1 CONTEXTUALIZATION CUES AND TEAMWORK IN ONLINE VIDEO GAME GAMEPLAY DISCOURSE By LUKE BRELAND A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUI REMENTS FOR THE DEGREE OF MASTER OF ARTS UNIVERSITY OF FLORIDA 2012
2 2012 Luke Breland
3 To my loving wife Katherine
4 ACKNOWLEDGMENTS I thank first and foremost Dr. Boxer and Dr. Wehmeyer, whose suggestions and guidance have been invaluable. I also extend my gratitude to all of the thoughtful and insightful discussions in online video game communities, especially with the number crunching nerds at Symthic; participation in these communities has been as educational as it has b een entertaining. I cannot ever thank my family enough for their support throughout my life and my academic career and my brother Finn for always being there to play a game and talk with me when ever my spirits need to be lifted Finally, I could never be the person I am today without my wife Katherine and her unwavering love an d support.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF FIGURES ................................ ................................ ................................ ......................... 7 ABSTRACT ................................ ................................ ................................ ................................ ..... 8 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .................. 10 Video Games as a Common Setting for Interaction ................................ ............................... 10 The Aims of This Study ................................ ................................ ................................ .......... 12 2 BACKGROUND ................................ ................................ ................................ .................... 16 Interactional Sociolinguistics ................................ ................................ ................................ .. 17 Research in Online Gaming Di scourse ................................ ................................ ................... 19 First Person Shooter Video Games ................................ ................................ ........................ 22 3 DATA AND METHODOLOGY ................................ ................................ ........................... 26 Participants ................................ ................................ ................................ ............................. 26 Data and Methods ................................ ................................ ................................ ................... 27 Privacy Concerns ................................ ................................ ................................ .................... 29 4 FEATURES OF CALLOUTS IN GAMEPLAY DISCOURSE ................................ ............ 31 Relevant Information ................................ ................................ ................................ .............. 31 Abbreviation and Special Terms ................................ ................................ ............................ 34 Repetition and Rapid Speech ................................ ................................ ................................ .. 37 Intonation ................................ ................................ ................................ ................................ 39 Summary ................................ ................................ ................................ ................................ 41 5 MANAGING PARTICIPATION ................................ ................................ ........................... 46 Participation Frameworks ................................ ................................ ................................ ....... 46 Response Cries ................................ ................................ ................................ ........................ 48 Rate of Speech and Repetition ................................ ................................ ................................ 50 Intonation ................................ ................................ ................................ ................................ 52 Summary ................................ ................................ ................................ ................................ 56 6 CONCLUSION AND LIMITATIONS ................................ ................................ .................. 60 APPENDIX
6 A TRANSCRIPTION CONVENTIONS ................................ ................................ ................... 63 B TRANSCRIPTS ................................ ................................ ................................ ...................... 64 LIST OF REFERENCES ................................ ................................ ................................ ............... 67 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ......... 69
7 LIST OF FIGURES Figure page 4 1 Pitch contour of Excerpt (3) Line 3. ................................ ................................ .................. 44 4 2 Pitch contour of Excerpt (6) Line 5. ................................ ................................ .................. 45 5 1 Pitch contour of Excerpt (10) Line 2. ................................ ................................ ................ 57 5 2 Pitch contour of Excerpt (10) Line 3. ................................ ................................ ................ 58 5 3 Pitch contour of Ex cerpt (10) Line 9. ................................ ................................ ................ 59
8 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requi rements for the Degree of Master of Arts CONTEXTUALIZATION CUES AN D TEAMWORK IN ONLINE VIDEO GAME GAMEPLAY DISCOURSE By Luke Breland December 2012 Chair: Diana Boxer Major: Linguistics This study provides an interactional sociolinguistic account of how gamers playing team based first person shooter (FPS) video games s uccessfully manage participation and cooperation across shared virtual environments. Recorded utterances of gamers during online gameplay are subjected to discourse analysis in order to answer two questions: 1) what features clearly mark utterances in vide o games as cooperative gameplay relevant communication; and 2) how do gamers take advantage of these features as tools for managing participation and cooperation ? The recorded interlocutors are players of two popular FPS titles on a home video gaming conso le: Halo : Reach and Battlefield 3 TM In order to answer the first question, recordings are se arched for the presence of call outs, a term used by online gaming communities to refer to distinct utterances that communicate information and commands relevant t o the ongoing video game match. Call outs are compared with other utterances to uncover potential contextualization cues and other properties of their use This study shows that common cues include repetition, rapid speech, relatively flat intonation, and t he use of abbreviated and specialized lexical items. The second question is pursued through discourse ana lysis of selected recordings, drawing on my own expertise as an active member of communities of practice associated with FPS video games.
9 The linguisti c and social interactions between players during gameplay are a regular part of the lives of an ever growing population, but these particular interactions are rarely the focus of sociolinguistic inquiry. This study contributes to filling the gap in socioli of these increasingly relevant yet understudied interactions.
10 CHAPTER 1 INTRODUCTION The use of multiplayer video games for entertainment and socialization is becoming increasingly prevalent as more people have access to high spe ed int ernet and video game consoles. Social intera ctions in gaming however, are currently underrepresented in linguistic literature This study attempts to apply the interactional sociolinguistics paradigm to a relatively novel but growing interactional c ontext: the communication of gamers in a shared virtual environment and disparate physical environments d uring goal oriented gameplay without a face to face component of interaction Th is research paradigm is primarily realized through an investigation of how contextualization cues (Gumperz 1982) are utilized as a multi purpose resource by interlocutors during online gameplay of popular first person shooter (FPS) video games. Video Games as a Common Setting for Interaction Video games are increasingly rele vant to the average person. According to the most recent results of a survey released by the Entertainmen t Software Association (ESA), 49 % of househo lds in America have at least one console for gaming; those households that fall in this group actually cont ain two consoles on average; 62% of gamers play with others; the average gamer age is 30 years old; and the average adult gamer has been playing video games for about 14 years (2012) The ESA also reports industry sales data from the NPD Group (formerly Na tional Purchase Diary) indicating that the gaming industry was responsible for about $24 .75 billion in revenue for the year 2011 Due to rapid increases in technology, gaming is swiftly becoming one of the dominant forms of entertainment in American life; where as gaming once required both a television and a dedicated game console along with games purchased on
11 physical media a new generation of smartphones, tablets, and other devices has brought ever more flexibility to increasingly complex games. Video game s and gamers, perhaps unsurprisingly, are increasingly the focus of research and discussion in many bran ches of science. McGonigal 2011 reports the application of video games and their design to real world problems, where researchers in the applied science s use games can run simulations and test models faster and more effectively than high powered supercomputers. Gee argues that socialization into the discourse of vide o games, and participation in their communities of practice, requires and develops skills that are desirable in the job market (2003) A dissertation by Soares Palmer has even investigated participation in gameplay discourse as a supplement to foreign lang uage learning (2010) So it is no surprise that online gameplay and interactions are beginning to attract the attention of linguistic researchers. As discussed in Gee 2003 the process of learning to communicate and cooperate properly in the d iscourse of a video game requires skills that translate into the e ducational and business worlds. Unfortunately the state of research on social interaction in video games is such claims have not seen widespread systematic investigation, and as a result there is not much definitive empirical data to corroborate them However, IBM commissioned a research team to conduct a study investigating the types of leaders and attributes that will emerge in the business world as the landscape shifts toward increasingly vi rt ual and distributed businesses. R eeves & Malone 2007 is a report to IBM which foun d that online video gamers are uniquely suited to leadership and collaboration in fiercely competitive virtual environments in other words, in the business world of the ver y near future.
