As everyone learned last month, if a guy jumps the fence and storms the White House, there are agents assigned to patrol the North Lawn, a sharpshooter backing them up, an attack dog intended to serve as a failsafe, another person at the door, and so forth. Instead of letting agents simply roam around, they put together protocols and procedures to minimize human judgment. Secret Service agents aren’t supposed to think … they’re supposed to patrol a space and react according to their training and their knowledge of the broader system’s design.
We do something similar in education (though without attack dogs, and — usually — without guns). We spend tons of time building systems to build procedures and protocols around important decisions. Sometimes they’re designed to prevent superintendents from using early education funds to build football stadiums. Sometimes they’re aimed at incentivizing teachers to set certain priorities or instruct a particular way.
The challenges of fixing schools and classrooms is generally tackled by means of systems. This is a serious challenge for the No Child Left Behind-era of American education reform, which generally seeks to collect more data on schools and use it to apply consequences to poor-performers. While the education reform movement is hardly monolithic, most reformers are enthusiastic systems-builders. There are good reasons for this — for instance, American education data are laughably incomplete — but the approach has serious limitations.
First, consider the purpose of a system. It’s meant to address the vagaries of humanity. That is, it makes life more predictable, routinized, stable, and unbiased. Call it original sin, call it human fallibility, call it whatever you like, but humans are often self-interested and unreliable — even when they have the best of intentions. This is a core American strategy at least as old as Madisonian institutionalism. We resist investing individuals—and their judgment—with consequential power, so we build systems to protect against the ways that they might screw up.
There are good reasons to build systems. They can give us valid, reliable information and perhaps a sort of objective fairness. They can guard against abuse and formally incorporate more data into how schools are run. They prevent public schools from becoming the province of any one individual. But there are limits to what we can achieve by building more protocols, more evaluations, more oversight, more checks and balances…more systems. After all, as the Secret Service recently found out, they’re still run, designed, and staffed by humans.
I think that Washington, D.C.’s famously controversial teacher evaluation system is solid proof of how this impulse looks in education reform today. Love her or hate her, former Washington, D.C. schools chancellor Michelle Rhee didn’t really make it easier for administrators to dismiss teachers. She made it easier for a system to capture their performance, process it, and then give them notice that they might be underperforming. And at the end of the day, she made it possible for the system to dismiss teachers repeatedly classified as “ineffective.” She still took plenty of heat for pushing that approach, but consider: Rhee’s reforms aimed to capture and evaluate teaching quality from a systems-eye view. If a teacher loses his or her job as a result, it’s largely because of the impersonal churn of procedures — not his or her administrator’s judgment.
That shouldn’t be taken as bestowing any blessing or blanket condemnation on Rhee’s work. But it’s a useful way to prompt reflection after years of these sorts of efforts to improve education. Or, to state the question outright, what’s more likely to improve teaching in the USA: comprehensive, higher-stakes accountability metrics for current teachers, or a substantive overhaul of how we recruit, train, and compensate them?
The latter path has a number of virtues, among them the prospect of loosening the various systems blanketing American schools today. That is, it’s easier to make the case for putting away pre-packaged curricula and detailed evaluation systems for highly-trained, effective professionals. The systems-driven approach to education reform arose in response to enormous educational inequity in a system where basic accountability seems vanishingly rare.
I think that meaningful data and accountability systems have their place in future education reform. And since those are still relatively rare, there’s some value in continuing to push for improvements on both. We just shouldn’t overestimate that potential. More comprehensive systems might improve American education somewhat, just like more fences and guns might make the White House a bit safer. At some point, though, the costs of adding more protocols, procedures, and oversight outweigh the benefits. Systems designed to guard against the weaknesses of our humanity still need to be inherently humane.
Conor P. Williams, PhD is a Senior Researcher in New America’s Early Education Initiative. Follow him on Twitter @conorpwilliams. Follow him on Facebook.
