The King's Problem
In which our protagonist reads the books, and the books do not lie, but they are not telling the whole truth.
The books were, as the king had warned, very long.
There were twelve of them, each one the size of a small chest, each one filled with columns of numbers in the careful hand of the royal accountants. Kenji sat at a table in the treasury room, with Pip orbiting his head, Lian beside him with her notebook, and Elara standing by the door, watching.
He read for two days.
He did not just read. He did what he had done at every job: he mapped the system. He drew diagrams. He wrote pseudocode for the treasury's flow of money. He traced where the gold came in (taxes, tariffs, royal mines) and where it went out (military, infrastructure, court expenses, debt service). He calculated what should have been in the treasury at the end of each year, and he compared it to what the books said was actually there.
The numbers did not match.
They were close. They were within a percent or two. But they were not exact, and in a system that had been running for decades, a one percent discrepancy compounded to a large number.
"There is about a five percent annual leakage," Kenji said, on the morning of day three. "Which means that over the last ten years, the treasury has lost roughly forty percent of what it should have. That is a kingdom-poverty-level amount of money."
"Where is it going?" the king asked.
"I do not know yet. The books are accurate. The accountants are not lying. The numbers are just incomplete."
"How can numbers be incomplete?"
Kenji thought about how to explain this. He thought about his old job, about the times a customer had complained that the system was wrong, and the system had been right, and the customer had been right, because the system was not measuring the thing the customer actually cared about.
"The books track what enters and what leaves the official treasury. They do not track what enters and what leaves the unofficial treasury. The unofficial treasury is money that is moved before it reaches the books. It is collected by someone, somewhere, and then it disappears."
"The corruption is upstream."
"The corruption is upstream. The books are downstream. You have been reading the books and wondering why the kingdom is poor. The answer is: you are reading the wrong books."
The king stared at him. Then he said: "How do we read the right books?"
"We trace the money. We follow it from the source, through every transaction, until it either reaches the treasury or disappears. We need access to the tax collectors, the tariff officials, the mine managers, and the court accountants. We need to interview them. We need to follow the gold."
"That will take months."
"It will take weeks. I have a faster method."
"Which is?"
"I will cast a spell. The spell will read the mana signature of every coin in the treasury, and it will track where each coin has been. The signature is like a fingerprint. It does not lie. It cannot be forged. And it will tell me, for every coin in the treasury, where it came from and how many times it has been moved."
The king looked at him for a long moment. Then he said: "You can do that?"
"I can do that. I have been practicing for two days. The spell is ready."
"Cast it."
Kenji closed his eyes. He held out his hand. He whispered: "Function create(name='audit', inputs={target=coin, scope=all_in_treasury}, outputs={transaction_history, mana_signature_map}, cost=medium, stability=very_high)!"
The spell was, in the language of his old world, a tracing query. A diagnostic that asked the system: where has this data been? Who has touched it? What is its provenance?
The treasury room shimmered. A thousand points of light appeared in the air, each one tracking a single coin. The lights moved, slowly, drawing a map in the air. They clustered. They diverged. They formed patterns.
And the patterns were not what anyone expected.
"Here," Kenji said, pointing. "These coins came from the southern mines. They passed through three intermediate hands before reaching the treasury. The intermediate hands were not the official tax collectors. They were someone else."
"Who?"
"I do not know their names. But the mana signatures are consistent. The same three signatures, on every southern coin. They have been intercepting the tax revenue, taking a cut, and then passing the rest along. The cut is being moved somewhere else."
"Where?"
"I am still tracing. But the pattern is clear. There is a network. There is a system. And the system has a name."
He looked at the map. He looked at the three consistent signatures. He looked at the way they connected to the court, to the palace, to the council chambers.
"Baron Vex," he said.
The king's face went pale.
# WORLD.md (updated)
## Status
- Treasury audit: spell cast, leakage found
- 5% annual leakage, 40% total loss over 10 years
- Three mana signatures consistently intercept southern tax revenue
- One signature identified: Baron Vex (corrupt noble, S3 primary antagonist)
## Lessons
- The books are accurate. They are just reading the wrong thing.
- Upstream corruption is invisible to downstream audits
- A mana signature is like a fingerprint. It cannot be forged.
- The system has a name. The name is Baron Vex.