Via the Facebook page of Making Thinking Visible (Project Zero, Visible Thinking) comes an interesting article from the NY Times, “Why Do Americans Stink at Math?”

It turns out a big chunk of the answer is, because American teachers stink at learning how to teach. This stinkage is illustrated by contrast to the Japanese, who ironically got jazzed about American innovations in teaching theory and practice during the ’80s, and implemented them at the same time they were going nowhere in the U.S. The article, which is adapted from Elizabeth Green’s forthcoming book *Building a Better Teacher*, argues that although the U.S. is a leader in conceptual innovation and extraordinary experimentation, we do a particularly bad job of general implementation because we fail to actually show teachers how to do the exciting new thing. This has happened over and over again. In contrast again, the Japanese made a commitment to the change and poured tremendous institutional and peer support into training up the educators. So in fact Green’s thesis is that it’s not that we stink at learning how to teach, but at teaching how to teach.

No doubt this is true, or at least it’s a perennial complaint. But there’s something a little odd about the argument. The consistent theme of each iteration of innovation is to take an experimental attitude to teaching, and to commit to an open-ended process of discovery. The Japanese teacher offered as model is Takeshi Matsuyama. “At the university-affiliated elementary school where Matsuyama taught, he turned his classroom into a kind of laboratory, concocting and trying out new teaching ideas.” The idea is to set up a discovery-oriented environment, then let students figure it out for themselves.

So, why does Green think teachers themselves need something other than this? I realize there are all sorts of strategies that ‘facilitate’ this process – I’ve developed many by doing, learned others by paying attention and reading and making connections. There’s much more for me to learn, and plenty I’ve forgotten that I shouldn’t have. I could tell all this to apprentices. But again, the point of the method is self-discovery through recursive experimentation and research and reflection. It’s really the opposite of ‘we need to show these people how to do this algorithm’, which is precisely the old model that we’re trying to get over. On this view, we don’t at all need to show teachers how to do this. We just need to set them to the task and let them sort it out.

Well in actual fact, that hasn’t worked. Instead, confusion reigns and the reform collapses back into old habits. Which, as Dave Mazella keeps saying, have the substantial merit of not working in familiar ways that define the norm, reinforced and perpetuated by what Green calls the “apprenticeship of observation.” And since it’s clearly the case that failure is endemically acceptable – normal, in fact – in the American education system, so things remain. Would teaching the teachers how to teach change that?

I’m not sure. It’s the disposition of discovery and risk that’s missing; that would seem to be built into our system, but it was in Japan too. And it would seem to be simple enough – it’s a one-page handout, a blog post – to convey the concept of moving from an “I, We, You” to a “You, Y’All, We” classroom framework. Try it, work with it. Here’s a problem: “Without the right training, most teachers do not understand math well enough to teach it the way [innovator Magdalene] Lampert does.” But Lampert’s method does not require the teacher to understand math, yet. It requires the teacher to understand the process of figuring math out, which, as the math-in-the-wild examples in the article show, is available to anyone who accepts the need to do so and puts their mind to it. Again, the idea that there’s some special training teachers need here seems off-base.

Green tells poignantly of teachers trying to do it right, but instead taking the new script and jamming the old one into it.

And how could she have known to do anything different? Her principal praised her efforts, holding them up as an example for others. Official math-reform training did not help, either. Sometimes trainers offered patently bad information — failing to clarify, for example, that even though teachers were to elicit wrong answers from students, they still needed, eventually, to get to correct ones.

How could she have known? Well, did her students figure something out or not? Did they start getting right answers or not? Why were their answers right or wrong? Really, she has to be told that eventually the point is to get to right answers? She’s looking for a recipe, rather than paying attention to what’s happening. It’s not hard to know if students are learning or not, if you pay attention and think a little.

What’s needed is curiosity and responsibility. When teachers have these, all is well, just as when students have them, all is well. The Japanese (and Finnish, and exotic flavor-of-the-month) example show that this can, to a degree, be generalized. I’m not sure what it would take to enable this in the American setting, but years of failed innovation suggest it’s not a one-variable problem.