity (Harlem), 295, 296
Rise Academy, 204–6, 209–16, 217, 218, 220,
295, 312
Kirkpatrick, Heather, 264, 267, 272
Knowledge Is Power Program, see KIPP charter
 schools
Koganei Elementary School, 340n
Kohlberg, Lawrence, 208
Kopp, Wendy, 153, 230
Kristof, Nicholas, 291
Kristol, Irving, 154

lab schools, 80–81, 85, 101, 124, 125, 144, 147,

193
in Japan, 127–30, 131, 132–35
see also professional development
sessions/schools
Lampert, Magdalene, 6, 7–11, 16–17, 20–21, 65–
69, 79, 80, 85–86, 91–96, 97, 98, 101, 102,
107, 109, 112, 113, 122, 123, 147, 157n–
58n, 162n, 172, 194, 195, 211, 213, 233,
246–52, 274, 275, 276, 277, 279, 280, 284,
311, 316, 320, 335n, 338n
BTR program and, 255–56, 258, 260, 261, 262,
263–64
classroom management challenges as handled
 by, 226, 227, 228
diagnosing source of student mistakes by, 8–11,
65
IAs concept and, 253, 255, 256, 260–61, 267,
284
at Italiaidea, 248–50, 251, 252, 253
at NewSchools Venture Fund conference, 250–
51, 252
pedagogical-content-knowledge research of, 68,
221, 222, 225, 247–48, 272
Summer Learning Institute and, 251, 255–56
“teacher quality” issue and, 246–48, 249–51,
252, 253
TKOT approach of, see TKOT approach
videotaped lessons of, 68–69, 92, 100–101,
113, 116, 314
language lessons, 128, 248–50, 251
Lanier, Judith, 80–87, 88, 91, 101–2, 107–9, 112,
150, 284, 337n
Lanier, Perry, 334n
Leadership Prep school, 187
learning:
general laws of, 26, 172
linear model of, 272–74, 275–76, 295
see also educational psychology
Learning to Teach summit, 264
Lee, Carol, 222–27, 228
Lemov, Doug, 150–52, 155, 156, 157, 158, 159,
161, 163–67, 169–73, 174–76, 180–81, 182–
85, 197–98, 199, 201, 203, 217, 218, 219,
228, 230, 239, 241, 272, 280, 293, 295, 316,
320, 343n
“build it/buy it” problem of, 181, 251, 252
“click” moment of, 153–54
on importance of “love” in teaching, 220, 320,
321
School Performance company of, 169–70, 172,
174
taxonomy project of, see “Taxonomy of
Effective Teaching Practices”
“Taxonomy 2.0” of, 278–79, 312
Lesley College, 67
lesson laboratories, 54–60, 67–68, 100, 335n
lesson plans, 46, 112, 131, 132, 136, 171, 224,
228, 242, 243, 296, 315, 316–18, 322
lesson structure, 116, 118–21, 133, 134, 141, 248,
249, 276, 318, 322
in Japan, 116, 118, 119, 133, 134, 141, 148,
194, 276, 278
single-problem format in, 93, 100–101, 119,
120, 131, 136, 138, 142–43
Levin, David, 167, 172, 217, 219
Lewis, Michael, 155, 317, 318
Life, 79
lightning-quick public correction, 187
linear model of learning, 272–74, 295
Lortie, Dan, 15–16, 163
Los Angeles Unified School District, 269, 289

