Digital ethnography of linguistic multitasking in World of Warcraft

[The following is a guest post by Lauren Collister, a Ph.D. candidate in sociolinguistics at the University of Pittsburgh.]

For a child of the 1980s like I am, what could be better than playing video games to get your Ph.D.? I am extremely lucky to be able to study a subject and a community that I love, while at the same time being able to exercise my scholarly research skills. I want to share the story of my dissertation because it’s an example of the ways that a project can evolve over the course of time due to the insights one can gain from doing a long-term ethnography.

For my dissertation work, I have been conducting a full-immersion participant-observation ethnography of World of Warcraft, the highest-population multiplayer online game in the world. I am a sociolinguist and a discourse analyst, so the question that originally drew me to the game was how the text-based language used in conversation was affected by the game world – whether by issues of timing, game events, or some social factor like avatar gender. I had a lot to learn.

Digital ethnographers like Boellstorff and Nardi emphasize the necessity of spending a lot of time in “the field” to really understand what’s going on in an online community. Digital linguistics work so often seems easy – chat logs are available to read long after the chat has happened, forums and blogs are publicly accessible, and researchers can perform a ‘scrape’ to collect data from Twitter or Facebook any time they want. What is missed, however, is the sense of community, and specifically the talk and action that happen simultaneously when one is a participating member of a group. This was the greatest lesson of my digital ethnography of World of Warcraft, and one which has shaped my research interests.

When I started the ethnography, I had never played a multiplayer online game before. I had played multiplayer games, certainly, and I had participated in my share of tabletop (i.e. non-digital) gaming, but I was primarily a console RPG gamer. An example of this type of game is the Final Fantasy series, which tells a narrative in which the single player participates. World of Warcraft was an entirely different experience.

World of Warcraft (which I’ll call WoW)is a massive digital world, with three continents and another entire planet to explore. More massively, it has millions of subscribers – twelve million at its peak – many of whom love the game and participate actively in its culture. All sorts of gaming styles are represented in the player base, from those who like to explore the world and learn its story to those who want to acquire the absolute best items and defeat the most difficult enemies.

At the beginning of my ethnography, I honestly had no idea what to expect. I had a friend who was already playing, and so I joined up with her for my journey through this virtual world. We communicated with each other and to our new friends in the game world through text chat. I wrote a Master’s Thesis on the impact of the game mechanics on (textual-) conversational flow and structure. Even as I wrote that thesis, I knew I was omitting a huge component of gameplay that I had only just discovered after a year of ethnography – voice chat, specifically that which occurred over third-party voice-over-IP programs like Ventrilo and Teamspeak.

I hadn’t anticipated voice chat being part of the linguistic picture when I started my research, but as I learned the game and became a competent player, I fell in with a much different crowd of gamers than I had ever encountered before, and these players had much different interaction patterns than those I had studied early in my work. This group is the more ‘hardcore’ gaming types, the ones who go after the rare items and defeat the most difficult enemies. In WoW, it was called the ‘raiding culture’. These players form groups of like-minded people, delicately balanced for different character strengths, and collaborate to execute complicated strategies for the ultimate goal of defeating an enemy ‘boss’.

Raid groups, particularly when I first encountered them in 2008, relied on each player to be expertly competent and fulfill their duties perfectly. I was a healer character, so it was my job to keep my friends alive while these bosses threw devastating spells or caused massive earthquakes; other players had different jobs, such as the tank characters who specialized in distracting the boss so it wouldn’t attack the healers, or the damage-dealers who were relied upon to use their most powerful spells and attacks. All of this required coordination, because if even one person in the raid group failed at their duty, everyone would die.