12 While t his comparison between business and video game environments may seem overly simplistic, Reeves & Malone 2007 evaluated online gamers according to the Sloan Leadership Model developed at MIT and found that these gamers displayed all as pects of leadership expertise delineated by the model but with a different distribution of emphases Their conclusion supports claims in Gee 2003 stating that video games are a key provider of opportunities to practice the kinds of leadership skills that are becoming more vital in the real world. Rather than working or studying, however, gamers are engaging in this process for entertainment. When gamers interact and collaborate in FPS games, their involvement may be developing skills that confer real world benefits. It is only natural, then, that sociolinguistics should endeavor to illuminate this understu died community with investigation of the processes that underlie the collaborative efforts of the participants. The Aims of This Study The goal of this st udy is to analyze communication during online matches of competitive FPS games. Of particular interest are those utterances between teammates that are geared toward cooperation and signaling threats to the immediate goals and interests of the team and indi viduals on the team; these types of utterances are typically termed callouts by competitive an d team oriented players, and their proper use is seen as an essential skill of a competent player. Given that callouts tend to be associated with combative encoun ters with the opposing team, a requirement of these utterances is that they convey appropriate information such as the number and location of enemies and which weapons they may be carrying in a very short amount of time. But are utterances related to gamep lay produced in the same manner across different FPS titles? If so, this may be a basis for confusion, as some of the most popular FPS games currently on the market offer vastly different gameplay experiences and require players to work toward goals in com pletely different manners. How do gamers know when they are being given a
13 callout, and when they are just hearing a teammate complain about the manner of his death? How do callouts help gamers effectively manage participation and activity in a shared virtu al environment? Callouts are of central importa nce to this study, so I will briefly delineate their form and function here. There is no truly universal formal definition of a callout, but a s they are understood in various FPS communities callouts communi cate information relevant to the h. This working definition can be further limited by stating that callouts are used during immediate and ongoing gameplay events such as defending an objective or encountering an enemy play er Collaborative gameplay discourse certainly takes place when there are no immediate enemy encounters, such as drawing up a strategy for holding a defensible but this kind of talk is not necessarily g For the most part, given the general gameplay goals of FPS video games, callouts frequently communicate information about enemy movements, actions, and locations ; callouts are not necessarily limited to these categories, however, and one might reasonably expect some callouts to take the form of ability to come out victorious in a given encounter One recognized hallmark of callouts in the gaming community is that they generally use specialized or s hortened terms to refer to landmarks in the virtual space, so that t eam members can swiftly and accurately reat in relation to themselves. The exact terms used are not universal across all play ers of the same game; s ome organized highly competitive gro ups of gamers ( typically called clans ) may develop and adopt a set of terms unique to their group. This lack of uniformity does not prevent randomly matched players from communicating successfully among themselves, som etimes using a variety of terms for the
14 same referent in virtu al space; this may be a product of the socialization into the norms of the community of practice around a given video game, though investigatio n of this process is not within the scope of this s tudy. Finally, it is worth mentioning that on teams where communication (and giving callouts) is ex pected, a form of uptake of a callout is also expected. If a player is warned about an enemy try ing to flank a nearby position it is expected that the playe r will attempt to neutralize the threat, and may also give a verbal acknowledgement of the callout. This uptake requires teammates to perceive the nature of the illocutionary force projected by callouts that is, a call to action that is not necessarily exp ressed in t and the actions they perlocutionary effect (Searle 1975). The term uptake ptake of a g iven callout is usually apparent in the audiovisual data recorded for this study and monitoring player uptake and response s can provide support for whether a given utterance was intended (or at least interpreted) as a callout I describe the actions of av atars in the virtual environment when relevant to my analysis. The concepts of frames and contextualization cues from Goffman 1981 and Gumperz 1982 respectively, offer a potentially fruitful method of analysis in investigating t his form of communication This study examines utterances to determine possible contextualization cues that frame This study draws heavily from concepts in interactional sociolinguistics and applies those concepts in keeping with the examples of past research that has shown success in analyzing gameplay interactions under this descriptive framework (e.g. Aarsand & Aronsson 2009, Keating & Sunakawa 2010, Piirainen Marsh 2010, Piirainen Marsh & Tainio 2009). In the process of analyzing video game interactions, t his study sheds light on the
15 nature of contextualization cues in a multiplayer online gameplay environment, and how players use these cues as interactional resources for teamwork. An analysis of communication during the online play of two different FPS gam es will attempt to answer two research questions: 1) h ow do gamers contextualize their utterances to clearly mark callouts; and 2) h ow do gamers manage the participation and non participation of themselves and others in ongoing gameplay events? To answer t hese questions, this study undertakes a discourse analysis of audiovisual recordings of first person shooter gameplay, relying heavily on the interactional sociolinguistics framework of analysis.
16 CHAPTER 2 BACKGROUND Within the tradition of sociolinguist ics, there is a rich and diverse body of work that examines interactions between many different types of interlocutors in different contexts. Even computer mediated discourse is seeing a great deal of attention. However, a review of the available literatur e quickly shows that oral communication between online gamers is almost completely and entirely unstudied in the most prominent linguistic academic journals with a only a few scattered online only journals that regularly treat the subject of video games and communication, e.g., Game Studies Journal of Virtual Worlds Research ( see ell, Grant ham, Wong, Workman, & Wang 2010, Wright, Boria, & Breidenbach 2002). Publications from both lesser known and well known print journals do focus on communication in t he context of gameplay, but typically with the focus on interactions between interlocutors sharing a physical enviro nment (e.g. Aarsand & Aronsson 2009, Keating & Sunakawa 2010, Piirainen Marsh 2010, Piirainen Marsh & Tainio 2009). Lacking a descriptiv e framework that is explicitly optimized for studying this type of online collaborative communication in a virtual environment without the potential for face to frames and participation frameworks (1981), and especially the concept of contextualization cues (Gumperz 1982). I have chosen this interactional sociolinguistic framework for its potential for fruitful application to the type of data I have collected, building on the examples set by the wo rk of Aarsand & Aronsson 2009 a nd Keating & Sunakawa 2010 The rest of this chapter will briefly discuss these basic concepts and highlight their application in related studies on interaction and video games. The chapter will then clo se with a brief overvi ew of FPS games and their mechanics, to familiarize the reader with concepts that gamers take for granted when communicating.
17 Interactional Sociolinguistics The interactions that occur between gamers while playing video games are a rich resource for a rese arch focus on interaction and contextualization. Interactional sociolinguistics is a framew ork espousing such a focus, stemming from important contri butions from Goffman 1981 and Gumperz 1982 among others. Goffman 1981 discusses frames by describing how pe ople develop interpretive schema through repeated and shared interactions and contexts and how they use these schema as frameworks for interpreting messages. The intended frame of an utterance is a vital component y on frames to correctly interpret the intended meaning and force of the utterance. Goffman also introduces the concept of a participation framework, which can be used to describe the shifting positions taken by participants in social interaction. These re lationships between messages and their context are not expressed overtly in explicit statements, however. The linguistic information people transmit and receive during interaction is underspecified, but they must have a way of expressing or understanding i ntended frames and meanings. Good players generally have a finely honed sense of situational awareness, gleaned from the information rich virtual environment of FPS games. Integrating their knowledge of gameplay with awareness of the flow of the match and the capabilities of their teammates can be essential. Part of this situational awareness must include the activities of their teammates, and so keeping track of changes in the participation framework becomes an important skill. This study shows that player s use features of their speech to signal their participation (or non participation) in ongoing gameplay events. Writing about how people make sense of utterances when it can seem that the linguistic signal itself contributes so li ttle information, Goffman state s:
18 A presumed common interest in effectively pursuing the activity at hand, in accordance with some sort of overall plan for doing so, is the contextual matrix which renders many utterances, especially brief ones, meaningful. (1981:143) Talk situated in an activity certainly relies on context for proper interpretation of meaning and intent. A string of utterances between teammates playing a video game, taken completely out of context and presented to a person unfamiliar with gaming, will lose much if n ot all of their intended meaning. Video game discourse is far from unique in this regard; context plays a critical role in eva luating meaning during much communication A central component of interactional sociolinguistics is analysis that takes into accou nt the contexts, activities, and communities within which discourse is situated. Not all talk situated in a given activity need necessarily relate to that activity directly, however, and there is still much about the meaningful context of an utterance that interlocutors appear to grasp without the aid of environmental or linguistic information. Gumperz 1982 introduced the concept of contextualization cues, which are linguist ic and extralinguistic features such as lexical choice and prosody, to describe the apparent mechanism by which interactants correctly i nfer the frame of an utterance. According to Gumperz 1982 contextualization cues signal a set of contextual knowl edge or activity type by which an utterance can be int conc ept of contextualization cues is closely linked to that of frames (Goffman 1981) The critical aspect of this paradigm is that the appropriate contextual presuppositions are signaled by surface features of the message, and not directly by the semantic mean ing of the message or indirectly through the activity or setting the message is situated in. The correct interpretive frame must be constructed between interlocutors; one must be able to supply the correct contextualization cues, while the other must have sufficient knowledge to know which interpretive frame is being signaled by the cues. Gumperz illustrated the importance and function of contextualization cues with examples of miscommunication that arise from misunderstandings of contextualization
19 conventi ons often because interlocutors from different communities of practice may use contextualization cues that signal different interpretive frames Gumperz takes care to note that these miscommunications do not register with the interlocutors as linguistic i ssues but rather as improper social or attitudin al behaviors. In a sense, the interlocutors are correct: a fluent knowledge of the appropriate contextualization conventions for a community requires proper socialization into that community. use of miscommunicati on to illustrate and explain relevant c ontextualization cues is not the only method for uncovering and accounting for these cues. The data I capture do not contain any events featuring miscommunication between participants, so the contextuali zation cues ac companying a given utterance need to be evaluated effect on other participants Callouts, the primary type of utterance under analysis in this study, are intentionally produced with an intended effect in mind: when one play er alerts his teammates to an enemy threat, that player is likely doing to in order to 1) cause teammates to become aware of the threat and 2) recruit the assistance of teammates in dealing with the threat. The interaction text, and meaning is an important focus of the interactive context, and effect, however, is central to speech act theory (Austin 1965 ; Searle 1975 ) Comparison betwee n utterances with similar linguistic or semantic content but different effects is one strategy employed in this study. Research i n Online Gaming Discourse Wright et al. 2002 provide an early discourse analysis of messages produced in networked multiplayer matches of a FPS game Counter Strike Wright et al. 2002 propose s a rough taxonomy of game talk seen online matches of the game ; the discussion of in game interaction does not include talk based around c ooperation or teamwork, but it demonstrate s that
20 ma stery of the linguistic features particular to Counter Strike d iscourse served as a distinguishing characteristic that set the inexperienced players apart from the veterans. Certain conventions of communication in a particular game, then, can serve as an index of socialization Ginet 1996). is one of the few studies that focus wholly on discourse with participants who are not physically co present but instead interact thro ugh the medium of a virtual environment In that study, all analyzed discourse is in the form of text based chatting. Chat messages, rathe r than vocal speech, also form the text of analysis in other studies involving video games ( e.g. Soares Palmer 2010 S teinkuehler 2006, Wright et al. 2002 ). The authors analyzed messages produced during an interactive, cooperative, competitive, multiplayer puzzle game developed specifically for the study. They found that messages produced during gameplay sessions tended t o differ from social usages, and were generally serious, short, and not particularly well formed. These messages were intended to fulfill a cooperative function, and the authors report that such communication was associated with success in their results. I n this sense bear some similarity to the callouts produced during online FPS gameplay in this study 2010 utilized text based rather than vocal data, however, so one should be careful wh en dra wing parallels between the two Nevertheless, t hese features do correspond roughly with features of callouts discussed in this paper. The interactional sociolinguistics (IS) framework has been helpful in understanding many instances of social interaction, and has also been applied to interactions involving v ideo games. Aarsand & Aronsson 2009 appl ies another concept from Goffman 1981 response cries in the analysis of how joint attention and cooperation are secured by participants playing video games.
