Becoming a Better College Teacher - Specialized Research AI at MIT

Becoming a Better College Teacher
(If You’re Lucky)

Harry Brighouse

This contribution is a narrative of how a professor attempted to improve as
a teacher over time. The narrator noticed the need for improvement through
teaching a new class badly, and learning that he had no reason to trust that he,
or many others, were teaching other classes better. The contribution describes
in some detail the steps he took, and continues to take, including observing col-
leagues, hiring a coach, reviewing videos of his classes, and participating in de-
partment workshops. There is no empirical evidence that he has improved, but
the narrator provides some reasons for optimism.

I magine that you call a plumber.

The plumber, a new hire, has never done any plumbing. She has nev-
er read any books about plumbing or attended any classes about how to
fix pipes or faucets or toilets or garbage disposal units. The house she grew up
in had running water, so it’s not as if she knows nothing about plumbing. And,
incidentally, she has been in the same room with some professional plumb-
ers when they were working. Unfortunately, she never saw the results of their
work; she always left before the water was turned back on, was not privy to re-
ports about whether the pipes and faucets subsequently leaked, and didn’t ask
the clients how satisfied they were with the outcomes. On further question-
ing, though, she reveals to you that she is actually a highly skilled baker.

On calling the firm you discover that they routinely hire new plumbers with
no experience or training, and don’t seek evidence about their potential to be-
come good plumbers. You learn that all the frontline employees are experts at
something else: they are trained as electricians, ice-sculptors, coopers, roof-
ers, literary critics, physicists, and more. But not as plumbers.

The firm does not assess employee performance based on results. Clients
fill in a short “customer satisfaction” form before the water is turned back on.
But, unless a plumber regularly receives truly awful ratings, they just file the
forms away. Pay raises are related to neither the results of the plumbing nor
the customer satisfaction ratings. After six years of employment, the firm fires

© 2019 by Harry Brighouse
Published under a Creative Commons
Attribution 4.0 International (CC BY 4.0) license
https://doi.org/10.1162/DAED_a_01758

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
4
8
4
1
4
1
8
3
1
3
5
1
d
a
e
d
_
a
_
0
1
7
5
8
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

some of the underperformers and gives unparalleled job security to the oth-
ers. Curiously, the main criteria for promotion, pay raises, and even job securi-
ty concern excellence in whatever they are actually trained in. Not plumbing.
You then discover that, far from being an outlier, this firm has exactly the
same hiring and promotion practices as all the other plumbing firms. You ask
your neighbors about their plumbers. Some have only had terrible experienc-
es, though many blame themselves. Several report one plumber for whom
they have the greatest praise because the drinking water was mainly clean and
nothing leaked again for days. And, rather surprisingly given the circumstanc-
es, a few can identify a single plumber whose work was impeccable and whom
they expect to remember for the rest of their lives.

N ow imagine that you are the plumber. If you are a professor in a re-

search university, it shouldn’t be too much of a stretch. I certainly
identify with her predicament. Of course, the analogy is highly im-
perfect. Baking is entirely unrelated to plumbing, whereas knowing how to do
philosophy is a prerequisite for being able to teach philosophy, and at least I
was trained in that. I had been the target of numerous attempts, many of them
successful, to get me to learn. So professors at least have something to build on.
But being able to do philosophy is only a prerequisite for being able to teach
it, and being a student does not automatically give one insight into teaching.

Most professors in research universities teach. Even the small proportion
whose research funding generates consistent “buy-outs” are hired on a tenure
line, which, at least after they get tenure, provides security; they keep their
jobs and salaries even if they don’t win grants. Most professors have received
little to no training as teachers, were hired for their potential not as teachers
but as researchers, and receive promotions and pay raises mainly for their per-
formance in research rather than in teaching. Once someone has tenure, their
teaching must generate numerous complaints in order for it to have negative
professional consequences of any significance. Few professors engage system-
atically in ongoing professional learning as instructors: they don’t read books
about teaching and learning, they don’t seek out more successful teachers and
observe them, and they don’t engage colleagues or professional observers to
help them improve their own instruction. Ask ten professors in research insti-
tutions–those who are expected to split their efforts equally between teach-
ing and research–how many of the last ten conferences and workshops they
attended and how many of the last ten publications they read were primarily
concerned with teaching rather than research. I predict that of those two hun-
dred conferences, workshops, and publications, fewer than one hundred will
be about teaching. I’d be surprised if the number were as high as ten.

