Education’s Great Stagnation

August 2026

Part 1: Education is Uniquely Stuck

Imagine a time traveler from 1916 trying to make sense of modern life.

"How does this glowing rectangle have mini people that talk? Where is that carriage's horse? Where’s all the smallpox? Why are women and non-whites voting? What do you mean 'open heart surgery', are you mad? What are 'nuclear' bombs? Where is all this clean water coming from? Why is your outhouse inside the building? What's 'space travel'? How can this machine do your farming work? What’s an assembly line? HOW IS THAT POT COOKING FOOD INSTANTLY?"

However, that same time traveler could walk into a school and say, "Ah, yes, the students have one teacher. The students rotate among classes organized by discipline and grade bands. The school grants credits. Sufficient credits will earn a degree. This is just the way I remember school!"

And you would have largely the same reaction if you were to walk into your kindergarten classroom.

With the exception of literacy, I doubt you can name any breakthrough in pedagogy, ever, and I'd bet you take the absence of such breakthroughs for granted.

This is astonishing given that

  1. Anybody can easily name 20 ideas to improve the education they had and
  2. Education seems like the most powerful lever to change the world.

The rich and powerful have tried to reform education. Bill Gates spent billions of dollars trying to transform education in the US. He assessed his own efforts as “no dramatic change[1]." Mark Zuckerberg and Priscilla Chan poured truckloads of money into a school, only to give up completely[2] a decade later. Successful startup founder and Googler Max Ventilla launched Alt School, promised a Silicon Valley-style shakeup backed by $176 million, then gave up completely[3] a few years later.

Of course, they’re not the only ones to try. After countless reforms[4], elections, Class Clowns[5], presidents and Prime Ministers, premiers and governors, panels, education ministers and education secretaries, commissions, budgets, studies, academic articles, books, school boards and superintendents, new buildings, charter schools, private schools, independent schools, vouchers, computers, networks, videos, online classes, correspondence classes, programs, philosophies, policies, curricula, standards, tests, textbooks, textbook versions, websites, Web 2.0, MOOCs, tinkering[6], AI, and fads and trends of every type... the reality is that the vast majority of educational systems work the same way they did in the 1950s.

Quality controls have greatly improved manufacturing. Computers are vastly better than they were 50 years ago. Television and movies are better than ever. The cost of air travel and even space travel have decreased dramatically. Medicine has improved by several revolutions. Nearly every aspect of society is radically better today[7] than it was in the past.

But your elementary school, high school, and post-secondary school are not.

I bet military trainers would recognize soldier pedagogy from 5000 years ago.

The remainder of this piece discusses some of the causes of this stagnation. The next essay will hypothesize as to how Bravo Math will finally overcome them.

Part 2: Eight Truths & Lies

Have you ever played “Two Truths and a Lie”?

Same idea here, but with 8 statements to evaluate. How true are the statements in terms of improving education?

  1. “Teachers should organize their lessons around learning styles.”
  2. “Continuous feedback supercharges learning.”
  3. “Education is all about building on each student’s innate strengths.”
  4. “We don’t need to remember facts if we can look them up.”
  5. "Technology will revolutionize education."
  6. “Learners should master one skill or one concept at a time.”
  7. “Busting teacher unions will unleash educational progress.”
  8. “The key to learning is doing.”

Take a minute to write down your answers and a brief explanation.

Here’s my take.

Those statements all vary from false to misleading. All of them.

Let's take another look.

