The plan · six phases · built in the open

From one trial to a verdict

The instrument is built and sealed; the first results are in. What follows is the honest path from a single coherence trial to a full comparative verdict against rival corpora — and the design that lets it run with no funding and no coders.

Build items8 / 8 shipped
Phase0–1 in progress
Tests115 green
Queue6 / 20,448 adjudicated

Where we are now

The instrument is finished. The trial has begun.

Eight build items took the design from a charter to a working, self-auditing instrument: it seals its own rules, judges contradiction-pairs, aggregates a residual ledger, checks itself for bias, and can already turn the same blade on a rival. The ninth task — actually working the 20,448-pair queue — is perpetual, and barely started.

[×]
Master preregistration sealed
blake2b hash 0xdd78…1874c345, run mizan-t1-coherence-v1 — sealed before any verdict.
[×]
Unit-test suite — 115 green
charter-hash stability, prereg tamper-detection, sweep round-trip, null computability, adversary JSON extraction.
[×]
Residual-ledger aggregator
verdicts → residual_ledger.json + the floor / bias check.
[×]
Report generator
queue progress, verdict distribution, floor status, battery → REPORT.md.
[×]
Compositional Distribution (T8)
F1–F5 taxonomy fixed in advance; folklore-load 30.8% vs claimed ~70%.
[×]
Rival-corpus loader contract (Layer R)
RivalCorpus base + Sefaria / NT skeletons + a synthetic stub + parity test.
[×]
Evidence-Verification Engine scaffold (Layer E)
FEVER-style claim→retrieve→classify interface + DATASETS.md license gate.
[×]
Comparative hook
runs the identical trial on any corpus; writes comparative_parity.json.
[~]
Adjudicate the queue — perpetual
6 of 20,448 contradiction-pairs judged. The loop pulls ≤6 pairs an iteration, adjudicates with the Coherence-Trial protocol, and records each verdict bound to the prereg hash — until a human stops it. Currently paused between unattended runs.

The road ahead

Six phases

Each phase is a self-contained deliverable with its own honest lose-condition. Green nodes are shipped; amber are in progress; the rest are planned, several already scaffolded.

Phase 0

Charter, preregistration spine & corpus acquisition

in progress

The neutral charter, the four verdicts, the symmetric decision rule, the Negative-Verdict Floor, and the nine-test battery — all written and hashed. The on-chain preregistration mechanism is designed and the founding documents are sealed locally. What remains: downloading the rival scriptures and the evidence databases (a license-gated, human step).

✓ charter sealed✓ prereg mechanism✓ EVE + rival-loader scaffolds○ rival/​evidence data○ anchor on AMANAH
Phase 1

The Coherence Trial v1 — AI-only proof of concept

in progress

Prosecutor hunts the tension, defender answers, neutral records — every step logged and bound to the sealed prereg, with the scrambled-text null as the floor. The instrument runs end-to-end; six pairs are judged and the bias floor passes. What remains: working the full queue and subtracting the empirical null at scale.

✓ instrument end-to-end✓ first verdict batch✓ comparative hook○ full 20,448-pair queue○ empirical null at scale
Phase 2

Compositional Distribution + Mechanism-Portability scaffolding

in progress

Answers Objections A and B from data already in hand. The folklore-load is measured (30.8%); the mechanism-portability frame — surface parallel vs genuinely cross-cultural mechanism, net of null — is scaffolded next on the Module-6 corpus.

✓ folklore-load measured (T8)○ mechanism frame (T7)
Phase 3

Full comparative confirmatory pass

planned

The whole battery, run identically across every rival corpus. Only here does “distinctive vs rivals, net of null” become sayable. Gated on the Phase-0 corpus acquisition.

○ Layer-R corpora loaded○ identical-pipeline diff
Phase 4

HARF crowd integration

planned

The Qur'an-internal subjective calls — “is this really a contradiction?”, “did this prophecy resolve?” — go to on-chain commit-reveal rounds. Dawid-Skene confusion matrices measure each annotator's bias instead of assuming it; honeypots down-weight rubber-stampers; a believer crowd conceding against its own preference becomes the highest-weight signal.

○ claim-verdict rounds○ bias-corrected consensus○ Scrutiny mode app
Phase 5

Scientific Foreknowledge + Anthropological Validity

planned

Tests 5 and 6, wired to the Evidence-Verification Engine against the Layer-E datasets — so external claims are verified against a retrieval record, not an LLM's memory. The engine scaffold already stands; it needs the datasets and the Red-Team Queue.

✓ EVE scaffold○ Layer-E datasets○ Red-Team Queue
Phase 6

Residual Ledger + public research module

planned

The capstone: after each ordinary explanation is spent, how much remains? The symmetric decision rule is applied, the combined-safeguard audit run, and the full provenance published. The aggregator already exists and this site is its first seed.