21 In t heir data, r esponse cries offer one method of managing participation frameworks, by signaling participation through attention to the video game being played. Their analysis also extends to the level of participation of other people who were co present duri ng the play of video games. Those who were playing video games had several methods of signaling inclusion in or exclusion from the activity frame. Other co present interactants could in turn monitor their attention and participation in accordance their des ired degree of (non)involvement. How gamers signal their participation in and attention to ongoing gameplay events without the interactional benefits of a shared physical environment will be a crucial aspec t of my analysis; this is a logical extension of t he investigation of Aarsand & Aronsson 2009 In their investigation of collaboration in gam eplay, Keating & Sunakawa 2010 introduce the term participation cues which take their name from the application of contextualization cues to participation framework s. Participati on cues in their study were an integral part of the discourse that took place during networked gameplay sessions of a Massively Multiplayer Online Role Playing Game, or MMORPG Analyzing these participation cues enables them to track the chan ging landscape of the participation framework of a group of people engaged in a cooperative gaming session. Since these gamers are simultaneously sharing the same physical and virtual environments, the management of participation across, within, and betwee n environments is a complex task. The authors found that rapid speech and repetition were two common participation cues used to engage the attention of teammates when a pl ayer was in danger. While I do not fully terminology, my study ext end s and support their work along with the work of Aarsand & Aronsson 2009 by examining contextualization cues as a resource for gamers to manage collaboration and achieve shared goals.
22 First Person Shooter Video Games In this section, I provide some backg round information for the particular games used in the analysis and some differences in their style s of play. This information is relevant to my analysis, and I provide it with the understanding that all readers may not be familiar with the knowledge and c onventions associated with gaming. This information forms a large part of the contexts and assumptions that players operate under while playing these games. As a life long member of communities of practice associated with video games, and as a gamer who ha s played the games under analysis since their release, I am qualified to write as an expert o n the characteristics of these games and the communities of pract ice that surround them. H owever, a full ethnographic account of the characteristics and practices associated with these games would be overly cumbersome and altogether unnecessary for the scope of this study. The data under analysis have been captured from the gameplay of two different games, both within the same genre: Halo : Reach (hereafter shortene d to Halo ) and Battlefield 3 TM Both of these titles are prominent and highly popular members of the first person shooter genre Battlefield 3 TM was the fifth bestselling video game of 2011 according to the ESA (2012) and as a result, they share many basi c similarities. They also have key differences, however, which affect the style and conte nt of the callouts in each game. First person shooters, commonly abbreviated to FPS in both written and spoken English, because the player assumes the point of view of the avatar that he or she controls and because these games contain gameplay centered on targeting and shooting enemies with an array of firearms. Although this gen re could be further subdivided and dissected ad nauseum it is enough to know that mod ern FPS games feature relatively realistic graphics and physics, online gameplay and intensely competitive (and predominately team based) gameplay. FPS video
23 games can v ary greatly in some respects within their genre, and the two most important differences for this study are briefly covered below. An important feature of FPS games that must be mentioned briefly is the health system. Simply put, in any shooter there is a c ertain amount of damage a player can take before dying. The amount of damage one can absorb varies across games, and the amount of damage one deals also varies across both games and the weap ons used within a game. A typical feature of modern shooter games is a health system where health regenerates over time if the player first takes nonfatal damage and then experiences an uninterrupted period of time where no damage at all is taken. The health system of a given FPS title (in coordination with the character istics of the weapons used) typically dictates the length and pacing of in game co mbat encounters between players. Halo and Battlefield 3 TM differ in their health systems. Halo a futuristic science fiction shooter, combines a health system with a regene rating shield system; no damage can be taken by the player until the shield has absorbed so much damage that it collapses. Once the shields are gone, a player may survive used and placemen t of the shots Shields can regenerate quickly, and a special ability available in some matches allows a player to temporarily become completely invulnerable (and immobile). Battlefield 3 TM on the othe r hand, is a shooter in a setting that closely emulate s the sights and sounds of modern warfare Weapons used in Battlefield 3 TM are directly modeled after current firearms used by real militaries across the globe. Players do not benefit from any regenerating shields, but there is a health system in Battlefie ld 3 TM Players can take up to 100 points of damage before death, and a gravely injured player who escapes combat intact will find his health automatically (but slowly) regenerating after a short delay. Damage in Battlefield 3 TM is
24 governed by a rather com plex set of factors, and a full analysis of the damage models in these games is certainly beyond the scope of this study. It is enough to state that combat encounters in Halo typically last several seconds longer than those in Battlefield 3 TM where death at close range can occur seemingly instantaneously. This statement about the pacing of combat in these two titles is not borne di rectly out of my o pinion. A mong online communities that discuss various FPS titles it is a well known property of these games that is usually taken for granted. Any debate typically centers around which style of gameplay is preferable, not whether Halo is slower than Battlefield 3 TM Another important difference between these games is their respective map s The term map in FPS games refers to a pre constructed (not random) virtual space within which a match can take place. Games typically offer many dif ferent maps of varying sizes, types of terrain, and themes. One sali ent characteristic of maps in Battlefield 3 TM is that the ma ps tend to be very large, and they are much much larger on average than th ose in Halo I argue that these often extreme size differences lead to differences in the strategic utilization of contextualization cues in callouts across these two games Callou ts are of central importance to this study, so I will briefly delineate their form and function here. There is no truly universal, formal definition of a callout, but as they are understood in various FPS communities callouts communicate information relev ant to the during the course of a matc h. Observation and experience suggests that t his working definition can be further limited by stating that callouts are used during immediate and ongoing gameplay events Collaborative gameplay discourse c ertainly takes place when there are no immediate enemy encounters, such as drawing up a strategy for holding a defensible position
25 For the most part, given the general gameplay goals of FPS video games, callouts frequently communicate information about enemy movements, actions, and locations; callouts are not necessarily limited to these categories, however, and one might reasonably expect some callouts to ta encounter. One recognized hallmark of callouts in the gaming community is that they generally use specialized or shortened terms to refer to landmarks in the virtual s pace, so that team members terms used are not universal across all players of the same game, however. Some organized highly competitive groups of gamers (somet imes called clans ) may develop and adopt a set of terms unique to their group. This lack of uniformity does not prevent randomly matched players from communicating successfully among themselves, sometimes using a variety of terms for the same referent in v irtual space; this may be a product of the socialization into the norms of the community of practice around a given video game, though investigation of this process is not a focus of this study.
