The Emergent Narrative theoretical investigation

This paper is part of a work in progress that aims to define a narrative theory adapted for optimal use in the Virtual Reality (VR) medium. In addition to theoretical results and findings on the development of an Emergent Narrative (EN) concept, the authors present their convictions regarding Virtual Storytelling (VS) and narrative consideration. Finally, an Emergent Narrative model is described along with a discussion on the novelty of the concept. Introduction In recent years, the AI community has shown a growing interest in storytelling and Interactive Storytelling (IS) and has been developing projects and theories in a number of areas [6, 7, 16, 17, 18, 20, 21]. Amongst them is the concept of Emergent Narrative (EN). Introduced by Aylett in 1999 [2], the concept has been developed at both the NILE00 and the NILE02 conferences [3, 12]. The research currently being conducted on EN is also implicated within the European Framework V project VICTEC (Virtual ICT with Empathic Agents) [23] and the European Network of Excellence project HUMAINE (Research on Emotions and HumanMachine Interaction) [11]. The EN concept aims at a potential answer to the narrative paradox [3] observed in 3D Virtual Environments (VE) and graphically represented Virtual Storytelling systems. EN seeks a character-based approach as opposed to a plot driven approach for IS and the construction of systems in which users actively participate in the narrative process in a highly flexible realtime environment, where authorial activities are minimised. The development of the story is managed by both the user and embodied Intelligent Agents and depends entirely on the interactions between each other and their environment. The research aims at finding and developing a narrative structure appropriate to the specific characteristics of Virtual Environments, combining the entertainment values of both storytelling and virtual experiencing. Furthermore, research into the Emergent Narrative concept has been targeted at its potential in terms of entertainment, education and pedagogy. It could potentially greatly impact the design and game-play of video games or interactive entertainment performances. In addition, since a number of educational methods use themes and storytelling to engage people’s emotions, one could see the study and use of narrative as a pedagogical tool for the development of Personal and Social Education (PSE) [10, 24, 25]. The Emergent Narrative concept could provide opportunities for educational and pedagogical material to be presented within a compelling, interactive and immersive Virtual Environment [23]. Part I: The Emergent Narrative Investigation The theoretical work involved with the development of an Emergent Narrative theory relates to previous and current research and theoretical work, most of which originate in different research domains. This investigation requires the researcher to consider stories and narratives from a high level of abstraction in order to capture their essence. Stories and narratives are extremely complex processes involving interactivity, character representations, narrative dynamics, user experiences, decision-making processes, participative narrative forms and affective and social behaviours. It is also necessary to consider how far the more classic and traditional narrative theories can be applied to this view of narrative. An Emergent narrative model is also multi-dimensional since several essential parameters such as space, time, narrative surface, user role and the nature of narrative controls have to be taken into account. 1.1: Interactive storytelling considerations As discussed in previous publications [4, 12, 13, 14, 15], classical narrative approaches [1, 5, 19] have little to contribute in practice to a theory of emergent narrative. There are a number of reasons for this. The role of the user is a key issue in interactive storytelling, with whether the user is considered as an author or a participant within the story having a major impact on theoretical approaches. The contradiction between authorship and participation is an important element of the narrative paradox previously mentioned. On the one hand an author seeks control over the direction of a narrative in order to give it a satisfying structure. On the other hand a participating user demands the autonomy to act and react without explicit authorial constraints. Casting the user either as a spectator, with no ability to act, or as the author him or herself avoids this problem, however it does not offer a solution for a participating user in real-time interaction within a narrative display. It also limits the storyline to a single entity. We argue that a serious consideration of the user as participant can actually present a solution to the narrative paradox, in the sense that it would encourage the emergence of several storylines while still leaving the user with the responsibility of conducting real-time meaningful actions within the unfolding narrative. The narrative paradox observed in 3D environments is a constraint only if the storyline of the narrative is one single entity to which all characters and the user must conform. Such problems do not occur when the storylines are multiple and opened to emergent behaviours and events. However, such a user approach implies that the story must be designed in a way that enables the narrative to emerge from the interactions between the different characters, the user and their individual storylines. Such a narrative design refers to the principles of storyfication,[3] in which the experience of a character is internally transformed into story, and argues for a dynamic and synthetic rather than a static and analytical approach to narratives. A process view of narrative can only at best provide a provisional instance of the overall storyline or plot. Although themes and character descriptions are authored, the storyline emerges from their interactions with each other and the user and cannot be completely predicted. Characters are motivated by their personality traits and personal characteristics and build the story as they interact with each other. This particular view of narrative is fundamentally opposed to the Aristotelian classical view of narrative [1] and argues for a character-based rather than a plot-based approach for interactive performances [4, 12, 13, 14, 15]. A character-based approach focuses on character definition rather than plot construction, and the authorial process relies on the articulation and control of interesting characters rather than the generation of plot events. 