👤 About the Author
Vladimir Propp
Leningrad State University
Vladimir Yakovlevich Propp (1895–1970) was a prominent Soviet folklorist and scholar who pioneered structural narrative analysis and formalist literary theory. For most of his academic career, he was affiliated with Leningrad State University, where he taught Russian folklore, linguistics, and philology. Propp revolutionized the study of narrative structure through his seminal work, Morphology of the Folktale (1928), in which he decomposed Russian folktales into thirty-one invariable narrative functions and seven recurring character types. His structuralist methodology posited that despite variations in surface characters and motifs, wonder tales share a constant morphological sequence. Propp's groundbreaking typological approach profoundly influenced structuralism, semiotics, narratology, and modern media studies worldwide, directly shaping the work of scholars such as Claude Lévi-Strauss, Roland Barthes, and Algirdas Julien Greimas.
Introduction
This article analyzes the influence of Vladimir Propp's morphology on contemporary computational narratology and artificial intelligence systems. The text examines whether formal fairy tale schemes can serve as an algorithmic foundation for plot generation.
The reader will discover why technical structural correctness does not guarantee artistic value. We will find that the boundary between generation and creativity lies in the machines' lack of intentionality and axiology.
Propp's Morphology as a Structural Foundation for Story Generation
Propp's theory provides AI with a ready-made skeleton in the form of a finite set of functions that maintain a constant order. Consequently, systems do not need to "invent" stories, but rather fill precise roles with specific characters and events.
This approach drastically limits the search space for the algorithm. Instead of infinite combinations, the AI operates within a cultural heuristic. Examples include systems such as ProtoPropp, which automate plot construction based on these rules.
However, simply mapping functions is not enough to create an engaging story. Formal coherence is a necessary condition, but it is not synonymous with aesthetic quality or emotional tension.
From Mechanical Structure to Modeling Intentionality
A believable story requires more than a sequence of events—it requires intentionality. The audience must understand why a character acts, rather than simply the fact that an action pushes the plot forward.
To avoid the "puppet effect," computational narratology introduces models such as BDI (Belief-Desire-Intention) and the IPOCL system. These allow for the modeling of agents' beliefs and goals, providing them with psychological depth.
Modern neural models (LLMs) differ from classical symbolic systems. Instead of explicit rules, they utilize statistical representations of patterns, which increases textual fluency but often weakens the global narrative architecture.
The Conflict Between Plot Structure and Agent Autonomy
In interactive systems, such as games, a conflict arises between the author's rigid plan and the player's free will. The solution is to treat Propp's functions as narrative goals rather than rigid scenes.
Instead of decision trees, narrative mediation and dynamic planning are employed. The system reacts to user actions by preconfiguring the path to the goal, maintaining plot integrity while allowing full agent agency.
However, a distinction must be made between technical optimization and the creative process. AI's ability to generate millions of combinations is not equivalent to artistry. Creativity requires axiological selection—the ability to recognize which variant possesses actual meaning.
Summary
Artificial intelligence can construct a narrative labyrinth with mathematical precision and find paths within it that are non-obvious to humans. However, the meaning of this journey does not derive from the number of parameters or computational speed.
The value of a story is born only in interaction with the recipient, who assigns meaning through the prism of their own experiences and values. It is the human need for meaning that transforms a dataset into a true experience.
Frequently Asked Questions
How can Vladimir Propp's theory be utilized in artificial intelligence systems to create plots?
Vladimir Propp's theory is used in AI systems for the automatic construction of plot structures by defining a set of permissible functions and the dependencies between them. It allows for the limitation of the search space for algorithms by imposing sequential and compositional rigors, which increases the likelihood of generating coherent and recognizable stories.
Why is simply mapping Propp's narrative functions in AI not enough to create a believable story?
Simply mapping Propp's functions ensures only compositional and functional coherence, ignoring the issue of character motivation and intention. Without modeling desires and beliefs, characters become puppets, and while the story may be causally possible, it remains psychologically unbelievable for the audience.
How can a rigid plot scheme be reconciled with the free will of characters and players in generative systems?
Situations should be designed so that the author's goals are achieved through actions that are rational from the perspective of the character and the player. Instead of rigid story trees, systems can treat the plot as a dynamic plan, where narrative functions become goals that can be realized in many ways depending on the user's actions.
Why is simply mapping plot schemes not enough to create an engaging story in AI systems?
Simply mapping plot schemes is insufficient because tension in a story depends on managing the audience's knowledge and the probability of alternative events. An engaging narrative requires modeling the psychology of the recipient and the strategic distribution of information, rather than just a correct sequence of functional events.
How does the way of modeling narratives in classic symbolic systems differ from modern language models?
Classic symbolic systems require explicit encoding of the world through precise rules and ontologies, making them highly interpretable. Modern language models, on the other hand, learn statistical regularities from vast text corpora, utilizing distributed representations instead of manually defined functions.
Why does simply ensuring structural coherence in AI-generated texts not make them creative works?
Structural coherence alone concerns only combinatorial and genre correctness, whereas creativity requires meeting the conditions of novelty and value. While generating text is a technical problem, assigning meaning to it and evaluating its aesthetic value are axiological problems that AI does not solve on its own.
How do interaction with the recipient and system complexity affect the meaning of a story compared to rigid structural rules?
The meaning of a story does not result from rigid rules, but is the outcome of the interaction between the generator, selection mechanisms, and the recipient's reactions. Through this process, emergence can occur—the appearance of complex narrative properties (e.g., emotion or a sense of justice) that are not explicitly written into the system's simple rules.
Is the ability of AI to generate an infinite number of plot combinations equivalent to a creative and artistic process?
No, because the number of combinations is not identical to the depth of discovery. Unlike technical optimization, in art there is no single, objectively defined objective function nor a mathematically indisputable answer to the question of which story is the best.
Why does the technically correct generation of a story by AI not make it a valuable work of art?
Technical correctness does not make a work valuable because coherence and formalization are not synonymous with meaning or creativity. The value of a story is determined by non-algorithmic factors such as culture, experience, collective memory, and human relationships.
