A comprehensive corpus of syllogistic reasoning following the classical Indian Nyāya tradition, designed for AI training and logical reasoning research.
This repository contains a carefully curated corpus of 339 Nyāya syllogisms covering diverse domains including Philosophy of Religion, Sanskrit Grammar, and general logical reasoning. Each entry follows the traditional five-part structure (Pañcāvayava) of Nyāya logic:
- Pratijñā (Proposition): The statement to be proven
- Hetu (Reason): The logical basis for the proposition
- Udāharaṇa (Example): Illustrative case supporting the reason
- Upanaya (Application): Connection between the example and current case
- Nigamana (Conclusion): Final restatement confirming the proposition
nyaya/
├── nyaya_corpus_clean.jsonl # Main corpus (339 entries)
├── corpus_analysis.ipynb # Comprehensive analysis notebook
├── sanskrit_staging_pipeline.py # Validation pipeline for new entries
├── HANDOFF_PROMPT.md # Development context and guidelines
├── corpus_statistics.json # Detailed corpus metrics
├── Datasets/ # Training and evaluation splits
└── staging_round_*/ # Historical development phases
- Total Entries: 339 syllogisms
- Cultural Distribution: 73% Non-Western, 27% Western traditions
- Primary Domains: Philosophy (45%), Logic (32%), Sanskrit Grammar (12%)
- Authority Sources: Pāṇini, Aristotle, Kant, Aquinas, and contemporary scholars
- Quality Assurance: 100% validation through multi-round staging pipeline
- Sanskrit Grammar: 39 entries covering morphology, syntax, and phonology
- Philosophy of Religion: 33 entries on theological reasoning
- Classical Logic: Traditional syllogistic forms and contemporary applications
- Rigorous validation pipeline with cultural sensitivity checks
- Proper authority attribution and scholarly grounding
- Balanced representation across logical domains and cultural traditions
- AI training for logical reasoning
- Cross-cultural logic comparison studies
- Educational resources for Nyāya logic
- Computational argumentation research
import json
# Load the main corpus
corpus = []
with open('nyaya_corpus_clean.jsonl', 'r', encoding='utf-8') as f:
for line in f:
corpus.append(json.loads(line))
print(f"Loaded {len(corpus)} Nyāya syllogisms")# Example: Examine a Sanskrit grammar entry
sanskrit_entries = [entry for entry in corpus
if 'Pāṇini' in entry.get('grounding_authority', '')]
for entry in sanskrit_entries[:3]:
print(f"Proposition: {entry['pratijna']}")
print(f"Authority: {entry['grounding_authority']}")
print("---")- Established core Nyāya structure
- Balanced cultural representation
- Quality validation framework
- Theological reasoning patterns
- Cross-cultural religious logic
- Contemporary philosophical applications
- Pāṇinian grammatical analysis
- Morphological and syntactic reasoning
- Traditional Indian linguistic logic
- Quantum logic applications
- Contemporary ethical reasoning
- Interdisciplinary domain expansion
The repository includes sanskrit_staging_pipeline.py for processing new entries:
python sanskrit_staging_pipeline.pyValidation Criteria:
- Schema compliance (5-part Nyāya structure)
- Authority grounding verification
- Cultural classification accuracy
- Logical coherence assessment
- Format entries according to Nyāya schema:
{
"id": "unique_identifier",
"pratijna": "Statement to be proven",
"hetu": "Logical reason",
"udaharana": "Supporting example",
"upanaya": "Application to current case",
"nigamana": "Concluding statement",
"grounding_authority": "Scholarly source",
"cultural_tradition": "Western/Non-Western",
"stage": "production",
"batch_id": "batch_identifier"
}- Add to
nyaya_corpus_staging.jsonl - Run validation pipeline
- Review and integrate approved entries
- Cultural Sensitivity: Respectful representation of all traditions
- Scholarly Rigor: Proper attribution and authority grounding
- Logical Validity: Adherence to Nyāya syllogistic structure
- Quality Assurance: Multi-round validation and peer review
Nyāya (न्याय) is one of the six orthodox schools of Hindu philosophy, primarily concerned with logic, epistemology, and methodology. The five-part syllogism (Pañcāvayava) represents a complete logical argument structure that includes:
- Hypothesis presentation (Pratijñā)
- Causal reasoning (Hetu)
- Empirical support (Udāharaṇa)
- Analogical application (Upanaya)
- Logical conclusion (Nigamana)
This structure provides a more comprehensive logical framework than the three-part Aristotelian syllogism, making it valuable for AI reasoning systems.
- Computational Argumentation: Training AI systems in structured reasoning
- Cross-Cultural Logic: Comparing reasoning patterns across traditions
- Educational Technology: Teaching logical reasoning through diverse examples
- Philosophy of AI: Exploring non-Western approaches to machine reasoning
The repository includes corpus_analysis.ipynb with comprehensive analytics:
- Statistical distribution analysis
- Cultural representation metrics
- Authority source mapping
- Domain clustering and patterns
- Quality validation reports
This corpus builds upon centuries of Nyāya logical tradition while incorporating contemporary scholarly standards. Special recognition to:
- Classical Nyāya philosophers (Gautama, Vātsyāyana, Udayana)
- Modern Nyāya scholars and translators
- Cross-cultural logic researchers
- Open source community contributors
This work is released under an open research license. See individual source attributions for specific scholarly works referenced in the corpus.
For research collaboration, corpus extensions, or technical questions, please open an issue in this repository.
Building bridges between ancient wisdom and modern AI through structured logical reasoning.