Sunday, August 12, 2018

On assigning grades for class discussion

Grading—that is, giving a numerical rating to our students’ work—is at best a necessary evil. I suppose there may be some teachers who enjoy this aspect of the job, but I haven’t met any.

As far as I can tell, there are three reasons why we think we must grade:

1. To motivate our students to work harder than they otherwise would work. Grades are both carot and stick.

2. To help sort our students into categories for the benefit of society: our grades will help the college admissions people determine who gets into the pre-med program at Harvard, who studies pre-med at Northwestern (#10 ranking according to this website: https://blog.prepscholar.com/best-pre-med-schools), and who goes to plumbing school. If I ever need an appendectomy, I hope the guy who performs my surgery, will not have had grade-inflating teachers who deprived my surgeon of his true calling of unclogging drains.

3. To give the students feedback, so they will know how they are doing and what they need to work on—and how hard they should work on it.

We teachers spend a lot of time grading our students. We agonize and we complain about it. We haggle with students and parents. We condemn grade inflation, and yet we keep giving in to the pressure and our grades keep going up.

And we are not alone in obsessing about grades. Education professors, pundits, bureaucrats and policy makers have a great deal to say on the subject. A very brief internet survey reveals the following headlines:

Confessions of a Grade Inflator: Between the grubbing and the blubbering, grading fairly is just not worth the fight. (Slate)

The Techy Teacher: Rethinking Grading (Educational Leadership)

Why Teachers Secretly Hate Grading Papers: For many, it's the most stressful part of the job -- partly because it's so hard to be fair. (Atlantic) (“Secretly”? Really?)

How to Escape Grading Jail (Chronicle of Higher Education)

Seven Strategies to make Grading Easier (Edutopia)

Grading students’ contributions to class discussions is particularly difficult. A quick perusal of a book put out by the Harvard Business School on “The Artistry of Discussion Leadership” (which has been on my bookshelf, mostly unread, for many years) contains the following paragraph:

Grading classroom participation remained a puzzle. I emphasized the importance of participation—it represented 40 percent of the students’ grade—yet I didn’t quite know what I was grading. How do you grade a good listener? How do you even know if someone is a good listener? How do you grade the overtalkative? If the teacher is the catalyst for discussion, isn’t the participation grade really the teacher’s self-evaluation as well? (59).

I’m hoping my surgeon didn’t have that teacher!

About 15 years ago during our last major curriculum review, the Exeter Curriculum Review Committee gave every member of the faculty a book with a different perspective on grading, and I did happen to read most of that one: Punished by Rewards by Alfie Kohn.

Kohn's book is an argument against grades. He says they are counterproductive, especially in relation to our first reason for grading: motivation. He argues that external rewards teach people that the thing they are being rewarded for isn’t intrinsically rewarding—thus, in the long run, they reduce motivation for learning rather than inspire it. He also argues against grading to achieve aims 2 and 3, ranking and giving feedback, and he cites lots of studies along the way. I take this social “science” with a grain of salt and have no time or inclination to do my own objective evaluation of the research. Most likely, though, I find the argument persuasive because it coincides with my own assumptions about human nature. This passage from the book resonates with me:

Children do not need to be motivated [to want to learn]. From the beginning they are hungry to make sense of their world. Given an environment in which they don’t feel controlled and in which they are encouraged to think about what they are doing (rather than how well they are doing it), students of any age will generally exhibit an abundance of motivation and a healthy appetite for challenge (198-199).

Another author who was invited to the school to rail against grades and whose way of thinking also resonated with me was Benjamin Zander, conductor of the Boston Philharmonic. He argues that grades constantly remind our students that they live in a “world of measurement,” and that what matters most in life is how you measure up against other people, that “life is about staying alive and making it through—surviving in a world of scarcity and peril,” and that this “keeps the universe of possibility out of view” (17-19).

If learning can be intrinsically motivating, as Kohn argues, class discussion adds another layer of incentive because participation in stimulating conversation is one of the finer pleasures of life, at least in my opinion. Every person has a built-in desire to participate in conversation with other people. The originators of the Harkness method seem to have appreciated this. In explaining how Exeter would use Edward Harkness’s generous grant, Principal Lewis Perry predicted that with the “conference method of teaching” “boys who seem to have few intellectual interests, who are shy or who think they are dull, can be made to have interest in their work.”

