Essential Elements of Narrative and Agency in Digital Interactive Narrative Games

ESSENTIAL ELEMENTS OF NARRATIVE AND AGENCY IN

DIGITAL INTERACTIVE NARRATIVE GAMES

Lindsey Joyce

APPROVED BY SUPERVISORY COMMITTEE:

______Monica Evans, Chair

______Kim Knight

______Roger Malina

______Miguel Sicart

______Frederick Turner

Lindsey Joyce

For my spouse, who lent strength and love in equal measure.

For my children, who inspired me through moments of doubt.

For my parents, who had faith when I lacked it.

For my sister, who never doubted and always listened.

For my friends, who understood and forgave.

For McKinley, who gave unwavering companionship.

and

For Bach, who wrote the melodies that kept me writing.

ESSENTIAL ELEMENTS OF NARRATIVE AND AGENCY IN DIGITAL INTERACTIVE NARRATIVE GAMES

LINDSEY JOYCE, BA, MA

DISSERTATION

Presented to the Faculty of

The University of Texas at Dallas

in Partial Fulfillment

of the Requirements

for the Degree of

DOCTOR OF PHILOSOPHY IN

ARTS AND TECHNOLOGY

THE UNIVERSITY OF TEXAS AT DALLAS

May 2017

ACKNOWLEDGMENTS

I have stood on the shoulders of giants. This text is proof of it. Somehow, I was lucky enough to assemble the Dream Team of committees. Without them, none of this would be possible. Over the last several years, this committee has guided and supported me in ways I am still discovering.

This text is a testament to them and stands as physical evidence of the enormous debt I owe, but can never hope to repay. I must first thank the chair of my committee, Dr. Monica Evans, who in the Fall of 2014 changed the direction and focus of my research with the simple question “Why are you here?” Though I could only guess at the answer then, Dr. Evans gave me her time, her patience, her respect, and her trust while I figured it out. Beyond that, she modeled the type of scholar and human I hope to be. I must also thank the other members of my committee: Dr.

Frederick Turner who indulged me in long conversations about video games and narrative that I will treasure forever and which have been recalled in this text more than once; Dr. Kimberly

Knight who never let me be comfortable with where I was, but instead continually willed me to be better and made me want to put it the work it would take; Dr. Roger Malina who changed the ways I think about media and publishing and academia, and whose stories about his own journey reminded me that missteps can be overcome with work and the right attitude; and Dr. Miguel

Sicart whose books challenged my assumptions about video games in ways that ripple throughout this text and who was kind enough to speak to me about soccer in a shuttle van in

New Jersey and, subsequently, spark the confidence it would take to ask him to join this committee.

I must also acknowledge my spouse and children who sacrificed so much, without complaint, while I pursued my degree. For every, “I’m sorry, I can’t right now” or “maybe in a minute,” or

“I wish I could, sweetie” I uttered over the last several years, I was consistently met with support and unwavering understanding. Together, they are my heart and deserve my everything.

Finally, I owe incomparable gratitude to my parents. Despite all my growing pains and all the stumbling blocks, they never once doubted my ability or the depth of my will. They have given their faith, fully and unfailingly, and I have leaned on it when there was little else to lean on. I know they will be as thankful as I am when I can finally take the next step.

February 2017

ESSENTIAL ELEMENTS OF NARRATIVE AND AGENCY IN

DIGITAL INTERACTIVE NARRATIVE GAMES

Lindsey Joyce, Ph.D. The University of Texas at Dallas, 2017

Supervising Professor: Monica Evans

Digital interactive games, or video games as they are more commonly called, have the substantial potential to involve players in the act of story creation. Though not all video games contain a narrative with which the player can interact, many do. These games, known as digital interactive narrative games, purposefully embed a narrative structure and allow a player to interact with narrative elements to produce a unique story product. Thus, while not all games should be considered narrative media, this text deals with those that can. To begin, this text addresses the counterproductive argument that suggest videogames are not a narrative form of media. Instead, this text posits that, if we acknowledge that digital games have narrative potential and that many game developers want to incorporate narratives into games, we can proceed with inquiries about how digital games support story creation.

While some Game Studies scholars argue that the inclusion of a narrative limits a player’s agency, this text suggests that such arguments misrepresent how narrative structures lead to the creation of stories and how those narrative structures support numerous type of player agency.

As such, this text examines narrative theory and its structures to show how those structures map

vii

to digital interactive narrative games. Next, this text engages with social agency theory to consider new ways of understanding the states and types of agency players possess in digital interactive narrative games. Based on these analyses, a set of evaluative criteria is generated to aid in the assessment of narrative and agency in digital interactive narrative games. These criteria are tested and applied in a series of three case studies of four digital interactive narrative games:

Mass Effect 2, The Elder Scrolls V: Skyrim, Kentucky Route Zero, and The Banner Saga.

Two primary conclusions are drawn in this text. The first is that players of digital interactive narrative games act as agents who perform as the narrator agent and the character agent found in traditional, non-procedural, narrative mediums. The second is that agency is not static in games, nor is agency of a single type. Rather, there are different types of agency a player can possess and maintain, and these occur at different times and to different degrees. The level of agency a player maintains will always be in flux; as the player interacts with the game, its rules, and its procedures, the player also interacts, mediates, and negotiates agency within the world the game creates. Similarly, the types of agency the player may experience are numerous. Given the focus of this text, two kinds of agency are directly considered: narrative agency and ludic agency. By neither oversimplifying the understanding of narrative nor of agency, this text establishes new ways and methods for understanding and analyzing digital interactive narrative games and the ways in which these games are and can be designed to contain a well-formed narrative while also establishing ways for the player to intervene and participate in the creation of a story.

TABLE OF CONTENTS

ACKNOWLEDGMENTS ...……………………………………………………………………..V

ABSTRACT…………………………………………………………………………………….VII

LIST OF TABLES ...…………………………………………………………………………….XI

INTRODUTION ...………………………………………………………………………………..1

PART ONE: THEORIES OF NARRATIVE AND AGENCY FOR DIGITAL INTERACTIVE NARRATIVE GAMES

CHAPTER 1: METHODOLOGY FOR CHAPTERS 2 AND 3 ………………………………... 9

CHAPTER 2: AGENTS IN SPACE AND TIME: FUNDAMENTALS OF NARRATIVE DESIGN …………………………………………………………………………………………15

CHAPTER 3: PROCEDURALLY CONSTITUTED AND SOCIALLY CONSTRUCTED AGENCY: AGENCY THEORY IN DIGITAL INTERACTIVE GAMES …………………….66

PART TWO: THE ANALYSIS OF NARRATIVE AND AGENCY IN DIGITAL INTERACTIVE NARRATIVE GAMES

AUTHOR’S NOTE CONCERNING CHAPTERS 5-7 ………………………………………..100

CHAPTER 4: METHODOLOGY FOR CHAPTERS 5-9 …………………………………….102

CHAPTER 5: CREATING COLLABORATIVE CRITERIA FOR THE ANALYSIS OF NARRATIVE AND AGENCY IN DIGITAL INTERACTIVE NARRATIVE ………………112

CHAPTER 6: ASSESSING MASS EFFECT 2 AND THE ELDER SCROLLS V: SKYRIM USING COLLABORATIVE CRITERIA FOR NARRATIVE AND AGENCY IN DIGITAL INTERACTIVE NARRATIVE ………………………………………………………………..131

CHAPTER 7: KENTUCKY ROUTE ZERO: OR, HOW NOT TO GET LOST IN THE BRANCHING NARRATIVE SYSTEM ………………………………………………………175

CHAPTER 8: THE COLLABORATIVE CRITERIA CHRONICLES: AN ANALYSIS OF AGENCY AND NARRATIVE IN THE BANNER SAGA …………………………………….189

CHAPTER 9: SUMMARY OF FINDINGS …………………………………………………...219

CHAPTER 10: FINAL THOUGHTS AND FUTURE CONSIDERATIONS ………………...238 BIBLIOGRAPHY ……………………………………………………………………………...248

BIGRAPHICAL SKETCH …………………………………………………………………….262

CURRICULUM VITAE

LIST OF TABLES

Table 1. Summary of Criteria ...... 135

Table 2. Analysis of Drama Management Criteria ...... 142

Table 3. Analysis of Narrative Experience Criteria ...... 150

Table 4. Analysis of Ludic Agency Criteria ...... 157

Table 5. Analysis of Narrative Agency Criteria ...... 165

Table 6. Analysis of Character Development Criteria ...... 170

Table 7. Amended Summary of Criteria ...... 192

Table 8. Analysis of Drama Management in The Banner Saga ...... 197

Table 9. Analysis of Narrative Experience in The Banner Saga ...... 203

Table 10. Analysis of Ludic Agency in The Banner Saga ...... 207

Table 11. Analysis of Narrative Agency in The Banner Saga ...... 213

Table 12. Analysis of Character Development in The Banner Saga ...... 216

Table 13. Final Analysis of Drama Management ...... 220

Table 14. Final Analysis of Narrative Experience ...... 223

Table 15. Final Analysis of Ludic Agency ...... 226

Table 16. Final Analysis of Narrative Agency ...... 230

Table 17. Final Analysis of Character Development ...... 233

INTRODUCTION

Game Studies as a field of research is still quite new compared to other fields of media study, including Art, Literature, Music, Theater, and even Film. While each of these media types, including games, undoubtedly share some aesthetic qualities, methodologies, and critical theories, scholars within the burgeoning field of Game Studies are generally more concerned with how it differs and what makes it theoretically and methodologically unique than with how it may be similar to other mediums. In other words, as a field of research, Game Studies is still anxious about validating itself as both unique and necessary.

This anxiety is par for the course. Once, Fiction was a field of study that had to validate itself against the likes of Poetry and History. Film Studies has only gained prominence as a valid field of research within the last century. As new media emerge, so too do the struggles to establish the methodologies and theoretical practices that will ground them in scholarly inquiry and institutions.

In the struggle to legitimize the field of Game Studies, scholars have often worked, for better or worse, to establish the worthiness of games scholarship by focusing on what sets digital games apart from other media. Consequently, the similarities between digital games and other media forms have always been viewed suspiciously. Comparisons between digital games and other mediums often include the caveat “yes, but…,” wherein the similarities are begrudgingly noted, but then complicated and made different again in relation to digital games. For example, while digital games employ methods and practices found in visual arts, sound design, theatrical performance, and film, games also provide something none of these other forms do: interactivity.

In digital games, there is an active player who spectates, mediates, and engages with the media throughout the duration of the gameplay experience. This interactivity is different from the interactivity theater scholars argue the audience possesses, and is also different from the interactivity English scholars argue the reader possesses. Digital games are coded to receive and respond to input from the player. The digital game requires an active player for the performance of the game to ever occur in a far more literal way than a movie screening requires an audience.

Thus, while Game Studies scholars do not deny that digital games share features and elements found in other media, they are more concerned with how these techniques are used to accomplish what other media forms do not.

Some might argue that this is all nothing but semantics. In fact, I would and do argue the same thing within this text. Yet, I also argue that the semantics, the distinctions in meaning and definition, are of profound importance to how we understand digital games as media objects and how we understand the theories that have, so far, been generated to analyze them.

For instance, within this volume, I discuss how Game Studies scholars have attempted to define games as a medium that is unique from all others while simultaneously borrowing concepts and terminologies from those other fields to do so. Despite this irony, the reasons for borrowing these terms is, at least somewhat, logical. By using already established and understood terms and concepts, Game Studies scholars automatically make their work more accessible to others by using them. Additionally, as a growing field, scholars have arrived at

Game Studies from numerous other fields of study. Bringing terms with them is not only logical, but doing so also permits scholars with different theoretical backgrounds to establish common ground from which to discuss digital games.

In terms of semantics, however, complications arise. Because Game Studies scholars commonly borrow terms from other fields of media study in order to establish Game Studies as a specific and unique form in its own right, game scholars also find themselves in the frequent predicament of 1) having to qualify their language use in such a way that readers can understand the familiar terms and concepts taken from other fields of research, and 2) having to ensure that readers can understand the unique differences concerning games and how each term is being applied to or used in the context of Game Studies specifically and uniquely. As a result, terms must continually be re-defined and qualified.

More problematic than the constant defining of terms, however, are the important contexts that are lost in the process. With each new definition comes the risk of erasing or negating the historical context behind the term. While the morphological changes of any word – the borrowing and redefining of its meaning – are a natural part of language evolution, the result of such evolution is best understood in retrospect and with the distance of time. Thus, while these terms may be understood well within their original and established field, Game Studies scholars cannot presume their colleagues are always aware of the historical contexts many borrowed terms carry. Again, because Game Studies scholars have arrived at the study of games from a plethora of other fields, there is a lack of theoretical or ideological similitude. Much can be lost in translation. As terms are borrowed and shared amongst Game Studies scholars, many have begun to lose the important and historical contexts that make them apt to borrow in the first place. Furthermore, having been borrowed and made familiar enough within the field of Game

Studies, many terms have also now begun to be misused and misapplied in ways that obstruct from the goal of using them: establishing similarity in order to highlight fundamental differences.

Most problematic of all, perhaps, is that as these terms lose the external context that made them apt, theoretical misunderstandings emerge which have limited our ability to establish Game

Studies as a valid field of study.

The ludology versus narratology debate is the perfect example of what is at stake. Though Janet Murray called this debate “closed” in 2005, scholars continue to discuss whether video games should be understood exclusively on their own terms as a uniquely interactive medium or whether video games could or should be understood as frameworks that could support narratives (“The Last Word”). Within the context of this debate, two terms have become of primary importance: narrative and agency. The ludologists, or scholars who view games though a mechanical or ludological lens, generally discredit the idea that games are a narrative form, arguing instead that games are, above all else, interactive simulations that provide a player with a level of agency that cannot be found in other mediums. Conversely, the narratologists, or scholars who view games through the lens of narrative, argue that digital games, while interactive mediums, maintain undeniable narrativity and, as such, suggest that narratology is a useful lens for approaching and understanding video games.

The issue, of course, is far more complex. Both the ludologists and narratologists were as right as they were wrong, yet neither the ludologists nor the narratologists could come to an understanding. Largely, this was due to semantics. Though common words, such as “agency” and “narrative,” were being brandished within the debate, the ways in which those terms were used and understood were not at all similar. In the words of George Bernard Shaw, the two sides were as “two countries divided by a common language” (708). As both “agency” and “narrative” were defined and redefined numerous times within this debate, the potential for common

understanding and shared intellectual progress decreased. Though both sides respected the medium of digital games and agreed that they were worthy of analysis, a specific methodology for how to do so became difficult to reach.

Entering the field of Game Studies, as I did, after the ludology versus narratology debate was deemed “closed” has provided me with the distance to notice the semantic differences embedded in the arguments made with respect to both narrative and agency. As such, this text begins, first, with clarification. My intent is to illuminate the semantic differences that existed within the debate and to reconnect the terms with their historical contexts in order to reveal that, when properly understood, both terms become accessible in deeper and more meaningful ways within the field of Game Studies. Once the work of re-contextualizing the terms is accomplished, my secondary goal is to show how a method of critical analysis can be built and employed to games based on a common understanding of narrative and agency.

Given these goals, in Section 1 of this text, I analyze how the terms “narrative” and

“agency” have been understood within the field of narratology and social agency theory, respectively. In Chapter 1, I establish the methodologies used to conduct these two inquiries. In

Chapter 2, I propose that narrative, as a concept, has been misunderstood and misapplied to and within the discourse about digital games and that this misunderstanding has, by consequence, resulted in needless bipartisan approaches to Game Studies. I begin by providing a summary of ways narrative has been defined differently by various scholars within Game Studies before analyzing how these definitions differ from those historically used by scholars of narrative.

Following this, I argue that digital games can employ narrative structures, but those structures differ from traditional non-digital, non-interactive narratives in two key ways: 1) a player agent

holds both the position of the narrator agent and the character agent simultaneously, and 2) the narrative structure is more directly affected and developed by narrative-space and story-space. In

Chapter 3, I suggest that definitions for agency within the field of Game Studies are far too restrictive, especially in lieu of how agency has been defined within various movements of social agency theory. After an analysis of agency within the context of social agency theories, I propose new ways of thinking about agency in relation to digital games.

In Section 2 of this text, I use the theories formulated in Chapter 1 and Chapter 2 to create criteria for assessment that can be employed to analyze how well a digital interactive narrative game creates a meaningful and complete narrative experience while also providing the player with a sense of narrative agency and ludic agency. These criteria can be found in Chapter

5. In the follow three chapters, I test the utility and merit of the assessment criteria by applying them to the analysis of four digital interactive narrative games: Mass Effect 2, Skyrim, Kentucky

Route Zero, and The Banner Saga.

Thus, having reviewed both the theories of agency within social agency theory and the theories of agency within the field of Game Studies, in Chapter 3, I return to grounded theory in order to propose new ways of thinking about agency within the framework of digital games. By identifying the theories of agency that have, to this point, not been given enough consideration within the field of Game Studies, I propose a more robust definition for what agency is and how it can be constituted in digital games.

Finally, this text concludes by reflecting once again on the criteria in order to propose how they could be refined in the future. Though the criteria are robust enough to act as a starting point, they are just that: a starting point. The goal of this text is, ultimately, to move beyond the

semantics that have limited Game Studies and to provide a pathway by which to move forward.

Having proposed one possible pathway forward, I am eager to continue refining the theories proposed here and look forward the opportunity to solidify Game Studies as a field full of potential and worthy of study.

PART ONE: THEORIES OF NARRATIVE AND AGENCY FOR DIGITAL

INTERACTIVE NARRATIVE GAMES

CHAPTER 1

METHODOLOGY FOR CHAPTERS 2 AND 3

My intentions for the first section of this text are twofold. My first aim is to show how concepts of narrative and agency have been misunderstood and misapplied within Game

Studies scholarship, frequently to the detriment of the field. My second aim is to introduce new theories for narrative and agency that maintain important historical context while still addressing games as unique forms of media.

In Chapter 2, I argue that Game Studies scholars have taken too broad a view of narrative and have generally erred in conflating narrative with story. This conflation, as a result, has restricted how Game Studies scholars are able to understand the construction of both narrative and story in digital games. This conflation is most problematic in reference to a particular type of game: the digital interactive narrative. Digital interactive narrative games are understood here as those games in which the player interacts meaningfully and continually both as a narrator agent and as a character agent within a well-developed narrative structure for the duration of the gameplay experience in order to produce a unique narrative experience and story outcome that is both logically and emotionally complete. Thus, in Chapter 2, I examine the definitions for and elements of narrative and story before considering each in relation to digital games. By redirecting our attention to the nuances and differences between narrative and story, I argue that Game Studies scholars can better understand how narrative works in games as well as how to better construct interactive narrative structures within digital games.

In Chapter 3, I argue that Game Studies has taken too narrow a focus in defining player agency, generally by wrongly equating agency with action or the ability to act. Extending from

this, I suggest that social agency theories reveal a number of considerations and criteria that make agency possible, but which Game Studies scholars have yet to recognize or consider. These additional considerations and criteria, though not yet studied in Game Studies, do alter how a player interacts with a digital game. Given this, the study of these agency theories present a new means by which to understand and better construct player agency in digital games.

Within both of these chapters, I have employed grounded theory and textual analysis as the primary qualitative methodologies. These qualitative approaches are most apt, given that my two aims in Chapter 2 and Chapter 3 are to engage with existing theories in the fields of narrative, agency, and digital games and to construct new theories about narrative and agency within the context of digital games. As such, within these two chapters, I make systematic comparisons between theories of narrative and theories about digital games, and between social agency theories and digital game theories. To do this, I closely examine the content, meaning, and discourse found within each of these three areas of study. Thus, while my methodology is most broadly situated within the framework of grounded theory and textual analysis, I also employ comparative analysis as a more focused method of grounded theory and content analysis as a more focused method of textual analysis.

Provided that my use of comparative analysis first required the gathering and analysis of data, I have employed content analysis first and comparative analysis second. In this respect, I have treated both the theories of narrative and agency as data sources that can be categorized into clusters of similarity and difference. To establish these clusters of similarity and difference, I conducted close readings within the fields of narrative theory, social agency theory, and digital game theory.

In my survey of narratology, I guided my close readings with three primary questions: What is a narrative? How is it defined? What are its essential elements? As I engaged with these questions, I created memos from the works of an array of narratologists and paid particular attention to those scholars who provided precise narrative rules or parameters within their theories. Though certainly not an exhaustive list of all narrative theories,

I engage specifically with the theories of H. Porter Abbott, Mieke Ball, Seymour

Chatman, Dennis Dutton, Gerard Genette, James Phelan, and Peter Rabinowitz in Chapter 2. I have selected to engage with the theories of these scholars primarily because they are representative of more contemporary forms of media. Each scholar has engaged with theories of narrative within the context of a world populated by emerging digital media. Their understandings of narrative are thus, directly or indirectly, informed by the digital and technological age.

Similarly, I asked the same questions about narrative while surveying Game Studies texts. Here, I identified texts written specifically about the use of narrative in games. As such, my resources include those generated by Espen Aarseth, Greg Costikyan, Gonzola Frasca, Henry

Jenkins, Michael Mateas, Jane McGonigal, Janet Murray, Mark Riedl, Frank Rose, and Marie-

Laure Ryan. In addition to identifying texts written specifically about narrative in games, I was also careful to include representative texts from both sides of the previously mentioned ludology vs. narratology debate.

Following the process of close reading and memoing, I used the practice of comparative analysis first to compare and contrast numerous theories of narrative created by narrative scholars, or narratologists, second to compare and contrast the theories created by Game Studies

scholars, and third to compare and contrast the non-games narrative theory with the narrative games theory. As I compared the theories of these scholars, four things became immediately clear: 1) narrative and story, through frequently equated in common conversation, are entirely distinct within the study of narrative 2) a number of Game Studies scholars wrongfully conflate narrative with story and in so doing have ignored substantial differences that exist between them 3) despite the agreement among narrative scholars that story and narrative are distinct, the elements of what constitutes each, or how they are defined, is still being debated despite 4) Game

Studies scholars treating the study of narrative as a concrete and static theory in relation to narrative in digital games.

Thus, extending from these findings, I return to grounded theory, but this time in practice rather than analysis. My goal for Chapter 2 is, therefore, to accentuate the often subtle but significant differences that exist between story and narrative and to show how those differences can help us better understand how narrative and story are constructed in digital interactive narrative games. Furthermore, by highlighting the distinction between story and narrative outside of digital interactive narrative games, I am also able to point to a number of ways in which narrative and story function differently in digital interactive narrative games. Primarily, I argue that the player agent who interacts with a digital interactive narrative game holds both the position of the narrator agent and the character agent simultaneously and that the narrative structure of digital interactive narrative games is more directly affected and developed by narrative-space and story-space.

The methodologies employed in Chapter 3 are similar to those used in Chapter 2. In my survey of social agency theories, I began by selecting which schools of thought within the

extensive study and history of social agency theory would be most lucrative to analyze. Within the confines of this work, accounting for the total of any one school of thought within social agency theory would be impractical, and by extension, accounting for all of social agency theory would be entirely impossible. Given this, I limited my analysis to agency theories found within the following schools of thought: determinism, compatibilism, structuralism, functionalism, existential and interpretive phenomenology, and science and technology studies. Next, I narrowed down each school of thought to two representative scholars whose theories would accentuate the fundamental ideals found within each school. Again, the work that remains to be done in comparing social agency theories with agency in games remains extensive. This volume seeks only to begin the work, cursory as it may be.

Having selected the schools of thoughts and the representative scholars from amongst each school, I then guided my close readings with four questions: How is agency defined within this school of thought? How is it constituted? How is it enacted? Is it static or variable, and if variable, in what capacity? These questions were precise enough to allow me to identify clusters of similarity and difference between the schools of thought, but also broad enough to allow for a comparison of these agency theories with those found within Game Studies and in specific relation to digital interactive games.

As I compared the social agency theories one against the others as well as with those theories of agency found within Game Studies, I identified four important observations: 1) though agency is frequently equated with the capacity to take action in Games Studies, this is rarely the case within social agency theories 2) within social agency theory, agency is generally considered to be a fluid method of making and remaking meaning, whether as a process of

forming an identity, combatting power structures, communicating across barriers, or engaging in social acts 3) while meaning making can lead to action, agency is also understood as an expression of selfhood within the field of social agency theories 4) though these last two aspects of agency can and do exist in digital games, they have not been given proper consideration or attention.

CHAPTER 2

AGENTS IN SPACE AND TIME: FUNDAMENTALS OF NARRATIVE DESIGN

The story of Cinderella has been told countless times. Those of us raised within Western culture and heritage know it by heart. We can tell the story ourselves: an orphaned girl, a wicked stepmother and two wretched step-sisters, a ball, a fairy godmother, a glass slipper, a wedding, a happily ever after. We have also been told the story over and over in countless mediums: picture books, novellas, poems, theatre, musical, animation, radio play, painting, collage, film, and so on. That is the thing about a great story: it can be told countless times and in myriad ways and still be recognizable. At least, that is the common perception - we believe that a good story is adaptable because it is so timeless, so classic. In reality, however, all stories are this way, not just timeless classics like Cinderella. This is because story, as a product, exists both within and outside of narrative confines.

Narrative, by comparison, is the form a story takes. While story is the product or result of a narrative, narrative is the structure, the mold, or the apparatus through which a story is communicated. It is, therefore, narrative rather than story that possess variability. A narrative can change, take new form, new shape, and structure, while the story it shapes and produces remains intact. The narrative can be visual, textual, or aural. It can be told from a variety of perspectives and in various tenses. In other words, narrative is pliable while story is constant.

It is true that this distinction between narrative and story has not always been understood as a given. Despite this, Marie-Laure Ryan points out that the “mediality” of narrative has been present at least as far back as Plato. By mediality, Ryan means the capacity for a medium to act as a system of communication or as the means of expression (“Narration in Various Media”).

Ryan goes on to suggest that narrative is now “widely regarded by scholars as a discourse that conveys a story.” (“Narration in Various Media”). Despite this general agreement among narratologists, far fewer Game Studies scholars are careful to address it. Instead, much like the lay populous who use the two terms interchangeably and synonymously, so too do game scholars who lack formal training in narrative theory. While the lack of distinction made by Game Studies scholars may be understandable given a lack of formal training in narrative theory, the failure to acknowledge the distinction between narrative and story has subsequently led to troubling and deeply-rooted misunderstandings about how narrative can and does work in games.

For instance, Game Studies scholars such as Eugenio Tisselli argue that interactive narratives, which are process-driven, offer little space for writers to develop plots or characters because games “formally acquire a mobile, emergent, and multidirectional nature” (6). By this,

Tisselli means that traditional narratives are static and linear. While the audience is free to interpret the text, change the speed at which they read, skip pages, and so forth, the written text with which they engage remains the same. Given this, the characters and the plot remain the same. By way of comparison, Tisselli argues that games are multi-directional rather than linear, that the player's input affects the processes of the game and can change its course. By extension, because the player’s interactions can alter the course of play, Tisselli suggests that the game is emergent rather than static. Given this, Tisselli suggests that games and narrative are at odds.

Similarly, Greg Costikyan argues that, “there is an immediate conflict between the demands of a story and those of a game,” and that, in order to provide a good story, gameplay has to be constrained. In other words, the player’s freedom must be sacrificed to maintain the cohesion of the story (“Games, Storytelling, and Breaking the String”). In Costikyan’s view, to blend story

and game is to do a disservice to both forms of media. Espen Aarseth too argues that games cannot be constructed using traditional narratology methods. His argument suggests that “when you put a story on top of a simulation, the simulation (or the player) will always have the last word" (First Person, “Genre Trouble” 52). Aarseth’s concern is that narratives cannot maintain their structure, their logic, or their sense of wholeness within a simulation that necessitates variation. Taken together, the arguments made by Tisselli, Costikyan, and Aarseth suggest – albeit wrongly as this chapter will illustrate – that story is not elastic enough, not malleable enough to maintain coherence or a sense of fulfillment when applied to an interactive system.

In fairness, however, it is worth noting that many narrative scholars equally misunderstand the nuances and complexity of video games. For instance, narrative theorist

Dennis Dutton argues that games are a regressive form of narrative and are comparable to a “child’s tea party with teddy bears” (133). He then adds that video games do not, in any way, improve upon narrative experience beyond their ability to add in “virtual reality effects” (Dutton

133). As such, Dutton fails to account for the complexity of digital game systems and processes beyond the graphical environments they can create. While less condescending in tone, H. Porter

Abbott also argues that games are not narrative. He bases this claim on the argument that, while stories can be understood reflectively outside of narrative confines, they must first be presented from within them. For Abbott, codes and formulas cannot be applied to narrative because narrative necessitates flexibility while code is rigid and fixed. He states, "For there to be any kind of success in narrative, the code and formulas that go into it have to be sufficiently flexible to permit all kinds of variations in the details" (Abbott 59). What Abbott fails to account for is the way a fixed code can generate infinite variability.

What these fundamental misunderstandings reveal is not that narrative and games are incompatible, but rather understandings of each have been oversimplified to the detriment of the both fields of study. The narratologists misrepresent games, and the games scholars misrepresent narrative. The narratologists believe the narrative form loses flexibility when codified, while the game scholars argue the inclusion of a narrative reduces the flexibility of the player to make choices. As it cannot be true that games are too rigid to support narrative and narrative is too rigid to be contained in games, it is worth reconsidering and rectifying the false logic that had led to such a misunderstanding of digital interactive narrative games.

Given these considerations, in this chapter, I investigate definitions for narrative and the components or elements of narrative provided throughout them which may be considered essential. While other scholars, such as Marie-Laure Ryan1, have attempted this work, none have done so specifically in juxtaposition with digital interactive games, nor has this work been done as a means to show how game, specifically, can maintain the components and structure of narrative. Thus, having established various ways of understanding narrative from the perspective of narrative theory, I move to consider how, and in what ways, definitions for digital interactive narratives have oversimplified the concept of narrative in order to account for the interactivity games introduce. In interrogating definitions for both narrative and interactive narrative, I will, thus, be able to clarify narrative aspects for scholars of games and gamic aspects to scholars of narrative. Beyond this clarification, however, I will also be able to show how, given more robust understandings of each, narrative structures in and of themselves contain interactive potential

1 In “Toward a Definition of Narrative.” The Cambridge Companion to Narrative. Ed. David Herman. Cambridge: Cambridge University Press, 2007. 22-36. Print.

that digital games can, and often already do, exploit without limiting either the game or the narrative experience. Using this as a bridge, I move my argument to new theoretical territory in order to argue that while digital interactive narrative games can and do maintain the components of narrative, the structures the narrative takes in digital interactive narratives do differ from more traditional narrative forms in two key ways. First, I argue that, while these games maintain the presence of a narrator agent and at least one character agent, it is the player who acts as each and both. Second, I argue that the narrative structure found in digital interactive narrative games places more importance on space as a narrative element than it does on time, despite the reverse being true in most textual and visual narrative media.

Narrative Structures and Their Components

Though narrative and story are now widely regarded by narratologists to be distinct from one another, there is less certainty about what, exactly, defines narrative. Thus, to begin, it is important to discuss how narrative, in the broadest sense, has been defined by narratologists and how these definitions differ or, in some cases, build upon one another. Such a starting point shows that, despite the long history of the narrative form, a solid definition of what a narrative is, scholars of narrative have yet to agree upon and what its component parts are. Though for narratologists this is an obvious point, it is an understanding that is less clearly acknowledged or articulated outside the field of narratology, and certainly, one that has escaped mention within the field of Game Studies. Additionally, the definitions assembled here clearly address the difference between narrative structures and story products, a nuance that, as stated previously, many Game Studies scholars have failed to address. Importantly, it should be noted that the

review of definitions that follows is not ordered by the date of publication. Rather than acting as a linear etymology, this section, instead, aims to show the breadth of definitions available, starting with those that define few essential elements and moving toward increasingly more complex definitions with more numerous qualifiers. Addressing the definitions in this order will provide the lens by which to show that digital interactive narrative games can be said to employ narrative not only by the simplest but also by the most complex of definitions. Additionally, the variety of definitions helps to identify the numerous elements that narrative theorists argue are essential components of narrative structure, and must, therefore, be identifiable in digital interactive narratives too. In other words, the definitions assembled here act as the backdrop against which to qualify digital interactive narrative games as narrative media – a point which, despite the name of the genre, has been contended by many scholars in Game Studies.

To begin, in 2002, H. Porter Abbott provides what is, perhaps, the broadest and simplest definition of a narrative. In The Cambridge Introduction to Narrative, he states, “As soon as we follow a subject with a verb, there is a good chance we are engaged in narrative discourse”

(1). However, while Abbott’s definition is incredibly simple, it is deceptively so. In his consideration of narrative, Abbott is also keen to draw a distinction between events and actions and warns against conflating the two, stating “‘Events’ is the key word here, though some people prefer the word ‘action’” (13). For Abbott, action is not necessary for a narrative, but an event is.

As an example, he argues that the statement “my dog was bitten by a flea,” a sentence composed in passive voice, possess no intrinsic action but is nevertheless a narrated event.

By comparison, several decades earlier in 1980, Gerard Genette defines narrative in

Narrative Discourse: An Essay in Method as “the representation of an event or sequence of

events, real or fictitious” (890). Genette’s definition makes no distinction between action and events, but by allowing for a “sequence” of events, Genette’s definition extends the parameters of what constitutes a narrative beyond those communicated in Abbott’s definition. Genette’s definition accounts for tense and tense shifts between multiple events, while Abbott’s definition does not. Additionally, by accounting for narrative as a “representation,” Genette’s definition also accounts for differentiation in perspective. To illustrate these expansions, let us return to the story of the dog, the flea, and the dog’s owner used in the previous paragraph. In the narrative structure “the flea bit the dog,” the story is delivered in past tense. The same story could also be told in present tense in the narrative form “the flea bites the dog.” Representation can also change depending on the narrative perspective. For instance, “I saw the flea bite the dog” is a story represented via a first-person narration while “the flea bit the dog” is a story represented via third person narration. Despite these changes in narrative form, the story remains the same: a dog is bitten by a flea. This distinction is important, as it clearly illustrates the difference between narrative and story. Genette’s definition accounts for numerous ways for a story to be constructed through changes in the narrative structure. Thus, despite the fact that Genette’s definition includes more qualifiers, his definition also expands the limits of narrative further than

Abbott’s definition does.

Writing in 1978 in Story and Discourse: Narrative Structure in Fiction and Film,

Seymour Chatman’s structuralist theory stipulates that “each narrative has two parts: a story

(histoire)…and a discourse (discours)” (19)2. While discourse, as Chatman states, is “the

2 Chatman also discusses the definition for narrative texts created by Russian formalists. In this definition, a narrative contains a fable, understood as “the set of events tied together which are communicated to us in the course of the work,” and a plot,

expression, the means by which the content is communicated,” story3 contains more components, including events, which are further segmented into actions and happenings, and existents, which are further segmented into characters and setting. Importantly, Chatman goes on to state that,

“clearly a narrative is a whole because it is constituted of elements – events and existents – that differ from what they constitute” (21). In other words, the narrative structure allows for the delivery of a story, but that story product is composed of actions, happenings, characters, and setting and can be transmitted or expressed through an array of forms that constitute discourse.

Chatman's definition also adds several additional qualifiers or considerations than the definition provided by either Abbott or Genette. To begin, Chatman's structuralist definition includes existents. Existents, according to Chatman, are the characters and the setting, and, as such, are contained in the space of a story product (96). To understand this, we can use the example: “Peter pulled down the sails.” In this example, Peter functions as the character, while the sails function as the items of setting. While we do not know if Peter is docked or out to sea, we can still deduce that he is on a boat. In terms of discourse, Chatman defines this as “the expression, the means by which the content is communicated” (19). He goes on to add that discourse has both a form of expression via structure and a substance of expression via manifestation such as verbal, cinematic, balletic, etc. (Chatman 26). Thus, the structure of “Peter

understood as “how the reader becomes aware of what happened” (19-20). Here, the fable is not dissimilar to the structuralist understanding of story, nor is the plot dissimilar to the structuralist understanding of discourse. 3 It is important to note that, although Chatman suggests stories are contained within narrative structure, that he still makes a distinction between the two. In other words, he does not make the two terms synonymous. That said, Chatman does make story synonymous with content. The content is that which generates the story product, or that which is communicated and be transmitted in any number of structures or discourses.

pulled down the sails” is written in the third-person, past tense while the manifestation is textual.

Chatman’s use of the “events” is also more complicated than either Abbott’s or Genette’s use of the term. While Genette includes events or sequences of events in his definition, Abbott stipulates that actions and events are different, and that narrative only requires events, though actions can be present too. Comparatively, the definition provided by Chatman stipulates that a narrative text requires events and that those events must either by actions or happenings. Thus,

Chatman’s definition does not negate Abbott’s, but rather clarifies and complicates it. Chatman notes that, while both actions and happenings are events that cause a change in state, an action is a change brought about by an agent while a happening is an event brought on by an object that alters the state of the agent (44-45). For example, “Peter pulled down the sails” is an action while “The boat was caught up by an enormous wave” is a happening (Chatman 45). Abbott’s definition would similarly account for the first sentence, “Peter pulled down the sails” as an action. That said, there are important differences between Chatman’s example for a happening

“The boat was caught up by an enormous wave” and Abbott’s example of the actionless event

“my dog was bitten by the flea. Though both examples are constructed in the passive voice, in

Chatman’s example, the wave is an object that alters the state of another object, the boat. In

Abbott’s example, there is one agent, the flea, changing the state of another agent, the dog, and this occurrence is witnessed by a third agent, the narrator. Thus, for Chatman, the narrative distinction between actions and happenings concerns the objects and agents that interact to bring about the event, while for Abbott – who does not address happenings as a narrative component at

all – the difference between actions and events is largely distinguished by a change in voice, from active to passive.

In the definitions provided so far, narrative is discussed as an artifact, something already constructed. By contrast, Mieke Bal’s definition of narrative accounts for a constructor too.

Published in 1985 in Narratology: Introduction to the Theory of Narrative, she claims narrative is “a text in which a narrative agent4 relates (‘tells’) a story in a particular medium, such as language, imagery, sound, buildings or a combination thereof” (Bal 5). Thus, while Abbott calls for a subject and a verb working together to represent an event, and while Chatman’s agent can be understood as the subject experiencing an event, Bal adds in the necessity for an agent which, she states, is “a function and not a person which expresses itself in a language that constitutes the text” (16). The subject, defined by Abbott, is not the same as the narrative agent, defined by Bal.

What Bal necessitates is, put simply, a narrator agent who delivers the narrative. The narrator agent’s main function is to tell the story within the confines of the narrative structure. It does not matter whether the agent is also a character or whether the narrator is omniscient or even trustworthy. It simply matters that the narrative contains a narrator agent. The addition of a narrator agent is important as it signals to the implied capacity of the narrator to perform or act.

Within the narrative structure, this agent’s key means of action are to mediate and reflect upon the developing story. This act of mediation is what allows for the differentiation between

4 The narrative/narrator agent should not be misunderstood as representative of the author. The author/writer and the narrator are entirely different entities. Barthes and Foucault have certainly complicated the idea of the author stating, respectively, that “it is the language which speaks, not the author” and that “in writing, the point is not to manifest or exalt the act of writing, nor is it to pin a subject within language; it is rather, a question of creating a space into which the writing subject constantly disappears.” (Barthes 143 and Foucault 206) Additionally, Bal also draws a distinction between the two, stating that, “several processes are involved in ordering the various elements into a story. The processes are not to be confused with the author’s activity – it is both impossible and useless to generalize about the latter” (8).

narratives that, otherwise, communicate the same story. Thus, by recognizing the narrator agent in her definition, Bal identifies another means to distinguish between narrative and story.

Finally, James Phelan and Peter Rabinowitz take Bal’s definition even one step further.

They define narrative as, “somebody telling somebody else, on some occasion, and for some purpose, that something happened to someone or something” (Phelan and Rabinowitz

“Introduction” 3). In this case, the narrative not only requires an agent – whether the subject agent defined by Abbott or the narrator agent defined by Bal – but also a receiver, a “somebody else,” who receives5 the representational telling of the something that happened (Phelan and

Rabinowitz “Authors, Narrators, Narration” 29). Here, a narrative requires an audience of some kind too. The addition of an audience to the definition of narrative is important too, as it suggests that narratives are always, in some capacity, interactive; they require both a narrator and an audience.

As the definitions assembled here exemplify, theories about narrative and the elements which define it are complex. Far more complex, certainly, than game scholars have acknowledged. In part, this lack of acknowledgment stems from incorrectly conflating story with narrative. Yet, understood as a structure, the concept of narrative becomes far more accessible to the study of games which are, likewise, a structured medium. By reviewing the simplest definition for narrative provided by Abbott, all a digital interactive narrative would require is a subject followed by a verb. Few would dispute that games meet these criteria. The question then

5 Certainly, the reader can do more than passively receive the story product or observe the narration. Readers can, of course, interpret, and as Stanley Fish notes, “interpretation is not the act of construing but the art of construction” (327). Through interpretation, readers make meaning. While the reader constructs the story via interpretative acts, it is important to maintain a difference, still, between the player and the reader. The player’s interpretive acts not only affect how she understands and constructs the meaning of the story product, but also how she will move through the game’s interactive narrative structure, and, subsequently, how the player interacts within that structure to produce the story which can then, again, be interpreted.

becomes, is there an element of narrative structures, as defined above, that games are unable to account for or support? I argue that there is not.

To illustrate the capacity for games to account for the narrative elements illustrated above, we can use Mass Effect 2 as an example. Looking back to Abbott’s definition, the game must include events. This is easy enough to account for. Mass Effect 2 is a space opera in which the main character, Commander Shepard, must assemble a team of fighters to take down an enemy which threatens to destroy the galaxy. The assemble of that team, without regard for other plot points, signify an array of events apparent in the game. Per Genette’s definition, the game must also account for tense and perspective. Here, the game also succeeds. Mass Effect 2 begins with Commander Shepard’s ship, the Normandy, being destroy. In the next scene, Shepard wakes up on an operating table. It is clear to the player that some time has passed between these two events and that the opening scene actually occurred in the narrative's past. This moment not only establishes the present tense the game primarily maintains, but also signals to moments in which the game reveals experiences that happened previously or in another tense. In terms of perspective, the game offers two. First, the game is presented to the player via a third-person perspective. Though the player controls Shepard, the player does not see through Shepard’s eyes.

That said, because the player does control Shepard, the player also maintains a first-person perspective. This is an idea to which I will return when discussion the player’s relationship to both the narrator agent and the character agent. Per Chatman’s definition, the game must be able to account for characters and setting. As already stated, the game meets these criteria – Shepard is a character and the setting is the galaxy in which Shepard exists. Of course, Chatman also identifies the need for events which can either be actions of happenings. Acting as Shepard, the

player can certainly bring about events, which qualify as actions. For instance, the player can engage in combat and kill the enemies she encounters. However, the player also experiences happenings, such that events are brought on by objects that change the state of the player. For example, the player frequently navigates unstable terrain which can cause the player to fall.

Here, the event happens to the player, rather than the player initiating the event. Finally, taking

Bal’s definition and the definition provided by Phelan and Rabinowitz, the game must also account for a narrator agent and an audience. While it is easy to establish the player as the audience, I also argue that the player is the narrator of the experience too. To support this argument, however, I must first engage with the issue of interactivity. For the moment then, we can say that, per the examples above Mass Effect 2 meets the criteria for narrative as established by the definitions assembled in this chapter. Where it is not yet clear how it does so, is due to the introduction of interactivity to the narrative experience. Given this, we can now turn to the definitions of interactive narrative which have been written by game studies scholars in order to consider how the introduction of interactivity alter or complicates the understandings of narrative provided thus far.

Interactive Narrative Definitions

Just as the definitions for narrative vary in terms of their complexity, so too do the definitions for interactive narrative. The definitions for interactive narrative, however, not only attempt – to varying levels of success – to clarify the components required to support a narrative, but also attempt to expand those definitions in order to account for the possibility of interactivity.

Largely, this is because games as a medium require two elementals that other narrative mediums

do not: a system of code6 and an interlocutor, or player, who interacts with that code.

Furthermore, these definitions for interactive narrative also attempt to account for the addition of interactivity without negating the parameters of what constitutes a narrative. This has not been easy work, especially when and where Game Studies scholars fail to account for the difference between story and narrative or fail to account for those narrative elements established by the more complex understandings of narrative, such as those offered in this chapter. Thus, where the definitions for interactive narrative have failed to account adequately for both narrative and interactivity, they have typically failed due to misunderstandings or oversimplification. In some respects, game scholars erroneously conflate narrative with story, treating both as a product that is generated through interaction with a medium. In other respects, the definitions err by employing the simplest understanding of narrative, which, although not technically incorrect, fail to account for the nuance and complexity that would most help game scholars understand where opportunities for interactivity exist, by definition, within narrative structures. Still, these definitions maintain utility in so far as they help document where thoughts regarding interactive narrative stand, and how game studies scholars have, up to now, approached the potential for games, as codified structures, to deliver story products.

For instance, in 2010, Mark Riedl defines interactive narratives as “an approach to interactive entertainment in which a system attempts to tell a story to an interactive participant”

(“A Comparison of Interactive Narrative System” 1). Riedl’s definition is not dissimilar to those of narratologists: his definition views narrative as an approach rather than a product, includes

6 Code here is understood as computer code, or the instruction in a computer program written by a programmer in a programming language. Code here is not meant to indicate or draw a parallel to semiotics. While that parallel certainly can be drawn, and is worth consideration elsewhere, it is outside the realm and intent of this chapter.

story as the outcome of the narrative approach, and addresses the need for a teller of the story.

In Riedl’s definition, however, the teller of the story is not defined as a narrator agent, but instead as a system. Importantly, Riedl’s definition destabilizes – though this was likely not his intention – the importance of a narrator agent. One might also interpret, from Riedl’s definition, that the system acts the narrator to the player. While Riedl is not wrong that the game system supports the interactive narrative structure, he is misguided in his assumption that the system replaces the narrator agent. As this chapter will discuss later, the player functions as the narrator agent and, as such, exists outside the systems presents in digital interactive narrative games. It would, therefore, be more precise to consider the game’s system as an addition to, rather than as a replacement of, what defines an interactive narrative.

