Theism Gradient

Start with the Guided Assessment panel. Use Next Unrated if you want a systematic pass through all 50 claims, or use the category and view filters if you want to begin with the area that matters most to you, such as prayer, healing, revelation, or scripture.

For a first pass, do not overthink each number. Give a provisional Confidence score, a separate Personal Substantiation score, and a short note where the rating feels unstable. The tool becomes more useful after enough claims have ratings for the dashboard, diagnostics, and dependency map to reveal patterns.

The five bands move from thinner to thicker commitments: Minimal Deism, Design Deism, Personal Theism, Interventionist Theism, and Specific Christian Theism. The point is to show where your ratings gain or lose support as the claims become more specific.

A profile can be strong in earlier bands and weaker in later bands. That does not mean the later claims are false. It means this tool is asking whether your current evidence and bridge premises support moving from a broad creator/source claim toward claims about Christian divine action, Jesus, scripture, Spirit, salvation, and transformation.

Each claim card asks for two ratings and an optional note. Confidence is how credible the claim currently seems. Personal Substantiation is how well you can personally defend that claim with reasons, evidence, bridge premises, and answers to live alternatives.

The annotation accordion inside each claim card gives a substantiation focus, bridge pressure, evidence checklist, and current read. Use those prompts before assigning very high scores, especially on interventionist and specifically Christian claims.

The tool separates C and P because a person can sincerely believe a claim more strongly than they can personally substantiate it. That gap is often where inherited trust, community reinforcement, private experience, hope, or delegated expertise enters the picture.

This is not meant to shame trust. It is meant to label it honestly. A high C and low P profile says, in effect, "I believe this strongly, but I cannot yet personally carry the evidential burden." A high C and high P profile says, "I believe this and can responsibly explain why."

Weight is the claim's effective contribution after combining Confidence and Personal Substantiation: sqrt((C/100) x (P/100)). If Confidence is high but Personal Substantiation is low, the claim receives less weight because the tool is discounting unsupported confidence.

Gap is max(0, Confidence minus Personal Substantiation), so unusually high Personal Substantiation does not cancel overconfidence elsewhere. Tension is how far the claim's effective support outruns its prerequisite bridge claims; unrated prerequisites count as zero current support. A large Gap points to possible overconfidence. A large Tension points to a possible skipped step in the chain of support.

Aggregate Position estimates how much rightward progress your rated profile has earned on the 1-to-5 gradient. It starts at 1, then adds the average effective support in Design Deism, Personal Theism, Interventionist Theism, and Specific Christian Theism. Minimal Deism anchors the baseline rather than moving the score rightward.

Theism Index converts that Aggregate Position to a 0-to-100 progress score. Average Substantiation Gap shows how much Confidence tends to outrun Personal Substantiation across rated claims, using only positive gaps.

Claims Rated is simply progress. The other three metrics should not be treated as grades. They are directional summaries that help you decide which parts of the profile deserve a closer look.

The Claim Scatter shows rated claims by category lane and Confidence. Each dot is a claim. Darker dots have stronger Personal Substantiation, and larger dots carry more effective weight.

Use the scatter to notice shape. Are later Christian claims clustered high while earlier bridge claims are low? Are many claims high in Confidence but faint because Personal Substantiation is low? Click a dot to jump directly to the underlying claim card.

The Category Profile summarizes average Confidence and Personal Substantiation inside each gradient band. It is useful for spotting transitions: for example, a profile may be strong in Minimal Deism and Design Deism but drop sharply at Interventionist Theism.

Clicking a category row filters the claim list to that band. Use this when you want to repair a specific section of the profile rather than scrolling through all 50 claims.

The View filter changes which claims are displayed: all claims, unrated claims, high-gap claims, dependency-tension claims, Christian action claims, highest-confidence claims, or lowest-substantiation claims. The Category filter narrows by gradient band. Search matches claim text, IDs, and tags.

Filters never change your saved scores. They only change what appears in the Claim Explorer so you can work on a narrower subset of the 50 claims.

Diagnostic Alerts identify high-pressure claims. A substantiation alert means Confidence is much higher than Personal Substantiation. A dependency alert means a downstream claim has much higher effective support than its prerequisite bridge claims.

Click an alert to jump to the claim. Then decide whether to lower Confidence, raise Personal Substantiation by adding better support, revise the note, or inspect the prerequisite claims that make the alert appear.

The Dependency Map shows downstream claims alongside the bridge claims they depend on. It is especially important for claims about prayer, healing, prophecy, scripture, Jesus, the Holy Spirit, salvation, and the core Christian picture.

Use it to ask whether a later claim is standing on earlier claims you have not rated, have rated low, or can only weakly substantiate. The map is not saying every dependency must be proven first; it is showing where the logical pressure lives.

