Implementing Harkness

IMPLEMENTING HARKNESS DISCUSSION IN AN AP LANGUAGE COURSE

A personal narrative documenting one teacher's experience with teaching students new to the method

Jodi Rice, September 2007

Day One - An introduction to a new discussion method

Day Two - How you read and write is just as important as how you speak and listen

Day Three - Preparing a more formal demonstration discussion

Brief interlude - Meet my classroom

Day Four - Introducing discussion tracking

With a fresh new crop of AP Lang students this year, and having had the practice last year of implementing a Harkness model in my Lang class, I hope this year to document some of the things that I do in order to train students to use this model to become more independent leaders of their own discussion.

I've mentioned before how crucial this training is to having students be successful in discussion. Too often we throw a discussion out to our students to pick up, and that often means that the more confident students take over, the insecure but outgoing students waste time, the shy students retreat, and the disengaged students further disengage. A good discussion has structure, and students should be conscious of what makes that structure work, whether you are using a full Harkness model, a Socratic seminar, or another model of discussion. Taking the time to instruct students and debrief with them may seem like taking time away from other aspects of your course, but it will pay itself back in more productive discussions later on.

DAY ONE - An introduction to a new discussion method

First, I let the students know during the course introduction that they will be more responsible for good discussion during the course, and that discussion has a structure that we will be learning throughout the first part of the year.

In my first full class with them, we start by reading an article. It doesn't really matter what the article is about, although having a topic relevant to the art of communication is helpful. I use an article called "Not from where you're standing". The article is about how often we don't pay attention to what actually is being said, but understand the shorthand anyway, which greases the wheels of communication, but is not in fact accurate communication (i.e. "Do you have the time?" is not the same as "Please tell me what time it is.").

The reading itself is done in a method I learned from Larry Scanlon, called by him "ghost reading". I explain this method to the students before I start: I begin reading the passage; at a certain, fairly arbitrary point, I simply stop reading. Anyone who wants to pick up then does, reads until she wants to stop, then stops. The next person picks up from there, and so on until the article is finished. There are no other rules, other than if two people pick up at the same time, they must figure out who will continue. There is no rule like "everyone must read once before someone can read twice".

After the reading, we debrief two things: the style of reading, and the content. In the first debrief, I ask students what it felt like to use that method of reading. We make observations about what happened -- did anyone not read, did anyone read more than once, were there any patterns, long pauses. Did the pauses create anxiety? Did this method create more or less anxiety than other methods (teacher calling on people, etc.)? How alert did the students have to be? What role did body language play? What happened when two people tried to speak at the same time? And so on.

In the second debrief, I ask students to pair up and spend 5 minutes talking about the article using two focus questions: What does this article reveal about the nature of communication? What do you consider to be important concerns about the ways in which we communicate? After 5 minutes are up, I ask them to share with the table -- but I AM NOT GOING TO CALL ON THE SPEAKERS. They must figure out for themselves who will speak, and use the same kind of strategies they used during the reading to know when to do so.

We discuss the article this way for about 5 more minutes, and then we debrief again -- once again about the effect of this particular kind of discussion, without my intervention. During the discussion, I have been taking notes about points I want to raise in the debrief, and NOT looking at the speakers. I ask them things like whether they felt the need to look to me for confirmation of what they were saying, rather than make that connection with the other members of the discussion. We talk some more about body language, about language cues that make connections between points, or that invite others in, or that confirm others' ideas. We evaluate whether any questions were asked. We consider whether agreement is the only way to make connections between points.

I then ask them to take notes about the most important points about discussion strategies as I summarize and list them. Then I give them 5 minutes at the end of class to choose 2 and write about them for their own reference (I will ask them to go back to these notes later on in order to reflect on their progress).

Back to top

DAY TWO - How you read and write is just as important as how you speak and listen

Building discussion skills isn't just about building speaking skills. It's also about giving students the tools to bring discussion matter to the table. To this end, I spend a good deal of time on annotation and focused analysis skills with them.

I'm lucky in that the teachers of the Grade 10 classes that feed into my Lang class spend time throughout that year teaching annotation. The last book they study in that year is Pride and Prejudice, and the teachers model and directly instruct a ton of annotation strategies for comprehension, vocabulary, character indexing, and so on. I capitalize on their knowledge, but I let them know that annotation is one of the skills I expect as part of their daily preparation for class. "In this class," I tell them, "'read' is synonymous with 'read and annotate'."