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
4
8
4
1
4
1
8
3
1
3
5
1
d
a
e
d
_
a
_
0
1
7
5
8
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

148 (4) Fall 2019Harry Brighouse

To state the obvious, the plumber’s incentives are all wrong: she is reward-
ed for her performance in something other than plumbing, despite the fact
that plumbing is her job. Professors in research universities are paid to re-
search and teach, but they, naturally, take the research more seriously because
they have trained in it and know they will be rewarded for it. Administrators
should change the incentives, thus creating self-interested reasons for profes-
sors to take teaching more seriously.1

But assume they don’t. Most professors already have non-self-interested
reasons to take teaching more seriously: both their students and the public
suffer from the effects of suboptimal teaching. And many professors like to
think of themselves as capable of making choices that align somewhat with
the general good, rather than entirely with their own self-interest. How can
they improve?

T he first stage in recovery is to admit there is a problem. The structure

of the profession makes the problem rather obvious, when you think
about it, but for many years, I didn’t. If you have fairly good command
of the material you are teaching, are okay at explaining things, have some pa-
tience, and have a friendly affect, you can go a long time without realizing you
are not teaching well. Add in the English accent at a Midwestern university,
and you might never notice at all. I was shaken from my complacency only by
the confluence of two events.

First, a friend sent me chapter six of former Harvard President Derek Bok’s
Our Underachieving Colleges to read for a research project we were planning.
Here’s the passage that made me blanch with embarrassment and immedi-
ately purchase the book to read in its entirety:

Teaching by discussion can also seem forbidding because it makes instructors un-
comfortably aware of their shortcomings. Lecturers can delude themselves that
their courses are going well, but discussion leaders know when their teaching is
failing to rouse the students’ interest by the indifferent quality of responses and
the general torpor of the class. Trying to conduct a discussion with apathetic stu-
dents is much like giving a bad dinner party.2

I was accustomed to talking a lot in my large lecture classes (with eighty-
plus students), knowing that the students (mainly juniors and seniors) could
punctuate my lecture by answering my questions, and to presenting materi-
al in my smaller classes for philosophy majors, knowing that those students
would regularly interrupt with queries. I was used to good student evaluations
of my teaching because, well, I am moderately well-organized, I key my talk to
the material they should have read, I’m reasonably friendly, and they like my

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
4
8
4
1
4
1
8
3
1
3
5
1
d
a
e
d
_
a
_
0
1
7
5
8
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

Dædalus, the Journal of the American Academy of Arts & SciencesBecoming a Better College Teacher (If You’re Lucky)

accent. But the passage hit home because I recognized my own talk as a way of
evading responsibility of ensuring they were fully engaged, and it crystallized
that the more I talk, the less I know what is going on in the students’ heads. I
wondered whether my high student evaluations might reflect the soft bigot-
ry of low expectations.3

Still: nothing might have changed had I not, that fall, been teaching, for the
first time, a First-Year Interest Group (FIG) seminar. The FIG program induces
groups of twenty first-year students to take three thematically linked courses,
one of which is a seminar just for them, together in the same semester. I taught
the central, twenty-person seminar on “Children, Marriage, and the Family,”
which attracted students with ambitions to become early childhood educa-
tors, nurses, social workers, and clinical psychologists, not philosophers.

The first few weeks of class were . . . awful. The readings were too difficult,
I had assigned too much, and while I talked from my carefully prepared notes,
the students stared in silence, trying to take notes, and wondering what on
Earth was going on. They were aliens who, as far as I could tell, might be think-
ing just about anything. How could I figure out what was going on in their
heads? And, until I did, how could I calibrate my talk to their learning needs?

At last I understood there was a problem.

E ven without Bok’s book, I would have known something was wrong

with the class. It was that bad. But I might well have persisted; I’d have
had nothing else to do. Knowing something’s going badly is good. But
it does not, in itself, spur improvement. I was motivated. But improvement re-
quires access to knowledge about how to do better. And when it comes to col-
lege instruction, gaining knowledge is not straightforward.

How do people learn complex skills? Think about playing the guitar. The
aspirant guitarist observes (and listens to) expert guitarists. She seeks out in-
struction. She tries to mimic some of what the experts do. She gets feedback–
some from her own ears and some from other people–then tries again. Then
she observes and listens again, mimics again, and gets more feedback. This is
roughly what professors do when they are learning to become, and trying to
improve as, researchers. As graduate students, they take seminars in which
they are inducted into the practices of research, and various skills are modeled
by their teachers and advisors. They read vast amounts of research by other,
already accomplished researchers and try to emulate their efforts. They pre-
sent at conferences and get feedback on their work from their colleagues and
teachers. As professors, they continue observing other researchers in their
field, interact with and learn from them, and continually seek feedback from
peers so that they can maintain and improve their skills.

