Xhairs Apologetics Audit Lab Inductive Symmetry Audit

First audit

Inductive Symmetry Audit

Test whether an apologetic conclusion relies on an inductive pattern that is treated as forceful only when it points toward theism.

How to use this audit

Start with the apologetic argument exactly as you would normally present it. The app then compares the inductive permission you grant to that argument with the permission you grant to nearby patterns that push against the same conclusion.

The point is not to force identical conclusions. The point is to make every asymmetry pay rent by naming a real evidential difference.

  • Step 1: choose the apologetic pattern closest to the argument being assessed.
  • Step 2: adjust the claim, evidence, rule, defense mode, and force slider until they match the actual argument.
  • Step 3: judge each parallel induction using the same evidential standards you want applied to the anchor.
  • Step 4: if a parallel is weakened or rejected, supply the differentiator that explains why.
  • Step 5: use the score, repair panel, summary, and AI prompt to decide whether the original conclusion is proportionate.

Treat the tool as a consistency audit. A high score means the argument may still be interesting, but it needs clearer limits, better differentiators, or a more modest conclusion before it should be treated as licensed.

Preset postures

Archetype lens

Apply a common stance profile, then inspect how the tension changes.

What counts as a good differentiator?

A good differentiator identifies a real evidential difference between the accepted induction and the parallel induction. It should change scope, evidence quality, defeaters, dependence, rival hypotheses, or specificity burden without simply restating the preferred conclusion.

The best differentiators are transferable: a fair critic could apply the same standard in other cases without quietly protecting only the preferred theological result. Weak differentiators usually rename the conclusion, appeal to mystery, or move from empirical evidence to necessity without a bridge premise.

Stronger differentiators
  • Independent evidence that the target domain really differs from the observed domain.
  • A quality difference such as replication, reliability, specificity, or source independence.
  • A defeater that undercuts the parallel without also undercutting the anchor rule.
  • A clear rival-hypothesis comparison showing why one explanation predicts the evidence better.
Weak differentiators
  • "God is different" without explaining which evidential variable changes.
  • Calling the preferred induction necessary while treating contrary inductions as merely empirical.
  • Moving from a modest explanation to a specific doctrine without added support.
  • Rejecting a parallel because it threatens the conclusion rather than because it is disanalogous.

When in doubt, ask: if this same differentiator were used against my preferred argument, would I still count it as legitimate? If the answer is no, the differentiator probably needs more independent support.

Compare archetypes

Same argument, different posture

Scores below show how the selected argument behaves under each preset stance.

Live tension

0 Low

Accepted anchor

Current stance map

Reading the stance map

The left score is the total tension created by the current settings. Each tile shows whether a parallel induction is accepted, weakened, or rejected, and how much tension that stance creates relative to the accepted anchor rule.

The anchor is the rule you are allowing to support the apologetic claim. The parallel tiles are nearby inductions that use comparable reasoning but pressure the argument in an uncomfortable direction. The tile number rises when a similar parallel receives much less permission than the anchor.

  • Accept means the parallel receives roughly the same kind of evidential permission as the anchor.
  • Weaken means the parallel matters, but you think a real evidential difference reduces its force.
  • Reject means the parallel should not count unless a strong differentiator explains why.

Click a tile to jump to the controls for that parallel. If a tile is high, the fastest repair is usually to accept more of the parallel pressure, lower the force of the anchor, or give a stronger differentiator.

Input

Argument profile

Choose a pattern, then adjust the claim and evidential pressure.

Setting the anchor

The anchor is the inductive assertion the apologetic argument wants to count as strong. Raising the force slider means you are asking that rule to do more evidential work, so rejected parallels will create more tension.

Use the claim field for the conclusion being defended, the evidence field for the observed pattern being emphasized, and the accepted rule field for the general inductive move that connects the evidence to the conclusion. Keep the rule as general as the argument needs it to be.

  • Inductive means the argument relies on observed regularities or analogies.
  • Abductive means the argument claims theism is the best explanation among rivals.
  • Modal, metaphysical, analytic, or rhetorical modes ask the rule to do extra work beyond ordinary observation.

The force slider is not a confidence meter for the whole religion. It measures how much work this particular rule is being asked to do in this argument.