26 CHAPTER 3 DATA AND METHODOLOGY Participants Fifteen total p articipants appear in the transcripts reported in this study. Participants ranged from random and unknown gamers to gamers I play with both regularly and infrequently For unknown gamers, there is no systematic and reliable method in the software or hardwa re of the game for screening my potential teammates by native language, region of origin or current residence, gender, or age. Given the nature of the types of matches played online, where there are only seconds from team assignment to the start of the mat ch, this study did not attempt to collect any sociolinguistic data about its participants. Even p articipants I play with regularly are not known by me personally, not even their given names. All that can be reasonably determined is that they are adult male s, with language backgrounds including predominantly American English varieties although Canadian English and Austr alian English speakers were also recorded Unfortunately, personal experience suggests that there is an extremely low likelihood of encounte ring vocal female gamers in FPS games and indeed only one female gamer was recorded during the course of this study The potential for variation and inclusion of participants from multiple speech communities and backgrounds is not problematic for this stu dy, however. Those who are active members of the communities of practice under investigation are, by definition, socialized into the expected social and linguistic behaviors that characterize their community of practice, and therefore are reasonably expect ed to share contextualization cues relevant to callouts regardless of cultural or linguistic background (Eckert & McConnell Ginet 1992) Any participants who are newcomers or are otherwise not at all socialized into the expected behaviors only increase the
27 potential for recording instances of miscom munication, which Gumperz found useful in his examination of contextualization cues (1982) Data and Methods Two different FPS video game titles, Halo : Reach and Battlefield 3 TM were played on a single video ga me platform, the Xbox 360 T wo different video games with very different styles of play were used in order to show that the features reported here are relevant for the larger community of practice associated with FPS video games, and not just characterist ic of fans of one particular franchise. recorde d, th rough separate but simultaneous processes to limit the impact of sound effects on the video game on the intelligibility of participant ut terances Video of th e online gameplay and the game audio (excluding speech betwe en gamers) were recorded onto a computer using a Dazzle D igital Video Creator 100 DVD R ecorder. Utterances from participants, excluding ambient game sound, were recorded into the computer program Audacity through a separate input For excerpts where visual data were required for maximum clarity of analysis, t he video and audio files were combined. In these cases, editing was only done to align the audio and video, and to redu etc.) in order to minimize its impact on the intelligibility of the speech of the participants. No editing was done to alter the audio data from the interlocutors, as these were recorded through a separate channel from the game audio. All speaker int eractions are intact and unedited in the recordings and no adjustments were made that would change the relative speed, volume, or quality of the utterances. To limit the impact of the on speech and behavior and to prevent my utterances as the primary researcher from guiding and shaping the overall level and tone of communication in these video game matches, I turned off the microphone on
28 my headset under various pre tenses (receiving a phone call, my wife had just gotten home, etc.) at the onset of each match sometimes before and sometimes just after recording had begun. Overall, recordings were made of about 24 online matches across both games totaling just over fou r hours of recorded audiovisual material. Some matches were virtually overflowing with communication, while others contained long periods of silence. Still others contained interlocutors whose microphones distorted their speech significantly, frequently ma sking the contributions of other speakers. In a few isolated instances, the recording equipment had to b e adjusted mid recording so that interlocutors could be heard clearly. These setbacks only affected a small portion of the data, leaving plenty of analy zable data for the purposes of this study. Not all r ecorded da ta a re transcribed A great deal of the recorded talk was not relevant to interest to this analysis, a nd only relatively clear and relevant interactions were chosen for transcription. This excluded interactions such as those containing a great deal of simultaneous speech, both for ease of transcription and for suitability of analysis. While issues such as speaker turn management during simultaneous speech are of in terest in sociolinguistics and Conversation A nalysis, such issues are not the focus of this stud y. Transcriptions follow a modified version of the Jeffersonian system of notation, which is com monl y employed in Conversation A nalysis. For transcription conventions used in this study, see Appe ndix A. The transcribed data were subjected to qualitative analysis i n keeping with studies in the interactional sociolinguistics paradigm, with particular atten tion paid to c ontexts and features relevant for the identification and analysis of contextualization cues. For some recorded data, the computer program Praat (Boersma & Weenink 2012) was used as a tool to
30 data being cons idered public behavior (IRB Protocol #2011 U 1155 ) As an extra precaution however, t he gamertags of participants represented in the transcribed data have been abbreviated anonymity.
31 CHAPTER 4 FEATURES OF CALLOUTS IN GAMEPLAY DISCOURSE Recall that described cooperative discourse in an online video game as serious, short, ill formed, and otherwise different from more social utterances. While these features describe text based co mmunication in t hat study, callouts can also be generally described in these terms in the audio data collected for this study. In this chapter I present transcribed data from two FPS video games to give descriptive examples of various features associated with callouts. I argue that these features can act as contextualization cues to signal a contextualization cue per section, I do not necessarily argue that these features nec essarily act as cues individually. In fact, Auer states that contextualization cues most effectively signal a frame when bundled together (1992) Judgment of whether an utterance can be pr operly labeled as a callout largely stem s from my own knowledge as a n expert member of the relevant communities of practice and further confirmation can be reached from any observed effect on the speech or behavior of teamm ates. Questionable cases are marked as such, or are not considered altogether Relevant Information Callouts mus t, of course, s goals and to the gameplay event that is the subject of the callout. Information that is typically relevant to an encounter with an enemy might include enemy player location, numbers, and equipment; the status of the enemy player(s) or the player issuing the callout (alive, dead, low health, etc.) ; and the apparent actions or intentions of enemy combatants. The brief excerpts in this section illustrate the types of information that can be p rovided in callouts.
32 Excerpt (1) was recorded in Battlefield 3 TM and features only one speaker. The speaker gives callouts as he works to accomplish an important objective for his team (in this case, safely a bomb on it). The information provided in each callout is relevant to his progress in accomplishing this objective, and such information own actions and statu s relative to the objective. EXCERPT (1) 1 AC : Got one on A 2 3 Aw this scope is awesome 4 ((pause)) 5 6 ((pause)) 7 Planted would be useful to his teammates in the context of this particular gameplay event. Useful information is contained in the clearest callouts which are marked with an arrow The information he communicates in th ese lines might be vital to nearby teammates in the event that he fails his assault on the objective. He details the amount and location of enemy re sistance at the objective A in L ine 1, the subsequent removal of the enemy threat in line 2, and status of the objec tive A and his actions toward it in L ines 5 and 7. The following example Excerpt (2) recording during a match of Halo is a little more complex. Three interlocutors give and respond to callouts, and these callouts report a variety of information. Enemy numbers, position, status, movement, and even weaponry are reported. The interlocutors also report their own status and level of threat, as well as the status of a hotly contes ted weapon. It is important to note that all of the callouts are relevant to ongoing events as
33 in the excerpt are to my gamertag and avatar which is why I have not s hortened it in this text; as I stated in the methodology section, my microphone is off and callouts directed to me are not responded to verbally. EXCERPT (2) 1 CK: They got a snipe at high snipe. 2 (7.8) ((Called out enemy is killed)) 3 P H : High snipe (is ) down. 4 HR : 5 6 P H : [(Gus) 7 8 To your left Gus 9 ((Gus and P die)) 10 I just lost snipe Every utterance produced in this highly communic ative exchange is a callout. In L ine 1 CK reports an enemy threat with a powerful weapon at the location where that weapon is typically found; the following seconds are silent as CK and P H uptake of the callout is initially signaled by his actions in supporting CK; as he reaches the downed sniper, he confirms to the rest of the team that the threat is elimina ted in Line 3 In L ine 4 HR reports the location and status of two enemies (one with low health) that have j ust killed him; the video shows I respond to his report by attacking one of the two enemies, but I am oblivious to the threat coming from my left even as P H reports it in L ines 6 7 and again, more urgently, in L ine 8 E xcerpts (1 2) demonstrate the wide va riety of information that tends to be communicated through callouts. These excerpts also show that callouts report ongoing or just completed events information which is at times acted on immediately or warn of immediate
34 threats. The information contained i n these utterances is relevant either to other (possibly nearby) teammates, or to the overarching objecti ves of the team (as in E xcerpt (1)). Abbreviation and Special Terms The life of an avatar in the two video games recorded for t Leviathan A player giving a callout while under attack is not necessarily just reporting information to keep the team up to date; the callout itself especially when it relays a location may also function as an indirect request for any nearby players to lend their assistance. Given how quickly a battle can be won or lost in FPS video games, an embattled player may not have much time to relay precise information. nell et al. report brevity as a prominent characteristic of the cooperative gameplay oriented discourse that occurred in their study, and the authors comment that short and effective communication likely arose from time constraints on the teams in the stud y, which required teams to complete a virtual puzzle in a certain amount of time (2010) This abbreviation in communication proved effective; the authors report that st 2010:31) Another source of abbreviation can be found in specialized terms used by players in callouts. As mentioned in Chapter 2, skilled players who are active members of a community of practice related t o a game are expected to be familiar with the maps found in that game. Part of this familiarity involves knowing and using certain terms that, in the context of gameplay on a particular map, refer to a specific place or area within that map a much faster a nd more efficient method of relaying information than a detailed description of the surro This section contains transcripts that show such abbreviation and specialization in lexical and syntactic forms.
35 The following E xcerpt (3) is brief, but gives a clear example of the kind of effective abbreviation used to communicate information in callouts. This was recorded during a match of Halo and the callout reports the arrival of two enemies at a specific location on the map. Thi s callout is characterized not only by dropped syntactic elements, but also by a two word term that is used to refer to this precise location. EXCERPT (3) 1 P H : 2 Ah crap 3 R M : U h : one top yellow (.) two top yellow R M ine 3 is the only callout in this excerpt, and it contrasts tellingly with P H (non callout) utterance s As two enemy players attack and kill R M he narrates their arrival and the relevant location. He does this in a highly abbreviated ut terance, which gives a simple number and location as t he enemies arrive sequentially. applies only to a specific room on this particular map a room on one of the highest floors of the virtual space, connected to one of the lowest floors by a tube filled with glowing yellow green energy; a player stepping into the open end of the tube at the bottom will be swiftly lifted to the Both rooms are pinpointed precisely by their designations, characterized by its relative height in the virtual space and the yellowish glow that spills out from the ends of the one way lift. Although it was not captured in my recordings, I have also heard the same two rooms referred to R M the match, his teammates gather at the lower end of the lift (bottom yellow) immediately following his utterance. Their reluct ance to rush to his aid likely stems from the fact that R M has already died at this point, and also because the narrow unidirectional lift empties into a rather small room and is a natural chokepoint for enemies to defend.