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
4
8
4
1
4
1
8
3
1
3
5
1
d
a
e
d
_
a
_
0
1
7
5
8
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

148 (4) Fall 2019Harry Brighouse

Professional learning needs an infrastructure. Aspiring guitarists have that: gui-
tar teachers, teach-yourself books, videos on YouTube of excellent guitarists,
and so on. Researchers, too: graduate school, feedback on grant and paper
submissions, specialized workshops and conferences, and department col-
loquia. But college teachers don’t. Unlike guitarists and researchers, college
teachers aspiring for excellence can’t even readily identify who the existing
experts are. Most disciplines lack both rigorous measures of student learning
and a systematic practice of trying to evaluate instructor quality. I don’t know
which of my colleagues are more successful than I am at producing learning
among their students over the course of a semester, let alone who is success-
ful at getting students to think better about trolley problems, or to understand
the purpose of thought experiments in ethical theorizing. The sparse profes-
sional learning resources around instruction are mostly generic: it is not ob-
vious how to apply lessons about pedagogy drawn from physics or mathemat-
ics to my own field. I’m a fairly typical professor in that I was enculturated
into a specific discipline (philosophy) and mostly teach within my broad field
of specialization (ethics, applied ethics, political philosophy). I want to learn
how to teach that better, to my students.

W hat next? It started with a book. So I read more books about col-

lege (and secondary school) teaching and learning in the hopes
of finding useful information. I wasn’t seeking some master plan
that would transform my teaching; I guessed it would be useful to find out
what is known about good and not-so-good teaching, generally. No literature
provides any precision on how to teach students effectively how thought ex-
periments work, let alone how to teach it to my students. But plenty provide
useful information about student learning, about the habits of successful col-
lege teachers, and, generally, about successful techniques. Shortly after I expe-
rienced my discomforting epiphany, my wife became involved in high school
improvement, through which I learned that many of the resources produced
for high school teachers can also be valuable for college teachers.4

It has been well worth devoting a good deal of time to reading about teach-
ing and learning, and I continue to do so avidly. But imagine learning to play
the guitar, or tennis, or to bake cakes, or to fix pipes by just reading a book.
The next move was to get feedback on my efforts to improve.

T he 2007 first-year seminar that forced me to face up to my inadequa-

cies was not a complete disaster. After three weeks or so, I began mod-
ifying my instruction considerably and spent a lot of time talking to
the students individually and in groups about the class, trying to gauge how

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
4
8
4
1
4
1
8
3
1
3
5
1
d
a
e
d
_
a
_
0
1
7
5
8
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

Dædalus, the Journal of the American Academy of Arts & SciencesBecoming a Better College Teacher (If You’re Lucky)

it was going for them, what was working, and what wasn’t. I devised in-class
exercises to make some sort of discussion happen, and to hold them more
accountable for the (too-difficult and too-voluminous) reading. The break-
through, though, came three years later, when I taught the class for a second
time. By then, I knew several of the 2007 students well, and in the summer of
2010, one of them, Emma, asked if there was anything she could do to help
with the 2010 version of the class. I knew exactly what I wanted from her.

Roger Federer is, reputedly, the greatest male tennis player of all time. But
he still has a coach.5 He’s not an outlier: top athletes and musicians normally
employ coaches to help improve their performance. However good, they need
someone to observe them, identifying strengths and helping them address
weaknesses. Researchers have coaches, too: A good Ph.D. advisor coaches
graduate students. Junior professors typically turn to senior colleagues, who
read their work, give feedback, suggest tweaks, help them uncover new op-
portunities, and advise about publication outlets. Experienced research-
ers have informal coaching networks of colleagues who routinely read their
work, helping them formulate problems and suggesting different techniques.
I asked Emma to be my instructional coach.

The director of the FIG program coincidentally knew Emma and offered us
a $500 budget. Emma’s job was to observe me once a week, take notes for a re- port on what was happening in the classroom, and then debrief for twenty to thirty minutes after class. It was the best use of $500 I’ve ever made.

The main benefit was the day-to-day criticism. Here are some examples:

• Week 2: “The material you’re covering is very challenging for freshmen. It is
good you are challenging them, and this is not too hard, but it would help them
a lot if you would sum up where the lecture and discussion have got to every
fifteen minutes.”

• Week 3: “Well . . . you didn’t do what we said last week.” This was, obviously,
the point at which I knew it was going to work well, because she proved she
would tell me when I was screwing up.

