Methodology
What is CANOPTICON
A forensic intelligence platform for Canadian political and economic structure analysis. It surfaces structural patterns by integrating primary documents, named entities, and verifiable claims into theses with explicit predictions, falsifying observations, and public correction logs.
Source grading (Admiralty Code)
Sources are graded A through F. A is primary documents (Hansard, OIC, SEDAR+ filings, peer-reviewed papers) with named authors and verifiable provenance. B is secondary reports with traceable sourcing. C is opinion or single-sourced reporting. D, E, F are progressively less reliable. Each thesis lists the grade distribution of its supporting evidence.
Toulmin claim structure
Every claim carries six fields: data (the evidence), claim (the assertion), warrant (the rule connecting them), backing (broader support for the warrant), qualifier (the calibrated language), and rebuttal (conditions under which the claim fails).
Calibrated language (IPCC scale)
Probabilistic claims use the IPCC verbal scale: virtually certain (≥99%), very likely (≥90%), likely (≥66%), about as likely as not (33–66%), unlikely (≤33%), very unlikely (≤10%), exceptionally unlikely (≤1%). When in doubt between adjacent qualifiers, the weaker is correct.
Root-striking discipline
Every investigation identifies its intervention level: symptom (managing effects), mechanism (proximate causes), or structural (root design). Theses cannot publish until they reach Why #4 or #5 in the Five Whys protocol — the structural condition or root design choice.
Editorial standards
Every published thesis ships with a methodology appendix listing sources used, grades, roles run, contradictions considered, coined phrase definition. The trail is the moat; critics must engage the evidence.
Corrections process
Every correction is public. The Corrections page lists what changed, when, why, and which item was affected. The Brier score on the prediction ledger is the operator's calibration track record.
Credibility-positive practices
We preserve contradictions rather than resolve them prematurely (machloket as legitimate posture). We bank predictions before outcomes. We name our operator. We publish the graph artifact behind every thesis so the evidence trail is independently traversable.