Eliezer Yudkowsky - Harry Potter and the Methods of Rationality

Which made it entirely plausible that in the tiny wizarding world, where science didn't seem much known at all, no one had ever tried the first, the simplest, the most obvious thing that any modern scientist would think to check.

The books were full of complicated instructions for all the things you had to do exactly right in order to cast a spell. And, Harry had hypothesized, the process of obeying those instructions, of checking that you were following them correctly, probably did do something. It forced you to concentrate on the spell . Being told to just wave your wand and wish probably wouldn't work as well. And once you believed the spell was supposed to work a certain way, once you had practiced it that way, you might not be able to convince yourself that it could work any other way...

...if you did the simple but wrong thing, and tried to test alternative forms yourself.

But what if you didn't know what the original spell had been like?

What if you gave Hermione a list of spells she hadn't studied yet, taken from a book of silly prank spells in the Hogwarts library, and some of those spells had the correct and original instructions, while others had one changed gesture, one changed word? What if you kept the instructions constant, but told her that a spell supposed to create a red worm was supposed to create a blue worm instead?

Well, in that case, it had turned out...

...Harry was having trouble believing his results here...

...if you told Hermione to say "Oogely boogely" with the vowel durations in the ratio of 3 to 1 to 1, instead of the correct ratio of 3 to 1 to 2, you still got the bat but it wouldn't glow any more.

Not that belief was irrelevant here. Not that only the words and wand movements mattered.

If you gave Hermione completely incorrect information about what a spell was supposed to do, it would stop working.

If you didn't tell her at all what the spell was supposed to do, it would stop working.

If she knew in very vague terms what the spell was supposed to do, or she was only partially wrong, then the spell would work as originally described in the book, not the way she'd been told it should.

Harry was, at this moment, literally banging his head against the brick wall. Not hard. He didn't want to damage his precious brains. But if he didn't have some outlet for his frustration, he would spontaneously catch on fire.

Thud. Thud. Thud.

It seemed the universe actually did want you to say 'Wingardium Leviosa' and it wanted you to say it in a certain exact way and it didn't care what you thought the pronunciation should be any more than it cared how you felt about gravity.

WHYYYYYYYYYYYYYYYY?

The worst part of it was the smug, amused look on Hermione's face.

Hermione had not been okay with sitting around obediently following Harry's instructions without being told why.

So Harry had explained to her what they were testing.

Harry had explained why they were testing it.

Harry had explained why probably no wizard had tried it before them.

Harry had explained that he was actually fairly confident of his prediction.

Because, Harry had said, there was no way that the universe actually wanted you to say 'Wingardium Leviosa'.

Hermione had pointed out that this was not what her books said. Hermione had asked if Harry really thought he was smarter, at eleven years old and just over a month into his Hogwarts education, than all the other wizards in the world who disagreed with him.

Harry had said the following exact words:

"Of course."

Now Harry was staring at the red brick directly in front of him and contemplating how hard he would have to hit his head in order to give himself a concussion that would interfere with long-term memory formation and prevent him from remembering this later. Hermione wasn't laughing, but he could feel her intent to laugh radiating out from behind him like a dreadful pressure on his skin, sort of like knowing you were being stalked by a serial killer only worse.

"Say it," Harry said.

"I wasn't going to," said the kindly voice of Hermione Granger. "It didn't seem nice."

"Just get it over with," said Harry.

"Okay! So you gave me this whole long lecture about how hard it was to do basic science and how we might need to stay on the problem for thirty-five years , and then you went and expected us to make the greatest discovery in the history of magic in the first hour we were working together. You didn't just hope, you really expected it. You're silly."

"Thank you. Now -"

"I've read all the books you gave me and I still don't know what to call that. Overconfidence? Planning fallacy? Super duper Lake Wobegon effect? They'll have to name it after you. Harry Bias."

"All right! "

"But it is cute. It's such a boy thing to do."

" Drop dead. "

"Aw, you say the most romantic things."

Thud. Thud. Thud.

"So what's next?" said Hermione.

Harry rested his head against the bricks. His forehead was starting to hurt where he'd been banging it. "Nothing. I have to go back and design different experiments."

Over the last month, Harry had carefully worked out, in advance, a course of experimentation for them that would have lasted until December.

It would have been a great set of experiments if the very first test had not falsified the basic premise.

Harry could not believe he had been this dumb.

"Let me correct myself," said Harry. "I need to design one new experiment. I'll let you know when we've got it, and we'll do it, and then I'll design the next one. How does that sound?"

"It sounds like someone wasted a whole lot of effort ."

Thud. Ow. He'd done that a bit harder than he'd planned.

"So," said Hermione. She was leaning back in her chair and the smug look was back on her face. "What did we discover today?"

"I discovered," said Harry through gritted teeth, "that when it comes to doing truly basic research on a genuinely confusing problem where you have no clue what's going on, my books on scientific methodology aren't worth crap -"

"Language, Mr. Potter! Some of us are innocent young girls!"

"Fine. But if my books were worth a carp, that's a kind of fish not anything bad, they would have given me the following important piece of advice: When there's a confusing problem and you're just starting out and you have a falsifiable hypothesis, go test it. Find some simple, easy way of doing a basic check and do it right away. Don't worry about designing an elaborate course of experiments that would make a grant proposal look impressive to a funding agency. Just check as fast as possible whether your ideas are false before you start investing huge amounts of effort in them. How does that sound for a moral?"

"Mmm... okay," said Hermione. "But I was also hoping for something like 'Hermione's books aren't worthless. They're written by wise old wizards who know way more about magic than I do. I should pay attention to what Hermione's books say.' Can we have that moral too?"

Harry's jaw seemed to be clenched too tightly to let any words out, so he just nodded.

"Great!" Hermione said. "I liked this experiment. We learned a lot from it and it only took me an hour or so."

"AAAAAAAAAAAAAAHHHHHHHHHHHHHHH!"

In the dungeons of Slytherin.

An unused classroom lit with eerie green light, much brighter this time and coming from a small crystal globe with a temporary enchantment, but eerie green light nonetheless, casting strange shadows from the dusty desks.