Today we’re talking about a model that influenced every discussion I ran in my classroom from my first year to my last, across grade levels, years, and countries. I’ve run hundreds of Harkness discussions - terrible ones, experimental ones, pretty ok ones, good ones, and absolutely incredible ones. Today I want to tell you how Harkness discussion changed the way I see group dynamics and why I can’t talk about class discussion without centering this model. I want you to try Harkness, or some spin off of it that fits your classroom space and size, and here’s why. Maybe you’ve heard me talk before about the new teacher conference I attended in Northern California when I was 22. At some point during that loaded weekend, someone handed me a sheaf of papers labeled “Harkness Discussions.” Inside, I found some example discussion charts, a summary of the model, and a dream.  Harkness was originally developed at Philips Exeter Academy, where a philanthropist named Edward Harkness made a gift to the school that was channeled into creating and implementing a model of discussion centering student voices. It sounds pretty simple - students sit in a circle, ideally - but in practice rarely - around a large oval wooden table, and talk to each other in class. They face each other, look at each other, acknowledge each others’ ideas, rather than all facing toward the teacher leading the way. By the time I started flipping through my packet in 2004, more than seven decades after that initial gift, teachers had been experimenting with and improving the model for a long time. I read everything I could find online, then decided to roll out a one month experiment in every class. I was waaay into experiments at that point, and my students were used to seeing my metaphorical jazz hands as I rolled out poetry slams, performance projects, transcendental showcases, and whatever else I cooked up late at night and on the weekend while I was working all. the. time.  So they were game enough when I explained what we’d be doing. I showed them a picture of a discussion chart and explained that a student observer would chart each discussion and give a compliment and a recommendation for improvement at the end of the discussion (not mentioning specific names). I explained that my role would be to help them prepare in advance for the discussion but not to moderate it during the actual conversation. I warned them about the vast potential for awkward silence, promised that they’d get through it, and also promised not to ruin everything by rescuing them. We talked about what could make a student-led discussion go well. And then we started.  During that first month of Harkness, I watched four different classes go through four very different evolutions.  F block skipped the floundering stage and went right to the “we’re awesome and we can rock this” stage. They had lots of kids who did the reading and wanted to talk, so after the initial observer comments that not everyone was talking (which is pretty much always the observer comment in every class in the first few Harkness discussions), things progressed quickly. With a little bit of help from me in chatting with observers before class, observations became more nuanced, and the class moved into the common next stages of Harkness, like helping students work on not interrupting each other, finding ways to subtly invite and support comments from students who were reluctant to speak, bringing more specific textual evidence into the conversation, making better transitions, and asking good questions. B block, on the other hand, floundered with the best of them. Maybe the trickiest transition into good Harkness that I ever saw over 25 different classes. Still, not to ruin the ending, but they got there by the end of the month. In D block I learned a lot about how to work with a slow-starting class. I integrated strategies like careful warm-ups to give students plenty to talk about, staring down at my notebook and writing “t