بِسْمِ ٱللَّٰهِ ٱلرَّحْمَٰنِ ٱلرَّحِيمِ
The Tahqiq Methodology.
Theo's reasoning is built on tahqiq — the classical Islamic methodology of rigorous verification. Every answer passes through five layers and twelve cognitive operations drawn from usul al-fiqh, the foundational science of Islamic jurisprudential reasoning.
See it in actionQuran first. Always.
Quran
The revealed word of Allah — the highest and most authoritative source. 6,236 ayat form the unassailable foundation of every analysis Theo produces.
Authenticated Hadith
Prophetic narrations verified through narrator chain analysis and computational grading. 449,415 narrations across 61 collections, each with attributed scholarly grading.
Reasoned Analysis
Scholarly reasoning methods — qiyas (analogy), ijma (consensus), and istihsan (juristic preference) — applied only when primary texts require interpretation.
Five layers. Every question.
Retrieve
Semantic and lexical search across all four IPSC indexes — Quran (6,236 ayat), Hadith (449,415 narrations), Rijal (197,868 narrator assessments), and Ilal (14,917 defect analyses). The retrieval layer identifies every relevant primary source before reasoning begins.
Investigate
14 corpus tools deployed across 5 iterative rounds: narrator credibility lookup, chain completeness check, grade verification, cross-collection correlation, rijal consensus analysis, ilal defect search, temporal sequencing, thematic clustering, verse-hadith linking, madhab position mapping, abrogation detection, scholarly attribution, chain topology mapping, and evidence strength scoring.
Generate
12 cognitive operations from classical tahqiq are applied systematically: the question is decomposed, premises identified, implications traced, contradictions detected, evidence weighed, scope determined, temporal context established, absence patterns recognized, analogies mapped, confidence calibrated, potential errors flagged, and reasoning strategy selected.
Verify
Chain-of-Verification (CoVe) checks every citation against the corpus. Each Quran reference is validated against the Tanzil.net standard. Each hadith citation is verified against the original collection text. Narrator assessments are cross-checked across rijal works. No hallucinated sources survive this layer.
Personalize
The response adapts to your preferred madhab (Hanafi, Maliki, Shafi'i, or Hanbali), knowledge level (student, intermediate, or advanced), and contextual needs. Comparative evidence is always available, but the primary presentation follows your preferences.
Twelve operations. Every answer.
Decomposition
Breaks complex questions into atomic sub-questions that can each be answered from primary sources. Ensures no aspect of the question is overlooked or conflated with another.
Premise Identification
Identifies the explicit and implicit assumptions embedded in a question. Surfaces hidden premises that could lead to reasoning errors if left unexamined.
Implication Tracing
Follows the logical consequences of a ruling or interpretation across related domains of fiqh. Catches unintended downstream effects before presenting a conclusion.
Contradiction Detection
Identifies apparent conflicts between sources — Quran verses, hadith narrations, or scholarly positions — and applies classical reconciliation methods (al-jam' wal-tawfiq).
Evidence Weighing
Ranks evidence by source hierarchy, chain strength, and scholarly consensus. Sahih narrations from mutawatir chains outweigh da'if narrations from singular chains.
Scope Determination
Determines whether a ruling is universal ('amm) or specific (khass), absolute (mutlaq) or conditional (muqayyad). Prevents overgeneralization from specific textual evidence.
Temporal Sequencing
Establishes the chronological order of revelations and rulings to identify abrogation (naskh). Later rulings may supersede earlier ones on the same matter.
Absence Reasoning
Recognizes when the silence of sources on a matter is itself significant evidence. Distinguishes between genuine scholarly gaps and intentional non-ruling.
Analogical Mapping
Applies qiyas — drawing parallels between a new question and established rulings when they share the same effective cause ('illah). The most structured form of Islamic legal reasoning.
Confidence Calibration
Assigns honest confidence levels based on evidence strength, scholarly agreement, and chain reliability. High confidence requires multiple sahih sources with scholarly consensus.
Error Recognition
Actively identifies potential errors in its own reasoning chain — logical fallacies, misapplied evidence, over-extrapolation — and corrects before presenting the final answer.
Strategy Selection
Chooses the optimal reasoning approach for each sub-question: direct textual evidence, analogical reasoning, scholarly consensus analysis, or principled analysis from maqasid al-shariah.
We do not issue fatwas.
Theo produces Evidence-Based Opinions (EBOs) — a fundamentally different category from fatwas. Understanding this distinction is essential to using Theo responsibly.
Evidence-Based Opinion
- Evidence presented openly
- Reasoning visible at every step
- Confidence calibrated with levels
- Every source cited and verified
- Human decides what to follow
Fatwa
- Conclusion issued directly
- Reasoning often summarized or hidden
- Authority-based certainty
- Sources may not be cited
- Human expected to follow
Both serve important roles. Fatwas come from qualified muftis with ijazah and deep training. EBOs provide transparent evidence for personal study and informed consultation with scholars. Theo empowers your understanding — it does not replace scholarly authority.
بِسْمِ ٱللَّٰهِ ٱلرَّحْمَٰنِ ٱلرَّحِيمِ
Ask Theo. See the reasoning.
I'm here. Ask me anything about Islam. Every source cited. Every chain verified.
Ask about any hadith, ruling, or scholar…449,415 graded hadith · 27,115 verified narrators · 95.2% Bukhari convergence