Three Truths Test
Classify claims by conviction, repetition, or independent verification
- Difficulty
- Moderate
- Time to result
- ~weeks to results
- Steps
- 5
- Confidence
- 97%
The Three Truths Test separates claims by the mechanism that makes them feel or become true. A personal truth is held deeply and may organize a person's life, but persuading others requires an additional act of persuasion. A repeated truth gains force because repeated exposure hijacks a useful human tendency to trust recurring patterns; propaganda exploits this pathway. An objective truth is established through experiments, reproducibility, peer review, and error correction that do not depend on one person's belief. The test asks what kind of support a claim actually has before assigning confidence or acting on it. It does not require attacking personal meaning. Instead, it prevents movement between categories without evidence, especially when conviction or familiarity is presented as though independent verification had occurred.
Origin
Tyson developed the taxonomy for a book chapter after noticing that scientists and religious communities use the word truth in materially different ways.
Core principles
- 01Deep conviction does not create transferable evidence
- 02Repetition can manufacture familiarity that feels true
- 03Objective claims require reproducible checks
- 04Bias is managed through independent error correction
How to run it
- 1
Define the claim
Rewrite the statement so it can be evaluated without slogans or ambiguous terms. Separate what is believed from what is being asserted about the external world.
Watch out An undefined claim can slide between categories during debate.
- 2
Test for personal truth
Ask whether the support is primarily inward conviction, identity, faith, or felt meaning. Respect that role while recognizing that it does not by itself verify the claim for others.
Pro tip Ask what evidence could persuade someone who does not share the identity or experience.
- 3
Test for repeated truth
Trace whether familiarity comes from repeated messages rather than new or independent evidence. Look for coordinated repetition, emotional framing, and dehumanizing effects.
Pro tip Count distinct sources, not repeated appearances of one source.
Watch out Familiarity can masquerade as corroboration.
- 4
Test for objective truth
Seek reproducible observation, independent verification, peer criticism, and mechanisms that expose bias or error. Lower confidence when only the original claimant can produce the result.
Pro tip Ask what result would disconfirm the claim.
- 5
Calibrate action
Match confidence and consequences to the claim's actual category. Require stronger objective support as the cost imposed on other people rises.
Watch out Do not use a personal truth to compel outsiders without additional evidence.
In the wild
Tyson uses belief that Jesus is a savior or Muhammad is the last prophet as examples of truths that can be deeply real to an individual. Another person, however, requires persuasion because the conviction is not shared evidence.
→ The belief is recognized as personally meaningful without being misclassified as independently verified.
A scientist performs an experiment and another person tries to verify it. If the result fails to reproduce, peer scrutiny may reveal a bias the original scientist could not see.
→ Built-in error correction lowers the chance that one person's bias becomes accepted truth.
Common mistakes
Treating categories as insults
Calling something personal does not make it worthless. The classification identifies its evidence mechanism and limits, not its emotional importance.
Counting echoes as sources
Many repetitions of one unsupported claim do not provide independent verification, even when the message becomes familiar.
Assuming scientists lack bias
The objective category depends on verification and error correction precisely because scientists are human and can be biased.
Is it for you?
Best for
It is best for evaluating consequential claims in research, media, organizations, and public debate.
Not ideal for
It is not ideal as a pretext for denying people harmless personal beliefs that are not being imposed as objective claims.
From the transcript
“recognizing that there are three categories of Truth”
“propaganda is the very essence of repeating something until you think it's true”
“perform an experiment I get a result someone else verifies the experiment or not”
From the episode
Episode 251: Neil deGrasse Tyson: Why Embracing Wonder is Critical For Personal Growth
Neil deGrasse Tyson