‘One can set out by refusing speech, or by making speech the province of the weak, of women, of powerless characters who speak because they can’t act.’ (Chion, 2009, 327).

As with the motion picture and, perhaps more pertinently, the animated cartoon, the move to voice acting in videogames introduced both new possibilities and new problems in the field of character development. Of course, the videogame’s dynamics of address are somewhat different to those of film and animation. For, in addition to situations where characters speak before, or within earshot of, the player/auditor, there are also instances in games where characters presume to speak for/as the player or intervene to speak to them, offering feedback, instruction or commentary.

Video game characters frequently say things that we wouldn’t say, or things that we feel like they wouldn’t say, or gesture outside the diegetic frame to advise us to ‘press x’, and while most gamers take this as a matter of course, some find the resultant forms of dissonance, frustration and bathos troublingly incompatible with those airy signifiers in which marketing blurbs tend to ground gaming’s appeal – agency, immersion, identification etc.

One of the problems is that, technological constraints notwithstanding, giving players a range of dialogue options requires a prohibitive investment of time and resources on the part of developers. Carolyn has already discussed one of the more popular workarounds on this very blog, analysing so-called ‘dialogue trees’ as a pragmatic, if rather limited, means of allowing players to exercise a degree of choice and assert an in-game personality. Dialogue trees offer a workable partial solution – albeit, as Carolyn points out, one underwritten by some questionable assumptions as to the nature of choice, identity and moral probity.

Another approach, sometimes used in conjunction with the dialogue tree, is to mute one’s hero(in)es altogether, a move that allows the developers to circumvent some of these issues while also, as Michel Chion notes, employing a tried and true method of virilizing cinematic characters. The silent protagonist is one of those video game design conventions that’s as familiar to players as it is gratingly incongruous to non-players. It’s also one of those design conventions that the Half Life games made a great case for, but which are looking steadily more threadbare with each passing year (under which category one might also file being ushered in a linear fashion through confined spaces and periodically subjected to meticulously choreographed scripted sequences). At best, then, the silent protagonist is a serviceable means of sidestepping a knotty problem, though in practice the protagonist’s silences can often be rather awkward ones.

For certain players, however, these gaps in the audio track are something else: an incitement to improvised oral performance. This was a possibility I’d never really considered until I began watching people roleplaying Skyrim. Roleplay is mode of gaming which aims to take the game seriously – or perhaps, as Bart Simon, drawing on Goffman (1967) might have it, to help the game to ‘save face’ by playing in such a fashion as to support its premises (2007, 167).  This commitment has both diegetic and ludic consequences: players concoct elaborate backstories and motivations for their avatars, consistent with the game’s setting and fiction; they purposefully make things difficult for themselves for the sake of ‘realism’, refusing to resort to cheats, exploits and hacks, or even to amenities offered by the game in the name of convenience (like ‘fast travel’ systems); they impose additional constraints felt to increase verisimilitude, in a form of what Parker (2008) calls ‘expansive play’. In the case of Skyrim, many players also make use of mods which enable or support particular kinds of play and performance – a mod that makes it a requirement for the player-character to eat, sleep and stay warm, say, or one that supplies an alternate starting scenario more consistent with the player’s impression of who their character ‘is’.

For some players, though by no means all, roleplay also means voicing the game’s protagonist, and a number of Youtube channels are devoted to videos by players who speak for/as their Skyrim characters. These videos are interesting for a number of reasons. For one thing, they cast light on the positioning of players in relation to a game’s characters and fiction, revealing how complex and contradictory the actual dynamics underwriting so-called ‘immersion’ and ‘identification’ tend to be. Some videos are voiced in the first-person while others vacillate between first- and third-person narration, switching too between taking the game’s posited reality at face value and treating the game as system or software (e.g. to point out things about their rig or a mod they’re using). Accounts of who a character is biographically (their upbringing and ethics) shade into descriptions of what they are mechanically (a magic user who specializes in light armour and one-handed weaponry, say).

If some players prove to be surprisingly deft and engaging narrators, ‘goofs’ and inarticulacies can be thought-provoking too: when a roleplayer, having been suddenly ambushed, takes the name of Christ in vain – rather than that, say, of Talos, Mara, or any other denizen of the Skyrim pantheon – I find myself wondering how one would swear in this world. English swearing, which remains substantially rooted in Anglo-Saxon, tends for example to focus on the body’s reproductive and excretory functions, while in Francophone Quebec, I’m told, blasphemy gets more of a look-in.