• Week 3: “You’ve had six sessions with them and you still don’t know all their
names. You should know all their names by now.” I knew eighteen of twenty-
two names but kept confusing two in particular, between whom, by the end of
the semester, I could not see the slightest resemblance. By the next session, I
knew all of their names.

• Week 4: “It is ok to cold call–in fact I wish more teachers would cold call. But
you need to tell them in advance that you are going to cold call–Marissa was
really put on the spot today. And when you do cold call, you have to make it

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
4
8
4
1
4
1
8
3
1
3
5
1
d
a
e
d
_
a
_
0
1
7
5
8
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

148 (4) Fall 2019Harry Brighouse

clear that if they don’t have anything to say that is fine.” One of the advantag-
es of her being there only one session a week was that I could refer to her ad-
vice during the subsequent session without her being there: so I asked wheth-
er they agreed, which they all did. I apologized to Marissa and told them that I
would feel free to cold call henceforth.

• The course devotes a couple of weeks to the tensions between multicultural-
ism and feminism. Two of the readings discuss specific practices within the
Hmong community as illustrations–and judge them quite negatively. The
subsequent year’s class, when I employed Emma again, had five Hmong stu-
dents, and she anticipated my anxiety about teaching those papers. My incli-
nation was just to drop those readings. Hers was to assign three of the Hmong
students to present (everyone had to do a group presentation in class). I fol-
lowed her advice. The Hmong students had not been vocal participants, but
when presenting on these readings, they were the experts in the room. The
other students knew even less than I did about Hmong culture and practices;
the Hmong students knew a lot. Incidentally all five Hmong students said the
readings represented their culture accurately, and that the judgments about
the practices in question were fair.

• The 2011 class was 25 percent Hmong, more than 50 percent non-White, and
more than 50 percent low income/first generation. The students would sit
in a crescent formation that was, after a week, more or less a rainbow, with
all but one of the Black students on one end, then Hmong, then Latina, then
White (and at the far end one Black man). Most class sessions involved small-
group discussions of four or five students. I would assign students randomly
to groups so that they would not always be discussing with their friends. But
the consequence in this class was that the loquacious White students were tak-
ing up nearly all of the discussion time within each racially mixed small group,
and then all of it in the full class discussion (because they always volunteered
as group reporters). Emma was able to think through the problem with me
and convinced me that the solution was to create racially homogenous small
groups. Indeed, this led to much more talk–and much more connection to the
class–from the non-White students. (You might ask why I didn’t group the
students according to how well they would work with and learn from one an-
other. That’s a good question and I have a good answer: I didn’t think of it.
Nor was I yet skilled enough to have learned which students would work well
together.)

Emma provided two things that made a big difference. One was a student-
centered perspective: she was only thinking about their learning and how they

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
4
8
4
1
4
1
8
3
1
3
5
1
d
a
e
d
_
a
_
0
1
7
5
8
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

Dædalus, the Journal of the American Academy of Arts & SciencesBecoming a Better College Teacher (If You’re Lucky)

were reacting, so when I was talking with one student, she could be observ-
ing the others and their responses to what was happening. Lacking the content
expertise, she could make judgments about how well they were learning. The
other was just a sounding board. I could pilot a new practice–cold-calling,
new discussion prompts, even new readings–confident that someone was ob-
serving and would actually tell me how well it succeeded or how badly it failed,
helping me think about whether to abandon it or modify it, and if so, how.

Before you try this at home, here’s some background. Emma majored in
nursing, not philosophy: like the students in the class, she was not an expert,
she had no interest in dazzling me with her own brilliance, and we both knew
that if we fell out it wouldn’t affect her professional or academic prospects at
all. She was one of the first students I spoke with when the 2007 class was not
going well, and in the intervening period, we had discussed her experiences as
a learner, both good and bad, in other classes.

But Emma was only an undergraduate student and not herself a great
teacher. Surely Roger Federer wouldn’t hire a twenty-year-old with limited
experience to coach him?

I am not, regrettably, the Roger Federer of college teaching. He is (I’m told)
the greatest (male) tennis player ever, whereas I was, at best, mediocre. Of
course, if I had the option of getting a professional instructional coach to ob-
serve me regularly, I’d jump at it. But I didn’t. Emma was not an expert teach-
er, and had no experience coaching, but–like many of our students–she was
well-positioned to deliberate usefully with me about instruction. As a senior,
she had taken twenty-four college classes, with different instructors of record,
many of them with teaching assistants as well. In those same three years, I had
observed just four teachers, in each case for just a single session, rather than
several times a week for fifteen weeks; she had seen, and thought about, more
teaching in the previous three years than I had in the previous twenty-five.
Since then, several other students have coached me and, starting in fall 2015, a
coach observed every single class I taught for a year. Their feedback has been
invaluable; indeed, so has the built-in requirement to stop and reflect on what
has happened.