madogiwa (“window gazers”), 131

Martin, Drew, 198–99, 204–7, 209, 216, 217,
218, 219, 295
Match charter schools, 166, 229
Match Teacher Residency, 229, 285–86
math education and lessons, 1–6, 8–11, 49–67,
69–79, 87, 88–90, 91–101, 102, 113, 124,
125, 126, 128, 147–48, 158, 167–69, 199,
221, 226, 227, 228, 248, 252, 253, 261–64,
274, 280, 305
Ball’s summer teaching labs on, 54–60
bar graphs in, 138–41
class discussions in, 3–6, 9–11, 51–54, 57, 58–
59, 60, 65, 71–78, 79, 92–97, 104, 105, 119–
21, 129–35, 168, 255, 256–61, 262
conjectures in, 53, 58, 60, 61, 65, 69, 72, 74,
276
diagnosing source of student mistakes in, 4–5,
8–11, 56–57, 58, 59, 61–63, 72–74, 76, 79,
94–95, 96–97, 101, 104, 105, 119
experimental curriculum in, 50–51, 52, 60, 92,
334n, 338n
finding visual representatives for teaching of,
63–65, 79, 134–35, 138–40, 296
guiding class to correct answers in, 11, 16–17,
 57, 58, 77, 95, 96–97, 119
IA concepts for, 255, 256–61
international achievement race in, 70, 89–90,
114–15, 144
in Japan, see Japanese education
NCTM standards in, 122, 123, 124, 127, 136,
147, 195
pedagogical content knowledge in teaching of,
61–63, 78–79, 221, 247, 255, 272, 280, 284–
85, 306
Polya’s problem-solving manual and, 135–36
proofs in, 10, 16, 53, 61, 73, 74, 76, 77, 121
reform efforts in, 70–71, 87, 90–91, 102–7,
122, 123, 136, 146, 195, 240, 243, 246, 283,
296
regrouping lessons, 103, 104, 132, 142, 276
textbooks in, 142–43, 144, 245
TKOT approach for teaching of, see TKOT
approach
U.S. students as falling behind in, 70, 89–91,
114–15
 U.S. vs. Japanese lesson structures in, 118–21
Mathematical Knowledge for Teaching (MKT),
78–79, 255, 280, 285, 306, 316
Mathews, Jay, 208
Matsuyama, Takeshi, 127, 128–29, 130, 131–32,
138, 193
McCaslin, Mary, 201
McCue, Scott, 158, 161, 162
McDaniels, Garry, 30, 31, 38
McIntosh, Amy, 14
McLeod, Lorraine, 301–3, 304
Measures of Effective Teaching (MET) project,
298, 301, 303, 304, 305
medical decision-making, 34–35, 36, 37, 38, 39,
83, 332n
memorization, 273, 274, 275–76, 295
Michigan, University of, 78, 112, 114, 233, 251,
280, 284, 309, 336n
School of Education at, 7, 78, 282, 309
Summer Learning Institute at, 251, 255–56
Teacher Education Initiative, 282–83, 284–86,
 309
Michigan State University (MSU), 31, 33–34, 38,
46, 52–54, 63, 65, 66, 68, 70, 80, 82, 86,
99n, 100, 105, 106, 111, 112, 147, 150,
157n, 233, 246, 264, 280, 281, 335n
College of Education at, 50, 61, 65, 82, 83–86
Elementary Intern Program of, 45–46, 47, 48,
49, 84
Institute for Research on Teaching (IRT) at, 67,
80, 85, 91, 284
Lanier’s failed reform efforts at, 85–87, 88, 91,
101–2, 107–9, 284
Mieze, Rousseau, 196–97, 198, 199–204, 216–21,
228–29, 244, 285
minicomputers, 338n
misbehavior, student, 175–76, 182–88, 192, 208,
209, 215, 228, 229, 273
on bus routes, 204–6
culture conversations for handling of, 211–16,
218, 227
Lampert’s approach in handling of, 226, 227,
 228
Lemov’s six levels of intervention for, 185–88,
191, 192
as “maladaptive coping strategy,” 225
nonverbal intervention for, 183–84, 185, 192
see also discipline
mistakes, student, 50, 77
diagnosing sources of, 4–5, 8–11, 56–57, 58,
59, 61–63, 65, 72–74, 76, 79, 94–95, 96–97,
101, 104, 105, 119, 225
and guiding class to correct answer, 11, 16–17,
57, 58, 77, 96–97, 277
Lemov’s “Punch the Error” response to, 273–
74, 278
as opportunities for learning, 11, 57, 76, 77,
277–78, 279
quick and firm correction of, 273–74
mitori (“observing students”), 148
“modeling,” 267–68, 270, 271, 284, 286, 302,
306, 310, 318
Moffitt, Susan, 236
Most Precious Blood parochial school, 157
Mumford, Venecia, 286