36 The utterance in L ine 3 would lik well formed American English, especially when presented devoid of context. This contrasts with P ine 1 which is not a callout and does not appear to be directed at any particular player ; he simply appears to be chastising himself for forgetting to equip the jetpack ability, which is favored by P H on his avatar. P ine 1 appears rel atively well formed, without so much as a drop ped pronoun to shorten his message. Specializ ed terms for referencing locations seem a little less abundant in recordings of Battlefield 3 TM than they do in recordings of Halo This may be because Battlefield 3 TM maps are larger by a great magnitude, and meant to represent urban or natural environme nts realistically; as a result, there might be many nondescript buildings with similar layouts in a small area or scattered all over the map. Prominent, colorful, otherworldly features are commonplace in the science fiction futuristic virtual universe of H alo ; this, combined with the relatively smaller map sizes, contributes to a relative ease of labeling specific places in each map. In Battlefield 3 TM are much less common. Player a ttention tends to gravitate toward the objectives, which always appear in the same place on a given map and are marked by br ightly colored, floating letters; these letters are always visible to the player regardless of distance from the objective or interv ening objects which might otherw Given the visual prominence of these objective letters and the importance of the locations that they mark, it is common in my recordings and my experience for players to use the letter marking an objective as a reference to th This usage is illustrated in E xcerpt (4): EXCERPT (4) 1 LV: [Losing D.
37 2 ? ?: [( ) 3 MD: 4 (1.0) 5 >Oh oh< D, D as in Delta, gotcha. 6 QM: Where where where 7 (1.0) 8 I see him. 9 ((Gun firing)) 1 0 Charlie, Echo, etc.). This is not al together u ncommon, especially during matches in which active objectives include B, C, D, and E all letters that might sound identic al due to channel distortions and background noise. In L ine 3 MD actual 1) as referring to B; after a moment he realizes h is mistake, verbalized in Line 5. MD also lets his teammates know that he thinks an enemy player is in the same objective area (Bravo), but he refer to a general area surrounding the actual objective a flagpole. The excerpts in this section provide a small sample of the ways that utterances can be abbreviated and marked with special terms during callouts. Novel lexical items, or n ovel uses of existing items, are not limited to video game communities of practice. Indeed, one would expect a given community of practice to appropriate or innovate words according to its needs. What is unique about this particular case is that the use of specialized terms, especially those that index a playing community. When these specialized terms are used because the speaker wants his or her utterance to be taken a s a callout, then they are being used as contextualization cues. Repetition and Rapid Speech Repetition and rapid speech can potentially act as contextualization cue s, signaling an utterance as a callout. Keating & Sunakawa describe rapid speech (usually a ccompanied by
38 Keating & Sunakawa view these as participation cues in their terminolog y, their study also of cooperat ive communication during the play of video games finds these cues during ong oing threatening encounters. My dat a agree with theirs about the co occu rrence of threatening events and utterances with repetition and rapid speech. While I do not adopt their ter minology of participation cues my analysis does not differ significantly from theirs (discussed further in Chapter 5). Excerpt (5) illustrates this theme of rapid speech and repetition in conjunction with immediate threats. In this example from Battlefiel d 3 TM the interlocutors are defending a lone location and actions, and his teammate responds to his callout swiftly Notice that when repetition and relativ ely rapid speech occur in this example, the players are in a threatening environment: EXCERPT (5) 1 GW: (Spawn) on me >spawn on me< 2 ((GW dies)) 3 4 AND shooting an LMG down that hall[way. 5 AN: m< (.) m < 6 GW: Nice. GW quietly utters Line 1 just before being killed by the enemies he su bsequently describes in Lines 3 4 The urgency communicated by his tone and repetitio n in Line 1 is completely absent in 3 uptake. callout, and the repetition and rapid speech are cues that he is participating in an ongoing threatening e vent of some sort. This is confirmed by the video recording of the event, which shows that AN is still killing e nemy players in the area even after his callout i n line 5. This example supports
39 description of pressured speech during threatening encounters in video games (2010) Intonation This final section of C hapter 4 describes the use of inton ation in callouts. The use of audio data, is not constant. This particular aspect of callouts also merits some acoustic analysis beyond my own impression. This was achieved by tracking and comparing pitch levels within and between utterances in the program Praat; data and visual representations of these pitch levels are provided in this section in addition to the transcripts of the utterances themselves. Try ing to compare intonation across players would likely be somewhat problematic at the very least due to potential idiolectal variation between different players. I have observed in the data that callouts are at times delivered in a flat intonation, relativ ; this relatively narrow ranged prosody also seems more likely during the tense and dangerous situations described for repetition and rapid speech For an example, first I return to E xcerpt ( 3). Recall that in this excerpt, player RM quickly gives a simple callout narrating the sequential arrival of two enemy players. The pitch of his voice during the callout does not appear to vary greatly: EXCERPT (3) 1 PH: 2 Ah crap 3 RM: Uh: one top yellow (.) two top yellow RM provides this callout just as he is confronted (and overwhelmed) by two enemies, but he does not make use of repetition or relatively rapid speech. He does, however, supply a callout characteriz ed by markedly flat pros ody; his utterance does not include a pronounced statement final fall in pitch common to English. Analysis of the utterance in Praat indicates that the
40 minimum and maximum pitch levels o ver the course of L ine 3 were 113.554 Hz and 1 45.844 Hz, respectively, meaning his pitch stayed within a range of 32.29 Hz across the utterance. For a visual representation of the pitch contour, see F igure 4 1 To get a better idea of whether this degree of pitch variation is common for this speaker, I turn to another utterance by the same speaker in a frame that does not indicate a callout. During the same match as Excerpt (3), RM makes a comment about a quick enemy encounter that has just concluded. The utterance is most definitely not a callout, and the overall intonation is much more expressive than that of Line 3 in Excerpt (3) Both the context of the utterance and its lexical content indicate that he is griping about an apparent slight by the game itself in not appropriately recognizing his contr ibutions to the preceding battle. The incident is transcribed in E xcerpt (6) below: EXCERPT ( 6 ) 1 ((PH is attacked by player from opposing team)) 2 PH: WOAH 3 (3.9) ((MI and RM fight and kill the attacker)) 4 MI: Got you= 5 RM: =Ah sure I get the assist outta that. Jee :pers. ine 5 against the enemy player who attacks PH. My recording of the incident shows that RM and MI both shoot at the enemy player at about the same time, and both players likely contribute roughly equal amounts of damage. Whether by luck of the draw or complicated calculations hidden in the e kill indicated by Line 4 a state of affairs lamented by RM. Line 5 seems to be more expressive and does not display quite so narrow a range as that discussed for Excerpt (3) L ine 3 above and pitch analysis in Praat confirms this assessment n a maximum of 193.068 Hz and a minimum of 78.664 Hz. This range of 114.41 Hz shows
41 somewhat more variation in pitch across the utterance in Line 5 than that showed by Excerpt (3) L ine 3 2 for the graphical representation of the pitc h contour in L ine 5 There are ample examples detailing such narrow ranged intonation in conjunction with callouts in the data, but I do not list them all here. Where my analysis characterizes an utterance as having flat intonation in Chapter 5, tracking capabilities. Summary Excerpts in this chapter demonstrate some of the common features associated with callouts in FPS titles. In addition to the style of deli very flat intonation, utterances mark ed by abbreviation and special terminology relative urgency communicated by speech rate and repetition callouts also tend to communicate only certain types of information, such as enemy position and status, player progress toward a goal, player status, an d ongoing events relevant to the goals of the team and of indiv idual players on the team I argue in this study that the se features act as contextualization cues, used to mark utterances for interpretation in a callout or teamwork based frame. In other wo rds, a complaint will differ from a callout in the responses of teammates. As a baseline for comparison with the callouts analyzed in this st udy, Excerpt (7) shows a collaborative exchange between teammates that is devoid of callouts and many of the features associated with callouts. The immediate and threatening nature of the gameplay context t hat typically results in a call out may have someth ing to do with the nature of the feat ures associated with callouts. When events or concerns a re either long term or past, t he style of the utterances is
42 very different. The excerpt below shows utterances that lack the abbreviation seriousness, and marked intonation seen in callouts. EXCERPT (7) 1 MI: Hey I 2 I have (.) cra zy good success when we hold this down. 3 4 And a shotgun. 5 PH: 6 RM: =Uh, >I got I got< [shotty. 7 MI: [Go up here, 8 Go up here to this little spot right here, 9 And you can sit right on top that ledge 10 11 (1.1) ((PH goes to the indicated position)) 12 RM: ((laughter)) 13 (1.2) 14 PH: Don e. 15 (1.5) 16 RM: [( )] 17 MI: With the exception of Line 6 the above exchange lacks the features of collaborative discourse and callouts from the studies and examples discussed above. Although this is a collaborative event a strategy is presented by M I utilizing the tools possessed by P H and R M M I utterances are fully realized without the heavy ellipsis Pronouns are r etained, and he even engag es in a little joking in line 17 None of the utter ances are interpreted as a call out, even though the exchange is given in the interest of obtaining term goal of holding a defensible posit ion. The key here appears to be the level of danger experienced by the team from their enemies; during the entirety of the exchange, the teammates are undisturbed and under no immediate threat. The lack of call outs here, and the abundance of call outs durin g combative situations, m ay suggest that the use of call outs is encounter based or associated with immediate and ongoing threatening events
43 The ostensible, community acknowledged purpose of giving callouts is to foster informative, cooperative communicati on among teammates so that the team will be up to date on various events and able to respond to threats in a swift, coordinated, and efficient manner, thereby increasing the possibility of winning the match. Callouts, then, are viewed by gamers as tools th While gamers may conceive of callouts as tools, what they may not realize is that they are using contextualization cues themselves as linguistic tools, shaping the flow of participation and a ctivity throughout a match. Callouts are licensed in the context of a threatening gameplay event; p layers may at times need to coordinate and signal their participation (or nonparticipation) in such e vents, and Chapter 5 will exp lore how players manipulate previously discussed features to achieve this.
