4 8 5 3 9 7 6

Take a few seconds and look at these numbers:

4 8 5 3 9 7 6

Now look away.

Can you remember them?

I came across this little experiment while reading Outliers by Malcolm Gladwell. In Chapter Eight, Rice Paddies and Math Tests, Gladwell explores a question I have heard in one form or another for most of my life: why do Chinese and other East Asian students seem to perform so well in mathematics?

We hear versions of it everywhere. At school. In movies. In jokes. In conversations about education.

“Asians are good at math.”

I've always thought of it as a stereotype. Perhaps one based on observable differences in academic results, but still a stereotype. It describes a group of people without really explaining anything.

Gladwell made me think about the question differently.

Four, eight, five, three, nine, seven, six

Try saying the sequence again in English:

Four. Eight. Five. Three. Nine. Seven. Six.

Now in Cantonese:

Sei. Baat. Ng. Saam. Gau. Cat. Luk.

The Cantonese words are remarkably compact.

Gladwell discusses research suggesting that the amount of information we can hold briefly in memory is partly constrained by how quickly we can say or rehearse it. Shorter number words can therefore make a string of digits slightly easier to keep in our heads.

It is a tiny advantage.

But the more interesting difference comes after ten.

English gives us eleven and twelve. Then thirteen, fourteen and fifteen. Eventually we arrive at twenty, twenty-one and twenty-two.

Chinese is almost boringly logical.

十一 — ten-one.

十二 — ten-two.

二十一 — two-ten-one.

二十二 — two-ten-two.

The language is practically showing you the arithmetic while you're saying the number.

As someone who speaks both Chinese and English, I had never really stopped to think about this. They were simply two different ways of saying the same numbers.

But perhaps they aren't quite the same.

One language asks a child to learn a collection of names. The other exposes more of the structure underneath them.

The mundanity of excellence

And perhaps the boring part is more important than it first appears.

It reminded me of an idea I've always liked: the mundanity of excellence.

We tend to look at exceptional results and search for exceptional explanations. Talent. Intelligence. Genetics. Some secret technique that separates one person—or one group of people—from another.

But excellence is often built from things that look remarkably ordinary when examined individually.

A slightly clearer number system.

A child practicing one more problem.

A teacher explaining an idea in a way that makes its structure visible.

A family treating study as part of the normal rhythm of life.

A culture in which persistence is expected rather than exceptional.

None of these looks extraordinary.

Repeated hundreds or thousands of times, however, small advantages can compound into outcomes that eventually look extraordinary.

Perhaps that is why “Asians are good at math” is such an unsatisfying explanation. It describes the outcome while hiding the machinery underneath it.

Is language an interface?

This was the part that stayed with me.

We normally think mathematics is universal. Two plus two is four whether you are in Sydney, Hong Kong, San Francisco, or Philadelphia . And mathematically, of course, it is.

But the interface through which a child first encounters mathematics isn't universal.

Language is part of that interface.

If one representation makes the underlying pattern slightly easier to see, then thousands of tiny interactions with that representation may matter.

Not because one child is inherently more intelligent than another. Not because being Chinese somehow comes with a mathematical gene. But because the same idea can become easier or harder to understand depending on how it is presented.

That sounds surprisingly familiar to anyone who designs software.

A complicated interface can make a simple system feel difficult. A thoughtful interface can make a complicated system feel almost obvious.

The capability underneath hasn't changed. The path to understanding it has.

Then there are the rice paddies

Gladwell goes further. He connects mathematical persistence with the agricultural history of rice-growing societies.

Traditional rice farming was extraordinarily demanding. A rice paddy rewarded careful attention, repeated effort and accumulated knowledge. Farmers could improve the productivity of relatively small plots through better irrigation, planting, maintenance and countless small adjustments.

The argument isn't that growing rice somehow makes your descendants good at algebra.

It is about the ideas cultures carry forward.

One of those ideas is that effort and outcome are connected.

If the field isn't producing enough, you work it differently. If the problem is difficult, you stay with it. If the answer is wrong, you try again.

There is something wonderfully mundane about that too. No secret. No shortcut. Just another attempt.

Of course, culture is complicated. Language is complicated. Education is complicated. There cannot be one neat explanation for the mathematical performance of millions of people across different countries, schools and generations.

The structure of Chinese numbers may contribute something. Teaching methods may contribute something. Family expectations, educational systems, practice and attitudes toward effort may contribute something else.

Which makes me think that perhaps the stereotype asks the wrong question.

Instead of asking:

Why are Asians good at math?

A much more useful question might be:

What environments make it easier for people to become good at math?

What we make easy to learn

I think about this a lot while building learning tools.

When someone struggles to understand something, our first instinct is often to look at the learner. Maybe they aren't concentrating. Maybe they haven't practiced enough. Maybe they're simply “not a math person.”

But sometimes the problem isn't the learner.

Sometimes it is the interface between the idea and the learner.

A strange word. An unnecessary abstraction. A badly designed explanation. A concept introduced before its foundations. A system that rewards memorizing the answer without revealing why the answer makes sense. Or an environment that makes getting something wrong feel like evidence that you aren't good at it.

Good learning design should do the opposite.

It should reveal structure. It should reduce unnecessary friction. It should connect new ideas to things we already understand. It should make another attempt feel natural. And whenever possible, it should make the complicated feel a little more obvious.

There is a temptation in technology to believe innovation must always look revolutionary. New AI models. New interfaces. New devices. New ways of doing everything.

Learning has reminded me that sometimes the most powerful improvements are much less exciting.

A clearer word.

A better example.

One less unnecessary step.

Five more minutes spent on the problem.

Then doing it again tomorrow.

The results may eventually look remarkable. The machinery underneath them can be surprisingly mundane.

4 8 5 3 9 7 6

Go back to those seven numbers.

Perhaps you still remember them.

Perhaps you don't.

Either way, what began as a simple memory test led me somewhere much more interesting.