Embedded Narratives and the Structure of Plans
The plans of characters in narrative, just as those of people in real life, are particularly rich in what I call "embedded narratives." Within plans, embedding is produced by transfers of control, i.e. by projected episodes in which the planner depends on the participation of sub-agents to achieve his own goal. Three types of constructs are shown to be necessary to the understanding of plans: (a) the actual plan of the main agent, i.e. the events he really wants to make happen; (b) the projected plans for the subagents, i.e. the plans by which these sub-agents, in the projection of the main agent, will be seeking to fulfill their own goal; and (c) the virtual plan of the main agent, i.e. what he presents to the sub-agents as being his intent, in order to secure their participation. Depending on the relations among these constructs, the attitude of the main agent towards subagents is sincere, deceptive, or doubly deceptive.The last section of the paper describes an attempt to simulate by computer the generation of plans with actual, projected and virtual components.
Return

The Heuristics of Automatic Story Generation
The intelligence of a story-generating computer program can be assessed in terms of creativity, aesthetic awareness, and understanding. The following approaches are evaluated with respect to these three criteria: simple transition networks, grammar-driven models, simulations, algorithms based on problem-solving techniques, and algorithms driven by so-called "authorial goals." The most serious deficiency of the discussed programs resides in the domain of aesthetic awareness. In order to improve on this situation, story-generation should not follow a strictly linear, chronological order, but rather proceed from the middle outwards, starting with the episodes that bear the focus of interest. The program should select as top-evel goal the creation of climactic situations, create the preparatory events through backward logic, and take the story to the next highlight, or to an appropriate conclusion through a guided simulation. This strategy is ilustrated in a "reverse-engineering," or generative reading of Little Red Riding Hood that simulates the reasoning of an imaginary computer program.
Return