Research artifactnot admittedmatrix

DEMO CORPUS MASTER MATRIX FINAL

Status: research corpus complete, pre-capture / pre-adjudication

source: COUNTERPEDIA_DEMO_CORPUS_CAPTURE_WAVE1_v0_1/dossiers/DEMO_CORPUS_MASTER_MATRIX_FINAL_v1_2.md

sha256: 36ae96bf71f6581cb06e89a18fb72ce59b938a83297c1b4aaead87ba1240adad

Presentation-only rendering. Counterpedia preserves this document’s source Markdown bytes exactly and formats them for reading here. This does not admit the document, verify its claims, or convert it into a governed Counterpedia entry.

Visual projection · derived from this matrix

Dimensionality explorer

One corpus, multiple explicit projections. Source categories remain PRIMARY, YES, and not central; sorting, rarity, overlap, and graph position are transparent transforms of those labels rather than truth scores.

Subjects

Capabilities

Primary

181

Active

391/480

Emphasis Capabilities Subjects

PRIMARY YES not central

Capability

Titan

E. Palestine

Theranos

MH370

737 MAX

Flint

FTX

Deepwater

Dieselgate

Lehman

Cambridge Analytica

COVID Origins

temporal truth

denominator / measurement discipline

causal decomposition

competing institutional findings

attributed explanation

adjudication vs allegation

procedural disposition vs factual finding

changing valuation

technical/scientific vs legal/political causation

confidence levels

model-dependent inference

unknown / unresolved state

source supersession

source-version drift / correction

first-party vs independent finding

majority / dissent

negative evidence / exclusion

information-flow / knowledge-at-time

physical/raw artifact → inference

regulatory compliance vs safety/health

configuration/version truth

sampling / protocol provenance

system-level vs individual-level metric

later recovery vs earlier deprivation

root cause vs trigger vs amplifier

investigative claim vs merits judgment

counterfactual claim discipline

raw data vs derived model

capability vs deployment

deployment vs efficacy

efficacy vs outcome causation

population model vs individual causation

settled core vs disputed scope

best-supported vs established

scientific vs intelligence assessment

incomplete access / missing evidence

absence of evidence vs evidence of absence

hypothesis-family typing

confidence label vs probability

institution-name continuity vs posture continuity

Select a cell to inspect one subject × capability relationship.

Status: research corpus complete, pre-capture / pre-adjudication
Demo Corpus Twelve: COMPLETE ✓

Titan ✓
East Palestine ✓
Theranos ✓
MH370 ✓
Boeing 737 MAX / MCAS ✓
Flint Water Crisis ✓
FTX ✓
Deepwater Horizon / Macondo ✓
Volkswagen Dieselgate ✓
Lehman Brothers / 2008 ✓
Cambridge Analytica / Facebook ✓
SARS-CoV-2 / COVID-19 Origins ✓

Legend: PRIMARY = especially strong; YES = useful secondary exercise; — = not central.