  1. "Teachers should organize their lessons around students' learning styles." There is literally no evidence[8] for the existence of learning styles. We misattribute to learning styles[9] what is better explained by prior knowledge or basic cognition. Yet overwhelming majorities[10] of people believe in learning styles, including 89% of teachers[11]. Psychologists, though, don't debate[12] this. They make April Fool’s jokes[13] out of it. Whoops!
  2. "Continuous feedback supercharges learning." If you provide feedback continuously or even at a very high frequency, then the best possible outcome is nobody knowing what the learner is doing or thinking independently[14] and when they relied on real-time feedback. Continuous feedback is a path to delusion, not a supercharger of learning.
  3. “Education is all about building on each student’s innate strengths.” Most evidence shows that most people can reach high levels[15] in most disciplines with the right training, another damning indictment of our educational system.
  4. “We don’t need to remember facts if we can look them up.” What we can learn next depends almost entirely on prior knowledge[16], i.e. what we remember. Data like phone numbers no longer need to be memorized because further learning does not rely on them. But if a chief engineer had to look up grade 1 math while designing a bridge, would you drive over it? Would the engineer even know what to learn next? If they knew, could they learn it in time to participate in designing the bridge later today? Of course not. Expert performance requires vast knowledge. Looking it up is woefully insufficient.
  5. "Technology will revolutionize education." From radios, to Thomas Edison's motion picture, to computers, to the internet and MOOCs, and now to AI, every generation of educational technologists has promised and failed to transform education[17]. We can see why by Veritasium’s fitness analogy[18]: “The world is full of heavy objects. And yet most people are not ripped.” The educational bottleneck is not transmitting information to students. Mental effort is. Ed tech evangelists never[19] make a compelling case that they will improve such mental effort in more than 5%[20] of students.
  6. "Learners should master one skill or one concept at a time." A 9-year-old student nailing 10 out of 10 exercises on a “Division Word Problems” quiz will convince most that they know division word problems, but sheds surprisingly little light on how well the student would do on an alternative quiz whose instructions were “Here are 20 word problems about a blueberry farm. Only solve problems requiring division.” Learners need to mix it up[21]. This mixup that cognitive scientists call interleaving, might be the most powerful pedagogical technique discovered in decades as it both reveals and develops a student’s discriminative contrast[22]. Yet it’s absent in virtually all technical textbooks and all technical courses.
  7. "Busting teacher unions will unleash educational progress." The 5 lowest performing states in the USA all have non-unionized teachers[23]. In fairness, there are some bad teachers protected by unions, but at a systemic level, this data disproves the notion that unions are the barrier.
  8. “The key to learning is doing." This even confused a physics Nobel Prize winner, Carl Wieman. He designed a hands-on lesson with a tethered ball so students could engage with the laws of physics in a sensory manner. They could analyze the ball’s motion by connecting the circular motion, velocity and acceleration, gravity, etc. The lesson was probably a great buzz of activity, one which most observers would approvingly call it collaborative and active learning. Except it caused literally no learning[24]. How? Because student cognition was: "Step 1 put the pole here, Step 2 press that button on the camera, Step 3 lift the ball this high, Step 4 record that measurement in cell D13 of Excel, wait, are we on track to finish on time?..." Such cognition was devoid of physics. And even if the students were trying to think in terms of physics, would they do so productively? Or would they entrench misconceptions? How much would they retain? Would they be able to transfer their learning? How would that compare to other forms of teaching? None of this is knowable from what students do *physically* because the proximate cause of learning is what students do *mentally*.

These mainstream beliefs underlie many of the failed attempts to improve education.

But as harmful as they are, beliefs alone do not cause millennia of stagnation.

There are institutional culprits, too.

Part 3: Sabotage and Gridlock

Have you ever heard a prestigious university announce:

“Over the decades, our teaching methods have improved by 1%/year so we can now educate any high school graduate. We no longer need selective admissions!”

Have you heard any private schools, any elementary school, any corporate training program, or any organization of any type, ever, even yours, say such a thing?

Did you love your most recent professional development?

I doubt it.

Why can’t the field of education sustain 1%/year improvements?

We can begin with teacher training, which excludes the most proven techniques for learning[25]. (It often includes nonsense about reading instruction[26].) So, they assume, like the rest of us, that they should trust their intuition when it comes to learning even though cognitive scientists adamantly say not to[27]. They study a lot of peer reviewed but non-useful[28] education research. They rarely learn[29] to manage unruly or immature learners.