Importantly, Riedl’s definition does address the addition of interaction. Riedl states,

“Interactive narrative allows the user to make decisions that directly affect the direction and/or outcome of the story being told by the systems” (A Comparison of Interactive Narrative System

1). Thus, according to Riedl, narrative systems deliver a story, but the player is an active agent in its construction7. Just as previous definitions stipulated the inclusion of a character or subject agent and narrator agent, Riedl stipulates the need for a player agent.

7 This active construction of interactivity is distinctly different from the activity that takes place between a reader and a text, or between an audience member and a film, or between an audience participant and a performance. This does not negate or challenge the importance of Reader Response theory or audience participation, but the type of activity described here is different by several significant degrees with respect to interactive games. In the words of Frank Rose, “video games are both nonlinear and interactive. They have different outcomes depending on the actions players take at crucial inflection points” (104). While a traditional narrative structure presents a singular and fixed path, games do not. While readers and audiences interpret the story product differently, the singular and fixed path of the text does not change. In games, the path changes with each playthrough of the game. Different actions can be taken and these actions yield different results and even, potentially, totally distinct story products. They contain a multiplicity that traditional narratives do not.

In The Art of Immersion, published in 2011, Frank Rose echoes Riedl’s claims. Just as

Riedl’s definition stipulates the need for an active participant, Rose also focuses on the importance of participation with respect to interactive narratives. He states, “The author8 starts the story; the audience completes it. The author creates the characters and the situation they find themselves in; the audience responds and makes it their own” (88). Extending from this, however, Rose also suggests that the idea of “audience” is outdated and should be replaced with the term “participants” (6). For Rose, the term "participants" is more apt than "audience" because it accounts for the interactive opportunities narrative structures provide. While Rose was talking about digital media in general, rather than narrative games specifically, his observation is important with respect to interactive narrative games. Rose is arguing that, as participants, the audience maintains a kind of agency too. As agents, the participant aids in the construction of meaning the story ultimately takes (57). Importantly, this shift highlights the difference between the type of relationship a player maintains and the agency they hold within an interactive narrative game compared to that of a reader, a listener, or an audience member. While reader response theory might argue that the audience constructs meaning, within interactive narrative games, the participant also helps constructs how the story product will be developed. Thus, the participant has a different type or level of agency in relation to the story and its development than the audience of other narrative forms. The participant, in the case of digital interactive narratives, can interact with and affect the experience and outcome of the game. In other words,

8 Rose’s use of the term “author” here is unfortunate. The concept of authorship has certainly been destabilized elsewhere, most famously by Roland Barthes in “The Death of the Author.” Despite this, however, Rose’s focus on the importance of the audience as participants in the construction of meaning is not only relevant in terms of Reader Response theories, but also within the extended conversation about interactive narrative games more specifically.

what a player does within the narrative structure can alter the story product in literal and visible ways.

Much like Riedl, Jane McGonigal contends that the relationship between a narrative and a participant or player is one of meaning-making. Writing in 2011, McGonigal states that a player’s interaction relies on narrative because it “sets the stage for meaning. It frames the player’s actions” (101). Again, McGonigal helps establish that the relationship a player holds with respect to an interactive narrative is different from the relationship the audience holds with respect to other narrative forms. Yet, while Riedl and Rose help establish how the participant engages with the narrative structure to affect the developing story and the final story product,

McGonigal focuses instead on how the narrative structure informs the participant and prepares the participant to make choices. In this case, the narrative structure delivers context and information, but the player has some level of ability9 to decide what to do with this information.

What the player opts to do may differ from player to player and, by extension, the resultant story products those players develop within the game’s narrative structure may also differ. Thus, as with Riedl, for McGonigal, the interactivity that occurs between the narrative structure of the game and the player is less about what the player can do in terms of the game's mechanics and more about how the player interacts with the narrative structure to derive and create meaning.

Marie-Laure Ryan’s criteria for narrative also extend to account for interactivity.

Published in Storywolds across Media in 2014, Ryan claims that a storyworld must contain

9 It is true that the player does never ever maintain full freedom of choice within the narrative structure of the game. As a closed system, the game and its rules will only make certain options available to the player. This is an idea to which I will return in the next chapter about agency theory in relation to games. For now, however, it is worth noting that, even if the options presented to the player are minimal, the player is still permitted to interact within the narrative structure to produce a story product in ways that a traditional reader or audience member cannot.

existents, setting, physical laws, social rules and values, events, and mental events

(“Story/Worlds/Media” 34-36). Per Ryan, these criteria apply to all narratives, including text- based, visual, performative, and yes, interactive ones. That said, Ryan also draws a distinction between intradiegetic elements and extradiegetic elements (“Story/Worlds/Media” 37).

Intradiegetic elements are part of the story world, while extradiegetic elements are not.

Importantly, Ryan notes that, in computer games, “the menus that offer the player a choice of actions and the statistics that report the player’s level of achievement” are extradiegetic elements

(“Story/Worlds/Media” 39). Importantly, theses extradiegetic elements are often present in the game to support the intradiegetic elements. As I will argue later, these menus are frequently the means by which the player changes the level of their interaction and position within the game and the storyworld, or to be more precise; these menus helps the player agent transition between the roles of narrator agent and character agent. Thus, as Ryan notes, these extradiegetic elements are par for the course in digital interactive narrative games. She also goes on to note that, rather than being unique to digital interactive narrative games, extradiegetic elements can be found in all narrative forms. For example, the narrator agent found in traditional narrative can be considered intradiegetic if the narrator agent is also a character agent, but the narrator agent can also be extradiegetic or removed from the storyworld, as a third-person omniscient narrator often is. Given this, Ryan’s observation supports the notion that game systems and their menus do not limit the potential for narrative structures to exist in games. Rather, they function as a natural element of narratives, and in the case of digital interactive narrative games, actually strengthen rather than weaken the player’s connection to that narrative and the means to interact with it.

Despite writing much earlier, in 1997, Janet Murray, author of Hamlet on the Holodeck, provides an understanding of digital interactive narratives that is more nuanced than many that came after. Murray’s definition also addresses the necessity for interactivity in digital narratives and multimedia, but argues that digital environments have four key properties: they are

“procedural, participatory, spatial, and encyclopedic” (71). The first two properties, she notes, are the elements of interactivity. The system of the game is codified and procedural but requires the participation of a player or, in Murray’s words, an “interactor” (152). Thus, while Riedl and

Rose have identified the player as a participant, Murray prefers to identify the player as an interactor. The use of this identifier, importantly, endows the player with a stronger sense of involvement. While a participant can be passive, an interactor is more inherently embedded in the construction and creation of an experience or product. By using this term, it is likely that

Murray is trying to draw a clear distinction between the ways a player interacts with a narrative and the way a reader interacts with a narrative. She writes, “We could perhaps say that the interactor is the author10 of a particular performance within an electronic story system, or the architect of a particular part of the virtual world, but we must distinguish this derivative authorship from the originating authorship of the system itself” (153). In other words, the interactor is more than a participant, but less than an author, as the interactor can only act within the confines and rules established by the system itself.

10 Murray’s use of the word “author” is no less irksome that Rose’s use, and both certainly signal to the wealth of information that could be gained if Game Studied scholarship paid more attention to concepts of narrative theory. That said, one might argue that Murray’s use of "author" is pointing more to the “active audience” theory of media and the audience’s ability to make meaning than to the authorship of a writer of a text. Given this, the distinction she is trying to make could be between the idea of an active audience and the ways in which a player can “instigate material change through action” in games (Galloway 3).

Regarding the other two properties outlined in Murray’s definition, she states that the spatial and encyclopedic are elements of immersion. The encyclopedia characteristic, she notes, holds real narrative promise. She says, “Because of the efficiency of representing worlds and numbers in digital form, we can store and retrieve quantities of information far beyond what was possible before” (Murray 83). In other words, a procedural narrative can draw from a wealth of potentiality based on its storage capacity. A printed, linear text has very limited storage space compared to a digital one. The same was true of film before digital recording processes were made available. The storage capacity, in relation to games, not only increases the ability for interactivity but also the potential complexity of the narrative form. As one example of how the narrative form is altered, Murray addresses the characteristic of space, an idea to which I will return later in this chapter. In relation to space, she writes that digital environments are

“characterized by their power to represent navigable space.” She continues, “Linear media such as books and films can portray space, either by verbal description or image, but only digital environments can present space that we can move through.” (Murray 79). In this way, Murray suggests the digital narratives can extend and utilize the elements of textual and visual narratives such that they become more, rather than less, present.

What these definitions show is that the key addition to interactive narrative games when compared to more traditional narrative forms is that of another agent – the player – who interacts with the narrative structure to create a story that holds meaning to them. Additionally, digital narratives privilege and utilize space differently, or at least at an increased capacity, then do traditional narrative forms. Importantly, neither of these qualities negate narrative as a form that can be implemented into games. These qualities do, however, change where narrative emphasis

is placed and how the form functions. Just a visual narrative forms function differently than aural narrative forms, so too do interactive narrative forms function in a unique way.

Given all these considerations, own understanding of digital interactive narrative takes all the definitions assembled here, both for narrative and for interactive narrative, into consideration, but also extends them such that new understandings for narrative in interactive games becomes possible. My definition, and the definition implied throughout the rest of this text, is one in which the player interacts meaningfully and continually both as a narrator agent and as a character agent within a well-developed narrative structure and storyworld for the duration of the gameplay experience in order to produce a unique narrative experience and story outcome that is both logically and emotionally complete. To begin, this definition supports the theory that narrative and story are not synonymous or interchangeable words or forms. As already noted, this is an important distinction to make as it opens new ways of understanding how a player interacts within a narrative structure to produce a story. If we understand narrative as a structure or a method through which a story is composed, assembled, and produced, we can better understand how to construct a narrative structure that allows for a participant to interact and construct a story product that has meaning to the player. Subsequently, this definition attempts to account for meaning by attributing it both to logic and to emotion. Additionally, my definition extends from the understanding that, while stories are delivered via narrative structure, they are not exclusive to that narrative structure and, as such, any narrative structure can be built to account for potentiality. Importantly, however, this definition also extends the understanding of interactive narrative in two new directions. First, this definition recognizes the different functions and purposes for both a narrator agent and a character agent and suggests that the player agent

functions as both. Second, my definition takes into account the need for events, both as actions and happenings, that occur within a storyworld and which must be realized both temporally and spatially.

While these two new considerations build on the definitions already outlined in this chapter, they deserve further analysis and validation. Thus, in the second half of this chapter, I will move my argument to new theoretical territory in order to show how the player agent of digital interactive narrative games mediates both the role of narrator and character and how this mediation produces opportunities for interactivity, rather than limiting the potential for interactivity, that Game Studies scholars have yet to acknowledge. Secondly, I will analyze how interactive narrative games utilize space and time differently as narrative elements that most textual or visual media and will show how the player agent’s relationship to the developing and resultant story are affected by the change in prominence of these elements.

The Narrator Agent and Character Agent

Before we can analyze the function of the player agent, it is vital to have a clear perspective and understanding of what narrative agents and character agents are and how they function. This requires more detail than the definitions for narrative provided in this chapter provide.

Put simply; narrator agents mediate the story for the audience as it is being developed while character agents act on and make choices within the developing story itself.

Comparatively, character agents are much less complicated in form, function, and construction than narrator agents. Character agents are present in the story to act on stimuli and be acted upon

by others (Phelan and Rabinowitz 34). The purpose, then, of character agents is to make choices and overcome conflict, to bring about and experience story events.

For instance, Cormac McCarthy’s The Road has two main character agents, the boy and the man, neither of whom are given names. In The Road, the man and the boy attempt to survive in a bleak and near-desolate post-apocalyptic landscape. As such, their choices and the actions they take are most centrally about survival against the elements and against other hostile humans.

By comparison, in Disgrace, by J.M. Coetzee, David Lurie, a South African professor of English with a penchant for sleeping with his students, is the main character agent. As a character,

Lurie’s main action is to deal with his moral choices and the repercussions thereof. In another novel, The Gathering, by Anne Enright, a middle-aged woman, Veronica, struggles to face her family and its history after her brother commits suicide. In each of these examples, the characters make choices and take action based on stimuli related to the circumstances in which they find themselves. The choices they take and the actions that result from those choices give the narrator agent something on which to report, interpret, and evaluate (Phelan and Rabinowitz 34). In other words, characters function to provide the narrator agent something to mediate for the audience.

By comparison, narrator agents are far more complex. Because the narrator agent can document, comment upon, and situate events in ways a character agent cannot, the narrator agent can be understood as “an instrument, construction, or a device” of the narrative rather than an entity of the resultant story the narrative produces (Abbott 68). The narrator agent is, thus, a mediator between the story as it is developed and the audience who witnesses its development.

Ultimately, the narrator agent functions to communicate and mediate the events, actions, or happenings of the character agents. Given this, “one of the most important choices an author

makes is about the kind of narrator they employ” (Phelan and Rabinowitz 33). In traditional narratives, an author must choose whether to use first-, third-, or (in rare cases) second-person narration. The author must also make choices about how the narrator will construct the temporality of the story being produced. The narrator can deliver the story in present tense, past tense, future tense, or in some combination thereof. Other choices include whether to make the narrator agent trustworthy or not, whether the narrator will be dramatized (a character through whom the narration passes) or undramatized (an outside agent who is often omnipotent), and so on (Lacey 109 and 113). These choices, Nick Lacey notes, are “integral to the narrative[s] development” (115). These choices immediately situate the reader in distinct and important ways with respect to how the developing story will be mediated and how it will be received and interpreted by its audience.

For example, both The Road and Disgrace utilize third-person narration, but that is largely where their similarities end. In The Road, the narrator agent is undramatized and omniscient. Narrating both actions and private thoughts, the narrator agent recounts, “They squatted in the road and ate cold rice and cold beans they’d cooked days ago…He hardly knew the month. He thought they had enough food to get through the mountains but there was no way to tell” (McCarthy 29). Later the narrator agent observes, “the boy found toys he’d forgot he had” (McCarthy 35). Both examples reveal the narrator agent’s access to the character agents’ thoughts: the audience is aware the man is thinking about food though he does not admit it through dialogue, just as the audience is aware the boy had forgotten he’d had certain toys despite the boy never saying so aloud. In each case, the narrator agent has omniscience.

By comparison, the narrator agent in Disgrace is undramatized but not wholly omniscient. In this case, the narrator agent only has omniscient access to the thoughts of the main character agent, David Lurie. For example, when Lurie asks his daughter’s girlfriend how his daughter has been doing, the narrator agent mediates the response as such: “Bev Shaw responds only with a terse shake of the head. Not your business, she seems to be saying” (Coetzee 104).

The inclusion of “seems” denotes speculation on the part of narrator agent, who, without access to Bev Shaw’s thoughts, cannot be sure this is what Bev intended to communicate. Furthermore, the narrator agent only recounts scenes in which David Lurie is present. Thus, the narrator privileges the perspective of Lurie over all other characters despite maintaining third-person perspective.

In both of these examples, the narrator agent is separate from the character agent. In other words, the narrator agent is not one of the characters in the story. As such, the distinction and the purpose of each agent are easy to decipher. Still, even in first-person narration, when the narrator agent also exists as a character agent in the story, there is a “noticeable difference between the narrator’s style and that of the actors” (Bal 9). Remembering that a narrator agent mediates and informs story while character agents act and respond to event, finding the distinction between agents in a first-person text becomes rather simple. Take this example from Ann Enright’s The

Gathering for instance:

Michael’s father was an artist and his mother was something else. I wasn’t used to

that either - most of the parents I knew were just parents - but he had this semi-

famous father and this mother who made appointments and met people and

dressed up to go out, and so he had all of that dragging behind him. It was hard

for him to know what he was going to do when he grew up, because he had been

grown up, at a guess, since he was ten years old” (81).

Here, the narrator agent functions to mediate both how the audience understands the character agent Michael as well as how the audience understands the relationship between the two character agents, Michael and Veronica. Veronica, as a character agent, is static in this passage; she takes no actions and makes no choices. That said, the narration of this moment by the narrator agent alters both the character of Veronica and that of Michael.

In each of these examples, the character agents exist inside the developing story while the narrator agent functions as a narrative tool that informs the audience's access to and understanding of the story. In The Road, Disgrace, and The Gathering, the narrative agent helps mediate the relationship intended between the reader and the character agent. As Abbott notes, a

“story is always mediated” (20). The narrator agent determines how a text will be mediated and, as a result, advances a specific understanding of the story for the reader.

Importantly, it is not only text-based novels that contain character agents and narrative agents. Films, photographs, paintings, plays, dance, etc. also contain them. In films and photographs, the camera acts as the narrator agent, while the entities depicted in the frame function as the character agents. Or, in the words of Stephen Heath, “frame space…is constructed as narrative space” (36). In paintings, perspective plays the role of the narrator agent11 while the entities depicted function as the character agents. Narrative songs contain both

11 Werner Wolf identified three types of narratives that are composed in static framed images: works that communicate a single moment within the image (monophase), works that communicate several moments within a single image (polyphase), and works that show a series of events in several images (422).

too. The Beatle’s song “Rocky Raccoon,” for instance, features the character agent Rocky

Raccoon and an unnamed third person narrator agent who explains Rocky's unfortunate demise at the hands of his rival (The Beatles). The key point here is that narrative agents and character agents are elements of the narrative form in general and are not elements unique to any given medium. As a result, narrative agents and character agents are also elements of interactive narratives.

Thus, the difference between the narrator and the characters can be expressed through a difference in purpose. So, what of the player agent? What is its purpose and how does it differ from that of the narrator agent and the character agent? For this, let’s first examine an example of an interactive narrative game.

The Player Agent

Banner Saga, a game developed by Stoic, is an interactive narrative game. It is also a turn-based, role-playing game (RPG). Banner Saga tells the story of a Viking epic in which the human race and race of giants, known as Varl, must join forces to defeat a great foe that threatens the existence of both their worlds. These characters are situated geographically on opposite sides of a map and for much of the game are unaware that they are both at war with the same earth-destroying enemy. During combat or fight sequences, the player acts as various characters simultaneously. Each character is given rotating “turns” in battle, much like pieces and opponents in a game of chess. Outside of these fight sequences, the player switches between and inhabits numerous roles, but only inhabits one at a time. While in each role, the player is presented with choices. These choices have ripple effects that alter the final story outcome, but

also how the various characters relate to one another as the story develops. Based on these choices the giants and humans may conquer the foe, or they may be defeated by it. Additionally, the choices the player makes while inhabiting various characters affect which characters live and which die. As a result, there are numerous ways for the story to take shape.

It is easy, using this example, to decipher the narrative, but less easy to separate or identify the narrator agents from the character agents. I argue that this is because the player agent functions both to inhabit the role of the character agent (or possibly character agents) and the narrator agent and to mediate the distinction between them. Rather than being akin to first-person narration in traditional text-based narratives in which the character agent and narrator are constructed to be the same entity, in digital interactive narrative games, the player holds both positions even while they remain distinct. As I will discuss later, the boundary between the two roles, though both roles are held by the player, is generally communicated or established through the game’s interfaces and systems. Additionally, I argue that the player’s ability to inhabit, mediate, and move between these two types of agents is one means by which the narrative structure becomes interactive within digital interactive narrative games. Put another way; the narrative becomes interactive based on the player agent’s need and capacity to fluctuate between the roles of the narrative agent and the character agent.

Thus, while narrative scholars such as Bal, Phelan, and Rabinowitz note that there is an important distinction between the narrator agent and the character agent, I am arguing that, in the interactive form of games, it is the player agent who acts as both. Thus, while both the narrator agent and character agent still exist and can be identified as distinct elements in interactive narratives, it is the player who interacts with and inhabits both roles. In this capacity, the player

agent must be able to act, make choices and engage in conflict as a character agent would while also mediating, reflecting on, and interpreting the story development as a narrator agent would. D. Fox Harrell and Jichen Zhu state that a player’s agency, or the power imbued to an agent, “centers upon the intentionality” of the player’s actions and that intentionality is exercised after critical thought is actualized through action (45). Within digital interactive narratives, the player agent’s intentionality shares two foci; as narrator agent, the player intentionally guides the production of the story by interacting with and mediating content that exists outside the storyworld, and as character agent, the player intentionally acts and reacts to objects, characters, environments, and events present within the storyworld.

A review of several digital interactive narrative games shows that the player interacts with the game both on the level of character agent and narrator agent. In The Banner Saga, for example, the player acts as the character agent while inhabiting the role of a single character on either side of the map (Stoic). For instance, the player agent will act as the human character,

Rook, leading a caravan of humans west in a desperate attempt to avoid the evil Dredge threat.

Alternately, in other moments in the game, the player agent will act as Hakon, one of the giant Varl, who is escorting a caravan eastward. While inhabiting these roles, the player is presented with scenarios and choices and makes those choices as whichever character agent the player is currently prompted to inhabit. Yet, while the player makes these choices from the position of a character agent, the player is also simultaneously aware of information that the character agent they inhabit is not privy to. For instance, while the player knows there is a caravan headed east and another headed west, the character agents do not know this. As a result, even while inhabiting a character, the player maintains a type of narrative remove from the

characters. So, while the choices presented to the player are oriented to the character agents, the player also maintains an outside narrative perspective too.

That said, the player’s access to narrative information is also limited; the player is not an omniscient narrator. For instance, while inhabiting the character agent role of Rook, the player must resolve conflicts with other traveling caravans, must distribute rations, negotiate scuffles between the men, decide when to stop and rest, and keep morale high. In these moments of choice, the player cannot anticipate how her choices will affect the overall outcome despite being privy to more information than the characters possess. Thus, the actions taken by the player as a character agent maintain dramatic appeal and suspense.

The movement between narrator agent and character agent also creates a unique type of interactivity. Importantly, the player may not be aware of this interactivity or her mediation and movement between the role of the narrator agent and character agent, despite it being present. In other words, the distinction between the two roles is more visible at the level of design than at the level of play12. Furthermore, the distinction is generally only made apparent to the player through the use of a game’s interfaces and the use of drama management systems. When designed well, these interfaces help the player to transition from role to role smoothly and to engage in both levels and as both agents, often simultaneously. The level of interactivity, then, is an element of design present in the narrative structure but is often less apparent in the act of play or the production of a story. Still, the narrative structure is not only interactive in the capacity that it provides the player with mutable ways to produce the story but is also interactive in the

12 One might argue that if players were aware of these two roles, they would easily be able to see the distinctions between them and would be able to recognize how the game’s systems and interfaced help them mediate both roles. However, players have not been groomed, to think this way or to make these distinctions, and therefore rarely do.

way it engages the player on the level of narrator agent and character by presenting the player with interfaces designed to meet the needs of each role.

In Mass Effect 2, the player also switches between the role of narrator agent and character agent throughout gameplay (Bioware). In this interactive narrative, the player agent inhabits the role of Commander Shepard. At the start of the game, Shepard’s spaceship is attacked and, though Shepard saves the crew, she dies in the attempt. Her body is rescued and revived by a paramilitary group known as Cerberus. Now begrudgingly indebted to Cerberus, Shepard agrees to spearhead a Cerberus mission to investigate a galaxy-threatening enemy known as the

Reapers. In addition to spearheading the mission, Shepard must assemble and gain the trust of an intergalactic team of highly trained and highly temperamental crewmembers. Thus, the player agent acts as the character agent Shepard, traveling around the galaxy to locate, recruit, and gain the trust of several alien species all whilst investigating and holding the Reaper threat at bay. That said, the game provides the player agent with information the character agent Shepard is not privy to. For instance, the player agent must make choices that carry moral weight throughout the gameplay experience. When the player agent makes a choice, they are scored on a scale of morality with good choices equating to a "Paragon" choice and bad choices equating to a

"Renegade" choice. While the player agent can see how the choices are scored, the character agent Shepard cannot. In these moments, the player agent functions as the narrator agent. The morality scale acts as a reflective element that informs the player agent as they move through the story experience. Functioning as the narrator agent, the player agent can think reflectively about past choices and how they were scored before making news choices.

Finally, the interactive narrative game Kentucky Route Zero illustrates how a player agent can simultaneously inhabit both the roles of character agent and narrator agent (Cardboard

Computer). In Kentucky Route Zero, Conway, a delivery truck driver in Kentucky, must deliver a package to an address he cannot find, but which he is told exists along Route Zero. On his journey to find “the zero,” Conway eventually picks up additional travel companions: Shannon, a local television repair woman whose migrant parents once worked in the coal mines in the town;

Ezra, a young boy who has lost his parents; and Junebug and Johnny, two musicians who entice

Conway, Shannon, and Ezra to come with them to a gig. While such a synopsis might lead one to imagine the player agent functions as Conway, this is only partially true. There are multiple protagonists in the game, and the player agent inhabits each of their perspectives, sometimes simultaneously, throughout gameplay. The player agent switches between characters via a dialogue selection menu. When the dialogue selection menu is displayed, rather than presenting the player agent with a set of dialogue options for a single character agent, as would be typical of most digital games, the player agent is, instead, presented with dialogue options for multiple character agents. As a result, the player agent must select not only what is said, but also which character agent speaks. Furthermore, as a result of this choice, the player agent functionally chooses which character agents will be developed and which will not. These choices are narrative choices and, when making them, the player agent functions as the narrator agent operating outside the context of and reflecting on the context of the unfolding story.

Time and Space in Narrative and Story

The collapse of the narrator agent and the character agent into the player agent within a digital interactive narrative system is not only relevant because it makes the relationship between the narrator agent and the character agent interactive, but also because it alters the spatiotemporal construction of narrative.

While the temporal construction of narrative has been the focus of a great deal of narrative theory, spatial construction has been given far less consideration. William Labov and

Joshua Waletzky argue that the temporal sequence of a narrative is an essential trait of the narrative form (“Narrative Analysis”). Ricoeur too states that the “ultimate referent” of narrative is its temporality (“Narrative Time” 169). This is largely because traditional narratives signal time in a variety of ways, including verb tense, adverbs, order (the arrangements of events in the story), duration (the time it takes to read the narrative compared to the time the story events lasted), and frequency (the number of occurrences in the story and the number of occurrences that are narrated) (Chatman 63). The temporality of narrative texts is also always two-fold: there is story-time, and there is narrative-time. As a result, time is always known and apparent to the reader. Characters agents express time, as does the narrator agent. Thus, while not all narratives express space explicitly, they all reveal time.

That said, story-time and narrative-time work differently, just as character agents and narrator agents do. While the expression of story-time can generally be taken at face value, the expression of narrative-time is far more deceptive; even when a narrator agent uses present or future tense, the story and its outcome exist separately from the time and space of narration

(Genette “Frontiers of Narrative” 895). By virtue of its capacity to deliver a story to an

audience, the narrator agent, as device of the narrative structure, develops the story toward a known or determinate outcome. In other words, the story must take place before it can be narrated. As such, narration is retrospective and, by extension, narrative-time is always reflective, always looking back, regardless of how it is presented to the audience. Furthermore, because the narrator is “almost always outside the story,” the narrator agent is always also temporally and spatially distant from the story, neither there in space or in time, but instead reflecting back on both (Chatman 55). Evidence of this can be seen in The Road,

Disgrace, and The Gathering.

The Road is a linear narrative in which one story event logically leads into the next in logical and temporal order. In other words, the narrative order is sequential. In this novel, the narrator agent also delivers the story through past tense. Given the past tense of the narration, it is clear to the audience that the narrator agent is reflecting back on events that have already transpired. At one point the narrator agent says, “Long before they reached the coast their stores were all gone. The country was stripped and plundered years ago and they found nothing in the houses and buildings by the roadside” (McCarthy 181). Here, the distance in time between the narrator agent and the story being delivered is clear. The narrator agent reveals that, though the boy and the man have not yet reached the coast and still won’t for some time, that they will eventually reach it. This is a narrative certainty communicated by the narrator to the reader, but which remains unknown to the character agents.

Despite the narrator agent’s position and perspective and the audience’s awareness of the narrator agent's use of past tense, the audience still experiences the linear narrative as if it is happening in the present tense. Unlike the narrator agent, the audience has no awareness of what

will transpire beyond what the narrator agent provides. For example, in one of the more intense moments of this post-apocalyptic story, the man and the boy stumble into a basement full of malnourished people who have been imprisoned by a group of cannibalistic survivors. The narrator agent recounts the event to the audience as such:

The boy clutched his coat. He could see part of a stone wall. Clay floor. An old mattress darkly stained. He crouched and stepped down again and held out the light. Huddled against the back wall were naked people, male and female, all trying to hide, shielding their faces with their hands. On the mattress lay a man with his legs gone to the hip and the stumps of them blackened and burnt. The smell was hideous…he turned and grabbed the boy…he’d dropped the light. No time to look. He pushed the boy up the stairs…he shoved the boy through the hatch and sent him sprawling. He stood and got hold of the door and swung it over and let it slam down and he turned to grab the boy but the boy had gotten up and was doing his little dance of terror. For the love of God will you come on, he hissed. But the boy was pointing out the window and when he looked he went cold all over. Coming across the field toward the house were four bearded men and two women.” (110-111).

Though the drama of the moment is focused, for both the reader and the character agents, on whether the man and the boy will escape the basement and the cannibals, the narration remains in past tense. Put differently, even though the narrator agent uses past tense, thereby signaling that these events have already occurred and that the narrator is, therefore, removed from the time of the events being narrated, the story unfolds with the immediacy of the present for both the character agents and the audience. The story product, on the level of narrative, is determinant.

The outcome is known and can, therefore, be relayed to the audience in any tense. For the characters, however, the experience is immediate, it happens, for them with no other time than the present. For the reader, for whom the outcome is unknown even while it is determinant, the experience maintains the drama of the present despite being presented in the past tense. Given this, we can argue that the story-time here is present and dramatic despite it past-tense narration.

Again, this establishes the difference between the two sets of time the audience experiences.

While the time of the story maintains a sense of the unknown and the dramatic, the time of the narrative structure occurs at a remove and a distance.

In comparison to The Road, the narration of Disgrace is delivered in present tense. That said, the narrator still possesses a reflexivity that belies the tense used. Despite the use of present tense, it is clear that the narrator agent knows the outcome of the story. For instance, when the narrator agent states, “At first they [his students] do not recognize him. He is halfway down the stairs before he hears the cry That’s him! Followed by a scuffle of feet,” the “at first” reveals the narrator agents temporal distance from the unfolding story (Coetzee 55). In spite of the narrator agent’s use of the present tense “is,” the audience knows the narrator agent possess knowledge that exists outside the context of the present moment the character agents are experiencing. The

“at first” reveals that though the students do not recognize him, they will do so momentarily.

Positioned in the present and within the context of the story-time, a narrator agent could not know the students would recognize Lurie momentarily. So, despite the narrator agent’s delivery of the story in the present tense, the audience is still aware of the narrator agent's position outside of the developing story’s time. Importantly, however, these two times are experienced simultaneously for the reader. Though the time communicated to the reader is bifurcated, the audience can understand the narrative-time and the story-time as distinct, even while receiving them at once.

Even in The Gathering, a story delivered in the first-person and in the present tense, the narrator agent is positioned outside the space and time of the story, reflecting back in such a way that the audience can interpret and think critically about the events and the characters. In fact, the

narrative framework of this novel demands it. In the opening pages, the main character agent,

Veronica, sets up the narrative framework of the story that follows. She states, “I would like to write down what happened in my grandmother’s house the summer I was eight or nine, but I am not sure if it really did happen. I need to bear witness to an uncertain event” (Enright 1). Thus, as the narrator agent, she positions herself outside the events of the story both in space and time: she remembers them, if unreliably, and is recounting them to the audience. In The Gathering, the narrator agent (ostensibly understood by the audience to be Veronica) recounts moments of

Veronica’s life as an adult and as a child, yet, the narrator uses the present tense to communicate both. Since Veronica cannot be both a child and an adult at the same time, it is clear that the narrator, despite also maintaining the character role of Veronica, exists in a time that is different from that of the characters. The audience, too, knows that both events have already occurred, that the narrator agent is reflecting back on them, and that the use of present tense can only be a narrative device. While it is harder to decipher Veronica as the narrator agent from Veronica as a character agent in those moments when the character is of adult age, it is still possible to separate the two. For instance, when the narrator agent says “for a week, I compose a great and poetic speech for my children about how there are little thoughts in your head that can grow until they eat your entire minds,” the use of “for a week” by the narrator agent clearly positions the narrator outside the temporal space and time of the character agent who composed the speech (Enright

175). The character agent, in the present, would not know the speech would be composed over a week but positioned retrospectively outside the time of the story, the narrator agent does.

Ultimately, and regardless of how the narrator agent delivers the story and regardless of how the narrator agent is positioned temporally in relation to the story, the narrator agent’s

function is to present content external to that of the story and to open occasions for the reader to reflect upon that unfolding story. Given this, there is a duality to time in every narrative structure. There is story-time, and there is narrative-time. These two types of time are present in all narrative texts and are as fundamental as character, narrator, and events. In this way, the dual function of time is paramount, no matter the narrative medium. Importantly, though, without this duality, there is also no narrative. As such, a defining equality of narrative is how the form fractures time.

Meanwhile, when space is described within a traditional narrative text, it is frequently described such that it is synonymous with setting and then “with background in the broadest sense” (Phelan and Rabinowitz 85). Taken, then, as background, space can also be “minimized without any obvious loss in ‘narrativehood’” (Buchholz and Jahn 540). In this regard, space becomes an element of narrative but not a function of narrative, in the same way as time. Going on step further, Bal goes so far as to argue that space is a disruption to narrative simply because extensive mention of it disrupts the time sequence of both narrative-time and story-time. She states, “chronology is always disrupted by spatial indications” (Bal 140). Likewise, H. Porter

Abbott dismisses setting as “inessential” to the components that form a narrative (59-60). While many of these theorists go on to complicate the assumption that space, especially in visual narratives, deserves more attention13, the fact remains that space, as a concept or element of narrative, has been rated as a lesser concern than that of time. As Buchholz and Jahn have

13 The study of Madame Bovary, for example, complicates theories of setting and space in relation to narrative. Teresa Bridgeman covers this well in her chapter “Time and Space” found in The Cambridge Companion to Narrative. Still, while Madame Bovary and other texts, such as Ulysses, complicate these theories in interesting ways, these texts can be considered the exception rather than the standard. Of course, these exceptions are important, but for the present purpose, the goals of this chapter are to establish the typical as a means for the comparison rather than to contend with the atypical.

summarized, “a story cannot have too much of temporal sequentially, but does grind to a halt when overloaded with spatial description” (540). Rather than space acting as an element of the narrative, within traditional text-based narratives, space acts as description, which according to some narrative theorists, is a non-narrative element because it contains no actions or events.

In the examples from The Road, Disgrace, and The Gathering, used previously in this section, recognition of each character agent's place in the space of the unfolding story is limited, and in each case, very barely disrupts the narration. In fact, in the example from Disgrace, the only indication of space is the use of the word “hallway,” while in The Gathering, the audience is provided only with “grandmothers house” and “summer” as a means to establish space.

Perhaps surprisingly, given McCarthy’s sparse style of writing, The Road provides the most detailed account of space. Here, the audience is provided with some context clues to establish space; the stone wall and the clay floor are enough to infer a basement or subterranean location of some kind. The audience is also told there are a hatch and stairs and that the air smelled

“hideous” (McCarthy 110-11). Beyond this limited description, however, the audience is left to imagine the space. As Chatman notes, “story-space is what the reader is prompted to create in imagination, on the basis of the character’s perceptions and/or the narrator’s reports” (104).

Of course, story-space functions differently in visual narrative forms such as film than it does in textual narratives. In film, story-space need not be imagined, as it is constantly on display. Given this, treatments of space in visual narrative forms such as film are more complicated. For instance, Bal argues that, even in film, space is “only a frame, a place of action” or an “acting place” (Bal 136). Likewise, Chatman suggests that, although space can be manipulated in film by scale, size, contour, texture, density, position, and degree, it remains a

background element except in those moments in which the character agents act upon it or are acted upon by it (Chatman 97). Thus, for Chatman, space becomes narratively relevant only when it is part of an event or action, both of which are elements of story-time rather than story- place. As he remarks, “story-space contains existents, as story-time contains events. Events are not spatial, though they occur in space; it is the entities that perform or are affected by them that are spatial” (96). For Chatman, character agents are spatial as a result of holding mass and taking up room, but the actions and events a character agent experiences are temporal. By contrast, Stephen Heath has argued that “frame space,” or that which is shown to the audience in the frame of the camera, “is constructed as narrative space” (56). Extending from this, Mark

Garrett Cooper has argued that, “the story does not happen in space so much as to space” (143).

What Heath and Cooper suggest is that, in film, space must be individualized in such a way that it is essential to narrative delivery. It must be realized for the audience in ways that it need not be in textual narrative. To illustrate this, Cooper provides as example using the narrative case of a murder. While a textual narrative may account for the action leading to the murder, a film will also need to “prepare a spot for the body to fall” (147). Subsequently, for Cooper, this space becomes an essential narrative element rather than inactive setting. One might complicate the arguments made by Heath and Cooper, however, by arguing that the frame itself is part of narrative structure, but that what happens inside it is not. In other words, the frame acts the mechanism for describing the scene, much as description in textual narratives, but that the setting only has meaning in so far as the existents experience an event within it. It is true, of course, that how a scene is framed might call attention to details that are important to the narrative – such as where a body falls – but this, again, is description rather than action itself. The frame, as a

narrator, informs how the audience views and makes meaning, but does not offset or outweigh the importance of time as a narrative element. Even as space becomes a more important element of narrative within film, the delivery of the story is still bound in time. Therefore, what the frame visualizes and makes possible is the ability for the audience to see events occur in time and space, but it is still the time, rather than space, that propels the story forward.

As this analysis shows, traditional narratives not only privilege story-time over story- space, but also distance the narrator agent from both story-space and story-time as a mean to provide narrative remove and the opportunity for critical thinking and reflection. That said, I have argued that digital interactive narratives collapse the role of the narrator agent and the character agent into a single player agent. For this to be so, the construction of story-time and story-space, as well as the narrator agent’s position outside of both, would also have to be collapsed. In the next section, I will illustrate how this is accomplished.

Time and Space in Digital Interactive Narrative and Story

In digital interactive narratives, the player agent must simultaneously be able to experience narrative distance while also experiencing story-time and story-space. This is accomplished in three key ways:

1. By making story-space a function, rather than an element, of the narrative

structure.

2. By making story-space one of the most interactive functions of story construction

and by privileging it over story-time.

3. By including a drama management system – a system in the game that “ensures a

well-formed story occurs each time” – that acts as the liaison between

story functions (story-space, story-time, and character agents) and narrative

functions (narrative-space, narrative-time, and narrator agents).

In digital interactive narratives, space is as important, if not more important, to the narrative structure as time. In her definition of cyberdrama (or what I refer to as the digital interactive narrative), Murray states that cyberdramas are “the enactment of a story in the particular fictional space of the computer” (“From Game to Cyberdrama” 4). Echoing Murray,

Henry Jenkins argues that games employ what he calls “environmental storytelling” (123). Thus, while in traditional text-based narratives space is generally considered to be descriptive set dressing, and while in film space is always visible and contains and frames action, it is only in interactive narratives spaces that the player moves herself through the narrative via her movement through space. Thus, the actions and events that occur in time occur due to the player's manipulation of space. Given this, space in digital interactive narrative games is arguably and uniquely a more important function that than of narrative than time.

When reading narratives, we imagine space, when watching narratives such as film or theater we view space, but when playing digital interactive narratives, we experience space in ways that are different by significant degrees than the way audiences access or understand space in other narrative media. As Jesper Juul notes, “it is hard for a player to be abstracted from and hence, to disregard the space in a game” (Half-Real 189). In other words, space is an interactive function of narrative in digital interactive narrative games. This is an incredibly important feature because it moves the narrative focus from a reflective one to an active one. In digital interactive

narratives, space is always present to the player, always visible and set before her. More than this, however, space is not static or unchangeable in digital interactive narratives, but rather is capable of being manipulated: the player agent can move through the game world and environment and engage with space directly. In the words of Alexander Galloway, digital games are an “active medium” in which the game’s “very materiality moves and restructures itself” (3).

For example, in digital games, the player agent can often pick up and move objects, travel across the game environment, find and use tools, and so on. Space, therefore, is a function of the game and is implemented into its rules (Juul Half-Real 188). As such, “the organization of spatial features in a game is critical to the design of a game's narrative space of possibility,” and to the player agent’s capacity for interactivity and agency (Salen and Zimmerman 390). In this way, space is more apparent to the player than time. Thus, while time is not typically an element the player can manipulate or interactive with (though there are some exceptions), space always is.

The focus on space does impact time, however, such that the player agent largely experiences time in the present and with the immediacy that a character agent would. Therefore, the game must also provide a means by which the player agent can remove herself from the space and time of the environment in order to reflect critically upon it as a narrator agent might.

Digital interactive narratives accomplish this via the use of drama management systems. Drama management systems function in interactive narratives as “omniscient story directors” (Magerko

2). These systems, which “bring about a dramatic experience for the human participant,” signal the player agent when to act as a character agent and when to reflect as a narrator agent (Riedl and León 1 Dialogue screens, morality systems, quick time events (QTE’s), and heads-up displays (HUDs) are all components of a drama management system. These systems generally

pause play, thereby removing the player agent temporally from story-time, and bring up unique interfaces that obstruct the player agent’s spatial view of the game environment, thus also removing the player agent spatially from story-space. While interacting with those interfaces which are linked to the drama management system, the player agent is effectively interacting at the level of narrative. The choices she makes and the way she engages with these systems alter the story, but the characters in the story do not acknowledge these interjections.

For example, in Mass Effect 2, at the character level, the player agent inhabits Shepard.

When engaged in conversation with other characters, a dialogue selection menu appears on the screen. This screen freezes story-time. While the player agent engages with the menu, the scene behind the menu remains frozen. Storytime is effectively at a standstill until the player agent makes a narrative choice which then reactivates story-time. Similarly, in Skyrim, the player agent has various interfaces that can be used to equip weapons, alternate between magical abilities, change clothes, mix potions, eat food, and so on. Accessing these interfaces, likewise, pauses story-time. While navigating these systems, the action and events of the story are frozen. The character agent, for instance, will not be attacked or engaged in conversation while the player agent interacts with these interfaces. Additionally, in the case of Skyrim, these interfaces also completely obscure the story-space too. Until the player agent exits these screens, she will not be shown the space of the storyworld. Thus, the drama management system acts as the barrier between narrative-space and narrative-time and story-space and story-time.

The components of the drama management system can also be used to document the choices made by the player and to provide feedback to the player based on the choices she has made. Yet, while these interfaces do score and report data, they do not mediate the story with

the same capacity as a narrator agent. Instead, drama management systems act as tools that assist the player agent. They help the player agent navigate between the roles of character agent and narrator agent and help the player agent mediate the difference between the space and time of the story and the space and time of the narrative. Thus, with the aid of a drama management system, the player agent can simultaneously act in the present space and time of the game while also possessing the ability to reflect on the experience with narrative distance and remove.

Aided by drama management systems, interactive narrative games can activate narration in such a way that it produces narrative agency and interactivity. In digital interactive narratives, the player agent can move through space and time in such a way that she can be in the story as the character agent and also outside the story interpreting and analyzing it as a narrator agent14.

Thus, while in traditional narratives the narrator agent and character agent are separated in space and time, the player agent navigates both the spatiotemporal planes of story and narrative at once. Much as the player agent switches between the roles of character and narrator, so too does the player agent switch between the space and time of story and narrative.

To illustrate these three points, we can return to Banner Saga, Mass Effect 2, and Kentucky Route Zero. In each game, space is a very real function of the narrative and is the central means by which the player agent conceives of her progress through the narrative structure. Quite literally, the narrative structure, as well as the player agent’s progression through

14 As suggested elsewhere, this is also different by degrees than the ways in which an audience may interact with textual or visual narratives. While in text and film, the reader has access to view and interpret the narrative and the resultant story from multiple levels, in digital interactive narratives the player agent is both narrator and character. The player's interjections and actions within these roles can alter how she navigates through the narrative structure and how the resultant story product is produced. In other words, the player has input, gives input, and the narrative system of the game responds to and accommodates this input.

that structure, is mapped out spatially. In The Banner Saga, the player agent is shown her location on a map and anytime the player agent is shifted between character agent perspectives, that change is represented on the map (Stoic). Similarly, in Kentucky Route Zero, the player agent must literally navigate the road, Rt. 0, using a map (Cardboard Computer). The subplot or story beats are represented as landmark destinations on the map. To progress in the story, the player agent must navigate from one space to the next along the narrative map. Mass Effect

2 also employs a map, though in this instance the map stretches across an entire galaxy

(Bioware). Each plant, space station, or ship acts as a narrative destination that, once visited, moves the developing story forward. In each case, the spatial structure functions as the means to progress the player agent through time and through story. In traditional literary narratives, time is primary means of narrative advancement; the structure of the narrative itself consistently denotes time through the tense in which the story is delivered. Comparatively, in digital interactive narratives, space is the primary means of narration. Here, the player exists in time, certainly, but the player progresses forward by interacting with and moving through space.

In each game, the fictional space also provides a staging ground where story events are enacted. While traditional narratives are concerned with the temporality of events, digital interactive narratives concern themselves with the spatiality of events. This alteration is essential to the game’s interactivity. As Jesper Juul notes, “space and level design in games are special areas where rules and fiction overlap” (Half-Real 163). In games, the player agent must frequently interact with her environment and, just as frequently, this interaction serves to develop story construction. In The Banner Saga, for example, the battles are turn-based, and during them, the player agent is given an aerial view of the battlefield (Stoic). The battlefield is arranged into

squares, much like a chessboard and each character, based on their battle style, can move across the squares in a specific pattern. Thus, confrontation in the game is incredibly spatial. The outcomes of these battles have consequences upon the developing story, and thus, the player agent’s understanding of the battle space is vital to her narrative progression through the game.

Comparatively, in Mass Effect 2, the player agent has more freedom to explore spaces and those explorations necessarily alter her perception of the story and her progression through the narrative structure (Bioware). For instance, even aboard Commander Shepard’s spaceship, the

Normandy, how the player agent navigates space has a direct impact on how she constructs the developing story. Aboard the ship, the player agent is free to roam the various floors and rooms.

