The right to the virtual city: Rural retreatism in open-world video games

Abstract

This article uses Lefebvre’s spatial triad and his concept of The Right to the City to categorise open-world video games as contested virtual spatial experiences, interconnected with the non-virtual spaces in which they are produced and played and replete with the same spatial, capital forces of alienation to be negotiated and maintained. We use qualitative gameplay data (n = 15), unpacking players’ journeys through Lefebvre’s conceived, lived and perceived spaces, to show, respectively, how open-world games can be (1) fundamentally about space, (2) spaces interconnected with the non-virtual world and (3) disruptive spatial experiences. In utilising The Right to the Virtual City and our players’ tendency to retreat into the wild spaces of our case study game, Red Dead Redemption 2, we evoke the same alienating forces of commodification and capitalism to which Lefebvre spoke, positioning open-world video games as both contested spatial experiences and opportunities to challenge spatialised inequalities.

Introduction

This article examines how open-world video games can be inherently spatial experiences, using Lefebvre’s spatial triad to unpick the spatial politics present in participants’ (n = 15) playthroughs of Red Dead Redemption 2 (henceforth RDR2), an open-world playable landscape set in the turn-of-the-20th-century United States. Players control a cowboy-avatar named Arthur Morgan, navigating their way through sprawling countryside vistas, barely trodden trails and emergent urban landscapes. Through our case study, we show open-world video games as fundamentally about space, replete and interconnected with the same rules, social forces and spatial politics as the non-virtual world – which led Lefebvre (1991) to describe play ‘as alienated and alienating as labour’ (p. 383).

Research using Lefebvre and video games to date has been almost exclusively theoretical, so following Murphy’s (2004: 228 emphasis added) assertion that ‘questions of how gamers identify while playing a game are incredibly complicated’, we offer an empirical test of players’ interactions in the game space of RDR2, which are explored in order to understand how ‘gamer identification sutures the gamer to the game’ (Murphy, 2004: 235). Lefebvre (1991) tells us that ‘social space [is] multifaceted: abstract and practical, immediate and mediated’ (p. 266) and he develops a set of tools not to categorise, but to decode social spaces – his ‘spatial triad’ (Lefebvre, 1991) – which includes conceived space, lived space and perceived space. The first is a top-down set of rules implemented by planners, designers and, in this case, software developers; the second is symbolic space, experienced through signs and symbols; the third can be understood as a bottom-up set of initiatives and directions made by the inhabitants of a space.

This article considers all three in sequence, showing, respectively, how open-world games can be (1) fundamentally about space, (2) spaces interconnected with the non-virtual world and (3) unsettling and confusing spatial experiences – through our participants’ interactions in the city, and their tendency to retreat into the wild. We conclude with our discussion titled The Right to the Virtual City, where we make our contribution to think of open-world games as primarily contested spatial experiences due to their playable arc as commodified, exclusionary, alienating encounters, and players’ resultant rejection of the built-up city environment.

Researchers routinely argue the importance of studying video games as spaces (see Jenkins, 2004), and in responding to Fraser’s (2012) argument that there is a ‘need and importance of a more thorough reconciliation of video game studies with Lefebvrian spatial analysis’, we seek to position game spaces in a symbiotic relationship with non-virtual space, revealing how the intersectionality of virtual and non-virtual spatial politics plays out in play (p. 101).

While it is important to acknowledge that contemporary Games Studies scholarship has moved beyond analysis of triple-A titles towards new developments in areas like indie gaming and ownership (see Lipkin, 2013; Parker et al., 2018, for example), gender politics (Fisher and Harvey, 2013), fan studies and modding communities (Stanfill, 2019), our return to triple-A titles is useful for two reasons. First, it is reflective of the significant impact that games like RDR2 continue to have: as a media product, RDR2 has sold something like 29 million copies at the time of writing (Makedonski, 2020), and the critical appraisal of the game situates it clearly in broader discussions of video games as problematic, reductive and damaging. For instance, in the context of historical (in)accuracy and representation (Donald and Reid, 2020). Second, there has been a renewed focus on triple-A titles and spatio-politics (see Holmes, 2019) that are worth attending to, particularly in developing critical tools for analysis of spatial problems in a virtual and non-virtual sense.

