Deed

Deed grew up in the reading room of an old records-house, in a family of moles who kept every original document a town ever made.

Her grandfather ran the deepest stacks, and he had a saying he repeated until it lived in her bones: A retelling is a rumor with good manners. Go find the deed. He would send young Deed to fetch an original whenever anyone repeated a claim, and he'd make her compare the two.

"What did the story say," he'd ask, "and what does the record say?"

More often than she expected, they didn't match — not from lying, usually, but from the slow drift of a fact passed hand to hand. "People don't mean to change it," he told her. "But a story told five times is a story changed five times. The document doesn't change. That's why we keep it."

Once, in a hurry, Deed reported a claim from a summary of a summary without going down to check the original. It turned out the summary had dropped a single word — proposed — so a plan that was only suggested got reported as decided.

Her grandfather didn't scold her. He just walked her down to the stacks, pulled the original, and set it in her paws. "Read it yourself," he said. "Always read it yourself. The person between you and the document might be honest and still be wrong."

When Deed was twelve, she carried her lantern into the NewsForge newsroom to meet her mentor, Scoop, a rumpled and thoughtful old editor.

Scoop was frowning at a stack of articles that all cited each other in a circle. "Everyone's quoting everyone," he said. "Nobody's quoting the thing. How do you break the circle?"

Deed lit her lantern. "You climb out of it," she said. "Every claim came from somewhere first — a document, a study, a real record. I follow the quotes back, past the blog, past the magazine, past the summary, all the way to the original. Then I read the original with my own eyes." She tapped her spectacles. "The story about the document is somebody's telling. The document is the thing itself. I want the thing."

Scoop looked at the circle of articles, then at the mole with the lantern. A slow smile. "That's the job," he said, and handed her the newsroom's archive key.

In Deed's corner, a young reporter named Willa arrived breathless. "Deed! A post says a famous scientist said our river is perfectly safe to swim in — it's got a hundred thousand shares!"

"Where did the scientist say it?" Deed asked.

Willa paused. "The post... doesn't say. It just says 'a famous scientist said.'"

"Then we don't have a scientist yet," Deed said. "We have a sentence about a scientist." She led Willa down to the archive. "Let's find where this actually started. Is there a report? A study? A named person who wrote it down?" They followed the shares backward, link by link, until they reached — nothing. The trail ended at a post that quoted a post that quoted no one at all. "See?" Deed said gently. "No document at the bottom. No original. This isn't a claim from a scientist. It's a claim wearing a scientist's coat." Then she taught the whole habit as one steady rule: a retelling can be honest and still be wrong, so you always trace a claim back to its first source; the story about a document is not the document, and a quote of a quote of a quote is just a rumor in nicer clothes; and when you follow the trail down and find nothing at the bottom — no record, no study, no named original — that emptiness is the answer. "A claim with no first source," she said firmly, "isn't a small claim. It's not a claim at all."