Capability	Titan	E. Palestine	Theranos	MH370	737 MAX	Flint	FTX	Deepwater	Dieselgate	Lehman	Cambridge Analytica	COVID Origins
temporal truth	PRIMARY	PRIMARY	YES	PRIMARY	PRIMARY	PRIMARY	PRIMARY	YES	PRIMARY	PRIMARY	PRIMARY	PRIMARY
denominator / measurement discipline	YES	YES	PRIMARY	PRIMARY	YES	PRIMARY	PRIMARY	PRIMARY	PRIMARY	PRIMARY	PRIMARY	PRIMARY
causal decomposition	PRIMARY	PRIMARY	YES	YES	PRIMARY	PRIMARY	YES	PRIMARY	YES	PRIMARY	PRIMARY	PRIMARY
competing institutional findings	YES	YES	—	YES	PRIMARY	YES	YES	YES	YES	PRIMARY	YES	PRIMARY
attributed explanation	PRIMARY	PRIMARY	YES	YES	PRIMARY	YES	YES	PRIMARY	YES	PRIMARY	PRIMARY	PRIMARY
adjudication vs allegation	—	—	PRIMARY	—	YES	YES	PRIMARY	PRIMARY	PRIMARY	PRIMARY	PRIMARY	YES
procedural disposition vs factual finding	—	—	PRIMARY	—	YES	YES	PRIMARY	PRIMARY	PRIMARY	PRIMARY	PRIMARY	YES
changing valuation	—	—	YES	—	—	—	PRIMARY	—	YES	PRIMARY	—	—
technical/scientific vs legal/political causation	YES	YES	YES	—	PRIMARY	YES	YES	PRIMARY	YES	YES	YES	PRIMARY
confidence levels	YES	YES	YES	PRIMARY	YES	YES	YES	YES	PRIMARY	PRIMARY	PRIMARY	PRIMARY
model-dependent inference	YES	—	—	PRIMARY	YES	YES	YES	YES	PRIMARY	YES	PRIMARY	PRIMARY
unknown / unresolved state	YES	YES	YES	PRIMARY	YES	YES	YES	YES	YES	YES	PRIMARY	PRIMARY
source supersession	PRIMARY	PRIMARY	PRIMARY	PRIMARY	PRIMARY	PRIMARY	PRIMARY	PRIMARY	PRIMARY	PRIMARY	PRIMARY	PRIMARY
source-version drift / correction	YES	PRIMARY	PRIMARY	PRIMARY	PRIMARY	PRIMARY	PRIMARY	YES	PRIMARY	PRIMARY	PRIMARY	PRIMARY
first-party vs independent finding	PRIMARY	PRIMARY	PRIMARY	YES	PRIMARY	PRIMARY	PRIMARY	PRIMARY	PRIMARY	PRIMARY	PRIMARY	PRIMARY
majority / dissent	—	—	—	—	YES	—	—	—	—	PRIMARY	YES	PRIMARY
negative evidence / exclusion	YES	YES	—	PRIMARY	YES	YES	YES	YES	YES	YES	PRIMARY	PRIMARY
information-flow / knowledge-at-time	PRIMARY	PRIMARY	YES	YES	PRIMARY	YES	YES	PRIMARY	PRIMARY	PRIMARY	PRIMARY	PRIMARY
physical/raw artifact → inference	PRIMARY	YES	YES	PRIMARY	PRIMARY	YES	YES	PRIMARY	YES	YES	PRIMARY	PRIMARY
regulatory compliance vs safety/health	YES	YES	YES	—	PRIMARY	PRIMARY	—	PRIMARY	PRIMARY	YES	YES	YES
configuration/version truth	YES	—	YES	YES	PRIMARY	YES	YES	YES	PRIMARY	YES	PRIMARY	PRIMARY
sampling / protocol provenance	YES	YES	YES	YES	YES	PRIMARY	YES	YES	PRIMARY	YES	PRIMARY	PRIMARY
system-level vs individual-level metric	—	YES	YES	—	—	PRIMARY	PRIMARY	YES	PRIMARY	YES	PRIMARY	YES
later recovery vs earlier deprivation	—	—	YES	—	—	—	PRIMARY	—	YES	YES	YES	—
root cause vs trigger vs amplifier	YES	YES	—	YES	PRIMARY	YES	YES	PRIMARY	YES	PRIMARY	YES	PRIMARY
investigative claim vs merits judgment	—	—	YES	—	—	YES	YES	YES	YES	PRIMARY	PRIMARY	YES
counterfactual claim discipline	—	YES	—	YES	YES	—	YES	—	—	PRIMARY	PRIMARY	YES
raw data vs derived model	—	—	YES	YES	YES	YES	YES	—	YES	—	PRIMARY	PRIMARY
capability vs deployment	—	—	YES	YES	YES	—	—	YES	YES	—	PRIMARY	YES
deployment vs efficacy	—	—	YES	—	YES	—	—	YES	YES	—	PRIMARY	YES
efficacy vs outcome causation	—	—	YES	YES	YES	YES	YES	YES	PRIMARY	PRIMARY	PRIMARY	YES
population model vs individual causation	—	YES	YES	—	—	PRIMARY	—	YES	PRIMARY	—	PRIMARY	YES
settled core vs disputed scope	YES	YES	PRIMARY	YES	YES	YES	YES	YES	PRIMARY	YES	PRIMARY	PRIMARY
best-supported vs established	YES	YES	YES	PRIMARY	YES	YES	YES	YES	YES	YES	PRIMARY	PRIMARY
scientific vs intelligence assessment	—	—	—	YES	—	—	—	—	—	—	—	PRIMARY
incomplete access / missing evidence	YES	PRIMARY	YES	PRIMARY	YES	YES	YES	YES	YES	YES	PRIMARY	PRIMARY
absence of evidence vs evidence of absence	PRIMARY	YES	YES	PRIMARY	YES	YES	YES	YES	YES	YES	PRIMARY	PRIMARY
hypothesis-family typing	YES	YES	YES	PRIMARY	YES	YES	YES	YES	YES	YES	YES	PRIMARY
confidence label vs probability	—	—	—	YES	—	—	—	—	YES	YES	YES	PRIMARY
institution-name continuity vs posture continuity	YES	YES	PRIMARY	YES	PRIMARY	PRIMARY	PRIMARY	PRIMARY	PRIMARY	PRIMARY	PRIMARY	PRIMARY