Such musings confine themselves to the level of what Chion calls ‘semantic listening’ (1994, 25) – the domain of linguistic content. Chion also identifies two other types of listening however: ‘causal listening’, which is directed toward identifying the source of a sound, and ‘reduced listening’, which is about cultivating an ear for texture and timbre in and of themselves. In the case of roleplay videos interesting things are often happening on these levels too.

On the causal plane, it is easy to find oneself making assumptions about the age, the gender, the class, the ethnicity, the intelligence and even the appearance of the person attached to the roleplaying larynx – and some Youtube users have no qualms about publishing uncharitable conclusions they’ve reached. Mods which use amateur voice actors (like Interesting NPCs, which adds a supporting cast of eccentrics voiced by volunteers) often face the problem of being swamped with willing actors capable of sounding like young(ish) Amercan males, while struggling to find anyone with what we might call the causal capacity to play older, ‘foreign-sounding’ and/or female characters. While it’s difficult to vouch for the accuracy of suppositions made on the basis of a voice recording, the granularity (or otherwise) of what different people infer can be striking: I can’t, for example, necessarily hear the Canadian accents that, according to many North American players, putatively US-bred characters often have in (the many) games made in Canada, though I’m better – or assume I’m better – at ‘placing’ people speaking UK English. If the internet once supported fantasies of transcending geographical limits and the confines of embodied identity (Wertheim, 1999, 17, 23), here materiality and geography creep back into the picture, albeit in a tantalizingly oblique, partial way. Videos’ soundtracks often bear unintentional traces of their circumstances of production: traffic noise, pets’ cries for attention, mobile phone alerts and PC interface cues, rustling pages, echoes, feedback, hissing sibilants and popping plosives intervene to remind us (as if we ever forgot) that we’re not really in a dragon-rife Nordic fantasy land. Such noises can afford us intriguing glimpses of the kinds of tech being used to play and record the game, and of the sorts of spaces where play takes place.

Beyond what is being said and who is speaking there are the nebulous qualities that render a voice distinctive, arresting or antipathetic at the level of reduced listening. For Barthes (1977, 179), certain voices simply possess ‘grain’, while for Brian Massumi it was not Reagan’s politics that endeared him to voters so much as ‘the timbre of his voice, that beautifully vibratory voice’ (2002, 41) (which is either something of a relief or even more alarming depending on how you look at it). Similarly, certain roleplayers can make the spectacle of someone picking virtual cabbages more engaging simply because their voices have a certain je ne sais quoi. With some players offering viewers hours upon hours of content, it is easy to imagine subscribers becoming acquainted with the nuances and idiosyncracies of their voices, sharing with habitual listeners to podcasts and radio shows the strange sensation of feeling like they know someone they’ve never seen, let alone met.

The ability to record and stream gameplay is being touted as one of the key features of new consoles,  and, inevitably, players are already testing the boundaries of such systems. Falling somewhere between Robert Browning’s verse and the Let’s Play video, the products of Skyrim‘s roleplay community suggests that the growing influence of services like Twitch might just be the catalyst for new forms of improvised vocal performance.

Barthes, Roland (1977). Image Music Text. London: Fontana.

Chion, Michel (2009) Film: A Sound Art. New York: Columbia University Press.

– (1994) Audio-Vision: Sound on Screen. New York: Columbia University Press.

Goffman, Erving (1967). Interaction Ritual: Essays on Face-to-Face Behaviour. New York: Pantheon Books.

Massumi, Brian (2002). Parable for the Virtual: Movement, Affect, Sensation. Durham: Duke University Press.

Parker, Felan (2008). The Significance of Jeep Tag: On Player-Imposed Rules in Video Games. Loading  2(3), http://journals.sfu.ca/loading/index.php/loading/article/viewArticle/44

Simon, Bart (2007). Human, All too Non-Human: Co-op A.I. and the Conversation of Action. Situated Play, Proceedings of DiGRA 2007 Conference: 165-9

Wertheim, Margaret (1999). The Pearly Gates of Cyberspace: A History of Space from Dante to the Internet. London: Virago.