Literature review

Video game space – which offers both a commentary on our contemporary world and an arena of play – has evolved significantly since Wolf’s (1997) taxonomy of spatial representations, towards using three-dimensional open-world geographies with scintillating graphics to tantalise an ‘affective’ experience from the player (Shaw and Warf, 2009). Dismantling these spaces into categories or typologies has been done frequently.

Konzack (2002) considers space through the virtual/non-virtual, as an interrelated layering of hardware, code, functionality, play, meaning, referentiality and socio-culture – myriad forces that shape spatial experiences. This suggests a binary between virtual and non-virtual, subdivided by physical/tangible things like hardware, and the less tangible nature of play itself. Indeed, Konzack (2002) goes on to argue how virtual space is distinct from the ‘playground’ – the world in which these factors exist and the game takes place, which ‘normally [. . .] are kept apart’ (p. 90).

Konzack’s space/playground dualism suggests a delineation but one where the player’s own non-virtual surroundings should be considered, which we will address through our methodological approach of ‘interactive elicitation’. Second, it demonstrates an ontological misunderstanding of the virtual/non-virtual as different. Shields (2003) explains that the virtual/non-virtual exist together as facets of the ‘real’: the virtual is ideally real, whereas the non-virtual – what he terms the ‘concrete present’ (p. 29) – is actually real. Therefore, the spatial distinction is ontologically unsound: the virtual and non-virtual are both real, but in different ways, so interconnections between the two are both possible and necessary, something we demonstrate empirically.

While these typologies and binaries appear helpful in delineating particular aspects of spatial overlap, typologies can also create confusion: Crawford (2015) suggests that ‘one reason that scholars cannot agree upon suitable terms to describe the objects of their study is because they do not necessarily agree on the fundamental nature of what it is they are studying’ (p. 574). Given this ontological position, our case study involves field testing why and how participants move between virtual/non-virtual interpretive frameworks, thinking through what these perspectives offer in terms of using video games to explore contemporary social issues.

There is some useful research that explores video games along these lines. Murphy (2004) maps how video game space is contiguous with the outside world of the gamer and is critical of potential distinctions between virtual/non-virtual space. Calleja (2007) warns against seeing the player and game as a binary, where ‘players plunge into the virtual environment’ (p. 83). Apperley (2006) states that ‘to privilege the categories of the visual and narrative is a failure to understand the medium’, where a pure focus on aesthetics and story obscures the particulars of gameplay and restricts us to a series of comparisons with more traditional forms of media ( p. 6). For Fernández-Vara (2011) ‘one of the clearest aspects that bridge stories and games is space’ – what is referred to as ‘environmental storytelling’ where spatial setting has an inseparable influence on narrative, and on players’ journeys through said narrative (p. 1). Equally, Jenkins (2004) uses ‘evocative spaces’ – similar to amusement parks – to show how the player draws upon previous cultural experiences to tell the story to themselves, rather than having it told to them – something that will be important in unpacking Lefebvre’s idea of symbolic ‘lived space’.

Following the ‘digital turn’ (Ash et al., 2018), Ash and Gallacher (2011) urge us to approach video game environments as ‘(techno)cultural practice’ (as opposed to just looking for ‘geographies in video games’, or at ‘geographies of video games’) that accounts for the cyclical production of games and gamers who influence each other in turn. This is not to argue that ludic architecture and virtual architecture are the same; however, as Adams (2003) attests, ludic space serves additional purposes including the deliberate construction of challenges and obstacles to manufacture a playable arc, something which Lefebvre (1991) has called ‘contested space’, whereby the corners of his triad rub against themselves.