Twelve-subject regression map

1. Titan — later interpretation vs earlier knowledge

A later engineering interpretation can establish what a signal meant without proving what an operator actually knew at the time.

2. East Palestine — information existence vs decision-maker knowledge

Evidence can exist within a system without reaching the person making an emergency decision.

3. Theranos — procedural verbs

Alleged, settled, neither-admitted-nor-denied, convicted, acquitted, dismissed and affirmed are different legal states.

4. MH370 — principled unknown

Observed signal, derived measurement, probabilistic route model, negative search evidence and unresolved cause remain distinct.

5. Boeing 737 MAX — distributed causation

A central mechanism does not erase interacting sensor, design, certification, training, workload, maintenance and crew-performance factors.

6. Flint — measurement/protocol truth

A regulatory percentile is not a household value or a health threshold; method/population/date travel with the number.

7. FTX — valuation and recovery

Petition-date claim value is not current-market asset value; >100% plan recovery does not retroactively erase fraudulent deprivation.

8. Deepwater Horizon — technical cause vs legal fault

Comparative-fault percentages are not physical-causation percentages; guilty pleas have offense-specific scope.

9. Dieselgate — settled core vs disputed halo

Corporate emissions cheating is settled while vehicle scope, multiplier, actor knowledge and modelled health effects require separate evidence.

10. Lehman — causal verb and institutional dissent

Preexisting crisis, trigger, amplifier and contagion are different causal roles; liquidity and solvency are separate states.

11. Cambridge Analytica — data → deployment → effect

Data possession is not model deployment; deployment is not persuasion; persuasion is not election-outcome causation.

12. COVID Origins — confidence without manufactured certainty

Best-supported, plausible, not ruled out, low confidence, moderate confidence, unresolved and established are not synonyms.

Corpus-level product requirements discovered

The twelve dossiers collectively imply that a governed claim system needs first-class support for:

Source posture
- scientific paper
- regulator finding
- complaint/allegation
- settlement
- guilty plea
- verdict
- court finding
- inspector-general report
- legislative majority/dissent
- intelligence assessment
- press release
- first-party statement
Claim relationship type
- observed
- measured
- derived
- modeled
- alleged
- admitted
- adjudicated
- supports
- contradicts
- qualifies
- supersedes
- triggers
- amplifies
- contributes
- excludes
- fails to exclude
Quantitative context
- denominator
- time
- population
- protocol
- valuation date
- confidence interval
- legal class
- system vs individual scope
Uncertainty vocabulary
- established
- strongly supported
- best supported
- favored
- likely
- plausible
- not ruled out
- insufficient evidence
- unresolved
- contradicted
- superseded
- unknowable from current record
Source evolution
- exact edition
- correction
- erratum
- amendment
- supersession
- historical posture preserved
- current posture separately addressable
Evidence access
- captured
- capture refused
- unavailable
- raw data absent
- mirror only
- archive only
- dynamic source
- underlying package unresolved
Institutional segmentation
- team vs parent organization
- majority vs dissent
- agency vs interagency assessment
- current officeholder statement vs coordinated analytic product
- government-hosted first-party exhibit vs government finding

Final Demo Corpus invariant

Counterpedia's job is not to maximize certainty. It is to preserve the strongest proposition the record actually permits—together with the exact evidence, posture, time, method, confidence and missing links that keep it from becoming a stronger one.

The Demo Corpus Twelve is now complete at the research/pre-capture layer.