Darshana Graph: A Parallel Commentary Corpus for Comparative Indian Philosophy, with Stylometric and Exploratory Graph Analyses
2026-06-16 • Computation and Language
Computation and LanguageDigital Libraries
AI summaryⓘ
The authors created the Darshana Graph, a large collection of over 125,000 texts from Hindu, Buddhist, and Jain philosophy, including many key scriptures. They focused on about 8,500 passages where different commentators from various schools explain the same verses, allowing for direct comparison of interpretations. They used simple measures to study writing styles and built a tool with a language model to map relationships between philosophical ideas, revealing disagreements and some challenges in automated analysis. The authors have made all their data and tools publicly available for further research.
DarshanaVedantaPali CanonsutraBrahma SutrasUpanishadsstylometrylarge language modelphilosophical commentarytext alignment
Authors
Joy Bose
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.