Raiding can be serious business, and it is an incredibly immersive experience. One of the ways that players facilitated play in raid groups was through the use of voice chat. My groups always used a third-party program called Ventrilo, although there are other similar programs like Teamspeak and Mumble and even a voice chat system built into the game itself (although players largely ignored that last one because of the game glitches it caused). To raid, each player was required to have the party’s preferred voice chat software installed and at least to be able to listen to instructions given by the group leader. Most people also had microphones so that they could talk, which let them ask questions, alert the group of a problem, or simply socialize. However, voice chat didn’t take the place of the text chat mode of communication – rather, players used both, sometimes at the same time, for different functions. That is, each mode of communication became specialized, and part of becoming a really good raider was learning the rules for communication.

I only realized how incredibly cool this linguistic multitasking was when I caught myself doing it. It was as simple as that – one day during my university’s winter break, I was in a raid group and found myself chatting to the group on voice chat at the same time as I was typing a private message to an in-game friend who wasn’t in the raid that night. I realized quite suddenly in the middle of that interaction that I was doing something really linguistically cool – and if I could do it, chances were that other people were doing it too.

Immediately, I turned on my ethnographer brain and started to observe the rules. I had some intuitions based on my own use of the modes – voice chat was being used for quick communications in the heat of battle, when the keyboard and mouse were being used to control the character. Text chat was being used to make a semi-permanent record of important things, like specific people assigned to specific tasks. But there was more to it than that – there was a social component too, and I had a feeling that it was about one’s intended audience.

I consulted with my advisor, read about multimodal discourse analysis (Sigrid Norris’s work was particularly influential), and recruited two participants for a small study. I gave my two participants (roommates at a university and friends of mine in the game) Flip video cameras, and had them record themselves while they played. By watching these videos, I saw how incredibly nuanced the players’ choice of mode of communication really was. To avoid face threats and ‘calling out’ players who were not following the strategy, my participants used private chat channels to alert the leader of the raid, who then sent private messages to the offending player. This preserved the offending player’s face, and contributed to a congenial environment where nobody felt pressured or threatened by the group – a strategy for maintaining enjoyment in an intense and social gameplay situation.

The roommates, sitting next to each other, talked quietly back and forth where no one else could hear them when they wanted to make fun of another player with a silly name. They tested jokes on each other, and then relayed the funniest ones to the entire raid group. And when the group happened on an incredibly rare item, the Fiery Warhorse’s Reins that summoned an awesome horse with flaming hooves, both players carefully navigated through the different communication modes to ensure that they were following the rules – both the encoded game rules but also the unspoken social rules – for the distribution of this coveted item.

Watching these videos gave me a perspective on how mode of communication functioned in this virtual world, something that was missing from my own previous work and which has not often been explicitly addressed in digital ethnographies. I decided to pursue this topic as the main premise of my dissertation. Using many different sources of data – the videos, interviews with participants, observations in the field, and autoethnographic examinations of my own behavior – I am mapping out the meanings of these different modes and how they function as a resource for doing social work in interaction.

I never would have come to this fascinating topic had I not spent a lot of time getting to know the game, its players, and the community. It also took an investigation of the skills I had learned as a player during the course of my ethnography.

References:

Boellstorff, T. 2010. Coming of Age in Second Life: An Anthropologist Explores the Virtually Human. (Princeton: Princeton University Press.)

Nardi, B. 2010. My Life as a Night Elf Priest: An Anthropological Account of World of Warcraft. (Ann Arbor: University of Michigan Press.)

Norris, S. 2004. Analyzing Multimodal Interaction: A Methodological Framework. (New York: Routledge.)

[Above is a guest post by Lauren Collister.]

Lauren B. Collister is a Ph.D. Candidate in Sociolinguistics at the University of Pittsburgh. She is broadly interested in language, identity, and the Internet, with specific interests in multimodality and virtual worlds. She can be contacted via e-mail (lbc8pitt.edu) or Twitter (@parnopaeus). More information about her work is available on her website (http://www.pitt.edu/~lbc8), Prezi profile (http://prezi.com/user/parnopaeus/), and blog (http://talesofanelinguist.wordpress.com).