Perry seems to have agreed with me about the intrinsic value of conversation and with the author who wrote an article with this title:  “Conversation: The Greatest, Most Lasting, Most Innocent, Most Useful Pleasure of Life.” 

The fundamental disagreement over grades, parallels the political divide in America today. If I subscribed to the Lord of the Flies view of human nature, and thought that people need to be coerced to do the right thing in every instance, including wanting to learn, then I would be more likely to conclude that grades are an essential motivational tool—and I would probably vote Republican. But the whole notion of Harkness and student-centered-learning reflects a more sunny view of human nature and of the inclinations and capabilities of young people. A book of essays about Harkness teaching written by Exeter teachers in the 1980s carries the title “Respecting the Pupil,” borrowed from Emerson, who wrote: “Respect the child, respect him to the end. … Be the companion of his thought, the friend of his friendship, the lover of his virtue.”

The 2003 Exeter Curriculum Review did not lead to the elimination of grades at the Academy, I’m sad to say. I don’t even recall there having been much conversation of the book at the time, much less a serious consideration of changing our approach to grades, though the book had an impact on those of us who took the time to read it. My own conclusion was that for pragmatic reasons we were not likely to be getting rid of grades as an institution any time soon--and that individual teachers could not unilaterally stop grading students--but that as much as possible we should try to de-emphasize grades, to think of them as a necessary evil that we must engage in solely for reason number 3: ranking our students for college admissions. This function is probably unavoidable because of the demands of parents, administrators, and college admissions departments.  And we may be doing students a favor by steering those who got lower grades at Exeter away from particularly competitive and demanding colleges and universities into ones where learning is more important than status. As for motivation and feedback, we should see grades as counterproductive and try to minimize the damage.

Unfortunately, trends have gone in the opposite direction. Our students have become more, not less focused on grades in the intervening years, and teachers feel more and more obligated to put more emphasis on explaining grades to anxious students and parents and to grade more facets of student performance. A good example of this is the growing emphasis on giving students grades for class discussion.

When I got to Exeter in the late 90s, most teachers were hesitant to give a specific letter grade to students for their class participation. The legendary history teacher Bruce Pruitt put it this way:

I expect students to participate and I don’t give them a grade for work in class. I move students whose work is in the middle either up or down on the basis of their class participation; otherwise I do not penalize quiet students. Often I move students who have been outstanding in class up a grade even if their average isn’t in the middle.


Bruce once said to me, echoing the Harvard Business School essay quoted above, that if a student doesn’t speak in his class, he blames himself and doesn’t feel justified in penalizing the student.  The late Exeter English teacher Peter Greer wrote: "I don't do anything formal in way of assessment" of class participation.  "I do it by a combination of memory, intuition, and feel.  I can't remember a complaint in that regard."  I was influenced also by my mentor, Mike Milligan, who wrote that he assigned no more than 20 percent of the grade to class participation, and assessed the student's discussion based on how they contributed “in a holistic sense, to the demeanor and rigor of the classroom experience. Hard to know what to do with the shy, but well-prepared student.”

Hard indeed. And yet, since Bruce and Mike left the school, things have gone in a different direction. Some teachers now base as much as 50 percent of the final grade on class discussion. Their arguments seem persuasive.  Milligan’s students were assessed mostly on how they did on papers—80 percent of their grade. Yet most of their time during the term was spent reading the daily assignments and participating in class discussions. Is it fair to give so much credit to students who happen to be good at writing, and so little to those who excel in verbal conversation? Perhaps not. But I’m just not convinced that I could come up with a fair way to assign a numerical value to what one student does at the table and then be able to justify it when challenged. And let’s face it, a grade is a number, whether you call it a B or an 85 percent. Some teachers even do use the 100-point scale for discussion grades, sometimes even making one-point distinctions among students. We all agree that discussion should not be assessed simply on the basis of who makes the most contributions to the discussions. So how then do you rate quality? Plus or minus for asking questions that reveal ignorance? When do you start subtracting points from students for citing the text too much? Or for speaking too frequently or at too-great length? And how do you manage to pay close enough attention to 12 different students while facilitating a class discussion and thinking about your next question? (See my piece on “The Improvisational Teacher” for an account of the ongoing inner dialogue of the discussion leader.) And finally, how much written feedback do you owe your students to explain how you arrived at 50 percent of their grade? From what I’ve seen, no teacher, even those who count discussion very highly, gives nearly as much feedback to this part of the student’s work as they do on the other fifty percent—writing.