44 Figure 4 1. Pitch cont our of Excerpt (3) Line 3
45 Figure 4 2. Pitch contour of Exce rpt (6) Line 5
46 CHAPTER 5 MANAGING PARTICIPATION To this point, discussion of contextualization cues and the kinds of utterances seen during online gameplay has focused on description of these meaningful devices. This is the first aim of my study, to describe the type and form of the interactional resources available to interlocutors in a virtual environment. The next important step is to examine how these resources are meaningfully applied in interaction and how they function in shaping the discourse and actions of the participants. Participation Frameworks contributions to studying interaction is the concept of participation frameworks (1981) according to Goffman; especially in contexts of group interaction, there will be bystanders and participants of various degrees. The pa rticipation framework of a given interaction co constructed by its interactants, and ever changing helps identify the status of each person in relation to that interaction. Not all are equal participants, and this can change over the course of the interact ion. This actually has implications for this study. When g amers use cooperative communication, they can draw on their knowledge of contextualization conventions in a strategic manner for various effects participation in a gameplay event. What happens when a player encounters an immediate threat that is not relevant to his teammates? Goffman distinguishes between participants who are or are not ratified in the various interactive relationships contained wi th a participation framework (1981) A person may become a ratified listener, for example, by having an utterance addressed to them. Gamers exploit the assumptions borne out of contextualization conventions so that the features of their speech
47 updates thei r status within the participation framework of the team during a match. During Excerpt (8) recorded on one of the larger maps of Halo the video shows H A driving in the y, separated from his teammates. E nemy players begin to spawn (appear in the virtual space, usually a few seconds after dying ) around his lo cation and begin targeting him. Other members of the team do have a visual of H A and his predicament, but are too far away to assist him. The ensuing exchange contains complaints, laugh ter, and encouragement, but nothing t hat would count as a callout. EXCERPT (8) 1 ((H A driving and surrounded by enemy players)) 2 P H : 3 H A : 4 (2.2) 5 P H : There we go HA 6 Drive like I know yo u can. 7 H A : ((laughter)) ( ) 8 ((H A dies)) 9 Aw, this is [garbage! 10 P H : [O:h, never mind. 11 H A : I hate life.= 12 A L : =Wo:w. 13 How are they get ting there so fast? A t this point in the match, the score is heavily lopsided in favor of our team and the in game clock is winding down. The enemy team has almost no chance of winning this particular match. PH momentarily aligns himself as a potential particip offering aid if HA can escape. There are too many danger s and obstacles however, and PH shifts from offering assistance to of fering encouragement in Lines 5 6 proficiency as a driver PH avoids all of the contextualization cues associated with callouts and als, effectively excluding himself as a potential participant. It may be that H A oal (winning the match). He is personally under immediate threat, but neither his
48 survival nor his defeat will c hange the outcome of the match. HA likely realizes this as he fails to directly or indirectly ask for assistance through callouts. I t could also be that he judges his teammates as too far away to assist him; th is possibility is suppor 13 ) as a commiserating response to H A ). Response Cries Response cries are not callouts, but they do have the potential to shape interaction in similar ways. My data i ndicate that respon se cries have the potential to be useful tools for securing and signaling joint attention and participation for in game encounters, a finding which agree s with Aarsand & Aronsson 2009 These respon se cries a re less useful as a cooperative tool, however, in situations where surprise encounters end too quickly for even response cries to cry seems not to depend wholly on his or her surprise, but rather on knowledg e of the nearness and orientation of teammates and their ability to help (or be helped) with the sort of situation that would typically trigger a response cry. In the context of an enemy encounter, especially a sudden one, a response cry could be a potenti al collaboration in the same manner as a callout Response cries are most definitely not callouts however, in that they give no actual overt information about the immediate gameplay event that prefaces the cry. I n the recorded video data, response cries only seem to occur when teammates a re in close proximi ty; this falls in line with a previous analysis of response cries during gaming as devices for a lerting and gaining the attention of others in a participation f ramework (Aarsand & Aronsson 2009). Excerpt (6) in particular prov ides an example of very effective use of a response cry. Ambushed by an enemy a very timely manner.
49 EXCERPT (6) 1 ((PH is attacked by player from opposing team)) 2 PH: WOAH 3 (3.9) ((MI and RM fight and kill the attacker)) 4 MI: Got you= 5 RM: =Ah sure I get the assist outta that. Jee :pers. P is issued as his avatar is attacked while standing near M I and R M in one room, while my avatar is in the adjacent room with a clear line of sight to the others. All attentions are directed elsewhere as the game has been calm and encounter free in the moments leading up to the attack T he attack comes as a complete surprise ; the enemy is able to enter the relatively small room and begin hurting PH before any teammate notices As soon as cry video from the encounter shows his teamm ates all i mmediately redirect their gaze to his location. R M and M I the two closest players to P H ake care of the offending enemy before PH is killed. C ompletion of thi s action is indicated in Line 4 by M ot you M be seen as eit her a taunt toward the enemy player he just killed or a call out uptake of sorts to P H so that P H kno ws the threat is over. The calm, short delivery of M rather than as a taunt. defending him from an imminent threat. The response cry even seems to function like a standard callout in this regard. The response cry not only carries the perlocutionary eff ect of diverting his a verbal uptake by MI in Line 4 The response cry here does not really meet the criteri a for a standard callout ; the cry supplies no overt information about a threat or gameplay event. The uptake of the teammates during this event unambiguously shows that this response cry is skillfully used as a tool similar to callouts. They co construct this interaction through their willing participation in this rapid call
50 a nd response. Other recordings show that response cries given by threatened teammates at a greater distance rarely receive this level of collaboration; they must be employed strategically in situations where they alert teammates who become ratified particip ants in the activity by virtue of their proximity and ability to take some sort of cooperative action. Rate of Speech and Repetition Keating & Sunakawa 2010 characterize s rapid speech and repetition in gaming communication as aspects of pressured speech du ring a threatening encounter analysis, rapid r ate of speech and repetition are participation cues that spur collaborative action centering on the threatened avatar. Excerpt ( 9 ) illustrates a similar dynamic during a match of Halo One pla yer from the right flank with a powerful weapon and suddenly spots an enemy near his position exchange with a callout location, and immediately updates his c allout with rapid speech and repetition as the enemy becomes a threat in his personal space. The remainder of the exchange is characterized by further repetition of his callout, both by the player and by his teammates. EXCERPT ( 9 ) 1 PH: Guy coming around by our teleport ? on the >right side right side<. 2 (4.4) ((P H attacked, trying to stay alive )) 3 [Need help right side. 4 AL: [(Yeah) we got it 5 HA: ( H ) (.) > oh god < 6 PH: Need help right [side 7 AL: [on the] right side= 8 V B : =Right side [right side, on it] on it 9 CH: [You need help? P H up spe ech i n Line 1 indicates he is updating the original call out to his teammates with some urgency due to the nearness of the threat. W he n he falls un der attack he repeats the call out with a request for
51 assistance in Line 3 His teammates at first appear to acknowledge and take action on his call out, indicated by A L response (Line 4 ) irectio n giving to AL in Line 5 PH, scrambling rather heroically to stay alive for so long, repeats his callout in Line 6 because he still has not received any help Video footage of the event shows that V B myself, H A (driving a vehicle), and A L (his passenger/ gunner) all focus on an enemy t about th P H repeats his call out in Line 6 with noticeable It is then that his teammates realize that the enemy attacking P H is not the one they have targeted, and that P H initial f ocus. Since P H is holding a powerful and strategic weapon, the initial target is immediately abandoned in favor of the most immediate threat to P H. Reali zation of this gaffe is indicated by V B and A L repetitive uptake of his call out in Lines 7 8 which coincides in the video wit h the shift of attention toward the true threat While P H does not give additional lexical locative information in h is repetitio n, his use of emphasis indicates the relative location of the The second and correct uptake of 8 The immediacy of the uptake indicates that emphasis was the primary signal used to correct the wayward teammates and steer them in the appropriate direction. The timing of the second uptake is not the only significant factor, however. The the repeti tion to his own words in Line 8 only confirms to PH that they indeed now have the correct lo cation (differing fr om the initial uptake in Line 4 ), it also communicates a sense of urgency that aligns them more properly as participants in the threatening event to PH, who has been under attack for a relatively long
