Ardalan SadeghiKivi


Using Referring Expression Generation to Model Literary Style
Nick Montfort | Ardalan SadeghiKivi | Joanne Yuan | Alan Y. Zhu
Proceedings of the Workshop on Natural Language Processing for Digital Humanities

Novels and short stories are not just remarkable because of what events they represent. The narrative style they employ is significant. To understand the specific contributions of different aspects of this style, it is possible to create limited symbolic models of narrating that hold almost all of the narrative discourse constant while varying a single aspect. In this paper we use a new implementation of a system for narrative discourse generation, Curveship, to change how existents at the story level are named. This by itself allows for the telling of the same underlying story in ways that evoke, for instance, a fabular or parable-like mode, the style of narrator Patrick Bateman in Brett Easton Ellis’s American Psycho, and the unusual dialect of Anthony Burgess’s A Clockwork Orange.