Replaying history: learning world history through playing

Replaying history: learning world history through playing Civilization III

Kurt D. Squire

Indiana University

Accepted by the Graduate Faculty, Indiana University, in partial fulfillment of the requirements for the degree of Doctor of Philosophy.

_______________________________________

_______________________________________

_______________________________________

COPYRIGHT PAGE

Kurt D. Squire

ALL RIGHTS RESERVED
 

Digital games is an emerging entertainment medium that an increasing number of educators are examining as tools for engaging learners. Yet, few models exist for how to use contemporary gaming media in formal learning environments. A commercial historical computer strategy game such as Civilization III is an intriguing artifact to examine in classroom contexts because of its wide appeal, design sophistication, and unique affordances as a world history simulation. Civilization III represents world history not as a story of colonial domination or western expansion, but as an emergent process arising from overlapping, interrelated factors.

The purpose of this study is to explore what happens when Civilization III, a complex computer game developed in entertainment contexts enters formal learning environemtns. This dissertation presents three naturalistic case studies in which Civilization III was used as the basis for a unit on world history in urban learning environments. I examine how the game engaged players, the social interactions that occur, how understandings emerge, and what role game play serves in mediating students’ understandings.

In all three cases, engagement was a complex process of appropriation and resistance, whereby the purposes of game play was negotiated among students’ identities, classroom goals, and the affordances of Civilization III. Civilization III engaged each student in unique ways, and this engagement affected the kinds of questions students asked of their games, the kinds of conceptual understandings that arose through game play, and the interpretations they made about history. History and geography became tools for game play and successful students developed conceptual understandings across world history, geography, and politics. These cases suggest the potential for using simulation games in world history education, but also the significant, unsolved challenges in integrating such a complex game within classroom settings.

_____________________________________________

_____________________________________________

_____________________________________________

_____________________________________________

Dedication

Dedicated to James Douglas and Janet Kretschmer, two people who taught **

Table of Contents

II. Game-Based Learning in World History 22

III. Methodology

IV. Case 1: The Media School

V. Case 2: Media Summer Camp

VI. Case 3: After-School Computer Club

VII. Conclusions

VIII. Implications

A growing number of researchers and scholars are acknowledging the cultural impact of digital games (Gee, 2003; King & Borland, 2003; Poole, 2001). Digital gaming is now an $18 billion global industry that many media scholars see as being a dominant "lively art" in the upcoming decades (Jenkins, in press). As Jim Gee (2003) argues, games are not only pushing the creative boundaries of interactive digital media but also suggesting powerful models of next-generation interactive learning environments. Those in the “edutainment” industry, as well as the teachers and students who support it, appear to agree. Year after year, social studies edutainment games such as The Sims, SimCity, Age of Empires, Railroad Tycoon and Civilization dominate the PC gaming sales charts (Squire, 2002). Many social studies teachers seem eager to exploit this new medium, as simulation games such as SimCity are installed on school computers throughout the country and thousands of teachers download the SimCity 3000 teacher’s guide (Bradshaw, 2002; Teague & Teague, 1995).

Despite the commercial success of — and educators' growing interest in — games like Pirates!, SimCity 3000 or Civilization, very little is known about how such games might be used as tools for learning. Although a growing number of educators, industry leaders, and political leaders have suggested that SimCity or Civilization could be used in social studies classrooms (Berson, 1996; Hope, 1996; Kolson, 1996; Lee, 1994; Prensky, 2001; Teague & Teague, 1995), there are to date no empirical research studies examining their effectiveness in classroom environments. Important questions persist about how teachers might use such simulations and how learners come to understand them. How might we leverage these games for use in formal or informal learning environments? What happens when you bring a complex world history simulation game such as Civilization III into the classroom? Does such a complex game — one that often positions players in situations where “academic” knowledge and understanding can be leveraged for real use in problem-solving — provide opportunities for supporting new kinds of learning? Or might the simplifications (hence distortions) inherent in any simulation reinforce, or even cause misconceptions about important historical, cultural, or geographical phenomena?