The player agent can elect to go to Shepard’s private quarters and can customize the room to make it feel personal. Alternately, the player agent can visit any or all of the crew. The player’s choice to do these things is also presented to her spatially. To visit the crew or to go to her personal quarters, the player must ride an elevator. Thus, spatially, she opts to go up or down elevators and corridors. Subsequently, once in the crew’s quarter’s, the player agent can elect to have conversations with the crew members. The result of such a choice is the development of characters and relationships. This development is important to the narrative outcome because

Shepard is told to earn the trust of the crew. To develop this trust, Shepard must locate each character in his or her unique space and engage with them. Even the activation of missions, ostensibly the events which propel the plot forward, are communicated to the player spatially.

Shepard must go to a command center and activate a map of the world. Only by selecting a point amongst the stars and planets in the galaxy – a place in space – can the narrative move forward.

Thus, by navigating the space of the ship, the player agent also engages with the narrative structure of the game and the story construction.

This is not to say that time is of no consequence in digital interactive narrative games.

Just as space is not inconsequential in film, time is not inconsequential in these games. What differs is the relevance of space in digital interactive narratives and how the function of space as a narrative element frequently supersedes that of time. Because the interactions games allow are navigational and focused on the interrogation and manipulation of space to prompt events, a focus on space is a narrative necessity. While it is true that some games incorporate time as a function that can be manipulated by the player (and in this case even the ability to pause might be considered a manipulation of time) not all games do. That said, space is always present and is always a function.

Finally, as the player agent moves through the space of each digital interactive narrative, a drama management system helps the player agent smoothly mediate the switch between narrative-space and narrative-time and story-space and story-time. In Banner Saga, the player agent can customize the character agents by building up their stats or increasing their item rank

(Stoic). These customizations are presented via character-customization screens which obscure the player agent’s view of the story space. These screens also pause the story, allowing the player agent as much time as needed to reflect on her choices as the narrator agent. When moving through these character screens, the player agent is operating outside story-time and story-space and inside narrative-time and narrative-space. Similarly, in Mass Effect 2, story-time and story- space are paused whenever the player agent engages with a dialogue screen, a HUD, or the

“mission computer,” an interface that show the player agent combat stats, morality stats,

equipped weaponry, etc. (Bioware). As noted previously, when a dialogue screen is displayed, the player agent must view her options from the perspective of the narrator agent and reflect on how these options will affect the developing story, especially when and where those dialogue options are color coded to signify a moral choice is being made. Because the character agent is unaware of these dialogue options, the dialogue screen is a signal to the player agent that she is no longer in story-space and story-time but in narrative-space and narrative-time. This same concept applies to the player agent’s use of the HUD and the mission computer too. In each instance, the player agent suspends story-space and story-time, pausing and removing them from view, in order to engage with the narrative structure of the game. The same concept applies to Kentucky Route Zero as well. Only, in the case of Kentucky Route Zero, the dialogue screen is the game’s drama management system (Cardboard Computer). As noted before, the dialogue screen in this game prompts the player agent not only to select what it said, but which character agent speaks. Thus, any time the player agent is provided with dialogue choices, she is thinking on the level of narrative-space and narrative-time. Once the dialogue option is selected, the player agent moves back to story-space and story-time. The switch between the spatiotemporal planes in Kentucky Route Zero is more rapid, with the shifts happening with each frame of dialogue. Still, it is important to note that these dialogue screens still suspend story-space and story-time. The player agent can only move in the game space after the dialogue screen disappears from view. Thus, even in this game, the drama management system acts as the mediator between narrative-space and narrative-time and story-space and story-time.

Conclusion

At the beginning of this chapter, I outlined several arguments scholars in Game Studies have made to discredit the narrative potential of digital games. Tisselli’s argument, for example, suggested that the multidirectional nature of games limited the potential for rich narratives.

Returning now, it is easier to see how Tisselli’s argument is misguided and how he underestimates the capacity for narratives to use the multi-directional nature of games to their advantage. Similarly, Costikyan’s critique that the demands of story are always in conflict with the demands of the game seems no less unfounded. Costikyan not only conflates narrative with story but in doing so ignores the fact that the functions of narrative, such that they are, generally support the functions of games. As Ryan has noted in “Toward a Definition of Game,” narrative is about problem-solving, conflict, interpersonal relations, human experience, and the temporality of existence (24). Narrative, in these ways, hardly works against the demands of a game, but rather, in conjunction with them. Finally, Aarseth’s concern that narratives cannot maintain their structure within a simulation should also now seem invalid. The structures and elements of narrative are no less present nor less malleable in a digital interactive narrative than in a textual or visual narrative. In fact, narrative structures are arguably more malleable when integrated into game systems than they are in any other medium.

As this analysis shows, digital interactive narrative games maintain the essential components of text-based narrative structures. They have a sequence of events, whether actions or happenings, that are experienced by characters within a particular time and space and which are mediated by a narrator who is removed from the space and time of the story. That said, while maintaining these essential structures, the inclusion of a player agent does necessitate a new

means of constructing these functions. Largely, this is accomplished by designing systems and user interfaces that allow the player to mediate and switch between story-time and story-space and narrative-time and narrative-space, as well as between the roles of narrator agent and character agent. These crossovers between roles, spaces, and times create opportunities for a very specific kind of interactivity: narrative interactivity. This concept of narrative interactivity is important for the very reason that it challenges and works against the idea perpetuated in Game

Studies scholarship that narrative and play are fundamentally different or at odds. What this chapter, instead, suggests, is that the roles of narrative in games are not a limitation to agency or interactivity, but another means by which to create it. Understood in this way, Game Studies should begin to consider the complexity of narrative structures, their component elements, and how those elements function within interactive mediums. As this chapter has shown, dismissing narrative as a detriment to interactivity is actually only detriment to Game Studies and our understanding of games as a complex medium.

CHAPTER 3

PROCEDURALLY CONSTITUTED AND SOCIALLY CONSTRUCTED: AGENCY THEORY IN DIGITAL INTERACTIVE GAMES

Introduction In the previous chapter, I discussed the collapse of narrative and story agents into a single player agent. Additionally, I argued that this collapse makes the player agent's navigation between narrative-space and narrative-time and story-space and story-time a source of interactivity.

The use of the word “agent” throughout Chapter 2 is intentional. While it would be easier to distinguish between the roles of the narrator agent, the character agent, and the player agent by referring to them simply as the narrator, the character, and the player, the addition of the word

"agent" implies that each of these roles is active. As active agents, each role also then logically holds or has the capacity to hold some degree and form of agency.

The concept of agency, however, is no less fraught or contentious within Game Studies than the concept of narrative. Additionally, just as there is no agreed-upon definition of narrative in games, there is also no agreed upon understanding or definition of agency in games.

Most broadly, game scholars have only generally agreed that agency and free will are non-synonymous in relation to games, but that is where the general consensus ends. That said, any definition that relies solely on an understanding of what it is not is hardly helpful in terms of understanding what it is. Nor, however, it is helpful for each Game Studies scholar to define agency according to his or her own preference, such that all understanding of the concept becomes subjective and without utility in broader argument.

In this chapter, I argue that we needn’t continue the cycle of redefining agency to the point that the concept itself becomes so thin and so fraught as to lose its relevance or utility.

After all, the construction of agency and how humans constitute it has, for centuries, been actively studied and continues to be a central theoretical issue within the fields of theology, philosophy, psychology, and sociology. Thus, to establish what agency is within games, we should attempt to understand its intricate and multifaceted history outside of games. In doing so, we can stop constructing definitions that exist in a vacuum, are different by slight and tedious degrees, and which suit individual purposes, and instead ground the understanding of agency within the rich bodies of theory that already exist. Then, where Game Studies scholars continue to disagree, they can do so from a place informed by history, context, and nuance.

Of course, the attempt to better understand agency in games by looking outside of Game

Studies does not imply that any one philosophical, sociological, or psychological theory of agency will apply entirely to games; nor should it. Instead, the implication is that by ignoring the available wealth of scholarship outside of Game Studies, the field of Game Studies itself suffers.

Thus, it is my intention with this chapter to engage – albeit topically – the history of agency theory as it exists and has developed across numerous fields and modes of inquiry. Having done this, this chapter also considers how these theories of agency lend themselves to games and how they might expand the definitions of agency currently in use by game scholars. Finally, I provide a new definition for agency and illustrate how this expanded understanding lends itself to better analyzing and understanding the construction of agency in digital interactive narrative games specifically and, more broadly, in all digital games.

Definitions of “Agency” within Game Studies

There are numerous definitions game scholars have created to define agency and how it functions in games. In most cases, these definitions attempt to delineate how and in what capacity a player can act or make choices within the closed system of a game. In this way, Game

Studies scholars have mutually agreed that, while agency in games is not synonymous with free will, the player’s actions and choices are inherently linked to the player’s experience of agency.

Janet Murray, for instance, defines agency as, “the satisfying power to take meaningful action and see the results of our decisions and choices” (Hamlet on the Holodeck 126). While

Murray’s definition is useful, it is limited by its focus on “action” alone. Murray correctly links agency to the capacity to make choices, but she also implies that choices are actions. As this chapter will demonstrate later, the attempt to equate choice with action creates a false paradigm that is complicated and challenged within a great deal of agency theory outside of Game

Studies.

Elsewhere, Katie Salen and Eric Zimmerman define agency as “making choices and taking actions” (670). Here, the two make a distinction between choice and action but imply that choice leads to action. This logic supposes that agency only exists when it leads to action, which is also a reductive understanding of agency. Salen and Zimmerman’s definition removes the opportunity for agency to be expressed through inaction, which is, of course, also an important means of expressing agency. While one might argue that inaction can be understood as a type of action, Salen and Zimmerman do not clarify this, and thus leave open the potential for agency to be understood in this reductive manner. By comparison, this chapter will address both action and inaction as important components of agency.

In another definition, Gonzola Frasca claims agency to be “the ability of the computer user to participate in simulations” (“Rethinking Agency” 167). While Frasca’s definition is simple, he fails to explain what he means by participation. Just as action and choice are not synonymous, neither are participation and agency. Again, as I will show later, even a cursory review of traditional agency theory shows that one can have agency without participating, just as one can participate without having agency. The two elements are not mutually exclusive.

As these definitions show, Murray, Salen, Zimmerman, and Frasca all present valuable, though incomplete, ways of understanding agency in games. Yet, their biggest flaw, and one they all share, is that each of the definitions focuses more on the actions, choices, and results presented to players by the game than on the players themselves. Given that agency theory outside of Game Studies is human-focused, it seems that agency theory within Game Studies should take more care to address the human element too.

Karen and Joshua Tanenbaum’s definition make the clearest attempt to address the player’s experience of agency. The two define agency as “a commitment to meaning” that “shifts the emphasis in an interaction away from the outcome of a choice and toward the intent which underlies that choice” (Tanenbaum and Tanenbaum 7). Unlike the other definitions, the

Tanenbaums’ suggest that, for a player to maintain agency in a game, the player’s choices must have both meaning and intent. The Tanenbaums’ change in focus is important, as it focuses more on the player as a critically thinking and rational human who happens to be playing a game, and less on the game and the actions, choices, and types of participation it makes possible. That said, the Tanenbaums’ definition fails to account for the ways in which a game’s closed system may limit or alter the player’s intent or attempts at meaning-making as a result of its code. In other

words, while Murray, Zimmerman, Salen, and Frasca focus too much on the game system, the Tanenbaums do not account enough for it.

My definition, used in previous publications, has been "one in which the player interacts meaningfully and continually with a well-developed narrative for the duration of the gameplay experience." Though the definition I used was, itself, an attempt to account for the numerous elements of agency presented by other Game Studies scholars, it also fails to account for the complexities of agency in games adequately. For example, my definition fails, much as the Tanenbaums' does, to account for a game's system, and instead focuses on the gameplay experience. Furthermore, my definition fails to consider that agency can be and is intermittent rather than continuous. Thus, while scrutinizing the definitions of others, it is only fair to identify the limitations of my own, too. Where agency in digital games is concerned, there is still much to learn.

The question then becomes, is there a definition that can account for the generalities and the specificities of agency as it pertains to the player, the game, and the interactive experience between the two? Furthermore, how do we account for agency in games in such a way that we account for both a game’s closed system of code, the choices and actions it makes permissible, the types of participation it allows, and the complex human being who interacts with it? While each definition attempts to account for one parameter of agency in digital games, no definition sufficiently accounts for the full complexity of agency in digital games. Simply cobbling the assembled definitions together would also fail to create a satisfying definition or understanding, as there still is more to agency than participating, interacting, decision-making, and meaning- making within the confines of the game experience. While each component is important, the

combination thereof does still not begin to account for how agency can be constructed in a game, how it function in the game, and how the player can interact in such a way that she can maintain it. In other words, agency as a concept is far more complex than we have considered it to be within Game Studies scholarship, and until we confront that complexity, our definitions for agency within the context of digital games will never be adequate.

Theories of Agency in the Social Sciences

Outside the context of games, the study of agency has a long and complicated history.

The subject of agency has been at the core of numerous theological, philosophical, sociological, and psychological inquiries over numerous centuries. As such, no single chapter could ever summarize its expanse. Given this, my goal is only to address – albeit topically – several key theories as a starting point. What follows will show not only how theories of agency have developed over time, but also the depth of insight Game Studies scholars can glean from even a cursory review of it.

Determinism and Compatibilism

Provided that “I think, therefore I am” is perhaps the most recognizable quotation penned in relation to the study of agency theory, it is as good a place as any to start recounting the history of agency studies. Written by René Descartes in Discourse on the Method in 1637, the phrase was an attempt by Descartes to summarize how man actualizes agency.

Descartes reasons that "if I persuade myself that there is an earth because I touch or see it, by that very same fact, and by a yet stronger reason, I should be persuaded that my thought exists; because it may be that I think I touch the earth even though there is possibly no earth existing at

all, but it is not possible that I who form this judgement and my mind which judges thus, should be non-existent" (280-281). Per Descartes, the physical world is deterministic and bound, but within that unchangeable construct, the human mind is free. This means that whether all cause and effect are predestined by some divinity, man is still a rational and free-thinking entity capable of actualizing himself through that very thought process. Extending from this,

Descartes also argues that because humans can perceive and understand, they can also act with intentionality on their thoughts and perceptions, whether or not those actions are part of a predetermined outcome. As such, Descartes suggests that acting with intentionality requires human will, which is derived from man’s capacity to perceive and understand. Thus, in a cyclical way, perception and understanding lead to willpower which leads to intentional action or inaction which, in turn, leads back to perception and understanding.

Determinist philosophies such as Descartes’s stress that human perception and intention are essential components of agency and that agency can be maintained even within a determinate system. This is an important observation to consider in terms of how we understand agency in games. While more contemporary thinkers than Descartes still debate whether the world is determinate or not, there is very little debate about whether game systems are determinate systems. Given that games are coded systems with strict parameters and rules, they can be considered to be deterministic. It is statistically possible, because the rules and variable are known, to map out every possible outcome of a game, even if one is unlikely to do so. While some may argue that this is not the case for those games that are procedurally generated, even those games make use of determinate algorithms. In other words, just as

Descartes’s theory understands the world as a closed system, so too can games be understood as

closed systems. That determinism, however, does not mean that agency within a closed system is implausible. If we extend Descartes’s theory to games, we might suggest that agency in games is derived from the player’s perceptions of her own will and her ability to actualize that will with intent within the closed system of the game. In this scenario, the player can understand that the system is determinate and still possess agency within that system.

Writing little more than one hundred years after Descartes, David Hume takes a compatibilist view of agency. Compatibilist philosophers argue that determinism and free will are not mutually exclusive, but rather can be present simultaneously. In other words, the concepts of free will and determinism work together compatibly. In An Enquiry Concerning

Human Understanding, Hume argues that moral responsibility, or the capacity to act and be held accountable for one’s actions, does not negate the possibility of determinism or vice versa. In other words, even within a determinate world, a man must take responsibility for his own actions whenever and if those actions are an expression of his intent (Hume 63-64). For

Hume, humans possess agency when they express their will and desire. Simply put, when a person chooses to do something of his own volition, his do so of his own free will. When a person is forced to do something or if choice is removed, then the person acted outside of his will.

For Hume, the compatibilist view matters in terms of moral responsibility, but within the context of games, the compatibilist view helps to articulate the negotiations between the player and the game system. Hume’s theory reminds us that the determinate rules of a game do not necessarily limit the player’s capacity for agency. In other words, the two are not mutually exclusive. But, as Hume’s theory points out, the relationship between the two is delicate. Just as

there are circumstances in which people in the world can be forced to act in ways that negate their agency, the system of a game can force a player to act in ways that negate their agency too. If we extend Hume’s theory to games, we can suggest that players lack agency when they are forced by the system to act against their own desires, wills, or understandings of the choices available to them. If the player’s actions do not reflect the player’s intent, it is likely the action is not a reflection of the player’s agency but, instead, a reflection of the game limiting it.

Importantly, for both Descartes and Hume, agency was grounded more strongly in intent than in the action – or inaction – that followed or actualized that intent. For both philosophers, it is a person’s intentionality that endows him with the ability to make conscious choices. Regardless of the type of systems – games or otherwise – in which those choices are made, Descartes’s and Hume’s theories suggest that if can actualize our intentions in some way, then we can maintain agency in any determinate system.

Of course, the determinist and compatibilist views of Descartes and Hume are not without flaw. Perhaps because of their belief in the overall determinism of the world, each speaks of agency as if it exists in a constant state and in equal measure for all humans. While their arguments are theoretically interesting, they are also difficult to observe in practice, as the parameters of the world and its circumstances are far too complex at the micro level of the minute-to-minute lived experiences of each individual. Thus, while it may be convenient to speak of humanity as a homogenous group it is far from practical.

Structuralism and Functionalism

Given that humans and the world are in a constant state of flux, it logically follows that agency is in a similar state of flux. The recognition of this essential truth eventually led to what has been called the "structure and agency" debates. Within this debate, structuralist and functionalist scholars examine whether individuals act as free agents who are independent from the society in which they live or in accordance with the dictates of the social structures in which they live.

As a methodological approach, structuralism attempts to understand humanity within the context of a larger, structured system. The work of structuralism is not to discover why humans act as they do, but instead to discover the underlying structures that prompt humans to act as they do. Within the structuralist movement, the theories of Claude Lévi-Strauss are perhaps the most well-known. Structuralist methodology, Strauss suggests, can only be applied to the study of societies of which the theorist, anthropologist, or scholar is not a part. In his memoir, Tristes

Tropiques, Strauss argues that individuals view their own society subjectively, but view the societies of others objectively (Badcock 83). By the nature of our experience and our position within a particular society, we naturally understand the society in which we live better than we understand societies that are foreign or otherwise isolated from our own (Badcock 82). As such,

Strauss argues that we cannot view our own society with objectivity. Instead, we view our own societies as inextricably linked to a complicated history with which we are entwined and empathetic. Given this, any attempt to map or to understand the underlying structures of the society in which we live are bound to be circular. "The application of our own theories to our own doings" can be nothing other than subjective and flawed (Badcock 84). Conversely, Strauss

argues that our view of external societies is detached (Badcock 83). Because of this detachment, we can expose the structural makeup of another’s society from an objective viewpoint. From this viewpoint, Strauss suggests, we can discern the values shared by individuals within a particular society. Discerning these values is important because the values of a society are

"defined by purpose" and "always imply some sort of commitment to choice" (Badcock 90). In other words, the values of individuals are subjective. For Strauss, this means that values not only define the number of choices an individual may conceive of, but also inform how an individual may opt for one choice over another or how the individual understands her choice before and after making it. In other words, both her choices and the values that inform them are a reflection of the societal structure in which she lives.

As we think about agency within the field of Game Studies, Lévi Strauss’s structuralist theory raises some important considerations. For one, Strauss prompts us to consider that a player will view their values subjectively while viewing the values of others objectively.

Furthermore, these values, are situated within societal structures in which the player either does or does not live. In turn, these values are directly transferrable to how individuals conceive of and commit to choices. For instance, developers of a game are likely to situate the game within their societal understanding and therefore within the values that inform that understanding. As such, the choices available to the player will be a reflection of those values and that society too.

This will be true whether or not the player is part of that society and value system. That said, the player will also bring a set of values to the game. If the game's values and the player's values are in harmony, then the player will share a similar conception of what her choices are in the game.

If, however, the game's embedded values and the player's values are not in harmony, the player is

likely to experience some disruption or feel some dissonance with how she is able to make choices in the game.

In comparison to structuralism, functionalism analyzes the compositions and evolutions of society and how, within the construct of a society, humans behave. Functionalist theories, therefore, argue that human behavior helps create social structure and shape society, but those structures and societies also frame human behavior in turn. Talcott Parsons argued, for example, that a “social system is made up of the actions of individuals" (190). These actions, however, are made available to individuals by the society in which those individuals act. In other words, individuals assess the possible actions available to them based on the society in which they live.

An individual in one society may conceive of a possible action that is inconceivable in another.

That said, the rules of society are not static; they change over time. What was inconceivable in one decade within a society may not remain so in subsequent decades. Importantly, however, functionalist theories suggest that the changes and evolutions that occur within a society are not random. Instead, the changes reflect the changing needs of the society based on the actions people within the society take, don’t take, and wish to take.

Just as different societal systems have different rules and, therefore, make different levels of agency possible, so do game systems maintain different rules and provide differing levels of agency to the player. While this statement seems logical, many attempts to define agency in games fail to account for it. What one game genre makes possible, another does not. Further, what one game makes possible within a given genre, another game in that genre will not.

Therefore, what Parsons's theory helps up remember within Game Studies is that any understanding of agency is in some capacity dependent upon the game being created, played, or

analyzed. Furthermore, the player and her circumstances are other variables. Just as a social system is made up of the actions of individuals, so too we could say that a game is made up of the actions of the player. While the system itself might have a set of rules, the player's interactions define the game's state of play. Here, Parsons's theory reminds us that while the code of a game may be fixed, gameplay isn't. The player is constantly negotiating the potentiality of her options within the constructs of the system. This means that the player is always mediating what can and cannot be accomplished based on the current game state. In other words, agency in digital games is, in some capacity, a negotiation between an agent and her current circumstance.

Therefore, it will differ as the situation does.

That said, regardless of which society agents find themselves in, Parsons argued that any unit said to contain agency must meet four criteria before it can act. First, there must be an agent or "actor" that can perform the act. Second, there must be a goal or finalization on which the action is focused. Third, there must be a situation necessarily different from the goal the agent is focused on that prompts the agent’s action. Fourth, there must be at least one alternative action not chosen, “so far as the situation allows alternatives” (Parsons 44-45).

Given that games are themselves structures in which an agent can function within a set of rules – much as people function within the rules of their societal structure – it is not hard to apply

Parsons's criteria to them. First, if we understand the player as the agent or actor, the game must allow the player to perform; the game must be interactive. Second, the player must be able to focus their action toward a goal, such as the completion of a level, a mission, a quest, or the game itself. Third, the game must provide situations that are distinct from the player's goal and must prompt the player to act. For instance, if the player's goal is to complete a level, the level

must provide reason and impetus for the player to take action toward that goal. Fourth, the player's action must be the result of choice, wherein the player's action is necessarily the negation of at least one alternative action.

Together, Parsons's third and fourth points are perhaps the most important to the study of agency in games. If we carry Parsons's theory over to games, we can understand his criteria to mean that the situation created by the game must not only provide the player with incentive to work toward a goal, but the player must, in turn, have knowledge of available alternatives and must, then, self-select a course of action based on the goal they are aiming their action toward.

Given these conditions, Parsons's theory goes a step beyond the ideas of intention and will as articulated by the determinists and compatibilists. In this case, for agency to be present, there must also be a choice, and the player making a choice must be able to act on it willfully and with intention while consciously choosing her course of action over at least one alternative. In other words, if a player is forced by the game to take one action in order to proceed, that action lacks the capacity for player agency.

In 2006, social cognitive theorist Albert Bandura built on Parsons's criteria in "Toward a

Psychology of Human Agency." Not unlike Parsons, Bandura suggests that humans and their capacity for agency are interwoven with and are, in many ways, inseparable from the societal conditions in which they find themselves. Bandura says, "most human functioning is socially situated. Consequently, psychological concepts are socially embedded" (165). Again, the functionalist implication here is that how we think and how we act are both affected by the society in which we live. Much like Parsons, Bandura also suggests that there are four criteria that must be met for a person within a particular society to think and act with agency. While both

scholars have proposed four criteria, there are distinct differences between each scholar’s set of criteria. Bandura's four include: first, the intentionality of the agent; second, the forethought of the agent, understood as the extending of the agent into the future as a means of and planning for the goal. In Bandura’s opinion, forethought helps the agent imagine potential outcomes that either motivate or alter the agent's course of action. Third, is self-reactiveness or the ability to

"construct appropriate courses of action and to motivate and regulate the execution." Fourth, is self-reflectiveness or the ability to self-examine past actions critically before taking new action.

Though both scholars are working within a formalist methodology, Parsons’s criteria are concerned with an agent externalizing a choice within a given situation in order to work toward a goal, while Bandura's criteria are more concerned with the internal psychological processes that prompt the agent to externalize that choice. Bandura's first point, which articulate the importance of intent, is reminiscent of the ideas expressed by Descartes and Hume. Intent, however, is not enough to constitute agency in Bandura’s estimation. To activate intention and maintain agency, an individual must be able to picture not only achieving her goals but also the means for doing so. She must project herself into a potential future and visualize the process of arriving there. The individual, moving forward and backward between reality and the imagined potential future, must also be able to react and reflect. Bandura thus suggests that human agency is a flexible process in which action and choice are the results of a much larger psychological process. Agency doesn't exist solely in the actualization or execution of intention but in an individual's ability to react and reflect to and on that actualization, both as the individual moves toward the intended goal and after the individual has reached it.

In terms of Game Studies, Bandura's criteria remind us that an important part of agency remains psychological and unseen. The execution of choice via action, if action occurs at all, is part of a much larger process that takes place in the individual’s mind. Given this, games must not only anticipate action by providing a player with a variety of options as she works to accomplish a goal, but they must also create choices that facilitate three types of critical thinking: forethought, self-reaction, and self-reflection.

Importantly, Bandura’s theory also reminds us that while we frequently discuss agency in terms of “action,” we must be cognizant of the many ways in which individuals can be active without physical movement. Frequently, when we talk about agency we talk about the potential for action, the attainment or realization of a goal, or the selection of options as a way to express choice. It is essential, however, not to forget that agency also exists in inaction and in the failure to realize a goal. Will, intent, reaction, and reflection can all exist in the absence of bodily movement or action. As such, there can be just as much choice in immobility as in mobility, and both can equally be the result of intent, and each can be reflected upon.

While the determinist and compatibilist theories too broadly account for human agency as something omnipresent and equally shared amongst all the individuals, the structuralist and functionalist theories of Strauss, Parsons, and Bandura help to establish the unique sets of criteria that create the possibility for agency within a society and for each individual agent within a society. That said, these theories fail to account for the larger whole that the structuralist and functionalist theorists maintain. For a theory or approach, then, that attempts to consider agency both from the top down and the bottom up, we can look to phenomenology, the study of human behavior, as a means to understand nature.

Phenomenology

Founded by Edmund Husserl at the start of the twentieth century, phenomenology is, at its most broad, the study of how our experiences and consciousness are structured. As a methodological approach, phenomenology subjectively considers how individuals experience and perceive a given phenomenon. While the initial phenomenological approach is subjective, the goal is more universal: by studying how the knowledge comes into being we can better understand how human consciousness is grounded.

In his book The Phenomenology of Mind, George Wilhelm Friedrich Hegel argues that the whole of history should be viewed and understood as an ongoing process in which humans move closer and closer toward rational self-determination. For Hegel, this rational self- determination serves to bring humanity closer to the "Absolute Idea," or toward the expression of an ultimate truth (21). As such, each individual is an actor, or an active entity, who helps move humanity closer to the freedom and self-expression of the Absolute Idea. Given this, we can view Hegel’s theory as the bridge between the deterministic and compatibilist viewpoints provided previously and the of more contemporary philosophies of the structuralist and functionalist theorists.

For Hegel, the Absolute Idea is expressed individually through each person but is also the result of humanity's collective efforts. Thus, the Absolute Idea is both subjectively expressed by every individual, but it is also objectively created by the whole of humanity. Hegel states,

“everything depends on grasping and expressing the ultimate truth not as Substance but as

Subject as well. At the same time, we must note that concrete substantiality implicates and involves the universal or the immediacy of knowledge itself” (75). Essentially, Hegel argues that

the subject, or individual, cannot be removed from the experience or the understanding of an occurrence.

To provide a simple example, two people watching the same film in the same room at the same time will still experience the film differently. Each person will bring to bear the measure of their full individual experience to the viewing. Given the differences in their past experiences, what resonates with one viewer may fall flat with another. What one views as important to the understanding of the film, the other may not. What one takes as metaphor or symbolism, the other may interpret as literal. Each viewer, as a result, has an entirely subjective experience while viewing the film despite their shared time and environment of viewing it.

The same is true of games, but to a degree further. Two people playing the same game will experience the game differently because of the unique experiences they bring to bear on it.

That said, each person will also have a distinct and unique game experience. Two people playing Tetris or Super Mario Bros will, despite interacting with the same game system, mechanics, and rules, have totally distinct play experiences. They will have played the same game title, while generating very distinct and different game play experiences. Thus, in terms of

Game Studies, the subjective experience of the individual not only defines their unique subjective experience of the game’s system, rules, aesthetics, and so on, but also how the player will define and experience the state of play in the game.

Extending from this, Hegel argues that in any conducted analysis of the event, the subject should not seek to remove him or herself from understanding but should allow their subjectivity to lead them toward a deeper understanding of the universe instead. This is accomplished by the subject becoming aware of himself as an object of analysis that is integral to the process of

being and understanding the world (Hegel 76). Therefore, in a phenomenological approach, the subject pays attention to herself and reflects critically on how her experience and her thinking leads to an understanding of any experience or any object of inquiry. Extrapolating from this, the subject must also consider how her subjective experience measures against those of others and why those subjectivities may be different.

In relation to Game Studies, this concept helps us consider how the player mediates her position as both the game's subject and the game's object. As the interactor with the game system, the player is subject to the game. But, the game also reflects the player's subjectivity – usually in the form of an avatar – to the player too, thereby making them a game object. This dual position, both inside the game and outside it as both subject and object, permits the player to consider her position critically.

Hegel goes on the argue, however, that an awareness of how one individual’s circumstances differ from another’s is not enough for real phenomenological inquiry or understanding. The next and essential step is for each individual to seek a broader understanding of humanity as a whole through these differing subjectivities. Phenomenology then, according to

Hegel, helps us to understand why individuals have different measures of agency at any given time and how those agencies are constructed and experienced differently on the individual and societal level. Importantly, according to Hegel, as we advance our understanding of ourselves and of the world, we become more rational. As we become more rational, we become aware of more choices and how our attempts to actualize any one choice may turn out. By extension of this, Hegel argues, we become freer.

Extending from this phenomenological basis, Martin Heidegger attempts to remove the barrier between subjects and their experiences, as well as between subjects and objects. In “Being and Time,” Heidegger argues that our relationship to the world is not solely mediated by our conscious thoughts as we interact with the world. Instead, he proposes that there are modes of existence in which we blend unthinking or subconscious activities alongside intentional activities (47). Commonly, this blend of subconscious thought and intentional action occurs when an individual engages with a task or with processes that are familiar. The ability to drive a car, for example, blends both unthinking and intentional activities for a driver familiar with and accustomed to the act. The driver knows where the brake and accelerator are without consciously thinking, just as the driver has a concept of where the hood and the tail of the car are in relation to other cars on the road, despite not being able to see both at once. In many ways, the driver mentally becomes the car, moving it through space as if it were an extension of the driver’s self. While operating the car, the driver maintains the intentional ability to navigate between two points too. The driver continues to think critically about the environment around the car and how best to get from point A to point B. In such moments, Heidegger suggests that we blend into the world and to our environment in such a way that subject and object are not clearly defined.

Having established both the conscious and unconscious relationships between subject and object, Heidegger goes on to extend his thinking about objects to humanity’s specific relationship with technology. In "The Question Concerning Technology," he suggests that we should stop thinking of technology as objects and instead conceive of technology as an inseparable part of humanity itself. Rather than being objects, technological tools are a means to

complete a task, and therefore, these tools are extensions of our consciousness and our intentionality. Thus, the essence of technology is not in the tool itself but in the thinking that brings the technology to being (Heidegger “The Question” 12). He states, “If we inquire, step by step, into what technology, represented as means, actually is, then we shall arrive at revealing.

The possibility of all productive manufacturing lies in revealing. Technology is, therefore, no mere means. Technology is a way of revealing” (Heidegger “The Question” 12). A human conceives of a need, brings a technology forward that aids in its acquisition, and from this gain new truth and understanding; she is revealed to herself. Within Heidegger’s mode of thinking, technology becomes an extension of humanity and an extension of the subject. Or, perhaps, to put it more correctly, the subject and object disappear, and there is simply humanity at work.

Heidegger’s philosophy regarding the relationship between humans and technology is not dissimilar from the concept of cognitive flow in digital games. Often cited within Game Studies,

Csikszentmihalyi’s concept of flow is bound up in feelings of pleasure and can be summarized as, “an emotional and psychological state of focused and engaged happiness, when a person feels a sense of achievement and accomplishment, and a great sense of self” (Salen and Zimmerman

336). While in a state of flow, the player experiences a sensation of extreme focus, the merging of action and awareness, a sense of control, and the distortion of time (Baron). Thus, both

Csikszentmihalyi and Heidegger note that when a human interacts with technology, the potential exists for a merging between the two, such that the distinction between the human and the technology is lost. The user of the technology views the technology as an extension of the self.

As with the automobile example used previously, many players could be said to reach a state of flow when they cease to consciously think of the buttons they need to push on the controller in

order to actualize their intentions in the game. But, Heidegger’s theory of flow can extend further in relation to Game Studies. Given that games are technological tools, players can enter into a state of flow not just with the controller but with the game itself. The ability for a player to connect with the game, with its mechanics, and with its playable elements is vital to the construction of the player’s agency. To maintain agency within the game, the player must feel one with it and must be able to move through it as if she were part of it. In Game Studies, this is generally referred to as "immersion." What Heidegger's theory suggests, however, is that without this state of flow or connectedness with the game, the player cannot move forward with agency.

Both Hegel and Heidegger remind us that the ways humans think about themselves, their relation to, and their interaction with the world and with objects are other essential components of agency. Neither our thoughts nor our attempts to act with intent are isolated from our experience with the world. Therefore, any construction of agency in game must not only account for the player’s thought processes and actions, but also for how those thought processes and actions will reverberate in the world. Considerations, such as these, have largely been at the center of Science and Technology Studies.

Science and Technology Studies

Within the field of Science and Technology Studies, theorists study how human values affect the innovations of science and technology and how those innovations, in turn, affect human values. Within this area of study, human agency is studied alongside technology as theorists attempt to understand how each one reconstitutes and informs the other.

For example, in How We Became Posthuman, N. Katherine Hayles considers the potential dangers, as well as some of the benefits, of posthumanist conceptions of humanity. As defined by Steve Nichols, the posthuman is a “speculative being," who is, “not a singular, defined individual, but rather can 'become' or embody different identities and understand the world from multiple, heterogeneous perspectives.” The danger of posthumanism, for Hayles, is that it fractures the connection between the mind and the body. It makes our humanity something speculative, immaterial, and ethereal. She writes, "Here, at the inaugural moment of the computer age, the erasure of embodiment is performed so that 'intelligence' becomes a property of the formal manipulation of symbols rather than enaction in the human life-world"

(xi). Hayles's key concern is that as we move further into the digital world and into posthumanist theories, the body is being lost from focus. Our discourse about human subjectivity has, as a result, become connected with a consciousness that functions as an informational pattern rather than as a consciousness that is both embedded and embodied

(2). Disgruntled by this, Hayles argues that the mind and body are inseparable and that information is not immaterial at all. She states, "What embodiment secures is not the distinction between male and female or between humans who can think and machines which cannot.

Rather, embodiment makes clear that thought is a much broader cognitive function depending for its specificities on the embodied form enacting it" (xvi). In other words, consciousness is an embodied state of being.

Elsewhere, in the keynote speech "Deeper into the Machine: Learning to Speak

Digital," Hayles returns to and extends this theory. Here, she argues that any technology's materiality exists both in its machinery as well as in its "incarnation of assumptions embedded in

its form and function" (375). In other words, as human-created tools, technology reflects and embodies our intentions and our agency. They are extensions of ourselves. For Hayles, this is the positive result of being posthuman: our embodiedness exists both within our singular bodies but also in the materiality of the technologies we share. Our interactions with technology are a collaboration that helps us reconstitute our assumptions, expectations, and intentions. Thus, while there is no technology without a human body to constitute it, Hayles also suggests that the posthuman uses technology as a means to constitute itself too. Importantly, however, Hayles also reminds us that these intelligences and technologies are linked to our embodiedness. The ways we collaborate and interact with the game and its intelligences make possible our thinking and our doing. Because Hayles argues that critical thinking is itself a form of action and agency in which we negotiate our state of being, possibility, and potentiality, she allows for the possibility that any conscious interaction between an individual and technology is a means to constitute or express one's agency in collaboration with others. This creates an ever-expanding feedback loop between our individual agency and our collective agency via our interactions with the technologies we have created and with which we interact.

In relation to games, Hayles argues that any media resulting from technology should be thought of as "collective intelligences that explore their conditions of possibility by trying to discover what they are good for. These attempts, in turn, feed back into technological innovation to transform their conditions of possibility" (How We Became Posthuman 371-372). In other words, when a player interacts with a game, they are necessarily interacting with the collective intelligences of others. These collective intelligences are not just those of the individuals who

developed the game, but of those the developers were informed by and of those the developers made allowances for as they developed the game. What Hayles suggests is that our interactions with media are always, then, a collaborative act. This is certainly truer in video games than in other media forms. That collaboration, however, is inextricably linked to our embodied consciousness and to the assumptions we make about the technology or media with which we are interacting. When we interact with a digital game, we are not only interacting with the code of the game, but with the machine that runs the code. Both of these technologies link us to other collectives while the technologies themselves reconstitute our own embodiedness. Thus, in our interaction with games, the player’s conception of her agency is always mediated by her interaction with the technology she uses to play the game as well as by those who informed the creation of the game.

While N. Katherine Hayles is concerned about posthuman constructions of information and consciousness, Donna Haraway is interested in where human organisms and technology collide. In Simians, Cyborgs, and Women: The Reinvention of Nature, Haraway constructs a form of agency which she then links to the cyborg, or a "hybrid creature, composed of organism and machine" (1). Thus, while the posthuman is constructed in the absence of a corporeal or material form, the cyborg is dependent upon it. Working in a symbiotic relationship, the cyborg requires both an animate, conscious body and technology. Thus, Haraway argues that the cyborg body heightens our connection with tools (178). Furthermore, Haraway argues that previous philosophies – such as the posthuman – have made the error of assuming a “deep and necessary split between nature and culture and between the forms of knowledge relating these two putatively irreconcilable realms" (8). Haraway goes on to suggest that the cyborg, an already

natural part of our lives, could reconstitute agency outside of social constructs of control.

Understood as a hybrid being, the cyborg is neither one entity, nor two, but instead, holds multiple identities. As a hybrid, the cyborg does more than mediate the boundaries of fractured identities and dualities. Rather, Haraway notes, the cyborg is part of those boundaries and is thus

"a way out of the maze of dualisms in which we have explained our bodies and our tools to ourselves" (181).

One position Haraway helps us consider in Game Studies is that the player and the game share a cyborg relationship. Given this logic, there is no longer the need to think of the player's agency as something that works in opposition to the rules and the system. Rather, both the game and the player create affordances for each other and, in some capacity, require each other in order for the experience to take place. The game, in terms of its hardware and code, exists without the player just as the player exists outside the play of the game. That said, the experience of the gameplay requires that both work in tandem. As such, the game and the player are unified in a system that both transcends each one while simultaneously reaffirming each other too.

If we are to take Hayles and Haraway at their words, then the technologies and tools that we use are as inseparable from consciousness as from embodiment. Given this, those same technologies and tools are also linked to our agency and how it is negotiated and constituted.

Actor-Network Theory, developed by Bruno Latour, is concerned with exactly how and to what capacity non-human objects affect individuals and their interactions within social systems. Latour states, "If action is limited a priori to what 'intentional', 'meaningful' humans do, it is hard to see how a hammer, a basket, a door closer, a cat, a rug, a mug, a list, or a tag could act…. By contrast, if we stick to our decision to start from the controversies about

actions and agencies, then anything that does modify a state of affairs by making a difference is an actor – or, if it has no figuration yet, an actant" (71). While Latour argues here for a broadening of our understanding of agency, he is also careful to clarify that objects do not necessarily determine action, but rather that they can influence, block, or make possible certain actions. For example, while having a car does not determine in and of itself how an individual might travel from one place to another, the ownership of the car does present the person with options beyond those of someone who does not have one. It presents the person, therefore, with different modes of experience, interaction, and agency. Even given the same starting point and ending point, two people traveling in their cars will still possess different levels of agency as they can elect their speed, route, position on the road, and so forth. In both cases, the object does not determine the action of the individual, but it does alter the possibility and the means by which the individual can actualize their intent.

While Actor-Network Theory certainly highlights the importance of the relationship between a player and the hardware that makes gameplay possible (the console, the controller, the disk, etc.), the theory is no less important when we consider how players interact with objects in a game's environment. In this respect, Latour reminds us that a game, by extension of the objects it allows the player to interact with, necessarily affects how a player can make meaning or express intent. Whatever limits the game imposes affect the means and the possibilities by which a player can actualize her intent. This seems like the most obvious observation one could make about player agency, and it is this observation that is so frequently at the heart of the discourse surrounding agency in games. A game that provides a player with agency, it is frequently argued, gives the player choices. These choices are, in turn, frequently reduced to the interactions the

mechanics of the game make possible. Of course, the question of agency extends far beyond this point, as I have already shown. Even in this respect, Latour goes on to complicate the simplicity of this assumption.

Extending Actor-Network theory further, Latour also suggests that individuals are not always conscious of their actions. He writes, "action is not done under the full control of consciousness; action should rather be felt as a nod, a knot, and a conglomerate of many surprising sets of agencies that have to be slowly disentangled" (44). In other words, our actions are not always intentional or rational. While other theorists would suggest that unintentional or irrational acts negate agency, Latour is less sure this is true. Rather, he suggests that when we act subconsciously, we do so because there are numerous sets of agency happening simultaneously.

Agency is itself not only flexible and intermittent, as previous theorists have suggested, but also of a type. In other words, there is no one form of agency, but many, each affecting the others.

Just as Hayles suggests that our interactions with technologies are always collaborative interactions between ourselves and the intelligences that bring the technology forth, Latour suggests that our interactions with objects are frequently wrapped up in a multitude of agencies. Extending from this, we can approach games as systems that allow for interwoven types of agency rather than as systems that allow for only a singular form of player agency. Furthermore, Latour reminds us that these agencies may or may not be in conflict and that the player may or may not be conscious of the agencies currently being negotiated.

For Latour, the necessity to unite the complex interplay between these types of agency is unnecessary (45). In fact, he strongly cautions against efforts to prescribe any agency to a particular means. "Action" Latour states, "should remain a surprise, a mediation, an event"

(46). That action can remain so is due to agency which, according to Latour, either does something (and exists) or does not (and doesn't). He states, "If you mention an agency, you have to provide the account of its action, and to do so you need to make more or less explicit which trials have produced which observable traces" (53). Tracking these observable traces is, then, the work of Actor-Network Theory. As such, Actor-Network Theory does not attempt to prescribe action to a set of pre-determined criteria. Instead, it attempts to find order within chaos rather than to apply order where it may not exist amidst chaos.

Certainly, it is true that the social systems with which Latour works are far too complex to map out. Social scientists cannot, nor should they, attempt to account for every agency or how it can be enacted. That said, given that games are far smaller systems with strict boundaries and rules, it is perhaps not inconceivable to account for the agencies present in any one game or the actions those agencies makes possible. While this mapping of agency is possible – if not always feasible – it is important to remember that it remains impossible to account for the individual agencies any one player brings to the game. Thus, one distinction we must mediate in

Game Studies is the difference between the agencies the player maintains as an individual outside of the game, and those the game makes possible. Furthermore, given these competing forms of agency, we must remember that the play experience of any player remains, ultimately, unknowable or indefinable. Yet, we might accept that the concept of flow, mentioned previously, is the point at which the agencies the game affords and the agencies the player brings to the game sync up. If, as Latour states, many actions are taken without a conscious awareness of the agency one holds, it is reasonable to conclude that the external agencies of the player are most in sync with the agencies permitted to her by the game system when the player does not

consciously need to think about the agency she does or does not hold. In a flow state, the player instead maintains a state of unconscious connection with the game and does not have to consciously negotiate between what is possible or what agencies are currently present to be negotiated or activated, but can instead simply work in harmony with the objects at her disposal to accomplish a goal.

Conclusion

The theories examined here are merely the tip of the iceberg. This chapter has not, for instance, considered the theories of agency found in poststructuralist theories, in feminist theories, queer theories, or disability theories to name but a few. While these areas of discourse would certainly extend the conversation further, the intent of this chapter is to broadly establish the value that social theories of agency can bring to the discussion of agency within the field of

Game Studies.

If we return now to the definitions of agency within Game Studies presented at the beginning of this chapter, we can see how insufficient they are. Within the field of Games

Studies, scholars tend to talk about agency in one of several ways: by discussing choice and how much of it a game gives to its player, by assuming agency is actualized by taking action, by suggesting that agency is the player's way of making meaning of and within the game experience, or by suggesting that participation and interaction with the game are equal to the player's level of agency. While the social agency theories examined here do not discredit the importance of action, choice, meaning-making, participation, or intent, they also expose considerations that, within Game Studies, have been ignored or not adequately addressed. These considerations include but are not limited to:

• Thinking of the game as both subject and object as well as of the player as both

subject and object.

• The unification of these subjects and objects into new amalgams of potential (i.e.,

the posthuman or the cyborg).

• The difference between conscious agencies and unconscious agencies and how

those agencies alter the player’s capacity to sync with the agencies of the game

into a state of flow.

• The multiple types and forms of agency present at any moment in the game.