In considering Lefebvre in video game research, Crawford (2015) neatly summarises the importance of his roots in cultural Marxism, pointing out that while most theories of virtual space frame play as an escape from ordinary life, Lefebvre (1991) acknowledges play as a ‘continuation of the established order’ (p. 383). Stemming from Huizinga’s (2016 [1949]) work on ‘magic circles’, games scholars (see Adams, 2003) have often drawn virtual space as a separate and distinct area of escape from everyday life, an approach we intend to counter through Lefebvre’s (1991) neo-Marxist notion that play and leisure are not separate, untouchable spaces but are ‘as alienated and alienating as labour’ (p. 383).

The application of Lefebvre in video game research in the past has been patchy, and this comes in part from a misreading of the original spatial triad. Nitsche (2009) splits game space out into five planes – rule space, mediated space, fictional space, play space and social space. He likens his ‘rule-based space’ to Lefebvre’s ‘spatial practice’ which Lefebvre, conversely, defines as not about rules but about a user’s daily reality. Nitsche also likens his ‘mediated space’ to Lefebvre’s ‘representations of space’, whereas for Lefebvre it is the ‘representational’ space where the image plane of shared signs and symbols overlays on top of the physical world. Aarseth (2008) has also employed Lefebvre’s spatial triad in a similar way but has not ‘clearly elaborated’ or fully considered how it maps onto digital play (Crawford, 2015: 578) making it theoretically interesting, but practically unhelpful.

This confusion demonstrates the need for a considered approach to applying Lefebvre’s typology, echoing Fraser’s (2012) concern that ‘there is a danger in reifying video games as themselves one aspect of his triad model of space’ (p. 100). In Lefebvrian terms, everyday life is not just played out in space, but is actually about space (Gottdiener, 1985) and how capital and labour alienate and dominate, so it is important to avoid treating the virtual as separate from the non-virtual.

Method

Montfort (2006: 17) asks ‘how is the game played, with how many people, doing what sorts of things?’ In helping to answer that, we conducted an empirical, qualitative analysis of how players’ engage with game spaces. Semi-structured interviews are often favoured in researching gameplay (see Atkinson and Willis, 2007; DeVane and Squire, 2008), whereas we use ‘interactive elicitation’ (Spokes and Denham, 2019) where gameplay is brought inside of the interview, with interviewing split into two stages before and immediately after gameplay. Sommerseth (2007: 765) argues that ‘realism in video games is dependent on the players’ embodied experience of play’, so witnessing gameplay itself and reducing recall bias are necessary benefits of our approach, weighted against potential drawbacks such as social desirability bias (Nederhof, 1985) which we have addressed at length in our earlier work (Spokes and Denham, 2019). Becker (2017) reminds us that ‘we should never confuse interviews with ‘being there’, seeing for yourself what happens and recording it soon afterward’ (p. 188) and our development of interactive elicitation effectively brings these two techniques together.

Our participants (n = 15) comprised of 10 males and 5 females, ranging in age from 20 to 25 years old: all self-identified as gamers – a prerequisite in the participant call – and either worked or studied at a University in the North of England. Participants were sampled opportunistically (Kemper et al., 2003) between January and June 2019, anonymised as P1–P15. Using Fereday and Muir-Cochrane’s (2006) hybrid inductive-deductive coding, data were analysed thematically (Boyatzis, 1998).

Participants were given identical instructions, beginning with 30 minutes of unstructured, unrecorded gameplay with RDR2. This initial playthrough offered the opportunity of control familiarisation (we used a PlayStation 4). Participants then began their recorded playthrough at an identical location in the city of St Denis – a port town similar to New Orleans. They were given identical instructions to ‘play the game however you would normally play it’. Of the 15 participants, 7 opted to do a structured mission and 8 chose to use their time to explore.