First year teachers are largely untrained in terms of thinking, learning, and behaviour! That alone clearly sets up teachers and students to fail.

But it gets worse.

Once teachers are in the classroom, they work alone. Teachers learning by observing and collaborating with other teachers is rare, unpaid, logistically difficult, and a violation of norms. This is why, for decades, the science teacher everybody dreads can work in room 307 while the winner of the Prime Minister’s Science Teaching Award works in room 308\.

Teaching can be tougher than rocket science[30], more demanding than first-year lawyering[31], and so relentless that teachers can't find time to go to the bathroom[32]. Teachers in much of the world are paid so little that they work a second or third job and are busy selling their blood plasma[33].

The vast majority of teachers want to improve. But this, too, is ridiculously difficult. Craft knowledge requires useful feedback and teachers largely work without it. Are you in touch with your kindergarten teacher? Do you know exactly what long-term effects their spelling pedagogy had on you? Did such information get fed back into the system to improve even a single kindergarten teacher’s craft?

Failing to collect and use analogous data in medicine is unimaginable. In education, it is the norm.

And how many teachers, especially new ones, have time to collect and act upon this kind of feedback? They don’t have time to pee!

And if you think teachers have enough time outside of class to work on their craft, you have never put in the hours necessary to get teenagers excited about logarithms. It’s a lot of hours. Unpaid hours. And you’ll probably fail.

Furthermore, there are likely hardwired biases[34] preventing us from learning from the more immediate feedback, such as how students behave during class, perform on quizzes and tests, etc. It is so easy to see learning and mastery where there is none.

Craft knowledge is usually not taught during teacher training and it is difficult to acquire on the job. Could teachers improve like doctors do, by seeking peer reviewed articles based on randomized controlled trials (RCTs) that carefully measure cause and effect? The Educational Endowment Fund (EEF) of England sought to do this on teachers’ behalf to help them assign homework properly. Anyone outside of education would assume that academia provided insight as to how the following obvious variables interact with homework:

Any guesses as to what the EEF discovered?

Well, EEF came back with - wait for it - absolutely nothing[35].

Nothing.

There is neither enough funding nor interest in controlled experiments to address any of those variables, at least not in any kind of causal and generalizable way. This leaves every teacher assigning homework based on folk pedagogy, akin to folk medicine doctors used in the 1700s.

And as it goes for homework, so it goes for nearly every other aspect of education[36]. I.e. Compared to, say, health care professionals, this forces educators to make stuff up as they go along, and this truly is the best they can do. That’s why you can iterate and research all you want, you have almost no chance of exciting teenagers over logarithms.

The evidence for many educational practices is scant because little money is spent on educational research. The pharmaceutical industry invests about 17% of revenues in R&D. The education sector has a comparable budget, yet invests about 0.1% of it[37] in R&D.

As to how the little money is spent, please ask a graduate student at a nearby faculty of education how the professors feel about controlled experiments. Go on, do it!

Last time I did so, the PhD student told me professors all but expressly forbade them. “Controlled experiments and statistics are for other departments. We do qualitative work here.”

Poor training. Lack of time. Little scientific insight. Minimal useful feedback. It is outrageously difficult to improve as an educator, or even to know if you have improved.

Things are no better at institutional levels

For example, say a brilliant educator creates a self-paced, interdisciplinary high school, one where math, English, socials, and science are combined into projects like “Launching a Business” and “Generating Clean Electricity for the School”. Students are delighted as they learn and accomplish more than they ever thought possible. Parents see their adolescent children transform. Life skills, character, community, social and emotional regulation all blossom. School feels meaningful and purposeful.

It’s the school of dreams.

But the government has graduation requirements, including English 12 and Math 11 marks. Post-secondary schools often require standardized test scores. Generating clean electricity probably doesn’t prepare students for either quite as well as courses who specialize in meeting those requirements.

The dream high school would have a hard time guaranteeing high school diplomas, even to hard-working students. Nor could it help its students get into competitive universities. Nor would such a weird non-diploma issuing school have any credibility with employers.

