So Why Are Some People Better At It?
I Didn't Buy It Either
A friend of mine is learning Korean.
900 days on Duolingo, 2-year tutoring. Even living in Korea. Still can't hold a conversation. Of course, I went on a whole rant.
"Stop studying rules. Just talk to people. That's how I learned English. Not from textbooks. Not from grammar drills. Language isn't something you learn. It's something you absorb. Pattern recognition. You pick it up by doing it, whether you're ready or not."
Explaining to a room full of people that language is 'acquired, not learned.' Very confident. But now...
She let me finish and went:
"OK. If it's all just pattern recognition, why does it come easy to some people and not to others?"
I was like "Well. It's about how much you practice...?" She looked at me. I would've looked at me too. We both knew we didn't buy it. The question stayed with me for weeks.
One Extra Question
One day, I was up late reading when I came across an article about chimpanzees and mirrors.
A chimp, in front of a mirror, for the first time. It freaks out. Another chimp. Rival. Threat.
A few days pass.
It starts making faces at itself. Checking its teeth. Using the reflection to look at its own back. It figures out something a dog never does.
That's not another chimp. That's me.
A dog sees movement and reacts. End of story. But the chimp sees movement, then asks what it is. One extra question. That's the whole gap between them.
'Wait... is that me?' Took him a few days. Took me a few years.
I stopped reading for a second.
The gap isn't about who sees more. It's about what you do with what you see. Or what you do with what you already think.
So what does it look like when you don't ask that extra question?
Here's a good one.
Ice cream sales go up. Drowning rates go up too. The numbers move almost perfectly together.
Wait... does ice cream cause drowning?
The data doesn't lie. But it doesn't tell the whole truth either.
Obviously not. Right? But the pattern is real. If you stop there, the conclusion feels right. And it's completely wrong. One more question fixes it.
Why do both go up at the same time?
Hot weather. More people at the beach. More swimming, more drowning. More heat, more ice cream. The sun was doing both things at once.
The pattern was real. It just wasn't the whole story. One more question, one more layer, and the picture changes completely.
Kurzweil's building from Part 2. 300 million workers stacked on floors, each one passing results up to the next. But the dog stopped at the first floor. The chimp climbed one more.
Why? What gives one animal more floors than another?
The neocortex. A rat's is thin, almost flat. A few floors. A chimp's is thicker, more folded. A human's is so crinkled that if you unfolded it, it would cover a dinner table.
Smooth brain, simple life. Wrinkled brain, existential crisis.
More floors, more chances to take what the floor below just found and ask "what is this, really?"
Nobody knows which came first. Maybe our ancestors kept asking deeper questions, and the brain grew more layers to keep up. Or maybe the layers came first, and deeper thinking followed. Science hasn't settled that yet. But it's hard to ignore: more floors, more depth.
Everything alive sees patterns. The difference is what happens after. Whether you take what you found and ask one more question about it.
In theory, though.
When you go deep enough, your brain starts losing what got you there.
Mine does, anyway. Do you remember what the first essay in this series was even about?
Putting It Into Words
I was writing Part 2 late one night. I'd just typed the last line. "That's what I think intelligence is." I leaned back and read it from the top.
It held up. Everything connected. I was about to close my laptop when something caught. A sentence I'd been fine with an hour ago suddenly had a hole in it. If it's all patterns, what about causes? What about the stuff nobody taught me?
Those questions hadn't been there before. Not while I was thinking about it. Not while I was planning the essay. They only showed up after I wrote the idea down and looked at it from the outside.
Part 3 came from those holes. And it kept happening. Every time I finished a section, the writing showed me what I'd missed. It wasn't the thinking that moved me forward. It was the writing.
Most of my ideas showed up after midnight. Most of them disappeared by morning. Should've been writing.
Why?
Here's how I think about it. Your brain is a desk. You can spread out four, maybe five ideas at once. Try to open a sixth and the first one slides off the edge. If depth means stacking question on question on question, you run out of desk space fast.
But the moment a pattern in my head became a sentence on a screen, it stopped being something I had to hold. It became a thing I could set down and look at. And once it was set down, my mind was free to ask the next question.
Words gave me more desk space.
But I wasn't just spreading out. I was also building up.
Part 1 asked what if I wasn't the dumb one. Part 2 stood on that question and built a theory: intelligence is pattern recognition. Part 3 tried to tear it down, and couldn't. This one, Part 4, is standing on all three, asking what makes some people go deeper than others.
None of them existed before I wrote the one before. I wasn't clearing desk space. I was building floors.
Language wasn't just helping me express my thinking. It might be what made the thinking possible in the first place.
In Arrival, learning a new language literally rewires how you think. A linguist learns an alien language, and the more she learns it, the more her mind changes. The language doesn't give her new words. It gives her new thoughts.
A psychologist named Vygotsky noticed the same thing watching children in the 1930s. Language isn't just how we talk to each other. It's how we talk to ourselves. Children mumbling while they play aren't being weird. They're organizing their own thinking.
Every animal communicates. But not every animal stacks thoughts. What's happening there?
Noam Chomsky had an answer I hadn't expected. Human language has something no animal communication system has: recursive structure. You can put a sentence inside a sentence inside a sentence. "The man who saw the dog that chased the cat that ate the fish." No limit. No ceiling.
That sentence is five layers deep. Try holding all five in your head without it. The words do the holding for you, so your brain can keep climbing.
Language isn't just how you share your thoughts. It's how you stack them.
Not being weird. Just building her first floor.
What's Upstairs
I think about the kid at his desk at 2 AM when I was seventeen. Doing everything the Korean education system told me to do. Memorize the formula. Apply the rule. Get the answer.
He wasn't lazy. He wasn't dumb. He was working harder than anyone in the room.
But nobody encouraged him to ask "why does this formula work?" Nobody questioned "what happens if you change this variable?" The answer was the answer. Why wasn't on the test.
College wasn't any different. I was a political science major at one of the top schools in Korea. You'd expect serious debate, papers that made you defend your thinking. Not at all. Most exams were multiple choice.
This is basically what happened to me. I'll tell you.
One I'll never forget. "Select all that are incorrect." Eight options. Options 2 and 5 were clearly wrong. But option 1 said: "None of the answers here are correct." If 2 and 5 are already incorrect, then "none are correct" is itself wrong. I checked 1, 2, and 5.
Got it marked wrong.
Maybe the professor meant it as a "none of the above" option, something outside the logic of the other choices. Fine. But he never said that. When students pushed back with logic, he pushed back with authority. The TA sided with him without blinking. He gave the same test for ten years. Same logic. Same answer: because I said so.
I stopped taking political science classes after that. Filled my schedule with science electives instead. At least in science, when the answer is wrong, it's wrong. Doesn't matter who's standing at the podium.
In Korean education, every time I asked why, the answer was either authority or "it's always been this way."
Maybe that's why school never felt right to me. It never told me there were more floors. Never encouraged me to experiment with my own hands. Never even let me ask "what if?"
I spent four essays arriving at three things.
(1) Intelligence is pattern recognition. (2) The difference isn't who sees more patterns. It's who stacks deeper. (3) And the tool that lets you stack, the thing that gives your brain more desk space, is language.
Patterns. Depth. Language.
Four essays to get here. (1) how we make sense of things, (2) how we stack patterns, (3) and what lets us climb them. One machine that does all three.
In December 2022, Open AI released a machine that does all three. It recognizes patterns across everything ever written. It stacks layers deeper than any human can hold in their head. And it doesn't just use language. Language is all it is.
They called it a Large Language Model.
Patterns. Depth. Language. Right there in the name.
I don't think that's a coincidence.