Conceived space (or ‘representations of space’)

Turning to our application of the spatial triad, the first facet – conceived space – is ‘conceptualized space, the space of scientists, planners, urbanists, technocratic subdividers and social engineers’ who ‘identify what is lived and what is perceived with what is conceived’; it is the ‘dominant space in any society (or mode of production)’ (Lefebvre, 1991: 33): Lefebvre argues that this form of space is the type which reinforces hegemonic control through top-down, design-oriented expectations and rules. For video games, Montfort (2006: 4) calls this ‘game form’ – ‘the simulated world and rules of the game’ – and seeing as the game has been comprehensively designed, every crevice of RDR2’s spaces are intrinsically ‘conceived’. We can use ‘conceived space’ to interrogate how game designers intended for space to be used, where space is constructed to divide the Old West from the new, and to alienate the cowboy-player from more-developed parts of the world.

First, participants felt that RDR2’s spaces were not conducive to non-violent play. P14 finds a reason why Arthur keeps getting into trouble:

[it’s] very one dimensional. So it didn’t give you, for example, options to be more charming or to be more lenient or to be more . . . [. . .] what’s the term? [. . .] less aggressive. So maybe that’s why he keeps getting into trouble. (P14)

there’s an interesting point where maybe the rules of the game is to basically survive those interactions enough times until you get to the end of it, you know, and he gets his gold or dies or whatever. (P1)

Despite the game having a scale signifying honourable/dishonourable play (implying intrinsically moral decision-making), P14 explains that the spaces they occupied were one-dimensional, ambushing their play and pushing towards violent action. P1’s comments are similar, where the playable story arc is presented as a series of dangerous trials they have to surmount in order to progress. This echoes what we (Denham and Spokes, 2018) have found in Grand Theft Auto (henceforth GTA), where players are systematically provoked by language, setting and space. P1 reminds us that this trend is not new:

What [games] have always done, – actually, particularly now, they say, ‘hey, you can do anything you want in this experience’, but really, it doesn’t work that way, you know? They want you to do certain things. Although, I guess, in this game, there is an attempt to try and say, ‘hey, do what you want’. (P1)

I’m being ambushed . . . oh dear . . . oh no. And I have none of my dead eye left, ¹ so I’ve got to just – either got to run away or . . . don’t think you can take any chances in the Wild West. (P4)

Notwithstanding these sorts of games being referred to as ‘open-world’, P1 sees that the sales pitch of ‘do what you want’ is stymied by the design of the game towards violent encounters – similar to Nitsche’s (2009) ‘rule based’ plane where developers overlay their intentions for play over a seemingly free space. These comments add weight to those critical of Huizinga’s (2016 [1949]) ‘magic circle’ model, or recent criminological examples in Atkinson and Rodgers’ (2016) ‘murder box’, where play offers an escape from constraints, or where violence is presented as predominantly player-driven, respectively. P4 summarises how the game acts upon the player’s use of space when cornered on their horse by bounty-hunters in a small woodland clearing: they have to respond quickly within the space available before being killed: realising that a gallop through dense woodland would be difficult, P4 has to fight:

It’s not a good farm. Oh, no, they’ve reported me for trying to be friends. (P9)

I’m just- I’m not familiar with this place, oh dear. I only talked to him! [And he’s reported you to the police for that?] Yeah! That’s just rude. (P2)

P9 is the first to address how the ‘game form’ (Montfort, 2006) alters from space to space by stumbling across a smallholding where the farmer is antagonistic and they have a bounty placed on their head within a few seconds (bounties are conspicuously instantaneous in the game, regardless of space). Along with P2, P9 realises that their company, and innocently intended solicitation, are more welcome in some spaces than others – this observation is especially noticeable in St. Denis where our participants set off from:

[You’re back in St Denis] Ahh, where people don’t like me. Like when we went into the theatre, a lot of people who were sitting there were [. . .] quite well dressed and . . . he does seem like he is, outside of the, that kinda inner social circle that was there. He’s quite a lonely guy. There’s a lot of characters I noticed . . . like, fancy ones, [. . .] I would imagine that would’ve had like a higher-class person, someone that would’ve owned that-. (P2)

