In this paper we investigate if, and then how interaction can define narrative in a computer game. When developing computer games that include stories or story elements it seems natural to use the traditional linear narrative and storytelling
methods without emphasizing that the computer is essentially different from any other media because of its interactive capabilities. Therefore it is important to investigate how it is possible to create forms of narratives that utilize the computers interactive capabilities. We approached the problem by investigating how it is possible to structure and objectify narratives so it can be understood by the computer. Through our investigation we propose that the solution to creating interactive narratives is to make procedural models that can be understood by the computer and theoretical models to work with when constructing such systems.
Keywords: Narrative, Interactivity, Computer games,
Storytelling is mankind’s tradition of communicating information, whether being educational, inspirational, entertaining or otherwise to get the listeners attention. We tell stories from our daily lives as a form of communicating our or other people’s experiences, ideas, or points of view.
Storytelling and the narrative therein has evolved in association with advances in our society and reflects on many of its changes (Mello, 2001, ¶ 2). As we sometimes observe; storytelling reflects social conditions, popular interests and traditions as in 19th century fairytales where characters dressed and spoke according to conventions of the story’s time and/or reflected the author’s hopes, fears or dreams. This is also true in 21st century narratives which both reflect contemporary and future interests and nostalgic flashbacks to ancient cultures of might and magic. Storytelling also serves the purpose of conserving social history, personal deeds and accomplishments; and through them we are allowed access to a grand spectrum of dramatic personal events, funny happenings and epic tales that changed the world as we know it (Mello, 2001; Latvala, 1999; Denk, 2006)?
When the storyteller or author sits down to tell or write a story he is usually doing so from his own point of view, he is presenting his perspective on certain topics, and uses his abilities to lead the reader through events in the story. The storyteller /author try’s to construct causes and effect relationship between events and the characters he creates to act in these events.
The method he uses to construct this is generally referred to as a narrative.
As a consequence, the author is, wholly or partly, responsible for the intellectual and artistic content of his story and as such has more or less total control over its content (NLC, 2002, p.21).
It is that total control that we, in this paper, want to challenge by investigating if established events and characters in a predefined story can be controlled and manipulated. By predefined story we mean, where events and characters have their own state or conditions for existence, and are subject to change when changes happen in their immediate environment, either through direct contact or alteration of objects in that environment. The changes are generated by interacting with the story, navigating and manipulation of elements therein, creating an interactive narrative.
In this paper we will describe our approach to interactive narrative and some theoretic models for describing and implementing interactive narrative. We ask the following question: Is it possible to simulate interactive narrative in a computer game? By providing answers to this question we hope to provide alternative methods of experiencing narrative through interaction.
The word Narrator, and its subsequent methodological denominator Narrative, derive from the Latin word “narrare” which has its Indo-European origin in the word “gnarus”, which means “to know.” (Meadows, 2003, p.5) Thus a narrative is a form of knowledge, and a narrator is one who has knowledge of something he is telling us.
In their book Film Art: An Introduction, Bordwell and Thompson (2004) conclude that narrative can be described as: “…a chain of events in cause-effect relationship occurring in time and space” (p.69).
While Cohn (1999) adds a more human
element, when she states that narrative is a:
“…series of statements that deal with a
causally related sequence of events that
concern human (or human-like) beings” (p.12)
Other definitions tend to be more towards more current media formats, explaining that a narrative is: “…a message that tells the particulars of an act or occurrence or course of events; presented in writing or drama or cinema or as a radio or television program”
(WordNet® -a, 2006).
While others are more specific: “How the plot
or story is told. In a media text, narrative
is the coherent sequencing of events across
time and space” (Boles, 2006).
These definitions are all justifiable and include elements that are of a traditional nature and can be found in traditional storytelling. All underline a causal approach to events in a story, while the some emphasizes events in time and space and human or human-like beings as part of a narrative. It can probably be argued that human or human-like characteristics are a part of countless narratives, since even abstract entities sometimes include recognizable human elements as a means to convey messages through a story.
According to these definitions we make sense of a story through narrative by identifying with its elements and linking them by cause and effect, time and space. Whether the time is backward or forward or events are remixed in seemingly random order, with hard established cause and effect relationships the fundamental purpose of narrative is to communicate meaningful messages to audiences or individual persons.
With all these combined elements a narrative can be considered the overall method chosen to convey the message within a story to the reader, viewer, or player depending on the media platform (Dansky, 2006, p. 1).
Narratives – Paradigm analysis
Our approach is that narratives can be analyzed and broken down into, at least, character types and the functions they perform in any given story regardless of demographic or social situation. We further suggest that these categories of character types and functions can be simulated as constants and variables depending on the task at hand. These categories of character types and functions are not necessarily simulated in detail but up to the point where it is perceived to be real.
Furthermore, we emphasized that building interactive narrative based on characters or entities is not entirely about building computational algorithms, but also concerns giving these characters and objects their own life by simulating what it means to exist in any given story orientated environment by simulating communication, emotions, and other complexities of human social interactions. We propose, based on the assumption that real life narrative and behavior – as it is with learning and intelligence – can be described precisely enough for it to be simulated by a machine. (McCarthy, Minsky, Rochester & Shannon, 1955, p.1)
This suggestion contradicts some authoritarian scholars in the field of narrative structure where they state that narrative is fundamentally a cognitive mental process that makes the human experience meaningful (Polkinghorne, 1988, p. 1) (McQuillan, 2000, pp. 7-8) and should, therefore, be at best a vague concept and hard to objectify. We counter-argue that the relationship between narrative and computer technology is at best ambiguous and that further research is needed to determine what happens to narrative when it is presented by a machine.