I do have to concede that the heavier emphasis on grading has led to more systematic efforts among my colleagues to give more concrete and regular feedback to students and give them suggestions for improvement. But, supposing that the feedback is all good. Can’t we deliver it without connecting it to a grade so as to avoid the negative impact of grades?

As corrosive as grading may be in general, I believe it is particularly problematic when assessing class participation. I worry that rubrics will steer every student toward the same sorts of behavior and can’t capture all the subtle nuances that go into a good discussion. Like a beautiful flower garden, a discussion is best when students contribute in a rich variety of different ways. Grading may lead to too much uniformity of student behavior and instill a sense of competition. I want my students to think of themselves as engaged in a collaborative process of moving their group toward deeper understanding of the topic. Assessment of class participation is more difficult, more subjective and harder to explain. When a student challenges my grade on a paper, we can look it over together and I can sometimes be convinced that I was wrong. When it’s a discussion grade, there’s nothing concrete that we can look at together to support my judgement or to show that I’ve made a mistake. When I know I am being observed and my performance is being assessed, I am more guarded, ill-at-ease, and less willing to take risks. I assume this is true of at least some of my students, too. I think a conversation goes better when the participants do not think they are under surveillance and being judged.

I’ve bucked the grading trend as much as possible, though increasingly I feel the pressure to succumb. A few years back, my department added this comment to our “Standard Operating Procedures”: “In particular, class participation shall be factored into the final grade for every student in History at Exeter, though the weight of this factor in relation to other graded work varies from teacher to teacher. Teachers should provide the student with class participation feedback at least once during the term as well as in the comment slip at the end of the term.”

So, I must grade class participation even though I agree with Kohn that “teachers and parents who care about learning need to do everything in their power to help students forget that grades exist” (206).

So, here are some of the ways I try to fulfill my grading responsibility but to de-emphasize grades so my students might pay more attention to history than to how they are doing in my class.

1. I try not to emphasize grades in my verbal comments to the students or use grades as threats by saying things like: “pay attention, because you’re going to be graded on this.” Or: “that was a weak discussion. If you don’t do the reading it will hurt your grade.” I do say, “you’ll need to understand these concepts to write the next paper.”

2. I don’t mention the discussion grade unless they ask.

3. But when they do, I explain my approach and the reasons for it, as outlined here. I have it spelled out on my web site. I’m negotiating the tension here between my desire to de-emphasize grades and my belief in the importance of transparency in grading. A student should not get smacked with a grade at the end of the term without any kind of warning. 

4. My grading policy is more carrot than stick. My goal in writing it was to reduce anxiety among an increasingly anxious people.

 5. My feedback is always qualitative and concrete, never numerical. I think it is impossible to be precise enough in judging a person’s participation in a conversation to be able to assign a number to it. Can you really explain the difference between a score of 85 and an 86?

6. I don’t grade on a curve. As Kohn says, “under no circumstances should the number of good grades be artificially limited so that one student’s success makes another’s less likely.” Yet we do have pressure to hold the line against grade inflation. (More of Kohn’s suggestions for de-emphasizing grades, which I found to be only of limited help, are on pp. 208-209).

I realize that other teachers at other schools, where there is less classroom autonomy, may find it more difficult to de-emphasize the grading of classroom discussions. Students and parents may be more insistent on getting constant updates on “how they are doing.” School policies may require more frequency and transparency in grades—even, God help us—in online grade books that are expected to be updated daily. Or you may agree with the author (not an Exeter teacher) of a piece in the Exeter Humanities Institute Discussion Dynamics reader who said: “To cause students to take Harkness seriously, they need to see evidence of this in the weight of their participation grade. Thus, teachers should heavily weigh participation.”

To those teachers who do decide to assign 40 or 50 percent of a student’s final grade to discussion, they better provide lots of qualitative feedback along the way. If a student writes four papers for an Exeter history teacher, they can expect to get four lengthy comments at the end of each paper, along with a lot of helpful comments in the margins. We put a tremendous amount of time into writing those comments, and all of them are designed to help the student do a better job on the next paper. If half the grade is determined by class participation, should they not get an equivalent amount of feedback on that? And teachers will need to pay a lot of attention to what each student is doing every day in order to gather the material they need to write that volume of comment.

I find that my comments on papers are able to touch on many of the various aspects of the student’s work: their mastery of content; their historical thinking skills; their composition skills; and their ability to write coherent sentences. Within each of these categories there are an almost endless number of skill subsets. Take content, for example. I can comment on accuracy, chronology, missing evidence, and use of contrary evidence because I have the time to look closely at and think deeply about each and every paragraph. The discussion-grading teacher will probably need to take some notes in class to be able to reflect critically and concretely to that degree on each student’s handling of content. They should also reserve some time after each class period or the end of each school day to jot down notes on each student’s performance. That would be very time consuming and hard to fit into a busy day.

One area of assessment you will want to evaluate is preparation for class. You want to give better grades to students who prepare adequately for class than those who don’t. Some students are so good at seeming to have done the reading that you can be easily fooled (their work on papers may be a better reflection of preparation—less easy to fake). One way around this problem is to give the occasional pop quiz.

Nothing I’ve said here should be interpreted as opposed to teaching students how to participate productively in a Harkness discussion. Our students are hungry for feedback and to learn how to be good conversationalists. Teachers should not hesitate to pass along good tips, on what not to do: Don’t hog the conversation. Don’t raise your hand.  Don’t interrupt. Etc. You will be doing them a big favor if you break them of such habits, and they will appreciate it, especially if they don’t think you will also be punishing them during the learning process with a bad grade. And we can tell them what they should do: Learn (and use) the names of the people in your group. Express humility. Practice active listening. Show interest in others’ ideas. Ask people to clarify or elaborate on something they said. Support claims with convincing evidence. Cite the text being discussed. Invite others—especially quieter participants—into the discussion.

References and resources for teaching discussion skills:

Rosamund Stone Zander and Benjamin Zander, The Art of Possibility: Transforming Professional and Personal Life (Penguin, 2000).

C. Roland Christensen, David A. Garvin, and Ann Sweet, eds., Education for Judgement: The Artistry of Discussion Leadership (Harvard Business School Press, 1991).

Alfie Kohn, Punished by Rewards: The Trouble with Gold Stars, Incentive Plans, A’s, Praise, and Other Bribes (Houghton Mifflin, 1993 & 1999) 

Meg Foley, “Sentence Stems to Enter Discussion,” Teaching Around the Oval blog, https://teachingaroundtheoval.org/2017/10/10/sentence-stems-to-enter-discussion/

On my website: "Discussion Templates." (https://www.billjordanhistory.com/discussing-history/discussion-templateshttps://www.billjordanhistory.com/discussing-history/discussion-templates

I could be wrong.  Two articles that give me some pause about the above: 

Shane Trotter, Grade Inflation is Ruining Education, Quillette, April 24, 2021.

Catherine Rampell, "Why getting rid of grades would help rich students — and hurt poor ones," Washington Post, May 11, 2017.

Saturday, August 11, 2018

Thinking Together, Thinking Alone: Introverts and the Discussion-Based Classroom

EXCOMM meeting at the White House Cabinet Room during the Cuban Missile Crisis on October 29, 1962.

Is discussion-based teaching, or "Harkness" pedagogy bad for introverts? What do we do, at a school like Phillips Exeter Academy which uses that method in every class, with that student who just does not want to participate? In her 2012 TED Talk, “The Power of Introverts,” author Susan Cain bemoans how schools are increasingly being redesigned to favor extroverts.

“When I was going to school,” she said, “we sat in rows of desks … and we did most of our work pretty autonomously. But nowadays, your typical classroom has pods of desks—four or five or six or seven kids all facing each other. And kids are working in countless group assignments. Even in subjects like math and creative writing, which you think would depend on solo flights of thought, kids are now expected to act as committee members. And for the kids who prefer to go off by themselves or just to work alone, those kids are seen as outliers often or, worse, as problem cases. And the vast majority of teachers report believing that the ideal student is an extrovert as opposed to an introvert, even though introverts actually get better grades and are more knowledgeable, according to research.”

