Beyond Structure to Communicative Function: The study of Embedded Storytelling in Kathasaritasagara

Embedded narratives, tales told by characters within larger frames, define the structure of Somadeva's Kathasaritasagara. Computational models of embedding, however, record who tells a story to whom, but not what the telling is intended to accomplish. This paper applies Gervas's (2021) stack-based model of embedded narrative, which formalizes Genette's (1980) narrative levels, to selected tales from the Kathamukha (“The courtesan who fell in love”) and Lavanka (“Guhachandra and His Celestial Bride and Indra and Ahalya”) Lambakas of the Kathasaritasagara. Although the model captures the text's structure, it misses the purpose of each story. Drawing on Austin's (1962) concept of illocutionary force and Searle's (1969) taxonomy of illocutionary acts, the study treats each embedded tale as a communicative act. It extends Gervás’s predicate: tell_story(narrator, narratee, story_id, purpose). The analysis shows that embedded stories with different structural designs can perform the same act: telling a tale with a purpose. The study thus offers a conceptual purpose-sensitive extension to computational narratology and demonstrates the value of Kathasaritasagara as a corpus for modeling nested narratives. Future research may evaluate the framework across larger corpora and examine its relationship with narrator reliability and reader response.