Editor’s note: This essay is an entry in Fordham’s 2025 Wonkathon, which asked contributors to answer this question: “What needs to happen next—at the state, district, and school levels—for the science of reading revolution to fulfill its promise and ensure that far more children learn to read well?” Learn more.
In 2019, as the “science of reading” label gathered steam, I was in a position to do something different. From an odd trifecta of fate sprung a question: Is it possible to create a mobile app that would help both expert and novice teachers? To work together at understanding excellent reading instruction? To move from a body of science to a common set of habits of mind?
The truth is, beyond first principles (which were actually somewhat unique), it wasn’t clear how this would actually evolve. None of us could then have known that the amazing Donna Trinca Schultz Hejtmanek would mold Facebook to create a “Science of Reading-What I Should Have Learned in College” group. Or that it would quickly surpass 5000, then 20,000, now ~250,000 members. Or that many other SoR groups would evolve to take on specific needs of the teaching communities. Nor could we have foreseen that the gifted storyteller Emily Hanford would take her curiosity about America’s reading problem, and her milestone Hard Words: Why aren't kids being taught to read? to a new level with the release of the teacher-jolting series Sold a Story. Even less could we have thought that Emily’s stories and the SoR groups would combine with the growing power of tired parents of dyslexic students, to pass such strong reading science laws as quickly as they did.
All of which is to say that motivation to change happened very fast. Much faster than the tools, standards, and systems that could define what—exactly—that change should be.
So fast that, by only 2025, the eminent Mark Seidenberg would feel compelled to write, “Doing all this instruction is wildly inefficient.” So fast that the same short time would impel plenty of other literacy experts to be ready to shun the very label, “science of reading.”
Yet this path was, in 2019, generally predictable, if not inevitable. Three million teachers, tens of thousands of education professors, many thousands more professional development providers and district consultants…it was certain that nearly every message the science had to give would be transformed and corrupted many times over.
To give just one of many examples of the confusion that’s so widely sown: the teaching of “blends.” Not of blending, an essential skill that should be explicitly taught and practiced. But “blends” as units of speech or text: “S blends,” “R blends,” “L blends,” etc.
One vendor, boasting seemingly well-credentialed expert advisors and proudly offering “SCIENCE OF READING X GAME-BASED LEARNING,” also offers “Race Track Blends”. Teaching and practicing these “Blends” will “enable children to identify blends as familiar units within unknown words....this then will make decoding unfamiliar words easier.”
Except that the science says it won’t. As UFLI’s Professor Holly Lane puts it, “Blends are not an actual linguistic construct.” “We blend sounds together…but we don’t teach blends. “Kids already have 26 letters and 44 phonemes to learn....to make more than 300 graphemes.” Why on earth, on top of having to learn all of that, would I want for kids to have to learn around 70 more things we call “blends”?
Many more such “science-based practices” regularly show up in the wild.
Thus, Structured Literacy still has to deal with lots and lots of intellectual waste.
All living things must deal with waste. Even the tiniest of prokaryotic cells. Just last year, I lost a longtime friend to the failure of his body (kidneys, liver) to deal with waste. That waste can kill is a lesson societies too often forget. While wastewater management goes back 8,000 years, plagues and pandemics have afflicted many societies since.
The cure for SoR? If teaching were psychology and the year were around 1952, the answer might be something like the Diagnostic and Statistical Manual of Mental Disorders (DSM). Yet we well know how that would go.
Coaching is now considered a primary solution.
This spring, I attended one of the most thoroughly professional sessions I’ve seen since entering education. Two leaders from the ReadOhio Literacy Coaching project presented Ohio’s new statewide coaching program, (See ReadOhio Coaching Report: Cohort 1 2023-2024 School Year.) I left elated!
Such state-directed coaching is certainly a recipe for lifting instruction, as we saw from Mississippi’s fantastic investments and results. It is a huge part of the infrastructure we must—at least until ed schools get their act together—build.
Yet this can only take us so far. It, too, is subject to the gaps in our common understanding of what outstanding and efficient reading instruction looks like.
Not that this challenge is new to SoR. Problems with the infrastructure of teaching go back to the ‘70s and ‘80s gatekeeping of orgs like the NEA. In newer infrastructure, from the glossy publications loved by foundations since the ‘90s, to the infinite blog posts of the 2000–2025 era, the conveyance of evidence-based methods has not kept up.
There is one other profession that has dealt well with garbage-collecting at this scale. Scale not just in member numbers and domain complexity, as teaching certainly faces. But also in the challenges of professional organization. Teaching autonomy, 16,000 districts, the behind-the-closed-door nature of actually delivering instruction, lack of tort law applicable to instruction: All these provide unique challenges.
True Open Source communities of practice could help to solve them.
This proven process of developing reliable intellectual assets has now been refined for 30 years by software engineering professionals. “Collaboration, not confrontation” is its ethos.
Workflow in an open source community is simply different from the discussions that normally take place around K–12 education. Better tools help enable this. Everything has a documented, linked history. There is a stability to intellectual assets that blog posts and podcasts never enjoy. Adopted procedures have a certain chain of custody, responsibility, and community buy-in.
Too, the culture of open source truly is to contribute to where you can, and be respectful where you can’t.
When I set out to build the Science of Reading mobile app (and its ecosystem), the goal was to build such a community. (And thus to help deal with accumulated intellectual waste.) In many ways, the effort went far further than I ever expected. Yet, in the biggest way, in building and modeling open source communities to teachers, it has completely fallen far short.
Structured Literacy’s biggest challenge remains.
It is no longer enough to fix K–3 foundations alone. (Though that is far from fixed yet!)
We must also cut instructional waste, reduce the opportunity costs of unneeded practices that don’t help kids read.
It’s past time for the experiments needed to create better, stronger sources of truth than reading instruction has yet had.
In short, to know better, and do better, to carve room for all the other learning that kids rightly deserve.
