Critical thinking for contested claims

Understand the claim.
Examine the evidence.
Test the explanation.

Critical thinking for claims, narratives and explanations. Learn to separate evidence from interpretation, test causality, and calibrate what the facts allow you to conclude.

Test the Claim evidence networkA central claim is tested against sources, evidence, explanations, context and calibrated confidence.SOURCESEVIDENCEEXPLANATIONSCONFIDENCECONTEXTCLAIM

Evidence assessment

Not a political score. A calibrated description of what the evidence supports.

False

Strong, independent evidence contradicts the precise claim.

Unsupported

The claim is possible, but positive evidence is missing.

Misleading

True elements are presented without decisive context.

Legitimate debate

The disagreement concerns values, causes, policy or priorities.

Unfalsifiable

No conceivable observation is allowed to weaken the explanation.

The governing principle

Critical thinking is not believing the opposite of the official explanation.

Institutions can lie, protect their interests and make serious mistakes. Alternative explanations can also be selective, exaggerated or unsupported. The task is to apply the same standards of evidence, causation and uncertainty to every explanation—including your own.

Learn critical thinking

One claim. Eight disciplined moves.

The method begins with precision and ends with a conclusion proportionate to the evidence. It works on official statements, alternative narratives—and your own assumptions.

01

State the precise claim

Replace vague actors such as ‘they’ or ‘the system’ with who allegedly did what, when and how.

02

Separate observation from interpretation

‘The company profited’ is an observation. ‘The company caused the event’ is a causal claim.

03

Find the fact kernel

Identify the true, plausible or historically grounded element that gives the wider story credibility.

04

Map the causal chain

Look for decisions, communication, money, capacity, coordination and actual implementation.

05

Generate competing explanations

Could incentives, error, chaos, group pressure or independent choices produce the same outcome?

06

Check source independence

Ten sites repeating one original assertion are still one source—not ten.

07

Search for disconfirming evidence

A robust explanation must be capable of losing strength when facts run against it.

08

Calibrate the conclusion

Use documented, probable, possible, unsupported and contradicted with deliberate precision.

Learn each move properlyEight guided lessons with worked examples, practice exercises and mastery checks.
Start the lessons

Test a claim

Turn a persuasive story into a testable case.

Use this private worksheet in your browser. Nothing is submitted. The value is in making every implied step visible.

Reasoning patterns

Recognise the move—not an enemy.

The expanded reference library examines recurring shortcuts in evidence, causality, framing, identity, influence and debate.

Causality

“They profited, therefore they caused it.”

A real financial interest establishes motive or incentive—not a decision, causal chain or implementation.

Use the causal-chain map →

Source reasoning

“Many sources report the same thing.”

Apparent confirmation can collapse when every article traces back to one unverified assertion.

Check source independence →

Self-sealing claims

“The lack of evidence proves the cover-up.”

If both evidence and its absence confirm a claim, the explanation no longer exposes itself to testing.

Ask the stop question →
Explore the complete pattern librarySearch and filter practical tests across six analytical domains.
Open the library

Applied cases

See the method risk a conclusion.

Twelve sourced cases test official failures, anti-establishment narratives, wartime stories and confident forecasts with the same visible standard.

Explore the cases

Influence & persuasion

Content, behaviour and attribution are different questions.

A message can be true and still be distributed manipulatively. It can resemble propaganda without proving who produced it.

01

Assess the content

Is the claim accurate, selective, misleading or unsupported?

02

Observe the behaviour

Is there coordinated timing, concealment, impersonation or artificial amplification?

03

Calibrate attribution

What evidence connects that behaviour to a specific actor, decision and operation?

Do not skip the ladder.Narrative resemblance is weak evidence of origin. Attribution requires a separate case.

Difficult conversations

Sometimes evidence is no longer the only issue.

Respect for a person does not require respect for weak reasoning. Challenge the claim rather than diagnosing the identity behind it.

Conversation stateWhat you may observeUseful response
Open disagreementQuestions are answered and both sides can name what might change their view.Examine the evidence together.
Identity defenceThe subject shifts, standards move and counterevidence feels personal.Ask one concrete question, summarise and pause.
Dominance or contemptRidicule, character trials, blame reversal or endless circles.End the evidence debate and set a short boundary.

Practical tools

Small tests that improve large decisions.

Use these as prompts before you believe, reject or share a consequential claim.

Evidence ladder+

Move from anonymous posts and decontextualised clips toward primary documents and independent evidence lines.

Symmetry test+

Would you accept the same evidence standard if the conclusion supported the opposing side?

Attribution ladder+

Separate resemblance, coordination indicators, analytical attribution and legally tested responsibility.

Fact-kernel test+

Acknowledge the true element before testing the much larger conclusion built around it.

Causal-chain map+

Make every implied step between an observation and a conclusion visible and testable.

Stop question+

What observation would make you reduce confidence in the explanation?

Research an unfamiliar claimFollow the full verification workflow and keep a private, downloadable research note.
Open the toolkit

Method & transparency

Do not trust the biography. Inspect the reasoning.

Test the Claim is intentionally presented without a personality-led brand. An author’s status, profession or identity cannot substitute for evidence.

Every substantive analysis should expose its sources, causal reasoning, uncertainty, update date and correction history. New evidence must be able to change the assessment.

  • Same standards in every political direction
  • Primary sources wherever possible
  • Fact and judgement clearly separated
  • Uncertainty stated rather than concealed
  • Corrections preserved, not silently erased