Their homework for today was to read the website "How to Mark a Book" and to read (and annotate) E. B. White's essay "Once More to the Lake". There's really no need to use this particular essay; I use it because it's nostalgic for the author's childhood summers, and the students are just returning to school, already nostalgic for their own. Also, White is a consummate stylist, his personal narrative is layered, has familiar memoir elements (they also studied memoirs last year), and feeds nicely into our first major essay, which is also a personal essay.

I start the class by asking them about their own strategies for annotation. For this discussion, I go back to the familiar format of calling on them when they raise their hands -- they've had enough for the time being of trying out the new method, and we'll come back to it later. They share their strategies -- all of which are valuable, and variable, depending on the nature of the student. I tell them that I will never dictate to them how to annotate, because they must find something that will work for them. Instead, I will sometimes tell them what to annotate, and, in general, expect them to annotate as much as possible. (I do warn those who like colour-coding that on the AP exam, no coloured pencils, markers, or highlighters are allowed, so they may want to find alternative strategies.)

Then we talk about the website. What surprised them about his discussion of annotation? What seemed to them to be a good idea? What did they think of the image of his annotated page?

Next: pick up your pens and pencils, your highlighters and post-it notes... we're going to walk through how I would annotate the first paragraph of White's essay! I read aloud, and ask them to shadow my annotation. (Note: some people do this kind of thing on an overhead, but I find that when you do that, students pay less attention to what you're saying and spend more time trying to copy exactly how you're annotating, down to the colours you use. So instead I just ask them to follow along on their page, listening carefully and using their own annotation strategies as they see fit.) I vocalize each and every annotation as I read. I go over the paragraph twice. I talk about circling, underlining, connecting, what I like, what catches my eye, why, what I'm writing in the margins, my process of finding the words I want to describe what I see...

By the end of just that first paragraph, they have so many annotations on the page you can barely see the original text! I tell them to take a breath, sit back, and look that over. Compare it to the image of the annotated page on the website. Now, I have practice in annotating and commenting on texts, so it's not really surprising that I can mark a text up to that degree. But I don't expect them to start out that way. However, if each person were to bring one different point to talk about, and we got to discussing and taking notes of everything that's said, it's entirely plausible that the collective annotation would be dense indeed! The students agree that this is much more reassuring than the expectation of having a densely-annotated page on their own right off the bat.

I give them five minutes to annotate the second paragraph individually. Then, in pairs, they compare notes, adding to their annotations. Already their pages are looking nicely marked up. We then spend some time discussing how to turn those annotations into nicely-composed sentences (using a model I construct, letting them emulate in their pairs based on their annotations, then workshopping a few examples up on the board/screen).

The process of annotating the text is a crucial one when teaching students the expectations for discussion. Students must understand that what they see in the text will form the foundation of the discussion itself, but that unless they bring those observations to the table, the discussion will falter, or revert to depending on a teacher-based interpretation or Socratic questioning session. Only this level of engagement with the text will allow them to come up with points for discussion, questions for clarification or deeper investigation, and a familiarity with the text when called upon to support their assertions. Practice with transforming their observations into actual points is the next level, which we will continue to work on. Following that is the ability to stand up to the teachers' challenges to their assertions, using the text, and then the ability to challenge one another's assertions on their own -- constructively, of course.

Back to top

DAY THREE - Preparing a more formal demonstration discussion

Until now, much of the introduction to this method of discussion has been informal, and fairly tentative. As we move forward with some of the materials the students have prepared over the summer, though, I will be expecting them to take control of a full-fledged discussion, which I will chart and they will observe and evaluate for themselves.

Part of the summer assignment involved selecting two "literary" memoirs from a list of four: On Writing by Stephen King; Reading Lolita In Tehran by Azar Nafisi; Down and Out in Paris and London by George Orwell; and The Autobiography of Malcolm X by Malcolm X and Alex Haley. All I required of them while they read was to keep an eye out for places in the text where the author or other people encounter increasing awareness or self-awareness, and to mark them for later reference. Now that they're back in school, I find out who has read what, put them into groups of approximately 4 or 5 around each book, and give them more specific instructions: they are to focus on their assigned book, and find and bring to class two or three passages of approx 2-4 pages long, in which the author writes about his or her increasing awareness of voice, audience, him or herself as a creator or audience with a purpose, or the importance of language, reading, writing and/or communication.

