Chapter 48: The Records Confirm Level 2
The notification arrived at 9:14 AM Friday.
FROM: Caltech Visiting Researcher Documentation Office TO: Dr. S. Cooper CC: Adam Carter SUBJECT: Academy City Exchange Documentation — Records Delivered
The requested records have been delivered to your department mailbox.
Adam read the notification on his phone, standing in his apartment with his coffee cooling in his hands. The records had arrived. The Ghost Index had done its work. The institutional documentation would confirm exactly what it should confirm.
He put down the coffee. Put on his jacket. Went to Caltech.
The walk took twenty-three minutes. The Witness Protocol registered Sheldon's behavioral signature from the moment Adam entered the building perimeter — the particular rhythm of someone who had received information and was processing its implications.
He has the records. He is reading them now.
Adam walked to the shared workspace.
He sat at his desk.
He waited.
---
Sheldon appeared in the doorway at 10:07 AM.
He was carrying a printed document — the Academy City records, sixteen pages, bound with a standard departmental clip. His expression was unreadable, which was unusual. Sheldon's expressions were typically transparent to anyone paying attention.
"Your records confirm Level 2 classification," he said.
"Yes."
"Telekinetic subtype. Research documentation specialty. Standard Academy City assessment protocol."
"Yes."
Sheldon walked into the workspace. He sat down in the chair across from Adam's desk — the same position he had occupied during the first interrogation.
"The calibration anomaly pattern in this building," he said, "is inconsistent with a Level 2 esper operating within standard parameters."
"The Academy City classification is the institutional authority on my ability level."
"I am aware of that." Sheldon set the records on the desk between them. "I am telling you the institutional authority is wrong."
There was a pause.
Leonard, at his desk across the room, had stopped typing.
---
Sheldon laid out his case methodically.
He went through each piece of evidence: the EM sensor logs with their forty-five data points of correlated micro-fluctuations. The calibration anomaly pattern showing magnitudes that exceeded Level 2 baseline by a factor of 1.7. The access log correlation at seven of ten. The whiteboard framework whose cognitive synthesis output was inconsistent with a Level 2 esper's documented processing capacity.
He went through the engineering diagnostic precision — the thermal coupling problem, the sensor arm diagnosis — that exceeded documented telekinetic range by an order of magnitude.
He went through the notation analysis, the facilities records, the timeline of correlated events.
"This evidence," he said, "is not consistent with Level 2."
Adam listened. He did not interrupt.
"The records are accurate within their own system," Sheldon continued. "I am not disputing that. I am saying the system's assessment methodology is insufficient to detect what you are doing. The Academy City protocol measures direct ability output under controlled conditions. It does not measure passive environmental interaction. It does not account for field-resonance effects in complex electromagnetic environments. It does not capture cognitive synthesis output that occurs below conscious threshold."
He paused.
"You are operating above your classification. The records do not show it because the records measure the wrong things."
Adam considered his response carefully.
"I have been transparent about my classification," he said. "I have not made claims beyond it. The Resonance Engine's passive engagement functions are not standard telekinetic output — they interact with environmental EM fields in ways that a standard assessment protocol would not capture. That does not mean the Level 2 classification is wrong. It means the assessment was conducted under conditions that did not include this building's instrument environment."
Sheldon was watching him intently.
"You are saying the classification is technically accurate in the context in which it was made."
"Yes."
"And you are not going to tell me what the accurate classification would be in this environment."
"I am telling you I was assessed as Level 2 by the institutional authority. That assessment is documented. Those records are in front of you."
Sheldon stared at him for a long moment.
"That is not a no."
The words landed in the quiet workspace.
He has identified the deflection as incomplete. He knows I am giving him technically true answers that do not extend past the question boundary. He knows the methodology argument is a redirect.
And he is right.
---
Leonard, at his desk, said: "Sheldon."
Sheldon did not look away from Adam. "I am conducting an investigation, Leonard."
"I know."
Leonard did not say anything else. He went back to his work.
He had chosen, for the third time, not to enter.
---
The conversation continued for another twelve minutes.
Sheldon asked more questions. Adam answered them with the same technique — technically accurate statements that confirmed the institutional record without extending past it. Each answer was true. Each answer was incomplete. Each answer left the same gap that Sheldon had identified.
Finally, Sheldon stood.
"I will continue my investigation," he said.
"I expected that."
Sheldon gathered the records. He walked to the door. He paused.
"For the record," he said, "I do not believe you have been dishonest with me. I believe you have been strategically accurate. Those are different things."
He left.
---
[PASADENA CENTRAL LIBRARY — FRIDAY AFTERNOON]
Adam did not return to his apartment after leaving Caltech.
He walked to the library instead — not to the locker room, just to the reading room. The zero-signature space where the Molecular Conductor's passive range did not register any sensitive equipment, where the Phase 2 Witness Protocol had no behavioral signatures to track, where he could sit in genuine quiet for the first time in weeks.
He sat at his usual table.
He thought.
He has been managing this for four months. The management has been nearly perfect.
It is not enough.
Sheldon is right. The records are technically accurate and substantively insufficient. The deflection is identified. The institutional cover is exhausted.
The next conversation will not have the same structural supports.
The next conversation will require a different response.
He pulled out his notebook. Found a blank page.
He wrote:
"What do I actually want to happen here?"
He stared at the question.
He had not asked it before. He had been managing the situation — deflecting, redirecting, maintaining the cover, protecting the gap between what they saw and what he was. He had not stopped to consider what outcome he was actually working toward.
Do I want Sheldon to stop investigating? That requires giving him a satisfying answer — which means either a convincing lie or a partial truth large enough to close the inquiry.
Do I want to maintain the cover indefinitely? That requires continuous management, continuous deflection, continuous exhaustion.
Do I want to tell him something real?
He sat with this question.
The reading room was quiet. The afternoon light filtered through the high windows. The zero-signature space offered no information, no behavioral signatures, no electromagnetic patterns to track.
Just the question.
Just the silence.
Just the particular mathematics of deciding what kind of person to be.
---
[ADAM'S APARTMENT — SATURDAY, 2 AM]
The Synthesis Core fired at 1:47 AM.
Adam woke with the warmth-at-sternum and reached for his notebook automatically, the motion practiced after months of 3 AM outputs. He opened to a blank page. Waited.
The output arrived.
Two lines.
He read them.
He read them again.
He sat in the dark apartment for ten minutes, the notebook open on his lap, the two lines visible in the streetlight filtering through the window.
Then he wrote, below the output:
"That is probably right."
He did not write the two lines themselves. He did not need to see them again to know what they said.
The Synthesis Core had answered the question he had asked at the library.
What do I actually want to happen here?
The answer was not a strategy. It was not a deflection technique. It was not a probability calculation or an evidence management protocol.
It was simpler than any of those.
And it was probably right.
Get Early Access to New Chapters
Thank you for reading. For those who want to skip the wait, my Patreon is currently 21 chapters ahead of the public sites.
Schedule: 7 new chapters released every 10 days.
Benefit: Gain a significant lead of 7 to 21 chapters depending on your tier.
Support the project and start reading the next arc now: Patreon.com/IsekaiStories