• The agency of the game as an interactive object and how the game's

agency alters the player's agency.

• The player's perception of what is and isn't possible within the determinant system

of the game and how those perceptions change throughout gameplay and through

the player's negotiations with the game's limits and agencies.

• The subjectivity of each player and how it alters their perception of what is and

isn't possible in the game or of how the player can make meaning of what is and

isn't possible in the game.

• The degree to which the agencies or the player and the game change during and

throughout gameplay.

• The importance of alternative means by which to work toward a goal.

• The importance for the player to feel compelled to work toward a goal.

• The critical thinking processes of meaning-making, such as forethought, self-

reactiveness, and self-reflection.

Given these considerations and how complex many of them are, it is fair to say that our understanding of agency in games is, as yet, topical at best. By extension, the definitions for player agency, including mine, fail to capture adequately what agency in games is, and by extension, how an interactive narrative game might construct or make room for it.

While Latour cautions against the impulse to define the conditions of agency in social systems, there is still a benefit in trying to articulate its parameters within the field of Game

Studies. I would, therefore, like to propose a new definition for agency in digital interactive games. For the remainder of this work, agency will be understood as the capacity of a socially situated player to engage meaningfully, either consciously or subconsciously, with a game through embodied actions and critical thinking processes in order to actualize intent via the execution of choices, either expressed physically or mentally, in order to achieve a self- selected goal present within the game state and space.

Still, there is much work left to be done, and this text cannot account for all of it. Continued scholarship by Game Studies scholars in relation to agency theory should continue in close consideration with broader social agency theories. While no social agency theory will be entirely transferable to game systems or to how players interact with them, the knowledge gained from these theories can help us to ask and to answer deeper and more complex questions about player agency in relation to digital games. Furthermore, the consideration of social agency theory extends our thinking about agency beyond the question of what the game permits and moves it close toward a more complex understanding of agency that considers the experiences players and creators bring to the game and its systems.

Moving forward, the focus of this text will be on just a small number of the considerations highlighted above. That is, this text is most interested in considering the numerous types of agency at work when a player engages with a game system that contains an interactive narrative system. In other words, I will proceed with Latour's understanding that there is no one type of agency, but instead, numerous types of agency that are constantly in flux and which are being negotiated by individuals and their situations and experiences. This text will also consider the critical-thinking processes of players, as proposed by Bandura, in relation to how the player makes meaning of and with the game and how this process leads to an actualization possessed by intent for the player.

PART TWO: THE ANALYSIS OF NARRATIVE AND AGENCY IN DIGITAL

INTERACTIVE NARRATIVE GAMES

AUTHOR’S NOTE CONCERNING CHAPTERS 5-7

The next three chapters were all published before the writing of this volume and have been replicated here with permission from each publisher.

Chapter 5, “Creating Collaborative Criteria for the Analysis of Narrative and Agency in

Digital Interactive Narrative Games” was published by The Computer Games Journal in Volume

4, Issue 1 in June 2015.

Chapter 6, “Assessing Mass Effect 2 and Elder Scrolls V: Skyrim: Using Collaborative

Criteria for Player Agency in Interactive Narratives” was published by Journal of Game

Criticism in Volume 2, Issue A in July 2016.

Chapter 7, “Kentucky Route Zero: Or How Not to Get Lost in the Branching Narrative

System” was published in the edited collection The Player Versus Story Divide in Game Studies:

Critical Essays. This collection was edited by Matthew Wilhelm Kappell and was published by

McFarland Press in November 2015.

I am grateful to each publisher for permitting the reproduction of these publications within this volume. That said, several caveats need to be made before proceeding.

These chapters have been reproduced here with no variation from their original publications. As each chapter was initially written as a stand-alone piece, some redundancies in information will occur.

Though these three chapters appear in Part Two of this volume, they were composed and published before the chapters found in Part One. As a result, these three chapters lack specific references to the ideas proposed in Chapters 2 and 3. While I believe the reader will be able to

100

draw the connections for themselves, I regret that in reproducing these chapters here, I cannot explicitly make the connections for the reader.

As each chapter went through a unique editing process, some small word changes may occur between the criteria I establish and use. Changes in formatting, graphs, and citation style are also a reflection of each publication’s requirement. These differences are not, however, detrimental to meaning and are similar enough to maintain their purpose. In Chapter 8, I provide the criteria again and using the wording I currently think best.

As a final note, Chapter 8 has been composed with the intention of revisiting the methodological framework and criteria outlined in Chapter 5 and employed in Chapters 6 and 7.

Chapter 8, therefore, clarifies where and how my thinking has evolved since these chapters were originally published. Additionally, Chapter 8 makes explicit reference to the ideas laid out in

Chapters 2 and 3. As such, Chapter 8 acts as the unifying piece between all chapters and the wording used within it should be considered the definitive or “author preferred” text.

101

CHAPTER 4

METHODOLOGY FOR CHAPTERS 5-9

In Part One of this volume, I employed several methodologies to create new theories for both narrative and agency in digital interactive narrative games. Now, in Part Two, my intent is first to generate a methodology and criteria for assessment that can be used alongside the theories generated in Part One, and second, to test the usability of this method and these criteria in a series of critical applications.

In Chapter 5, I generate a set of criteria that can be used to assess the presence, structure, and functions of narrative and agency within digital interactive narratives. While the work of the previous section was to identify the limitations of current narrative theories and agency theories in relation to digital interactive narrative games, the goal of this chapter is to work in collaboration with recent digital games scholarship to create critical criteria which can be employed to analyze particular qualities of narrative and agency in relation to any digital interactive narrative game. That said, these criteria were created by adhering to a specific methodology and set of guidelines.

While I provide a detailed description of my process in Chapter 5, a quick summary of that process is warranted here first. The process for creating the criteria for assessment began with content analysis and close reading. As I read through articles created by narrative scholars, social agency scholars, and digital interactive game scholars, I memoed any occasion in which a scholar identified a criterion they felt was necessary to the creation of a narrative or the creation of agency. Having then assembled this list of criteria, I conducted a comparative analysis of them in order to locate any criterion which no less than two scholars, myself

102

excluded, had identified. Next, having minimized the list to a collection of mutually established criteria, I used deductive categorization to organize those criteria into five specific categories for assessment: narrative experience, ludic agency, narrative agency, character development and interactions, and drama management system requirements. I have developed the criteria into these five areas of assessment so that, based on one's need, one can conduct an analysis of a digital interactive narrative game using any or all of the areas of assessment. The result, then, is a set of five categories for assessment, each with a unique set of criteria by which someone can qualitatively assess the absence of presence of those criteria within a digital interactive narrative game.

In Chapters 6, 7, and 8, I conduct a series of critical analyses in which I assess the five categories according to their criteria within the following digital interactive narrative games:

Mass Effect 2, The Elder Scrolls V: Skyrim, Kentucky Route Zero, and The Banner Saga. Each of these titles was primarily selected because it can be classified as a digital interactive narrative game. Beyond this, these titles were also selected for analysis because they differ from one another considerably regarding production value, play style, narrative genre, game environment, and drama management systems.

To begin, Mass Effect 2 and Elder Scrolls V: Skyrim are both AAA games, meaning both games were developed by large teams and were backed by large budgets. By comparison,

Kentucky Route Zero and The Banner Saga are both independently developed games, meaning their development teams and budgets were substantially smaller. This distinction is important and helps to determine the merit of the criteria at both ends of the development spectrum. By

103

assessing both AAA and indie games, I aim to show that the criteria do not contain any biases toward a particular value of production.

Beyond the size of the development team and funding, the games also differ in terms of their play styles. Mass Effect 2, for instance, is classified as a single-player, action role-playing game presented in third-person perspective. Skyrim is also an action role-playing game, but differs from Mass Effect 2 in that it is an open-world game and allows the player to switch between first- and third-person perspective. Kentucky Route Zero, on the other hand, is an episodic point-and-click adventure game presented in a third-person perspective. Finally, The

Banner Saga is a tactical turn-based role-playing game presented in third-person and aerial third- person perspective. Importantly, in the case of The Banner Saga, the player cannot choose when to switch between the two perspectives as the play can in Skyrim. By selecting games that represent a variety of play styles with different perspectives, I aim to show that the criteria are not limited or biased in any way that privileges one play type or player perspective over another.

In addition to representing a spectrum of play styles and player perspectives, the four games also contain very distinct plots and narrative genre types. In narratological terms, Mass

Effect 2 is a space opera in which the protagonist, Shepard, must assemble an intergalactic and interspecies crew to defeat the Reapers, a race of organic-synthetic hybrids who threaten all other lifeforms in the galaxy. Skyrim is a fantasy tale in which the protagonist is a “Dragonborn,” or one who can absorb the souls of the dragons that plague the land and use this absorbed power to defeat Alduin the World-Eater, a dragon prophesied to end the world. Kentucky Route Zero can best be classified as a tale of magical realism. Here, the plot is simple: a character named

Conway must deliver a package to someone in Kentucky who lives somewhere on the stretch of

104

road known as Route Zero. Of course, given its magical realism classification, Route Zero is not merely a road, but a pathway to places that exist out of place and time with the regular world.

Finally, The Banner Saga is a Nordic epic or legend in which the heroes, a group of giants and humans, must wage war against an enemy known as the Dredge while simultaneously holding off an evil darkness that threatens to eat the world entire. By selecting an array of narratological genres, each with a unique plot, I aim to show that the criteria are not biased toward a particular narrative style or a particular set of story elements.

As noted in Chapter 2, the environment in a game is of critical importance and functions as more than a setting. The player moves through the spaces and environments in the game and interacts with objects within that environment. As such, in selecting games, I was careful to select games with distinct environments and specific means by which to interact within those environments. Mass Effect 2’s environment can most broadly be defined as a “space” environment. In Mass Effect 2, the player moves between three basic environment types: space ships; highly inhabited city marketplaces; and nearly desolate, enemy-held outposts. Alternately, as the fifth installment in The Elder Scrolls series, Skyrim deals with a particular region of a much larger world called Tamriel. This region, not surprisingly called Skyrim, is in the Northern part of the world and, as such, presents environments and landscapes akin to those found in the upward reaches of the Northern Hemisphere: mountains, snow, tundra, and grasslands. While

Skyrim presents an immense open-world environment, the player can still be found in one of four general environments: big cities, rural towns, dungeons and catacombs, and open terrain. The scale of Kentucky Route Zero’s environment is much smaller, as all environments in the game exist within a small pocket of rural Kentucky. The environments the player interacts with are

105

also, then, much smaller and consist of homes, churches, businesses, gas stations, and similar landmarks. Finally, the environment of The Banner Saga is not dissimilar to that of Skyrim, as it too provides an icy northern landscape that stretches across a large kingdom or country. That said, the environment in The Banner Saga is more desolate and uninhabited given the near- apocalyptic plot. While the player moves through cities, rural towns, and open terrain, nearly all of these are sparsely inhabited or already abandoned.

Of course, environments are important in games because the player interacts with them.

As such, my consideration of environment also extended to the mechanics available to the player that allow him to interact within the game environment. While I discuss this much more thoroughly within the individual analyses found in Chapters 6 7, and 8, I will draw some cursory distinctions between the mechanics here to show the methodology I employed in selecting them.

In Mass Effect 2, the player has some liberty in choosing where they go, but their options are always limited by where they are located within the game’s branching narrative structure.

Furthermore, once certain locations have been selected, the player cannot leave until the mission is complete. This is most common in missions that require combat. Beyond these environmental navigations, the player can interact in two key ways: dialogue and combat. Dialogue is presented using a radial dialogue menu while combat is akin to that of a third-person shooter. In Skyrim, the player is not restricted in how they navigate the open-world except while in a boss encounter.

Otherwise, the player is free to travel throughout the environment at her own discretion and whim. In Skyrim, the player can also interact through dialogue and combat, much like Mass

Effect 2, but the player has more ability to specialize combat through the use of unique weapons, clothing pieces, skill trees. In Kentucky Route Zero, the player is considerably restricted in how

106

they move. In this point-and-click game, the player can select where on the screen the character agents will move to, but the only other interaction available to the player is interaction via a dialogue selection menu. In The Banner Saga, in order to maintain the game’s linear narrative structure, the player is restricted in where they go. Outside of combat, which is turn-based and much like a chess match, the player’s only means of interaction with the environment is via one of two interfaces: a dialogue interface and a morale interface that allows the player to select when to rest, when to eat, when to train, and when to move the caravan along the narrative structure’s determinate path.

Finally, in selecting these four games, I considered how each game employs the use of a drama management system. The drama management system in Mass Effect 2 is overt and easy to detect. One component of the drama management system is the radial dialogue menu. This menu presents dialogue options to the player. These options often correspond with the game’s morality scale. When a player selects a dialogue option, she receives feedback from the drama management system indicating how her dialogue choice scores on the game’s morality scale.

This process is also used in the game’s QTEs. When a player make a choice during a quick-time event, the drama management system will provide feedback to the player as to whether her choice was morally good or morally bad. Beyond the immediate feedback presented to the player when she inputs a selection, the player can also view her morality status at any time by accessing the character interface screen, another component of the game’s drama management system. In

Skyrim, the drama management system is more hidden than the drama management system in

Mass Effect 2 but is still present and detectable. In Skyrim, the player has multiple character interfaces with which to interact, and all could be said to impact the player’s perception of the

107

game’s unfolding drama. One of these interfaces is, again, a dialogue selection menu. In the case of Skyrim, no morality is attached to dialogue options, and no feedback is given to the player beyond the continuation of the conversation. Using additional character interfaces, the player can track and select quests, change attire and weaponry, consume potions and foods, and select skill upgrades and magical abilities. While the players can customize their experiences through these character interfaces, these interfaces do not provide feedback to the players that assess or catalog their choices in the game. In Kentucky Route Zero, the only interaction the player has with the drama management system is via the dialogue selection menu. The game, however, does not present the players with any feedback based on their dialogue selection, beyond the appearance of more dialogue. Beyond the dialogue selection menu, players are not presented with any other interfaces or the ability to see how the game is tracking, judging, or altering the story they are developing through the dialogue input. In The Banner Saga, a Heads-Up Display (HUD) is always visible to the player. This HUD, displayed at the top of the screen, is also a representation of the game’s drama management system. It relays key information to the player, which the player can then use to affect the game’s unfolding drama. The display tells the player their current renown (gained from defeating enemies), the number of fighters, non-fighters, and Varl currently in the caravan (important for combat and the rationing of supplies), how much food the caravan has, the caravan’s morale level (which affects the amount of willpower available during combat), the number of days since the player last rested (affects morale), and the option to camp or rest. While the game also presents the player with a dialogue selection menu, it is this HUD that provides the player with constant feedback about her status in the game.

108

After I had selected these four games, the last step in my methodological process was to play the games and assess them based on the established categories and criteria. While I would consider this process of analysis similar to that of “close reading,” my position as both the player and the assessor makes this system of analysis just as closely akin to hermeneutic phenomenology. Given this, my analysis remains conscious of the fact that I am always already existent in the world as both a scholar of games and as the player of these four specific games.

Beyond that, however, my analytical process also remains aware of the multiple roles a player is asked to inhabit, especially the joint roles of narrator and character, and the additional worlds and environments the player inhabits at the same time.

Without access to the codes that make up each game, the analyses of the games can only come from the experience of playing them. Given this, in my phenomenological approach, I also collected experiential material by recording my play sessions of the games and watching recordings made by other players. Such recordings are referred to as “Let’s Plays.” Broadly defined, Let’s Plays are audiovisual recordings that capture digital play. Additionally, the player also records his or her audio commentary while in the act of play. Such commentary often reacts to or critically reflects upon the game content and the act of play itself. Given this, Let’s Plays act as a rich medium for phenomenology as they capture particular experiences of play as well as the variances between those experiences. In addition to recording my play sessions, I collected

Let’s Plays made by other players for each of the four games. This was an important part of the process for several reasons: 1) each play session of a game is unique, and thus, the Let’s Plays of others allowed me the observe differences in gameplay styles and outcomes. 2) In digital interactive narrative games, the outcome of a playthrough, based on the player’s input, is also

109

unique, and thus, the Let’s Plays of others allowed me to observe how a player’s input and choices did or did not alter the outcomes of the story experience. 3) Text-based reviews and walkthroughs do not document player’s reactions or critical analyses with the same immediacy, and thus, the Let’s Plays of others allowed me to observe the second-to-second feedback and commentary of the players as well as their more reflective comments about the total experience.

In other words, Let’s Plays capture the unique performativity of games. As Kris Ligman observes, Let’s Plays allow us to capture and document personal performance, reaction, and interaction with the medium itself. Not quite like any other medium due to its interactivity, the study of digital interactive games requires consideration of the game as well as of the player in the action of play. Let’s Plays capture this otherwise ephemeral act, and were, therefore, an important medium of study in my process.

Taken together, these Let’s Plays acted as the documents against which I observed and applied my analysis of the five categories of assessment and their accompanying criteria. The results of these analyses make up Chapter 6, an analysis of Mass Effect 2 and Skyrim; Chapter 7, an analysis of Kentucky Route Zero; and Chapter 8, an analysis of The Banner Saga.

Part Two of this text concludes with Chapter 9. This chapter compiles the analyses of the games together in order to analyze what the criteria reveal about each area of assessment and to propose future considerations for the continued development and refinement of the criteria.

In the future, I recognize that it will be most beneficial to have multiple people play each game and to employ the criteria to assess the games. For the purposes of this initial research, however, I am the only player. In this way, while the criteria I created to assess the games were compiled to represent the views of numerous Game Studies scholars, the actual analysis of each

110

game was conducted solely by me and is, therefore, biased to my interpretation. This does not, of course, limit the merit of the analyses, but does highlight the potential for future growth. By asking other scholars to engage with these criteria, I am confident the criteria will change for the better. I welcome and look forward to the opportunities to re-examine these criteria and continue to grow this method of analysis.

111

CHAPTER 5

CREATING COLLABORATIVE CRITERIA FOR THE ANALYSIS OF NARRATIVE AND AGENCY IN DIGITAL INTERACTIVE NARRATIVE15

Abstract Although the intermingling of game and narrative in digital media may seem an easy match to the lay game consumer, the relationship between ludic play and narrative play is quite complicated. Given that a key feature of play is agency, and that a key feature of narrative is the listener/reader/consumer’s position outside the text, how can the player maintain a sense of agency in the narrative as well as in the state of play? Individual scholars and game developers have been attempting to successfully balance ludic and narrative agency for some time now, but their experiments tend to yield different (though not incompatible) criteria for analysis. As a result of such isolated experimentation, little progress in this ﬁeld has been achieved. Rather than contributing to such isolated scholarship and development, I intend to show that by establishing collaborative criteria, useful and progressive data can be harvested that will aid not only in the evaluation of current digital interactive narratives but in the production of future ones as well.

Keywords Agency, Interactive narrative, System design, Criteria, Analysis, Experimentation,

Collaborative, Ludic play, Games

15 From The Computer Games Journal, “Creating Collaborative Criteria for Agency in Interactive Narrative Game Analysis” Volume 4, Issue 1, 2015 p. 47-58, Lindsey Joyce. ©Springer. With permission of Springer

112

1 Introduction

The relationship that exists in digital interactive narrative games between user agency (the capacity of the player to make meaningful choices in the game) and the game’s narrative (the structure and function of story embedded in the game) has been the subject of extensive research.

Such discussion at one point became such a source for scholarly contention that it is now known as ‘‘the ludology versus narratology debate’’. In some capacities, the debate was more about modes of study. Some scholars have claimed that games are distinct; for example, Johan

Huizinga’s understanding of play was that it is ‘distinct from all other forms of thought in which we express the structure of mental and social life’. Others claim that games are stories; Janet

Murray argued that games and stories mutually share two essential features that bind them— contest and puzzle—and she considered the study of games as a unique and separate ﬁeld of study from that of other media (Huizinga 1955; Murray 2004). More importantly, the debate has also addressed these concepts in reference to an interlocutor or consumer of the media itself. It is within this context that the difﬁculty between narrative and agency arises.

The central agreement that many scholars have retained is that the relationship between narrative and agency is an entangled one; to increase the strength of the narrative is to decrease that of agency, and vice versa. Despite this inverse relationship, there still exists the desire to create interactive narratives, in which the player is both presented with a story, and is allowed to make meaningful choices that impact or alter the outcome of that story. Both game studies scholars and game developers alike have been attempting to create digital interactive narrative games that provide the player with a game experience that successfully balances the two.

Perhaps not surprisingly, given the complication of reconciling agency and narrative, several

113

previous attempts have made little headway. While speciﬁc problems may be addressed by any one game, the overall experience of such games generally remains unbalanced and unsuccessful in other capacities. For example, in Bethesda’s Skyrim (2011) the player has a high level of ludic agency and narrative agency, but the narrative ceases to be coherent—the more the player engages with the system, the more static the non-player characters seem to appear. Alternatively,

Bioware’s Mass Effect 2 (2010) has rich character interactions and a cohesive narrative experience, but the game stiﬂes the player’s perception of ludic and narrative agency by using heavy-handed drama management systems. Riedl and Leon (2008) deﬁned drama management as ‘a technique by which a virtual world and computer-controlled characters within the virtual world are manipulated in order to bring about a dramatic experience for the human participant’

(Riedl and Leon 2008). Here, drama management systems should be understood as those systems, such as branching narrative, morality scales, heads-up displays (HUDs), and menu screens, which track, alter, and report narratively important information to the player.

The collective stumbling over how to reconcile narrative and agency cannot solely be blamed on the difficulty of their reconciliation. The blame must also be apportioned among scholars and game developers who, despite their earnest attention to create interactivity in games, are less interactive in their communications with one another, and may not be fully aware of the collective research that has accumulated in the field. The study of digital games and of interactive narrative is still a new field of research, and researchers’ self-criticism of their own work may impede progress, or at the very least slow progress significantly, rather than spur it forward. In addition, although Murray (Murray 2013) declared in 2013 that ‘the

Ludology/Narratology discussion has moved on’, the rift it created has not yet been ﬁlled. If the

114

focus could be shifted away from contention and disagreement towards collaboration and agreement among researchers and developers, a solid foundation for game creation and analysis could be realised.

With this spirit of collaboration in mind, and with the aim of drawing more attention to agreements in the ﬁeld (rather than disagreements), I have conducted a cross-examination of key research in the ﬁeld of interactive narrative. Through this examination, I have compiled a list of essential criteria that should be both considered and included in the creation of digital interactive narratives in order to best balance the relationship between story and agency. This list of criteria can be applied to more meaningful and useful analyses of digital interactive narratives, especially of AAA games that have a large impact, wide distribution, and diverse user-base.

2 Criteria

In order to establish a list of criteria for consideration in the analysis of digital interactive narrative games, I focused my search using the following approaches:

1. Experiencing from the Player’s Perspective Because I do not have access to the source code

for any AAA games, my analysis must rely on criteria that can be examined from the

player-side of the game experience. Given that the aim of digital interactive narrative games

is to engage the player in the story, focusing on the criteria from the player’s perspective is

appropriate. Furthermore, this is probably the best way to induce a collaborative analysis, as

many scholars tend to research interactive narrative by producing their own criteria and

applying them to their own unique experiments. Such experiments exist in a vacuum and are

rarely experienced by a public audience in the way AAA games are. Therefore, an analysis

115

of criteria in games to which a large audience has access is useful if games in such spaces

are to be improved.

2. Seeking a Second Opinion Only criteria, which have been agreed upon by at least two

scholars as being important to the overall success of the interactive narrative experience,

have been included.

3. Requirements I grouped the criteria into ﬁve key areas of requirement: (1) narrative

experience, (2) ludic agency, (3) narrative agency, (4) character development, and (5)

interactions and drama management requirements. These requirements are discussed further

in the next Sect.

4. Producing a Concise List of Criteria I rearticulated the essential criteria into a concentrated

list that addresses the established criteria into a more concise list, without sacriﬁcing

individual scholar’s intent or meaning.

3 Requirements

Before applying these criteria to interactive narratives published within the realm of AAA games, an extended explanation of each of the ﬁve key areas, and the criteria established within them, is necessary. Thus, while the logical next step in the creation of these criteria is to use them as analytical tools in order to gauge the balance of agency and narrative in speciﬁc games, this paper seeks only to establish usable criteria.

116

3.1 Narrative Experience

It is important to analyse the criteria that are essential to the creation of the overall narrative experience, as the other criteria must work to support the outcome of the narrative experience.

The contention between ludology and narratology also stems from complications in deﬁning criteria for narrative experience. Some scholars such as Gonzalo Frasca have argued that play should be prioritized over narrative, and that the interactive elements take precedence over narrative experience. For instance, Frasca argued that The Sims (Maxis/EA 2000) is an interactive narrative, whereby the narrative is constructed solely by the player. In his review of the game, Frasca (2001) stated that the family albums crafted by players can be read as stories, and went on to argue that, ‘Unlike other action video games, where players record their performances on video in order to analyze or show off their skills, the family album storytellers do not focus on the game itself, but use the feature as a narrative tool’. Conversely, other scholars have prioritized the narrative experience over the play experience. For example, Julie

Porteous, Fred Charles and Marc Cavazza experimented with an interactive experience, whereby the user could select certain personalities, relationships, landmarks, etc. for several non-player characters. Once the user selected these elements, a narrative was generated and played out for the user, much like a TV show. Thus, the user only engages in play at the front-end of the experience, after which (s)he idly watches the narrative unfold (Porteous et al. 2013).

For the purpose of this study, neither concept has been adopted for deﬁning interactive narrative. Interactive narrative will be understood here as one in which the player interacts meaningfully and continually with a well-developed narrative for the duration of the gameplay experience. To this end, many scholars agree. Riedl (2010) argued that a successful interactive

117

narrative should ensure ‘a well-formed storyline occurs each time’ the player engages with the system, but he also stipulated that all events in the narrative should be signiﬁcant to the ﬁnal narrative outcome. In other words, the story should be well developed enough so that no amount of interactivity will lead to a broken, inconsistent, or incomplete narrative outcome. Games and playable media scholars Mateas and Stern (2000) referred to such narrative experience as

‘artistically whole’. Of course, the importance of an artistically whole experience for the player is important, but it is equally important to remember that in the construction of an interactive narrative, multiple and artistically whole experiences must be created and accounted for. The multiplicity of outcomes must not only allow for the mutability of a single play experience, but also of multiple play experiences, since replay value is an important commercial concern. Mateas and Stern (2000) argued that a good interactive narrative experience should support six or seven unique experiences before the game begins to feel ‘stale’. Although this magic number has not been proven, the importance of replay value remains an important concern for many researchers.

Finally, in addition to being complete, the narrative should also be coherent enough for the player to interact meaningful within it. The logic of the narrative directly correlates to the player’s ability to engage meaningfully. Players should be able to understand, via the narrative, not only what they want to do, but—in the context of the narrative—why they want to do it

(Mateas 2004). The narrative should be constructed in such a way, that it is complete and coherent enough to enable critical and meaningful actions by the player.

Based on these considerations, I have established the following criteria as essential components in the analysis of digital interactive narrative games:

118

• There should be different well-formed/artistically whole, dramatic and narratively coherent

stories, complete with a relevant resolution that occurs smoothly during each play-through

and that should support replayability.

• The narrative should be clear enough to help players understand what to do and why they

should do it, based upon their own critical reﬂections.

3.2 Ludic Agency

Having established the essential criteria for the narrative, it is important to consider how the player will interact with the narrative by discussing the essential elements that support both meaningful ludic agency (in this section) and narrative agency (in Sect. 3.3).

I use the term ‘ludic agency’ to refer to the player’s capacity to move about and to interact with the environment crafted by the game, in ways that are not necessarily, though they frequently might be, linked to the narrative. Ludic agency is arguably an easier form of agency to achieve, especially in games that are not inherently narrative. The interactivity of most digital games, whether narrative or not, is dependent on ludic agency, but the issue does become more complicated when the player is supposed to interact within a narrative. The inclusion of a narrative necessitates ludic functions that work in conjunction with that narrative.

Unfortunately, there are many games that fail to align ludic agency with narrative agency, causing what has been termed a ‘ludonarrative dissonance’ in the game (Hocking 2007). Ludic functions that may directly affect narrative interaction, and which may be managed seamlessly by a drama manager, are those ludic systems that are presented to the player in order that (s)he may interact with the narrative. For example, dialogue screens, morality systems, quick time

119

events (QTEs), and heads-up displays (HUDs) are all ludic elements that accommodate interaction by the player with the narrative.

Margaret Archer, a respected sociologist who studied the capacity for human agency, argued that most occasions for agency stem from embodied interactions. She claimed: (1) that our concerns for physical well-being stem from interaction with the environment, (2) that our concerns for person achievement stem from interactions with objects/objectives, and (3) that our concerns for self-worth stem from interactions with other subjects (Archer 2001). Although

Archer is not a scholar of digital games or interactive narratives, her research on human agency is valuable in conjunction with the understanding of ludic agency in games. Archer reminds us that interactions in the game space with the environment, objects, and subjects (NPCs) are central to the player’s sense of agency within the game space. Furthermore, the means by which each of these types of agency is achieved in a game is through ludic interaction: commonly, the player moves about the space using the directional pad, and interacts with objects and engages with NPCs using the input buttons or trigger buttons.

The types of interactions Archer described are reﬂected in the research of Mateas and

Stern (2000) who referred to such interactions as ‘‘embodied interactions’’. Mateas equated

Archer’s understanding of agency to ludic functions:

Embodied interaction matters. Though dialogue should be signiﬁcant…it

should not be the sole mechanism. Embodied interaction such as moving

from one location to another, picking up an object, or touching a character

should play a role in the action (Mateas 2004).

120

The attention paid by both Archer (in terms of subject-to-subject interaction) and by Mateas and

Stern to the importance of dialogue as a means to express agency is also important. While Archer stipulated that dialogue between two subjects should be focused on ‘predominant concerns with pre-eminent emotions’, Mateas and Stern translated her argument into a game-speciﬁc one.

Dialogue in a game should chime with the player’s predominant concerns and emotions, and the player should not be ‘offered an occasional small number of obvious choices that force the plot in a different direction’ (Archer 2001; Mateas and Stern 2000). To further support this argument,

Karen and Joshua Tanenbaum insisted that agency is not about selecting options, but is instead about expressing intent and receiving a satisfying response to that intent (Tanenbaum and

Tanenbaum 2009). In other words, dialogue should be free from, rather than constrained by, ludic elements that project limited options on frozen (rather than narratively dynamic) menus. As narratology scholar Ryan (2009) observed, such limited options do not only hinder the expression of intent, but also create of ‘loss of ﬂuidity’ that ‘takes a toll on the player’s immersion’.

Given the joint considerations paid to ludic agency by these scholars, I propose the following criteria, by which ludic agency in interactive narrative digital games can be assessed:

• Players should not be offered merely a small number of obvious choices that are forced

upon players by the plot. Although dialogue should be an important component, it should be

used to align predominant narrative concerns with pre-eminent emotions, instead of

predictable choices that propel the story forward. Where possible, dialogue should be

removed from menu systems altogether.

121

• Players should be able to enjoy meaningful embodied interactions with the environment,

objects, and subjects (NPCs) in conjunction with the narrative.

3.3 Narrative Agency

The criteria outlined above, however, do not guarantee a successful interactive experience with a narrative. Criteria for narrative agency, i.e. criteria that work in conjunction with those that promote ludic agency, must also be established.

In contrast to creating ludic agency in a game (an adept feature in many digital games), creating narrative agency is fairly complicated. Video game designer and programmer, Johnson

(2013), states that the challenge of interactive narrative games is to leave ‘room for the player, not just within the rules and mechanics and the systems, but within the story as well’. To do this, the author of the narrative must simultaneously attempt to deliver a cohesive narrative, while allowing the player to influence the outcome of that narrative (Sharma et al. 2010). As stated before, this requires writing not one but several cohesive narratives that, to the player, fit together seamlessly regardless of the choices (s)he makes to influence it. Thus, the narrative must be mutable in all iterations so that the player feels immersed in the story. As Riedl and Leon (2008) observed, such seamlessness allows for the particulars of a certain vignette or story portion (i.e. a story fragment existing along a specific and activated story branch) to make narrative sense within the context of the whole.

While immersion, especially ludic immersion, is possible to achieve in most other game types, it is essential in digital interactive narratives, because without it narrative agency cannot be achieved. Despite its importance, this is also complicated to achieve because it depends on the

122

‘engagement of the imagination in the construction and contemplation of the story world that relies on purely mental activity’ (Ryan 2009). In other words, this type of immersion cannot be presented ludicly, but must be stimulated in the player mentally through narrative agency.

Interactive computing scholars Riedl and Bulitko (2013) both argued that it is through immersion that the player ‘believes that [(s)he is] an integral part of an unfolding story and that [his/her] actions have meaningful consequences’. While Riedl and Bulitko rightly articulated the importance of meaningful consequences, scholars Fox Harrell and Jichen Zhu pointed out that for consequences to matter to the player, (s)he must be able to think critically and to care about those consequences too. Harrell and Zhu (2009) stated that agency ‘centers upon the intentionality’ of the player’s actions; that intentionality is exercised after critical thought and is expressed through action. Karen and Joshua Tanenbaum mirrored Harrell and Zhu’s point by indicating that a focus ‘on the idea of meaning allows us to shift the emphasis in an interaction away from outcome of a choice and toward the intent which underlies that choice’ (Tanenbaum and Tanenbaum 2009).

Therefore, players must be able to maintain the perception that their choices are meaningful, because if they cannot then they are less likely to deliberate choices carefully or critically, and their actions will not be the result of intent. The methods for facilitating meaningful choices are linked to the ludic functions in the game discussed above. Nevertheless, it is worth repeating that when choices are presented to the player through whatever ludic means necessary, the players must be able to understand their options (i.e. why should would want to make a choice), and be able to deliberate between options. It is also important to stress that the

123

need for players to care about—to be emotionally engaged in—their choices is as important as the logic they employ to make them.

Emotional engagement is central to narrative agency in an interactive narrative. As proliﬁc writer and game designer Greg Costikyan observed:

Because a player is involved in the creation of the experience, because the

experience of play is at least as much his product as that of the game

designer, the emotions he feels can affect him much more deeply than the

surface, empathic response you feel when viewing or reading about

characters in a story (Costikyan 2000).

Narrative agency is thus dependent on critical thinking through an emotional engagement with an immersive narrative experience, in which the player perceives the game to respond meaningfully to his or her choices and feelings. To accomplish such a feat, the player should feel like an active agent in the narrative.

In many digital games the player assumes the role of a prescribed protagonist, and that protagonist makes commentary, and portrays emotions over which the player has little control

(but must submit to for the sake of the game’s strict narrative experience). Mateas (2004) observed that such games have a ‘third-person reflection’, meaning that the player can be immersed in the narrative only through third-person, and not first-person, reflection. Not surprisingly, Mateas (2004) argued that interactive narratives should include ‘first person engagement’, whereby the player does not have the feeling of playing a role that only allows for third-person reflection. In other words, if the player assumes a third-person role, the choices made in that role are still made via first-person engagement. To draw a parallel to acting, the

124

player assumes the role much like an actor, but has control over the delivery of that role.

Similarly, Riedl (2010) argued that the user should be an active participant (an actor) in the experience. Thus, the player must be an active agent in the story, who reﬂects on and cares about choices in the ﬁrst-person in such a way that (s)he is immersed in the narrative, and thus the narrative maintains meaning.

Finally, it is worth noting that these criteria again mirror criteria for human agency, as outlined by Margaret Archer:

The reﬂective powers which give us our capacity to acquire personal

identities are dependent upon our ability to prioritize our concerns in the

world…if our personal identity represents our unique subjectivity, then

who we are is formed by our…reﬂections upon the world, meaning its

natural, practical, and social orders (Archer 2001).

Archer conveyed the importance of first-person reflection on players’ ability to reflect and to be concerned about the narrative world in which they find themselves. Archer’s research also validated the need for critical reflection. She argued that the human capacity for critical reflection is essentially the capacity to evaluate context, imagine alternatives, and to collaborate on outcomes (Archer 2001). It is essential for the player to be able to think critically in the collaborative construction of the narrative space. To encapsulate the essential components required to create and maintain narrative agency, the following criteria are proposed:

125

• The user must have an active role in the narrative, whereby (s)he inﬂuences events both

during and at the end of the game, and as a result should not feel that (s)he is playing a role

that forces one to reﬂect on how a character (instead of oneself) would react.

• The user should be able to reﬂect critically upon and evaluate social contexts in order to

envision and bring about narrative alternatives.

3.4 Character Development and Interactions

The consideration of character is, logically, a key concern in the development of a narrative, and should be as much of a key concern in the development of an interactive narrative. The development of character in interactive narratives, however, must be dually concerned with how to develop the non-player characters the player is meant to interact with, as well as the way in which the player develops one’s own sense character through those interactions.

While some scholars like Frasca (2003) have argued that the more agency a player is given, the ‘less personality the character will have’ [i.e. that this way of thinking unnecessarily restricts the creative and critical capabilities of the player as well as the capacity for NPCs (non- player characters)], this freedom does not limit character development—it enhances it. Of course, the enhancement of the player-character necessitates that the NPCs in the game be incredibly developed so that the player’s engagement with them derives emotion, meaning, and purpose. Therefore, in the development of interactive narratives, where development of the player-character is reduced, development of NPCs should be increased.

It is worth noting that the enrichment of NPCs is not only dependent on the concept of

‘character’ itself (such as it would be understood in text, ﬁlm, etc.), but is also dependent on AI

126

systems. While the capacity of AI continues to improve, the criteria listed here envision ideal circumstances of both AI capability and a deeper understanding of good character creation.

These criteria are therefore something to strive for, if they have not yet been fully attained.

To this end, in the creation of NPCs, scholars have agreed on several key elements to enrich the interactive narrative experience. The most important criterion, and that one from which the other criteria follow, is that the NPCs should be believable characters. Riedl describes character believability through behavior traits. He wrote, ‘The behaviors of a character must support, and not violate, the suspension of disbelief that the character could be real’ (Riedl

2010). Bryan Loyall, scholar and Director of BAE Systems Knowledge-Based Modelling and

Planning, concluded that the believability of a character can be simpliﬁed down to player respect: ‘The characters in the world need to seem real to the participant…they need to be believable enough that the participant cares about them…real enough that the player respect them.’ For Loyall this ensures that the player is immersed in the narrative and its consequences, and will therefore be less likely to toy with or attempt to break the narrative experience intentionally. Furthermore, for Loyall (2004) this means developing strong personalities rather than weak personalities for characters, since players are more likely to respect ‘solid’ characters rather than treat them as narrative toys. Mateas (2004) argued that for NPCs to be believable they must have rich personalities, emotions, and social interactions, and that the player’s interactions with NPCs will be more believable if they are focused on emotional entanglements between the

NPC and the player character, or between an NPC and another NPC. Again, the focus on emotional interactions encourages meaning and critical thought processes from the players and helps to immerse them more fully in the narrative experience.

127

Given these considerations, the proposed criteria for character development in a digital interactive narrative game are as follows:

• Believable characters should be developed so that the users care enough about them to

respect them. This entails developing characters with strong personalities and emotions that

are rich enough so that users can infer their thoughts.

• Interactions between NPCs to NPCs and between users and NPCs should be about

relationships and emotional entanglements rather than as props to propel action.

3.5 Drama Management Interactions

A drama management system is a system in the game that, according to Mark Riedl and Carlos

Leon, ‘ensures a well-formed story occurs each time’. When programmed into a game’s structure, the drama management system manipulates the virtual world and the NPCs to bring about a dramatic experience for the player (Riedl and Leon 2008). The drama management system tracks and presents the mutable narrative situations, interactions, and choices to a player based on the information it has derived from the player’s input and experience up to that point. In other words, the drama management system is that system that accounts for the interactions of the player in such a way that the narrative outcome is logical and complete.

When used well, the drama management system allows for the smooth mutability and adaptability of the interactive narrative, and improves the overall experience of the player. When used poorly, the system can make the player overtly aware of the game’s manipulations of the narrative, and drastically reduces the player’s immersion and agency. In their research article

128

‘Drama Management and Player Modeling for Interactive Fiction Games’ (2010), scholars

Sharma Manu, Stantiago Ontanon, Manish Mehta, and Ashwin Ram articulated the issue clearly by stating that, ‘Too much agency can result in losing control of the plot, but limited agency of the player can make the player aware of the medium and therefore reduce the engagement.’ The drama management system is, in a metaphorical way, the illusion that masks the magician’s trick and as such, ‘the user should not be aware of the existence of the experience manager or its interventions’ (Riedl and Bulitko 2013).

Therefore, while a drama management system aids in the overall success of an interactive narrative game, it should not be detectable by the player. This means, in any analysis of these criteria in current games, one can only judge those systems with drama management systems that are too obvious and which do not meet the essential criteria outlined below:

• A drama manager must be present in the system and must improve the overall user

experience, but should also be dependent on user experience and should adapt to speciﬁc

user strategies, in order to ensure a smoothly formed and easily mutable story.

• The drama manager should be hidden. The user should not be aware of the existence of the

drama manager or its interventions in the story.

4 Conclusion

As the research ﬁeld of digital games is still relatively new, and as the sub-ﬁeld of interactive narratives within digital games is even newer, there continues to be a pressing need for collaborative rather than isolated efforts toward improving the balance of narrative and ludic

129

agency in games. The criteria established in this paper reﬂect those upon which scholars, tackling the issues of digital interactive narrative games from different angles and with different intents, agree. This does not mean, however, that the limitations of these criteria should not be questioned, analyzed, or altered, but rather that these criteria can help to improve our focus on improving the development of digital games.

The merits of the criteria should be tested for utility in order to see if they do indeed aid game developers in the creation of better games, and aid scholars in their analysis of the success of those games. It is necessary to apply these criteria and best practices to previously published games, as this will reveal which interactive narratives have proved successful and which ones need to be improved. Additionally, the study of these criteria may be best tested by using already published AAA games to help validate not only the criteria itself, but also the use of such criteria in relation to the demands of a mass consumer public interested in interactive narratives.

130

CHAPTER 6

ASSESSING MASS EFFECT 2 AND THE ELDER SCOLLS V: SKYRIM USING COLLABORATIVE CRITERIA FOR NARRATIVE AND AGENCY IN DIGITAL INTERACTIVE NARRATIVE16

Introduction

I’ve just woken up on an operating table. I have no idea where I am, how

I’ve gotten here, who the woman is yelling at me over the intercom, or why alarms are sounding.

It seems reasonable under these conditions to demand answers from the first person I encounter, but when I do “+2 Renegade” flashes on the screen. Still I feel disoriented, confused, and impatient with my current circumstances, so after one too many ill-explained directives I make clear that I’m tired of taking orders. When I do, another “+2 Renegade” displays on my screen. This doesn’t seem reasonable or fair. What I had read and selected as panicked and confused dialogue options on my computer screen were delivered by the voice actor in Mass

Effect 2 as harsh and unfeeling admonishments. At that moment, it became clear to me that the avatar I was playing in the game wasn’t anything like me, despite my being in control of what she said and where she went. The game had already misrepresented my intentions, had already alienated me, had already made my interactions with the system less meaningful than

I’d hoped. The story wasn’t playing out as I expected and my agency within it was already compromised.

…

16 From Journal of Games Criticism, “Assessing Mass Effect 2 and Elder Scrolls V: Skyrim: Using Collaborative Criteria for Player Agency in Interactive Narratives.” Volume 3, Issue A, 2016, Lindsey Joyce © Journal of Games Criticism. With permission of Journal of Game Criticism.

131

The relationship between a player’s agency1 and a game’s narrative2 has been the subject of extensive research and scholarly debate. At one point the discussion became so popular and contentious within the game studies community that it earned its own name: “the ludology versus narratology debate.” In some ways, this “debate” was less about arguing sides than it was about defending modes of study and areas of discourse. While some scholars attempt to cement games as a singular form of study free from the shadow of non-games theories and modes of discourse3, others seek to understand games through association with other media types and theories4, including film, theatre, and text. In this instance, the debate was about staking scholarly claim over games; should games be studied within English Departments, Computer Science

Departments, Media Departments, or elsewhere? In another instance, however, the debate was concerned with the differences between ludology and narratology as methodological approaches.

Within this context, the argument is not about whether game studies should become its own mode of study, but about how ludic functions and narrative functions can be used in a game, whether the two can coexist, and which is the more essential function.

Within this context, the central – though unsatisfactory – argument is that the relationship between narrative and agency is fundamentally in tension5; to increase the strength of a narrative is to decrease the player’s sense of agency, and vice versa. While scholars largely agree on this score, there are numerous and disparate claims for how to decrease this tension, if decreasing it is possible at all. Within this context, agency is also frequently valued over narrative. While narrative may or may not be present in a game, agency is assumed to be a component of all games and, as such, is considered a defining feature that outweighs that or narrative. The logic here is that games can exist without narrative but not without agency. This misguided logic,

132

however, works in direct opposition to attempts to balance the two. Furthermore, if this logic were as sound as it appears, one could expect that few games would place a high priority on narrative, but this isn’t the case at all. In fact, the opposite is true: the number of interactive

6 narrative games has increased over the years rather than decreased.

This increase is indicative of a number of things, not least of which is the potential power of the interactive narrative form when developed and executed well. Interactive narrative games allow for two things other narrative and game forms do not: the meaningful representation and engagement of the player with the game’s procedures and the ability for that representation to lead to different story outcomes. The importance of player representation should not be understated – though it is a concern the ludology versus narratology debate largely ignores – and is perhaps the most important outcome of interactive narrative games from a cultural perspective.

In traditional narrative forms such as novels, films, and theatre, the reader or viewer frequently lacks representation. A female reader can no sooner change the gender of a male protagonist in a book than an African American viewer can change the race or skin tone of the lead actor on a movie screen. Of course, this problem exists in games too7; players cannot change the appearance of Lara Croft in Tomb Raider or of Nathan Drake in the Uncharted series, much as they may wish to do so. By comparison, in interactive narrative games, the player can more opportunity to inform how characters a represented and developed. Even when the game dictates certain attributes of the player-character, such as age, gender, race, sexuality, etc., players are still endowed with the ability to perform and subvert the stereotypes and tropes associated with identities through their choices and interaction with the narrative and its outcome.

