DH 2026

Daejeon, July 27–31

Poster

Encoding Pedagogy: Interpreting the Upanishads through Small-Scale Semantic Annotation

Yashee Singh
Free University of Berlin, Germany · yashee.singh@fu-berlin.de
Frank Fischer
Free University of Berlin, Germany · fr.fischer@fu-berlin.de

Introduction

This poster presents the initial conceptual and encoding phase of a doctoral project that utilizes small-scale semantic annotation to highlight the pedagogical structures embedded within the Upanishads. Since ancient Indian philosophical texts were composed primarily as extended teacher–student dialogues, the Upanishads function as repositories of pedagogical method, ethical reasoning, and reflective inquiry. Their dialogic structures constitute, as Black argues, the primary vehicle through which philosophical knowledge is transmitted, characterized and socially enacted (Black, 2007). The project reframes these texts through a Digital Humanities (DH) lens, investigating how interpretive annotation can model these pre-colonial pedagogical features at a small-data scale. The specific corpus for this pilot phase includes key dialogues from the Kena and Katha Upanishads (Olivelle,1998), which offer examples of didactic exchange.

Background and State of the Field

Existing digital repositories for Sanskrit texts like GRETIL, SARIT, and the DHARMA project have developed robust TEI infrastructure for textual preservation, critical apparatus, and transliteration. However, no existing project applied TEI’s analytic mechanism to encode pedagogical or instructional moves within Sanskrit dialogues. The only computational study of the Upanishads (Chandra and Ranjan, 2022) performs topic modelling on English translations. The foundational premise of this research is that a definable pattern of education and literacy exists within the Upanishadic structure (Surpi, 2021). This project fills a critical gap by moving beyond preservation toward semantic exploration of the embedded pedagogical relationships, grounding the work in the discourse of effective communication through ancient wisdom (Tiwari, 2023).

Methodological Approach

The pilot corpus comprises two texts: the Kena (34 verses), and Katha (119 verses), Upanishads selected because together they represent the full range of Upanishadic pedagogical modes: direct disciple questioning in the Kena, and extended narrative teacher-testing in the Katha (Naciketas–Yama dialogue). This small-scale approach privileging fine-grained semantic annotation over corpus breadth within a bounded pilot follows the methodological tradition established by Pierazzo (2011), who argues that deliberate selectivity in encoding constitutes a scholarly argument in itself, and instantiated in TEI practice by Truan and Romary (2021), who demonstrate that a restricted corpus enables the depth of annotation that large-scale projects structurally cannot sustain.

The methodology involves following components:

  • Close Reading: Each passage is first examined for its instructional function (what pedagogical work the passage performs).
  • XML/TEI Encoding: Sample passages are encoded using <sp> and <speaker> for teacher-student turns, <seg> for semantic units, and <interp>/<interpGrp> elements. The pedagogical moves derived from close reading (analogical explanation, ethical guidance, reflective intervention) would be semantically encoded in the TEI pipeline. This pipeline follows the Women Writers Project's guidance on encoding philosophical dialogue in non-dramatic prose (WWP Encoding Guide) and the precedent of Platonic dialogue encoding in the Perseus Digital Library. The project highlights how annotation itself constitutes a critical act where each encoding decision articulates an interpretive stance on how pedagogical meaning is structured. 

Early Outputs Presented in the Poster

The poster will display:

  • A preliminary representation of pedagogical categories derived from close reading.
  • Encoded excerpts from selected Upanishadic dialogues that illustrate how TEI can model teacher–student relationships and instructional strategies.
  • Visualization of annotation layers showing how dialogic and pedagogical structures manifest within the text’s organization.
  • Reflections on methodological choices, including the limits, affordances, and ethical implications of digitally representing indigenous pedagogical traditions.

Conclusion

By applying DH methods to a non-Western pedagogical tradition systematically absent from existing digital Sanskrit infrastructure, this project enacts what Risam (2018) theorizes as expanding the cultural record of digital humanities not by rejecting established encoding frameworks, but by directing them toward indigenous intellectual traditions that those frameworks have not previously been used to model. In centering pre-colonial Indian pedagogical traditions as both object and methodological focus, the project broadens the cultural and intellectual diversity represented within DH scholarship.

This pilot forms the methodological groundwork for the subsequent dissertation stages, which include:

  • Refining the instructional ontology.
  • Expanding the corpus to other classical texts.
  • Developing an interactive digital archive that allows users to search, compare, and visualize teaching techniques across the texts.

This poster presents the initial conceptual, methodological, and encoding work for this multi-year study, contributing both to DH methodology (small-scale annotation) and to the broader visibility of non-Western intellectual traditions.

References
  1. Birgit Kellner et al. “SARIT: Search and Retrieval of Indic Texts.” https://sarit.indology.info/ 
  2. Black, Brian (2007): The Character of the Self in Ancient India: Priests, Kings, and Women in the Early Upanishads. Statue University of New York Press.
  3. Chandra, Rohitash, and Mukul Ranjan (2022): “Artificial intelligence for topic modelling in Hindu philosophy: Mapping themes between the Upanishads and the Bhagavad Gita.” in: PLoS ONE, vol. 17, no. 9, 2022, https://doi.org/10.1371/journal.pone.0273476.
  4. “DHARMA.” Hypotheses, https://dharma.hypotheses.org/.
  5. “Dramatic Speakers.” Women Writers Project, Northeastern University, https://wwp.northeastern.edu/research/publications/guide/html/drama_speakers.html.
  6. KadekSurpi, Ni (2021): “Hindu Literacy: The Basic Concept of Education and the Literacy Pattern in Upanisads.” in: KnE Social Sciences, vol. 5, no. 7, pp. 86–96. https://doi.org/10.18502/kss.v5i7.9323.
  7. Lorenz, Ekkehard, and Oliver Hellwig (2020): GRETIL – Göttingen Register of Electronic Texts in Indian Languages. https://gretil.sub.uni-goettingen.de/gretil.html
  8. Olivelle, Patrick (2008): Upanisads. OUP Oxford.
  9. Pierazzo, Elena (2011): “A Rationale of Digital Documentary Editions.” in: Literary and Linguistic Computing, vol. 26, no. 4, 2011, pp. 436–47. https://doi.org/10.1093/llc/fqr033.
  10. Risam, Roopika (2019): New Digital Worlds: Postcolonial Digital Humanities in Theory, Praxis, and Pedagogy. Northwestern University Press.
  11. Tiwari, Devendra Nath (2023): “Connecting the Dots: Effective Communication through the Upanishads Ancient Wisdom.” in: The Asian Thinker, vol. 4, no. 20, 2023, pp. 107–112. https://theasianthinker.com/wp-content/uploads/2024/08/12.-Connecting-the-Dots-Effective-Communication.pdf.
  12. Truan, Naomi, and Laurent Romary (2021): “Building, Encoding, and Annotating a Corpus of Parliamentary Debates in TEI XML: A Cross-Linguistic Account.” in: Journal of the Text Encoding Initiative, no. 14, 2021. https://doi.org/10.4000/jtei.4164.