Once in class, they work in their small groups: they share their passage choices and narrow them down to one passage for the group. In cases where I have two groups focused on any given book, the different groups need to ensure that they are all choosing different passages. Once the passages are selected, each group provides me with a photocopy of their passage, making sure it is clean of any annotations. Each student in the class will get a package that contains copies of ALL the group's passages. Each student is to read and annotate the entirety of that package.

The packages form the basis for formal discussions that I set up over the course of about two upcoming classes. I'll lay out the framework here, though, so that you can see where I'm headed with it.

Each discussion lasts approximately 15-20 minutes, and is followed by a class debrief. The discussion groups themselves are composed of the core "discussion leader" group, who has chosen the passage under discussion for that session, plus 4 or 5 additional participants, chosen from amongst the rest of the class. I don't reveal who those participants are until the day of the discussion, so each student must prepare as though she is to be a part of the discussion.

The rest of the students -- those who are not directly participating -- have roles assigned to them when they arrive in class, also. Some track the course of the entire discussion, using a variety of Harkness tracking sheets I acquired from the Exeter Humanities Institute (I explain to them how each is used, and they choose which they want to use, making sure that different ones are used by each student). Others track individual students using a fishbowl tracking sheet, also something I adapted from an EHI document. The fishbowl tracking sheets allow observing students to provide individualized feedback to the discussion members in a private way, while the other sheets allow a more global type of feedback. I also track the discussion as a whole, and run the debrief at the end of each discussion. Some of the tracking sheets may be found here and here.

The discussion leaders are responsible for making sure the discussion runs smoothly, but they are not to dominate the discussion. On the contrary, they must make sure that everyone gets a chance to contribute, and that the discussion remains organic, yet relevant and focused. I tell them that to an outsider, it should appear as though there is no difference between the discussion leaders and the other participants. Because everyone has prepared the same passage, and the focus is on that passage, it should make no difference if someone has read the entire book while someone else hasn't. Everyone can get into the discussion to the same degree.

Note: Last year I had one student -- a very good one -- who, early on, forgot to prepare for the discussion she was in charge of. I required her to participate regardless. At the end of the discussion, during the debrief, I asked her how it felt to be unprepared and still required to participate. She said she was able to pick up on some things and thus contribute a little bit, but she felt really regretful that she hadn't been prepared, because everyone else was so engaged and she felt left out. Seems a no-brainer to us teachers, but for a student to articulate this understanding is significant. The more students come to realize this important connection between meaningful preparation and active learning and come face-to-face with it through experience and the instruction/debriefing process central to learning good discussion technique, the more they internalize it and drive themselves to do it IN ORDER to have fun, stimulating classtime -- a carrot, rather than a stick.

I will write more about these formal discussions (I call them Round Tables) fit into the overall process in a later entry. My actual Round Tables are scheduled for the first week of October (right now it is September 15). September for me tends to get broken up by various absences (mine, for the Jewish High Holidays; the students', because of their roles as leaders in several beginning-of-year outdoor ed programs), making it hard to develop groupwork consistently until everyone is back in class for good.

Back to top

My classroom: The result of a parent-association curriculum-materials grant:

The clip-on power bars are for the laptops that are standard at my school. I don't have a blackboard; instead, I have a SmartBoard and an overhead-mounted projector, and I use my laptop from my seat to project material on the SmartBoard instead of writing on a blackboard. If I want students to interact with that material, I can have them use the interactive functionality of the SmartBoard. Notice that technical difficulties one day sent me back to good old chart paper and markers, though! This room was pretty much gutted to make room for this furniture, so it's a bit bare -- my plan for making it a bit more inviting includes plants on the windowsills, though I have a notoriously brown thumb!

The room is very long and narrow, with lots of space at the end where I'm standing to take this photo. If I wanted to expand the table for more people, I could add the two tables meant for the centre to the length of the oval. However, when I tried this configuration, it was clear to me that it made the table far too long for a proper discussion. The students sitting along the long sides would not be able to see others on the same side as them, and definitely would not be able to hear one another properly.

I usually sit in the middle of the long side under the windows, which gives me the best opportunity to see and interact with the largest number of students.

Today we experimented with charting a discussion; the charts are posted on the bulletin board down the left-hand side of the SmartBoard.