✓ ledger aggregator○ public verdict dashboard

The nine tests · live status

The battery

TierTestQuestionStatus
Tier 1The Coherence Trial
4:82
Does the corpus contain genuine internal contradictions when assaulted adversarially — at a density above/below matched rivals, net of the empirical null?active
Tier 1The Prediction Ledger
30:2-4
Of the Book's dated/falsifiable forecasts, how many resolved — with a complete hit-AND-miss accounting (no curated wins)?planned
Tier 1The Preservation Record
15:9
Empirically (not theologically), how stable is the transmitted text vs rival textual histories?planned
Tier 1The Inimitability Challenge
2:23
(Aesthetic literary judgement.)excluded
Tier 2Scientific Foreknowledge Audit
41:53
For each physical-world claim: prediction-lock (4 conditions) + Red-Team Queue (5 tests) + miss-accounting. PREDICTED is hard-won; default is CONSISTENT_WITH.planned
Tier 2Anthropological Validity Assessment
external
Does a contested norm survive the 4 AVA tests — distinguishing 'society abandoned it' from 'disproven' from 'self-defeating'? Bidirectional: can return genuinely ANACHRONISTIC.planned
Tier 3Mechanism Portability
external
Is a narrative's underlying mechanism cross-culturally recurrent (MPM, independence-verified, net of null) rather than a mere surface parallel (SPM)?planned
Tier 3Compositional Distribution
external
What is the measured function-tag distribution (folklore-load F1-F5), Qur'an vs rivals?measured
Tier 4The Residual Ledger
external
After each ordinary explanation is spent, what fraction of the corpus remains unexplained? The honest endpoint.planned

How it runs with no funding and no coders

The governance triangle

The blocker was real: credible adjudication seemed to need paid blind coders, an adversarial academic, and anti-bias preregistration. Four assets already in hand replace all three — and turn believer-bias from a fatal flaw into a measured quantity.

AI as tireless adversary

Two preregistered roles — prosecutor and defender — run on every claim, with a neutral adjudicator, every step logged. Adversarial at corpus scale, with no stake in the afterlife.

🗄

External evidence, not opinion

Moral, scientific and cross-cultural calls are answered against large pre-collected human-annotation datasets (FEVER, SciFact, Social-Chem-101, D-PLACE) via a verification engine — not LLM recall.

👥

The crowd, with bias measured

HARF's commit-reveal + Dawid-Skene turns believer-bias into a measured, subtractable quantity; honeypots catch rubber-stampers; concessions-against-interest carry the most weight.

🔗

The chain as preregistration

Charter, prompts and plans are hashed and anchored on AMANAH before results exist. Silent re-runs and retro-fitting become impossible by construction — by math, not trust.

The data — honest acquisition status

Two layers, mostly still to gather

Comparison needs rival scriptures (Layer R); external assault needs evidence databases (Layer E). The loaders and the license gate are written; most of the data is a deliberate, license-aware human download — not yet done, and not pretended otherwise.

Already on disk

Reused, not rebuilt — these carry Phase 1–2.

Module 6 — comprehensive verse analysis · 6,236 verses · 14 frameworks · already used by T8on disk
QURAN-NLP + quranic-corpus morphology · local · text / lemmas / morphologyon disk

Layer R — rival corpora (the comparison)

Each loaded under the identical loader and put through the identical assault.

Hebrew Bible · Sefaria API / exportsto acquire
New Testament · Sefaria / SBLGNTto acquire
Homer — Iliad & Odyssey · Perseusto acquire
Analects of Confucius · ctextto acquire
Dhammapada · public domainto acquire
Stoics — Epictetus, Aurelius · Perseusto acquire
Pre-Islamic Mu'allaqāt · public domainto acquire

Layer E — evidence databases (the assault weapons)

Pre-collected human judgment at scale — the breakthrough for the no-coders constraint.

FEVER — 185k claims vs Wikipedia · permissive · verification engineto acquire
SciFact / SciFact-Open · permissive · scientific claimsto acquire
Social-Chemistry-101 — 4.5M moral judgments · permissive · anthropological validityto acquire
D-PLACE — 1,291 societies · open · cross-culturalto acquire
Cliopatria — polities 3400 BCE–2024 CE · open · historicalto acquire
World Values Survey (WVS) · redistribution-limited — license auditto acquire

Fixed before the data — the verdict the whole thing can reach

What would make the objection win

Objection largely holdsNet of null and against benchmarks: the Qur'an's profile is statistically INSIDE the rival-corpus range (not distinctive) on the tested dimensions; negatives sit at/above the floor; and the four ordinary explanations jointly account for the large majority of the corpus (small residual).
Objection largely failsNet of null and against benchmarks: the Qur'an's profile sits OUTSIDE the rival range on multiple INDEPENDENT dimensions; this survives the empirical null, the combined-safeguard audit, and adversarial review; and the residual is substantial and not attributable to the ordinary explanations.
IndeterminateResults are mixed or underdetermined. Reported as the finding, not spun. The most likely honest outcome given underdetermination.
← See the findings so far