Nagano school, 136, 137

National Assessment of Educational Progress, 155
National Council of Teachers of Mathematics
(NCTM), 122, 123, 124, 127, 136, 147, 195
National Governors Association, 283
National Institute of Education (NIE), 30–31, 35,
38, 39, 67
Nation at Risk, A (study), 90, 91, 115
natural-born-teacher fallacy, 6–7, 9, 11, 13, 18,
19–20, 24, 26, 44, 46, 294, 308, 319
Negron, Joe, 295–97
neriage, 148
“New Math” program, 87
New Orleans Charter Middle School, 162, 167,
169
NewSchools Venture Fund conference, 250–51,
252
New Teacher Project, The, 292
New York (state), 294, 307
New York, N.Y., 14, 153, 292, 314
charter schools in, 167, 168, 170, 172, 197,
295, 296
Education Department of, 14
Nixon, Richard, 30, 39
No Child Left Behind (NCLB), 91, 146, 237–38,
287, 288, 289, 290
No Excuses (Carter), 162n
“no-excuses” movement, see entrepreneurial
education movement
No Limits bus, 204–6
nonverbal intervention, 183–84, 185, 192
“Normalizing the Error” technique, 278
North Star Academy, 167, 169, 180

Obama, Barack, 11–12, 13, 17, 290, 298

“100 Percent” technique, 177, 181–82, 184, 192
operant conditioning, 221–22
Opportunity bus, 204, 205
O’Reilly, Jane, 152, 153
O’Toole, Stacy Birdsell, 228–29, 244, 285
Oublier, Mrs. (pseudonym), 102–4, 105, 106–7,
122, 235

Papy, Georges, 334n, 338n

Papy-Lenger, Frédérique, 334n, 338n
Parker, Francis, 19–20, 21, 39, 81, 82
Pastore, Patrick, 184, 185, 197
pedagogical content knowledge, 49, 61–63, 68,
78, 247, 255, 272, 280
principals as lacking in, 306
in teaching discipline, 221–28
in teaching English classes, 221, 225–27, 306,
317
in teaching math, 61–63, 78–79, 221, 247, 255,
280, 284–85, 306
pedagogy, 25, 27, 49, 55, 66, 100, 116
“progressive,” 157–58, 159
see also educational psychology; science of
teaching
Peiser, Brett, 166, 167, 180
performance incentives, 70, 169, 181, 287, 291,
293, 295, 296, 297, 298, 299
Phillips, Alexander, 202
Phillips, Josh, 174–75
Phillips, Vicki, 298
Piaget, Jean, 259
place value lessons, 103, 104, 263
PLATO (Protocol for Language Teaching
Observations), 300–305, 350n
Polya, George, 122, 124, 127, 129
problem-solving manual of, 135–36
“Positive Framing” technique, 188, 190
“post,” in class discussions, 271
poverty, 17, 40, 41, 43, 154–55, 156, 158, 215,
231, 291
Practical Introduction to Evaluation and
Measurement, A (Gage and Remmers), 23
“practice-based teacher education” movement,
312
praise, giving of, 188
“press,” in class discussions, 271, 304
Princeton University, 153, 161, 192
principals, 46–47, 124, 143, 166, 168, 170–71,
175, 176, 202, 228, 234, 244, 285–86, 306
“inappropriate” discipline employed by, 198–
99, 204
see also Fry, Jessie
private individual correction, 186–87
private schools, 15, 67, 163
“process-product” paradigm, 28–30, 31
contradictory research findings in, 36–37
professional development sessions/schools, 11–
12, 54, 99, 101, 106, 107–9, 110, 143, 167–
68, 243, 272, 281, 310
see also lab schools
“progressive pedagogy” movement, 66, 157–58,
159
progress traps, 206, 209
psychology, 26, 31, 32, 84, 150, 272
public apologies, 210–11, 216
public lessons, 54–60, 127–30, 131, 132, 135,
137, 138, 140–42
 see also lab schools
public schools, 11, 15, 70, 82, 112, 222, 256, 279,
314
attrition rates of, 203
chaotic atmosphere of, 158, 159
federal funding of, 41, 154–55, 156
low-income vs. affluent, 17–18, 19, 40, 144
student enrollment numbers in, 27, 241
suspension and expulsion rates in, 200, 207
“zero-tolerance” approach in, 207–8
see also education, American; charter schools;
specific schools
“Punch the Error” technique, 177, 178, 181

