With over 50 million characters in over 1500 texts, the Chinese Buddhist Canon is a complex literary collection. Besides the Buddha himself, there is a myriad of characters including bodhisattvas, deities, disciples of Buddha, monks, lay Buddhists as well as kings. This paper analyzes the hierarchy among these characters by examining their verbal interactions. Exploiting techniques from natural language processing, we extract all direct speech from the text, and examine the relation between the speakers, listeners, and the quotative verbs used for reporting the speech. We show that a number of the quotative verbs indicate relative status between the speaker and the listener. We then use their usage patterns to induce a hierarchy of the characters in the Canon.
展开▼
