Burden of Proof

We Americans have uneasy relationships with many of our most cherished laws and traditions. For instance, that pesky First Amendment-- can't we just limit freedom of speech to people who aren't stupid and wrong?

We also have trouble with the whole "innocent until proven guilty" thing. We all agree that you shouldn't jump to conclusions, but, man, when you just know that somebody is guilty, why should we have to bother with all this convoluted due process crap?

The framers had a pretty good idea how quickly things can go south without the presumption of innocence. Because when you presume, guilt, the whole focus, the whole purpose of the process completely changes.

Remember pressing? You might remember it from the case of Giles Corey, one of the Salem residents accused of witchcraft. Corey would not confess to witchery, and so the authorities tried to get a plea out of him by simply piling more and more slabs of rock on top of him. Instead of confessing, Corey died.

See, if we start with the assumption that a person is innocent, then the process involves figuring out the truth, whatever it might be. But if we start with the assumption that the person is guilty, then the process is about getting him to admit it, and since we already "know" that he's a guilty, guilty Bad Guy, anything we do to get a confession out of him is okay. Tricks? Sure-- we're trying to "catch" him being guilty, not find out what actually is going on. Torture? Doesn't matter-- it's just a down payment on his punishment. In a system that presumes guilt, we may never get at the truth, because we aren't even looking for it. We just head directly into the punishment phase.

It's all about burden of proof. If we assume that I'm innocent until we know otherwise, then you shoulder the burden of proving that I'm not. If we assume I'm guilty, then I have the burden to somehow prove that I'm not.

The attempt to change teacher evaluation in this country doesn't just represent a change in focus or technique. Some reformsters are trying to shift the burden of proof. "More than half the students in New York State failed the Big Test," exclaim Campbell Brown et al. "That means that more than half the teachers in New York State must suck." And so we set out to design a system in which teachers are assumed to be incompetent until they can prove otherwise. And that means a gotcha system, a pressing system, a system that is not interested in getting an accurate picture of what is going on in the classroom. "Accurate picture?" scoffs CBET. "We have an accurate picture-- crappy teachers are everywhere and they're stinking up the joint. Now, prove you're not one of them."

I believe some reformsters believe that student test results are actual useful data (they're dead wrong, but they believe it). But I also believe that some reformsters like using test data because it will give them the results that they already presume are true. They already know the "truth" (public school teachers are terrible), so a "good" evaluation system is one that "proves" what they already "know."

This shifting the burden of proof to teachers blinds the system, because we're no longer trying to find out what's actually happening in classrooms. We're just trying to catch teachers being "bad."

Worse, the same attitude trickles down through the system. American students are all terrible, right, because they're trapped in failing schools. We alllllll know that, right? So when, for instance, New York claims that almost three quarters of NY students are failures, reformsters don't leap up and say, "What the hell! Are you sure that's right?" and demand a more careful look at where those figures came from. No, it just confirms what they already "know" to be true. If somebody (say, one of the students who was just labeled a failure) wants to prove they're not a failure, the burden is on the student.

This is the essence of bad assessment, particularly for young children. Part of the idea of authentic assessment (for you young folks, that's an approach to assessment that was just gathering steam when No Child Left Behind came along and stabbed it in the heart) was that for teachers to approach assessment by asking, "What would be the best way for me to allow the student to reveal what she knows or can do?" High stakes standardized testing says to the child, "Prove to me that you're not a loser."

As I said, particularly rough on small children. Barring any kind of abusive home life, it has never occurred to a seven-year-old that she sucks, let alone that she should be prepared with an affirmative defense to prove she doesn't.

For a small child, the idea that the world sucks and she also sucks is serious news. But for children of any age, establishing that they're in an adversarial relationship with an education system that considers them broken and stupid unless they can prove otherwise is not helpful for anyone. Nor is it useful to employ a system of tests designed to "reveal" as much failure as possible. It's teacher 101-- success experiences create more success.

Tell students that they're failures over and over and over again and many will buckle under the heavy burden of proving that they aren't. And in the crazy world of reformsterism, reformsters speak as if they understand the importance of expectations, and then support a system that says plainly to students, "We expect most of you to fail."

A justice system that puts the burden of proof in the wrong place will collapse under its own badly distributed weight. It will fail to do its job, fail to find the truth, fail to support the innocent. An education system that makes the same mistake will yield similar results.