The science of reading is having a moment. Pundits are trumpeting the Southern Surge and celebrating academic outcomes in (gasp!) Mississippi and Louisiana. Outlets like the New York Times and Washington Post are rife with stories advising blue-state officials in New York and California to look south for educational inspiration. Congressional Republicans are holding hearings on what Washington should do to support the science of reading.

Here’s a rule of thumb: You know school reforms have made it big when they’re discovered by those who don’t normally have much to say about education. Well, if folks are going to hop on an education bandwagon, phonics-based early literacy may be the sturdiest vehicle around. As edu-fads go, this is a good one.

It’s also something of a reprise.

After all, it was 26 years ago this spring that the National Institute of Child Health and Human Development issued its National Reading Panel report, which made the case for the science of reading and emphasized the need for explicit instruction in phonemic awareness and systematic phonics instruction.

Those recommendations were the foundation of the Bush administration’s Reading First initiative, which sought to ensure that K–3 reading materials and instruction were science-based. Launched with high hopes, the $6 billion effort was soon undone by operational headaches. It flopped so badly and was memory-holed so efficiently that many who’ve embraced today’s “science of reading” charge aren’t even aware that they’re retracing the footsteps of their Bush-era predecessors.

What went wrong with Reading First? Pretty much everything. The machinery to evaluate instructional materials was rickety. Reading First deputized reading labs at three universities to vet materials, a structure ill-equipped to fend off vendor-supplied junk science. When officials at the U.S. Department of Education sought to deal with the result, they stumbled into allegations of misconduct and favoritism. Oh, and vendors proved adept at repackaging the same old materials and trainings as suddenly consistent with the new requirements.

The result was an ambitious push to overhaul reading instruction that ultimately delivered underwhelming results. With much of the nation gearing up to take the science of reading for another spin, it’s worth asking how states can increase the odds that the result will be more Mississippi than meh.

There are obvious lessons from Reading First, things like improving product evaluation and shoring up guidelines for federal officials. But such lessons, while useful, are also limited—especially for an effort that’s mostly playing out in the states. And they don’t really illuminate why good ideas, even those backed by high-caliber research, so often fall flat in schooling.

After all, the disappointment of Reading First wasn’t just a matter of program design or technical acumen. It reflected a more fundamental challenge.

Policy can make people do things, but it can’t make them do them well. Said another way, policy is a blunt tool that works best when compelling action is enough. That’s why policy works reasonably well if the task is issuing Social Security checks or setting noise ordinances. It’s much shakier when the action is more nuanced, like changes in instruction, curriculum, or classroom culture. The failure to appreciate this has tripped up a slew of seemingly sensible reforms, from teacher evaluation to school turnarounds.

In education, bets on policy are safest when dealing with “musts” and “must nots,” as with things like compulsory attendance, annual assessments, class size limits, and graduation requirements. These tend to be clear-cut and quantifiable. If you want to require that school choice programs get funded, or that high schools offer career apprenticeships, there’s no substitute for policy.

Policy is far less reliable when it aims for complex endeavors concerned more with how things are done than whether they are. Compulsory attendance doesn’t mean students will learn anything. Funding a choice program doesn’t mean it will be accessible or competently managed. High schools can “offer” apprenticeship programs without providing meaningful placements or supervision.

Again: Policy can’t make people do things wisely or well. And, in education, it’s usually the quality of the thing that matters most—as with teacher evaluation, school improvement, or reading instruction. Equipped with only the blunt instrument of policy, though, public officials face enormous pressure to make the world a better place.

It can be useful to offer a concrete example of how this plays out.

Picture a state legislator who visits a school with a terrific new teacher induction program. She wants to ensure that other schools offer something similar. What can she do? Well, she can require that all schools adopt the new teacher induction program. Of course, the schools she knows need it most may not take it seriously. So, she includes a provision that requires that all schools hold a new teacher orientation session and assign a mentor to each new teacher.

But now our legislator fears educators may treat the mandatory meeting as a joke and mentoring as busywork. So, she requires that orientation cover eleven specified topics, mentors meet weekly with their charges, and supervisors fill out a two-page report on each mentoring session. Now some supporters are starting to get frustrated about red tape and undue rigidity, but our legislator remains restless. She fears her measures still aren’t enough to ensure more than box-checking compliance. So, she adds more provisions . . .

You see the problem.

This is why education can seem like it’s drowning in rules—rules that are quite consciously crafted to stop stupidity and malfeasance. As they say in Silicon Valley, “That’s not a bug, that’s a feature.”

So, back to the science of reading bandwagon. What’s all this mean for the current enthusiasm for emulating Mississippi and Louisiana?

Well, there are three simple mantras I keep reciting. They may sound brutally simple—even banal. They’re all painfully obvious to any hardened ed reform veteran. On the other hand, they probably need to be shouted from the rooftops, because I’ve heard from a lot of now influential ed reform novices (including big-name pundits and Republican officials) who’ve never had cause to give them much careful consideration.

First, erecting a process to select “high-quality” instructional materials doesn’t mean you get high-quality materials. It’s not just that Mississippi or Louisiana have a process for making good choices; it’s that competent people run those processes, and the results are honored. I mean, what states say, “Hey, we’re looking for ‘low-quality’ materials!”? The problem is either that staff don’t know how to identify good materials or that, in practice, schools largely ignore their determinations.

Second, hiring instructional coaches and delivering mandated trainings doesn’t mean pedagogy will improve. It matters what those coaches do, what those trainings cover, whether teachers take them seriously, and whether there’s follow-through. Some of the more credulous accounts of the Southern Surge would have readers think that instructional coaches are a novel idea. They’re not.

Third, policies like ending social promotion, restricting the use of discredited teaching methods like “three-cueing,” and revamping required trainings are crucial for shifting classroom culture and expectations. Yet, many who seek to bring Mississippi’s lessons to blue states seem to wistfully hope they can settle for changing instructional practice while steering clear of those measures which will annoy schools of education and teachers unions. Unfortunately, changing entrenched practice requires breaking some hearts.

Regular readers know I’m skeptical of edu-fads and think you do well to bet against them 80 percent of the time. Even so, a push for sensible, evidence-based, back-to-basics reading instruction is one reform that’s full of real promise. The question is whether those who are eager to hop upon this shiny, widely celebrated bandwagon are looking for a joyride or are willing to do the work.

Frederick Hess is an executive editor of Education Next and the author of the blog “Old School with Rick Hess.”