Simulating Interactive Narrative
If we are to reach the state of interactive narrative one needs to generate changes at the core of the narrative structure. If through interaction one is able to decide how the plot progresses by starting and stopping actions and events, and in what order events occur while maintaining a comprehensible plot or story he has achieved the state of interactive narrative. Narrative is there to produce meaning and comprehension, if that fails there is no narrative; only confusion.
Based on this, we propose that the plot in an interactive narrative scenario be described on two levels: the author level and the player level. Furthermore, we propose that a procedural and theoretical model is needed to describe in detail the elements that are essential for a narrative to be considered interactive.
By describing the plot at two different levels a form of balance could be achieved between these key elements in the game narrative:
1.High-level plot – predefined elements are under the author’s control
2.Low-level plot – the elements are freely altered by the player
At the high-level, the author could retain the freedom to define the key episodes of the story while their actual implementation in the player time-frame is left to the game engine. While at the low-level the player performance can be dynamically analyzed and logically applied to the lower level details of the how and why of a specific episode or game sequence. Figure 1 shows a model describing the two-level plot.
We expect that the overall effect will be that the plot becomes more dependent on the player who gains a part of the control on the development of the story. Moreover, the player is provided with a less predictable, more life-like experience with a different selection of the details in each episode and possibly the (involuntary) triggering of different sequences of episodes altogether.
Interactive Narrative Model
Interactive narrative needs a model of the narrative structure in any given computer game. This model should be based on analysis of traditional narrative structures and include the key elements or principals needed to successfully execute or reach a state of reciprocal course of action that triggers interactivity.
The challenge is to find or construct a model that is suited to the interactive nature of computers. This model should be able to describe the procedurals of interactive narrative and not just a list of its elements.
We work under the impression that interactive narrative should be based on a realistic simulation. This means that one needs to replace the action of the author statically writing the plot, with dynamic software capable of simulating the writing that is supported by theoretical structure to ensure that the storyline keeps coherent and intriguing.
Therefore, it is necessary to develop a form of procedural model or platform for narrative, where the player engages in interactive communication with the gaming world resulting in a state of interactive narrative or as close to interactive narrative as the simulation allows. This platform would naturally be a computer game with an intriguing storyline or a journey through multiple rising actions, main events, and conclusions manipulated through interactive action by the player.
Figure 2 shows a conceptual model of one possible structure of the narrative elements and their relationship to other elements and functions within the game framework.
1.Through the game the player makes sense of the gaming world and brings his perception to it through? the story/game mechanics
2.Narrative causality and logic:
Here the rules, including all possible NPC actions at any given time in the narrative, are executed. Gaming world includes the narrative causality and logic and is designed and made by the game developer.
3.Narrative sequences and events:
Here the player makes a choice in the narrative; these choices are influenced by all possible actions provided by the execution of NPC rules.
An adaptive process where the player is affecting and being affected by narrative events. Player is able to affect narrative sequences of events and consequently their perception of events changes.
5. The narrative causality and logic is a constant predefined element but still contributing to narrative sequences of events as perceived by the player.
We suggest that a model of the narrative is needed to provide visual representation and overview of the components needed for an interactive narrative to be comprehensible and support the plot structure provided in the story. Such a model includes physical elements such as: human like characters, environmental objects (houses, cars, and trees etc.) and subjective elements like emotions, textures, light and shadows, character identities and other references to real entities.
We furthermore propose that narrative is not only a cognitive process, but can be presented as concrete units and rules that are subject to changes through interaction.
As a consequence we propose that the term interactive narrative describes the result of the interaction between – predefined elements in the game formalized and developed by the author – and the subjective perception that the player will form while interacting with them.
The problem is that a comprehensive plot is dependant on a successful execution of the narrative and the relationship between events that need to be established in a meaningful way. The danger is that if we leave this relationship to chance or random interactive functions, we risk that no narrative will be established.
However, by analyzing traditional narrative functions and elements it is possible to isolate elements within narratives that can be considered action based and subject to active engagement and justifies the implementation of interactivity into narratives within the framework of an event based story.
Having established a relationship between interactivity and narrative we conclude that interactive narrative is a real possibility given that all parameters are met. We further conclude that extensive procedural and theoretic models are needed to explain and execute a state of interactive narrative in a computer game.
Bordwell, D., & Thompson, K. (2004) Film Art: An Introduction, New York: McGraw Hill, 2004 (7th ed.), p.69
Cohn D. (1999) The Distinction of Fiction,
Baltimore: Johns Hopkins Univ. Press, 1999, p.12
Dansky, R. (2006). Introduction to Game Narrative, In Bateman, C. Game writing: Narrative Skills for Videogames. Boston Mass: Charles River Media. Retrieved 10 November, 2007, from
Denk, K. M. (2006). Making Connections, Finding Meaning, Engaging the World:
Theory and Techniques for Ignatian Reflection on Service for and with Others, p.17.
Latvala, P. (1999) Finnish 20th Century History in Oral Narratives. Folklore vol.12. December 1999. Retrieved 10 November, 2007, from
McCarthy. J., Minsky, M. L., Rochester, N., & Shannon, C.E. (1955) A proposal for the Dartmouth summer research project on Artificial Intelligence. Retrieved December 6th, 2006, from: http://www.formal.stanford.edu/jmc/history/dartmouth.pdf
Meadows. M.S. (2003). Pause & Effect: The Art of Interactive Narrative. Indianapolis USA: New Riders
Mello, R. (2001). The power of storytelling: How oral narrative influences children’s relationships in classrooms. International Journal of Education & the Arts. 2(1). Retrieved December 6th, 2006, from: http://www.ijea.org/v2n1/index.html
WordNet® -a (2006) Cognitive Science Laboratory Princeton University. Retrieved December 6th, 2006, from: