When asked to evaluate the French Revolution, Zhou En-lai supposedly said, "It is too soon to say." As we mark the tenth anniversary of No Child Left Behind, the law's defenders are making the same argument.

So, reporters must follow the convention of saying that "the jury is still out," and that NCLB "did shine a light on underperforming minority groups," before they recount the law's failure to increase scores on national standardized tests. For instance, theNational Research Council's blue ribbon panel concluded that twenty years of test-driven accountability produced minimal increases in student performance (about 0.08 of a standard deviation) while prompting many unintended negative effects.

As the true believers grasp at any straw to defend NCLB, we should take the time to savor the ways that they twist themselves into pretzels while pleading innocence. For instance, the Huffington Post's Joy Resmovits reported some jewels from these advocates for test-driven accountability. Dianne Piche, who helped write the law, said, "the lack of progress shouldn't be pegged to NCLB itself. 'The law is not a person ... it's a piece of paper.'" 

Yes, education bureaucrats who worked with pieces of paper can claim innocence due to their lack of knowledge of poor schools, but a decade ago educators predicted precisely how and why NCLB would backfire. The phenomenon that Piche and company calls "student performance" is not necessarily something real. It is just bits of electrical bursts in computer systems. That data may or may not reflect actual learning. What gets measured, however, gets done, meaning that, inevitably, teaching and learning took a back seat to gaming the numbers. 

Charles Barone, Democrats for Education Reform's director of federal policy, was another author of the law, and he has a long history of condemning teachers who do not appreciate his handiwork. A value of NCLB, Barone argued, is that, "under the law, watching movies and assigning irrelevant or no homework was no longer acceptable." But Barone has now admitted, "It's [NCLB] not serving much purpose absent of revision." 

As Barone's desire to micromanage instruction made clear, NCLB's architects had little idea of what it would take to overcome the legacies of some of our nation's most enduring evils. NCLB was an attempt to solve society's problems on the cheap. For instance, Eric Hanushek, a senior fellow in education at the Hoover Institution of Stanford University, has admitted that NCLB forced principals and teachers to do something that they had no idea how to do -- systematically overcome intense concentrations of poverty. But Hanushek, an economist, credited NCLB for, "a new attitude along the lines of 'we might not know what to do, but we've got to do something.'"

As educators predicted, some experiments worked, while others resulted in the narrowing of the curriculum, rote instruction, nonstop test prep, fabricating test score gains, and pushing out difficult-to-educate students. By the way, Hanushek also had an answer for the National Research Council's conclusion that graduation examinations (which were expanded during the NCLB era) increased dropout rates by 2%. Hanushek minimized the harm, replying that it "assumes this consequence [dropping out] does considerable harm to the affected students." He then speculated that the economic losses for those types of marginal students would not be great.

Mark Schneider, of the American Institute of Research, should know better but he has participated in the most laughable spin regarding NCLB. Schneider and others acknowledged that improvements in student outcomes on the most reliable assessment, the federal NAEP, slowed after NCLB was passed. But they redefined the 1990s, during the Clinton era prosperity, as a part of the "NCLB era," meaning that the law's authors claimed credit for improvements that occurred before they wrote the law. The 2002 law was credited with improving test scores in 1998 because they were all a part of the era of "consequential accountability." But now, Schneider has agreed that, "overall gains in performance during the NCLB era are not large."

Schneider also spilled the beans on what "reformers" believe is the key to improving schools. "We had the screws on the states," he said, but politics interfered. So, there we have it. Data-driven accountability hawks concluded that NCLB failed because it was not coercive enough.

The latest hope for true believers in test-driven reform is that Secretary of Education Arne Duncan can do by fiat what they could not do through the democratic process. As Charlie Barone commented, the failure to renew NCLB has allowed Duncan to impose his version of NCLB-type accountability. Duncan has doubled-down on the most coercive mandates. Duncan has freed 90% of schools from NCLB, while he demands that the law's flawed metrics be used to fire teachers. Duncan has tried to micromanage the lowest performing schools, creating an NCLB on steroids for our most challenging schools while liberating the schools where most voters send their children.

And that is the legacy of NCLB. It has turned reform into something that social engineers do to educators and students. Its educational politics of destruction are being perpetuated by Arne Duncan's NCLB II. It will be interesting to see who Duncan blames after his bubble-in path to "reform" meets the same fate.