Ideally, deans would invest in creating a cadre of skilled instruction-
al coaches. Alternatively, training a cadre of students to provide the service
throughout the college would help improve instruction, and would be an in-
vestment in those students’ futures. If you are a dean, consider those two op-
tions. If you are an instructor, though, don’t wait for the dean to act. Find one
or two thoughtful students with whom you have a good relationship and pay
them (out of your own pocket, if you are in a position to do so) to do for you
what Emma, and others, have done for me.

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
4
8
4
1
4
1
8
3
1
3
5
1
d
a
e
d
_
a
_
0
1
7
5
8
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

148 (4) Fall 2019Harry Brighouse

I regularly get students to observe me now. Someone observed every single

class session I taught during the 2015–2016 academic year. Sometimes
colleagues say, “It’s very courageous of you to ask for feedback.” It isn’t. I

want to improve. They’re undergraduates. I have tenure.

I was also influenced by pedagogy scholar Tony Wagner’s The Global Achieve-
ment Gap.6 In one chapter, he describes a workshop for K–12 teachers. Wagner
(or one of his colleagues) led small-group discussions of video-recordings of
classes. The aim was for previously unacquainted teachers to develop a com-
mon language for discussing instruction, and to come to some sort of interper-
sonal agreement on standards of practice. Like most teachers (and nearly all
professors), his participants had spent very little time observing other teach-
ers and were not practiced in rigorous, detail-oriented discussion of what
works and what doesn’t. Initially, the reactions to what they were observing
were very diverse; there was no agreement about whether what is being done
is good or bad teaching. But over the course of the workshop, the participants
would develop a common understanding, and a language for expressing it.

Not content with only funding Emma, the director of the FIG program facil-
itated biweekly discussions of teaching and learning among instructors from
the program. Typically, five to ten instructors would discuss a problem of prac-
tice such as grading, prompting discussion, or whether to disclose one’s opin-
ions about the controversial issues one is teaching. These discussions went
well but, inspired by Wagner’s book, I wanted to get concrete and discuss actu-
al instruction. If you propose something like this, you must be the first volun-
teer. A professional videographer recorded part of a lesson. More than twen-
ty colleagues attended the discussion (I also invited two students from the
class whom I thought could give context, though mainly I wanted them to ex-
perience a faculty discussion about teaching and learning). By this stage, I had
gained some confidence: all of the students were engaged in the class, I was able
to induce all of them to talk, their presentations and written work were of high
quality, and I believed that there was a good deal of discussion in the classroom.
I was wrong. Sure, during a twenty-minute segment, nearly every student
spoke (even the one student who, when I asked their permission to be record-
ed, had said “That’s fine, but I won’t talk”). But, as one colleague cheerfully
pointed out, it wasn’t a discussion. I’d ask a question, someone would answer
it (either voluntarily or because they were cold-called), and I’d dialogue with
that student. Most of the rest were listening. But the focus was on me, not on
the ideas, and not on each other. It was like a series of ping pong games, in
which each of them was playing with me, but none were playing with each oth-
er. And it was easy to see that I was the person preventing discussion from hap-
pening. My nonverbal cues encouraged them to focus on me, rather than each

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
4
8
4
1
4
1
8
3
1
3
5
1
d
a
e
d
_
a
_
0
1
7
5
8
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

Dædalus, the Journal of the American Academy of Arts & SciencesBecoming a Better College Teacher (If You’re Lucky)

other, and I was too eager to validate what each of them was saying by respond-
ing, rather than opening up the discussion to comments from other students.7
Again, recognizing a problem is only the first step. I didn’t know how to
make good, full-classroom discussions happen. If I’d known how to do that, I
would have been doing it already!

I was being trapped by the impulse Bok identifies that leads many col-

leagues to reject class discussion altogether. Professors value rigor and
know that the best guarantee of optimal rigor is to use all the airspace
themselves. Our talk is rigorous, while our students’ talk is sloppy. But else-
where in this issue of Dædalus, Carl Wieman, echoing former Harvard Pres-
ident Charles W. Eliot, observes:

The most basic principle that every teacher should know about teaching this sort
of thinking is that the brain learns the thinking that it practices, but little else.
To have students learn to recognize relevant features and make relevant decisions
more like an expert in the field, they must practice doing exactly this. The longer
and more intense the practice, the greater the learning.8

In STEM, problem sets and labs go some way to facilitating the necessary
practice. But in the interpretative social sciences and the humanities, students
practice only by writing and by discussing. Reading or listening to someone
talk about philosophy, sociology, literary criticism, or psychology is not prac-
ticing, it is just observing an expert practicing. Nor is taking (often inexpertly
designed) multiple choice tests. Watching Roger Federer play tennis, and an-
swering multiple choice tests about what he does, would not suffice for be-
coming even a modestly competent tennis player. You have to practice. And
then practice more.

