Seven principles for rigorous investigation.

Investigative inquiry has two failure modes. The first is paranoia: every coincidence is connected, every refutation is part of the cover-up, the system can never be wrong. The second is paralysis: nothing is ever provable, so nothing is ever said, and the powerful are protected by the very rigor that was supposed to constrain them. The methodology below is built to keep an investigator out of both failure modes simultaneously.

The first six principles describe how a pattern is elevated into a finding. Falsifiability asks whether the hypothesis even has investigative shape — whether it names the observations that would refute it. The simple-explanation test requires that the simplest documented explanation be honestly sought before any inference earns the right to be called a finding. Pattern vs. finding draws the line between scaffolding and structure — what points you to look somewhere is not the same as what holds up once you do. The two-phase research structure formalizes the boundary into a workflow with two distinct epistemic standards. The public-record standard says what evidence is admissible into Phase Two — and how that record sometimes has to be built rather than found. The institutional distinction says what kind of evidence sustains a finding: documented conduct, never inferred association.

The seventh principle, the retirement of hypotheses, is the symmetric case: how to honestly stop pursuing a hypothesis the documentary record will not bear out. Without it, the methodology only moves in one direction, and a discipline that can only move in one direction is not a discipline.

Read them in sequence the first time. Return to them in any order after. The threshold the first six collectively define — the point at which evidence makes denial costly enough to matter — is the topic of The Standard.