Epistemic Framework

The analytical tools we use to examine claims, evaluate evidence, and reason about belief — without demanding all-or-nothing conclusions.

Bayesian Reasoning

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.

The Five-Step Process

1

Define prior plausibility

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?

2

Define expected evidence if claim were true

If the claim is true, what evidence would we expect to find? What observations should follow logically from the claim being accurate?

3

Define expected evidence if claim were false

If the claim is false, what would we expect to see instead? What alternative explanations exist, and what evidence would they predict?

4

Evaluate observed evidence

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?

5

Update rationally

Adjust confidence proportionally to the strength of the evidence. Strong evidence warrants larger shifts; ambiguous evidence warrants smaller ones. No leap required.

This matters for religious claims because it allows honest engagement without the false binary of "completely true" or "completely false." Most claims deserve nuanced confidence levels, not tribal verdicts.

The Falsifiability Principle

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:

Is the claim falsifiable?

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.

Does the claim retreat to unfalsifiability under pressure?

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.

Are material tokens used to shield metaphysical claims?

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.

The Token + Unfalsifiable Retreat Pattern

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.

Three Evaluation Questions

1

Is a tangible artifact used as evidence?

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.

2

When the artifact fails, is the claim reframed to a non-testable domain?

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.

3

Does authority remain intact despite empirical failure?

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.

Why This Matters

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.