quantitative metrics, 155, 156–57, 169–70, 181,

194, 343n
“Quick Images” (IA), 256–61

“Race to the Top” grant program, 298–99

reading education and lessons, 18, 48–49, 51,
100, 114, 158, 172, 228, 268, 276, 283, 305
national achievement levels in, 155, 240
Reddy, Ranjana, 204, 207, 208, 210, 211
“restatement,” in class discussions, 271
“revoicing,” in class discussions, 271, 304
Ribicoff, Abraham, 40
“Right Is Right” technique, 177, 178, 181
Rise Academy, 204–6, 220, 295
evolution of discipline at, 209–16, 217, 218,
312
Rochester Prep, 180, 182, 184, 186, 188, 197
Rockoff, Jonah, 293, 294
Romer, Roy, 247
Rosenberg, Seneca, 230–32, 233, 235–37, 238–
40, 241–46, 250, 251, 294–95, 328n
Rosenshine, Barak, 28–29
Rowe, Mary Budd, 37
Roxbury Prep, 161–62, 166, 167, 168–69, 174–
75, 180
Royce, Josiah, 27
Rudall, Evan, 161–62, 166, 167, 180
Rundquist, Sylvia, 98–100, 105, 106, 215
Sanders, William, 44n
Sax, Cantor, Riff (film), 223–24
Scholastic Aptitude Test, 23, 292, 300
School of the Future, 315
School Performance, 169–70, 172, 174
Schoolteacher (Lortie), 15–16
“school-to-prison pipeline,” 208
Science Curriculum Improvement Study, 334n
science education and lessons, 49, 51, 100, 114,
125, 128, 194, 221, 228, 264, 305, 306, 334n
science of teaching, 20, 26–27, 31–32, 39, 87,
188, 272
initial disregard of, 25–27, 81–85, 86
“process-product” paradigm in constructing of,
28–30, 31, 32, 36–37
see also educational psychology; teacher
education and training
Setagaya Elementary School, 127–30, 131, 132,
147–48
Setagaya Ward, Tokyo, 127, 132
Short History of Progress, A (Wright), 206
Shulman, Lee, 31, 32–40, 44, 67, 70, 80, 84, 85,
101, 234, 264–65, 272, 281, 284, 286, 337n
medical decision-making studies of, 33–35, 36,
38, 39, 83, 332n
“pedagogical content knowledge” concept of,
63, 78, 221, 317
Silberman, Charles, 88
Singapore, 144, 240, 341n
single-problem format, 2, 3, 53–54, 59, 79, 93,
100–101, 116, 119, 120, 131, 136, 138, 142,
253, 277
Skinner, B. F., 26, 31
SLANT position, 183, 185, 187
Snow, Richard, 31, 32
Snyder, Andy, 314–15, 316–18, 320–21, 322
Solomon, Jesse, 251–52, 253, 255, 256, 264
Spartan Village school, 45, 46–48, 49, 53, 55, 60,
61, 65, 66, 84, 96, 98, 107, 113, 163, 215,
246–47, 251, 280, 281, 286, 313, 338n
teaching labs project at, 67–68, 100, 101, 109,
110, 172
 TKOT reforms as unsustainable in, 109–12,
169, 234, 339n
Spencer, Margaret Beale, 225
Spillane, Jim, 240
Sposato, Charles, 285–86
staff meetings for teachers, 109–10, 111, 157
Staiger, Douglas, 288, 289, 290, 291, 292, 293,
294, 299
standardized tests, 43–44
SAT, 23, 292, 300
standards movement, 70, 71, 91, 146–47, 155,
237–38, 242, 311, 342n
Common Core standards in, 283, 291, 295–97,
311
entrepreneurial education movement influenced
by, 156, 170, 290
performance evaluations in, see evaluations,
teacher
testing emphasized in, 13, 146, 155, 156, 169,
237, 242, 244–45, 288, 311
see also accountability movement
Stanford University, 30, 31, 32, 35, 36, 38, 80,
122, 264–65, 332n
STAR position, 183
STEM fields, 70
Stevenson, Harold, 114, 115, 164
Stigler, James, 113–14, 115–16, 122, 123, 145,
146–47, 155, 164
TIMSS study led by, 117–18, 119–21, 122,
144, 145–46, 147, 164, 194, 226
Storey-Fry, Jessie, see Fry, Jessie
“Stretch It” technique, 177, 178–79
student teaching, 85, 86, 108, 128, 129–30, 132,
265
study groups, 135–36
Success for All, 241
Summer Learning Institute, 251, 255–56
suspensions, 200, 205, 207, 217
Suzuka, Kara, 69
Sykes, Gary, 36
Syracuse, N.Y., 170, 172, 175, 177, 181
Takahashi, Akihiko, 123–32, 135–38, 142, 145,
193, 195
Taubman, Alfred, 90, 91, 108, 112
“Taxonomy of Effective Teaching Practices,” 150,
174, 175, 176–79, 180, 181–82, 184–95,
197–98, 219, 230, 239, 242, 273, 280, 319
approach to student mistakes in, 273–74, 278–
79
for managing classroom culture and behavior,
219–20, 221, 228