Defense mode
Understanding defense modes

Defense mode describes what kind of move the argument is making from evidence to conclusion. This matters because different moves carry different burdens. A modest induction from observed patterns is not the same as a claim about necessity, metaphysical structure, or analytic truth.

  • Inductive: The argument says observed regularities or analogies support a probable conclusion. This is the lightest mode, but it must still treat similar regularities with similar permission.
  • Abductive: The argument says theism is the best explanation. This requires comparison against live rivals, not merely showing that theism is compatible with the evidence.
  • Modal: The argument moves from what appears regular, intelligible, or unavoidable to what must be true. This requires a bridge from observed patterns to necessity.
  • Metaphysical: The argument makes a claim about the deep structure of reality, such as grounding, ultimate explanation, or being itself. Ordinary examples can motivate this, but they do not automatically settle it.
  • Analytic: The argument treats the key rule as definitional or conceptually true. This can clarify terms, but it may stop doing evidential work unless an additional premise connects the definition to reality.
  • Rhetorical: The argument leans on obviousness, intuition, incredulity, or persuasive framing. This can be psychologically powerful, but it is the riskiest mode because it can hide missing premises.

Choose the weakest mode that honestly captures the argument. If the argument only needs ordinary pattern recognition, use Inductive. If it claims rival explanations fail, use Abductive. If it says the conclusion is necessary or built into reality, use Modal or Metaphysical. Higher-burden modes increase the score because they ask the anchor rule to do more work.

8
Risk 0

Result

Symmetry status

Interpreting the score

The score measures leftover asymmetry after your differentiators are counted. It is a pressure reading, not a truth verdict. A high score means the favored induction is being allowed to count more than similar rejected inductions.

Low scores usually mean the anchor and parallels are being treated with comparable permission. Moderate scores suggest the argument may be repairable by clarifying scope or improving a few differentiators. High and severe scores indicate that the preferred conclusion is receiving special treatment.

  • Lower the score by accepting relevant parallels, reducing anchor force, or supplying independent differentiators.
  • Raise the score by giving the anchor high force while rejecting similar parallels with weak or circular reasons.
  • Use the bars to see which parallel claims are carrying the most tension.

A high score does not show that the conclusion is false. It shows that this route to the conclusion is carrying an unresolved consistency burden.

    Repair

    Proportionate conclusion

    Weaker claims usually survive the audit better than doctrinally loaded ones.

    Using the repair panel

    Repairs preserve what the evidence can reasonably support while removing unsupported specificity, scope drift, or modal inflation. A repaired claim may be less dramatic, but it is usually harder to accuse of cherry-picking.

    The cards below are generated from the selected argument pattern and the current tension profile. They are not rebuttals; they are ways to restate the argument so its conclusion is proportionate to the evidence being used.

    • Modest claim lowers the conclusion from proof to pressure, question, or possibility.
    • Scope control keeps observations inside the domain where they were actually observed.
    • Bridge premise names the extra support needed before moving to a theological conclusion.
    • Immediate pressure point appears when one rejected parallel is creating especially high tension.
    • Burden shift appears when the defense mode has moved from induction into modal, metaphysical, or analytic territory.

    Use these repairs as candidate revisions, not as final answers. A good repair should keep the legitimate evidential pressure while refusing to infer more than the evidence licenses. If the repaired claim feels too modest, that is often a sign that the original argument owed additional bridge premises.

    Summary

    Audit summary report

    A compact record of the current score, stance ledger, flags, and repairs.

    AI prompt

    Ready-to-paste exploration prompt

    A self-contained prompt that updates with every current stance and differentiator.

    Using the AI prompt

    Paste this entire prompt into an AI assistant. It includes the current argument profile, parallel stances, score drivers, flags, repair options, and output instructions so the assistant can examine the tensions directly.

    The prompt is regenerated whenever you change the pattern, edit the argument, move the force slider, change a treatment, or revise a differentiator. Copy it after the settings match the stance you actually want to explore.

    The generated prompt asks the assistant for a diagnosis, the top tensions, a stronger differentiator, a repaired claim, and follow-up prompts you can paste as the next step in the conversation.