133

More pragmatically, the increase in the development of interactive narrative games is also indicative of a simpler truth: that the presumed tension between narrative and agency is misguided. As stated previously, there is a flawed logic that persists on both sides of the ludology versus narratology debate: that games can exist without narrative but not without agency. It is, of course, true that games can exist without an embedded narrative. Games such as Tetris and Minecraft prove this to be true. The flawed logic, instead, resides in how we conceive of and approach narrative specifically within game structures. When narrative becomes a procedural element or function of a game, it also becomes a source of agency, rather than a limit to it. As a result, interactive narrative games maintain two forms of agency: narrative

8 9 agency and ludic agency .

When viewed this way, the arguments of the ludology versus narratology debate take on new complexity and possibility. Furthermore, when viewed this way, it becomes possible to analyze both sources of agency in an effort to find how they coexist in interactive narrative games and how they affect players’ interactions and efforts to make meaning of the game experience. I argue that, when approached this way, not only can a balance be found between narrative agency and ludic agency, but also that once found, this balance will ensure better player representation and more meaningful game experiences within the construct of digital interactive narrative games.

Method

In an effort to complicate the intersection between narrative agency and ludic agency, I surveyed the field of game studies and created a set of evaluative criteria broken into five areas

134

of assessment that should be considered and included in the creation and study of digital interactive narrative games. While ludic agency and narrative agency are each one of the five areas for assessment, the other three areas for assessment are also essential components of interactive narrative games and, subsequently, affect a player’s capacity to maintain ludic agency, narrative agency, or both. The five areas of assessment, their definitions, the essential criteria for each area, and the sources upon which the criteria were assembled are as follows:

Table 1. Summary of Criteria Area of Definition Criteria Sources Assessment

Drama The degree to • A drama management system should • Mark Riedl and Management which the improve the player’s experience by Carlos Leon (2008) Interactions game’s systems ensuring a well-formed story occurs, ensure that a but should also be dependent on player • Manu Sharma well-formed experience and should adapt to specific (2010) story occurs player strategies. each time the • Mark Riedl and game is • The drama manager should be Vadim Bulitko (2013) played. hidden. The player should not be aware of the existence of the drama manager or its interventions in the story.

Narrative The player’s • There should be different well- • Julie Porteous, Fred Experience overall formed/artistically whole, dramatic and Charles, and Marc perception of narratively coherent stories, complete Cavazza (2013) story creation as with a relevant resolution that occurs well as their role smoothly during each play-through and • Mark Riedl (2010) within its that should support replayability. creation • Michael Mateas and • The narrative should be clear enough Andrew Stern (2003) to help players understand what to do and why they should do it, based upon their critical reflections.

135

Ludic Agency The player’s • Players should not be offered merely • Clint Hocking capacity to a small number of obvious choices that (2007) move about and are forced upon players by the plot. to interact with Although dialogue should be an • Margaret Archer the environment important component, it should be used (2001) crafted by the to align predominant narrative concerns game in ways with pre-eminent emotions, instead of • Michael Mateas and that are not predictable choices that propel the story Andrew Stern (2003) necessarily, forward. Where possible, dialogue though they should be removed from menu systems • Karen Tanenbaum frequently are, altogether. and Joshua linked to the Tanenbaum (2009) narrative. • Players should be able to enjoy meaningful embodied interactions with • Marie-Laure Ryan the environment, objects, and subjects (2009) (NPCs) in conjunction with the narrative. Narrative The player’s • The player must have an active role • Soren Johnson Agency capacity to in the narrative, whereby (s)he (2013) meaningfully influences events both during and at the influence the end of the game, and as a result, should • Manu Sharma outcome of the be able to reflect and think critically (2010) narrative. about choices within the context of the narrative before making them. • Mark Riedl and Carlos Leon (2008) • The player should be able to reflect critically upon and evaluate social • Marie-Laure Ryan contexts in order to envision and bring (2009) about narrative alternatives. • Mark Riedl and Vadim Bulitko (2013)

• Fox Harrell and Jichen Zhu (2009)

• Karen Tanenbaum and Joshua Tanenbaum (2009)

• Greg Costikyan

136

(2000)

• Michael Mateas (2004)

• Margaret Archer (2001)

Character The capacity to • Believable characters should be • Gonzola Frasca Development which non- developed so that the players care (2003) and Interaction player enough about them to respect them. characters and This entails developing characters with • Mark Riedl (2010) the player- strong personalities and emotions that character are are rich enough so that players can • Bryan Loyall (2004) emotionally form emotional attachments. developed and • Michael Mateas three- • Interactions between NPCs to NPCs (2004) dimensional. and between players and NPCs should be about relationships and emotional entanglements rather than as props to propel action.

In the creation of these criteria, it was important to establish common ground between current game studies scholars and theories rather than to perpetuate the continued and exasperated taking of sides within the ludology versus narratology debate. As such, the criteria I formed were dependent upon agreement by at least two scholars and reflect points of consensus rather than of contention within the field. My method for creating these criteria and a more extensive discussion of them can be found in a previous article, “Creating Collaborative

Criteria for Interactive Narrative Game Analysis” (Joyce, 2015). Having already formed the criteria, it is now my intention to apply them to two digital interactive narrative games: Mass

137

Effect 2 (Bioware, 2010) and The Elder Scrolls V: Skyrim (Bethesda Game Studios, 2011). In doing so, I will show how ludic and narrative agency function distinctly in games, but also how they overlap to create a complex narrative game experience.

I have elected to analyze Mass Effect 2 and Skyrim because both contain an interactive narrative wherein the narrative responds to and is altered by the player’s input and choices. Both games are also high-selling AAA games with a broad and diverse audience. Each of the games can also be categorized as an action role-playing game. Additionally, each game offers the player customization options in relation to race, gender, age, height, weight, etc. and utilizes experience points as a means to progress character skills. Both games can also be played in third-person perspective, though Skyrim also allows play from the first-person perspective. Amongst these similarities, there is also an important difference. While Mass Effect 2 follows a more linear narrative in a relatively small game environment, Skyrim’s narrative and its game environment remain open to exploration by the player. The games maintain enough in common for comparative analysis of the criteria, but also a sufficient degree of variability on which the criteria can be tested.

My analysis will begin by applying criteria in drama management interactions before moving onto narrative experience, ludic agency, narrative agency, and character development, to Mass Effect 2 and Skyrim. While my previous article provides extensive detail about how I established each area of assessment and the criteria within in, I have provided brief summaries here for necessary context. Following these summaries, I present an analysis concerning the specified criteria in Mass Effect 2 and then in Skyrim. Following the individual discussion of each game, I will conclude each section with a discussion of findings.

138

Analysis

Drama Management Systems

A drama management system is a system in the game that, according to

Mark Riedl (2010), “ensures a well-formed story occurs each time” (p. 1). Programmed into a game, the drama management system manipulates the virtual world and the Non-Player

Characters (NPCs) to bring about a dramatic experience for the player. The drama management system presents the mutable narrative situations, interactions, and choices to the player based on the information it has derived from the player’s input and experience up to that point. In other words, the drama management system is the system that accounts for the interactions of the player in such a way that the narrative outcome is reasonable, without major narrative inconsistencies, and complete. Additionally, the drama management system tracks and records the input given by the player. Sometimes this record is made visible to the player, and sometimes it isn’t. For instance, in The Walking Dead (Telltale Games, 2012) a player may be a told a character will “remember” something the player has elected to do or say, but the game provides no further data to the player. By comparison, in Catherine (Atlus, 2011), the player’s choices are recorded and presented to the player on a scale that shows an angel on one side and a devil on the other. Player’s see this scale when they make a choice but can also access and view the scale at their leisure. When used well, the drama management system allows for the smooth mutability and adaptability of the interactive narrative and improves the overall experience of the player. By contrast, when used poorly, the drama management system can make the player excessively aware of the game’s manipulations of the narrative and drastically reduce the

139

player’s immersion and perception of agency. Therefore, while a drama management system aids in the overall success of an interactive narrative game, it should remain undetectable by the player. This creates a problem for analysis; if a good drama management system is invisible, then any system that can be pinpointed and analyzed is already, simply by being detectable, unsuccessful. To work well a drama management system:

• Should improve the overall player experience

by adapting to the player’s strategy and ensuring a smoothly formed and easily

mutable story occurs.

• Should not be visible or obvious to the player. The player should not be aware of the

existence of the drama manager or its interventions in the story.

Mass Effect 2

The drama management system in Mass Effect 2 is overt and easily detected, and as a result, it significantly reduces the capacity for player agency in the game. The drama management system does adapt to the player's strategy, and the drama management system does alter character interaction based on the player’s previous choices and input, but the mutability of the story is limited, as I will discuss later, and the machinations of the drama management system are clearly visible. In fact, the drama management system is clearly visualized for the player in numerous ways, and as a result, the player can easily track how the system is intervening in and guiding the story toward its conclusion. Primarily the drama management system is visualized through the dialogue system and the visual feedback systems in the game. Most narrative choices are selected via the dialogue system while the visual

140

feedback systems report on and track the player's moral progress throughout the narrative.

Morality takes a central role in the outcome of the narrative in Mass Effect 2. The player’s morality (either Paragon if good or Renegade if bad) affects how NPCs engage with and treat the player and opens and closes certain narrative possibilities. When the player makes a choice in the game, the drama management system reports to the player how their choice scaled on the morality system. Additionally, the player can view their current morality standing by accessing the user interface screen. The overtness of the game’s drama management system impacts almost every other key area of play I analyze here, and as a result, the high level of detectability of the system is the biggest setback to providing the player with agency in the game. Because the morality system is easily decipherable within gameplay, the need for critical reflection is reduced, and the narrative experience is compromised. Because the player is offered a small number of obvious choices that map along an easily decipherable binary of good/Paragon, bad/Renegade the ludic agency is compromised. With such a rigid dialogue system, the player cannot envision or act upon narrative alternatives. The player’s narrative agency is, therefore, limited.

Skyrim

In Skyrim, the drama management system is more hidden than the system found in Mass

Effect 2 but still isn’t integrated well enough to be undetectable. In Skyrim, the drama management system focuses less on narrative drama than on the drama of play. Where gameplay is concerned, the progression of the character is smoothly mutable throughout the game: There are myriad ways a player can customize a character, via appearance and skill sets.

141

For instance, a player who chooses to sneak through dungeons will more quickly level up in stealth proficiency than a player who does not. While such skill progression in the game is automatic, experience points are earned so that players can customize other skill sets via a skill map. As a result, the drama management system seamlessly adjusts to the player’s personal style of play. That said, the drama management is not as well hidden in terms of the game’s narrative structure. The game procedurally generates, or generates content algorithmically rather than manually, both quests and environments for the player to explore, and though the player’s skill sets and play preferences may advance or level up as a result of these procedural generations, they fracture the possibility for narrative cohesion. The narrative cannot account for the procedural generations in a coherent or cohesive way. These random encounters, though ludicly engaging, are not accounted for by the game’s drama management system in terms of narrative.

No narrative changes or feedback are given to the player as a result of them. Therefore, the more sidequests or randomly generated encounters the player engages with, the less coherent the narrative provided by the drama management system becomes.

Discussion of Findings

The results of applying the evaluative criteria for this area of assessment are summarized below:

Table 2. Analysis of Drama Management Criteria Mass Effect 2 Skyrim Drama Management Absent/Low/Medium/High Absent/Low/Medium/High

The drama manager improves user experience Absent Low

The drama manager adapts to user experience High Medium

The drama manager is hidden from the user Absent Absent

142

In Mass Effect 2 and Skyrim, neither drama management system succeeds in balancing narrative and play. In Mass Effect 2, the drama management system is too overt, obviously manipulating the story, and simultaneously disrupting and pausing play. Conversely, the system in Skyrim, though better hidden, limits cohesion and development of the story in the game. Thus, the drama management system in each game privileges one element of the game’s interactive narrative experience, either story or play, over the other. Without a drama management system that can equally manage both these elements, each game fails to be a successful interactive narrative, based on the evaluative criteria applied. More theoretically, however, the visibility of the drama management systems acts as a reminder to the player that her agency is constructed and, therefore, less meaningful. While in a balanced system “there is play, a system, and a gap between the player and the system where interpretation takes place,” games with visible drama management systems limit the space of interpretation for the player by revealing themselves as the arbiters and interpreters of the player’s input (Sicart, 2013, p. 36).

Narrative Experience

In an interactive narrative game, the story should be sufficiently well-developed so that no amount of interactivity will lead to broken, inconsistent, or incomplete narrative outcomes. In other words, “a well-formed storyline occurs each time” the player engages with the system, and all events in the narrative should be significant to the final narrative outcome (Riedl, 2010, p.1).

The storyline should not only have significant narrative outcomes, but each possible outcome should be part of an “artistically complete” story (Mateas and Stern, 2003, p.1). Yet, the story must also seem mutable. Developers must write multiple narrative beats and outcomes for the

143

player to experience and have a sense of control within, and those component narrative pieces must also come together to form a whole given the player’s input. This multiplicity is important both for the player’s sense of agency in the game and to the replay value of the game: an important commercial concern. Finally, in addition to being complete and replayable, the narrative must also be coherent enough for the player to interact meaningfully with and within it. The logic of the narrative directly correlates to the player’s ability to engage meaningfully. The player should be able to understand, via the narrative, not only what they want to do, but why, in the context of the narrative, they want to do it (Mateas, 2004, p.4).

In other words, the narrative should have a complete and coherent construction that enables both critical thinking and meaningful action by the player. Based on these considerations, I established the following criteria as essential components in the analysis of digital interactive narrative games:

• A unique, well-formed/artistically whole, dramatic, and narratively coherent story,

complete with a relevant resolution, occurs smoothly with each playthrough and

supports replayability.

• The narrative should be clear enough to help players understand what to do and, upon

critical reflection, why they should do it.

Mass Effect 2

Although the Mass Effect series contains three games with an overarching and continuous narrative, each game, including Mass Effect 2, contains an artistically whole and coherent story of its own. Whether the story derived from each playthrough of Mass Effect 2 can

144

be considered unique is, however, debatable. At the beginning of the game, the player selects one of six player classes, which affects the player's combat style. In addition to this initial customization at the start of the game, the player is also presented with moral choices throughout the game. These choices range from saying, “thank you” or “this is a waste of time” to a scared man trying to offer intel, all the way to allowing a crewmember to murder someone or not. Depending on the choices the player makes, she will be gauged as a Paragon (good) or a Renegade (wicked) character in the game, and NPCs will respond and interact with the player based on his current state of morality. For instance, the choice to say “thank you” earns the player two paragon points, while opting to say, “this is a waste of time” results in two renegade points. By comparison allowing the crewmember to kill another earns 15 renegade points, while stopping the murder earns 15 paragon points. The points are generally balanced, with any choice resulting in the same number of points to either their paragon or renegade score.

Despite the potential for such customizations to create a different narrative with relevant resolutions in each playthrough, the narrative outcome is largely the same no matter how the player functions in combat or what choices he makes. While small subplot beats may differ, and while certain characters may live or die, the result of the narrative specific to the player- character, Shepard, remains the same, and will always culminate with Shepard detonating the enemy’s, known as Collectors, base. While there is, arguably, one other ending – one in which

Shepard dies – this ending cannot be carried over into Mass Effect 3 and should be viewed as a fail state. In this way, the game always ends with the destruction of the Collector base, and, therefore, the difference in play experiences cannot alter the ending to any great extent.

145

Additionally, the outcomes of the individual missions in the game remain largely the same.

Many of the missions involve Shepard persuading potential crewmembers to join in the mission.

Regardless of the player’s actions or how the player persuades them, they will join the crew. The differences in play, as a result, exist not so much in the plot as in character interaction. As

Eric Brudvig (2010) observes, although "the heroes are what drive the story...the arc of the main tale isn't in itself, exceptional.” The lack of complexity or uniqueness in the narrative's conclusion could perhaps be blamed on the need to prep players for the continuation of the story in Mass Effect 3, but considered as a single entity, the narrative experience in Mass Effect

2 doesn't offer much variety regarding unique and relevant conclusions. From moment to moment the narrative provides mutability on the micro level, but the same level of mutability does not translate to the narrative on the macro level. For instance, whether the player allows the crewmember, Garrus, to murder someone or not, he becomes a loyal crewmember. The choice by the player to allow or stop the murder, then, has little narrative impact overall.

Furthermore, the final chapter of the game, despite all mutability that leads up to it, differs little from player to player or playthrough through playthrough. The player will, regardless of her choices or moral standing, defeat the universe-threatening enemy; all that differs is which characters are still alive after the battle. But seeing as the characters left alive are only visually represented and not commented upon within the dialogue that follows the battle, even this variance seems inconsequential within the context of Mass Effect 2. In the end, the mutability offered throughout the game means little to the narrative outcomes and the narrative outcomes are hardly unique or distinctly different from one another.

146

Whether the narrative is coherent enough to help players understand what to do and why to do it is also debatable. While in general the narrative is coherent enough to provide context and background that help to inform player decisions, there are occasions when a player may need or want more context but is incapable of getting it from the game’s visual interfaces. The player can only scroll through a limited set of speech options, and while the dialogue system allows the player to analyze the choices she is given, the player does not have the option to gather further information that might help inform her decision. Given this, the player has no recourse but to proceed though conversations when she may need more information to make a critically informed choice. The player must make a choice that affects the narrative whether or not she fully understands the options presented to her. The dialogue restrictions and the inability for players to exit conversations and gain more information before selecting choices have to do with the importance of the game’s morality systems, which are central to Mass Effect

2’s narrative mutability.

In Mass Effect 2 the narrative experience hinges largely on how the player scores along its morality system. The player is led to believe that her choices, whether Paragon or Renegade, will alter the course and outcome of the story. Beyond scoring a different morality, however, the outcomes of the player’s choice are actually too similar to generate a truly unique experience. For instance, on the games “loyalty” quests, which are taken to assure a crew member stays loyal to Shepard, the player can complete the quest making either Renegade or Paragon choice and still maintain the crewmember’s loyalty. In Mass Effect 2, players ultimately lack agency over how they progress through the narrative or gameplay; the moral choices they are limited to, how their selection of those limited choices are measured, and

147

the conclusion are all dictated by the system. This limitation – perhaps largely a result of the overt drama management system discussed in the earlier section – affects the overall narrative experience and the capacity for players to think critically about the narrative. Without this critical capacity, the player lacks true meaningful narrative agency.

Skyrim

In Skyrim, a single playthrough of the game can span well over a hundred hours, and thus, replayability is nearly an irrelevant concern; the developers purport the game to be, in effect, endless thanks to procedurally generated encounters and quests (Schrier, 2011). There are multiple ways by which a character can be uniquely developed using skill sets available to players in the game. However, because character development is dependent upon skills, this also means the player’s character development lacks depth and is largely a ludic concern. The game does, however, contain a main quest line, and it is through this quest line that the player can experience the central narrative of the game. Yet, the narrative of the main quest line remains the same, regardless of how the player chooses to play the game, the skill sets she elects to develop, or the choices she attempts to actualize. An elf character leveled in destructive magic and stealth will experience the same main quest conclusion as an orc character leveled in heavy armor and melee weapons. As Justin McElroy (2011) states, “[the player] discovers he or she is half- dragon. He or she has to stop the dragon. These two facts are pretty much the only absolutes." However, while the main quest line is static, the magnitude of quests and side plots provide the opportunity for far more narrative customization than the main questline alone provides. Outside the main quest line, any player’s experience of the game will certainly be

148

unique. On the other hand, the game’s narrative variety and magnitude are also a detriment to the games ability to logically or cohesively tie narrative strands together. There is so much to do and experience in the game that the development of a cohesive story with a relevant resolution – especially in a game that claims to be endless – is nearly impossible. But as Oliver

Banham (2011) observes, the "driving factor" behind Skyrim is not story. "Gameplay," he says,

"is the more compelling force” (Banham, 2011). In other words, the game makes little attempt to stitch the player’s narrative experiences together. Instead, the player must do so on his own in an effort to complete the story experience.

Even without narrative cohesion or the possibility of narrative completion, the game still provides sufficient background information for the player to understand what to do and why her character should or shouldn’t do it. While the narrative is not cohesive or entirely coherent when taken as a whole, it does attempt to provide coherency and the capacity for critical consideration from moment to moment. Players have access to a wealth of knowledge.

NPCs present the player with novel and useful information, as do the books, scrolls, journals, letters, etc. scattered about the game environment. In most cases, the player is also free to put choices on hold until they have acquired the information needed to make a choice. For instance, when prompted to join the Thieves Guild, a player can ask NPCs question about it, listen in on conversations between multiple NPCs, read books on the subject, explore the guild’s halls, etc. Having done so, the player can then determine whether to join or not. Additionally, when asked to join another guild elsewhere in the game, the player will be able to assess whether membership in both guilds – such as the Thieve Guild and Dark Brotherhood – is worthwhile.

Despite these two guilds being at odds with one another, the game does not make it impossible to

149

join both guilds. Instead, it merely delineates the consequences of doing so. If a player in both guilds is caught murdering, he will be ejected from the Thieves Guild, and lose the benefits that guild provides. In this way, Skyrim does prompt the player to think critically about the choices he makes and how those choices could impact his overall narrative experience of the game.

Discussion of Findings

The results of applying the evaluative criteria for this area of assessment are summarized below:

Table 3. Analysis of Narrative Experience Criteria Mass Effect 2 Skyrim Narrative Experience Absent/Low/Medium/High Absent/Low/Medium/High

Game provides at least one complete story High Medium

Game provides possibility of different story Low Absent outcome

Each story outcome is coherently constructed Low Absent

Each story outcome has a relevant resolution Low Low

Multiple story outcomes make the game Medium Low replayable

The narrative structure is rich enough for user Medium High to make choices

Though neither Mass Effect 2 or Skyrim provides a truly unique narrative conclusion upon repeat playthroughs, each game does provide at least one cohesive story

150

outcome. While Mass Effect 2’s tight linear structure helps maintain cohesion, Skyrim’s open structure gives players more narrative choice and freedom to explore those choices. While both games have a high replay value, the main narrative experience in both generally stays the same.

While players can change classes, play style, and their choices, the main mission to either “kill the dragon” or “detonate the base” remain the same. The difference is that in Mass Effect 2 the

NPCs matter and, depending on the choices the player makes, those NPCs may or may not survive the mission. These subtle changes in the game’s outcome – despite the need to destroy the base in each – increase the game’s replayability. In Skyrim, NPCs are of little consequence and do not change the experience significantly upon repeat playthroughs.

Ludic Agency

The term "ludic agency" refers to the player’s capacity to interact with the environment crafted by the game in ways that are not necessarily but frequently are linked to the game’s narrative structure. For ludic agency to contribute effectively to the inclusion of a narrative, the ludic functions must work in conjunction with that narrative. Ludic functions that may directly impact narrative interaction, and which may be managed seamlessly by a drama management system, are those ludic systems that are presented to the player so that they may interact with the narrative. For example, dialogue screens, morality systems, quick time events

(QTEs), and heads-up displays (HUDs) are all ludic elements that allow the player to interact with the narrative. Mateas (2004) refers to these types of ludic interactions as “embodied interactions” and argues, “Embodied interaction matters. Though dialogue should be significant…it should not be the sole mechanism. Embodied interaction such as moving from

151

one location to another, picking up an object, or touching a character should play a role in the action” (p. 30). The importance of embodiment and embodied actions has a long history in the study of agency, both within and outside the study of games. Inherent in the discussion of agency are issues of power and control, and, more importantly, the distribution of that power and control between the socially constructed body and the individual body. Within this relationship, the individual’s agency is reduced by the construction of the sociopolitical power, or power that is outside the control of the individual. The process by which socially prescriptive identities are constructed defines and–to a philosophical degree–immobilizes the individual.

Rather than being active bodies with self-determination and agency, the individual becomes hampered by the confines of the system and its dictates. Speaking of an agentless body, Butler

(2001) states, “the body is figured as a mere instrument or medium for which a set of cultural meanings are…related” (p. 38). Meanwhile Meynell (2009) notes of the mind that, “the agent’s position in social hierarchies influences what she can know, what she wants, and what moral rights and obligations she might have” (p.7). These subjugating systems tell individuals (in either spoken or unspoken ways) who they are and the limitations that come with that identity. In other words, the system creates powerless subjects rather than empowered agents. It’s not hard to see how these issues translate to games. As codified systems, players are confined to and, to some degree, defined by the rules and limitations of the game system. Agency theorists propose that to resist sociopolitical constructions of the self, one must transition from socially dictated subject back to an individual, mobile, and embodied self. The embodied self must act with intention and operates “outside” the boundaries of socially constructed subjects. When Mateas indicates that embodied interactions matter, he is drawing on the history of agency theory and recognizing that

152

in any system, sociopolitical or codified, individuals cannot maintain agency without embodiment.

Dialogue is an important extension of embodiment, and though it is frequently represented textually in games, it is still a ludic function presented to the player through the game’s mechanics. For the player to maintain agency, she must be able to communicate her mind through an embodied action: a speech act. Dialogue is the function by which the player can do so. As such, the dialogue in a game should align with the player’s primary concerns and emotions. Dialogue should not, as Mateas and Stern (2003) note, be “offered an occasional small number of obvious choices that force the plot in a different direction” (p.1), because agency is not about selecting options; it is about expressing intent and receiving a satisfactory response to that intent (Tanenbaum, 2009, p. 38). In other words, dialogue should be free from, rather than constrained by, ludic elements that project limited options on frozen menus rather than narratively dynamic ones. Given the joint considerations paid to ludic agency by scholars, I developed the following criteria by which to assess ludic agency in interactive narrative digital games:

• Players should be offered more than a small number of choices that obviously force

plot points

• Though dialogue should be an important component, it should be used to align

primary narrative concerns with pre-eminent emotions, and per the criteria above,

should resist offering easily decipherable choices presented only to propel the story

forward. Where possible, dialogue should be removed from menu systems

altogether.

153

• Players should be able to have meaningful embodied interactions with the

environment, objects, and subjects (NPCs) in conjunction with the narrative

Mass Effect 2

Mass Effect 2 fails to meet all but one of the criteria strongly. The game restricts the player to only a few choices at a time, each of which is easily detectable as the Renegade or

Paragon choice, and clearly serves to push the plot forward. Furthermore, the choices presented to the player are aligned with the game’s morality system and will dictate how the player will scale on the morality spectrum. In this way, the player is offered a small set of choices which clearly map to the binary system of morality. Adding to this, the options presented to the player are easily decipherable; the average player can clearly distinguish between the Renegade choice and the Paragon choice with little difficulty. The game even goes so far, in certain cases, to color-code the Paragon choice as blue, and the Renegade choice as red to alleviate any challenge in choosing how to act. Given these limitations, the choices provided to the player are both easily decipherable and clearly provided as a means to move to the plot toward specific outcomes. To make matters worse, these obvious choices are presented primarily through a dialogue system which pauses play and further disrupts narrative flow. Although Mass Effect

2 does focus conversations around the characters’ emotional complexities, the placement of most narrative development and plot progression within a dialogue system is counter-intuitive in an action game. The result, Bart Robson (2010) observes, is that "the game plays like a story- oriented tactical shooter, funneling Shepard and his entourage through conversations and choices before plunging them into firefights and chase sequences.” The game divides itself between

154

choice-oriented conversations that develop the plot and shoot-out scenes in which the actions of the player have little effect on the narrative. Additionally, embodied interactions in the game are mostly limited to combat, loot, and gear changes, and computer hacking. Any other embodied interactions take place in cut scenes. While the choices the player makes may activate those cut scenes (a sequence in a video game that is not interactive and which suspends gameplay), the player has no agency within the scenes themselves, and they serve only as cinematic interludes with the player positioned as a passive viewer. As a result, the game keeps ludic agency entirely separated from narrative agency. What happens ludicly has little bearing on the narrative, and the narrative choices have little bearing on the ludic engagements. This lack of agency also explains why the romantic relationships Shepard can develop feel awkward: the player has not been able to embody interactions between Shepard and any of the crew, and any scenes where they interact romantically feel voyeuristic rather than personally engaging. The separation of ludic agency from narrative concerns drastically minimizes the level of meaning the player can derive from the game. Perhaps more importantly, without embodied interactions reflected in the narrative and without narrative choices reflected in the embodied interactions of the game, Mass Effect 2 suggests that the mind and the body maintain separate agency.

Skyrim

In Skyrim, the player is free to experience the narrative beats, of which there are many, in any order he chooses. Players can abandon side quests at any time or can elect never to complete the main quest at all. In other words, the game never forces itself upon players, thereby giving

155

them a great deal of ludic agency. The open-world structure also provides more opportunity for embodied interactions than does Mass Effect 2. For example, while traversing the landscape, the player may end up in conversations or altercations with a great many NPCs or animals, including deadly snow bears and dragons. The player can also elect how to interact; the player can battle using magic or weapons, can run away, can pickpocket people or loot areas, or can stealthily creep by without even being noticed. Each of these actions is an embodied interaction the player is free to make.

The freedom Skyrim provides comes with several limitations, however. One such limitation is the shallow emotional depth of the characters with which the player interacts. While

I will discuss this more in a later section, I mention it here because the emotional shallowness reduces the importance of individual characters and interactions in the game and negatively affects the player’s ludic agency through dialogue. The dialogue between player and NPCs is focused on the NPCs’ needs and goals–usually, he or she needs a specific item that is down at the bottom of a deep cave where only the player is daring enough to go–rather than articulating aspects of his or her emotional complexity. As a result, the dialogue does not align with primary narrative concerns nor with pre-eminent emotions. Additionally, while the game avoids offering the player limited choices that force the plot forward, the game also lacks a cohesive plot. As mentioned previously, this is a central quest line, but this quest line is mandatory and the more alternate quests a player takes on, the less clear the plot of the central quest becomes for those players who choose to pursue it. While the game allows the player a high level of ludic agency, the game fails to adequately respond to this agency with logical or relevant narrative beats. Thus, the player can move and interact freely, but the game provides little feedback,

156

context, or purpose that makes these interactions meaningful in terms of the narrative. Because the game lacks a narrative structure that makes its ludic agency relevant or purposeful, the player in Skyrim is also made static and powerless. As Catriona Mackenzie (2009) suggests, in an interactive narrative ideally “we create our identities and shape our characters by appropriating our past, anticipating our future experiences, and identifying with or distancing ourselves from certain characteristics, emotions, desires, and values” (p.107). Because Skyrim fails to acknowledge the consequence of past or future experiences, and because the characters lack emotions, desires, ad values that carry weight in the game, the player remains powerless in Skyrim.

Discussion of Findings

The results of applying the evaluative criteria for this area of assessment are summarized below:

Table 4. Analysis of Ludic Agency Criteria Mass Effect 2 Skyrim Ludic Agency Absent/Low/Medium/High Absent/Low/Medium/High

A small number of choices don't force plot Absent High

Obvious choices don't force plot Absent High

Dialogue is emotionally-driven High Absent

No dialogue system at all Absent Absent

Embodied interactions with environment Low High

Embodied interactions with objects Low High

Embodied interactions with subject Low Low

157

Per the established criteria for ludic agency, Skyrim more successfully provides ludic agency than does Mass Effect 2. While Skyrim meets many of the criteria for ludic agency, the ludic agency given to the player lacks purpose or synthesis with the narrative. On the other hand, while Mass Effect 2 provides a more engaging narrative experience and while the player may have a higher level of narrative agency in Mass Effect 2 than in Skyrim, the player in Mass Effect 2 is also far more restricted in terms of ludic agency than in Skyrim. In fact, Mass Effect 2 only excels over Skyrim by providing more emotionally driven dialogue whereas Skyrim's dialogue is comparatively flat. In Mass Effect 2, the number of choices the player is given clearly align with the game’s binary morality system, making the choices both small in number and easily decipherable. Thus, while the choices are emotionally driven, they clearly move the plot along a binary path. Given that the narrative's development is linked to the conversations between characters, it is through constraining dialogue options that the player mechanically influences the narrative development of the game. Unfortunately, this dialogue system is highly restrictive and doesn't allow for the player to have many embodied interactions in the game. In Skyrim, by comparison, the player can have numerous embodied interactions with the environment, with the NPCs, and with objects in the world, but these interactions matter little in terms of the plot. Thus, while the plot isn’t forced by a small number of obvious choices, the game lacks a strong sense of plot altogether. The player is left adrift in a jumble of narrative beats that fail to come together into a logical and complete story.

158

Narrative Agency

While ludic agency is the player’s capacity to move about and to interact with the environment crafted by the game, narrative agency is the player’s capacity to meaningfully influence the outcome of the narrative. Though seemingly ambiguous, the key to narrative agency is the level of “meaningfulness” in the player’s interaction and influence over the narrative. Narrative agency has been defined countless time by numerous scholars, as have the ways it derives a sense of meaningfulness. For example, Janet Murray (1997), states the meaningfulness is derived from “exerting power over enticing and plastic materials” (p. 153). In terms of narrative agency, Murray’s definition implies that the narrative must remain malleable or mutable. For Emma Westecott (2008), meaningfulness occurs when a player is “in control of where [they] go and how [they] progress.” While movement through the space of the game may be more of a ludic concern, control over how a player progresses is, within interactive narratives, linked to narrative progression. For Isaac Karth (2015), meaningful agency is the “control that facilitates the player’s transition into a deeper understanding of the ergodic system” (p. 2). In

“Agency Reconsidered” Wardrip-Fruin et. al. (2009) say agency is meaningful when “the actions players desire are among those they can take as supposed by an underlying computation model”

(p. 1). My own definition states that if a choice in a game is to be meaningful, the game must provide the player enough information to make a choice, but must do so without diminishing the meaning of that choice; the choice should neither provide over-simplified alternatives nor outcomes. Choice becomes meaningless when the decision-making process is curtailed by an unambiguous or heavy-handed system that does the thinking required in the decision-making process for the player (Joyce, 2014).

159

In terms of narrative agency, these criteria for meaningfulness necessitate the player’s understanding of the game and its narrative as well as the ability to act on that understanding to bring about a result. The developers of the game narrative must then simultaneously attempt to deliver a cohesive narrative while allowing the player to influence the outcome of that narrative (Sharma, Ontañón, Mehta, & Ram, 2010). The narrative must be seamless and mutable in any and every playthrough of the game so that the player feels immersed in the story (Riedl,

2010, p.2). Immersion at the narrative level is complicated to achieve because it depends on the

"engagement of the imagination in the construction and contemplation of the story world that relies on purely mental activity" (Ryan, 2009, p.54). In other words, this type of immersion cannot be presented through mechanics, but must be stimulated in the player mentally. In other words, narrative agency comes from the capacity to think critically about the player’s position, narratively, in the game, her relation to other characters in the game, and how a potential choice may alter either of those things and, therefore, the outcome of the game too. It is through immersion that the player “believes that [he is] an integral part of an unfolding story and that

[his] actions have meaningful consequences” (Riedl & Bulitko, 2013, p.1). For consequences to be meaningful, they must be the result of player intent derived from critical thought and expressed through action (Harrell & Zhu, 2009, p.45). Additionally, emotional engagement is central to narrative agency in game play “because the experience of play is at least as much [the player’s] product as that of the game designer, the emotions he feels can affect him much more deeply than the surface, empathic response you feel when viewing or reading about characters in a story” (Costikyan, 2000). To accomplish such a feat, the player should be encouraged to feel like an active agent in the unfolding of the narrative. The player should also feel that she is the

160

protagonist and, therefore, playing a leading rather than supporting role, regardless of whether the game itself provides a first or third person perspective (Mateas, 2004, p.8). Based on these considerations, I established this list of criteria for narrative agency:

• The player has an active role in the narrative whereby she influences events that occur

as well as how the story ends, and as a result should feel immersed in the player-

character such that the player identifies clearly with it.

• The narrative should be constructed in such a way that it prompts critical thinking and

reflection from the player. The player should be able to reflect critically upon and

evaluate social contexts in order to envision and bring about narrative alternatives.

Mass Effect 2

In Mass Effect 2, the player does influence story development. The story develops depending upon which narrative choices the player selects. Yet, while the player is influential in this development, the strictly binary narrative choices, which in turn map to a binary morality system, limit just how influential the player can be. Restricted to what amount to “good” or

“bad” decisions, the game fails to provide nuance or the necessity for critical thought and reflection. Thus, the game does not stimulate the mental activity required to achieve deep narrative agency.

Whether or not the player feels immersed in her character is also uncertain. During the first playthrough of the game, players inhabit the role of Shepard and play the game mentally in the first-person despite the game’s visual third-person perspective. The more times a player plays through the game, however, the more abstract the concept of player-as-Shepard becomes.

161

For instance, during the first playthrough, the player is apt to be more invested and feel more connected to the choices they make. With repeat play sessions, however, the player is always selecting choices from a place of increased knowledge and memory and, therefore, from decreased suspense. The knowledge the player possesses and brings to the narrative in repeat playthroughs lacks the necessity of critical reflection that the first play experience attempts to provide. While this could also be said of other games, Mass Effect 2’s rigid binary morality system further limits suspense. If the player learns which choice is a Renegade choice they also, by proxy, learn which choice is Paragon. That said, the first play experience also impedes opportunities for critical reflection in ways that limit narrative agency.

The game constructs a rich and immersive world in which the player’s understanding of social context, inter-species relations, galactic history, and politics are all important to how they make choices. But because the drama management system in the game is so overt, the player is less likely to become immersed, think critically about choices, or investigate narrative alternatives.

Although the player’s sense of character is connected to a moral purpose in the game, the dialogue system over-simplifies morality. The moral choices are generally easy to decipher and are sometimes even color-coded to reduce all nuance or ambiguity. By limiting the information provided through the dialogue wheel, and by negating the player’s option to exit conversations to gain more information before committing to a choice, the player’s agency and commitment to meaning-making are reduced. Additionally, because the system only presents a small and select number of choices via the dialogue system, narrative alternatives are also impossible. While the player can select from the narrative choices presented via the dialogue system, the player is still ludicly restricted by the game’s strictly linear narrative

162

structure. The player can neither move her avatar nor exit the dialogue system without completing the conversation. The narrative structure is rigidly linear such that the story it sustains fails to adequately account for the player’s need to critically reflect on choices as part of the meaning-making process. The story then, despite the choices a player makes while crafting it, remains the authorial work of the game rather than of the player.

Skyrim

In Skyrim, the player can have a truly unique experience. With so many options and with such a large environment, the possible sequences in which a player can experience the narrative beats of the game are seemingly limitless. Additionally, the game is designed so that the player and the character are one, or as Oliver Banham (2011) states, "Your character is the ultimate expression of your play-style." The character is an empty canvas waiting to be filled by the player. The player designs the character, forming an attachment to it through this process. But the character also lacks a backstory, making the character a narratively clean slate as well.

Because the player is so involved in creating the character, she does not need to consider how the character would act or react when presented with choices; any choices made are authentic to that player-character’s experience as a result of the player’s involvement in the character’s aesthetic and behavioral creation. In this way, the player has a high level of narrative agency and is free to make up his own stories as he explores the environment. This freedom allows the player to complete quests in a variety of ways and allows the player to consider and execute one of any of the multiple solutions that she personally identifies as best. While this does enable the player to bring about narrative alternatives within a singular quest, the player’s capacity for narrative

163

agency changes if the main quest line is considered the “key” narrative experience. The main quest, as previously mentioned, does not change regardless of other quests and circumstances the player has engaged with or encountered. There are no narrative alternatives within this quest.

Because this quest isn’t mandatory, it’s within reason to argue that narrative agency throughout the game is high in terms of the player’s ability to influence events that occur throughout gameplay.

Despite the player’s ability to influence events that occur, the player rarely needs to reflect critically or to evaluate social context in the game. As stated previously, players are rarely faced with a choice that has serious ramifications in the game or which is irreversible. For instance, to return to the Thieves Guild and Dark Brotherhood example, a player who is in both guilds and is caught murdering can activate the “Reparations” quest to atone for the murder and remain in the guild. Similarly, a player who joins the Companions to become a werewolf can later seek a cure for the lycanthropy. The game almost always provides a means for the player to both have an experience and evade the potential penalties for it. By providing players with every opportunity to explore the game’s full potential, the game sacrifices long-term consequences and the narrative relevance for choices. Additionally, the need for critical reflection is reduced, as there is always a deus ex machina available to the player. Players can do what they want, but what they do will rarely have a meaningful impact on the game’s overall narrative.

Discussion of Findings

The results of applying the evaluative criteria for this area of assessment are summarized below:

164

Table 5. Analysis of Narrative Agency Criteria Mass Effect 2 Skyrim Narrative Agency Absent/Low/Medium/High Absent/Low/Medium/High

User has active role High High

User influences the events that occur High Medium

User influences how the story ends Low Absent

User is not playing a role Medium High

User needs to reflect Low Absent

User can bring about narrative alternatives Absent Medium

The two games are relatively equal in terms of the degree of narrative agency is concerned. In both games, the player has an active role and influences events that occur; in Mass

Effect 2 the player influences events by making moral choices primarily via dialogue choices, while in Skyrim the player influences events by selecting when and in what capacity they will accomplish quests. In the first playthrough of each game, the player can inhabit her role, though in Mass Effect 2 the ability to inhabit the role of Shepard may become more abstract with repeat play experiences. Important distinctions still exist between the games, however. For instance, in Mass Effect 2 the narrative experience is meant to give the player the narrative agency required to think critically about choices and to influence the story’s end, but the drama management system inhibits the player’s ability to do so. On the other hand, in Skyrim, there is very little need to reflect at all because very few actions have irreversible consequences, and even if a player does reflect on choices, the choices they make do not influence the story’s end, such that there is one and provided the player pursues it. Narrative agency in Skyrim is greatly

165

overshadowed and buried beneath the high level of ludic agency offered to players. The scale of the game environment and the number of possibilities for engagement within that environment are too large to maintain a coherent and meaningful story, and as a result, according to a game critic, “It’s difficult to ever feel completely satisfied with a play session of Skyrim” (Onyett,

2011). Ultimately, neither game provides satisfying narrative agency or the capacity to think critically about the story as it unfolds in the game’s narrative structure.

Character Development and Interactions

Character development is integral to immersion and suspension of disbelief, helps players emotionally connect to and understand the unfolding story, and adds complexity to the player’s interaction. As Mark Riedl (2010) notes, “The behaviors of a character must support, and not violate, the suspension of disbelief that the character could be real” (p. 2). The development of character in interactive narratives, however, must be dually concerned with how the player develops his own sense of character through interaction with NPCs as well with how the NPC’s character must itself be constructed. That said, in interactive narratives, the less the game restricts and defines the role of the player-character, the more freely the player will experience narrative agency and an increased sense of personal player-character development

(Westecott, 2008). For the player-character to be sufficiently developed to a level that necessitates critical thought and is emotionally driven, the player-character should be able to interact with well-developed and believable NPCs. There is, therefore, a correlation between how well NPCs are developed and how well the player can develop her own sense of character and meaning in the game. Adding to this, Bryan Loyall (2004), scholar and Director of BAE

166

Systems Knowledge-Based Modeling and Planning, boils believability of a character down to player respect, writing, “The characters in the world need to seem real to the participant…they need to be believable enough that the participant cares about them...real enough that the player respects them” (Loyall, 2004, p.5). Furthermore, NPCs must have rich personalities, emotions, and social interactions and that the player’s interactions with NPCs will be more believable if they are focused on emotional entanglements between the NPC and the player character or between an NPC and another NPC (Mateas, 2004, p.8). Given these and other considerations, digital interactive narrative games should contain:

• Believable characters that are developed such that the player cares enough about them

to respect them and such that they have strong enough personalities for players to

connect with them emotionally.

• Interactions between NPCs to NPCs and between the player and NPCs should

be primarily about relationships and emotional entanglements and should not be used

as prop to propel action.

Mass Effect 2

The character relationships in Mass Effect 2 are very well developed, or as Bart

Robson (2010) observes, "The characters are the game...the lifeblood of Mass Effect 2.” Each character in the game is a three-dimensional character with strong emotions and a strong personality. Not only are the characters well developed, but the characters and their interpersonal relationships with one another also develop and change throughout the game. The smaller-scale environment of the game and the crew’s confinement to their ship, the Normandy, when not on a

167

mission provide a small enough microcosm for characters to be developed; the player spends enough time in close quarters with the NPCs to learn their histories as well as their sense of ethics and morals. Robson (2010) notes that "the game builds a rich history for each [character], making time spent with them much more interesting.” The result of the investment in character enables the player to infer the thoughts and probable reactions of the NPCs. It also helps the player-character form invested relationships with the NPCs, such that the player is concerned for and emotionally engaged with the characters. Additionally, interactions with characters are generally focused on emotions and feelings. Each character is three-dimensional; they have surface traits, backstories that involve emotional depth and moral turmoil, and articulate specific worldviews. The NPCs also engage in social conflict, and one of the player's jobs in the game is to negotiate the treacherous relationships and conflicting world views of the crew. However, although the interactions between characters are entangled in complex emotions and give the character enough depth to be more than narrative props, the interactions are confined to the rigid structure of the dialogue menu and the drama management system, and thus, are primarily responsible for moving the plot forward. In this regard, the characters lose some sense of depth and meaning; it is too obvious that the characters are plot devices meant to force plot and moral choices. Rather than remaining complex and three dimensional for the sake of narrative depth and narrative interactivity, the NPCs become an extension of the morality system. They are present to prompt dialogue choices and actions with consequences that clearly map to either a

Renegade or Paragon result. Thus, interactions with NPCs could be seen as manipulative with every interaction resulting in a quid pro quo or morality status points.

168

Skyrim

Characters in Skyrim are less developed than the expansive environment in which they exist. Due to the scale of the environment, there are too many NPCS in the game to sufficiently develop them all to a level that prompts emotional connection or respect. Furthermore, the player-character in Skyrim generally moves through the game as a solitary protagonist, spending little time with any one NPC in the game. Though NPCs may tag along on some missions to offer help, and though the player can hire a companion to travel along, these characters are also poorly developed and are restricted to a specific number of speech-acts that will be repeated for as long as the NPC is present. As the Executive Editor at IGN, Charles Onyett (2011), observes, "the followers are meant to be tools of battles instead of ever-evolving personalities."

