Charter Sectioning Engine · method: Inversion Quotient v2
Mark: VoidSection™, asserted on the basis of first public use. This repository's commit history is the dated record. The instrument is offered under this mark by Nebelkrähe (Occultus Theoretician).
An instrument that sections a charter against its record.
In contract law, a void section is a provision still printed in the document but no longer operative. Institutions accumulate them: the mission statement outlives the mission, the form survives the telos. This instrument locates a charter's void sections and scores the inversion — the distance between the document and the conduct.
One assessment is a section — a single cut through the specimen. The product is serial sectioning: repeated cuts across reporting periods reconstruct the volume, the way histology rebuilds tissue and tomography builds the scan. One cut is an opinion; the series is the imaging.
Inversion Quotient v2 — full methodology in research/03_INVERSION_QUOTIENT.md:
- Five-point inversion scale per dimension: 0 aligned · 25 drifted · 50 reform-captured · 75 substantially inverted · 100 fully inverted.
- Disqualifying dimensions by entity class (a child-serving institution's child protection, a fiduciary's client-first duty, a truth-claiming institution's truth), with three-zone handling: a clear trigger overrides the aggregate; the threshold-adjacent zone requires dual reporting — strict and conservative interpretations printed side by side with the evidence that would move the score — because a single number would hide a genuinely contested judgment.
- Mechanism specification: drift, reform capture, founding capture, inversion, and constitutional-design effects are different failure modes and are named separately, per dimension.
- Component-specific scoring where variation is wide. No single number where the evidence does not support one.
What the incumbent raters do not do:
- Rubric pre-registration. Every rubric hashes; the hash publishes before any entity is scored under it. The scoring criteria cannot quietly move after the fact.
- Mandatory falsifiers. Every scored dimension carries, in writing, what evidence would move it up and what would move it down.
- Minimum evidence binding, both sides. A score that cannot cite its record does not publish.
- Misfire-inclusive. Score revisions print permanently. A rater that only shows its current opinion is promotion, not measurement.
- Instrument under public test — not validated until its own ledger says so.
Pre-registration. No entity has been scored. The committed engine, rubric definitions, and demo profile (a synthetic entity) constitute the sealed instrument before first use — the criteria on the record before any result exists. Sequence from here, gates not dates: pilot entity hand-scored → serial backfill → extraction assist → public profiles.
The method descends from a documented research program with a primary-source floor of
26 April 2026. Lineage documents and the capture hash-ledger:
research/PROVENANCE.md.
The instrument audits records and claims, not persons. Its output is a score, never a certification. Assessment is per-entity behavioral audit on the documented public record. Consistency is not correctness — resolution is.
All rights reserved. Published for pre-registration and public audit of the method; not licensed for reuse.