That simulation games can potentially distort the phenomena they are meant to model has been widely acknowledged. In his critique of SimCity, city planner Kenneth Kolson (1996) notes that SimCity distorts the powers of a mayor in public planning, discounts the historical importance of race and ethnicities in the evolution of cities, and overestimates the appeal of public transportation to most Americans. Similarly, Barkin (2001) notes that in attempting to capture, quantify and operationalize the dynamics of culture, Civilization III offers an ostensibly problematic concept of culture drawn from French and German theories of culture that is foreign to any anthropologist. This problem of simplification/distortion of “reality” in games is exacerbated by the fact that edutainment products are typically developed and marketed as entertainment products first, and then appropriated for use in classrooms second. Other tensions, such as the tension between playing the game as a bounded semiotic system versus reflecting on the game as a model representing some more substantial phenomena in the world beyond it, may very well be endemic to the medium.

Contemporary digital gaming models such as that underlying Civilization III are potentially powerful learning tools that are understudied as a viable educational resource. Studying learning in digital games might teach instructional technologists valuable lessons about how to design interactivity, support online collaboration, or engage users. Understanding how such games are used in formal learning environments might productively inform the design of educational games (Games-to-Teach Team, 2003). At the very least, educational technologists could benefit from paying closer attention how players are already interacting with such “edutainment” games and how they are being used in classrooms (Squire, 2003).

Despite the lack of formal inquiry into the potential for digital games to support learning, there is a long tradition of using paper-based games and simulations in social studies classrooms (e.g. Clegg, 1991). Unfortunately, most uses of games have been atheoretical; rarely, if ever are they tied to contemporary notions of how people learn or the broader goals of social studies education. Digital games, which bring with them new affordances, possibilities, and potential problems, have yet to be seriously studied in classroom contexts.

Building on this past research in (largely paper-based) games in social studies education, I argue that educators need to examine not just the game – player system, but the broader social contexts of game play. Cooperative and competitive social arrangements frame game play activity. In some cases, the social context of game play – the kinds of reflection activities, discussion, collaboration, and competition that emerge in game play are as important as the game itself in determining what activity emerges and what learning occurs. Prior research has assumed a priori what the learning goals and outcomes of game-based learning environments should be, treating games as content transmission systems as opposed to tools to think with. I argue that games might be more conducive to constructivist instructional approaches, whereby learning is an inferential, interpretive process and learning outcomes are intricately tied to the goals, intentions, and motivations of the learner (e.g. Bednar, Cunningham, Duffy & Perry, 1996).

Indeed, if the activity outside of the game (discussions, research, knowledge sharing) are as important as the game itself, then educational game researchers need a theoretical model which accounts for both student-game interactions and student-student interactions. I argue for a cultural-historical approach to understanding learning in game-based learning environments, as it allows researchers to examine not only the role of the game in learning, but how social structures mediate activity.
Theoretical Foundation

Underlying these debates about the potential of games to support learning are theoretical questions central to instructional design, educational technology, identity and learning, teaching world history, and the learning sciences more generally. Both proponents and critics of digital game-based learning have habitually assumed objectivist epistemologies and transmission models of learning, whereby the game contains fixed meanings which are broadcasted to a passive game-playing recipient (e.g. Prensky, 2001; Provenzo, 1991). How players infer meanings from game play, construct understandings about game worlds, and then relate these experiences to non-gaming experiences is not entirely clear; where do players draw lines between fantasy and reality? How do players know when a game is realistic and when it is not? How do players explore game worlds as systems and how do they treat these understandings of game systems?

Most educational game research has treated game play as isolated psychological phenomena, ignoring the broader social contexts of game playing and social relationships that envelope most gaming experiences (e.g. Grossman, 2000, Malone, 1981). Treating the learning context as an interaction between an isolated player and a game as an isolated system is problematic on several levels; games are frequently competitive endeavors where players test skills against other players, cooperative exercises where players work together to solve problems (whether it be in single player or multiplayer games), or simply excuses for friends and families to socialize. Minimally, game play as social practice can be characterized by the social purposes it serves, the social relationships which become folded into game play, and the formal and informal communities that arise in support of game play.