P2 explains that the city of St. Denis is not intended for them (or Arthur as the avatar). This draws attention to the sorts of commodified spatial inequalities that Lefebvre highlights as part of ‘the right to the city’. Lefebvre (1995 [1968]) argues that urban life has been repressed from a co-created, organic social milieu towards a repressive commodity, at once alienating and angled towards excessive wealth – characteristics that P2 draws out in their interpretation of St. Denis as a distinctly classed space where grizzled cowboys are not welcome. The developer’s nod to America’s changing landscape, typified by turn-of-the-century industrial prosperity, is clearly acknowledged by participants through the impacts of wealth polarisation and alienation within the conceived space of the city. P2 sees ‘violent contrasts between wealth and poverty’ (Lefebvre, 1995 [1968]: 67) – only one participant chose not to immediately leave the city, and those who referenced it did so as a waypoint for the consumption of new clothes and weapons – what Lefebvre (1995 [1968]: 86) calls the distinction between ‘use value’ and ‘exchange value’ spaces, ‘spaces bought and sold, the consumption of products, goods, places and signs’:

I think they did such a good job of making it so utterly terrible and frightening I just don’t like being there. As soon as I got there [. . .] I felt ridiculously uneasy about being there and I feel as though Arthur as well, the way he’s characterised doesn’t particularly suit that environment. [. . .] I think the setting and the time that the game’s sort of designed in is conducive to that because there’s that like, these stories of these old Western sort of bandits and things. (P4)

P4’s description of St Denis as ‘utterly terrible’ for his avatar as an outlaw resonates with Lefebvre’s (1991: 116) comment that conceived space is where the ‘history of ideologies resides’. As he concludes, a ‘. . . conceived space is a place for the practices of social and political power; in essence, it is these spaces that are designed to manipulate those who exist within them’ (Lefebvre, 1991: 222), and through P4’s ‘stories of old western bandits’, we can see how that manipulation begins to work in ‘lived space’.

Lived space (or ‘representational space’)

Lefebvre’s (1991) concept of lived space refers to the decisions made by people who live in a space (i.e. perceived space, coming next), but ‘space as directly lived through its associated images and symbols’ (p. 33). It is the space of symbolic representational comprehension, aligned with Jenkins’ (2004) ‘evocative spaces’ or what Nitsche (2009) refers to as ‘evocative narrative elements’. This is analogous to Lefebvre’s (2000) notion of ‘thought’, where people develop ideas in relation to space based on shared reference points (p. 135). In games research, Montfort (2006: 4) calls this ‘reception and operation’ – ‘accounting for what the player does in interpreting the meaning of the game; connecting the experience of the game to the other experiences of gaming, art, literature or life’.

Participants identified evocative narrative elements in RDR2’s persistent use of symbolic cues from a shared reference point of ‘Spaghetti Western’ films. Despite our not asking about film, 14 of 15 participants referenced films or film-characters:

uhh, he seemed like a very sort of typical action hero tough guy. Chuck Norris esq almost. (P6)

I don’t know because my idea of cowboy is Woody out of Toy Story, that is my extensive knowledge of cowboys. [. . .] you know that rugged, he doesn’t care what anyone says . . . he’s just kinda, you make trouble, I’ll give ya trouble. (P2)

Django. This could be like a Quentin Tarantino game. Like when you think of Django and how they shoot and things it’s just like that. It’s just all animated [. . .] It makes you think of Django and Gangs of New York. (P7)

These character references are disparate – from the martial-arts actor Chuck Norris, to Woody from Toy Story and the gunslinger Django. Django Unchained and Gangs of New York are both set in a similar era but follow entirely different cross-sections of history – from slave emancipation to territorial gang-rivalry, respectively. Despite the obvious differences between these figures and settings, they are used in an identical way, titillating players’ ‘transgressive imaginations’ (O’Neil and Seal, 2012) to import the cultural schema of cowboy as a ‘tough’, ‘rugged’ sharpshooter with pinpoint accuracy and overflowing charisma. On top of these divergent references, the most common were to Clint Eastwood or John Wayne and in these comments, we can see how the cowboy genre intersects space:

Uhm, yeah just like the taverns and the you know the western setting of a film kind of rolling into a town on a horse, people, that kind of seemed to signify a lot of old Westerns in my opinion yeah. (P12)

He seems like Clint Eastwood film-type cowboy. I don’t know if that’s authentic properly but it’s kind of what I’d expect. [. . .] I’d say it’s probably based a lot on pop culture just with like the way he opens doors. He does it like they’re saloon doors even if they’re not. I’ve walked in and out of that train station and both times he’s opened the doors like that in and out and I’m like I don’t think the hinges go both ways! (P9)

P12 calls on the Western genre as a spatial signifier, describing how Arthur travels like a cowboy – a lone-wolf rolling into towns with residents sat outside taverns eyeing his progress. P9 agrees that Arthur is based on popular cultural tropes, speaking specifically on the way he enters and exits spaces: Arthur pushes through or aggressively swings open doors – just like saloon doors in a film – even if he is at the train station and the hinges do not appear to move in both directions. These signifiers are what Lefebvre (1991) calls ‘representational space’, what Konzack (2002: 96) has called ‘referentiality’ where ‘these characteristics are signs, ornaments or game structures that have originally been used in other media or other games, [given] new meaning by transferring meaning from where it originally appeared’. The ‘new’ meaning that participants identify is a feeling that the player is directly inhabiting an established filmic universe where Arthur occupies space as an outlaw, his Western attributes rubbing up against spaces rapidly turning against him. P13 sees these signifiers as a deliberate act by the game’s developers:

they have the kind of Western theme playing when you’re riding for long periods [. . .]. It’s just trying to give you that feeling of being in that type of world. (P13)

It’s interesting, though, because it’s the historical context it’s in makes it more – I don’t know. With GTA, it’s obviously a satirical take on our society, but it’s one that has enough familiarity that we can at least – I don’t know – make sense of. I don’t know what people lived like here really, apart from Sergio Leone movies and Clint Eastwood films and all that kind of stuff. (P1)

P13 considers RDR2’s music, with cameos from Willie Nelson – a central figure in the ‘Outlaw Country’ genre – as a signifier and references its purpose-built Western aesthetic to help players situate themselves in the historical environment they have not lived through. Indeed, with the lyrics: ‘cruel cruel world I’m moving on. I’ve been living too fast, and I’ve been living too long – cruel cruel world I’m gone’ (Willie Nelson) and ‘may I stand unshaken amid, amidst a crash of the world’ (Rocco DeLuca) – RDR2’s music is decidedly on point for its turn-of-the-century Americana narrative.

P1 is the first of many to make a direct comparison with GTA, the most successful franchise in the developer’s catalogue, which has a similar playable dynamic of ‘micronarratives’ (Jenkins, 2004) across an open-world environment. They point to the importance of representational space in RDR2 to contextualise play and improve immersion, whereas GTA, being set in the modern world, gets by with fewer filmic cues. The effect of this is to slightly alter gameplay, but this can also prove problematic if the signifiers do not fully align with expectations:

You cannot be in the Wild West without a saloon. No women there? How do I start a discussion and be charming and stuff? [. . .] is he bothering her? I’m going to be a white knight. Doesn’t anyone in this town need my help or something? Clint Eastwood walk into the town and immediately he finds people. (P14)

P14 spent the majority of their playthrough confused, the lived space of filmic representations failing to mesh with the conceived space of unbridled violence, particularly in relation to a highly gendered experience of space: where Clint Eastwood was able to roll into a town where his services are required and immediately charm women, P14 finds the ‘urban problematic’ (Lefebvre, 1995 [1968]: 65) – a ‘habitat’ that his kind should not ‘inhabit’, where any attempts at ‘charm’ are met with the conceived overlay that cowboys lack the right to the city and should be exploring the countryside elsewhere.

