The analytical tools we use to examine claims, evaluate evidence, and reason about belief — without demanding all-or-nothing conclusions.
Most religious discourse operates in binary mode: either a claim is absolutely true or it is rejected entirely. This creates a false dilemma that traps honest thinkers between total acceptance and total denial.
Bayesian reasoning offers an alternative. Instead of asking "Is this true or false?" we ask "How confident should I be, given the available evidence?" This allows beliefs to be held with degrees of confidence that can shift as new evidence emerges — without requiring the emotional violence of all-or-nothing conclusions.
Before examining specific evidence, establish how plausible the claim is based on background knowledge. What would you expect to be true before looking at the data?
If the claim is true, what evidence would we expect to find? What observations should follow logically from the claim being accurate?
If the claim is false, what would we expect to see instead? What alternative explanations exist, and what evidence would they predict?
Examine what evidence actually exists. Does it more closely match what we would expect if the claim were true, or what we would expect if it were false?
Adjust confidence proportionally to the strength of the evidence. Strong evidence warrants larger shifts; ambiguous evidence warrants smaller ones. No leap required.
A claim that cannot, even in principle, be shown to be wrong is a claim that carries no information about reality. If no possible observation could count against it, then no observation counts in its favor either.
When evaluating any claim — religious, institutional, or otherwise — we apply three diagnostic questions:
Can we identify, even hypothetically, what evidence would count against it? If no possible observation could disprove the claim, it is not making a testable assertion about reality.
Some claims begin as testable assertions but, when confronted with disconfirming evidence, are reframed into non-testable domains. This retreat pattern is itself diagnostically significant.
Tangible artifacts (buildings, texts, relics, institutions) are sometimes presented as evidence for non-material claims. When the artifact fails to support the claim, the claim is shifted to a domain where artifacts are declared irrelevant.
This is a distinctive pattern identified across religious and institutional contexts. It describes a three-part dynamic in which a tangible artifact is used to support a claim, but when the artifact fails under scrutiny, the claim retreats to an unfalsifiable domain — while institutional authority remains intact throughout.
A physical object, text, institution, or observable phenomenon is presented as supporting the truth of a larger claim. The artifact grounds the claim in the material world.
When scrutiny reveals problems with the artifact, the claim shifts: "It was never about the evidence — it's about faith." The claim retreats from testable to untestable ground.
The institution's authority is not diminished by the failure of its evidence. The system that made the original claim retains its credibility, despite the claim having failed its own evidential test.
These tools allow for honest examination without demanding all-or-nothing conclusions. They replace the binary framework that traps so many people — "either believe everything or reject everything" — with a graduated approach that respects both evidence and the complexity of human experience.
This connects directly to one of the project's core commitments: Bayesian updating is superior to binary reasoning. Evidence should shift confidence gradually, not demand all-or-nothing conclusions.
The goal is not to tell anyone what to believe. The goal is to provide rigorous tools for thinking clearly about claims that matter deeply — and to demonstrate that intellectual honesty and personal meaning are not enemies.