For educators interested in harnessing the power of games to support learning (e.g. Games-to-Teach, 2003; Media X, 2003; Prensky, 2001), this challenge of how to account for both the person-tool interaction and the broader social contexts in which gaming is situated and game meanings are created is crucial (Squire, 2002). Elsewhere (e.g. Squire, 2002), I have argued for adopting a socio-cultural learning perspective to understand gaming (In What Video Games Have to Teach Us About Learning and Literacy, Jim Gee also draws from socio-cultural learning theory in describing how learning occurs through gaming). Socio-cultural learning theory (defined more precisely in the theoretical section) offers game-based educators several insights into learning through game play: (1) Knowledge is described not as facts to be memorized but as tools which mediate activity (Barab, Hay, Barnett, & Squire, 2001; Vygotsky, 1978); (2) Socio-cultural learning theory encourages researchers to view game play not as purely a human-computer interaction phenomena, but as a socio-cultural one mediated by classroom microcultures and broader social contexts, including classroom culture; (3) Socio-cultural learning theory provides a framework for understanding students’ goals and intentions and how these contribute to trajectories of students’ identities, and (4) a language (a theory of signs, or semiotics) for thinking through how knowledge is represented in games and how this knowledge develops in a learning environment. Of particular interest to me is how socio-cultural learning theory might provide a language for examining classroom practice mediated by game play and situated within classroom cultures.

Activity theory, a neo-Vygotskian socio-cultural theory emerging from the Russian School of psychologists offers a particularly interesting lens for educators interested in examining game-based learning environments. Activity theory takes human work as its unit of analysis. For activity theorists, work is organized by an object, which shapes activity and reciprocally is influenced by human actors, as mediated by tools and social institutions. By taking work as the unit of analysis, activity theorists examine the tools, signs, and language which mediate human interaction with object, as well as the social structures, including community norms and divisions of labor which frame activity. As such, activity theory takes the person performing in social contexts, including the social and political environs in which they are situated as the minimal meaningful unit of analysis. Importantly, activity theorists regard humans and the objects of their activity in dialectal relations, shaping and reshaping one another through time.

Activity theory is an intriguing theoretical framework for understanding gaming because it focuses researchers’ attentions not only on how a tool such as Civilization III mediates learning of social studies, but also focuses researchers on how game play is mediated by social structures, which might include school cultures or informal groupings. Given that games are profoundly social experiences (King & Borland, 2003), it is critical that game researchers focus not just on human and computer interactions, but on how emergent game cultures shape gaming activities and the impact that these activities have on cognition. By examining the object, or focus of activity, activity theorists are also interested in how participants view and understand activity, particularly participants’ objects, goals, or motives. Emerging theory in game studies suggests that gamers approach games in unique ways, and one cannot assume a priori to know a player’s goals and intentions while gaming (e.g. Bartle, 2003).

Influenced by Hegel and Marx, activity theorists are very interested in the material conditions of work, and adopt an historical approach to understanding activity (Engeström, 1999). Humans, their tools, signs, and language – as well as the community norms and structures in which they are situated – are understood historically by investigating their use in actual settings, frequently through traditional ethnographic, historic, or qualitative case study techniques (Engeström, 1999). Activity theorists enter activity settings, observing and interviewing participants and generating narratives of what activity emerges (e.g. Engeström, 1999). Critical to an activity theory approach is understanding how activity systems are viewed from multiple vantage points and teasing out contradictions among differing activity systems, particularly the contradictions that emerge when activity systems overlap.

One might anticipate several contradictions, such as contradictions between using games for enjoyment vs. using games to master social studies, or collaborative communities of inquiry vs. competitive gaming structures. Game-based Educational technologists working in other settings (e.g. Barab, Barnett, Yamagata-Lynch, et al., 2002) have used contradictions to understand change and innovation in a system, finding that contradictions can be a useful tool for refining design experiments.

Specifically, this dissertation examines what classroom practices emerge and how learning occurs when Civilization III is used as the basis for learning about world history in two learning environments (1) an Humanities enrichment course in a Media and Technology Charter School (Media case) and an after-school computer club program. Using qualitative case study techniques, I examine the following five research questions:

1. 
   What practices and contradictions emerge when games are brought into formal learning environments, particularly, how do gaming practices (e.g., competition, learning through failure) intersect with the practices and culture of formal schooling?
   