While some NPCs may offer tidbits of background about who they are, these tidbits lack emotional resonance and are rarely exchanged in the context of a shared experience between the player-character and the NPCs. For instance, followers generally have basic dialogue that the player can initiate through the “talk” mechanic, but the followers also have a cycle of comments they make when questing with the player, and these comments are made without prompting by the player. These comments are not relative to where the player is or what the player has done, and act more as awkward non-sequiturs than as conversations that form emotional bonds. The redundancy of both the cycle of comments and those that can be initiated makes the characters in the game static and flat rather than dynamic and well developed. Rather than characters, the

NPCs are more accurately props for play. Interactions with NPCs in Skyrim serve to initiate and end quests – most of which the player-character will undertake alone, and the player’s mission is, more often than not, to retrieve an item. The player can then elect to retrieve and keep these

169

items or bring them back to NPCs to exchange for an alternate item. Thus, NPCs function primarily to help the player acquire enticing goods that add to the play–but not the story– experience. Justin McElroy (2011) notes that "The rewards for taking on these quests are often very practical, like a new weapon or a precious stone," and are not meant to enrich the player's relationship with any NPCs. Relationships in this way are typically based on the exchange of goods and services in Skyrim. The player retrieves something the NPC needs and in return is given something for his effort, at which point the relationship ends.

Discussion of Findings

The results of applying the evaluative criteria for this area of assessment are summarized below:

Table 6. Analysis of Character Development Criteria Mass Effect 2 Skyrim Character Development Absent/Low/Medium/High Absent/Low/Medium/High

User cares about characters High Absent

User respects characters High Absent

Character have strong personalities High Absent

Characters have strong emotions High Absent

User can infer character's thoughts High Absent interactions are about emotional High Absent entanglement relationships are not props for actions High Absent

Based on the evaluative criteria, the character development in Mass Effect 2 far surpasses the character development in Skyrim. If the player derives a pleasant experience from ludic

170

agency in Skyrim, they derive it from the rich character development in Mass Effect 2. Mass

Effect 2 is centered on emotionally rich relationships and the complexities that inherently arise when individuals of different backgrounds and beliefs must cooperate. Character development is the core of the experience in Mass Effect 2, and the player feels her interaction carry emotional weight because the characters respond to her acts of agency, even if the narrative experience as a whole does not dramatically change as a result of those actions. While it is obvious that NPCs in Mass Effect serve primarily as plot devices for driving the story forward, the mutability of the

NPCs based on the player’s input in the form of dialogue choices and QTEs remain emotionally resonant within the developing story. What the player does in Mass Effect 2 matters to the characters the player interacts with, if not to the final narrative outcome. The same cannot be said of Skyrim, which sacrifices the potential depth of a finite set of characters for a large number of characters and large environmental scope (McElroy 2011).

Overall Discussion of Findings

As the preceding analysis suggests, neither Mass Effect 2 nor Skyrim meets all the required criteria of the five areas for assessing interactive narrative games. However, Mass

Effect 2 meets more criteria overall. Whereas Skyrim succeeds in providing a high degree of ludic agency, Mass Effect 2 provides a low degree. Whereas Mass Effect

2 provides a high degree narrative agency and character development, Skyrim provides a low degree. This overall assessment shows that where one game largely succeeds, the other fails. Ultimately, Skyrim’s immense game environment works against its narrative experience rather than for it – limiting narrative cohesion and relevance, the ability to develop characters, and means by which to add meaning to the amount of content delivered to the

171

player. Comparatively, Mass Effect 2’s environment is small enough for characters to be developed within it and for a mutable and replayable story to unfold, but the confines of the environment stifle exploration in a way that traps the player into the structure of its branching linear narrative. Yet it is also impossible to say that one particular configuration between environment and narrative structure is better than the other since neither game achieves a balance between ludic and narrative elements.

In terms of the final outcome of the narrative experience, the two games are quite similar. Mass Effect 2 only significantly exceeds Skyrim’s success in terms of its character criteria. Otherwise, the games generally score similarly across the five areas of analysis: drama management systems, narrative experience, ludic agency, narrative agency, and character development and interaction. Most significantly, neither game successfully hides its drama management system. In each case, this visible system diminished the players’ experience by reminding them that the game maintains the highest level of control over the experience. Thus, a key finding of this analysis is the central role played by a game’s drama management system in shaping the balance between ludic and narrative agency in digital interactive narratives. Taken together, these two games also reveal the complication between balancing story and play; Mass Effect 2 provides a more engaging narrative experience, but fails to balance that experience with an equal measure of ludic agency. Skyrim, on the other hand, offers so much ludic agency that the narrative experience becomes disjointed and incoherent. Furthermore, the two games begin to reveal that without balancing narrative with agency, the player loses a sense of power and meaning in the game experience as a whole.

172

Conclusion

This analysis demonstrates the usefulness of applying these criteria (drama management systems, narrative experience, ludic agency, narrative agency, and character development) for comparative analysis of digital interactive narratives. Although the analyses performed here are broad, just touching on each of the sub-criteria when a richer analysis could be conducted, the observations produced are still relevant and important to the understanding of agency in digital interactive narratives, and did, as anticipated, help to identify specific limitations of each game's design and systems. Furthermore, the analysis reveals the complex relationship between ludic agency and narrative agency, but also reveals how the two coexist, affect one another, and alter the game experience in key ways.

Primarily, the analysis reveals that Mass Effect 2 provides greater narrative agency while Skyrim provides greater ludic agency. Neither game, however, was found to meet successfully all the criteria necessary to provide balanced ludic and narrative agency in the overall narrative experience. In both cases, the drama management system fails to appropriately balance the narrative and ludic elements present in the game. In light of this observation, further investigation of drama management systems in other digital interactive narratives would be worthwhile, and may help pinpoint essential areas for improvement in future game development. This suggests that these evaluative criteria can be used to identify potential weaknesses and areas of improvement as a game is being developed. Additionally, while the criteria did illustrate the difficulty of balancing ludic and narrative elements, the categorical approach of the analysis helped to outline why and in what areas the ludic and narrative elements

173

clashed. The criteria establish a common ground or starting point from which analysis of games may take place within game studies.

The criteria are not, however, without limitation. It became clear while applying the criteria, that compartmentalizing the criteria is difficult; what limits ludic agency may be the result of the drama management system, and what limits narrative agency may be the result of character development, and so on. In this way, it’s important to consider the criteria holistically as well as individually. Additionally, the analyses performed here are broad surveys rather than in-depth considerations. A more in-depth analysis of each game may produce more useful findings and results. Even with these limitations, the criteria proved useful, and their continued use may aid in the development of future digital interactive narratives and the study of games. Through further and continued application, the criteria can continue to be refined in order to advance the study and development of digital interactive narrative until such a time that ludonarrative balance is finally achievable.

As evidence by the passion in the game studies field around the ludology versus narratology debate, what is at stake with striking the balance between ludic and narrative elements is, of course, more than the development of a successful interactive narrative games, but also the ethical balance of power between game and player, and the creation of a digital form that promotes critical thinking in relation to complex choices, empathy with others in terms of those choices, and a better understandings of our own agency as players. To properly balance ludic agency and narrative is to balance collaboration between the game and player; to create a story that did not previously exist and which no person could tell alone.

174

CHAPTER 7

KENTUCKY ROUTE ZERO: OR, HOW NOT TO GET LOST IN THE BRANCHING NARRATIVE SYSTEM17

The key stumbling block in understanding and developing digital interactive narratives is not that ludic and narrative elements are incompatible – a false argument to begin with – but rather that games scholars have collectively spent too much time arguing unnecessary sides and too little time establishing common criteria on which to assess games. In a previous article,

"Creating Collaborative Criteria for Interactive Narrative Game Analysis, “I set out with the lofty (and ongoing) task to begin establishing such criteria. By cross-examining key research in the field of interactive narrative, I compiled a list of essential criteria that should be considered and included in the creation of digital interactive narratives and which attempts to better balance the relationship between story and agency. Having previously used these criteria to access agency in two highly development and promotionally budgeted (so called AAA) games, Mass

Effect 2 and Elder Scrolls V: Skyrim. I now aim to test these criteria against the independently developed game by Cardboard Computer, Kentucky Route Zero. I have chosen Kentucky Route Zero not only to test the utility of the criteria outside of the AAA game space, but also to prove that the criteria can be used to test the success level of games, whether

AAA or not, that attempt to push or alter boundary conventions, as Kentucky Route Zero does. In this way, the criteria serve not only to enable post-production analysis, but also as a means of

17 From The Player Versus Story Divide in Game Studies: Critical Essays© 2016 Edited by Matthew Wilhelm Kapell by permission of McFarland & Company, Inc., Box 611 Jefferson NC 28640. www.mcfarlandpub.com

175

questioning the potential success of innovative design choices against a set of essential criteria of consideration.

Kentucky Route Zero’s non-standard approach to both character development and player perspective within a plot that is ostensibly one of small consequence makes it of particular interest for analysis. Not only do the minimal point-and-click mechanics manage to achieve a great deal, but the narrative also shows that adventure and meaning need not be derived from the high and epic stakes common to most games. The plot of Kentucky Route Zero, can be summarized as such: Conway, a delivery truck driver in Kentucky, must deliver a package to an address he cannot find, but which he is told exists along Route Zero. On his journey to find “the zero,” Conway eventually picks up additional travel companions: Shannon, a local television repair woman whose migrant parents once worked in the coal mines in the town; Ezra a young boy who has lost his parents; and Junebug and Johnny, two musicians who entice Conway,

Shannon, and Ezra to come with them to a gig. While such a synopsis might lead the player to assume they will play as Conway, this is only partially true. Conway may arguably be the central protagonist, but there are multiple protagonists in the game and the player will take on each of their perspectives, sometimes simultaneously, throughout game play. The game switches between characters via dialogue selection menus; one of the key ludic elements of the game. In other words, Players are not only presented with dialogue for Conway, as they would be in most single-protagonist or single-perspective games, but are offered dialogue for several characters throughout gameplay. This unique approach to narrative and ludic design, as well as the interfaces used to manage the drama and character development, situate the player not as a single character, but instead as author. It is from this position of author, I argue, that player is able to

176

experience greater levels of agency. By using the criteria I have previously established and replicated here, I intend to show how Kentucky Route Zero’s approach to player perspective and character development successfully balances both ludic and narrative agencies in a way that many single-protagonist games often fail to do.

Narrative Agency

The narrative agency in Kentucky Route Zero is interesting because of its multiple playable characters. The player seems to occupy two spaces in the narrative: in one space, the player acts as and through the characters and their perspectives, and in the other space, acts as the omniscient narrator or author who, despite lacking actual omniscience, orchestrates the overall story. The player’s space is thus one of character development and specificity, but also one of totality: the narrative experience is crafted around many perspectives and players must keep each in mind as they move through the game. Michael Mateas and Andrew Stern note that in game narratives designed to provide agency, “the player should not have the feeling of playing a role…[but] Rather the player should be able to be themselves as they explore the dramatic situation” (“Towards Integrating Plot” 2). Kentucky Route Zero’s use of multiple narrative roles and spaces for the player makes providing this agency more feasible; freed from a single character perspective, the player has more opportunity to explore and think about the dramatic situation. By elevating the player above a single-perspective experience, I argue that Kentucky Route Zero actually enriches the capacity for narrative agency. The player isn't confined to a single viewpoint through which to access and assess the narrative, but is present through all perspectives. In Kentucky Route Zero, the player is each of the characters singularly and collectively, and must not only consider choices as unique and important to a single

177

character, but also of how those choices will impact the other characters and the narrative whole.

This is not only accomplished by the way dialogue is presented, but also by the lack of feedback systems and the lack of certainty about where any dialogue choice will lead (Joyce “Kentucky

Route Zero” 19). Many interactive narrative games incorporate feedback systems that “tell” players about their choices rather than “showing” them their effect of their choice through narrative integration. Games like Bioware’s Mass Effect tell players if choices are good

(Paragon) or bad (Renegade), while Telltale’s games like The Walking Dead signal important choices have been made by telling players “[character name] will remember that.” Choice systems focused around dialogue over-use feedback to a fault; choices become a reflection of the system's structure and authority rather than of the player's agency and authority. In Kentucky

Route Zero there is no feedback beyond the continuation of the dialogue. Players can only guess where any particular dialogue choice will take the story, but their choices nonetheless feel organic based on how they want to develop a specific character. In turn, this allows players to reflect critically upon each choice, to consider how they have been steering a character's development up to a certain point and how, given that development, the characters would react. Importantly, it also presents players with the opportunity to consider how they want the character to react in reference to how they have developed the other characters. In other words, players must negotiate each character’s development within the social context of the other characters they are simultaneously developing (Archer 308). Mateas and Stern argue that a player should not feel constrained by a role in a game, but should instead be able to inhabit the space as themselves (“Integrating Plot” 2). While this is useful advice in single-protagonist games, Kentucky Route Zero takes a different approach to narrative agency by freeing the player

178

from any single role at all. While one limitation of this approach may be the inability to deeply connect with a single character or “become” the protagonist, Kentucky Route Zero’s approach necessitates critical thinking from the level of story. Playing an authorial role, players must maintain consideration for the personal and interpersonal motivations of all the characters alongside consideration for how the interplay between them informs the plot as a whole. Because players inhabit the perspective of more than one playable character, they can think critically on a level not allowed by single-protagonist game systems. They think on the level of authorship rather than on the level of character.

Ludic Agency

A key function of ludic agency is the player’s ability to have embodied interactions in the game space. Michael Mateas bluntly states, “Embodied interactions matter,” and goes on to say that while dialogue should be significant, as it most certainly is in Kentucky Route Zero, it should not be the sole means of engagement (“A Preliminary Poetics” 30). Given that Kentucky Route

Zero is a point-and-click game, the only way for players to interact with the system is to click a single button on their mouse. In this regard, there are certainly games that offer more meaningful embodied interactions with game environments. Kentucky Route Zero’s unique approach to narrative agency, however, lends itself to an important understanding of ludic agency. Despite not being able to move the character with the same level of freedom provided in other game types, the physicality of the characters is an important concern in how players manipulate the narrative and construct their own unique ludic experience. For instance, when Conway is hurt in an abandoned mine, players are presented with options on how to respond to the pain: they can elect to have Conway say, “I can walk on it, but it’s

179

slow,” or “I can walk on it, but it’s painful.” Then later, when Conway visits a doctor, players not only have the opportunity to create the narrative that is constructed around

Conway’s injury, but also how he responds to a surgical procedure that replaces his injured leg with a holographic one. When asked about his new leg, players are given three options as a response: “This isn’t my leg,” “It seems better, but…it wasn’t worth it,” or “It was just the heat.

The shingles were cracked in the sun, and one of them slipped out under his feet.” Both the players’ response in the mine and the players’ response about the surgery construct their ludic experience. Though players’ embodied actions are restricted by the game’s point-and- click mechanics, their perception of embodied action and meaning is heightened through this narrative control. Players do not relate to Conway’s relationship with his leg because they feel the mechanical limitations it forces on them (such as when a character in a game walks with a limp for a small amount of time as in Skyrim, impeding physical progress).

Instead, they relate because of the narrative and the choices it presents to players that allow them to construct a response for Conway with which they can relate and empathize.

This empathy toward characters, and the degree to which the player can embody them, is enhanced by the delivery and complexity of choices and dialogue. There are no easy choices in Kentucky Route Zero. Perhaps because the stakes of the narrative are so small, nuance is of more importance than opaqueness. In large games, such as the Mass Effect series, the stakes are high: the universe hangs in the balance. For this reason, Mass Effect, and games like it, often overly simplify their presentation of choices in order to make sure players, on some level, know the consequence of a choice before they’ve made it. The concern for the players’ possible frustration over an un-intended outcome resulting from a misunderstood

180

choice overrides the potential for narrative nuance and, as a result, the choices are reduced to moral challenges absolved of the weight of real ethical dilemma (Sicart 45). Players are apt to play for an outcome rather than for the experience that precedes it: save the world as the good guy, or ensure its demise as the bad guy (Smith). Kentucky Route Zero has no such pretense. There is no ethical dilemma and thus no need to reduce choices in the system to bifurcated moral challenges. In Kentucky Route Zero, although dialogue choices are still scripted, the system does not “manipulate [the dialectic]” in the same way morality-based games do (Sicart 43). In Kentucky Route Zero, the universe doesn’t hang in the balance, and if anything does hang in the balance, players have hung it there themselves through their unique construction of the narrative. For instance, the player can construct a story in which Conway is making his last delivery because his business is closing down, or a story in which he is happily retiring; the weight of the narrative and the importance of Conway’s delivery are chosen by the player through this narrative choice. Additionally, the player can construct a story in which Conway focuses on his relationship with his employer’s husband, or on his relationship with his employer’s son, both of which offer different dramatic portrayals of Conway’s past and how it may (or may not) impact his character in the present. The game, and the narrative the player constructs, is focused on the journey and the details the player is interested in pursuing more than it is interested in specific plot points that could “force the plot in a different direction” (Mateas and Stern, “Towards Integrating Plot” 1)

To this end, the narrative and ludic systems aren’t focused on narrative outcomes, as many games are. In Kentucky Route Zero, all choices ultimately lead in one direction, and on that level, all players will arrive at the same conclusion. The journey to that pre-determined

181

conclusion differs in two important ways: in dialogue choices and how those choices map out the journey to the conclusion of the game. In other words, while the ending is the same, the choices players make in how they develop characters leads to different experiences and scenes prior to the conclusion. In this way, the choices players make are, in one capacity, futile; the plot will progress along the same string of pearls and in a very linear fashion (Costikyan “Games,

Storytelling, and Breaking the String). Yet, in Kentucky Route Zero, every narrative move is centered on character emotions, about why a character says what he or she does. Such choices, or those choices which players interpret as having consequence for the character, have been found to increase the players’ perception of agency (Fendt et al. 10). In this way, Kentucky Route

Zero is a ludic metaphor for prioritizing the journey over the destination, as the narrative is one of character development placed primarily in the players’ hands.

Character Development and Interaction

Given that most games allow players to inhabit only one narrative protagonist, the demands of character development rest more heavily in the hands of the game’s developers. As a result, the NPCs (Non-Player Characters) have to be made believable and real by the developers in order for the player to interact with them. In this way, the analysis of character development and interaction is more commonly about how well a game has succeeded in constructing believable agents. Kentucky Route Zero, however, puts character development more directly in the hands of the player. It is worth noting that, although the game offers multiple protagonists, it still dictates when players can alter their perspective between those characters. For some stretches in time, players may only be prompted with dialogue options for Conway or for Shannon, and at other times, both simultaneously. When presented with the latter scenario, players must select

182

which character speaks. In other words, players must not only choose what will be said and by whom, but which character will be developed more deeply as a result of that choice. This navigation between characters within the system adds dimension and complexity to the characters by allowing players to develop them at their own selection. While some players may give Shannon more opportunities to talk, other players may choose to develop Conway more deeply; in either respect, their understanding of the narrative experience will be altered. Rather than being props the player cannot alter and which are used to propel the narrative forward, the protagonists in Kentucky Route Zero have greater capacity for complexity; the player can elect when and how to flesh them out. Because players build the characters up from both sides, they are not only believable, but their dialogue carries the weight of emotional entanglement and complexity within a coherent narrative (Riedl 2).

Of course, the options for character development are still scripted by the developers –

players cannot input their own dialogue. The game, however, still grants more agency by allowing players to select from emotionally rich options focused on character development over plot progression, and by allowing them to select dialogue for more than one character. Rather than reducing narrative branches to binaries such as “X responds with anger” or “X responds with sympathy,” Kentucky Route Zero presents more cryptic and nuanced choices, the result of which is the heightened capacity for players to project their understanding and interpretation onto the narrative itself. For instance, players are presented with several different ways to develop Shannon’s character, most of which offer different reflections on her life growing up as a migrant worker in a mining town. When presented with the reasons

Shannon wasn’t able to have a dog as a child, players can select “My folks worked alternating

183

shifts [in the mine] for a while” or “My dad was allergic.” Given the first choice, players will then be prompted to follow up with another statement from Shannon saying, “No time to care for a dog” or “And then they finally got their shifts in sync.” In this interaction, players have more agency to understand the choices, and the weight of those choices, as they want rather than as the game dictates. Each selection by the player realizes Shannon’s character in a different way.

Shannon either becomes a child whose parent was allergic to animals (a not uncommon occurrence) or a child whose parents were always working until (and this is only implied by the narrative) they were lost in a mine collapse. The characters feel believable because players are crafting not just how a character responds in the present moment, but also what their backstory is. As Haywire Magazine’s Editor-in-Chief, Joe Köller, observes, “Kentucky Route Zero turns their full creative power to words, they become a tool not just for announcing future intent, but willing into existence the particulars of your backstory, moving events by casually noting them and establishing your views by expressing them.” Players craft the existence of the characters across planes in time in such a way that the characters become more complex, multi-dimensional, and real. While Mass Effect prompts players to select a backstory for Shepard at the start of the game, the ramifications of this one-time choice are never dramatized in the narrative that follows. It feels like a hollow choice or like lip-service paid to the importance of fully realized characters. Additionally, players of the Mass

Effect series are only ever reminded of Shepard’s past deeds by NPCs. This narrative background, meant to add dimension to Shepard as the protagonist, doesn’t resonate in the same way as the background stories in Kentucky Route Zero. Players of Mass

Effect don’t construct Shepard’s background story; they merely select it once before it

184

becomes the total experience of Shepard’s past. In Kentucky Route Zero, character background is more malleable: the player develops it in slow fragments over the course of the game and in conjunction with the construction of the character’s present experience. Thus, in Kentucky Route

Zero players not only narratively construct what the past was, but why it has bearing on the character and his/her actions in the game’s present. That these actions occur simultaneously brings the multi-dimensionality of the characters directly to the foreground of the game, and makes it the ludic function by which players advance.

Drama Management

The drama management system of Kentucky Route Zero is perfectly hidden in the sense that players are not given any non-dialogue feedback surrounding their choices, and as a result, the story they create is smoothly mutable. Because the main mechanic of the game could arguably be called a "story" mechanic, or a truly “ludonarrative” mechanic, the system managing the story and the way players interact are inseparable from one another. As a result, the need to balance two opposing systems, one narrative and one ludic, is made null. There can be no interventions by the drama manager with the system since the system is the creation of that story.

Furthermore, beyond the presentation of dialogue, the system remains invisible. No feedback is given to players other than the development of the story itself. Additionally, the story has no metric system which assesses or tracks the player's choices. The system, therefore, stays out of the way so that players feel like the story they are creating is uniquely their own. They are the author of it. The drawback to the invisibility of the drama manager is that, unless the game is played more than once, players may not be able to distinguish just how subtly the game alters the players experience around their choices. In other words, without a comparative experience to

185

gauge the differences that occur between playthroughs, the player could feasibly doubt that the system is altering at all as a result of their choices. Without an awareness for how the system is tracking or responding to the player’s choices, the game seriously risks damaging or erasing the potential agency it attempts to provide. This raises the question about how ideal an invisible drama manager actually is, or alternately, the question of whether players have been trained by other interactive narrative games to expect some measure of feedback that assures them the system is actually a narratively interactive one.

Narrative Experience

Despite the invisibility of the drama management system, a new story experience is constructed with each playthrough, provided that players alter their dialogue choices in subsequent playthroughs. Additionally, even if players continue to question whether the system alters the narrative trajectory based on their choices, the game’s replay value also remains high (Mateas and Stern “Towards Integrating Plot” 1). By selecting different choices, the way characters are developed – and thus the player’s understanding of otherwise similar events – will be different. The story changes with each playthrough because the characters within it change.

The design of Kentucky Route Zero and the way it invites players to interact and maintain agency through character development rather than story outcome also increases the opportunity for and importance of critical reflection (Sicart 56). Players don't have to act because they are thinking of the experience/outcome they want, but are instead free to act based on how they perceive the subtleties of characters and how such subtleties may amount to more meaningful choices. Rather than being limited to “right” and “wrong” ethical and ludic binaries, player can make choices in Kentucky Route Zero because, upon critical reflection, it feels right in that

186

moment and within the context of the story experience the player is crafting. It's about the subtlety and nuance that eventually lead to deep and rich character development and story experience.

Conclusion

Overall, this analysis of Kentucky Route Zero shows that, despite the game’s unique approach to character development through multiple protagonists (or perhaps because of it), the game provides an increased level of agency to players by treating them less as inhabitors of a single role and instead as the authors, or in the very-least as co-authors, of the overall game experience. Narratively, players simultaneously think on the level of single characters and on the level of the narrative whole; how their choices affect both that character, the other characters, and their understanding of those characters within the narrative itself. Ludicly, the game compensates for its minimal mechanics by focusing part of Conway’s development on his body.

Additionally, the ludic impetus is not on forward plot progression and momentum, but instead on character enrichment. In tandem, the ludic and narrative capacities for increased agency create a rich system that maintains the weight of choice and the importance of critical reflection. The game’s system, however, is still not perfect. While the drama management system is hidden in an attempt to heighten the player’s immersion, its invisibility doesn’t necessarily improve the overall user experience, such that the player may question how and if the system is altering that experience at all.

The assessment of Kentucky Route Zero using the established analytical criteria also helps to cast a light on two concerns going forward: firstly, that the criteria of assessment works in the analysis of indie games, including those that take a non-standard approach to narrative and

187

interactivity and secondly, that although the criteria work against such innovative games, that their application also reveals possible limitations of the criteria. For instance, the application of the criteria against Kentucky Route Zero reveals that, in some cases, the criteria may be too limiting, especially where it presumes the player acts through a single protagonist. In other cases, it challenges the assumption that a complete invisible drama management system is ideal; if players are completely unaware of its existence, they may question that any choice is recognized by the system at all. While some might argue that this reveals some drawbacks to the criteria, I view this as an essential part of the criteria growth. The criteria I have established are not meant to be static, but evolving. As games evolve so too must our understandings of how agency works within them. As game developers continue to push the boundaries of player interactivity and agency, the criteria we use to assess them must expand as well in a constant feedback loop. This back and forth is important not only for the analysis of games, but also for the grounding of game specific theories in the field. What is simultaneously at stake is the improvement of games and the theories we use to interpret, understand, and improve them. The continued growth and refinement (or ebb and flow) of these analytical criteria is essential to that improvement. The merits of the criteria and of their use continue to be lucrative not only to our understanding of how agency is provided in games, but also in the continued effort to establish critical systems for analysis in the field of game studies.

188

CHAPTER 8

THE COLLAORATIVE CRITERIA CHRONICLES: AN ANALYSIS OF AGENCY AND NARRATIVE IN THE BANNER SAGA

The previous chapters in Part Two outlined and employed a series of collaborative criteria to analyze three digital interactive narrative games: Mass Effect 2, The Elder Scrolls V:

Skyrim, and Kentucky Route Zero. These chapters, however, were all previously published and, as such, represent different phases of research regarding the theories of narrative and agency in digital interactive narrative games. Given this, it is the intention of this chapter to accomplish three specific goals.

First, this chapter revisits the criteria initially made in Chapter 5 and clarifies how the definitions for and consideration of those criteria have evolved throughout the subsequent research process. While no major changes to the criteria have been made that would limit the validity of the criteria established in Chapter 5, this chapter highlights small changes that add more nuance and complexity to those criteria.

Second, this chapter establishes how the criteria laid out in Chapter 5 and employed in

Chapters 6 and 7 directly relate to the theories provided in Chapters 2 and 3. As such, this chapter aims to clearly unite the ideas and theories expressed in Part One of the text with the methodologies and critical analyses employed throughout Part Two.

Third, in the service of the other two goals, this chapter includes the critical analysis of the independently developed digital interactive narrative game: The Banner Saga. Taken together, these three goals cohesively revisit the ideas laid out in this volume while also bringing the work of this text to a close.

189

Before proceeding with the analysis of The Banner Saga, it is important to provide a basic synopsis of this digital interactive narrative game’s structure and plot arch. The Banner

Saga, an indie game developed by Stoic, is a digital interactive narrative akin to a Nordic epic or legend in which the heroes, a group of giants and humans, must wage war against an enemy known as the Dredge while simultaneously attempting to stave off a world-eating darkness that threatens to destroy the world. In conjunction with this plot, the game has two basic modes of play, each which has a specific set of mechanics (Buchanan).

In one mode, referred to here as the “combat mode,” the player agent interacts with the world via a turn-based tactical combat interface. Using this interface, the player takes turns in combat with the computer-controlled enemy. In each turn, the player can move one of several characters on the battlefield, each of whom possesses a unique combat ability and restrictions on how that character can move about the board. The combat mode, in this way, is not dissimilar to a game of chess. As Ben Reeves described it, “players control a team of a half-dozen characters all with their own special attacks and weaknesses and plot out the best way to clear off a chess board of attacking foes” (“Fight for the Top”).

In the other mode, referred to here as the “dialogue mode,” the player agent’s primary means of interaction with the game is through a dialogue selection menu. In these moments, the player agent could be inhabiting one of several character agents, including characters from both the human race and the race of giants known as Varl. While in dialogue mode, the player has no control over which character agent they inhabit as this is dictated by the game system and narrative structure.

190

Despite having two modes, the choices the player agent makes while in each mode can still affect the narrative experience and the story product. As Griffin McElroy notes, “Heroes that fall in battle will require a few days rest before they can fight at full strength – and those who fall too many times will die permanently. Battles aren’t the only places where your heroes can fall.

The Banner Saga follows the caravan of several factions…their stories tied together by a thread of survival” (“The Banner Saga Deals Death”). Given this, The Banner Saga is a rich game for analysis using the amended collaborative criteria outlined in this chapter. Using The Banner

Saga, this chapter accounts for the limitations of the criteria addressed in Chapters 6 and 7 while also clarifying how the criteria can be rearticulated and expanded such that those limitations are minimized.

Revisiting the Collaborative Criteria Alongside the Banner Saga

Each of the previous chapters in Part Two of this volume have articulated the criteria of assessment established in Chapter 5 in similar but different ways. As this was due to the preference of the publications and not a reflection of any authorial change in concept, this chapter proposes a final wording for the criteria of assessment. While these criteria should continue to evolve and change over time, the criteria outline in this chapter should be recognized as the definitive criteria as the reflect the fullness of the research conducted up to this point.

Without belaboring the point further, the table below articulates the five areas of assessment and their criteria in their final form:

191

Table 7. Amended Summary of Criteria Area of Definition Criteria Sources Assessment Drama The degree to • A drama management • Mark Riedl Management which the system should improve the player and Carlos Leon Interactions game’s agent’s experience by ensuring that • Manu Sharma systems ensure a well-formed story occurs, but • Mark Riedl that a well- should also be dependent upon and and Vadim formed story responsive to the player agent’s Bulitko occurs each experience and should adapt to time the game specific player agent’s strategies is played. throughout gameplay. • The drama manager should be hidden. The player should not be aware of the existence of the drama manager or its interventions in the story.

Narrative The player’s • The game should support • Julie Experience overall replayability by providing the Porteous, Fred perception of opportunity for different, well- Charles, and the game’s formed, artistically whole, Marc Cavazza narrative dramatic and narratively coherent • Mark Riedl structure as stories, complete with a relevant • Michael well as the resolution for each play-through. Mateas and player’s • The narrative structure should be Andrew Stern perception of clear enough to help player agents his/her role understand what to do and why within the they should do it, based on their story being own critical reflections. developed within that structure.

Ludic The player’s • Player agents should not be offered • Clint Hocking Agency capacity to merely a small number of obvious • Margaret move about choices that are forced upon them Archer and to interact by the plot. Although dialogue • Michael with the should be an important component, Mateas and environment it should be used to align Andrew Stern crafted by the predominant narrative concerns • Karen game in ways with pre-eminent emotions, instead Tanenbaum and

192

that are not of predictable choices that propel Joshua necessarily, the development of the story Tanenbaum though they forward. Where possible, dialogue • Marie-Laure frequently are, should be removed from menu Ryan linked to the systems altogether. narrative. • Player agents should be able to enjoy meaningful, embodied interactions with the environment, objects, and subjects (NPCs) in conjunction with the developing story.

Narrative The player’s • The player agent must have an • Soren Agency capacity to active role, whereby she influences Johnson meaningfully story events throughout gameplay • Manu Sharma influence the and as a result is able to think • Mark Riedl outcome of the critically about and reflect on and Carlos Leon narrative. choices within the context of the • Marie-Laure narrative structure and developing Ryan the story before and after making • Mark Riedl them. and Vadim • The player agent should be able to Bulitko reflect critically, invest • Fox Harrell emotionally, and evaluate social and Jichen Zhu contexts and values in order to • Karen envision and bring about Tanenbaum and alternatives built within the game’s Joshua narrative structure. Tanenbaum • Greg Costikyan • Michael Mateas • Margaret Archer

Character The capacity to • Believable characters should be • Gonzola Development which non- developed so that the player agents Frasca and player care enough about them to respect • Mark Riedl Interaction characters and them. This entails developing • Bryan Loyall the player- characters with strong personalities • Michael character are and emotions that are rich enough Mateas emotionally so that player agents can form developed and

193

three- emotional attachments. dimensional. • Interactions between NPCs and NPCs and between player agents and NPCs should be about relationships and emotional entanglements; NPCs should not be props used merely to propel action.

Drama Management Interactions

• A drama management system should improve the player agent’s experience by

ensuring that a well-formed story occurs, but should also be dependent upon and

responsive to the player agent’s experience and should adapt to specific player agent’s

strategies throughout gameplay.

• The drama manager should be hidden. The player should not be aware of the existence

of the drama manager or its interventions in the story.

Chapter 5 stipulated that “a drama manager must be present in the system and must improve the overall user experience, but should also be dependent on user experience and should adapt to specific user strategies, in order to ensure a smoothly formed and easily mutable story."

Given the important distinction between story and narrative, and how agents perform within each, a clearer expression for this criterion can be expressed as, “A drama management system should improve the player agent’s experience by ensuring a well-formed story occurs, but should also be dependent upon and responsive to the player agent’s experience and should adapt to specific player agent’s strategies throughout gameplay.” The substitution of "player agent" for

"player" is important here, as the use of this term acts as a reminder that the player agent

194

functions as and mediates the roles of both the narrator agent and character agent. The drama management system must, therefore, account for and support both of these roles, such that the player feels involved in the creation of the story product from the level of narrative and involved in the drama of the unfolding story with the immediacy of and from the perspective of the character agent.

In The Banner Saga, the drama management system is not hidden, and yet the player agent remains largely unaware of how the drama management system is intervening in the creation of the story product or of how it is responding to the player agent’s input.

While in dialogue mode, the drama management system is always visible to the player agent in the form of a Head’- Up Display (HUD). This HUD remains at the top center of the game screen and acts as a constant reminder of several key factors that affect how the story product develops, including the player agent’s current renown (gained from defeating enemies), the number of clansman in the caravan (important in regard to combat and the rationing of supplies), the number of supplies (the caravan’s access to supplies affects the caravan’s morale), the caravan’s morale level (which affects the willpower available during combat), the number of days since the caravan last rested/camped (important to maintaining morale and willpower), and the option to rest/camp. These stats are linked to the game’s narrative structure and affect how the development of the story within the structure, yet what they project to the player agent is, ultimately, no more than general knowledge.

Additionally, while in dialogue mode, the drama management system is also visible via the dialogue selection menu itself. In dialogue mode, the player agent interacts with the game solely through the dialogue menu. This menu appears at the center of the screen and pauses play

195

until the player agent makes a dialogue selection. After the player makes a dialogue selection, one or several things can occur: the player agent could be prompted with more dialogue choices; the player agent could enter into combat; or any of the elements shown on the HUD could change, including the morale of the caravan, the number of supplies, or the number of people in the caravan. These potential occurrences represent the only means of feedback provided to the player agents as a result of their dialogue selections.

Importantly, the feedback offered in dialogue mode always has immediate relevance to

the story experience the player agent is producing and thus could be said to adapt to the

player’s specific strategies. While a game like Mass Effect 2 provides the player with feedback

regarding the player agent’s moral status in the game, the information provided to the player in

The Banner Saga remains pragmatic and practical. For instance, the number of people in the

caravan, the number of supplies available, and the caravan’s morale all have relevance both to

the player agent as narrator agent and the player agent as character agent. While other games,

such as Mass Effect 2, provide the player as narrator agent with information to which the

character agents are not privy (such as a morality score) the information provided by the HUD

in The Banner Saga is information of which the character agents are also aware.

Unfortunately, the same cannot be said of the combat mode in The Banner Saga,

however. When entering combat, the player agent is presented with a character selection

screen. Here, the player agent assembles a combat team before entering any battle. Using this

interface, the player agent can upgrade skill trees for each character. While this allows the

player agent to specialize specific strategies and means of combat play, the information located

in this interface has no relevance on the level of character and does not alter how the characters

196

act or respond in dialogue. Regarding this, Mark Filipowich suggests that, in combat mode,

“people are reducible to their utility” (“Resource Based Humanity). While Filipowich makes an

astute observation here regarding what some may consider to be a fragmentation between the

two modes of play in the game, I suggest that this interface is merely one interface in the game

with which the player agent engages solely as the narrator agent.

Additionally, while in combat mode and upon entering combat, the player has access to a

unique combat interface. This interface displays where each character is within the turn

rotation, and shows each character’s combat stats, including attack combat, willpower, and

special ability points.

Despite the fact that dialogue mode and combat mode are entirely distinct from one

another in terms of their interfaces, what happens on the battlefield does affect what happens

while in dialogue mode and vice versa. The outcome of a battle determines the number of

casualties sustained by the caravan. This, in turn, has a ripple effect on the caravan’s ability to

sustain supplies and maintain morale. This, then, is likely to alter how the player agent makes

choices while in dialogue mode. The choice made in dialogue mode can alternately affect the

amount of willpower and morale the characters take into combat. Thus, combat mode

maintains an important link to the dialogue mode and, therefore, to the narrative structure of

the game.

Table 8. Analysis of Drama Management in The Banner Saga The Banner Saga Drama Management Absent/Low/Medium/High The drama manager improves player agent’s High experience

197

The drama manager adapts to player agent’s High experience

The drama manager is hidden from the player Absent agent

Narrative Experience

• The game should support replayability by providing the opportunity for different, well-

formed, artistically whole, dramatic and narratively coherent stories, complete with a

relevant resolution for each play-through.

• The narrative structure should be clear enough to help player agents understand what to

do and why they should do it based on their own critical reflections.

Chapter 5 stated that “there should be different well-formed/artistically whole, dramatic and narratively coherent stories, complete with a relevant resolution that occurs smoothly during each play-through and that should support replayability.” Additionally, the chapter stipulated that

“the narrative should be clear enough to help players understand what to do and why they should do it, based upon their own critical reflections.” The first criterion has been reworded here to stress that this criterion is reliant upon the demands of replayability. While replayability is often used to sell a game to consumers, as the number of hours a player can engage with a game helps to validate a game’s cost, replayability is given precedence here because the replayability of the game highlights the central difference between digital interactive narrative games and other narrative media: to replay isn’t to experience again, but to experience anew. While a reader’s perception of a novel will certainly change with repeat readings, the drama and action within the text remains the same. While one production of Romeo and Juliet may differ from another, it is

198

certain that the young lovers will die. As established in Chapter 2, this is the difference between narrative and story, and in most mediums, the story product remains the same regardless of how different one narration of it may be from another. In digital interactive narrative games, however, the player can replay all the same choices over and over in order to replicate and recreate the same story product if they choose to, but the player may also alter her choices in such a way that a different narrative experience and story product are produced. As Greg Costikyan has lamented, the disappointment with so many digital interactive narrative games is that they tend to be “beads-on-string” narratives, in which the beads represent “small areas where there is some freedom of action,” but where the string is the overall “linear progress” of the story experience

(“Games, Storytelling, and Breaking the String” 8). In other words, the choices made make no difference in terms of the resultant story outcome. The challenge and opportunity presented by digital interactive narrative games, therefore, is to ensure that the narrative structure provides enough variety and enough difference that, upon repeat playthroughs, the story product is not only different but also as coherent, well-formed, and artistically whole as every other story product. In Narrative Design: Working with Imagination, Craft, and Form, Madison Smartt Bell likens the work of constructing a narrative to the work of carving a sculpture from a block of stone (213). In both instances, what happens is a carving away of excess so that what remains is what is essential for the piece to take shape. The creation of digital interactive narratives is different then, in that it imagines the many rather than one sculpture waiting to be carved from the rock, and it makes them each a possible result.

The narrative structure of The Banner Saga is clear enough for the player agent to understand what to do and why to do it, both on the level of narrator agent and character agent.

199

Additionally, given the minimal feedback provided to the player agent, most of the player agent’s choices are dictated by her critical reflections and understanding of how the story has developed so far and how she would like it to develop in the future. For instance, while in dialogue mode, the player agent, regardless of which character agent they currently inhabit, is likely to reflect on the morale of the caravan, the number of supplies available, the number of fighters available, and so on when selecting any dialogue option. In the desolate landscape of the game, there is strength in numbers, but the player agent also risks causing potential mutiny should the number of people in the caravan outnumber available supplies. As Reeves notes,

“every bend is filled with tough choices and difficult encounters that will leave your heroes constantly wrestling to find food and supplies” (“Fight for the Top”). Therefore, the player agent must consider what is ethically right for the good of the entire caravan when making choices.

The only input to help the player agent gauge such choices is her understanding of the story and the stats reflected to her by the HUD. As an example, in one randomly-generated event, a group of starving people approaches the player agent’s caravan seeking food. The player agent can elect to welcome them to the caravan, despite the caravan’s mistrust of the outsiders; offer the strangers supplies without welcoming them to the caravan, despite potentially being low on supplies; or dismiss the strangers without offering them anything. Here, there is no wholly good outcome. If the player agent welcomes the strangers, the caravan’s morale will be weakened. If the player agent offers supplies, the caravan may go hungry, which also leads to decreased morale; or the player can ignore the strangers who, later in the game, attack the caravan and steal supplies anyway. Of course, the player agent cannot anticipate these outcomes when making a choice and must, therefore, reflect upon what she feels is the most ethical choice for the caravan.

200

Throughout the game, the player is, therefore, compelled to care for the caravan as a whole and to consider the circumstances of the characters when making choices.

The game also supports replayability, though with some limitations. Throughout gameplay, the narrative structure allows for a variety of events which alter gameplay and story development. For instance, within every chapter or level of the game, a series of random events can occur. While some of these events have a high likelihood of repeating in subsequent playthroughs, others are programmed to occur less frequently (“Additional Events - The Banner

Saga Wiki Guide”). Additionally, some events will happen in every playthrough, but depending on the choices the player agent makes, the outcome may be different. For instance, depending on the player agent’s choices, some characters may live or die, or some characters may become active heroes or characters who can be used in combat mode. The most prominent example takes place in the game’s final chapter/level. Here, the player agent will find himself in the town of

Boersgard. Most of the citizens in the town have fled, and the Dredge army will soon arrive at the gates. Over the course of the next five days, the player agent must keep the city from being overrun by small bands of arriving Dredge, find supplies to keep morale high, help build ships at the dock, investigate riots, find space for people to stay, and so on (“The Banner Saga

FAQ/Walkthrough for PC by Mythril Wyrm - GameFAQs”). What the player agents elect to do, and the order in which they elect to do them, will affect the status and circumstances – in terms of morale, energy, and the number of fighters – of the character agents the player agent can use in the final battle. These events, whether randomly generated or programmed into every playthrough of the game, are part of the narrative structure of the game and, as a result, have an effect not only on gameplay and strategy, especially in terms of combat mode, but also on the

201

story being developed. That said, despite all the narrative variance offered throughout gameplay, the story product can only be resolved in one of three ways. In one scenario, the player agents find that the choices they have made throughout gameplay have rendered it impossible to defeat the final boss. In another scenario, the player agent selects the character agent Alette to shoot a special arrow into the final boss, Bellower, in which case Alette dies. In the last of the potential scenarios, the player selects the character agent Rook to shoot a special arrow into Bellow, in which case Rook dies. Of these potential endings, the first ending could be considered the “fail” ending in which the player has effectively lost the game. The latter two options, both considered win states, are effectively the same in terms of gameplay: a major or “main” character agent dies.

Yet, despite their overall similarity, the emotional impact of each will be resoundingly different for the player agents based on how she has engaged with the narrative structure to develop the story through gameplay. Of course, given that the game only provides these three endings in every playthrough of the game, the endings are likely to lose emotional resonance with repeat playthroughs. Thus, while the game offers a rich narrative structure full of variance, this variance amounts to little with respect to the final outcome of the story, which must resolve itself in one of only three ways. With respect to The Banner Saga as an individual game, this lack of variety may be said to limit replayability, but it is also important to note that Stoic made The Banner

Saga with the full intent of producing a sequel. As the player agent’s choices in the first game carry over to the second, the ending of The Banner Saga may be limited in order to maintain narrative coherence in the sequel. Taken on its own, however, The Banner Saga limits its replayability by limiting the ways in which the story product can be concluded.

202

Table 9. Analysis of Narrative Experience in The Banner Saga The Banner Saga Narrative Experience Absent/Low/Medium/High Game provides at least one complete story High

Game provides possibility of different story outcomes Low

Each story outcome is coherently constructed High

Each story outcome has a relevant resolution Medium

Multiple story outcomes make the game replayable Medium The narrative structure is rich enough for the player agent to make High informed choices

Ludic Agency

• Player agents should not be offered merely a small number of obvious choices that are

forced upon them by the plot. Although dialogue should be an important component, it

should be used to align predominant narrative concerns with pre-eminent emotions,

instead of predictable choices that propel the development of the story forward. Where

possible, dialogue should be removed from menu systems altogether.

• Player agents should be able to enjoy meaningful embodied interactions with the

environment, objects, and subjects (NPCs) in conjunction with the developing story.

The criteria for ludic agency remain more or less the same as in previous chapters. Where changes have been made, they were made to draw a clearer distinction between story and narrative. As mentioned in Chapter 2, some of the ludic functions that directly affect story construction, and which may be managed seamlessly by a drama management system, are those

203

that are presented to the player in order that she may interact with the narrative structure. For example, dialogue screens, morality systems, quick time events (QTEs), and heads-up displays

(HUDs) are all ludic elements that accommodate interaction by the player with the narrative

(Joyce "Assessing Mass Effect 2"). Again, as mentioned in Chapter 2, these ludic systems establish the difference between the player’s role as narrator agent and character agent and allow the player to mediate, interact with, and switch between these two roles.