Would you hire someone without a high school diploma for an important job?

The school will face opposition from politicians, superintendents, employers, parents, taxpayer associations, etc. “This isn’t a real school!”

Furthermore, building this school from scratch is so hard that it burns out school staff, who already have a high turnover rates. No team can stick with the project.

Even the narrower element of self-pacing - every child learning at their own pace, mastering content before moving on - is no miracle. A self-paced course means the student is often working in solitude and not with other learners, so positive norms don’t spread. There is less collective excitement. Students miss interacting[38] with the passionate expert. There is less opportunity to communicate or collaborate. The weaker and less motivated students fall just as far behind[39] as they would in a teacher-paced class. Rampant cheating arises because having 200 students writing the chemistry test at 200 different times guarantees some will tell others too much about the test in advance. This is why self-pacing lacks an impressive track record.

“People often ask what I wish I’d known before starting AltSchool," Max Ventilla said, "and I say: However difficult you think working in education is... multiply that by 10. Life at a startup is hard, but education is exponentially harder.”

Such institutional forces fight like hell against change, as documented by Larry Cuban[40].

Teacher professional development is super difficult. Institutional demands, such as graduation requirements and expectations of a “real school” make systemic changes nearly impossible.

And none of this even addresses the greater black box question of “What makes a good education?”

We have all criticized educators and the education system. Hopefully we can also recognize why hospitals improve faster than schools, why hard working teachers can struggle for years, and why education reformers face such daunting odds.

In Revolutionary Hypotheses[41], I share my ideas for overcoming them.

Notes

Another barrier is socioeconomic status, a daunting obstacle in the way of improving educational outcomes. But we know that it is not the only barrier to educational revolutions for two reasons.

  1. Organizations that basically exclude the underprivileged, such as many employers and universities, do not claim big advances in teaching either.
  2. Schools vary dramatically in how they help (or harm) low-income students and this has been measured by random assignment data[42].

I will write about how Bravo Math will tackle this problem in a later post.

References

[1] Nicholas Kristof, "Bill and Melinda Gates's Pillow Talk" (The New York Times)
https://www.nytimes.com/2015/07/19/opinion/sunday/nicholas-kristof-bill-and-melinda-gatess-pillow-talk.html

[2] Fast Company: Zuckerberg and Chan school closure
https://www.fastcompany.com/91584151/mark-zuckerberg-and-priscilla-chans-school-is-the-latest-billionaire-education-experiment-to-collapse

[3] Larry Cuban, "Goodbye AltSchool, Hello Altitude Learning"
https://larrycuban.wordpress.com/2019/07/08/goodbye-altschool-hello-altitude-learning/

[4] David Tyack & Larry Cuban, Addicted to Reform
https://thenewpress.org/books/addicted-to-reform/

[5] Class Clowns (Education Next review)
http://educationnext.org/clown-school-review-class-clowns-how-smartest-investors-lost-billions-education-knee-book/

[6] David Tyack & Larry Cuban, Tinkering Toward Utopia
http://www.hup.harvard.edu/catalog.php?isbn=9780674892835&content=reviews

[7] Steven Pinker, Enlightenment Now
https://en.wikipedia.org/wiki/Enlightenment_Now

[8] British Psychological Society: Learning styles myth
https://digest.bps.org.uk/2017/07/20/oh-dear-even-people-with-neuroscience-training-believe-an-awful-lot-of-brain-myths/

[9] Daniel Willingham, Learning Styles FAQ
http://www.danielwillingham.com/learning-styles-faq.html

[10] American Psychological Association: Learning styles survey
https://www.apa.org/news/press/releases/2019/05/learning-styles-myth

[11] Frontiers in Education: Teacher belief in learning styles
https://www.frontiersin.org/journals/education/articles/10.3389/feduc.2020.602451/full

[12] Daniel Willingham interview on learning styles
https://www.youtube.com/watch?v=rhgwIhB58PA

