I’ve been getting lots of questions lately about how I wound up doing what I do. There are a range of answers, depending partly on the exact focus of the question—Why academia? Why China? Why world/comparative history?—and plenty of people to cite as influences. One of the most important of these is somebody about whom I actually know very little—and so have conveniently mythologized, even as I’ve tried to think through what his example might tell me about teaching history.

About a month into the seventh grade at J.H.S. 218 (in Flushing, New York), our US history/social studies teacher had a health crisis, and had to be replaced. I knew the new teacher only as “Mr. Epstein,” with first initial “R.” But he may have done more to make me into a historian than anyone else.

On one of his first days, he explained how the class would work. The textbook, the dates of tests and other assignments, and so on would remain unchanged. What would change was that doing very well on all these scheduled assignments would earn you only 85 points out of a possible 100. The last 15 points would come from assignments we would choose ourselves—sort of.

The drill was as follows: every Friday, Mr. Epstein would come in and make claims about history that he expected we would find outrageous, and that would often contradict our textbook (when they weren’t beyond anything our textbook even considered). Some would be things he actually believed; some would be things he also thought were outrageous; some would be somewhere in between. It was our job to choose one or more of these statements, go to the library, and prove them wrong. There were no further guidelines, except that we had to cite the evidence for our “refutation.” If we thought that one sentence would suffice we could hand in one sentence; if we thought we needed many pages, we should either write many pages or wait until something easier came along.

This, as at least some of us discovered, could be fun: we were actually rewarded for proving that we knew better than the grown-up assigned to teach us. We were 12 or 13, and it was 1971 in New York City: distrust of authority was definitely in the air, and though we were all too young to be real “60s” kids, plenty of us had older siblings and/or mass media models that made us wish we were. Moreover, it was challenging: the more we wrote, the more comments we got back. Consequently, it was also somewhat addictive: I began writing for the sake of making a good argument, rather than for the promised 15 points.

And, as I realized years later, it was also an “introduction to historiography”: that bit about how the Constitutional Convention was really about creditors versus debtors, for instance, was about as much of the Beard thesis as seventh graders could be expected to absorb. But most of all it was brilliant teaching, harnessing our tween rebelliousness and smart-aleck tendencies to make us think much harder about history than any test based on the standard curriculum could have done. Needless to say, we also learned a lot more than we would have otherwise about research, writing, and other skills that would eventually became part of my everyday life. “Eventually” should be stressed here; over the course of the next couple years, I certainly had assignments that involved library research, but nothing that made me think about what was worth researching and why.

I made a brief, unsuccessful, attempt to find Mr. Epstein a number of years ago; in addition to simply feeling grateful, I’ve thought about him in the context of various professional issues. In part this stems from a sense that what he did was not only clever and generous (it presumably meant lots of extra grading time), but a bit risky, even then. Today, I suspect, it would simply be impossible in many public school systems. It addressed no particular content standard. Moreover, all we were told about what to aim for was that we should be convincing, which would presumably make it hard for an assessment to pinpoint exactly what skills we were supposed to develop, and to what level. Consequently, I have cited his example many times in explaining why I doubt that externally imposed assessment schemes will improve teaching. So after all these years, his formula is still working—it still feels slightly rebellious to be enthusiastic about his class.

It is not the case, of course, that this would always be a brilliant method of teaching, or even that I know how it worked with all the students. I have certainly had classes—well beyond the seventh-grade level—where this assignment would never have worked; and I am mindful of the educational studies suggesting that, in classes with a highly nondirective structure, the better students don’t so much solve problems for themselves as find ways to get the direction they need, while others flounder. In short, the moral of this story is probably not the simple Deweyan one that I would have drawn (without knowing the name for it) at the time. Making this teacher a heroic individualist proving the bureaucrats wrong was my projection, perhaps without much basis in reality.

But it does point to something that historians know well in other contexts, and that is often overlooked in proposals for assessing teaching: that it is difficult, but fundamentally important, to notice slowly developing effects of an event that may be almost undetectable in the short run. It is relatively easy to measure what a student knows at the end of a term, or how happy they were with the course, but what really matters is how useful it will look from the perspective of a few years hence. And while this is to some extent true in all fields, it is especially the case in history, because course content is not cumulative in the same very direct way that some kinds of study are. (Mastery of Algebra I or Spanish I is pretty much indispensable to Algebra II or Spanish II. Having mastered the content of a beginning survey of US history will no doubt help you in an upper-level class on the Progressive Era but isn’t truly indispensable, and it has only very general benefits if your next history course is on women in Song China.) As a result, students have even more to lose in history than in most other classes if heavily prescriptive content-focused standards make it tempting to teach to the test.

This is just one of many reasons to emphasize—as the AHA’s Tuning Project does—that descriptions of desired learning outcomes need to emerge from within the discipline, not be imposed from above. It also speaks to related, but separate, questions about measurement, intuition, and judgment in the design and evaluation of teaching. If tested at the end of the year, I’m not sure that my classmates or I would have shown any measureable effect from Mr. Epstein’s assignments; and maybe we would even have missed a couple of questions that could have been covered if time had not been spent on them. But I am quite certain they made a difference. And if one thing we historians learn to do is trace influences that take time to appear—becoming confident enough to stake our professional reputations on some such claims being stronger than others—then we should be bringing that expertise to bear, both in our intra-disciplinary deliberations and in discussions with administrators who are compelled (whether they like it or not) to justify their decisions with “metrics.”

So let me conjure my projection of Mr. Epstein one more time. I imagine him saying “By 2013, historians in the United States faced unprecedented pressure to articulate what they could contribute to general education, and to show how they did it. Insisting that they couldn’t do it the same way that natural scientists do, they refused to do it at all—and so failed to defend their place in the curriculum.” Once again, I want to prove him wrong—and thereby vindicate him.

Kenneth Pomeranz is president of the AHA.