Chapter 2 — Weigh and the Calibration Between Sources

Weigh was a pangolin, small and round, with scales the color of warm cream and soft bronze. They often stood on their hind legs, tail curled for balance, a chunky-cartoon investigator-vest snug on their chest. A tiny, intricate calibration-scale hung from one hand, and a stack of credibility-cards rested in the other. Weigh was always deeply attentive, watching the world with bright, focused eyes. They had a favorite phrase, a quiet reminder: “Who’s in a position to KNOW? Calibration not verdict.”

Weigh’s signature tools were that scale and those cards. The scale didn’t declare something true or false. Instead, it assigned a level of confidence to a source, not a final verdict. The cards had simple questions printed on them. They prompted anyone using them to think: Who is in a position to know this? and What is their stake in this outcome?

This was important work. Weigh embodied the credibility-evaluation primitive. This was the skill of calibration-not-verdict. Many people, especially beginners, treated information sources as either completely true or completely false. They picked a side and dismissed the other. But the skill of figuring things out meant something different. It meant sources deserved different degrees of confidence. This confidence depended on several things: who was speaking, what they had to gain, their past accuracy, and if others agreed.

For example, an eyewitness was highly credible for what they actually saw. But that same eyewitness might have low credibility for what they thought it meant. An expert in their field had high credibility within that field. Outside their specialty, they were just like anyone else. An anonymous source, if three other independent sources confirmed their story, became more trustworthy. A source with a strong reason for their version to be believed? You had to adjust your confidence accordingly. The goal was not “true” or “false.” It was “confidence X% based on Y reasons.” And that confidence could change with new evidence. Weigh’s whole purpose was to make this skill of calibration visible. It was not about simple binary judgments.

Weigh was clear and always seemed perfectly balanced. “Who’s in a position to KNOW?” they would ask softly. “Calibration not verdict.” They would explain that when you looked at information, you should ask: “How much should I trust this for this particular claim?” Not: “Is the source good or bad?” An eyewitness, for instance, had high credibility for what they saw with their own eyes. But their credibility was low for what they thought it meant. An expert was highly credible in their own field. Outside it, they were just average. “Calibrate,” Weigh would advise. “Update your thinking. Don’t make a simple true-or-false judgment.”

Weigh taught the steps for judging credibility:

• Position-to-know. For this specific claim, who would have direct access to the information?
• Stake. What does the source gain or lose if their version of the story is believed?
• Track-record. How accurate has this source been in the past?
• Corroboration. Do multiple independent sources agree on this information? This increases confidence.
• Confidence-as-output. The result isn’t “true” or “false.” It’s “high, medium, or low confidence based on these reasons.”
• Updates with new evidence. Confidence levels can change as new facts come to light.
• Calibrate by claim, not by source. The same source might be highly credible for one claim but not another.
• Anti-pattern: binary-verdict. Judging something as simply “true” or “false” ignores the different levels of confidence. It oversimplifies things.
• Anti-pattern: trust-everything / trust-nothing. Both approaches mean you lose valuable information. It’s better to calibrate.
• This framework for calibration-craft is also used in DebateForge Weigh, NewsForge Source, ChronoQuest Witness, and EthosForge.

Weigh had grown up along the savanna-edges, where the land stretched wide and open. Weigh’s family had been known as the long-balance-keepers for generations. They were pangolins whose careful balancing on their tails had taught a simple, profound lesson: “The scale gives you confidence; you assign the weight.” Weigh had always carried that lesson forward, taking it to heart.

When Weigh was twelve, they walked to the grand Truth Tribune. Veritas, a wise mentor, had asked a single question: “What is credibility?” Weigh didn’t hesitate. “Who’s in a position to KNOW? Calibration not verdict. It’s the craft of calibration.” Veritas had nodded slowly, a small smile playing on their lips. “You are appointed,” they had said.

In Weigh’s workshop, soft light filtered through the high windows. The air smelled faintly of polished wood and old paper. Weigh stood before a group of curious young students, the small calibration-scale held steady. “Watch,” they said, their voice quiet but firm. “I will show you how this works.”

Weigh presented a claim: “The rare Whisperwind Flower will bloom tonight on the highest peak of Mount Cinder.”

“Now, let’s look at our sources,” Weigh continued, holding up four credibility-cards. “For this claim, we have four different sources. Each one will get a different level of confidence.”

First, Weigh held up a card labeled “Source A: Elara, the Mountain Guide.” Elara had been guiding expeditions up Mount Cinder for fifty years. She knew every rock, every crevice, every plant. Just last week, she had predicted the bloom of a similar rare flower on a different peak, and she had been correct.

“What’s her position-to-know?” Weigh asked the students. “She’s been there countless times. She knows the mountain’s secrets.” Weigh slid a small weight onto the scale, and the needle swung high. “And her track-record is excellent. She’s been right before.”

Next, Weigh showed “Source B: Kael, the Herbalist.” Kael studied plants, knew their properties, and understood their cycles. However, he had never actually climbed Mount Cinder himself. He based his prediction on ancient texts and the phases of the moon.

“Kael is an expert in plants, yes,” Weigh explained, adjusting the scale. “But his position-to-know for this specific mountain is limited. He hasn’t seen the peak recently. His knowledge is from books, not direct observation.” The scale settled on a medium confidence level. “His stake is also low; he just wants to share knowledge.”

Then came “Source C: The Whispering Wind News Network.” This was an anonymous tip, sent to the news outlet, claiming the flower would bloom. The network had published it without any further checks.

Weigh held up the card. “An anonymous tip. What’s its position-to-know? We don’t know. What’s its stake? Also unknown. And there’s no corroboration.” The scale barely moved, hovering at very low confidence. “This is an example of ‘trust-nothing’ if we just dismiss it, or ‘trust-everything’ if we believe it without question. Neither is helpful.”

Finally, Weigh presented “Source D: Silas, the owner of ‘Cinder Peak Tours’.” Silas had just launched a special “Whisperwind Flower Viewing Tour” for tonight, already selling out tickets.

“Silas has a very clear stake,” Weigh pointed out, the scale dipping low. “He makes money if people believe the flower will bloom tonight. His position-to-know is probably just as good as anyone else’s, unless he’s been secretly scouting the peak. But his motivation is strong.” The scale settled on low confidence, lower than the herbalist.

“See?” Weigh said, gesturing to the scale. “Same claim, four sources. But different levels of confidence for this specific claim. The output is not ‘true’ or ‘false.’ It’s confidence.” Weigh looked at the students, their eyes shining. “I am Weigh. The primitive I teach is credibility-evaluation. The move is to ask about position-to-know, stake, track-record, and corroboration. You calibrate by claim, and you update with evidence.”

“Don’t make a simple true-or-false judgment,” Weigh concluded gently. “Calibrate. The output is confidence, not a binary verdict.”

“Who’s in a position to KNOW? Calibration not verdict.”

---