1.2: Participative narrative findings Since classical narrative theories and models are mainly analytical, they do not offer practical mechanisms to deal with interactivity and user integration. The investigation of the Emergent Narrative concept, in accordance with its premise of the user as a participant rather than a spectator or author, has been mainly oriented towards the study of interactive and participative narrative and performance forms and models. Genres such as Role Playing Games (RPGs), interactive and street theatre or video games are all of potential relevance to the development of an Emergent Narrative theory. RPGs lie mainly in the realm of entertainment although RPGs can be used for pedagogical purposes [10] and the lack of resources such as research works, literature or publications makes its investigation challenging. There are also a wide range of games and mechanisms for which it is difficult to produce comprehensive definitions of types and categories. The selected approach for this study is empirical and aims at identifying narrative patterns, elements or factors influencing the creation, development and unfolding of dramatic narratives and stories, using the knowledge acquisition software KAT [8]. The success of such an approach depends heavily on the quality and level of expertise of the expert involved: the results displayed in this paper are based on knowledge elicitation sessions conducted with experienced RPG Game-Masters and writers over a 12 month period. In the study three different types of RPGs were considered for certain specific reasons. Firstly, board RPGs were studied. Board RPG is played with fictive characters, sometimes actually with board-sized models, and is organised as a campaign; the game is composed of several sessions and usually involves a quest with a group of largely cooperating characters. The interest of this genre lies in their episodic nature and the way they handle narrative events and character development. Secondly, conflicting RPG was researched. Conflicting RPG is a variant played with conflicting character goals and personalities over a short period in a single session and is of interest for its management of narrative tension and narrative set up. Finally, Live RPG was analysed: Live RPG is played in the real world in the same type of spirit as historical re-enactment, and is relevant for its management of narrative controls on a real-time basis. When first investigating the field of RPGs it became apparent that they shared a lot in common with the concept of Emergent Narrative and that a lot could be drawn from existing practice. Areas of the theory that are the most susceptible to be influenced by RPGs are the ones that are concerned with story management, narrative control and character definition, roles and motivations. The way the story unfolds is emergent and fits perfectly with our vision of an Emergent Narrative theory. However, the level of cognition necessary for the GameMaster role in managing and supporting this process seems beyond reach as far as Artificial Intelligence and intelligent agents are concerned. Thus the study of board RPGs on its own does not provide enough elements for a realistic theory of emergent narrative. The time dimension and particularly real-time interaction is a very important element that dramatically affects the form, nature and level of narrative control. Although board RPG is directly relevant to the investigation in terms of character development and the handling of narrative events, it proved to be irrelevant as far as narrative controls were concerned. However, conflicting RPG and Live RPG handled the real-time element successfully and similarly, distinguishing between narrative controls that occur before the RPG game session and those that occur during the game session. In both cases, most of the narrative controls are defined and set up prior to the game taking place, emphasising the idea of the narrative as a hypothesis based on the types and personalities of the characters, their roles and the nature of the environment that surrounds them. The narrative control over the game’s set-up is limited to the definition of the environment, tasks and roles. Since there is no way for the game writers to predict with accuracy the players’ state of minds or histories prior to pre-designed narrative events, it is virtually impossible to guarantee how they will affect the players. The plot is highly provisional and its meaningfulness is almost totally dependent on the way the character roles have been written. Such an approach encourages the development of sub-plots and individual or relatively small group storylines. The design generally consists of a hypothetical storyline composed of several hypothetical plot elements, each of which involves several groups of a number of characters. Plot elements should be inter-related with at least another plot element, their interaction ultimately defining the storyline of the game session [Figure 1]. Figure 1: Storyline decomposition for interactive participative narrative forms Since there are inevitably a large number of events that occur but are not predicted or anticipated and may even go unnoticed by the game writers or players, conflicting and Live RPGs generally conduct a debriefing session at the end of the game. Here each player describes their individual storyline, along with their understanding of the overall game plot to the other players. This session helps individuals to understand the role they played, either consciously or unconsciously, in the unfolding of the game narrative, and the reasons for other players’ behaviours, thus refining and completing the individual storification processes.. Interventions from the Game-Master that allow some sort of narrative control during the game session are limited to the timing and unfolding of narrative and dramatic events and the use of dedicated agents that act for the interest of the dramatic unfolding and obey directives emitted by the Game-Master. However, such practice is not general and would only limit direct interaction with the players to an influential rather than an authoritative status. The study of interactive theatre and street theatre were also essential in the consideration of an Emergent Narrative theory in bringing know-how, rules and mechanisms in defining characters for interaction. Specific attention must be brought to characters if they are to be designed for an interactive environment, for example they must be given a back-story, a history, and a set of motivations. Such attention would bring depth to characters and help to define their believability and credibility. Video games were also studied in order to identify mechanisms that would bring emotional depth to characters [9]. However, the study of video games surprisingly failed to deliver practical narrative elements for a theory of Emergent Narrative. Although it is undeniable that the way characters and environments are presented in video games is of interest to a theory of Emergent Narrative for graphical environments. In fact, this has a stronger impact on content development than narrative structure. Video games have shown, in general, a classical and conservative approach to narrative considerations. To conclude, certain elements have been identified that are essential to an Emergent Narrative theory within participative and classical narrative approaches. However, it must be noted that what is theoretically feasible is not always practically realisable; the fact of practical applications of part of the theory in the field of RPGs is encouraging concerning the success of such an approach. The findings of the investigation could be briefly summarised within the following set of rules [Table 1]. Hypothetical plot line Hypothetical plot elements