Our society so values extroverts over introverts that introverts frequently try to fit in by making “self-negating choices” and forcing themselves to act like extroverts, she says, citing personal experience.

All this emphasis on extroversion and society’s embrace of a “new groupthink, which holds that all creativity and all productivity comes from a very oddly gregarious place” is not just bad for introverts, it’s also bad for humanity in general, she argues.

According to “insights of contemporary psychology,” Cain says, “We can't even be in a group of people without instinctively mirroring, mimicking their opinions. Even about seemingly personal and visceral things, like who you're attracted to, you will start aping the beliefs of the people around you without even realizing that that's what you're doing. And groups famously follow the opinions of the most dominant or charismatic person in the room, even though there's zero correlation between being the best talker and having the best ideas.” History proves that she is right to be worried about this sort of groupthink (more on that later).

So does all of this mean that the move toward discussion-based teaching is misguided?

I don’t think so. Even Cain acknowledges that there is value in teaching kids to work in groups. She just thinks that they should also be encouraged to spend time thinking as individuals apart from those groups. A discussion-based pedagogy does not prevent that from happening. In fact it encourages it. We still require our students—both the introverts and the extroverts—to sit quietly and read. You really can’t have a Harkness discussion with students who haven’t read something. At Exeter, for every 50 minutes of class discussion, we ask students to read and study for 75 minutes on their own. So essentially we are having them do exactly what Cain recommends: “to go off by themselves, generate their own ideas freed from the distortions of group dynamics, and then come together as a team to talk them through in a well-managed environment and take it from there.” And after that, we have them go off on their own again to write, drawing on what they’ve read and thought independently about the subject, revised by whatever they got out of the group discussion.

And it is worth mentioning, too, that the whole idea for developing a “conference method” of teaching actually came from concerns about the quiet kid in the class.  Edward Harkness donated millions of dollars to Exeter in the 1930s for the purchase of conference tables and so the school could hire enough teachers to shrink class size to 12 so “the diffident boy” could be included.

But still, I take Cain’s concerns about the dangers of “groupthink” quite seriously. In his path-breaking analysis of group dynamics among policy-makers, Irving Janis identified problematic social dynamics (much like what Cain describes) that led to very bad policy decisions, notably the Kennedy administration’s ill-fated Bay of Pigs invasion in 1961.

“I use the term ‘groupthink,’” he wrote, “to refer to a mode of thinking that people engage in when they are deeply involved in a cohesive in-group, when the members’ strivings for unanimity override their motivation to realistically appraise alternative courses of action” (9).

No doubt the same sort of dynamic could take hold in a group of teenagers discussing the Bay of Pigs invasion. And that’s where the teacher’s role becomes key—to foster what Cain refers to as a “well-managed environment” in which the softer voices can be heard and dissenting voices are welcomed and not squelched (and students aren’t encouraged to ape the teacher’s views).

Fortunately for life on Earth, immediately after the Bay of Pigs fiasco, President Kennedy set up a commission of inquiry to look into what went wrong in the decision-making process and to set up a “well-managed environment” for making decisions during future crises. The commission recommended and Kennedy accepted a series of “sweeping changes in the decision-making process” in time for the Cuban Missile Crisis, in which an early consensus for aggressive confrontation was replaced with a more measured, conciliatory approach that avoided World War III. The major changes which Kennedy had put in place included 

  • encouraging skepticism about any and all ideas expressed; 
  • explicitly welcoming “frank and freewheeling” discussion; 
  • encouraging new visitors to the group, who tended to remain silent, to add their voice to the deliberations; 
  • occasionally breaking the advisors into subgroups; and 
  • giving advisors time and encouragement to meet at times without the boss (Kennedy) present. 
The new committee that was formed with these instructions became known as the Ex-Comm, and later evolved into the group that now meets in the White House “Situation Room” for deliberations involving geopolitical crises.

The new guidelines, Janus argues, made it possible for the Ex-Comm to “obtain the morale gains of high cohesiveness without the losses caused by groupthink” (140-142).

Like the president, teachers in the discussion-based classroom should encourage group cohesiveness while avoiding groupthink, and might consider adapting some of the Ex-Comm’s rules of engagement to their classroom environment. For now, though, I would like to offer a few suggestions for what teachers can do to help introverts and other reticent students to get involved in class discussions.

Making classrooms introvert-friendly

1. Don’t assume a quiet student is an introvert. Shyness and slow processing may also inhibit student contribution and getting shy or slow-thinking extroverts into the conversation may require different approaches than those that work for introverts. In the start of school questionnaire teachers might ask  students to tell them which of these categories fits them best.

2. Avoid, if at all possible, penalizing quiet students with a bad grade. See my post on grading discussion, in which I quote an expert on discussion-based teaching, who says: “If the teacher is the catalyst for discussion, isn’t the participation grade really the teacher’s self-evaluation as well?” Perhaps you have not created an environment that is welcoming to every type of learner.

3. Arrange one-on-one meetings with quiet students where you can ask them why they aren’t speaking, offer suggestions for how to prepare for class and get into the discussion, and ask them how you might help them (Ask: "Would you mind if I called on you?").

4. Use writing assignments in class. This is especially helpful for the slow processors, some of whom are also introverts. You can ask students to write a paragraph for homework and then call on the quiet students to read theirs at the start of class (see my History 420 syllabus for an example of how I’ve done this). You can begin class by asking students to write quietly for a few minutes to warm them up and gather their thoughts. You can stop a discussion in the middle of class and have them write about something, then share a bit of their writing or the ideas they came up with while writing. (My use of writing in class discussions has been influence by participation in Bard College's Institute for Writing and Thinking summer seminars for teachers, which I recommend highly).

5. Use small group break-out sessions. Reticent students are more likely to participate in a smaller group and if you really need to grade class participation you can listen in and give them credit for the work they do in those settings.

6. Appoint quiet students as scribes. When small groups report back to the full class, have the quiet student deliver the report. Or ask them to take notes on a whole-class discussion and summarize at the end.

7. When setting up groups, put introverts together. I used to try to engineer small groups so that they each had an equal number of strong and weak, talkative and quiet students, especially when those groups would be competing against each other in a debate or some kind of role-playing exercise. The problem with that, of course, is that one or more of the dominant students ends up doing all the talking. This spring in my law course I tried grouping students of similar abilities and learning styles together and the “weaker” groups did surprisingly well in the oral argument enactments they performed. The quiet students weren’t weak after all, they had just been overshadowed by less reticent students.

8. Most importantly, establish an inclusive, collaborative, low stress environment in your classroom. Use METICs to get student input on the class dynamics (that is, get meta about the discussion from time to-time so students can tell you things that they noticed but you missed about the classroom dynamic). Consider setting up a suggestion box where students can drop you notes about how the class is going. Introverts would be more likely to make use of this.

I’m hoping that readers of this blog will respond in the comments section with their own suggestions about how to help introverts, shy students, and slow-processors to feel more at home at the Harkness table.

References:

Susan Cain, “The Power of Introverts,” TED Talk, February, 2012. She also wrote a book: Quiet: The Power of Introverts in a World that Can’t Stop Talking (Broadway Books, 2013). https://www.ted.com/talks/susan_cain_the_power_of_introverts/up-next

Irving Janus, Groupthink: Psychological Studies of Policy Decisions and Fiascoes, 2d ed. (Houghton Mifflin, 1982)

See also, two pages on my course website: Helping Quieter Students and Finding Your Voice.

Saturday, August 4, 2018

The Improvisational Teacher

Improvisation

At this year’s EHI conferences, East and West, the teachers, most of whom were new to Harkness and assigned to lead discussion-based classes for the first time, put tons of effort and lots of good thought into preparing for the classes: reading carefully the documents we had given them and coming up with creative ways to launch discussions. They formulated thoughtful questions and sometimes quite elaborate preliminary exercises, often dividing the participants into pairs or small groups at the start of the 45-minute sessions or having them do something up at the chalk (or white) board. By midweek, though, I noticed that mostly they did not intervene much after these fine beginnings. In a couple of instances they didn’t intervene at all. If I could use a nautical analogy—they gave the Harkness boat a good firm push out into the lake of discussion, but then stayed on the shore and waved good-bye.

This got me thinking. What does it mean to lead a Harkness discussion? What is the teacher doing in a student-centered class beyond just sitting and listening? Where most teaching can be scripted, like those opening exercises, the Harkness class is improvisational. Preparation is useful, but it’s not enough. So I wrote the following note to the teachers at EHI-West, in La Jolla, Calif.