as tool in teacher-training reform, 181–82, 190–
91, 192, 193, 272, 273–74, 278–79
video analysis used in creation of, 179–80, 181,
182, 184, 187, 189–91, 192
“Taxonomy 2.0,” 278–79, 312
teacher education and training, 7, 20, 25, 27, 43,
60–61, 67, 70, 80, 84, 91–92, 105, 229, 231,
236, 251, 252, 256, 268–69, 280, 281–82,
285–86, 291
in charter networks, 229, 242, 243, 245–46,
278, 285–86, 312
in facilitating class discussions, 100, 128, 132–
34, 254, 255, 260–61, 269–72, 284, 304,
305, 310
inconsistencies of instructional guidance in,
230–32, 233, 235, 238
at Italiaidea, 250, 251, 253–54, 255, 278
in Japan, 127–130, 131, 132–135, 136–137,
193; see also jugyokenkyu
learning TKOT approach in, 93–95, 98–99,
100–101, 107
marginalization of, 25–27, 81–85, 86
mentors in, 15, 129, 131–32, 142, 232, 242,
243
MSU’s Elementary Intern Program for, 45–46,
48, 49
pedagogical content knowledge as important in,
61–63, 68, 78–79, 221, 225, 226, 227–28,
247, 255, 272, 284–85, 306
professional development sessions/schools in,
54, 99, 101, 106, 107–9, 110, 111–12, 243,
272, 281, 310
 public lessons and teaching labs in, 54–60, 67–
68, 80, 81, 100, 101, 127–30, 131, 132–35,
137, 138, 140–42, 144, 147, 193, 335n
reform efforts in, see teaching reform
see also education schools
“teacher quality” issue, 11–14, 15, 17–18, 19, 40,
147, 181, 249–52, 281–82, 283–86, 290,
294–95, 297–313
accountability argument in, 12, 13, 15, 17–18,
43–44, 70, 181, 287–305, 310–12
Ball’s efforts in creating infrastructure and,
281–82, 284–86, 291, 296–97, 307–10, 311–
13
complexity of, 246–48, 253
and difficulties in creating high-level teaching
at scale, 181, 244, 246–48, 249–50, 281,
310, 312
Gates and, 297–98, 300, 304, 310
Hamilton Project and, 290, 291–93, 294, 296,
297, 298, 299–300, 308
Hanushek’s teacher-effectiveness studies and,
 41–44, 150, 290, 292
in low-income vs. affluent schools, 17–18
performance evaluations in addressing of, 292,
293, 295, 297–308, 310–11
training-reform efforts and, see teaching reform
“value-added” calculation and, 43–44, 283–94,
296, 297, 306–7, 308
teachers, teaching:
building strong relationships between students
and, 220–21, 286
certification-exam pass rates for, 294
decision-making studies on, 35–36, 37, 38, 39
diagnosing source of student mistakes in, 4–5,
8–11, 56–57, 58, 59, 61–63, 65, 79, 94–95,
96–97, 101, 104, 105, 119, 225
essays published by, 143
evaluations of, see evaluations, teacher
importance of “love” in, 220, 320, 321
lack of communication between, 125–27, 163
leading class discussions in, see class
discussions
Lemov’s taxonomy project on, see “Taxonomy
of Effective Teaching Practices”
“modeling” in, 267–68, 270, 271, 284, 286,
302, 306, 310, 318
myth of natural-born, 6–7, 9, 11, 18, 19–20, 24,
44, 294, 308, 319
parental preferences in, 232, 233
parent communication and relationships with,
131, 160, 166, 232, 285–86
pre-lesson prep work of, 99–100, 131, 220–21,
267, 269, 315, 316–18, 322
as private “ordeal,” 15–16, 163, 243
“quality” reform debate in, see “teacher
quality” issue
recruiting of, 14–15, 25, 180–81, 240, 242,
293, 295
reform efforts in, see teaching reform
salaries of, 43
science of, see science of teaching
size of workforce in, 12, 14, 27, 282, 328n
soft bigotry showcased by, 177
 student misbehavior and, see misbehavior,
student
as studied craft, 9, 19–21, 81, 321
tenure of, 83, 91, 107, 171, 293, 299
TKOT approach in, see TKOT approach
turnover in, 232–33
“value-added” calculation for measuring of, 43–
44, 150, 287–94, 296, 297, 306–7, 308, 333n
volunteer study groups of, 135–36
see also education, American; specific school
subjects and teachers
teacher’s guides, 51, 92, 95, 106
Teachers Institute, 252
Teach for America (TFA), 15, 167, 172, 182, 206,
230–32, 236, 281, 285
educational infrastructure created in, 238–40,
241, 242
recruiting of teachers by, 240, 242