Perceived space (or ‘spatial practice’)

Similar to Ash and Gallacher’s (2011: 356) ‘geographies of videogames’ – and highlighting the critical, sometimes rebellious ways in which ‘videogamers do not always straightforwardly accept these representations’ – Lefebvre’s ‘perceived space’ is the ‘bottom up’ experience of the player which ‘embraces production and reproduction, and the particular locations and spatial sets characteristic of each social formation’ (Lefebvre, 1991: 33). In exploring spatial practice, it is important to acknowledge Crawford’s (2006) work on the illusion of being in control as a key appeal for video games – what we are analysing here is not the player necessarily being in control, rather, the way that the spatiotemporality of lived and conceived space are altered through the activity of play and player choices, making up the ‘practical basis of the perception of the outside-world’ (Lefebvre, 1991: 117), or in this case, the game world. Spatial practices include working out what representations of space and representational spaces mean, and how they interact with quotidian experiences in the virtual city.

Despite conceived space pushing players towards violence, the overwhelming spatial practice was to explore and take advantage of manipulating this filmic world in experimental ways:

It does look quite run down. So the other one was, I take it, a bigger city? I do love how space in games is developing. In fact, when I play any game like this now, I tend to just explore. (P1)

[We haven’t completed very much of the game so most of it is greyed out] oh right. Well I’m gonna go to all of these places then! (P11)

P1 sees the spatial politics of St Denis as oppositional to the more run-down shack-like towns they encounter later in their playthrough. This affirms the playstyle of exploration rather than pursuing the game’s more regimented narrative structures or engaging in violence. Participants embraced the vast, enterprising, ‘land-of-opportunity’ setting that Western films propagate and took pleasure in interacting and testing this previously off-limits space in new ways. When P11 saw that most of the game map was still greyed out and unexplored, they made it their mission to defog unchartered territories. These spatial practices also included a desire for many participants to abstain completely from violence, including P1, P8, P9, P11, P14 and P15. Although, their work to eschew the game’s conceived intentions is not entirely free, notably when exploration impacts the game’s programming:

Just exploring. Trashing that farm although not on purpose. I thought you’d be able to pet the other animals like so you could pet the horses. (P9)

It’s kind of unsatisfying when – no, it’s kind of like . . . what’s the word, like . . . a bit gutting when you find the edge of the boundaries. (P3)

P9, in their exploration of the game’s wilder spaces, found a smallholding which they mistook for a petting-zoo (in their attempt to approach and touch sheep). This resulted in an accidental ‘trashing’ of the place, letting animals out of their pens, and provoking the farmer until a bounty was placed on their head. P3 displays how this trumping of the conceived over the perceived is ‘unsatisfying’, and the reality that the spaces cannot be freely interacted with, or that the player is not in control (Crawford, 2006), becomes clear. Still, players pushed back against events that unfolded against their will:

Explore as many different scenarios as possible, I think . . . without it getting too messy. If that was me on my own file I would’ve been so upset, straight away, disappointed that I got so- I would’ve probably reloaded back up or whatever because it’s just like . . . not what I wanted to happen. (P3)

I’d say the really fun part is just riding around just seeing all the different scenes of the game. Cos it’s an incredible world that they’ve built. The less fun parts are where you’re just stuck behind cutscene walls and you have to just wait to come back in and play. (P9)

P3 got ‘messy’ when ambushed at the train station by characters whose reason for violence was unclear. This was not concerning during our short gameplay, but P3 is clear that to avoid tarnishing their own game they would have started over. This is an example of the conceived consequences of a space rubbing against the perceived intentions of players for how that space should be used (in P3’s case, to simply buy a train ticket) framed by the cinematic influences of their lived spatial experience and their desire to explore. P9 echoes this in a different way through ‘cutscenes’, small snippets of one-directional cinema that guide the player through a game-designer-led narrative. Unwittingly, players can stumble into lengthy filmic episodes against their want for exploration. What this represents is a direct pushback by participants against conceived programming, despite that corner of the triad playing a dominant role in shaping their play experience. Back in the context of St Denis, P4 says,