While the wording of the criteria has changed little, it is also important to consider here how these criteria link back to the theories of agency considered in Chapter 3. Compatibilist theories of agency, for instance, could support the argument that players lack agency when they are forced by the system of the game to act against their desires, wills, or understandings of the choices of available to them. Therefore, to support ludic agency, the player agent should not be offered false choices that negate the player agent’s ability to feel his or her choices matter with respect to developing the story. In other words, the player should be able to maintain the perception that her intent matters, and likewise, as Functionalist theories could support, she should be able to actualize that intent in a variety of ways in order to achieve a goal.

Additionally, as Hayles’s theory of posthumanism reminds us, our thinking mind is part of the active body, each inextricably linked to the other (4). Therefore, the player’s ability to think critically must be linked to the player’s ability to interact in such a way that the player agent feels she is embodied in the digital world of the game. Without the ability to feel existent in the space the game creates, or in other words for the player to feel part of the causes and effects of the game environment, both her critical thinking and attempts to actualize goals will lose the feeling of intentionality or possibility.

204

Regarding all five areas of assessment, The Banner Saga most obviously fails to meet the criteria established for the assessment of ludic agency. Despite the variety of narrative experiences the player agent can construct throughout gameplay, the player agent is still “on rails” ludicly throughout gameplay. In other words, the player agent’s location and movement through the game are not in the player agent’s control. As reviewer Daniel Tack observes,

“travelling through the wilderness from town to town represents a significant portion of the gameplay, full of stops that require input from the player to move on” (“The Banner Saga

Review”). Here, Tack pinpoints one issue with the ludic agency. The player’s trajectory from space to space is outside of her control despite the input required of her as she moves between those spaces.

The only time the player agent controls her movement is during combat mode, and even then, the player agent is restricted to the chess-like combat environment. Otherwise, and despite that the narrative is structured to be a travel epic, the player agent cannot control her movement across the game’s landscape. Each location is predetermined by the narrative structure.

Additionally, the game system also dictates which character agent the player agent will inhabit during dialogue mode sequences. The player cannot move between character agent roles of her own accord. Furthermore, the player agent cannot interact at will without other characters. Only while at camp, which is activated when the player agent elects to let the caravan rest, can the player agent opt to speak to other characters. Even in these moments, the player agent cannot talk to any character she wishes but is instead given a set of options for characters who are available for conversation. In this way, the narrative structure overrides ludic agency, and the characters available to speak with at camp prioritize the narrative concerns of the game over the player’s

205

preference. The player agent also cannot interact with the environment. In addition to being on rails, the player cannot manipulate the settings or spaces in which they find themselves, especially in dialogue mode. While in dialogue mode, the only means of interaction is with the dialogue selection menu and the HUD. In combat mode, the player is still restricted to a tiled combat board on which the player agent simply selects the tile to which they want a character agent to move and the skill they want the character agent to use once on the tile. In general, throughout The Banner Saga, the player agent can only ever manipulate the character they currently inhabit, and therefore, cannot be said to enjoy meaningful embodied interactions with the environment, the objects, or characters they are not inhabiting.

Despite these limitations, the choices provided to the player are rarely obvious in terms of the overall plot and its trajectory toward such a limited number of potential story conclusions.

While the player agent is constantly forced to make choices that will help the developing story to progress, these choices maintain the immediacy of the current moment and, therefore, do not feel as though they are obviously forced upon the player in the service of the plot. As Filipowich observes, “The game does not consistently reward the same morality…mercy makes a hero one day and a fool the next. The ‘Hard Choices’ of the game actually feel like they’re participating in the narrative thrust of the game” (“Resource Based Humanity”). For instance, there is an event in

Chapter Two of the game when the player agent, currently controlling the character agent Rook, encounters two men, Hogun and Rogun, both of whom wish to join Rook’s caravan. The people of the village Hogun and Rogun come from do not want the two men to leave, claiming that they need them to stay and defend the village. Hogun and Rogun, however, know the village is doomed, whether they stay or not. Given this situation, the player agent is presented with three

206

choices: tell the men and the villagers to handle the situation themselves, side with the villagers, or side with Hogun and Mogun. Each choice leads to a different outcome in which the player agent gains a different level of Renown, a different number of supplies, and the possibility of

Hogun and Rogun, two accomplished fighters, joining the caravan. Again, however, the player agent cannot anticipate these outcomes, and thus, when deciding whether to get involved and in what capacity to get involved, the player agent must do so, most prominently, from the level of the character agent. The player agent’s foresight or access to information outside the character perspective is so limited that the player agent is compelled to make choices from the level of knowledge available at the character level. That said, because the dialogue options are relevant to the player agent’s current circumstance and have an immediate effect on the player agent’s current status in the game (as well as an unknown effect on the future), these choices can also be said to align the predominate narrative concerns with pre-eminent emotions. Offering perhaps the highest proof of praise, reviewer Leif Johnson said, “The Banner Saga made me care” (“The

Banner Saga Review”).

Table 10. Analysis of Ludic Agency in The Banner Saga The Banner Saga Ludic Agency Absent/Low/Medium/High

A small number of choices don't force plot Medium

Obvious choices don't force plot High

Dialogue is emotionally-driven High

No dialogue system at all Absent

Embodied interactions with environment Absent

207

Embodied interactions with objects Low

Embodied interactions with subject Low

Narrative Agency

• The player agent must have an active role whereby she influences story events

throughout gameplay, and as a result, is able to think critically about and reflect on

choices within the context of the narrative structure and developing the story before and

after making them.

• The player agent should be able to reflect critically, invest emotionally, and evaluate

social contexts and values in order to envision and bring about alternatives built within

the game’s narrative structure.

The changes made to the criteria for narrative agency have been made, once again, to point distinctly to the difference between narrative and story as outlined in Chapter 2. Of course, this distinction is incredibly important within the discussion of agency related to narrative.

Chapter 5 stipulated that the player agent should influence events, but it is important to clarify here that events are a story component. Thus, while the player may guide and influence events both as narrator agent and as character agent, the resultant events are merely contained within a narrative structure but result in the development of the story product. Following this line of thought, the player agent should be able to think critically about and reflect on choices within the context of the narrative structure and within the context of the developing story before making them.

208

Chapter 5 also suggested that the player agent should “not feel as though (s)he is playing a role that forces one to reflect on how a character would react.” This criterion should be clarified, too, as it applies only to those moments in which the player agent inhabits the role of character agent. While acting as a character agent, the player should feel connected to that role.

That said, the player agent should also be able to remove himself from the role of a character in order to guide and influence the developing story from the external role of narrator. In other words, the narrator agent should be free to reflect on the developing story objectively rather than subjectively, as a character agent would. Yet, while it is important for the player to be able to maintain both roles, the ability to maintain a position of narrative remove helps the player agent became more familiar with character motivations, emotions, and actions. Thus, rather than limiting the intimacy a player can maintain with a character agent, the ability to engage objectively at the level of narrator agent increases it. Of course, the level of intimacy a player maintains with the characters and within the developing story is also dependent upon her ability to think and engage critically with the game.

Chapter 5 also stipulated that “the user should be able to reflect critically upon and evaluate social context in order to envision and bring about narrative alternative.” As noted in

Chapter 3, much of a player’s agency in a game is derived from the players’ perception of their will and their ability to actualize their intentions in the game. Provided that this perception is psychologically stimulated in the player, we must consider that the player’s sense of immersion cannot always be stimulated by ludic mechanics. In many cases, the perception of will and agency will be derived, instead, by how frequently the game incites the player to engage critically and actively with the narrative’s structure and within the developing story. The player

209

must be able to mediate what can and cannot be done in such a way that the player feels immersed in the unfolding story and trusts that her actions and choices will reverberate in the game world.

As noted in Chapter 3, a player’s perception of narrative agency will likely be heightened if the game anticipates and responds to the player’s values. Of course, it is no easy task to create a game that can account for and anticipate any one player’s set of values or morals. A game, however, can adopt a set of ethics and, in doing so, require the player to adhere to them (Sicart

The Ethics of Computer Games 22). Yet, so long as the player can understand the code of ethics employed by the game, the player can anticipate and adhere to them in such a way as to maintain a sense of agency in the game world. Thus, the player’s perception of agency is linked to how well the game system makes its ethical structure apparent and comprehensible to the player. As

Miguel Sicart notes, games create argument through their rules, and “the player’s understanding of that model makes a game have meaning” (Beyond Choice 34). In order to act with meaningful intent, the player must be able to think critically within the context of the rules dictated by the game. A player who understands the rules and how they reverberate through the game world can, by extension, engage on a deeper level emotionally and intellectually within the narrative structure and, ideally, interact in such a way that she feels like an active participant in shaping the story product.

To that end, the narrative variety offered to the player agent within the The Banner

Saga’s narrative structure is largely influenced by the choices the player makes while in dialogue mode. In turn, the dialog presented to the player is focused on ethical dilemmas. While in dialog mode, the player agent must reflect on the dilemma, the means of addressing it as provided by

210

the game in the dialog selection menu, and the potential outcomes any means of addressing it may have. Therefore, the player’s role and ability to influence story events actively are contingent upon the player’s understanding of the game’s ethics. Most broadly, the rules of the game communicate to the player that her choices, though made individually, have repercussions for all. In his review of the game, Mark Filipowich observes, “The [player’s] long march across the continent is motivated by a duty toward…the clan. The best hope is doing right by [the clan] and minimizing loss, not saving the world or winning the day” (“A Family that Plays Together”).

As noted previously, any choice the player makes immediately alters the status of the caravan in terms of willpower, supplies, morale, and renown. In The Banner Saga, the HUD acts as a constant reminder of the game’s ethics and of how the player agent’s choices influence the narrative experience and the story events. While a game such as Mass Effect 2 tracks and scales the level of morality of the character agent, Shepard, The Banner Saga instead tracks the general welfare of the caravan. In this way, the game focuses on the welfare of all over the welfare of one. It is clear that the player agent’s ethical goal is to maintain the welfare of the caravan above all else.

It is logical, then, for the HUD to remain visible whenever the player agent is in dialog mode as the information contained there supports the player agent’s ability to think critically about the choices presented in the dialog selection menu. Furthermore, the visibility of the HUD increases the player agent’s emotional investment, while the dialog menu presents the player agent with the opportunity to evaluate social contexts and make valuated choices. As mentioned in the discussion of narrative experience, The Banner Saga offers the player numerous ethical dilemmas to which there is no entirely good or entirely bad outcome. Matthew Buchanan notes,

211

“You are confronted by story encounters where a seemingly shrewd choice costs you dearly or where indulging in a fit of scruples yields riches” (“Can Games Teach Us to Die”). Yet, when such dilemmas are presented to the player agent, she must consider what she knows regarding the circumstances of her caravan and what she knows of the game world – i.e. the social contexts created in a world on the brink of destruction when people are scared and hungry – when deliberating the choices that the dialog selection menu provides. In Griffin McElroy’s words,

“the consequences of your decisions are often quite dire – it’s not just your trail of peasants that can be swallowed up as you travel. Sometimes, heroes – even main story characters – can be assigned heroic duties, like sacrificing themselves to buy your caravan some time” (The Banner

Saga Deals Death”). Because the player agent’s choices affect the stats of the caravan, the player agent is asked to emotionally invest in the well-being of the caravan, but the dialog presented to the player agent often complicates this primary objective. For instance, when the player agent encounters NPCs who wish to join the caravan, the NPCs will often plead or share emotional stories of what has happened to them. Thus, the player agent is frequently put into a position where her feelings of sympathy for individuals who have done no direct harm may, ultimately, harm the caravan she is meant to care for. Thus, the player is not only apt to feel emotionally entangled in her choices but will need to assess her choices based on the social context of the dilemma with which she is presented. For instance, will the caravan view her as a harsh leader if she rejects those in need simply because supplies are low, or will the caravan resent her for welcoming more people to the caravan because supplies are low? The game makes such dilemmas common place and uses them as a means to keep the player agent actively involved in how the story product is shaped.

212

Table 11. Analysis of Narrative Agency in The Banner Saga The Banner Saga Narrative Agency Absent/Low/Medium/High Player agent has active role High

Player agent influences the events that occur High

Player agent influences how the story ends Low

Player agent is not playing a role Medium

Player agent needs to reflect High

Player agent can bring about narrative Medium alternatives

Character Development and Interactions

• Believable characters should be developed so that the player agents care enough about

them to respect them. This entails developing characters with strong personalities and

emotions that are rich enough so that player agents can form emotional attachments.

• Interactions between NPCs and NPCs and between player agents and NPCs should be

about relationships and emotional entanglements; NPCs should not be props used

merely to propel action.

Compared to the criteria outlined in Chapter 5, I have changed very little. As with the other criteria in this chapter, I have merely changed “player” to “player agent” as a reminder of the dual roles the player agent holds as detailed in Chapter 2. Though the modification here is minimal, the distinction between the character agent and the narrator agent roles are, again, of

213

the utmost importance. The analysis of Kentucky Route Zero in Chapter 7 suggests that the criteria for character development and interactions are too limiting because they do no account for the player's potential to inhabit multiple character roles. If, however, we maintain a distinct understanding of the player agent as an agent who mediates the experience from both the level of narrator and character, it also becomes possible to account for the player agent holding more than one character role. Thus, as suggested in Chapter 2, we need to expand our thinking from a single player who inhabits one character to a player agent who can inhabit multiple character agent roles as well as a narrative agent role. In doing so, the criteria remain applicable and flexible enough to account for a variety of player agent roles.

The Banner Saga is an apt example to show the flexibility the criteria maintain in accounting for a variety of player agent roles. In both the combat mode and the dialog mode, the player agent will inhabit more than one character agent role.

Combat in the game is turn-based. Additionally, combat is rotational. Before a battle, the player selects several characters to bring into battle and creates the rotational order for the turns these characters take. As such, for each turn, the player agent inhabits a single character agent.

Once in battle, the player agent is provided with an aerial view of the battlefield. From this vantage point, positioned outside the realm of the battlefield, the player maintains a certain sense of objective removal, as would the narrator agent. Additionally, however, for each turn, the player agent also inhabits a single character agent, acting as that specific character throughout that specific turn. That said, when not currently inhabited by the player agent, the characters in the battle continue to act as NPCs, complete with system-generated dialogue and physical combat reactions. This allows for the player agent to remain emotionally attached to each

214

character agent both while she inhabits it and while she does not. The NPCs never become marionettes such that they only exist when activated by the player. Instead, the characters maintain a sense of life whether the character agent inhabits them or not. This way, when the player agent does inhabit a character agent, she does so with a clear picture of whom that character agent is based on what has occurred and what has been said in the moments since she last did so.

The player agent will also inhabit several character agents while in dialogue mode. At the start of the game, for instance, the player inhabits Ubin, one of the giant Varl. Appropriately,

Ubin is known by the nickname “narrator.” While the player will not inhabit the role of Ubin long in terms of dialogue mode, it is through Ubin that the player agent learns the narrative structure of the game and its mechanics. Once initiated to the general workings of the game and once made aware of the central plot of the narrative structure, the player agent moves to inhabit

Rook, a human hunter. In terms of the gameworld, Rook and the caravan he will come to lead start on the far eastern side of the world map. Meanwhile, the giant Varl Hakon and the caravan he comes to lead, of which Ubin is a part, starts on the far western side of the world map. Thus, when the player agent is switched between these two character agents, she is also changing her physical location in the gameworld as well as the caravan and the NPCs she is leading. Each caravan maintains its own number of supplies and people as well as its own level of morale and willpower. Given this, the player agent forms emotional attachments to two distinct groups of characters. The characters in each caravan are unique, and their personalities are strong enough that they complicate and affect how the player agent makes decisions with respect to that specific caravan. As Daniel Tack observes in his review of the game, “the act of promoting characters [in

215

combat mode] carries with it some degree of risk, as the character could be lost forever during storyline [dialogue mode] development….it certainly drives home the theme that choices matter and have serious repercussions” (“The Banner Saga Review”). For instance, Rook’s caravan has a rebellious character, Ekkill, whose actions make him hard to trust. Meanwhile, Hakon’s caravan includes a human prince who is obstinate in most conversations, but whom the player agent feels compelled to keep happy in order to maintain peace between the humans and the giants. In this way, each caravan has a unique set of characters and personalities for the player agent to interact with. Furthermore, the interactions between the player and the characters are generally focused on emotions. While the effect of any interaction will be pragmatic – a loss or gain of supplies and morale – emotions are the impetus behind both interactions. In dialogue mode, most characters speak to the player agent in such a way as to share their current feelings or to relay how others in the caravan are feeling. Thus, when the player agent selects dialogue options, she does so in such a way that she is aware of and can consider both the emotional and pragmatic ramifications of all options.

Table 12. Analysis of Character Development in The Banner Saga The Banner Saga Character Development Absent/Low/Medium/High Player agent cares about characters Low

Player agent respects characters Low

Characters have strong personalities Medium

Characters have strong emotions High

Player agent can infer character’s thoughts Low

216

Interactions are about emotional entanglement High

Relations are not props for action Medium

Conclusion

In general, there are very few criteria The Banner Saga fails to meet across the five areas of assessment. In fact, of all the criteria, the game failed to meet only three: the drama management system was not invisible, the game did not enable embodied interactions with the environment of the gameworld, and the game used a dialogue selection menu.

Interestingly, two of those criteria leave little room for gray either; the game either does or does not have a visible drama management system and the game either does or does not use a dialogue selection menu. While The Banner Saga did have a visible drama management system and used a dialogue selection menu as one of its key mechanics, I would argue that these elements are incorporated into the game in a far less intrusive way than in other digital interactive narrative games. In the case of The Banner Saga, the HUD is a visible reminder of the drama management system, but that HUD is also the tool by which the player can assess her circumstances and make decisions within the game. In the case of The Banner Saga, the visibility of the drama management system does not fragment the roles of narrator and character, but instead, keeps them united. Additionally, the use of dialogue as a primary mechanic is logical in The Banner Saga. The dialogue system is used to align the predominant narrative concern (to care for and keep the caravan safe from harm) with emotions that conflict with this goal. In other words, the dialogue system shows emotional complexity in the way it encourages the player agent to make choices in which there is no absolute moral good or bad. As noted, other systems

217

in the game limit the actual emotional depth of the NPCs, but it is not the dialogue system that does so.

The analysis of the criteria of assessment also highlights the numerous ways in which the criteria intersect and affect one another. For example, the goal or plot created within the narrative structure will likely affect how the player agent relates to the characters she encounters. In The

Banner Saga, the player is compelled to place the needs of the caravan as a whole above the needs of individual characters. With this priority in mind, the player’s ability to relate to characters as complex individuals – rather than as extensions of the collective – is affected. As such, the criteria for character development and the criteria for narrative experience become linked.

Given these considerations, it is clear that the criteria should allow for more flexibility.

This flexibility should exist not only in how each singular criterion is defined and assessed but also in how the criteria relate to one another. Of course, while the criteria should continue to be refined and adapted, as was always the intent, it should be noted that the criteria continue to be a useful starting point for assessment and analysis. These criteria work to establish a context or background against which to notice the intricacies and nuances that exist in relation to each digital interactive narrative. Additionally, as the next and final chapter shows, the criteria also provide a cohesive lens through which to compare and understand digital interactive narrative games as a genre too.

218

CHAPTER 9

SUMMARY OF FINDINGS

In the previous three chapters, the criteria for assessment established in Chapter 4 were used to analyze four digital interactive narrative games, each of which was intentionally developed to include a narrative structure with which the player can interact to produce a unique story product. Two of these games, Mass Effect 2 and Skyrim, are AAA games, produced and funded by large production companies. The other two, Kentucky Route Zero and The Banner

Saga are indie games, produced by small independent developers without the financial backing or support of a large company. Each of the four titles also has a distinct play style, plot, environment, and set of mechanics. Each digital interactive narrative is incredibly different from the others. In fact, what unites these four games is only that each can be defined as a digital interactive narrative. As such, these four games were selected precisely to test the quality and value of the criteria across a spectrum of digital interactive narrative games.

Additionally, the previous chapters illuminate how the criteria can be used to identify specific qualities about each game. The application of the criteria to Kentucky Route Zero, for example, reveals that the game’s unique narrative structure and the type of choices presented to the player in the dialogue selection menu greatly increase the player agent’s level of narrative agency and the means by which characters are developed. Conversely, the same assessment revealed that the dialogue selection menu in Mass Effect 2 is too easily decipherable in terms of the game’s dependence on a binary morality system and, as such, decreases the player agent’s level of ludic and narrative agency.

219

As Chapter 5 also illustrated, the criteria are beneficial in a comparative analysis as well.

Using the criteria as a means to compare Mass Effect 2 to Skyrim reveals that Skyrim’s open- world environment increases the player agent’s level of ludic agency, but also that the open- world environment does not adequately support the narrative structure of the game. On the other hand, in Mass Effect 2, the player agent has less freedom to explore the world, and yet, this decreased ludic agency does not increase the player’s narrative agency. Thus, the comparison of these two games reveals that ludic agency and narrative agency are not necessarily in conflict.

Having used these criteria to assess these four games individually and to assess Skyrim and Mass Effect 2 comparatively, it is now my intention to compile the analyses of the game together, to analyze what the criteria reveal about each game when compared within one area of assessment, to consider what the criteria reveal about each area of assessment in relation to digital interactive narrative games as a whole, and to provide considerations for how the criteria might continue to be developed and improved in the future.

Drama Management

Results

Table 13. Final Analysis of Drama Management Kentucky Route The Banner Mass Effect 2 Skyrim Zero Saga Drama Management Absent (0), Low (1), Medium (2), High (3)

The drama manager improves player agent’s Absent Low High High experience

220

The drama manager adapts to player agent’s High Medium High High experience The drama manager is hidden from the player Absent Absent Medium Low agent .33 .33 .88 .77

Based on the criteria of assessment, Kentucky Route Zero most clearly meets the criteria for an

effective drama management system across the digital interactive narrative games analyzed in

this text. The success of the drama management system in Kentucky Route Zero is largely

dependent on its near invisibility. In the other three games, the drama management system is too

visible to the player. In Kentucky Route Zero, however, the system that manages the drama is

also the key mechanic with which the player interacts – the dialogue selection menu. Given this,

Kentucky Route Zero is unique in that “the system managing the story and the way players

interact are inseparable from one another” (Joyce “Kentucky Route Zero” 24). Additionally, the

drama management system in Kentucky Route Zero limits feedback given to the player, such that

the only feedback provided are new dialogue options the player can use to continue crafting the

story product. Comparatively, The Banner Saga is similar to Kentucky Route Zero so long as the

player is in dialogue mode. While in dialogue mode, the player agent is given minimal feedback,

but what feedback they are given is directly relevant to the player’s position as narrator and

character. That said, the feedback is omnipresent via the HUD, so the player agent is also always

reminded of the system that is mediating and tracking her experience, even if the ways the

system does so aid her ability to think critically about what to do next. Additionally, while in

combat mode, the game provides an additional screen showing the stats of the combat-eligible

221

characters. Here, the drama management system totally fragments the role between narrator agent and character agent. When interacting with this system, the player agent operates fully in the narrator role, viewing all the characters objectively and from outside the drama of the developing story. Similar interfaces can also be seen in Mass Effect 2 and Skyrim. In the case of

Mass Effect 2, the drama management system also consistently provides the player with feedback that not only decreases her need to think critically, but also further fragments the player’s ability to move between the roles of narrator agent and character agent seamlessly. In Skyrim, the drama management system adapts to the player’s play style in terms of mechanics and, in this way, enhances the player’s ludic experience. That said, the drama management system does little to enhance the narrative experiences the game offers. In other words, the drama management system does not support or adapt in order to aid in the production of a coherent or emotionally resonant story product.

Future Considerations

Provided that the drama management system was detectable in each of these four games, it is perhaps unreasonable to suggest that drama management systems should be hidden. Instead, the criteria should be amended such that they assess whether a drama management system, either visible or not, smooths the transitions between the player agent’s interventions as the narrator agent and as the character agent. In other words, the drama management system should allow the player agent to act on the level of narrator agent without fragmenting the immersion she maintains as the character agent, and vice versa. The drama management system should, therefore, allow the player agent some level of remove and objectivity, but not so much so that

222

the player agent loses a sense of emotional investment in the development of the story on the

level of character. Ultimately, a drama management system that improves the player’s

experience will do so by considering how and in what capacity the player agent interacts on the

level of narrator and where the player agent will interact on the level of character. Additionally, a

drama management system that improves the player's experience will do so by supporting the

transition between and balance of these two roles such that they feel seamless to the player agent.

Narrative Experience

Results

Table 14. Final Analysis of Narrative Experience Kentucky Route The Banner Mass Effect 2 Skyrim Zero Saga Narrative Experience Absent (0), Low (1), Medium (2), High (3)

Game Provides at least one High Medium High High complete story

Game provides possibility of Low Absent Medium Low different story outcome

Each story outcome is Low Absent High High coherently constructed Each story outcome has a Low Low Medium Medium relevant resolution Multiple story outcomes Medium Low Medium Medium make the game replayable The narrative structure is rich enough for the player Medium High High High agent to make informed choices .55 .38 .83 .77

223

While all the games could be said to provide at least one complete story, Kentucky Route

Zero best meets the criteria of assessment for narrative experience in terms of variance and replayability. One might assume that the game succeeds above the others in this category because it is more like a Choose Your Own Adventure book than it is like a game. While the other three games have clear win states, Kentucky Route Zero does not. While death or the loss of the game is a possible experience in the other games, in Kentucky Route Zero it is not.

Similarly, there are no stats or metrics in Kentucky Route Zero; there is nothing to level up and no numerical way to judge one character, object, or situation against another. The game is meant, above all else, to provide the player agent with an interactive narrative experience, and with this sole focus in mind, the game succeeds above the others.

For all its focus on story creation, the narrative structure of Kentucky Route Zero is still not robust enough to score a high rate of replayability. As with Mass Effect 2 and Skyrim,

Kentucky Route Zero’s narrative structure results in a similar story resolution with every playthrough, despite the variance of choices offered to the player throughout gameplay. That said, where Mass Effect 2 and Skyrim scored low in terms of different story outcomes, Kentucky

Route Zero scores a medium. The difference, here, is a reflection of the unique way Kentucky

Route Zero reveals its narrative structure to the player. Because the player agent can select which character agent speaks as well as what that character agent says (to the degree that the game provides them the dialogue option to do so), the player agent has more direct input on how the story will be told from the narrative level. By extension, how the player agent comes to understand the story product will be different despite what is otherwise a similar resolution to previous playthroughs. In Mass Effect 2, for example, the player agent can elect never to speak to

224

or take a certain character in combat in one playthrough, and then to always do so in another.

Despite this variance, the player agent will always inhabit the character agent Shepard. In

Kentucky Route Zero, however, the player agent has some ability to decide who speaks and who is spoken to, thereby altering the narrative delivery and the story product, but not the story resolution.

Interestingly, Mass Effect 2 has the lowest rating in terms of a narrative structure that supports rich choices. While all three of the other games earned high scores, Mass Effect 2 earned only a medium. Mass Effect 2 scores lower than the others because of its drama management system. While the game provides the player agent with many choices, those choices adhere to such a strict binary on the game’s morality spectrum that they fail to maintain complexity or richness and negate the player agent’s need to think critically as a narrator agent or to invest emotionally as the character agent.

Future Considerations

Though initially conceived to act as the amalgam of all the other areas of assessment, in the future it would be beneficial to consider exactly how the other areas of assessment lead to a cohesive narrative experience and potential story products. Through the analyses conducted here, it became apparent quite early that the criteria in one area of assessment were linked to and complicated by criteria in other areas of assessment. While it is perhaps easy to analyze each criterion separately or even to analyze the criteria for each area of assessment as relevant only to that area of assessment, the work of understanding the interplay between the criteria is much harder. Moving forward, it is that work that should become the focus.

225

Ludic Agency

Results

Table 15. Final Analysis of Ludic Agency Kentucky Route The Banner Mass Effect 2 Skyrim Zero Saga Ludic Agency Absent (0), Low (1), Medium (2), High (3) A small number of choices Absent High High Medium don't force plot Obvious choices don't force Absent High High High plot

Dialogue is emotionally- High Absent High High driven

No dialogue system at all Absent Absent Absent Absent

Embodied interactions with Low High Low Absent environment Embodied interactions with Low High Medium Low objects

Embodied interactions with Low Low Medium Low subject

.28 .61 .66 .47

Once again, Kentucky Route Zero earned the highest score based on the established criteria for assessment. In this case, the results are perhaps initially surprising, especially in comparison to a game like Skyrim, which was purported to offer the player extensive environmental freedom in a never-ending procedurally-generated story and gameworld.

Additionally, ludic elements are often made synonymous with the mechanics that enable the player agent to interact with and in the gameworld, and Kentucky Route Zero offers very few mechanics for the player to use. Furthermore, the criteria for ludic agency suggest that dialogue

226

systems should not be used, and in Kentucky Route Zero the dialogue system is the primary means of interacting within the game’s narrative structure. So how, then, does Kentucky Route

Zero come out on top?

First, we can compare the game in terms of the dialogue each offers. In Mass Effect 2,

The Banner Saga, and Kentucky Route Zero the dialogue provided in the game – either textually or audibly – is emotionally focused. In Mass Effect 2, the dialogue reveals the complex emotions involved in trying to find camaraderie amongst a group of very different people, who have collectively been assigned the mission to save the universe. In The Banner Saga, the dialogue is emotionally charged with a sense of despair – the world is ending, and there is no real way to stop it – but also a sense of duty – until the world ends, all that remains to do is protect our humanity. In Kentucky Route Zero, the characters are all, in some way, lost, and the dialogue serves as a means to narrate how they came to be so and how the characters might come to find themselves again. In Skyrim, however, the dialogue lacks emotional complexity and cohesion.

Likely, this is because the game lacks a coherent narrative. Because the player agent can go where she likes and complete missions in any order she likes, the narrative structure of the game cannot maintain an emotional arch that ties the dialogue of characters to the story the player agent develops.

Skyrim also provides the highest rates of embodied interactions; however, without a sense of purpose or emotional impetus, these embodied interactions remain hollow gestures. Yes, the player agent can move freely around the gameworld at her discretion just as she can touch, move, and consume objects, but the gameworld responds little to the player’s ability to do so in meaningful ways. The game allows the player to do anything, but in doing so, does not

227

acknowledge certain contradictions that would otherwise make the player agent’s embodied interactions more meaningful. Thus, while the other games score lower in terms of what embodied interactions they permit, where they do permit embodied actions, those actions carry a stronger sense of meaning.

Interestingly, however, one could argue that Skyrim’s lack of unifying emotional and narrative arch is what enables the game to score as well as it does in terms of choices that do not force the plot. Because Skyrim lacks a rigid narrative structure and does not, for that matter, ever require the player to complete the main quest line, there is little occasion for the game to force the plot (such that there is one) in any direction. Comparatively, Mass Effect 2’s unifying and emotional narrative arch is perhaps too focused, as the game frequently forces the plot in a given direction by offering the player both extremely limited and obvious binary choices. Thus, in

Chapter 5, the analysis of these two games side-by-side might lead to the assumption that ludic agency negates narrative agency and vice versa, but the additional analysis of Kentucky Route

Zero and The Banner Saga show that this is not the case. Both Kentucky Route Zero and The

Banner Saga have emotionally focused dialogue and branching narrative structure, and yet neither of these games forces the plot by offering too few or too obvious choices. While the choices in Mass Effect 2 are clearly mapped to the game’s Paragon/Renegade binary, neither

Kentucky Route Zero nor The Banner Saga uses such a dominant morality system. Instead of having each choice speak to a spectrum of morality, the choices in The Banner Saga and

Kentucky Route Zero speak for themselves. The ramifications of these choices are also more complex than morality, especially in The Banner Saga where any given choice affects not one, but several elements of the player agent’s caravans.

228

Future Considerations

Each of the four games employs a dialogue selection menu as an important mechanic.

Furthermore, the dialogue selection menus are the central mechanic used to link the player agent to the narrative structure. The input provided by the player agent when selecting a piece of dialogue is also, in all the games but Skyrim, visibly linked to the drama management system. In other words, dialogue selection menus are a means by which the drama management system receives input and adapts to the player’s experience. In the future, therefore, it is worth considering drama management systems in relation to ludic agency and vice versa. As the criteria for ludic agency note, dialogue should align with the concerns of the story being developed within the narrative structure as well as with the preeminent emotions of that developing story. Such an understanding of ludic agency could be used to assess the design and presentation of dialogue selection menus as well as the dialogue presented within them. In other words, it is too easy to dismiss dialogue as a limit to ludic agency, but dialogue may restrict ludic agency by degrees based on these more complex attributes. That said, in the future, it is also worth considering how dialogue selection menus disrupt embodied interactions or how dialogue may be presented to the player in such a way that the focus of play and interaction remain on the embodiment of character. Finally, given the link between dialogue selection menus and the drama management system, future consideration should also assess how embodied interactions are, or could be better, linked to drama management systems.

229

Narrative Agency

Results

Table 16. Final Analysis of Narrative Agency Kentucky Route The Banner Mass Effect 2 Skyrim Zero Saga Narrative Agency Absent (0), Low (1), Medium (2), High (3)

User has active role High High High High

User influences the events High Medium High High that occur User influences how the Low Absent Medium Low story ends User is not playing a role Medium High High Medium

User needs to reflect Low Absent Medium High

User can bring about Absent Medium Medium Medium narrative alternatives

.55 .55 .83 .77

Again, the data shows that Kentucky Route Zero meets the criteria at a higher rate of

success than the other three games. In this case, however, the ratings across the games are closer

than in any other area of assessment. Provided that each of the games was designed intentionally

to provide the player agent with a narrative experience, it is perhaps not surprising to see a

smaller number of difference between the ratings for the criteria related to narrative agency.

In each of the four games, the player agent has an active role, though the type of role the

player agent most frequently or strongly inhabits differs from game to game. For instance, one

might argue that the player is more frequently an active character agent in Mass Effect 2, while

the player agent is more frequently an active narrator agent in Kentucky Route Zero. While both

230

games permit the user to inhabit both character agent and narrator agent roles, the games differ by the degree to which the player agent maintains these roles actively.

The player agent’s input also influences the events that occur in each of the four games.

That said, Skyrim earned a slightly lower score than the other three games. In Mass Effect 2,

Kentucky Route Zero, and The Banner Saga, the narrative structure of the game is rigid enough to maintain logic and cohesion within the story being developed. The same cannot be said of

Skyrim. While the gameworld responds to the player’s choices, the narrative structure does not respond to the events in as meaningful a way.

Ultimately, each of the four games scored poorly in terms of the player agent’s ability to influence the outcome of the game’s story product. Due to the game’s use of procedural generation of content, Skyrim fails to provide a satisfying story conclusion. That said, the world is meant to be unending and limitless, and so the game’s failure to provide a conclusion is to be expected. Comparatively, in terms of the environment and the event that takes place in it, the conclusion of Kentucky Route Zero will remain the same regardless of the player’s input throughout gameplay. In this respect, the player agent has no capacity to influence the outcome.

Yet, the player does have enough flexibility in terms of how she develops the character agents within the game’s narrative structure that the understanding of the outcome will be different in terms of its narration and in terms of the player’s perception of the characters.

Most interesting, however, is the similarity between Mass Effect 2 and The Banner Saga.

Both games earned a low score here and subsequently both also earned medium scores for the player agent’s capacity to connect with the character agents and to make decisions on the level of character. The similar scores between the two games would suggest a strong correlation between

231

these two criteria. In both games, the player agent is presented with objective information that removes the player agent from the role of character agent too severely. In Mass Effect 2 the morality system fragments the balance between the roles of narrator and character, while in The

Banner Saga the omnipresent HUD and the character stat screen remind the player of the objective utility of characters in terms of maintaining the caravan’s stats. Thus, one might conclude that the need to balance the player agent’s ability to shift seamlessly between the role of narrator and character is essential and that the interfaces provided to the player agent should provide objective information without losing emotional or dramatic resonance.

Future Considerations

It would be beneficial to think more critically about the player agent’s ability to bring about narrative alternatives. For instance, in every game but Skyrim, there is limited variability in the final outcome of the story product. In other words, the games provide few alternatives for how the developed story is ultimately finalized. Thus, each of the games provide a higher rate of variability in terms of the narrative beats the player agent encounters during gameplay than in the conclusion of gameplay. The question then becomes whether or not the choices present throughout gameplay amount to sufficient alternatives if the story product concludes in only limited ways despite those choices. In other words, do the alternatives exist in the number of choices presented to the player or in the number of outcomes? Either option seems to be a reductive way to conceive of the player agent’s narrative experience as a whole, but the questions still warrant consideration in terms of how we continue to push our understanding of narrative agency.

232

Character Development

Results

Table 17. Final Analysis of Character Development Kentucky Route The Banner Mass Effect 2 Skyrim Zero Saga Character Development Absent (0), Low (1), Medium (2), High (3) Player agent cares about High Absent High Low characters Player agent respects High Absent High Low characters Characters have strong High Absent High Medium personalities Characters have strong High Absent High High emotions Player agent can infer High Absent Medium Low character’s thoughts

Interactions are about High Absent High High emotional entanglement

Relations are not props for High Absent High Medium action

1.0 0 .95 .71

As the data shows, the sheer expanse of the gameworld in Skyrim once again limits its

capacity to focus. In this case, the game fails to provide rich characters with complex emotions

or qualities the player agent can relate to, respect, or make inferences from. As a result, the NPCs

in Skyrim are little more than hollow props used to propel the character forward through self-

selected missions. While The Banner Saga met the criteria more completely than Skyrim, The

Banner Saga placed more emphasis on the care of the caravan than on the care for particular

characters, and as such, it is hard for the player agent to care about the characters or respect them

233

as individuals. Additionally, the interactions in The Banner Saga are less about how individuals may respond and more about how the caravan will be affected. Thus, while the interactions are about emotional entanglement, those entanglements are presented as ethical quandaries that will change the state of the caravan, rather than as entanglements that help the player come to know the characters better.

Comparatively, Mass Effect 2 earned the highest rating for character development. In this one area of assessment, the morality system of the game acts as a benefit. Because the interactions in the game are meant to lead the player to make choices based on an understanding of the game’s morals, those interactions are necessarily charged with rich emotions. The interactions with characters are meant to carry the weight of morally hard choices, and often the game complicates those choices for the player by using NPCs to express conflicting opinions and judgments. Some choices will lead certain characters to like the player agent more while others will like the player agent less. In this way, the player agent gets a sense of each character’s individuality. Ultimately, while the character interactions are rich and lead to emotional respect and entanglement with the characters, it is the omnipresence of the drama management system that decreases the weight of complexity the character interactions are meant to create. In Mass

Effect 2, it would be inaccurate to say the characters are props for action, but it would be accurate to say they are props for morality.

Kentucky Route Zero also earned a nearly perfect score in terms of accommodating the criteria for character development. Yet, in the case of Kentucky Route Zero, the characters are not made complex in order to increase the drama of the player’s choices or the subsequent morality points associated with those choices. Instead, the characters are rich in Kentucky Route

234

Zero because the primary system with which the player interacts is focused on the advancement of characters. In the case of Kentucky Route Zero, the trajectory of the story product may be somewhat rigid, but the capacity to alter characters through the act of narration is flexible. That said, in Kentucky Route Zero the player agent must always develop one character at the expense of another. By selecting dialogue for one character rather than another, the player agent’s understanding of each is altered. Therefore, the level of character development per character is dependent on the player. In Mass Effect 2, the player also has the ability to develop or not develop a relationship with certain characters, but the game does not necessitate the player to choose to engage with one character over another. In Kentucky Route Zero, however, the game’s structure necessitates a choice between the development of characters.

Future Considerations

Future considerations about character development should consider how a game develops character within its particular gameworld and with reference to the plot the narrative structure presents to the player agent. As the analyses of the four games in this text reveal, characters do not exist independently of a gameworld but are, instead, an extension of it. Characters add context to the environment as well as to the narrative structure. In this way, characters, who provide one means for presenting embodied interactions that connect the player agent to both the narrative and ludic elements of the game, act as a bridge between ludic and narrative agency.

Without these opportunities for meaningful and emotionally resonant interactions with characters, the player agent would likely lose a sense of attachment to the gameworld or the consequences and meaning of the story being developed. Thus, the criteria for character

235

development should be expanded to include the ways in which the characters support the logic of the gameworld in relation to the plot. If the characters have complexity and dimension within the context of the world they inhabit, the player’s interactions with them will likely carry more emotional weight, and those interactions will be less likely to act as props used only to move the plot further.

Conclusion

The criteria used in the analysis of Mass Effect 2, Skyrim, Kentucky Route Zero, and The

Banner Saga were initially established as a means to find points of agreement and commonality from which to assess the quality of narrative and agency present in digital interactive narrative games. These criteria were formed by considering the ideas and theories presented by narratologists and ludologists working in Game Studies, but also included considerations from broader theories of narrative and social agency.

The criteria established in Chapter 5 have provided lucrative insights about how agency and narrative function in Mass Effect 2, Skyrim, Kentucky Route Zero, and The Banner Saga.

That said, each analysis also provided useful insight about how the criteria might be altered to support richer and deeper analyses. There are certainly ways for the criteria to be refined. As this chapter has provided potential next steps or considerations with respect to each of the key areas of assessment.

Future considerations for drama management systems should include the examination of how they permit player agents to act on the level of narrator agent without fragmenting the immersion players maintain as the character agent, and vice versa. Future considerations for the

236

ludic agency should the assessment of dialogue selection menus and how they may or may not disrupt embodied interactions. Additionally, it would be lucrative to consider how dialogue may be presented to the player in such a way that the focus of play and interaction remain on the embodiment of the character. With respect to narrative agency, future considerations should assess whether or not the choices presented throughout gameplay amount to significantly different narrative conclusions. If the game limits the number of conclusions, the criteria should be able to assess whether the structure offered enough variety to the player throughout gameplay to create different story products despite similar story conclusions. Considerations for character development should be expanded to include the ways in which the characters support the logic of the gameworld in relation to the plot. Finally, in the future, considerations for the overall narrative experience should be conducted last and should specify how the interplay of the other areas of assessment support the narrative experience.

That said, the continued use of the criteria to analyze digital interactive narratives will, and should, help refine the criteria further. Together, the refinement of the criteria and their subsequent application to the assessment of other digital interactive narrative games should provide a consistent feedback loop that leads to better and more useful understandings of narrative and agency in digital interactive narrative games.

For the criteria to evolve, however, collaboration will be needed, too. The formation of these criteria began with the spirit of collaboration, and it is from this spirit that they should grow. The observations of ludologists and narratologists both have value and merit and, when placed together, can create useful means by which to understand games and their capacity to tell stories and engage players.

237

CHAPTER 10

FINAL THOUGHTS AND FUTURE CONSIDERATIONS

It is fair to say that digital games are distinct and different enough from other media to necessitate their individualization into a unique area of study. Though games employ and use aspects and techniques found in other media such as film, theater, narrative, sound, animation, and so on, there will always remain, despite the use of these things, a difference in how they similar aspects and techniques are employed and how they are experienced. As such, it is these points of difference which have been the focus of conversation within the burgeoning field of

Game Studies.

Yet, while it is easy to say that games are a unique medium worthy of study as such, the formation of games into a distinct mode and area of study is slow. Where the formation and articulation of Game Studies has faltered, it has done so by moving too quickly to disassociate the study of games from other media.

Projecting forward into the future, one can imagine a time when Game Studies is understood without the need for constant reference to the media and the studies of those media that have preceded it. Given the newness of the field, however, that time is not yet here, and we do the study of games a disservice when we act as if it is. So, while it is true that the study of games deserves distinction, and while the field will eventually advance its theories and practices to the point that the relevance of their study is just as distinct and apparent, the work of game scholars is still, for better or worse, enmeshed in acts of comparison. As such, efforts to differentiate and solidify game studies too quickly from other types of media come at the

238

expense of understanding exactly what makes games different in the first place. At this stage, game scholars must grapple with sameness in order to understand the differences more clearly.

The ludology versus narratology debate stands as the prime examples of what can happen when we attempt, too quickly, to demarcate digital games as a medium independent of and distinct from other media. The greatest conceit of the ludology versus narratology debate is, of course, that the ludologists and the narratologists were actually making the same arguments. Yet, without a shared understanding established between them and without an appreciation for the contexts each side employed to generate their understandings, the debate continues tediously and to the detriment of progress.

Without intending to, Gonzola Frasca succinctly defines the real issue at the heart of the ludology versus narratology debate in his essay “What is Ludology? A Provisory Definition.” In it, he writes that “one of the reasons most researchers try to use narratology to explain videogames is that there is a lack of a formal discipline that focuses on games...Thus, in order to understand videogame, we first need to understand games.” Later in the essay, he goes on to state, “The current state of videogame research is mainly driven by scholars who try to explain computer games through previously existing media. While I do not necessarily discard these approaches, I think that they are incomplete and that by studying videogames as something else than games, they are denying its main potential” (Frasca “What is Ludology”) Frasca, thus, acknowledges the work of non-ludologists while prioritizing the research of games above it.

Unfortunately, in prioritizing the focus of the field entirely to games scholarship and theory – an area he readily admits is underdeveloped – he diminishes the importance of other media and areas of research. Additionally, this approach to the study of games also limits the potential for

239

the study of games to grow naturally into its own domain by extending outward from previously established methods and areas. In ignoring such overlaps, significant opportunities are and have been lost.

The narratologists are, of course, not without some blame either. As this text has shown, digital games can and do maintain narrative structures and do possess narrativity. In this way, the study of narrative in relation to digital games is reasonable and valid, much as the study of performance theory, sound theory, and film theory are valid means of interrogating how digital games are constructed and experienced. That said, shoehorning the medium of digital games into any one of these areas of research still ignores the ways in which the medium of games functions differently than theatre, film, sound, or narrative. Digital games cannot be equated to any other medium, and any effort to understand them exclusively in reference to another medium will also always be limited.

The study of games as isolated to and from other media is limited, but so too is the study of games solely within the context of another type of media studies. Ultimately, if Game Studies is to progress and validate itself, it must do so by understanding games in relation to other media and by using this understanding as a means to establish what makes games unique and how games function uniquely as media.

In order to progress, then, Game Studies should embrace the practice and philosophy of wayfinding. Wayfinding, or a means of navigating without precise directions or a clear path forward, relies heavily on a method called orientation. The act of orientation, by extension, depends upon understanding one’s position in relation to where one is, how one got there, and where one wants to go. In other words, the act of orientation takes stock of an ultimate goal or

240

destination by first being aware of current circumstance and how one arrived at those current circumstances. Wayfinding as a practice, then, is both reflective and progressive simultaneously.