[13] Rachel Marks, April Fool's learning styles post
https://x.com/rachelgmarks/status/1509774117305786369

[14] Robert Bjork on feedback and learning
https://bjorklab.psych.ucla.edu/wp-content/uploads/sites/13/2016/07/RBjork_inpress.pdf

[15] Peak: Secrets from the New Science of Expertise
http://peakthebook.com/index.html

[16] Daniel Willingham, "How Knowledge Helps"
https://www.aft.org/sites/default/files/periodicals/LookItUpSpring2000.pdf

[17] History of educational technology promises
https://www.youtube.com/watch?v=GEmuEWjHr5c

[18] Veritasium, fitness analogy
https://www.youtube.com/watch?v=0xS68sl2D70

[19] Hack Education: History of the Future of Education
http://hackeducation.com/2015/02/19/the-history-of-the-future-of-education

[20] Education Next: The 5 Percent Problem
https://www.educationnext.org/5-percent-problem-online-mathematics-programs-may-benefit-most-kids-who-need-it-least/

[21] Scientific American: The Interleaving Effect
https://www.scientificamerican.com/article/the-interleaving-effect-mixing-it-up-boosts-learning/

[22] Anna Stokke podcast: Discriminative contrast
https://www.annastokke.com/transcripts/ep-63-transcript

[23] Business Insider: States without teacher unions
http://www.businessinsider.com/states-where-teachers-unions-are-illegal-2011-2

[24] Carl Wieman study on active learning
https://www.sciencedaily.com/releases/2018/01/180103101105.htm

[25] NCTQ, Learning About Learning report
https://www.nctq.org/dmsView/Learning_About_Learning_Report

[26] APM Reports: Sold a Story
https://features.apmreports.org/sold-a-story/

[27] Cognitive scientists on learning myths
https://youtu.be/t4VHdXRKRBY?t=10s

[28] Education Next: Making education research relevant
https://www.educationnext.org/making-education-research-relevant-how-researchers-can-give-teachers-more-choices/

[29] Teacher preparation interview/resource
https://www.youtube.com/live/mN8kpD3GhBc?si=JPqx_hvmpB-7QUlT&t=508

[30] Slate: Teaching isn't rocket science
http://www.slate.com/articles/life/education/2013/12/teaching_in_america_s_highest_need_communities_isn_t_rocket_science_it_s.html

[31] Washington Post: First-year teaching vs. lawyering
https://www.washingtonpost.com/news/answer-sheet/wp/2014/02/22/you-think-you-know-what-teachers-do-right-wrong/?arc404=true

[32] Washington Post: Teacher burnout
https://www.washingtonpost.com/news/answer-sheet/wp/2014/12/12/teacher-the-day-i-knew-for-sure-i-was-burned-out/?arc404=true

[33] TIME: Teaching in America
http://time.com/longform/teaching-in-america/

[34] Hardwired biases and learning
https://www.youtube.com/watch?v=oxZzoVp5jmI

[35] Anna Stokke: Homework evidence
https://www.annastokke.com/ep-37-resources

[36] Filling the Pail: Education's Dark Matter
https://fillingthepail.substack.com/p/educations-dark-matter

[37] Research spending in education
https://youtu.be/A7cEfgwpTxE?

[38] EdSurge: Self-paced learning
https://www.edsurge.com/news/2020-06-16-what-a-forgotten-instructional-fad-from-the-60s-reveals-about-teaching

[39] EdSurge: Self-paced learning outcomes
https://www.edsurge.com/news/2020-06-16-what-a-forgotten-instructional-fad-from-the-60s-reveals-about-teaching

[40] Larry Cuban blog
https://larrycuban.wordpress.com/

[41] Bravo Math: Revolutionary Hypotheses
http://bravomath.com/essays/revolutionaryhypotheses.html

[42] Waiting for 'Superman' / random-assignment discussion
https://www.imdb.com/title/tt1566648/