52 amount of time by this point. By us ing repetition to display levels of urgency matching PH, AL and VB have contextualized their utterances to signal that they have truly placed themselves in the same communicative frame as PH. rvations by Keat ing & Sunakawa 2010 In their primary framework, PH, AL, and VB all use repetition and rapid speech as cues for the appropriate participation frame. Effective communication and interaction (modeled in this study as teamwork) does not occur until all of the interlocutors have cued the same participation frame. This analysis does not differ signif icantly from mine except in terminology Keating & Sunakawa might say that the interlocutors used participation cues to signal the appropriate participation frame th rough which they could co construct a successful interacti on; my analysis is that the interlocutors used contextualization cues to signal the appropriate interpretive frame through which they could successfully understand their interaction. Both interpreta tions appear to be valid judging by the events in Excerpt (9). My analyses in this study take advantage of this by assuming somewhat equivalent roles for contextualization and participation cues and frames; given that these powerful tools for understanding are somewhat underspecified concepts in the literature (Goffman 1981, Gumperz 1982), they license a wide degree of interpretive power a not altogether undesirable state of affairs for microa nalysis. As a result, I use whichever terms and concepts are most suitable for interpreting a given interaction, while assuming a large degree of similarity in how these processes are pragmatically realized by interlocutors. Intonation Flat intonation may be one of the least obvious cues signa ling a callout in a threate ning gameplay event. Intonation has scalar rather than binary values, and not all interlocutors express the same d egree of variety in pitch across similar utterances. There do appear to be conte xtualization conventions relevant to the degree of pitch varia tion in signaling an utterance
53 as a potential callout. Utterances given in a narrower pitch range appear to be associated with giving callouts Callouts do not necessarily require a relatively flat intonational delivery, however, so what is the purpose beh ind it? I argue that it is another tool for efficiently navigating this information rich environment relevant positions in the participation framework of a given gameplay event. Gamers who communica te effectively can manipulate intonation across an utterance to create fi ne grained distinctions in how they relate to ongoing gameplay events. Young Greek women studied in Archakis & Papazachariou 2008 similarly take advantage of prosodic contextualizatio n conventions in peer interactions. In their case, they may manipulate intensity rather than pitch in order to signal their degree of involvement in the interaction. Excerpt (10) shows a gamer in Battlefield 3 TM using variation in pitch range across his ut terances to signal different degrees of his participation in ongoing events relevant to the goals of his team. He has just recently begun using a MAV, an in game miniature remote controlled hovering drone that can be used to scout objectives and enemy posi tions without putting oneself in danger. In this excerpt, GW is able to inform his teammate AN about enemy positions lexically while simultaneously informing AN about his participation status prosodically. EXCERPT (10) 1 ((GW is piloting a small airborne spy drone)) 2 GW: There we go now 3 There are four 4 AN: >Alright ten four< 5 (1.0) 6 On my way over there [now 7 GW: [U::m, 8 (1.4) 9 y good on A. prosody when giving enthusiastic meta
54 In Line 2, t here is a markedly steep intonational ri pitch continues until it rises again at the tail end of his utterance. Pitch analysis in Praat shows that the pitch during t he utterance peaks at 273 .125 utterance ( 112 .524 This shows a range of 160 .601 Hz across the whole utterance, and this variation is not limited to a sharp rise pitch climbs back to about 2 20 Hz on the last word of Line (see Figure 5 1). variable pitch in Line 2 is juxtaposed with relevantly much more level intonat ion during his callout in Line 3; there is almost no pause at all between Lines 2 3. This close temporal pitch contours all the more noticeable. Line 3 is unmistakably a callout. GW is informing his teammate AN that he sees four enemies keeping watch over the objective B that AN is approaching, a roughly man sized e game) that must be destroyed by planting a bomb on it, as in Excerpt (1). It takes time to successfully arm the bomb once the crate has been reached, and then the attackers briefly become the defenders as the MCOM begins blaring an alarm while the opposi ng team has a limited amount of time to defuse the bomb on the MCOM. Since it will take some time before the bomb detonates, this callout relays important information about a gameplay event highly relevant to AN; if he wants to successfully destroy objecti ve B, he will first have to eliminate the four enemies guarding it. For Line 3, shows a m aximum pitch of 227 .281 and a minimum of 95 .630 Hz just before the brief pause T his represents a range of 131.651 Hz, a difference of almost 29 Hz compared to the pitch range of Line 2. The visible pitch contour in Figure 5 2 shows that most of the pitch variation in Line (emphasis tends to raise pitch in addition to ampl itude) and the falling intonation before the pause. The
55 pitch after the brief pause occupies a narrower range, and represents the majority of the utterance. Since most of the pitch variation occurs during the first second of an utterance that lasts about f our seconds, interlocutors during the match would he ar the callout in Line 3 as possessing a flat intonation relat ive to the non callout in Line 2. The third utterance of interest here is Line 9. As AN reports that he is on his way to objective B, GW takes a moment to check the status of objective A and finds that it is relatively free of enemies. His utterance communicates information relevant to an ongoing in game event (attacking the objective) and is therefore a callout, but its intonation is somewhat d ifferent from the callout in Line 3; it would be somewhat ambiguous as an example of either flat or expressive intonation. The information GW reports in Line 9 definitely rates the utterance as a callout The intonation in Line 9 is not as expressive as Li ne 2, which is apparent in the pitch contour in Figure 5 3 ; if anything, it is closer to the flat intonation of Line 3 Pitch analysis in Praat shows maximum and minimum pitches of 258.419 Hz and 116.68 Hz respectively, which is a range of 141.789 Hz : na rr ower than that of Lin e 2, and only a little wider than that of Line 3. In Excerpt (10) G is controlling an airborne reconnaissance vehicle that looks like a miniature helicopter, and is called a MAV. The vehicle in this game is controlled remotely, so a p concealed place is relatively free from the en 9 These utterances are still ca llouts they communicate information that is invaluable to the teammates who are attacking the A and B obj ectives but they features associated with a threat to the speaker such as repetitive, abbreviated, or rapid speech ine 4 does co ntain such features, however AN is signaling his participation in an attempt to attack the B objective, an event which he knows (thanks to G W
56 callout) will involve fighting at least four enemy players. This certainly qualifies as an imminent threat relevant to AN. GW uses his control of intonation to contextualize his unique role as a participant in the same event as AN. By adopting pitch ranges in his callouts that are narrow relative to his non callout, GW aligns himself a s a relevant participant in a threa tening gameplay event. The event is not threatening to him however, and this is marked by his calm delivery and more fully realized, unabbreviated utterances. He is not indicating that he is a non participant in the event; rather, he is communicating that the only support he can provide is informational, as his avatar is distant from the action. In this way, GW skillfully manages intonation along with other contexualization cues to supply the maximal amount of relational and tactical information to his tea mmate. Summary Gamers who play together over an Internet connection may not share the same physical space, and avatar interaction within the virtual environment is quite limited compared to the complex range displayed by humans in everyday face to face int eraction. Communities of practice centered on first person shooters have developed rather effective linguistic and paralinguistic tools for use in interaction through a persistent virtual environment, however. Gamers can take advantage of their shared know ledge of a cooperative gameplay frame by manipulating certain features of their utterances to continually manage participation of themselves and others throughout the activity. Threatening gameplay events can crop up at a action, so this is quite an efficient use of discourse by video game players.