We can make students write outside of class and we can make them (pre-
tend to) read. But professors should know that most students will not discuss
the material outside of the classroom because they are not in the habit of doing
so, and even if they wanted to, they can’t because they don’t know their class-
mates. They might come to know their classmates, of course, but only if profes-
sors make that happen inside the classroom. Classroom discussion is essential
for students to master the content and skills we care about; and for that to hap-
pen, the professor must be willing to sacrifice some rigor. My impulse to give rig-
or undue priority over engagement was preventing discussion from happening.
Fortunately, I was able to observe other teachers who did know how to
run an actual discussion. The first time was rather fortuitous. I invited then–
graduate student Paula McAvoy, who had previously been a high school social
studies teacher, to teach my class an issue she had written a paper about. They

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
4
8
4
1
4
1
8
3
1
3
5
1
d
a
e
d
_
a
_
0
1
7
5
8
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

148 (4) Fall 2019Harry Brighouse

were assigned the paper, and Paula trusted them to read it. After making them
introduce themselves by naming something they loved, she spent ten minutes
reviewing the main argument, and then set the students to a complex small-
group discussion assignment that required them to engage with and debate
the ideas and arguments in the text. Students reported back to the full group,
after which Paula led a discussion engaging all twenty students. From observ-
ing her, and other teachers, more often, I’ve learned a great deal about how to
make real discussion happen.

I now give very explicit instructions to students, making clear the expec-
tation that they address their classmates, not me, even though I will usually be
the one setting the agenda and facilitating. Until this is the norm, I frequent-
ly stand behind the student who is speaking so they’re forced to look at other
people as they talk. Discussions usually focus on some problem or prompt that
I have devised, which relates to a problem that arises in the reading, or is di-
rectly about the reading itself, and which is either on a handout or (if it is short)
on a slide. The questions are usually very specific but sufficiently open-ended
that reasonable people can disagree (and about which I anticipate disagree-
ment in the class). I use “think, pair, share” and, in smaller classes, cold-call-
ing liberally, to ensure that all students participate. (Emma was right about
cold-calling; my students seem fine with it after they have come to trust that it
really is okay to say they don’t want to speak just now. One student told me at
her graduation that after the first class session of freshman year, she called her
mother and said “I hate Brighouse and I’m going to drop the class because he
says he’s going to cold-call.” But by graduation, she no longer went bright red
when talking in class. Another student recently told me that cold-calling was
“life-changing” because, having made exactly the same phone call in her own
first week, she now contributes confidently to all her classes. Voluntarily.)

I have to curb my tendency to jump in with either interrupting reassurance
or some interesting, pedagogically valuable comment. So I’ve engineered
some sort of gestalt switch in my head. When a student speaks, instead of
thinking that I am depriving her (of assurance or of some valuable thought)
by not responding, I think to myself that I am depriving her precisely by re-
sponding: preventing interaction with her peers, the reasons they can give to
her, and the opportunity to surprise and be surprised by them. If the conver-
sation ebbs, or if some particular strand is, in my opinion, played out, I step
in to prompt the discussion with further questions, and often with low-pres-
sure cold-calling. My rule of thumb is that, on average, at least four students
should speak before I contribute again.

Running a discussion this way–that is, running a discussion–is more men-
tally taxing than engaging in twenty-one separate consecutive conversations.

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
4
8
4
1
4
1
8
3
1
3
5
1
d
a
e
d
_
a
_
0
1
7
5
8
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

Dædalus, the Journal of the American Academy of Arts & SciencesBecoming a Better College Teacher (If You’re Lucky)

The instructor is simultaneously trying to read twenty-one minds, keep ev-
erything on track, interpret what the students are saying, remember what
needs to be highlighted at the end, and be sensitive to the needs of each stu-
dent (some of whom need drawing out, others reining in). It’s especially dif-
ficult with first-year students who are, at my institution and others like it,
disposed to be deferential and, just because they know less and are inexperi-
enced, have less to say than their older peers.