Preset Comparison shows how your current profile differs from built-in stance profiles such as Skeptical Naturalist, Minimal Deist, Cautious Christian, Experiential Christian, and Doctrinal Christian. It helps you see whether your actual ratings match the label you might naturally give yourself.

The Apply buttons overwrite the current profile with that preset, so export first if you want to preserve your own ratings. Use presets as comparison tools, not as answers.

Use Brief for a quick summary, Full for the most complete internal review, Skeptical Audit for pressure-testing, Pastoral Reflection for a gentler self-examination, and AI Prompt Only when you mainly want material to paste into another assistant.

The report changes as you update ratings. If you want to preserve a specific stage of your thinking, copy the report or export the profile before revising scores.

Export Profile downloads a JSON file containing your saved responses: Confidence, Personal Substantiation, notes, and profile metadata. Import Profile loads a compatible JSON file and replaces the current local profile with that file's ratings.

This is useful for backups, comparing versions over time, or moving a profile between browsers. It is not a public submission system.

Use the AI prompt after you have rated enough claims for the profile to show real structure. The prompt carries the current score summary, category profile, strongest commitments, gaps, dependency tensions, and repair options into an AI assistant.

The best use is focused: ask the assistant to examine one high-gap claim, one dependency chain, or one category transition rather than asking it to settle the whole worldview at once.

Reset clears the saved local responses for this browser. If you have ratings you may want later, export the profile before resetting.

Reset does not change the app's list of 50 claims or the preset lenses. It only clears your local scores and notes.

This audit helps you inspect the shape of a Christian-theism profile rather than merely declare whether Christianity is true or false. It asks how far your current ratings move from thin claims, such as a source or creator of reality, toward thicker Christian claims about prayer, healing, revelation, Jesus, the Holy Spirit, salvation, and transformation.

The app is especially interested in whether those moves are supported by bridge premises. A bridge premise is the additional claim needed to move responsibly from one level of the gradient to the next. For example, moving from "a creator exists" to "the Christian God answers prayer" requires more than general belief in a creator.

Neither. It is designed to be rigorous enough for skeptical pressure and fair enough for serious Christian self-examination. A Christian can use it to identify beliefs that are well anchored, beliefs that are sincerely held but underdeveloped, and beliefs that depend on other claims that need more attention.

The app does not assume that low personal substantiation means a claim is false. It means only that the claim may not yet be carrying the evidential weight your confidence assigns to it.

Confidence is your current assessment of how credible the claim seems. It can include argument, experience, testimony, community trust, scripture, philosophical reasoning, and your broader worldview. It is the "how likely or believable does this seem to me?" score.

Confidence should not be used as a pure loyalty marker. If a claim feels important to your identity but you cannot yet say why it is well supported, that is exactly the kind of case where Confidence and Personal Substantiation should be separated.

Personal Substantiation asks whether you can personally make a responsible case for the claim. It is not a measure of sincerity, devotion, or spiritual seriousness. It asks whether you can explain the evidence, name the bridge premises, handle the strongest rival explanations, and say what would count against the claim.

A high Personal Substantiation score means you are not merely borrowing confidence from a pastor, book, family, community, or preferred theological system. It means you can personally articulate why the claim deserves the role it plays in your web of beliefs.

That pattern is not automatically irrational. Many people responsibly trust specialists, communities, historical traditions, or lived experience beyond what they can personally defend in detail. The app flags the pattern because it becomes risky when a high-confidence claim is used as a major premise in further reasoning.

If the gap is large, you have three honest options: lower Confidence, raise Personal Substantiation by doing the work, or explicitly mark the claim as held by trust rather than personally demonstrated evidence.

A substantiation gap is the positive distance between Confidence and Personal Substantiation. If you rate Confidence at 80 and Personal Substantiation at 45, the gap is 35 points. If Personal Substantiation is higher than Confidence, the app records the gap as 0 rather than letting it cancel overconfidence elsewhere.

The gap is not a verdict of falsehood. It is a prompt to ask: Why am I more confident than my own present ability to substantiate the claim? Is that because I am relying on good delegated evidence, or because I have not inspected the claim carefully?

Dependency tension appears when a downstream claim has much higher effective support than its prerequisite bridge claims. Effective support is the geometric combination of Confidence and Personal Substantiation, so a prerequisite with high Confidence but low Personal Substantiation will still be treated as weak support.

Unrated prerequisites count as zero current support. That does not mean the prerequisite is false; it means the app has no entered support for it yet. This makes skipped bridge claims visible rather than quietly treating them as if they had already been established.

This does not prove the downstream claim is wrong. It shows where the argument may have skipped steps. The most useful repair is usually not to abandon the downstream claim immediately, but to inspect the weakest bridge claim and decide whether it can be strengthened.