2. 
   How does Civilization III engage players in formal learning environments?
   
3. 
   How does learning occur through game play, specifically, how does playing Civilization III remediate students’ understandings of history?
   
4. 
   What are the pedagogical potentials (affordances) of using games (specifically Civilization III) in world history classrooms?
   
5. 
   How should we design learning activities and environments when using games in formal learning environments?
   

Consistent with Cobb and colleagues (Cobb, Stephan, McClain, & Gravemeijer 2001), this study is a design experiment designed to examine what happens when Civilization III is used as the basis for learning world history in three contexts. The first case is a month-long unit on world history, as a part of a ninth grade humanities class at a Media and Technology Charter School in inner-city Boston. In the second case, a subset of these students participated in a week-long, half-day computer camp investigating the potential of using Civilization III to learn about social studies. The third case is an after-school computer club sponsored by the YWCA but occurring at a suburban, working class Boston middle school. All three cases were convenience samples, chosen for their willingness to participate in this experimental program and ability to illuminate research issues, and each case involved approximately 20 hours of instructional time.

I use Stake’s (1995) case study techniques to address these research questions. Stake’s case study technique is particularly useful because it is responsive to the particularities of a case, including the unintended consequences. With no real empirical research on what happens in game-based learning environments, little is known as to what will happen when games enter classrooms, let alone what types of learning occurs. There are other important questions about how classroom cultures will appropriate gaming media, how non-gamers react to game-based learning units, how games compete with other learning activities for students’ attention, or how girls take to game-based learning environments (See Cassell & Jenkins, 1998 for a discussion of gender and gaming). Stake’s methodology emphasizes the importance of not over-prescribing data collection and research procedures, but of allowing data collection to emerge in response to emerging themes. I use Stake’s case study methodology (1995) for each, using observations, interviews, and document analysis to build narrative accounts of each classroom.

In each case I was a direct participant. Although I had planned to participate in each case as little more than an observer, local needs demanded that I play an active role in shaping classroom activity. As a result, I hired a research assistant to assist in data collection and analysis. Consistent with Cobb, Confrey, diSessa, Lehrer, and Schauble, (2003), I approach this design experiment equally as a teaching experiment, whereby interacting with participants and the case yields fruitful data about the design of learning environments. Researchers can modify the learning environment (e.g. introducing new learning materials, manipulating social arrangements) in order to illuminate research themes. In this these cases, I try to make this cycle of manipulating the environment and examining results as explicit as possible, so that the reader can perhaps vicariously experience some of the decision making process I experienced. Negotiating this role was often tricky, and I try to give the reader a sense of these struggles in each case study. Because the research questions involve examining what practices emerged when Civilization III was brought into the classroom, as opposed to directly comparing a game-based learning environment to a traditional environment, this study avoids some of the more obvious threats to validity, such as “tainting” the research environment. At the same time, my participation in the case makes the applicability of these findings to other contexts somewhat limited, as I am not the typical teacher. These limitations are explored further in the next section.

This study is designed to examine what happens when Civilization III is used as the basis for a unit in learning world history. Very little is known about what happens when a game as complex, abstracted, and simulation-based as Civilization III is used to study world history, and although myself and others have argued for different models for thinking about how games can be used to support learning, little is known about how these approaches play out in practice. Importantly, this study examines one game (Civilization III) being used for select purposes in three very specific settings. As such, this study has limited applicability how other games (such as Europa Universalis, Patrician, or 1602 AD) might be used in world history, how games such as Colonization might be used in colonial history, or how games such as Hidden Agenda might be used in modern history. I believe that these cases should provide useful insights for educators exploring such models, but caution against extrapolating too far from these results.

One of the biggest limitations of this study is the samples chosen. These samples were chosen for convenience – specifically, for accessibility and willingness to experiment with an innovative unit. All of these cases involved students from working-class backgrounds, populations of students who are known to resist learning history (Loewen, 1995). Many of the Media students were highly resistant to authority, creating some tense moments as teachers and researchers tried to require outside learning activities. Whereas students in traditional, middle-class or upper-class contexts might be expected to engage willingly in outside research, readings, or discussion activities, students in the Media case were reluctant to engage in such activities (See Chapter IV). At the same time, this case is particularly illuminative of the tensions between students’ and teachers’ intentions; as the case study shows, students who were not interested in playing the game – or more interested in playing the game than studying history – were quick to make their opinions known to researchers. Consistent with the case study approach (Stake, 1995), I attempt to highlight the particularities of this case and support the reader in generalizing my findings to his or her own learning contexts as deemed appropriate.

