Massively Multimodal Communication and Space: A Case Study of Video Game Livestreaming

Abstract

The aim of the PhD project is to provide a qualitative linguistic analysis of live webcasts based on a case study of two Twitch streamers and their broadcasts. At the same time, questions and concepts regarding spatial references, the usage of space, and the overlap of spaces are also included in this analysis of the relatively young genre of live audio-visual webcasting.

Live video streaming has become quite popular in recent years, mainly due to the advancements in computer technology and the emergence of new online platforms specialized in the live broadcasting of large e-sports events and private video gaming sessions (Hamilton et al. 2014). In addition to YouTube Live, an internet platform called Twitch provides its users with the opportunity to broadcast their gaming experience directly to a potentially large audience. This platform, initially a sub-category of the general interest streaming site Justin.tv, was launched in June 2011 as an independent website due to the ever growing e-sports community.

Even though live streaming platforms seem to become increasingly attractive to both streamers and viewers, as is indicated by the rising numbers of active users each year (Twitch 2015), research dealing with this kind of virtual environment is still scarce. Among the few papers that have been published so far, we find, for example, a study by Cheung and Huang (2011) regarding spectators at major e-sports events. The authors analyzed the roles of audience members who were physically present during such events and, as a result, divided them into nine so-called spectator categories, each defined by a specific set of attributes. Two years later, Smith et al. (2013) take this categorization up and, based on the analysis of various online communities, show that six of the nine categories are also applicable to the physically absent viewers found on platforms such as Twitch and YouTube Live. A recent study by Hamilton et al. (2014) focuses on the emergence of online communities around specific streaming channels on Twitch. Based on the analysis of their data, which consists of field notes regarding the observation of users and spectator numbers, as well as interviews with participants, the authors argue that Twitch can function as a so called third place, a concept that was first introduced by Ramon Oldenburg who defines it as a public place “that host[s] regular, voluntary, informal, and happily anticipated gatherings of individuals beyond the realms of home and work” (Oldenburg 1997: 16, as quoted in Hamilton et al. 2014: 1318).

The argument of Twitch being a third place will also be taken up in the PhD project as one component of the research. Contrary to Hamilton et al. (2014) though, the current study will not only consider ethnographic data, but will also include an analysis of the participant’s verbal data in form of speech and writing in order to gain insights into the question of whether the development of a community and the affiliation with it can be observed on a linguistic level as well.

In addition to the possible channels of communication provided by the Twitch interface, some streamers also make use of third-party features, programs and platforms which are either actively embedded in the stream (e.g. TeamSpeak, subscriber pop-ups, follower alerts) or passively given reference to without being shown in the streaming window (e.g. Twitter, Facebook, the streamer’s own website). Consequently, these additional means of communication allow streamers to communicate with their possible audience through a whole range of different channels, which are not directly bound to the webcasting platform. For example, by speaking into a microphone and broadcasting the spoken content live via the Twitch interface, streamers can comment on what happens in the game or their immediate environment, they can answer any questions that other users pose in the Twitch chat, or they can engage in interactions with fellow players and other community members via Twitch and TeamSpeak simultaneously. However, streamers can also engage in written communication and interaction with other current online participants via the Twitch chat, the in-game chat, and even the TeamSpeak chat. This use of different communication channels may eventually lead to streamers constantly switching between the immediate physical space and different virtual spaces during their live broadcast. Consequently, in addition to looking at communication-related issues, the study of a mixed-media environment such as Twitch also calls for research questions that consider issues such as the usage(s) of virtual and physical space in such a live streaming context and, in connection to this, how space and spatiality influence the linguistic output of the participants. As an example, we can look at the streamers who are located in a physical environment during their live transmission and film themselves with a video camera while they play a video game. Next to the digitalization of a streamer’s actions in the physical space by immediately uploading the video feed to the streaming platform, we can discover another modification of this space. While some streamers transfer the entire rectangular image that is recorded by their webcams, thereby showing themselves and part of their immediate environment, other streamers position themselves in front of a so-called green screen and modify the image transfer in a way that only their body and specifically selected objects are visible as an overlay on top of the video game stream. Additionally, sometimes third-party programs are used, which provide users with gif images, small video sequences, as well as statistics regarding subscribers, followers, etc. These digital objects can also be positioned as an overlay anywhere in the video game screen. Thus, what the audience gets to see as the final product in the Twitch video window may be a modified version of a virtual space, which consists of several layers of overlapping digital images. Consequently, in their communicative acts, both the webcast producers and the recipients can refer to and make use of one or more spaces at the same time.

The data primarily consists of screen recordings of live streams on Twitch, which represent the broadcasts of specific streamers over the course of several weeks. Additionally, screen recordings of the TeamSpeak interface are used to assign each speech act to a specific user, whose name lights up whenever he/she communicates via the program. Also, a chat-recording program is employed in order to obtain a chat log in a separate file, which then serves as a raw chat transcript. From these recordings, a corpus containing transcriptions of spoken and written data, as well as descriptions of gestures, viewing directions and body/avatar positions is created. The analysis of this corpus and the audio-visual material will be conducted by drawing on methods and concepts from conversation analysis (see Sacks 1992; Schegloff 2002 [1970], 2007), (multimodal) discourse analysis (see Baldry and Thibault 2006; Kress and van Leeuwen 2001), computer-mediated communication (see Georgakopoulou 2011; Herring 2007), chatspeak (see Cherny 1999), space in interaction (see Hausendorf 2013), space in computer-mediated communication (see Beisswenger 2013), and audience design (see Bell 1991; Goodwin 1986).

PhD candidate

Nathalie Meyer

Supervision

Christa Dürscheid, Andreas Jucker

Funding

URPP Language and Space