S ince 2016, my department has held monthly “brown bag” meetings on

teaching and learning. In the wake of an uptick in reported racist inci-
dents on and around campus, our chancellor called for departments to
discuss initiatives around diversity and inclusion. Our response–institut-
ing the meetings about instruction–is not as orthogonal to the call as it may
seem. Instructional quality is the most neglected–and perhaps the most se-
rious–equity issue in higher education. Good instruction benefits everyone,
but it benefits students who attended lower-quality high schools, whose par-
ents cannot pay for compensatory tutors, who lack the time to use tutors be-
cause they have to work, and who are less comfortable seeking help more than
it benefits other students. Philosopher Jennifer Morton, also a contributor to
this Dædalus volume, emphasizes the importance to her first-generation and
low-income students at City College of embodied and engaged interaction
with the professors and with each other in a well-managed classroom:

I often require that my students defend a position in front of the classroom. For
many, this is the first time they have spoken in front of a crowd of students from
differing socioeconomic and ethnic backgrounds. The experience is terrifying, but
as one Latina student told me, even though her face still “lights up red” when she
speaks, she is now able to raise her hand and contribute to class discussions. By
the time that student graduates and walks into her first job interview, she will have
learned to manage her fear of speaking her mind. For students from low-income
families who manage to overcome the tough odds, college is the first place where
they will be asked to defend a position and to engage in vigorous intellectual de-
bate. It is also likely to be the first place where they have to consistently engage with
middle-class students and professors and navigate middle-class social norms.9

More generally, lower-income, first-generation, and minoritized students
are more vulnerable to harm from low-quality instruction because they have
fewer academic resources to fall back on.

The idea behind the faculty discussion group and the brown bag is encap-
sulated by this comment by former University of California, Berkeley, Educa-
tion Dean Judith Warren Little on K–12 school improvement:

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
4
8
4
1
4
1
8
3
1
3
5
1
d
a
e
d
_
a
_
0
1
7
5
8
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

148 (4) Fall 2019Harry Brighouse

School improvement is most surely and thoroughly achieved when: Teachers en-
gage in frequent, continuous and increasingly concrete and precise talk about
teaching practice (as distinct from teacher characteristics and failings, the social
lives of teachers, the foibles and failures of students and their families, and the
unfortunate demands of society on the school). By such talk, teachers build up a
shared language adequate to the complexity of teaching, capable of distinguish-
ing one practice and its virtue from another.10

My department is small: there are about twenty faculty and some thirty
active graduate instructors. Most of my colleagues, like most professors, see
their department as the home of their professional life on campus; professors
are generally skeptical that disciplinary outsiders can provide useful insights
about how to teach our content well. And few professions are more status-
conscious than academia: if you want systematically to change faculty be-
havior, you need to operate in the discipline, and colleagues with high status
must be involved.11 In the typical meeting, one or two people (often one fac-
ulty member and one graduate instructor) present some ideas about a specif-
ic problem of practice (for example, how to make discussions more inclusive,
how to incorporate discussion into large lectures, how to reach absentee stu-
dents, or what sorts of comments are useful when grading papers) and mod-
erate a whole-group discussion. Attendance is voluntary, but the meetings av-
erage about fifteen participants, including graduate instructors, junior profes-
sors, and tenured professors as well as highly respected researchers. Until the
brown bags, my department had, like most departments in research institu-
tions, no formal forum for discussing instruction. Coordinating the meetings
is now recognized as committee service. What we have instituted is imper-
fect, but attendance has not declined over time.

Judith Little continues:

Teachers and administrators frequently observe each other teaching, and provide
one another with useful (if potentially frightening) evaluations of their teaching.
Only such observations and feedback can provide shared referents for the shared
language of teaching, and both demand and provide the precision and concrete-
ness, which makes talk about teaching useful.

We haven’t reached that point. Yet.

I am much more serious about teaching than I used to be. I spend more

time talking with students, and have developed strategies for engag-
ing and reaching out to the less-advantaged students who are much less
likely to seek my support and help than the students for whom the culture of

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
4
8
4
1
4
1
8
3
1
3
5
1
d
a
e
d
_
a
_
0
1
7
5
8
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

Dædalus, the Journal of the American Academy of Arts & SciencesBecoming a Better College Teacher (If You’re Lucky)

academia is a second home. Am I actually a better teacher, though? I think
so. But then I would think that, wouldn’t I? Whether because the learning
we most care about can’t be measured, or because (as I suspect) we just hav-
en’t bothered figuring out how to measure it, we lack high-quality measures
of learning, so I can’t go back and compare the learning that was happening in
my classes before 2007 with the learning that happens now.