This approach would enable Game Studies scholars – whether ludologists, narratologists, or otherwise – to mutually work toward the goal of distinguishing games, and digital games most specifically, as a unique medium. This approach, however, enables scholars to work toward this goal without ignoring the ways in which the uniqueness of games is most clearly understood within the context of those media games wish to distinguish themselves from. In other words, we can embrace sameness while seeking difference. We can orient games as a media both like and unlike other media and, via this orientation, move the study of games forward into new and more distinguished territory.

If we proceed in the spirit of wayfinding, games scholars will not only lead the study of games forward as a unique medium, but will do so in ways that appreciate the history, theories, and conversations from which is has grown. In the effort to propel Game Studies forward too quickly, not only has the field unnecessarily segmented itself, but it has also fragmented the vocabulary and the definitions that would otherwise establish a shared understanding from which to progress.

This text has, therefore, attempted to reconnect the study of games to theories of narrative and theories of agency in order to establish clear definitions and understandings for both and, ultimately, to provide a means forward. As such, ideas presented in this text are the result of a proper orientation between the fields of study that predate the focused study of games, an understanding of the current issues at stake in the study of games, and an outlook toward where the study of games intends to move. This text, therefore, acknowledges how games can be

241

understood within the context of other media theories while also remaining a distinct media form.

In Chapter 2, I suggest that when narrative is understood as a structure and mechanism for the delivery of a story, rather than the story product itself, than narrative theory becomes easier to understand and account for in relation to digital interactive narrative games. I argue that those elements which have been outlined by narratologists to be essential to the construction of a narrative structure do remain intact within digital interactive narrative games. Still, with the introduction of interactivity, some elements are presented differently in digital interactive narrative games than they are in other textual or visual narrative media. In Chapter 2, I suggest two ways this happens. First, I introduce the theory that the narrator agent and character agent found in other narrative media are both roles the player agent mediates. Secondly, I argue that while most narrative forms communicate narrative in a primarily temporal way, digital interactive narrative games place the emphasis on the spatial delivery of narrative. Both of these arguments advance the discussion of narrative in relation to digital interactive narrative games while maintaining the integrity and the elements of narrative structure as a concept and theory.

These theories present new ways of understanding digital interactive narratives as a form of media that, despite being narrative, engages the form in unique ways that are distinct and dissimilar to other media.

In Chapter 3, I consider the ways in which much of Game Studies scholarship has vastly oversimplified the concept of agency by equating it with the capacity to act. In some ways, this is an understandable oversimplification. After all, players interact with a set of rules that permit them to either take or not take a particular action. Still, it is no less true that such an

242

understanding of agency is reductive in terms of the agency players bring with them to games or how that agency might be supported, challenged, or mediated by a game’s system and within a game’s storyworld. Of course, it is also true that some Game Studies scholars have attempted to move the discussion of agency beyond this simplified reduction. That said, most of the advancements have happened within the vacuum of Game Studies scholarship. By ignoring the vast amount of scholarship related to social agency and how it is constructed and mediated in complex social systems, Game Studies has missed the opportunity to consider how that scholarship might apply to the consideration of agency in games. Simply put, the theory of agency in games could be advanced significantly by considering the ideas expressed within the theories of social agency. By way of example, Chapter 3 analyzes several theories of agency from only a few of the many schools of thought related to social agency theory. Even in this cursory overview, no less than ten ways of rethinking agency in relation to digital games emerged. Of the ten, this text attempts to give continued treatment to two of these developments: first, the suggestion that agency isn’t of one type, but rather of many, and second that agency is an act of making meaning through critical thought processes, which may or may not lead to action. The rest remain for consideration in the future.

In addition to reconnecting game-specific understandings of narrative and agency back to theories of narrative and agency that exist outside the context of games, this text also provides a set of criteria that allow one to assess the presence and quality of a narrative structure and the capacity for agency provided to a player agent in digital interactive narrative games. Thus, taking the concepts generated in Chapters 2 and 3 into consideration, Chapter 5 provides a set of criteria which can be used to assess the intricate ways in which a game system attempts to provide the

243

player with a well-formed narrative, while also establishing ways for the player to intervene and participate in the creation of a story product by providing various types of agency and methods of meaning-making to the player. Taken together, Chapter 5 establishes ten criteria organized into five areas of assessment: drama management interactions, narrative experience, ludic agency, narrative agency, and character development and interaction. In addition to incorporating the considerations for narrative and agency identified in Chapters 2 and 3 in more nuanced ways, the criteria are also meant to enable one to determine how, and to what level of success, any game supports or includes these elements. Given this, the criteria in Chapter 5 provides a new means of analysis, and a methodology for that analysis, for digital interactive narrative games.

The criteria established in Chapter 5 should, however, suggest a new starting point for the study of interactive narrative games rather than an endpoint. The application of the criteria to the assessment of Mass Effect 2, Skyrim, Kentucky Route Zero, and The Banner Saga show the utility and validity of the criteria as they are, but this does not mean the criteria should not be refined or extended as new theories and scholarship emerge. In fact, they can and should be. That said, this text should exemplify the growth that can come, first, from clarifying points of disagreement and, second, establishing the points of agreement that are often obscured by such disagreement. The hope, therefore, is that the continued growth of the criteria will come from collaborations between narratologists and ludologists alike.

Going forward, there are several logical next steps for developing this research and the criteria for assessment further. The most obvious of these would be to consider other ways in which agency might function in games and whether those types of agency are presented in digital interactive narrative games. For instance, this text establishes the player agent as someone who

244

inhabits both a narrator agent and character agent role in digital interactive narrative games.

Future research might consider how this idea is complicated when the player is both the subject

(acting) and object (being acted upon) of the game’s system of rules. Interrogation of this theory may or may not align with the concept of actions and happenings in narrative theory, but if it does, the implications may reveal yet new ways for narrative and agency to coexist in digital interactive narrative games. Alternately, one might consider the degree to which the agencies of the player and the game change during and throughout gameplay. Extending from this, one might consider whether the player agent's level of agency is more powerful when functioning as the narrator agent or the character agent and whether the reverse is true of the game’s level of agency. Of course, these are just two suggestions for how we might continue to think about agency in relation to interactive narrative games. There are certainly others, and each would be lucrative.

Additionally, as the study of narrative and agency continues, it would be worthwhile to re-think about the criteria of assessment. While some criteria may change or new criteria may be added, it will also be worthwhile to consider whether the five areas of assessment are worth maintaining or if there are better approaches and methodologies for assessing them. As Chapters

6-8 show, the criteria have utility but, often, the criteria from one area of assessment informed or overlapped with considerations of another. While this was expected, approaching the criteria from one area to the next can lead to a fragmented understanding of the game. It is possible then that future use of this methodology might restructure how the criteria are grouped based on what the most frequent overlaps or connections are between the criteria in games. For instance, it is likely that the presentation of the drama management system to the player agent will have an

245

effect on the narrative experience. That said, the drama management system may also affect the level of ludic agency the player experiences. Alternately, future use of the criteria might maintain the areas of assessment in order to focus one’s initial analysis, but might then forego the rigidity that delineates the criteria into areas of assessment when reporting the findings. In this case, the report of one's findings could occur in whatever order most makes sense based on which criteria informed the others and where they overlapped. For the purpose of clarity, and for the purpose of testing the utility of the criteria and the areas of assessment, this text has maintained the same methodology for the analysis of each game. That said, in the future, this approach does not necessarily need to be maintained.

Of course, it is also important to note that, while this text has worked exclusively with digital interactive narrative games and exclusively with issues regarding the presumed contention between narrative and agency, there are certainly other important considerations that lie outside these parameters. For example, similarly comparative analyses to those conducted in this text but with respect to performance theory and the concepts of participation, both actor and audience, would also be lucrative. Furthermore, consideration for the constructions of space within film theory would also provide useful insight. Indeed, such studies would like only deepen the discussion in relation to narrative and agency while also extending understandings of games further too. In other words, by embracing the numerous overlaps between games and other media and media theory, we increase, rather than decrease, the means by which to understand games as a distinct and worthy medium worthy of scholarship.

Regardless of the path forward and how the research extends from here, we must be careful not to repeat the mistakes born from the ludology versus narratology debate. As this text

246

has aimed to illustrate, the language and the semantics guiding the discourse around digital games matters immensely. We must be careful with our words and with our meanings. The redefining of terms is a natural linguistic process, but it should also be a slow process guided by revelation. It should not, by contrast, be a fast or individualized process by which scholars define a word in order to make it work for their unique purposes, only for another scholar to take it up and do the same. Nor should it be a process that seeks to distinguish games as unique without first doing the necessary work of articulating how such differences are understood in the context and study of other media. In other words, we cannot distinguish games without direct engagement with those forms we wish to distinguish it from. Instead, we must wayfind. We must be careful with our words and cognizant of the weight they carry and the contexts, theories, and histories on which they rely. By mutually building the language we use to analyze and assess games, we also pave the way for progress. As this text shows, agreements exist in seemingly disparate arguments once we take the time to reflect, contextualize, and orient ourselves. From this, great potential springs.

247

BIBLIOGRAPHY

As several chapters in this text were previously published, and because those chapters have used more than one citation style, the bibliography has been organized by chapter. Subsequently the bibliographic citations use the same style as the citations in each chapter.

Introduction

Joyce, Lindsey. “Space, Time, and Agency: Fundamentals of Interactive Narrative Systems.” n. pag. Google Scholar. Web. 25 Jan. 2017.

Murray, Janet. “The Last Word on Ludology v Narratology (2005).” Janet H. Murray: Humanistic Design for an Emerging Medium. N.p., n.d. Web. 24 Aug. 2015.

Shaw, Bernard, and Archibald Henderson. “A Dialogue on Things in General.” Harper’s Magazine May 1924. Harpers. Web. 20 Jan. 2017.

Chapter 2

Abbott, H. Porter. The Cambridge Introduction to Narrative. Second. Cambridge: Cambridge University Press, 2008. Print.

Aarseth, Espen. “Genre Trouble: Narrativism and the Art of Simulation.” First Person: New Media as Story, Performance, and Game. Ed. Noah Wardrip-Fruin and Pat Harrigan. Cambridge, MA: The MIT Press, 2004. 45–55. Print.

Bal, Mieke. Narratology: Introduction to the Theory of Narrative. Toronto: University of Toronto Press, 1985. Print.

Barthes, Roland. Image, Music, Text. Trans. Stephen Heath. New York: Hill and Wang, 1977. Print.

Bioware. Mass Effect 2. N.p., 2010. Xbox 360.

Buchholz, Sabine, and Manfred Jahn. “Space in Narrative.” Routledge Encyclopedia of Narrative Theory. Ed. David Herman, Manfred Jahn, and Marie-Laure Ryan. London: Routledge, 2005. 504–544. Print.

Cardboard Computer. Kentucky Route Zero. N.p., 2013. Microsoft Windows.

Chatman, Seymour. Story and Discourse: Narrative Structures in Fiction and Film. Ithaca, NY: Cornell University Press, 1978. Print.

248

Coetzee, J.M. Disgrace. New York: Penguin Books, 1999. Print.

Cooper, Mark Garrett. “Narrative Spaces.” Screen 43.2 (2002): 139–157. Print.

Costikyan, Greg. “Games, Storytelling, and Breaking the String | Electronic Book Review.” Electronic Book Review. N.p., 28 Dec. 2007. Web. 9 Nov. 2014.

Dutton, Denis. The Art Instinct: Beauty, Pleasure and Human Evolution. New York: Bloomsbury Press, 2009. Print.

Enright, Anne. The Gathering. New York: Grove Press, 2007. Print.

Fish, Stanley. Is There a Text in This Class: The Authority of Interpretive Communities. Cambridge, MA: Harvard University Press, 1980. Print.

Foucault, Michel. Aesthetics, Method, and Epistemology. Ed. James Faubion. New York: The New Press, 1998. Print.

Galloway, Alexander. Gaming: Essays on Algorithmic Culture. Minneapolis: University of Minnesota Press, 2006. Print.

Genette, Gérard. “Frontiers of Narrative.” The Critical Tradition: Classic Texts and Contemporary Trends. Ed. David Richter. New York: Bedford St. Martin’s Press, 1989. 890– 899. Print.

---. Narrative Discourse: An Essay in Method. Ithaca, NY: Cornell University Press, 1983. Print.

Harrell, D. Fox, and Jichen Zhu. “Agency Play: Dimensions of Agency for Interactive Narrative Design.” AAAI Spring Symposium: Intelligent Narrative Technologies II. N.p., 2009. 44–52. Google Scholar. Web. 19 Nov. 2014.

Heath, Stephen. “Narrative Space.” Screen 17.3 (1976): 68–112. Print.

Herman, David et al. Narrative Theory: Core Concept and Critical Debates. Columbus, OH: The Ohio State University Press, 2012. Print.

Jenkins, Henry. “Game Designs as Narrative Architecture.” First Person: New Media as Story, Performance, and Game. Ed. Noah Wardrip-Fruin and Pat Harrigan. Cambridge, MA: The MIT Press, 2004. Print.

Juul, Jesper. Half-Real: Video Games between Real Rules and Fictional Worlds. Cambridge, MA: The MIT Press, 2005. Print.

Labov, William, and Joshua Waletzky. “Narrative Analysis.” N.p., n.d. Web. 13 Dec. 2016.

249

Lacey, Nick. Narrative and Genre: Key Concepts in Media Studies. Houndmills: Macmillan Press, 2000. Print.

Magerko, Brian. “Evaluating Preemptive Story Direction in the Interactive Drama Architecture.” Journal of Game Development 2.3 (2007): 25–52. Print.

McCarthy, Cormac. The Road. New York: Vintage Books, 2006. Print.

McGonigal, Jane. Reality Broken. N.p. Print.

Murray, Janet. “From Game-Story to Cyberdrama.” First person: New media as story, performance, and game 1 (2004): 2–11. Print.

---. Hamlet on the Holodeck: The Future of Narrative in Cyberspace. New York: The Free Press, 1997. Print.

Phelan, James, and Peter Rabinowitz. “Authors, Narrators, Narration.” Narrative Theory: Core Concepts and Critical Debates. Ed. David Herman. Columbus, OH: The Ohio State University Press, 2012. 29–38. Print.

Ryan, Marie-Laure. “Narration in Various Media | the Living Handbook of Narratology.” N.p., n.d. Web. 10 Mar. 2017.

---. “Introduction.” Narrative Theory: Core Concepts and Critical Debates. Ed. David Herman. Columbus, OH: The Ohio State University Press, 2012. 3–9. Print.

Ricoeur, Paul. “Narrative Time.” Critical inquiry 7.1 (1980): 169–190. Print.

Riedl, Mark O. “A Comparison of Interactive Narrative System Approaches Using Human Improvisational Actors.” Proceedings of the Intelligent Narrative Technologies III Workshop. ACM, 2010. 16. Google Scholar. Web. 8 Nov. 2014.

Riedl, Mark O., and Carlos Leon. “Toward Vignette-Based Story Generation for Drama Management Systems.” Proceedings of the 2nd International Conference on Intelligent Technologies for Interactive Entertainment (INTETAIN), Workshop on Integrating Technologies for Interactive Stories. N.p., 2008. Google Scholar. Web. 26 Nov. 2014.

Rose, Frank. The Arts of Immersion. New York: W.W. Norton & Company Inc., 2011. Print.

Ryan, Marie-Laure. “Story/Worlds/Media: Tuning the Instruments of a Media-Conscious Narratology.” Storyworlds Across Media: Toward a Media-Conscious Narratology. Ed. Jan-Noel Thon and Marie-Laure Ryan. Lincoln: University of Nebraska Press, 2014. 25–49. Print.

250

Salen, Katie, and Eric Zimmerman. Rules of Play: Game Design Fundamentals. Cambridge, MA: The MIT Press, 2004. Print.

Stoic. The Banner Saga. N.p., 2014. Microsoft Windows.

The Beatles. Rocky Raccoon. Apple Records, 1968. Audio Recording.

Tisselli, Eugenio. “Narrative Motors.” Ed. Philippe Bootz and Sandy Baldwin. Morgantown: West Virginia University Press, 2010. Print.

Wolf, Werner. “Pictorial Narrativity.” The Routledge Encyclopedia of Narrative Theory. Ed. David Herman, Manfred Jahn, and Marie-Laure Ryan. London: Routledge, 2005. 422–426. Print.

Chapter 3

Badcock, C.R. Levi-Strauss (RLE Social Theory): Structuralism and Sociological Theory. Taylor and Francis, 2014. Print.

Bandura, Albert. “Toward a Psychology of Human Agency.” Perspectives on psychological science 1.2 (2006): 164–180. Print.

Baron, Sean. “Cognitive Flow: The Psychology of Great Game Design.” N.p., n.d. Web. 20 Jan. 2017.

Descartes, René. Descartes: Key Philosophical Writings. Ed. Enrique Chávez-Arvizo. Ware, Hertfordshire: Wordsworth Editions Limited, 1997. Print.

Frasca, Gonzola. “Rethinking Agency and Immersion: Video Games as a Means of Consciousness-Raising.” Digital Creativity 12.3 (2001): n. pag. Print.

Haraway, Donna. Simians, Cyborgs, and Women: The Reinvention of Nature. New York: Routledge, 1991. Print.

Hayles, N. Katherine. “Deeper into the Machine: The Future of Electronic Literature.” Culture Machine 5 (2008): n. pag. Google Scholar. Web. 11 Oct. 2016.

---. How We Became Posthuman: Virtual Bodies in Cybernetics, Literature, and Informatics. Chicago: University of Chicago Press, 1999. Print.

Hegel, Georg Wilhelm Friedrich. Preface and Introduction to the Phenomenology of the Mind. Ed. Lawrence Stepelvich. New York: Macmillan Publishing Company, 1990. Print.

Heidegger, Martin. Being and Time. New York: Harper, 1962. Print.

251

---. The Question Concerning Technology. New York: Garland Publishing Inc., 1997. Print.

Hume, David. Dialogues and Natural History of Religion. Ed. J.C.A. Gaskin. Oxford: Oxford University Press, 1993. Print.

Latour, Bruno. Reassembling the Social: An Introduction to Actor-Network-Theory. Oxford: Oxford University Press, 2005. Print.

Miyamoto, Shigeru. Super Mario Bros. Nintendo, 1985. NES.

Murray, Janet. Hamlet on the Holodeck: The Future of Narrative in Cyberspace. New York: The Free Press, 1997. Print.

Pajitnov, Alexey, and Vladimir Pokhilko. Tetris. Nintendo, 1984. NES.

Parsons, Talcott. Toward a General Theory of Action. Harper and Row, 1962. Print.

Salen, Katie, and Eric Zimmerman. Rules of Play: Game Design Fundamentals. Cambridge, MA: The MIT Press, 2004. Print.

Tanenbaum, Karen, and Joshua Tanenbaum. “Commitment to Meaning: A Reframing of Agency in Games.” Digital Arts and Culture 2009 (2009): n. pag. Google Scholar. Web. 9 Nov. 2014.

Author’s Note

Joyce, Lindsey. “Assessing Mass Effect 2 and Elder Scrolls V: Skyrim.” Journal of Games Criticism. N.p., n.d. Web. 18 Dec. 2016.

---. “Creating Collaborative Criteria for Agency in Interactive Narrative Game Analysis.” The Computer Games Journal (2015): n. pag. CrossRef. Web. 7 July 2015.

---. “Kentucky Route Zero: Or, How Not to Get Lost in the Branching Narrative System.” The Play Versus Story Divide in Game Studies: Critical Essays. Jefferson: McFarland, 2016. 17–27. Print.

Chapter 4

Bethesda Game Studios. The Elder Scrolls V: Skyrim. N.p., 2011. Xbox 360.

Bioware. Mass Effect 2. N.p., 2010. Xbox 360.

Cardboard Computer. Kentucky Route Zero. N.p., 2013. Microsoft Windows.

252

Ligman, Kris. “Gamasutra - Opinion: Let’s Play Crackdown Is an Attack on Game Culture.” N.p., n.d. Web. 3 Feb. 2015.

Stoic. The Banner Saga. N.p., 2014. Microsoft Windows.

Chapter 5

Archer, M. (2001). Being Human: The Problem of Agency. Cambridge: Cambridge University Press.

Bethesda Game Studios. (2011). The Elder Scrolls V: Skyrim [Xbox 360].

Bioware. (2010). Mass Effect 2 [Xbox 360].

Costikyan, G. (2000). Where stories end and games begin. Game Developer, 7(9), 44–53.

Frasca. (2001). The Sims: Grandmothers are cooler than trolls. Game Studies: The International Journal of Computer Game Research, 1(1). Retrieved from http://www.gamestudies.org/0101/frasca/

Frasca, G. (2003). Simulation versus Narrative. In M. Wolf & B. Perron (Eds.), The Video Game Theory Reader (pp. 221–235). New York: Routledge.

Harrell, D. F., & Zhu, J. (2009). Agency Play: Dimensions of Agency for Interactive Narrative Design. In AAAI Spring Symposium: Intelligent Narrative Technologies II (pp. 44–52). Retrieved from http://www.aaai.org/Papers/Symposia/Spring/2009/SS-09-06/SS09-06-008.pdf

Hocking, C. (2007, October 7). Ludonarrative Dissonance in Bioshock - Click Nothing. Retrieved November 9, 2014, from http://clicknothing.typepad.com/click_nothing/2007/10/ludonarrative-d.html

Huizinga, J. (1955). Home Ludens: A Study of the Play Element in Culture. Boston: The Beacon Press.

Johnson, S. (2013, May 8). Should games have stories? Retrieved November 26, 2014, from http://gamasutra.com/view/news/191861/Should_games_have_stories.php

Loyall, B. (2004). Response. In N. Wardrip-Fruin & P. Harrigan (Eds.), First Person: New Media as Story, Performance, and Game (pp. 2–9). Cambridge, MA: The MIT Press.

Mateas, M. (2004). A Preliminary Poetics for Interactive Drama and Games. In N. Wardrip- Fruin & P. Harrigan (Eds.), First Person: New Media as Story, Performance, and Game (pp. 19– 34). Cambridge, MA: The MIT Press.

253

Mateas, M., & Stern, A. (2000). Towards Integrating Plot and Character for Interactive Drama. In In Working notes of the Social Intelligent Agents: The Human in the Loop Symposium.

Maxis. (200AD). The Sims.

Murray, J. (2004). From game-story to cyberdrama. First Person: New Media as Story, Performance, and Game, 1, 2–11.

Murray, J. (2013, June 28). The Last Word on Ludology v Narratology (2005). Retrieved January 15, 2015, from http://inventingthemedium.com/2013/06/28/the-last-word-on-ludology- v-narratology-2005/

Porteous, J., Charles, F., & Cavazza, M. (2013). Authoring Plan-based Narratives via a Social Network. In Proceedings of the 23rd International Conference on Automated Planning and Scheduling (pp. 14–17). Rome, Italy.

Riedl, M., & Bulitko, V. (2013). Interactive Narrative: An Intelligent Systems Approach. AI Magazine, 34(1), 67–77. https://doi.org/http://dx.doi.org/10.1609/aimag.v34i1.2449

Riedl, M. O. (2010). A comparison of interactive narrative system approaches using human improvisational actors. In Proceedings of the intelligent narrative technologies III workshop (p. 16). ACM. Retrieved from http://dl.acm.org/citation.cfm?id=1822325

Riedl, M. O., & Leon, C. (2008). Toward vignette-based story generation for drama management systems. In Proceedings of the 2nd international conference on intelligent technologies for interactive entertainment (INTETAIN), workshop on integrating technologies for interactive stories. Retrieved from http://www.cc.gatech.edu/~riedl/pubs/itis08.pdf

Ryan, M.-L. (2009). From Narrative Games to Playable Stories: Toward a Poetics of Interactive Narrative. Storyworld: A Journal of Narrative Studies, 1(1), 43–59.

Sharma, M., Ontañón, S., Mehta, M., & Ram, A. (2010). Drama management and player modeling for interactive fiction games. Computational Intelligence, 26(2), 183–211.

Tanenbaum, K., & Tanenbaum, J. (2009). Commitment to meaning: a reframing of agency in games. Digital Arts and Culture 2009. Retrieved from https://escholarship.org/uc/item/6f49r74n.pdf

Chapter 6

Archer, M. (2001). Being Human: The Problem of Agency. Cambridge: Cambridge University Press.

Atlus. (2011). Catherine [Xbox 360].

254

Banham, O. (2011, December 13). The Elder Scrolls V: Skyrim -. Retrieved June 30, 2015, from http://www.thunderboltgames.com/review/the-elder-scrolls-v-skyrim

Bethesda Game Studios. (2011). The Elder Scrolls V: Skyrim [Xbox 360].

Bioware. (2010). Mass Effect 2 [Xbox 360].

Brudvig, E. (2010, February 8). Mass Effect 2 Review. Retrieved June 30, 2015, from http://www.ign.com/articles/2010/02/08/mass-effect-2-review-2

Butler, J. (2001). Gender Trouble: Feminism and Subversion of Identity. Hoboken, NJ: Taylor and Francis.

Cheng, P. (2007). Waiting for something to happen: Narratives, interactivity and agency and the video game cut-scene. In Situated Play. Proceedings of the DiGRA 2007 Conference. Tokyo: DiGRA (pp. 15–24). Retrieved from http://www.digra.org/wp-content/uploads/digital- library/07311.24415.pdf

Costikyan, G. (2000). Where stories end and games begin. Game Developer, 7(9), 44–53.

Frasca, G. (2003). Simulation versus Narrative. In M. Wolf & B. Perron (Eds.), The Video Game Theory Reader (pp. 221–235). New York: Routledge.

Hamilton, K. (2011, November 9). Skyrim Will Have Infinite Randomly Generated Content. But Will It Ever Feel as Real as Minecraft? Retrieved June 30, 2015, from http://kotaku.com/5858056/skyrim-will-have-infinite-randomly-generated-content-but-will-it- ever-feel-as-real-as-minecraft

Hocking, C. (2007, October 7). Ludonarrative Dissonance in Bioshock - Click Nothing. Retrieved November 9, 2014, from http://clicknothing.typepad.com/click_nothing/2007/10/ludonarrative-d.html

Johnson, S. (2013, May 8). Should games have stories? Retrieved November 26, 2014, from http://gamasutra.com/view/news/191861/Should_games_have_stories.php

Joyce, L. (2014). Agency in Meaning and Intent: Limitations of Morality Systems in Interactive Narrative Games. In VG6. Oxford: Inter-Disciplinary.net. Retrieved from http://www.inter- disciplinary.net/critical-issues/wp-content/uploads/2014/05/joycevgpaper.pdf

255

Joyce, L. (2015). Creating Collaborative Criteria for Agency in Interactive Narrative Game Analysis. The Computer Games Journal. https://doi.org/10.1007/s40869-015-0004-x

Juul, J. (2013). The Art of Failure: An Essay on the Pain of Playing Video Games. Cambridge, MA: The MIT Press.

Karth, I. (n.d.). Ergodic Agency: How Play Manifests Understanding. Engaging with Videogames, 205.

Loyall, B. (2004). Response. In N. Wardrip-Fruin & P. Harrigan (Eds.), First Person: New Media as Story, Performance, and Game (pp. 2–9). Cambridge, MA: The MIT Press.

Mackenzie, C. (2009). Personal Identity, Narrative Integration, and Embodiment. In S. Campbell, L. Meynell, & S. Sherwin (Eds.) (pp. 100–125). University Park, PA: The Pennsylvania State University Press.

Magerko, B. (2007). Evaluating preemptive story direction in the interactive drama architecture. Journal of Game Development, 2(3), 25–52.

Mateas, M., & Stern, A. (2003). Integrating plot, character and natural language processing in the interactive drama Façade. In Proceedings of the 1st International Conference on Technologies for Interactive Digital Storytelling and Entertainment (TIDSE-03). Retrieved from http://users.soe.ucsc.edu/~michaelm/publications/mateas-tidse2003.pdf

McElroy, J. (2011, November 10). The Elder Scrolls V: Skyrim Review: Paths of desire. Retrieved June 30, 2015, from http://www.engadget.com/2011/11/10/skyrim-review/

Meynell, L. (2009). Introduction: Minding Bodies. In S. Campbell, L. Meynell, & S. Sherwin (Eds.), Embodiment and Agency (pp. 1–24). University Park, PA: The Pennsylvania State University Press.

Murray, J. (n.d.). The Last Word on Ludology v Narratology (2005). Retrieved August 24, 2015, from http://inventingthemedium.com/2013/06/28/the-last-word-on-ludology-v-narratology-2005/

Onyett, C. (2011, November 10). The Elder Scrolls V: Skyrim Review. Retrieved June 30, 2015, from http://www.ign.com/articles/2011/11/10/the-elder-scrolls-v-skyrim-review

Perlin, K. (2004). Can there Be a Form Between a Game and a Story. In N. Wardrip-Fruin & P. Harrigan (Eds.), First Person: New Media as Story, Performance, and Game (pp. 12–18). Cambridge, MA: The MIT Press.

256

Riedl, M., & Bulitko, V. (2013). Interactive Narrative: An Intelligent Systems Approach. AI Magazine, 34(1), 67–77. https://doi.org/http://dx.doi.org/10.1609/aimag.v34i1.2449

Robson, B. (2010, February 2). Mass Effect 2 - PC review at Thunderbolt. Retrieved June 30, 2015, from http://www.thunderboltgames.com/review/mass-effect-2-2

Ryan, M.-L. (2009). From Narrative Games to Playable Stories: Toward a Poetics of Interactive Narrative. Storyworld: A Journal of Narrative Studies, 1(1), 43–59.

Schreier, J. (2011, November 9). Skyrim Will Have Infinite Quests, Director Says. Retrieved March 27, 2016, from http://www.wired.com/2011/11/skyrim-infinite-quests/

Sharma, M., Ontañón, S., Mehta, M., & Ram, A. (2010). Drama management and player modeling for interactive fiction games. Computational Intelligence, 26(2), 183–211.

Sicart, M. (2009). The Ethics of Computer Games. Cambridge, MA: The MIT Press.

Sicart, M. (2013). Beyond Choice. Cambridge, MA: The MIT Press.

Sterling, J. (2011, November 10). Review: The Elder Scrolls V: Skyrim. Retrieved June 30, 2015, from http://www.destructoid.com/review-the-elder-scrolls-v-skyrim-215507.phtml

Tanenbaum, K., & Tanenbaum, J. (2009). Commitment to meaning: a reframing of agency in games. Digital Arts and Culture 2009. Retrieved from https://escholarship.org/uc/item/6f49r74n.pdf

Telltale Games. (2012). The Walking Dead [Xbox 360].

Walker, A. (n.d.). The Choose Your Own Adventure Books Were The First Interactive Games. Retrieved April 14, 2016, from http://www.kotaku.com.au/2014/11/the-choose-your-own- adventure-books-were-the-first-interactive-games/

257

Wardrip-Fruin, N., Mateas, M., Dow, S., & Sali, S. (2009). Agency reconsidered. Breaking New Ground: Innovation in Games, Play, Practice and Theory. Proceedings of DiGRA 2009. Retrieved from http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.190.1393&rep=rep1&type=pdf

Westecott, E. (2008). The performance of digital play. In Forum: Special Issue 2 Play. Retrieved from http://www.forumjournal.org/site/issue/special/play/emma-westecott

Chapter 7

Archer, Margaret. Being Human: The Problem of Agency. Cambridge: Cambridge University Press, 2001. Print.

Bethesda Game Studios. The Elder Scrolls V: Skyrim. N.p., 2011. Xbox 360.

Bioware. Mass Effect 2. N.p., 2010. Xbox 360.

Cardboard Computer. Kentucky Route Zero. N.p., 2013. Microsoft Windows.

Costikyan, Greg. “Where Stories End and Games Begin.” Game Developer 7.9 (2000): 44–53. Print.

Fendt, Matthew William et al. “Achieving the Illusion of Agency.” Interactive Storytelling. Springer, 2012. 114–125. Google Scholar. Web. 14 Nov. 2014.

Heron, Michael James, and Pauline Helen Belford. “Do You Feel Like a Hero Yet? Externalised Morality in Video Games.” Journal of Games Criticism 1.2 (2014): n. pag. Web. 9 Nov. 2014.

Joyce, Lindsey. “Agency in Meaning and Intent: Limitations of Morality Systems in Interactive Narrative Games.” VG6. Oxford: Inter-Disciplinary.net, 2014. Web.

Köller, Joe. “Blank Slate.” Haywire Magazine. N.p., 14 Apr. 2013. Web. 15 Nov. 2014.

Mateas, Michael. “A Preliminary Poetics for Interactive Drama and Games.” First Person: New Media as Story, Performance, and Game. Ed. Noah Wardrip-Fruin and Pat Harrigan. Cambridge, MA: The MIT Press, 2004. 19–34. Print.

Mateas, Michael, and Andrew Stern. “Integrating Plot, Character and Natural Language Processing in the Interactive Drama Façade.” Proceedings of the 1st International Conference on Technologies for Interactive Digital Storytelling and Entertainment (TIDSE-03). N.p., 2003. Google Scholar. Web. 9 Nov. 2014.

---. “Towards Integrating Plot and Character for Interactive Drama.” In Working Notes of the Social Intelligent Agents: The Human in the Loop Symposium. N.p., 2000. Print.

258

“Moral Decisions (Concept) - Giant Bomb.” N.p., n.d. Web. 8 Nov. 2014.

Sicart, Miguel. The Ethics of Computer Games. Cambridge, MA: The MIT Press, 2009. Print.

Smith, Ed. “Catherine: How Scoring Systems Kill The Mood | Medium Difficulty.” Medium Difficulty. N.p., 13 Apr. 2012. Web. 8 Nov. 2014.

Telltale Games. The Walking Dead. N.p., 2012. Xbox 360.

Chapter 8

“Additional Events - The Banner Saga Wiki Guide.” IGN. N.p., n.d. Web. 30 Dec. 2016.

Archer, Margaret. Being Human: The Problem of Agency. Cambridge: Cambridge University Press, 2001. Print.

Buchanan, Matthew. “Can Games Teach Us To Die?” wg420b. N.p., n.d. Web. 15 Jan. 2017.

Costikyan, Greg. “Games, Storytelling, and Breaking the String | Electronic Book Review.” Electronic Book Review. N.p., 28 Dec. 2007. Web. 9 Nov. 2014.

---. “Where Stories End and Games Begin.” Game Developer 7.9 (2000): 44–53. Print.

Filipowich, Mark. “A Family That Plays Together.” bigtallwords. N.p., 21 June 2016. Web. 15 Jan. 2017.

---. “Resource Based Humanity.” bigtallwords. N.p., 6 May 2015. Web. 15 Jan. 2017.

Frasca, Gonzola. “Simulation versus Narrative.” The Video Game Theory Reader. Ed. Mark Wolf and Bernard Perron. New York: Routledge, 2003. 221–235. Print.

Hayles, N. Katherine. How We Became Posthuman: Virtual Bodies in Cybernetics, Literature, and Informatics. Chicago: University of Chicago Press, 1999. Print.

Hocking, Clint. “Ludonarrative Dissonance in Bioshock - Click Nothing.” Click Nothing. N.p., 7 Oct. 2007. Web. 9 Nov. 2014.

259

Johnson, Leif. “The Banner Saga Review.” IGN. N.p., 14 Jan. 2014. Web. 15 Jan. 2017.

Johnson, Soren. “Should Games Have Stories?” Gamasutra. N.p., 8 May 2013. Web. 26 Nov. 2014.

Joyce, Lindsey. “Assessing Mass Effect 2 and Elder Scrolls V: Skyrim.” Journal of Games Criticism. N.p., n.d. Web. 18 Dec. 2016.

Loyall, Bryan. “Response.” First Person: New Media as Story, Performance, and Game. Ed. Noah Wardrip-Fruin and Pat Harrigan. Cambridge, MA: The MIT Press, 2004. 2–9. Print.

McElroy, Griffin. “The Banner Saga Deals Death, Both in Combat and Storytelling.” Polygon. N.p., 24 Dec. 2013. Web. 15 Jan. 2017.

Porteous, Julie, Fred Charles, and Marc Cavazza. “Authoring Plan-Based Narratives via a Social Network.” Proceedings of the 23rd International Conference on Automated Planning and Scheduling. Rome, Italy: N.p., 2013. 14–17. Print.

Reeves, Ben. “Fight For The Top 50 2014 – The Banner Saga.” Game Informer. N.p., n.d. Web. 15 Jan. 2017.

Riedl, Mark, and Vadim Bulitko. “Interactive Narrative: An Intelligent Systems Approach.” AI Magazine 34.1 (2013): 67–77. Web.

Ryan, Marie-Laure. “From Narrative Games to Playable Stories: Toward a Poetics of Interactive Narrative.” Storyworld: A Journal of Narrative Studies 1.1 (2009): 43–59. Print.

260

Sharma, Manu et al. “Drama Management and Player Modeling for Interactive Fiction Games.” Computational Intelligence 26.2 (2010): 183–211. Print.

Sicart, Miguel. Beyond Choice. Cambridge, MA: The MIT Press, 2013. Print.

---. The Ethics of Computer Games. Cambridge, MA: The MIT Press, 2009. Print.

Smartt Bell, Madison. Narrative Design: Working with Imagination, Craft, and Form. New York: W.W. Norton & Company Inc., 1997. Print.

Stoic. The Banner Saga. N.p., 2014. Microsoft Windows.

Tack, Daniel. “The Banner Saga Review – Of Mead And Massacres.” Game Informer. N.p., n.d. Web. 15 Jan. 2017.

Tanenbaum, Karen, and Joshua Tanenbaum. “Commitment to Meaning: A Reframing of Agency in Games.” Digital Arts and Culture 2009 (2009): n. pag. Google Scholar. Web. 9 Nov. 2014.

“The Banner Saga – Characters.” N.p., n.d. Web. 15 Jan. 2017.

“The Banner Saga FAQ/Walkthrough for PC by Mythril Wyrm - GameFAQs.” N.p., n.d. Web. 30 Dec. 2016.

Chapter 9

Joyce, Lindsey. “Kentucky Route Zero: Or, How Not to Get Lost in the Branching Narrative System.” The Play Versus Story Divide in Game Studies: Critical Essays. Jefferson: McFarland, 2016. 17–27. Print.

Chapter 10

Frasca, Gonzola. “What Is Ludology? A Provisory Definition.” Ludology.org. N.p., n.d. Web. 15 Apr. 2017.

261

BIOGRAPHICAL SKETCH

Lindsey Joyce was born in Baltimore, Maryland on July 9, 1982. She holds a Bachelor of Arts and a Master of Arts in English from West Virginia University. Lindsey has held faculty positions at West Virginia University, Park University, Richland College, and The University of

Texas at Dallas.

262

CURRICULUM VITAE

Education Ph.D. in Arts and Technology, The University of Texas at Dallas, Dallas, TX. Expected May 2017 M.A. in English, West Virginia University, Morgantown, WV. August 2010 B.A. in English, West Virginia University, Morgantown, WV. May 2008

Peer-Reviewed "Agency in Meaning and Intent: Limitations of Morality Systems in Publications Interactive Narrative Games." Ed. Lindsey Joyce and Brian Quinn. Oxford: Inter-Disciplinary Press, 2016. “Assessing Mass Effect 2 and Elder Scrolls V: Skyrim Using Collaborative Criteria for Player Agency in Interactive Narratives” Journal of Game Criticism (2016) "Kentucky Route Zero: Or How Not to Get Lost in the Branching Narrative System” The Play Versus Story Divide in Game Studies: Critical Essays. Ed. Matthew Wilhelm Kapell. Jefferson, NC: McFarland Publishers, December 2015. "Creating Collaborative Criteria for Agency in Interactive Narrative Game Analysis.” The Computer Games Journal (2015): n. pag. CrossRef. Web. 7 July 2015. “The Medium is the Message: The Last of Us as Argument Against Misrepresentations and Stereotypes of Gender in Games.” First Person Scholar. 9 July 2014. http://www.firstpersonscholar.com/an-uneven- partnership/ Alternative & Pop “Flip This House: Tips for Bridging the Tech Gap Between School and Publications Home” EdSurge News (2016). "Videogames for a Good Kid" Five Out of Ten Magazine. Ed. Alan Williamson. Oxford, Five Out of Ten, April 2015. "Losing Luck, Finding Control" Five Out of Ten Magazine. Ed. Alan Williamson. Oxford, Five Out of Ten, April 2015. "Pulling the Strings" Haywire Magazine. Ed. Joe Koeller. Haywire, February 2015. “How Does Wayward Manor Fit Into Neil Gaiman’s Oeuvre? It Doesn’t.” Killscreen. 13 August 2014. http://killscreendaily.com/articles/how- does-wayward-manor-fit-neil-gaimans-ouevre-it-doesnt/ “The Impeachable Optimism of Cosmochoria.” Killscreen. 23 June 2014. http://killscreendaily.com/articles/unimpeachable-optimism- cosmochoria/ Presentations “Constructing Space, Creating Time, Considering Bodies: Structures of Narrative Systems” – Extending Play Conference, New Brunswick 2016 “Constructing Space, Creating Time, Considering Bodies: Structures of

Narrative Systems” – Videogame Cultures & the Future of Interactive Entertainment, Mansfield College, Oxford, September 2016 “Personal Criticism: Bringing Personal Experience and Narrative into Critical Essays” – Writefest, Houston 2016 “How to Create and Cultivate Your Personal Brand” – Writefest, Houston 2016 "Let's Plays: Curating a New Type of Discourse for Digital Play Experiences" - Digital Frontiers, Dallas 2015 "Ethical Values and Valuation: New Sites and Systems of Interaction in Play Space" - CGSA, Ottawa 2015 “Assessing Mass Effect 2 and Elder Scrolls V: Skyrim using Collaborative Criteria for Player Agency in Interactive Narratives.” – Extending Play Conference, New Brunswick 2015 "Kentucky Route Zero: Or How Not to Get Lost in the Branching Narrative System,” RAW Graduate Symposium, Dallas 2015 “Agency in Flow: Creating a Collaborative Criteria for Interactive Narrative Game Analysis,” Society for Literature Science and the Arts, Dallas 2014 “Assessing Mass Effect 2 and Elder Scrolls V: Skyrim using Collaborative Criteria for Player Agency in Interactive Narratives.” Rice Graduate Symposium, Houston 2014 “Agency in Meaning and Intent: How Dialogue and Visual Feedback Systems Limit Agency in Games with Morality Systems,” Videogame Cultures & the Future of Interactive Entertainment, Mansfield College, Oxford, July 2014 “The Failure of Prudence,” West Virginia University Graduate Student Colloquium, Morgantown, March 13, 2010 “Rhythm Scientist of Musical Thief: An Analytical Interview which Questions the Plagiarism of Sound and the Agency of Influence in DJ Culture,” Society for Literature science and the Arts, Atlanta, November 2009 “Coming Around: A Re-Vision of Horticulture and Culture in Leslie Marmon Silko’s Gardens in the Dunes,” Eighth Native American Symposium, Durant, November 2009 “Coming Around: A Re-Vision of Horticulture and Culture in Leslie Marmon Silko’s Gardens in the Dunes,” West Virginia University English Graduate Student Colloquium, Morgantown, April 4 2009 “Rejecting the Tropes of Femaleness: The Female Body as a Means of Resistance in Nervous Conditions and Woman at Point Zero,” West Virginia University Undergraduate Literary Symposium, Morgantown, April 2008

Edited Curations “The Week in Video Game Blogging.” Critical-Distance. Critical- Distance.com.

"Blogs of the Round Table." Critical-Distance. Critical-Distance.com

“This Month in Let’s Plays.” Critical-Distance. Critical-Distance.com

Teaching Experience Senior Lecturer, The University of Texas at Dallas 2016-Present • Instructor of Digital Content Design and Freshman Seminar • Researcher in Narrative Systems Research Lab Graduate Teacher, The University of Texas at 2014-2015 Dallas • Instructor of Writing and Research in New Media Adjunct Professor, Richland College Fall 2014 • Instructor of Composition and Rhetoric Adjunct Professor, Parker University 2012-2013 • Instructor of Composition and Rhetoric and American Literature Senior Lecturer, West Virginia University 2010-2012 • Instructor of Composition and Rhetoric Graduate Intern, Center for Literary Computing 2009-2010 • Provider of digital media, interactive experiences, and web design services Graduate Teaching Assistant, West Virginia 2008-2010 University • Instructor of Composition I and Composition II

Industry Digital Marketing Lead, Istation 2015-2016 Experience • Developed, maintains, and tracks marketing campaigns • Analyzed and optimized content for effectiveness • Recommended changes or new ideas based on performance • Conceptualized how content can be used in multiple formats • Maintained the content database • Sourced content by interviewing internal and external customers • Researched tips, best practices, and news through social media • Filtered the best material to be shared with audiences

Managing Editor, Five out of Ten Magazine 2015-2017 • Turns submissions into publishable articles • Contributes feature article as needed • Promoted the magazine via social media • Assists with overall direction of the magazine

Managing Editor, Haywire Magazine 2014-2015 • Develops brand and product output • Manages team of 4 on-staff writers and 4 on-staff editors • Receives and responds to pitches from new/guest writers

Program Director, Set For College 2012-2015 • Designed, develops, and implements tutoring programs • Designs SAT and ACT curriculum • Manages marketing and communications

National Writing Project Director 2010-2012 • Creates and maintains partnership with k-12 teachers in 7 counties • Develops new strategies for the writing classroom in k-12 • Maintains affiliations with National Writing Project network • Applies for annual federal funding and oversees expenditures • Maintains advocacy team at national level

Service Founding Member, Women in Games Network 2016 Inaugural Organizing Member, Jam the System 2016 Game Jam Committee Member, Game Production Lab 2016 Proposal Reviewer, Extending Play 3 2016 Committee Member, ATEC Dean Search 2015-2016 Graduate Student Representative, ATEC Colloquia 2014-2016 Treasurer, English Graduate Organization 2009-2010 Student Representative, Graduate Student 2009-2010 Organization

Professional DIGRA Affiliations SLSA CCCC ICA Inter-Disciplinary.Net ELO