Dr. Quark

Dr. Quark had learned this as a young researcher who once tried to answer everything by staring.

She'd been curious whether a certain pond-bug preferred warm water or cold. So she'd watched. And watched. She filled notebooks with careful observations of bugs in one pond — beautiful, detailed, and completely unable to answer her question, because she'd never set up a cold pond to compare against the warm one. She had mountains of looking where she'd needed a test.

Her mentor flipped through the notebooks. "Gorgeous observing," she said. "And it can't answer you, because 'does it prefer warm or cold' is a comparison — and you only built one side." She'd drawn three doors on the whiteboard. "There are kinds of science questions, and each has its lab. What is this / what's happening — you observe. Does changing this change that — you experiment, with a fair comparison. What would happen if / why does the pattern hold — you build a model and test its predictions. Pick the lab before you pick up a single tool. You just spent a month in the wrong one."

Young Quark had felt it — a month of lovely, useless watching. But it stuck: name the kind of investigation first. Observing, testing, or modeling. The kind is the door.

She came to the Science Labs grown, and took the clipboard from the old researcher.

"You won't teach them biology or physics or chemistry," the mentor said. "The lab-folk do that, each in their room. You stand at the junction and sort the questions into the right kind of investigation."

"That's the whole game most kids are missing," Quark said. "They know facts. What they can't do yet is look at a question and say this one needs a test, not a look — or this one needs a model, because you can't experiment on a whole galaxy. I don't teach the science. I make sure they walk into the lab that can actually answer them."

A boy arrived, exasperated. "My question is 'why is the sky blue,'" he said. "Do I set up an experiment? I don't have a sky."

"And that," Quark said, "is exactly the clue that tells you the kind. You can't run a fair experiment on the whole sky — you can't make a second, red sky to compare. So it's not an experiment question." She pointed to the third door, the one full of soft-glowing diagrams. "It's a model question. You build an explanation of how sunlight and air scatter light, and then you test the model's predictions — 'if my model's right, then sunsets should be red, and they are.' You can't change the sky, so you change your model and check it against what the sky already does." She grinned. "The impossibility of the experiment wasn't a dead end. It was a signpost — it told you which lab you needed."

The boy looked at the glowing diagrams. "So 'I can't test this directly' doesn't mean 'give up.' It means 'this is a modeling question.'"

"Now you're reading the junction," Quark said.