That said, suppose that you were choosing between two plumbers from
the remarkable firm I described in the introduction. Here’s all that you know
about them: Both were trained as terrific bakers, and neither has been trained
at all as a plumber. One has simply followed the incentives. The other has read
a good deal about plumbing, regularly observes other plumbers, employs a
plumbing coach, gets colleagues to observe her, and frequently meets with
other smart plumbers who, despite having been hired through a similarly bi-
zarre process, are serious about consciously trying to improve their plumbing
skills. If you wanted a really good cake, you might toss a coin. But if you want-
ed the best chance of getting your pipes fixed, I’m guessing you’d choose the
latter.

about the author

Harry Brighouse is the Mildred Fish Harnack Professor of Philosophy and Carol
Dickson Bascom Professor of the Humanities at the University of Wisconsin-
Madison. He is the author of Educational Goods: Values and Evidence in Decision-
Making (with Helen Ladd, Susanna Loeb, and Adam Swift, 2018) and Family
Values: The Ethics of Parent-Child Relationships (with Adam Swift, 2014) and edi-
tor of The Aims of Higher Education: Problems of Morality and Justice (with Michael
McPherson, 2015).

author’s note

Thanks for comments from Tim Brighouse, Bob Moon, Emma Prendergast,
Diana Hess, Gina Schouten, David O’Brien, Mike McPherson, Sandy Baum,
Lynn Glueck, Chris Bertram, Jennifer Morton, Jennifer Noyes, Paula McAvoy,
the participants in a discussion at the American Academy of Arts and Scienc-
es, and various discussants at Crooked Timber. Thanks also to Alan Sidelle,
Jennifer Jennings, Albertine Schellenberg, Lauren Swance, Katherine Nahn,
Michelle Barichello, and, most of all, to Emma Marston and Greg Smith.

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
4
8
4
1
4
1
8
3
1
3
5
1
d
a
e
d
_
a
_
0
1
7
5
8
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

148 (4) Fall 2019Harry Brighouse

endnotes

1 See Willian G. Bowen and Michael S. McPherson, Lesson Plan: An Agenda for Change in
American Higher Education (Princeton, N.J.: Princeton University Press, 2016); and
Robert Zemsky, Gregory R. Wegner, and Willian F. Massy, Remaking the American
University: Market-Smart and Mission-Centered (New Brunswick, N.J.: Rutgers Univer-
sity Press, 2005).

2 Derek Bok, Our Underachieving Colleges: A Candid Look at How Much Students Learn and
Why They Should Be Learning More (Princeton, N.J.: Princeton University Press,
2006), 125.

3 They did. Maybe they still do.

4 She has subsequently become an instructional coach, but only after the experiment I

go on to describe.

5 See Atul Gawande, “Personal Best,” The New Yorker, September 26, 2011.

6 Tony Wagner, The Global Achievement Gap: Why Even Our Best Schools Don’t Teach the
New Survival Skills Our Children Need–And What We Can Do about It (New York: Basic
Books, 2010).

7 I rewatched it with current students who have recently taken the class and was grati-

fied that they were shocked at how bad it was.

8 Carl Edwin Wieman, “Expertise in University Teaching & the Implications for
Teaching Effectiveness, Evaluation & Training,” Dædalus 148 (4) (Fall 2019).
Charles W. Eliot, in his inaugural address as president of Harvard, explained,
“The lecturer pumps laboriously into sieves. The water may be wholesome, but it
runs through. A mind must work to grow.” Charles W. Eliot quoted in Bok, Our
Underachieving Colleges, 123.

9 Jennifer M. Morton, “Unequal Classrooms: What Online Education Cannot Teach,”

The Chronicle of Higher Education, July 29, 2013.

10 Judith Warren Little, The Power of Organizational Setting: School Norms and Staff Develop-
ment (Washington, D.C.: U.S. Department of Education, 1981), 12–13. Warren is
not saying that the talking and the shared language are, themselves, sufficient for
continual improvement. They are necessary components, and signs, of continuous
improvement.

11 See Mary Sue Coleman, Tobin L. Smith, and Emily R. Miller, “Catalysts for Achiev-
ing Sustained Improvement in the Quality of Undergraduate STEM Education,”
Dædalus 148 (4) (Fall 2019).

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
4
8
4
1
4
1
8
3
1
3
5
1
d
a
e
d
_
a
_
0
1
7
5
8
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

Dædalus, the Journal of the American Academy of Arts & SciencesBecoming a Better College Teacher (If You’re Lucky)
Download pdf