Start by separating meaningful experience from evidential permission. A prayer experience can be personally meaningful even if it does not strongly establish that God intervened. A healing testimony can be sincere even if it lacks medical confirmation. A guidance impression can be spiritually important even if it may also be explained by intuition, reflection, or community influence.

Higher ratings are strongest when the claim includes specificity, independent confirmation, alternative explanations, durable outcomes, and a fair accounting of comparison cases. The key question is: What makes this more than coincidence, selection effect, suggestion, hindsight, or ordinary human judgment?

Private experiences can count as data for the person who had them, but the app asks you to be clear about what kind of data they are. An experience may justify humility, openness, gratitude, or personal trust without licensing a highly specific public conclusion.

When rating such claims, ask whether the experience contains information you could not otherwise have known, whether it produced durable moral fruit, whether it is corroborated, and whether psychological or social explanations have been considered without defensiveness.

Possibility means a claim has not been ruled out. Plausibility means the claim has some fit with what you already believe or observe. Evidential permission means the available evidence is strong enough to responsibly use the claim in further reasoning.

Many theological claims are possible. Fewer are plausible. Fewer still are substantiated well enough to carry heavy argumentative weight. The app is designed to keep those categories from quietly collapsing into each other.

Yes. Testimony is a normal and often legitimate source of belief. The question is what strength it should be given in a particular case. Stronger testimony is specific, early, independent, costly, consistent, and aware of live alternatives. Weaker testimony is vague, socially reinforced, repeated within one community, or insulated from disconfirmation.

If a claim depends heavily on testimony, you do not need to rate it low by default. But you should reflect that dependence in your Personal Substantiation score unless you can explain why the testimony is especially strong.

Preset lenses are not answers. They are comparison profiles. Apply one to see what a stance looks like numerically, then inspect where diagnostics appear. You can use them to ask whether your own profile is closer to skeptical naturalism, minimal deism, cautious Christianity, experiential Christianity, or doctrinal Christianity.

The most useful question is not "Which preset am I?" but "Where does my current profile become less consistent than the profile I think I hold?"

Diagnostic Alerts are triage signals. They point to claims where Confidence may be outrunning Personal Substantiation or where a downstream claim may depend on weaker prerequisites. Click an alert to inspect the relevant claim, then read the annotation and dependency chips before changing the rating.

A good audit response is not always to lower the flagged claim. Sometimes the better move is to strengthen a prerequisite claim, add a note that names delegated trust, or split a broad claim into a more modest version.

One large question like "Is Christianity true?" hides too much structure. A person may have strong reasons for a minimal creator, weaker reasons for divine communication, and much weaker reasons for a specific claim about healing or prophecy. The 50-claim format lets those differences become visible.

The goal is not to fragment belief into trivia. It is to reveal which claims are doing the most work and which transitions in the chain require the most careful support.

Start with the claims that feel central to your own faith or doubt. You do not need to rate all 50 in one sitting. The "Next Unrated" button is useful for a systematic pass, while the filters are better when you want to focus on a particular category such as prayer, revelation, or interventionist theism.

A good first pass can be rough. After that, use Diagnostic Alerts, the Dependency Map, and the final report to decide where a more careful second pass is needed.

Confidence can include your awareness that a strong public case may exist. Personal Substantiation should track what you can personally explain and defend right now. If a specialist or apologist could defend a claim better than you can, note that in your rationale rather than silently giving yourself their substantiation score.

This distinction matters because a borrowed case may be legitimate, but it should be labeled as delegated trust. The app is trying to prevent delegated trust from masquerading as personal mastery.

No. Human knowledge depends heavily on delegated trust. We rely on historians, scientists, translators, doctors, pastors, textual scholars, and communities. Delegated trust becomes a problem only when it is treated as if it removes the need for proportion, caveats, or awareness of disagreement.

If your rating depends on delegated trust, ask whether the source is competent, independent, transparent about uncertainty, responsive to objections, and not merely repeating what your community already prefers to hear.

Scripture can function as authority, testimony, historical source, theological interpretation, or devotional anchor. Those are not the same role. When rating scripture-related claims, ask which role scripture is playing and whether the rating depends on already assuming the Christian conclusion.

For example, "scripture says this" may be enough inside a confessional setting, but the audit asks what makes scripture reliable for the specific claim being rated. That may require arguments about transmission, genre, interpretation, historical context, theological coherence, and rival readings.

Historical miracle claims usually require several kinds of support at once: source quality, proximity, explanatory comparison, prior plausibility of divine action, and a reason to prefer miracle over mistake, legend, symbolic interpretation, fraud, grief experience, or social development.

