Darshana Graph: A Parallel Commentary Corpus for Comparative Indian Philosophy, with Stylometric and Exploratory Graph Analyses

2026-06-16

AI Summary

This study addresses the lack of large-scale, structurally aligned cross-school annotated corpora in Indian philosophy, which has hindered systematic comparison of how different darśanas interpret the same canonical texts. The authors construct a parallel annotation corpus comprising 125,000 records, including 8,500 aligned interpretations of identical sūtras by eighteen commentators across philosophical schools, marking the first public release of such cross-darśana aligned data. Their methodology integrates transparent stylometric metrics—such as citation density, refutation rate, and syntactic complexity—with a constrained large language model pipeline guided by predefined relational lexicons, supplemented by deterministic validation and embedding-based contrastive analysis. Findings reveal a moderate negative correlation between citation density and refutation rate, significantly heightened refutation among doctrinally affiliated schools, and notable genre-based variation within the Pāli Canon. The resulting relational graph effectively captures inter-school interpretive divergences while also highlighting current limitations in relation extraction.

Abstract

We introduce Darshana Graph, a corpus of over 125,000 text records spanning classical Hindu, Buddhist, and Jain philosophical traditions, drawn from public-domain and openly licensed translations of sources including the Bhagavad Gita, Brahma Sutras, principal Upanishads, the Pali Canon, and core Jain texts. Its distinctive contribution lies in a structurally unique subset of roughly 8,500 Hindu and Jain records in which the same root verse or sutra is aligned across eighteen historical commentators representing five schools of Vedanta and other darshanas, enabling direct comparison of how independent interpretive traditions read identical source material. To our knowledge, no publicly available resource provides comparable cross-commentator alignment at this scale. We present two analyses built on this corpus. First, a transparent stylometric comparison requiring no machine learning measures argumentative style through scriptural citation density, explicit refutation rate, and sentence complexity. It finds a moderate negative correlation between citation density and refutation rate, a marked increase in refutation rate across three commentators in a related doctrinal lineage, and measurable genre-level differences within the Pali Canon itself. Second, we describe a constrained large language model pipeline that extracts typed philosophical relationships between concepts using a predefined relation vocabulary and deterministic post-hoc validation. The resulting graph surfaces cross-school disagreement patterns while also revealing important extraction limitations, including cases where an independent embedding-based analysis disagrees with the graph-derived findings. We release the full corpus, extracted relationship graph, and all source code.

Problem

• parallel commentary corpus
• comparative Indian philosophy
• cross-commentator alignment
• philosophical interpretation
• stylometric analysis

Innovation

• parallel commentary corpus
• stylometric analysis
• constrained LLM pipeline
• philosophical knowledge graph
• cross-school comparison