At every moment in a Harkness discussion you will be confronted with decision points. A series of questions will run through your head. Should I intervene or not? If I intervene, what should I say? Should I push the students toward my pre-determined plan? Should I let go of my plan and let the students follow a thread they have stumbled onto? Have they exhausted this topic? Should I transition them to a new one? What do I do about those quiet students? How can I draw them in? Can I get that Harkness warrior to stop talking without hurting his feelings? What am I missing in the group dynamic? Is anyone feeling left out? Did everyone hear the great point that kid made? Should I ask her to repeat it? Should I ask her to read more of the quote? Will that interrupt the flow of the discussion? How can I steer the conversation without interrupting the momentum that seems to be building? Did Johnny do the reading? Why isn’t his text marked up? Should I call him out on that? Were Pat’s feelings just hurt by what Alex said? Should I do something? They don’t seem to understand a key concept; should I just step in and explaining it or wait and see if they figure it out on their own? Or can I just let go of that? Will they need to know that to understand the next reading assignment? Am I talking too much? Is Bobbie talking too much? Should I steer them away from that tangent? Was that introduction of outside information relevant? Should I help them to make a connection that brings them back to the main point, or should I tell them to ignore it as irrelevant? Should I let go of the content we haven’t discussed yet? How important is it? What about the document they didn’t discuss? Should I return them to that important point that came up a while back that was insufficiently developed? Have they failed to discuss any ideas that will be relevant to the next paper they are going to be writing? Is there enough time left to get to a satisfying conclusion? When and how do I move the class toward a good end point? Or do I just let it go? They just seem to be digging into what’s most important but the bell is about to ring. Should I cut it off now or wait for the bell? How should I end the class? Should I sum up? Remind them of a few important points that came out? Give them instructions for the next day?

After reading a draft of this, my colleague, Meg Foley, wrote back to say: “I think your point, Bill, is to say that sometimes as we reflect on those questions the answer is ‘Yes, I will intervene because of this.’ We don't just sit silently on those questions.”

Right! I think that our EHI leaders had a tendency to answer “No” in too many cases, and that our discussions would have been even better if they had sometimes (though certainly not always) answered “Yes,” and steered the conversation more.

If we pick up on the boat metaphor, after you launch the vessel out onto Harkness Lake, you need to jump in, and grab the tiller so that boat does not float aimlessly in circles or run aground as the students all pull the oars in different directions. And, by the way, if you've spent a lot of time on those preliminary boat-launching exercises, which give teachers a reassuring sense of control as well as some respite from the stress of improvisational teaching, you might not have left enough time for the boat to make much progress toward an interesting destination.

Students don’t regard intervention during the class as preventing a student-centered discussion, but as an essential element to make it successful.  Meg asked a representative sampling of Exeter students to comment on what the teachers did in class and wrote about the results in her excellent “Teaching Around the Oval” blog. “By and large, the students did not mention ‘activities’ or ‘strategies’ in the way you might imagine,” she wrote. “They essentially said: teachers are effective in teaching us Harkness by letting us go at it and then nudging and questioning us, when we need it.”

In the same blog post, Meg writes that student comments about their teachers also focused on the “general manner of the teacher,” as also being more important than the sorts of activities or strategies that teachers used to a more scripted class plan are likely to favor.

In my mind, that means patience; a willingness to take a back seat; attentiveness and engagement; open-mindedness, humility, and a willingness to express uncertainty and admit to not knowing some things—perhaps even self-deprecation; sometimes saying the words “I could be wrong”; a sense of humor; inquisitiveness; a respect for and eagerness to learn about opposing views (being, as New York Times editor Bill Keller put it in complimenting a reporter, “voraciously open to contrary argument”); enthusiasm for intellectual work and for the subject matter at hand; being a unifying presence, a builder of community. Above all it means having an attitude of trust in your students and faith in their intellectual capacities—as the title of a book on teaching at the Harkness table by Exeter teachers in the 1980s put it, “Respecting the Pupil."

As Emerson said, in a passage quoted in the frontispiece of that book, “Respect the child, respect him to the end. … Be the companion of his thought, the friend of his friendship, the lover of his virtue.”