A high rating is strongest when those components are not blurred together. The audit is especially interested in whether a miracle claim is being supported by historical evidence itself or by prior confidence that God acts miraculously.

A strong bridge premise is independently defensible, relevant to the specific move being made, and not merely a restatement of the desired conclusion. It should explain why evidence from one category licenses movement into the next category.

For example, moving from "the creator can communicate" to "the Christian God revealed truth in scripture" requires more than the bare possibility of communication. It needs a reason to think communication actually occurred in this tradition and that the relevant texts reliably preserve it.

That can happen. Some people arrive at Christian conviction through prayer, worship, transformation, or perceived divine guidance before they have a developed cosmological or design argument. In that case, the dependency map may feel reversed compared with your personal story.

The audit is not denying that experience can be primary in a person's biography. It is asking whether the later claim can carry evidential weight without support from the bridge claims it logically depends on. Personal order of discovery and logical order of support are related, but not identical.

Lower Confidence when the claim seems less well supported after you inspect the alternatives, dependencies, and defeaters. Raise Personal Substantiation when the claim still seems credible and you can now articulate the case more responsibly.

A third option is to revise the claim. Sometimes the best repair is not "keep the same claim at a lower number" but "state a more modest version." For example, "God sometimes heals miraculously" may become "some healing reports deserve further investigation and are not easily dismissed."

They should. The app is not trying to freeze your beliefs. It is a snapshot of your current confidence, substantiation, and dependency structure. Export a profile before major revisions if you want to compare your thinking across time.

Changes are especially informative when they reveal a pattern. If your Confidence falls but Personal Substantiation rises, you may be becoming more modest and more precise. If both rise, you may be strengthening the claim. If Confidence rises while Personal Substantiation stays low, ask what changed besides desire, community pressure, or emotional salience.

Yes, but it should be framed carefully. The most productive group use is not to shame low scores or reward high scores. It is to compare which claims different people find most substantiated, which bridge premises feel weakest, and which kinds of evidence are doing the most work.

Group use works best when people can say "I hold this by trust," "I need to study this," or "I think this claim should be more modest" without being treated as disloyal.

The app does not settle metaphysics, adjudicate every historical dispute, measure spiritual maturity, or decide what God has actually done. It also cannot know whether your private experiences are veridical. It only organizes your own ratings and shows tensions among them.

Its strength is structural clarity. Its weakness is that it depends on honest self-rating. If you inflate Confidence and Personal Substantiation to protect a preferred conclusion, the app cannot stop you. It can only make the pattern easier to notice.

A good outcome is not necessarily a lower or higher Theism Index. A good outcome is a more honest map of what you believe, why you believe it, which claims are central, which are speculative, and which need stronger bridge premises.

You may end with a more modest Christian profile, a more carefully defended Christian profile, a deistic profile, or an agnostic profile with specific open questions. The common good is that your web of beliefs becomes less blurry and less protected from inspection.

Aggregate Position is a progress estimate, not a simple average of category numbers. The app first calculates effective support for each rated claim as 100 x sqrt((C/100) x (P/100)), then averages effective support within each gradient band.

The position starts at 1. The Design Deism, Personal Theism, Interventionist Theism, and Specific Christian Theism band averages each add up to one point of rightward movement. If every rightward band had full effective support, the position would be 5.00. If only Minimal Deism were supported, the position would remain near 1.00.

Theism Index is the same result converted to a 0-to-100 scale: ((Aggregate Position - 1) / 4) x 100. Because the app discounts claims when Personal Substantiation lags, a high-confidence but weakly substantiated Christian claim will not move the index as much as a personally defensible one.

The final report is for review, comparison, and reflection. Use the brief report when you want a quick profile, the skeptical audit when you want pressure-testing, the pastoral reflection when you want a gentler self-examination, and the full report when you want the most complete export.

The AI prompt is designed to continue the audit in a more conversational setting. Paste it into an AI assistant and ask it to focus on one or two tensions at a time. The best follow-up work usually happens claim by claim, not by trying to settle the entire Christian worldview in one pass.

Yes, if used as an honesty tool rather than a weapon. The goal is not to strip away every belief that cannot be proven on demand. The goal is to distinguish mature confidence from inherited confidence, personal conviction from public evidence, and central commitments from claims that need more careful support.

A healthier profile may become more modest in some places and stronger in others. That is not failure. It is the normal work of bringing belief, evidence, testimony, experience, and humility into better alignment.

Your ratings are stored locally in this browser's storage unless you export them or paste the report into another tool. Exported JSON files and copied reports are under your control, so treat them as personal reflection data.

If you reset the app, export first if you want a backup. Imported profiles replace the current saved ratings with the contents of the selected JSON file.