Teaching as Leadership framework of, 242
Teaching, Multimedia, and Mathematics
(Lampert and Ball), 335n
Teaching and Learning in Japan (Rohlen and
LeTendre), 164
Teaching Gap, The (Stigler and Hiebert), 146–47
teaching reform, 83, 85, 86, 87–89, 101–2, 106,
107, 123, 147, 150, 181, 237–38, 241, 250,
251, 252, 278, 279, 280, 281–313
Achievement First efforts in, 242–44, 245–46
Ball’s efforts in creating infrastructure for, 281–
82, 284–86, 291, 296–97, 307–10, 311–13
Common Core standards in, 283, 291, 295–97,
311
in entrepreneurial movement, 157–158, 161,
162–168, 169–72, 175, 194–195, 238–239,
241–246, 250–252, 253, 255–264, 272, 273–
274, 278–279, 281, 285, 291, 293, 295, 312
facilitating academic discourse and, 264–67,
269–72, 304, 305, 310
as “fuzzy math,” 122, 283
Grossman’s “core practices” for English
teaching and, 267–68, 271–72, 284
“high-leverage practices” in, 284–86, 309–10
 IAs concept in, 253–55, 256–61, 267, 284, 310
Lampert’s complicated proposal for, 246–48
Lanier’s ill-fated ambitions in, 85–87, 88, 91,
101–2, 107–9, 284
Lemov’s taxonomy as tool in, 181–82, 190–91,
192, 193, 272, 273–74, 278–79
parental relationship building in, 285–86
PLATO training sessions in, 300–305, 350n
resistance to change in, 105, 108–10, 111–12
TKOT approach in, see TKOT approach
see also education reform; “teacher quality”
issue
Teaching with Problems and the Problems of
Teaching (Lampert), 247–48
TeachingWorks, 309–12
Teach like a Champion (Lemov), 228, 278
tenure, 83, 91, 107, 171, 293, 299
tests, testing, 43–44, 146, 155, 158, 236, 237,
240, 241, 242, 281, 291–92, 299
academic rigor problem and, 244–45
charter students’ performances on, 168–69, 252
 emphasis on, in reform movements, 13, 146,
155, 156, 169, 237–38, 242, 244–45, 246,
288, 311
international scores in, 70, 89–90, 114–15,
116–17, 144, 145, 287
NCLB law’s role in creation of new, 237–38
scoring gap between black and white students
in, 289
textbooks, 100, 105–6, 131, 142, 144, 235, 240,
245, 268, 295
in Japanese education, 142–43, 144
medical, 34, 35
Thai, Kevin, 200, 202, 204, 217
Third International Mathematics and Science
Study (TIMSS), 116–23, 144, 145–46, 147,
164, 194, 226
“This Kind of Teaching” (TKOT), see TKOT
approach
Thorndike, Edward, 25–26, 27, 31, 81–82, 273
Title I, 154–55
TKOT approach, 68, 93–97, 100–107, 147, 157n–
 58n, 162n, 211, 213, 216, 250, 281, 312
academic rigor achieved in, 276–77
conjectures as “fence posts” in, 58, 66, 72, 137
diagnosing student mistakes and thought
processes in, 4–5, 8–11, 56–57, 58, 59, 61–
63, 65, 72–74, 76, 79, 94–95, 96–97, 101,
104, 105, 119, 225, 284, 285
engaging students in, 138–41
failed attempts in implementing of, 102–12,
122, 147, 150, 169, 195, 234, 246, 339n
and Grossman’s “core practices” for English
teaching, 267–68, 271–72
guiding class to correct answer in, 11, 16–17,
57, 58, 77, 95, 96–97, 116, 119, 254
IAs in, 253–55, 256–61, 267, 284, 310
importance of listening in, 94–95, 97, 277, 319
Japanese teaching techniques as similar to, 113,
116, 119–21, 122–23, 129–30, 131, 132–35,
136, 137, 138–42
methods for facilitating discussion in, 133–35,
269–72
 mistakes as opportunities for learning in, 11,
57, 76, 77, 277
single-problem approach in, 2, 3, 59, 79, 93,
100–101, 116, 119, 120, 131, 136, 138, 142,
253, 277
teacher preparation required in, 99–100, 131,
267
training teachers in use of, 93–97, 98–99, 100–
101, 107
use of term, 66–67, 335n
videotaped lessons on, 68–69, 71–78, 79, 92,
100–101, 113, 116, 314, 336n
visual representations used in, 63–65, 79, 134–
35, 138–40, 296
see also class discussions
Tokyo, 125, 127, 148, 195, 314, 341n
Tokyo Gakugei University, 127, 135, 142, 341n
Toll, Dacia, 167
Tough, Paul, 215
Toyota, 165
Tschang, Chi, 168, 198–99, 202, 204, 218, 219
tsumazuki, 148
Tuesdays with Morrie (Albom), 318
“turnaround” schools, 256