57 Figure 5 1 Pitch contour of Excerpt (10 ) Line 2
58 Figure 5 2. Pitch contour of Excerpt (10 ) Line 3
59 Figure 5 3 Pitch contour of Excerpt (10 ) Line 9
60 CHAPTER 6 C ONCLUSION AND LIMITATIONS To this point, various examples from two different video games have shown the following characteristics to be associated with callouts: flat intonation abbreviated and specialized utterances and terminology lexical information s pecifying self or enemy location and status, rapid speech, and repetition. Several examples were presented, however, where utterances were rated as callouts but lacked some of these features. A gamer has a variety of choices when deciding how to signal the communicative intent of an utterance; what is being signaled when some features are chosen instead of others? It is important to consider that the contextualization cues associated with pressured speech (repetition, rapid speech, brief utterances) are con flated with callouts. A great many callouts occur during events that have an elevated level of threat to either the individual players pressure situations would use such pressured speech to ask thei r teammates how their work or family lives are going. It is important to remember that contextualization cues alone did not result in a given speech sample being rated as a callout; callouts were also determined on the basis of their apparent functions and uptake by teammates. Further directions for this research would certainly include comparison to the features of discourse during other high pressure situations, such as between law enforcement officers or armed service members, or even between participant s in a high pressure environment that does not involve combat, such as air traffic control rooms. So are these particular cues nothing more than the natural by products of high pressure situations? Recall Excerpts ( 8 10 ); the presence or absence of such pr essured speech appears to participation) in a gameplay event. The type of information being transmitted by players is important, but so too is the
61 relevance of the ongoing situation to each player on the team. In this way players are able to manage their participation in various events as the game progresses, an important skill for the great deal of collaboration necessary for a successful team. B ody orientation in space relative to other ga mers fulfills this function in Keating & Sunakawa 2010 but t he gamers recorded in this study were unable to rely on such cues to participation frames since they did not share the same physical space during gameplay. Instead, the orientation of avatars rel ative to the action and the cues of pressured speech and callouts supplied by the players give a relatively detailed representation of the changing participation frameworks across the match. My data indicate that response cries a re useful tools for securin g and signaling joint attention and participation for in game encounters, a finding which agrees with Aarsand & Aronsson 2009 These response cries were less useful as a cooperative tool in situations where surprise encounters end too qui ckly for even brie f cries to effectively secure the attention of on his or her surprise, but rather on knowledge of the nearness and orientation of teammates and their ability to he lp (or be helped) with the sort of situation that would typically trigger a response cry. My study does have several limitations. This is not a quantitative analysis, and the speech and behavior patterns described here cannot therefore be generalized onto the larger population. These online gaming interactions are a great source for future work with a focus on quantitative analyses. Another limitation is a relatively small number of participants; there were many players over the course of the recorded games who were participants in the in game actions but did not wear a microphone and were therefore extremely limited in their communication, and wholly invisible in this analysis. Still more participants were recorded giving callouts or just engaging in
62 social chat, but the total number of participants in the excerpts analyzed here is fifteen. Of those fifteen participants, several appear in multiple excerpts in this study. Since excerpts were selected for analysis based on their clarity and relevance to the fe atures under investigation, the clearest communicators became powerful voices in this analysis. The communicative styles described here are common throughout first person shooter communities of practice however; further analysis featuring a broader partici pant base should confirm the features discussed here. In the work and social arenas, successful interaction and cooperation is greatly affected by effective communication and management of participation and footing toward certain situations. This is also t rue in the arena of online gaming, which is increasingly an important site for social interaction for both adults and children. In light of the sparse (but growing) attention paid to this area and type of interaction, the goal of this study has been to exa mine callouts and contextualization cues in the context of gameplay during two different FPS video games. Further studies in this area could have important consequences for our understanding of how interlocutors manage participation and communication acros s multiple events, especially without a face to the communicative skills and resources utilized during online gaming could find application to our understanding of success ful interaction in a changing workplace environment that is becoming increasingly based in the virtual world
63 APPENDIX A TRANSCRIPTION CONVENTIONS [ Indicates the beginning point of overlapping speech = Indicates no pause from one utterance to another (. ) Used to indicate a very brief pause (0.0) Numbers in parentheses indicate the length of a pause (in seconds) ( ) A closed set of parentheses with no text indicates unintelligible words (text) Text in parentheses is used for unclear speech Indicates the previous sound is cut off : Colons are used to denote prolongation of the preceding sound >text< Text between angled brackets is produced relatively rapidly for that speaker text Text between degree signs indicates relatively quiet speech text Under lining indicates emphasized speech TEXT Text in all capitals indicates loud volume Used for falling intonation ? Indicates rising intonation Indicates a continuing intonation ((text)) de scriptions of non verbal actions This arrow indicates a line in the transcript that is of particular relevance to the analysis A downwards arrow precedes a particularly marked fall in pitch An upwards arrow precedes a particularly marked rise in pitch
64 APPENDIX B TRANSCRIPTS EXCERPT (1) 1 AC: Got one on A 2 3 Aw this scope is awesome 4 ((pause)) 5 6 ((pause)) 7 Planted EXCERPT (2) 1 CK: They got a snipe at high snipe. 2 (7.8) ((Called out enemy is killed)) 3 PH: High snipe (is) down. 4 HR: 5 6 PH: [(Gus) 7 8 To your left Gus 9 ((Gus and P H die)) 10 I just lost snipe EXCERPT (3) 1 PH: 2 Ah crap 3 RM: Uh: one top yellow (.) two top yellow EXCERPT (4) 1 LV: [Losing D. 2 ??: [ ( ) 3 MD: 4 (1.0) 5 >Oh oh< D, D as in Delta, gotcha. 6 QM: Where where where 7 (1.0) 8 I see him. 9 ((Gun firing)) 1 0 EXCERPT (5) 1 GW: (Spa wn) on me >spawn on me< 2 ((GW dies)) 3 4 AND shooting an LMG down that hall[way.
65 5 AN: 6 GW: Nice. EXCERPT 6 1 ((PH is attacked by player from opposing team)) 2 PH: WOAH 3 (3.9) ((MI and RM fig ht and kill the attacker)) 4 MI: Got you= 5 RM: =Ah sure I get the assist outta that. Jee : pers. EXCERPT (7) 1 MI: Hey I 2 I have (.) cra zy good success when we hold this down. 3 4 And a shotgun. 5 PH: 6 RM: =Uh, >I got I got< [shotty. 7 MI: [Go up here, 8 Go up here to this little spot right here, 9 And you can sit right on top that ledge 1 0 11 (1.1) ((PH goe s to the indicated position)) 12 RM: ((laughter)) 13 (1.2) 14 PH: Done. 15 (1.5) 16 RM: [( )] 17 MI: EXCERPT (8) 1 ((HA driving and surrounded by enemy pl ayers)) 2 PH: 3 HA: 4 (2.2) 5 PH: There we go HA 6 Drive like I know you can. 7 HA: ((laughter)) ( ) 8 ((H dies)) 9 Aw, this is [garbage! 10 PH: [O:h, never mind. 11 HA: I hate life. = 12 AL: =Wo:w. 13 How are they ge tting there so fast? EXCERPT (9 ) 1 PH: Guy coming around by our teleport on the right side >right side<.
66 2 (4.4) ((P attacked by enemies)) 3 [Need help right side. 4 AL: [(Yeah) we got it 5 HA: oh god< 6 PH: Need help right [side 7 AL: [on the] right side= 8 VB: =Right side [right side, on it] on it 9 CH: [You need help?] EXCERPT ( 10 ) 1 ((G W is piloting a small air borne spy drone)) 2 GW: There we go nda getting the hang for it now 3 There are four 4 AN: >Alright ten four < 5 (1.0) 6 On my way over there [now 7 GW: [U::m, 8 (1.4) 9
67 LIST OF REFERENCES Aars and, Pal Andre & Aro nsson, Karin ( 2009 ) Response cries and other gaming moves Building intersubjectivity in gaming. Journal of Pragmatics 41: 1557 75. Archakis, Argiris, & Papazachariou, Dimitris (2008). Prosodic cues of identity construction: Intensity in Greek young women's conversational narratives. Journal of Sociolinguistics 12(5):627 47. Aldo Di Luzio (eds.), The contextualization of language 1 38. Amsterdam: John Benjamins. A ustin, John (1965). How to do things with words New York: Oxford University Press. Boersma, Paul, & Weenink, David (2012). Praat: doing phonetics by computer Computer program. Version 5.3.32 Online: http:// www. praat.org. Eckert, Penelope, & McConnell Gi net, Sally (1992). Think practically and look locally: Language and gender as community based practice. Annual Review of Anthropology 21:461 90. Entertainment Software Association ( 2012 ). E ssential facts about the computer and video game industry. Online : http://www.theesa.com/facts/pdfs/ESA_EF_2012.pdf Gee, James Paul ( 2003 ) What video games have to teach us about learning and l iteracy New York: Palgrave Macmillan. G offman, Erving (1981). Forms of t alk Philadelphia: University of Pennsylvania Press. Gu mperz, John (1982). Discourse s trategies Cambridge: Cambridge University Press. Keating, Elizabeth, & Sunakawa Chiho ( 2010 ) Participation cues: Coordinating activity and collaboration in complex online gaming worlds. Language in Society 39: 331 56. McGon igal, Jane (2011 ). Reality is broken: Why games make us better and how they can change the world. New York: The Penguin Press. Grantham, John; Wong, Wyatt; Workman Kevin; & Wang Alexander ( 2010 ) dint u say that: Digital discourse, digital natives and gameplay. Journal of Virtual Worlds Research 3(1): 2 36. Piirainen Marsh, Arja ( 2010 ) Bilingual practices and the social organisation of video gaming activities. Journal of Pragmatics 42: 3012 30. Piirai nen Marsh, Arja, & Tainio, Liisa ( 2009 ) Collaborative game play as a site for participation and situated learning of a second language. Scandinavian Journal of Educational Research 53(2): 167 83.
68 Reeves, Byron, & Malone, Thomas (2007). Leadership in games and at work: Implications for the enterprise of massively multiplayer online role playing games. Report for IBM. Online: http://www.seriosity.com/downloads/Leadership_In_Games_Seriosity_and_IBM.pdf Searle, John (1975). Indirect speech acts. In P. Cole & J. Morgan (eds.), Syntax and seman tics 3:59 82. New York: Academic Press. Soare s Palmer, Dionne ( 2010 ) Second language pragmatic socialization in World of Warcraft. Davis, CA: University of California dissertation. Steinkuehler, Constance A. ( 2006 ) Massively multiplayer online video gam ing as participation in a discourse. Mind, Culture, and Activity 13(1): 38 52. Wright, Talmadge; Boria Eric; & Breidenbach, Paul ( 2002 ) Creative player actions in FPS online video games: Playing Counter Strike. Game Studies 2(2). Online: http://www.gamest udies.org/0202/wright/
69 BIOGRAPHICAL SKETCH Luke Breland was born in 1985 and grew up in West Virginia and Alabama before graduating from the University of Alabama with a degree in anthropology in 2007, marrying his wife and best friend Katherine soon af ter After a year in Arkansas and two years in China teaching English with his wife and brother, Luke enrolled in the graduate linguistics program at the University of Florida, where he finally feels at home. Luke received a Master of Arts in linguistics a t the University of Florida in 2012, and is currently enrolled in the linguistics program as a doctoral student. He probably plays video games a little more often than he should, and he would really, really like to have an authentic Chinese meal again.