In the after-school case, game players were perhaps more amenable to augmenting game play with other learning activities, but having students do readings or other activities would have run contrary to the purposes of the camp. At the same time, these participants were unlike participants in many after-school environments. They attended regularly, had little choice among other activities, and seemed to adopt the social mores of school, perhaps because the camp occurred on school grounds and in a computer lab. Designers of after-school settings such as Boys and Girls Clubs may find that the social mores of this case bear little resemblance to those they face, and that there is little transferability from this case to their situations.

Role of researcher

In both cases, I played an active role in shaping the learning environment. I devised activities, offered help on game play, devised just-in-time lectures, and tried to connect students’ games to historical events. I have spent thousands of hours playing Civilization III and nearly that same amount thinking through this dissertation. It is unreasonable to expect that a typical teacher would have the experience or energy to do the same. I have attempted to capture what I learned from these experiences in unit plans and through suggestions for designing curricula with Civilization III (see Appendices A and B for sample unit plans using Civilization III in other areas of world history); nevertheless, if the teacher is a critical component of a game-based learning environment, then my role needs to be accounted for. (This issue is further explicated in design research in general in Barab & Squire, in press).

At the same time, I came into each of these cases as an outsider to the school or camp cultures and was disadvantaged in terms of integrating the game into school and classroom cultures and anticipating how game play would meet students’ needs. One can imagine that a world history teacher who plays Civilization III may be able to better integrate game experiences into the curriculum, anticipate students’ misconceptions, or understand how to negotiate moment-to-moment classroom interactions. In truth, permanent school faculty would have a much deeper knowledge of how to integrate the game in such ways.

Each of these cases was a fairly substantial unit, lasting 4-6 weeks and including a minimum of 20 teacher-student contact hours. At the same time, Civilization III is a complex game to learn and a single game can take dozens of hours to play. These limitations on contact hours and students’ inability to take games home to play meant that students had relatively little time to experiment with the game. A dedicated Civilization III player might spend 20 hours playing Civilization III in a weekend; these students had relatively little time to learn the game interface, experiment with alternative strategies, or explore the game more generally. One can imagine how a unit that lasted the duration of a semester, a learning environment with more flexible time allotments, or educational programs where students had laptops or access to home computers where they could play the game, might develop differently.

There is a tendency for many researchers to treat “games” or “game-based learning environments” as a meaningful category or variable with little respect to the specific games or game genres that are being studied. Civilization III is a turn-based resource management strategy game where players exploit natural resources, build civilization and city improvements, set tax rates, and negotiate with other civilizations. Civilization III is an open-ended game meant to be played in a multitude of ways and support multiple game strategies. As an emerging medium, games are often treated monolithically, as if the practice of playing Quake, a first-person action game is the same as playing Civilization III, a relatively slow-paced strategy game (See also Games-to-Teach Team, 2003). Much the same way that one would not want to do a case study of students learning to read with the Bible and then generalize to books in general, one would not want to take this study and generalize the findings to games in general.

Chapter II provides a background for using Civilization III to support learning in world history. I examine the practical, intellectual, and theoretical issues behind studying world history. I cover the history of research of using games and simulations in social studies education, and present a theoretical argument for the potential benefit of using games in world history education, using activity theory as a lens for discussing how learning might occur through game play and how game-based learning environments can be investigated.

Chapter III presents my methodology. I describe the structure of the design experiment, discussing the role of the researcher and the affordances of Civilization III as a tool for studying world history. I also detail my methodology for generating case studies and the analysis procedures I used for generating assertions and analyzing activity systems.

Chapters IV, V, and VI present the three case studies. In Chapter VII, I offer my conclusions, and Chapter IX presents my proposed implications for the design of game-based learning environments.