UChicago Charter School, 335n

Uncommon Schools charter network, 180–81,
239, 250, 295
unions, teacher, 30, 110, 290, 293
“uptake,” in class discussions, 271–72, 304, 305

“vagueness dictionary,” 29
“value-added” calculations, 43–44, 150, 287–94,
296, 297, 306–8, 333n

Wakabayashi Elementary School, 341n

war on poverty, 154–55, 291
welfare programs, 154, 155
Whitehead, Alfred North, 87
Williams, Darryl, 191, 193
Williams College, 153, 196, 202
Williamson, Peter, 267, 271
Wilson, James Q., 159
Wilson, Suzanne, 87, 106
Wood, Molly, 166
word problems, 50
Work Hard. Be Nice (Mathews), 208
World Association of Lesson Study conference
(2011), 341n
Wright, Ronald, 206
writing lessons, 18, 158, 231, 268, 276, 277, 302,
316

“zero-tolerance” approach, 207–8

Zimmerli, Bob, 188–91, 193, 197
 Author’s Note

All children’s names in this book have been

changed to protect their identities. Except where
indicated otherwise in the Notes section, I
provided these pseudonyms.
 Copyright © 2014 by Elizabeth Green

All rights reserved

First Edition

Distance charts from Magdalene Lampert’s

Teaching Problems and the Problems of
Teaching © 2001 are used with permission of
Yale University Press

Fraction boxes courtesy of Deborah Loewenberg

Ball & Meghan Shaughnessy.
Copyright (2012). Reproduced with permission of
Mathematics Teaching and
Learning to Teach Projects.

Dividing circles courtesy of Deborah Loewenberg

Ball. Copyright (1990). Reproduced
with permission of Mathematics Teaching and
Learning to Teach Projects.

Multiplication diagrams courtesy of Deborah

 Loewenberg Ball. Copyright (1989).
Reproduced with permission of Mathematics
Teaching and Learning to Teach Projects

For information about permission to reproduce

selections from this book,
write to Permissions, W. W. Norton & Company,
Inc.,
500 Fifth Avenue, New York, NY 10110

For information about special discounts for bulk

purchases, please contact
W. W. Norton Special Sales at
[email protected] or 800-233-4830

Book design by Chris Welch

Production manager: Louise Parasmo

ISBN 978-0-393-08159-6
ISBN 978-0-393-24415-1 (e-book)

W. W. Norton & Company, Inc., 500 Fifth

 Avenue, New York, NY 10110
www.wwnorton.com

W. W. Norton & Company Ltd., Castle House,

75/76 Wells Street,
London W1T 3QT