Back to top

DAY FOUR - Introducing discussion tracking

Tracking a Harkness discussion serves several purposes. For one, it gives the students a strong visual representation of many important principles, like equal participation, types of contributions, patterns in speaking order, length of "stage time", etc. It can also be a useful tool for a teacher trying to gauge how well students are adapting to the process, and whether there are gaps or students who are dominating the discussion or falling through the cracks of it, so that such issues can be addressed either with the class as a whole or individually with those students.

The classic Harkness diagram is the one you can find in this article, (with pen marks in blue). The oval represents the table, with the participants seated around it, and lines extend from each participant to others around the table in an even distribution. Obviously, a real discussion won't look like this, but it's the ideal. There are other ways of charting the discussions -- you can develop whatever works for you. 

First, though, I formally outlined my expectations for class discussions. I have to say, this was definitely one of the most boring ten minutes I've had so far in this class. Too much of me talking! But it's a quick way to summarize and make explicit some of the things I have been talking about and modeling over the last few classes.

Once we have those clear, I show them some sample charts, representing past classes. The oval chart I show them is probably indicative of a "B"-graded discussion. I don't tell them that, though. Instead, I ask them for their observations, and they point out things like the lack of participation of one member, the fact that much of the discussion is clustered on one half of the oval, that most, but not all students made page references to support their points. Explaining that a grade for a discussion would be based not on individual achievement, but on collective achievement, I asked them whether they thought this discussion was an "A". Universally, they decided that it wasn't. "It's good, but it's not ideal," explained one student. I asked them what they'd rate a "good", then -- a "B", apparently. "C" got described as "could be good if...", and we had to find a more reasonable descriptor for a "D" than "miserable failure." We settled on "definitely lacking something."

After talking about the oval, we talked about a couple of the other charts -- what they can and cannot show, observations and criticisms and assumptions about the discussions they showed. Then I told them that we would be having a discussion... and that four intrepid volunteers would be charting it. Two eagerly volunteered, while two more followed suit after thinking about it for a moment: would they get more out of participating, or charting? Did the charts look too daunting? There was a bit of laughter around the table as the students commented on how complicated it all looked. But I reassured them that whatever method they found worked for charting, they should use. It would quickly become clear to them just what they could manage.

The class had been asked to read a four-page scene from Saint Joan, so they had some material prepared. I participated in the discussion, which I don't usually do, but I wanted to model some strategies, including taking notes and requesting page references. I probably only spoke about three times, though. The rest of the discussion was handled by the students, who, once they got going, didn't want to stop, even when I tried a couple of strategies for bringing the discussion to a close. For example, after a small lull, I asked if anyone would like to summarize the main points of the discussion. That summary reminded someone of a point they wanted to bring up, and the discussion picked up again. We didn't have a ton of time for debrief.

But debrief is one of the crucial steps, so I ended up forcing a close to the discussion to give the four observers time to give their observations. Some brief comments of note:

And that last point is part of the reason why the debrief is so important. We will have more of these "training" discussions over the next couple of weeks, and each will give the students the opportunity to internalize strategies that move them from where they are now -- hovering between a "C" and a "B" -- to where they want to be -- the ideal "A" discussion.

Back to top

* * *

A video in which a teacher (not me) explains Harkness learning to her class.

Excerpted from "Harkness table" at Wikipedia:

The Harkness table refers to a style of teaching wherein students sit at a large, circular table with their teachers, in use at many American boarding schools and colleges. It encourages classes to be held in a discoursive manner. The name for the method comes from the oil magnate and philanthropist Edward Harkness, who presented the schools with a monetary gift in 1930. He described its use as follows:

What I have in mind is [a classroom] where [students] could sit around a table with a teacher who would talk with them and instruct them by a sort of tutorial or conference method, where [each student] would feel encouraged to speak up. This would be a real revolution in methods.

Harkness tables are used most notably by Phillips Exeter Academy, which highlights the tables' use in its curriculum on its website, as well as other American independent schools, [. . .]

Harkness learning can vary [. . .] There are general principles and goals, however, that go along with this method. The main goal is to encourage students to come up with ideas of their own and learn good reasoning and discussion skills. Depending on his or her style, the teacher may interact very little, interjecting only to guide the discussion.

Back to top

I had the opportunity to discuss the design of the table with the VP and requested a modular variation of the classic Harkness oval, so that it could be disassembled for various purposes (like tests) where I wanted the flexibility to change seating. The centre was meant to be filled in with two more long tables; however, they were incorrectly sized and will need to be shaved down to fit the gap. In the meantime, this works well.