I just ran straight out of it [. . .] I wanted to experience things that you can’t experience now. We can all go to see these things, [. . .] there’s nothing really stopping us from going to cities, we can all go out into the countryside but it’s not as though we have the freedom necessarily to hunt these great herds of animals. (P4)

I think if you were going to do well in the game it would probably be [. . .] like going to the theatre and doing all the little things you can do. Sometimes I will, like I’ll go to them kind of places and pay for things. But I just prefer doing it on my own, just going my own way. (P8)

Although we have already problematised the idea that game spaces can be a ‘free’ experience, P4 implies that the countryside of late-19th-century America is freer than the city, or less of a ‘conceived space’. They allude to the fact that cities dominate their non-virtual lives, and that rural spaces seem to be less designed, or have more ‘freedom’, due to their lack of familiarity. P4’s words, especially on hunting, go far beyond exploration into their desire to inhabit a filmic schema of what a Wild West cowboy could and should be doing. P8’s response to the question ‘did you do well?’ further distances the city space from the countryside. For P8, the city is an aspirational space – the conceived space of planners and architects (Lefebvre, 1991). It is a place you need to go to in order to be a decent player, to achieve, or to ‘pay for things’. Echoing Lefebvre’s (1995 [1968]) sentiments in The Right to the City, the fruits of the city are deeply classed, exclusionary and focused on exchange value – cowboys like Arthur Morgan lack the right to inhabit these new, industrial, capitalistic spaces, and P8 would rather go their own way.

The right to the virtual city

By unpacking RDR2’s different spatio-political experiences using Lefebvre’s (1991) triad and associated concepts related to the ‘right to the city’, it is possible to reflect on the open-world genre as (1) fundamentally about space, (2) interconnected with the non-virtual world both ontologically and empirically and (3) as unsettling and confusing spatial experiences. This is important because as Crawford (2015) attests ‘focusing on play as an isolated space, centres it, and runs the risk of ignoring its wider social context’ (p. 577).

First (1), we find that RDR2 as an example of open-world games, is fundamentally about virtual and non-virtual spaces. P15 made it their mission to visit these spatial critiques in their gameplay:

[on prisoners building the railway line] I feel like I want to go see that because I feel like, I thought that was like a lot of what the game was about. (P15)

Open-world games are often characterised as primarily violent experiences, and while violence is indeed present here, we can see in RDR2 a focus on interactions with and critiques of space. P15 highlights how turn-of-the-century industrialisation is ‘what the game was about’, using their gameplay to highlight the migrant-built transcontinental railroad that fuelled the growth of cities (driving ‘outlaw’ cowboys further to the periphery) – a transition referred to by P4 as a ‘metaphor for America going forward’. We made this argument in ‘conceived space’, where players felt excluded from the urban environment due to their position as cowboys and gave these spatial politics as a reason for centring their play in the countryside.

In this way, RDR2 exemplifies the tendency in role-playing games more broadly for players to be characterised as outsiders (Vella and Giappone, 2018). RDR2 plays upon the fundamental game oppositions of inside-outside and city-wild. Vella and Giappone (2018) cite the walled cities in The Witcher III as an example of how the function, use and opportunity differences between developed and developing space is used to position the player as outside of something that they should have to ‘earn’ (p. 9) access to. In RDR2, the cities are not physically enclosed, but it is the lived space of representation that positions the urban as an affront to the sensibilities of the outlaw cowboy.

Second (2), we find that open-world spaces can be inherently connected to the non-virtual world in which they are played. P1, at the beginning of their exploration, immediately picked up on the inseparable nature of virtual and non-virtual space. The problem with this game is that you have

People working 80-90 hour weeks, essentially, which Rockstar definitely are guilty of as a company, means that the level of detail in this game I find problematic. Even though it’s clearly beautiful, incredibly detailed and visually exquisite in like all sorts of places, it compromises or undermines that vision for me. (P1)

Having a critical awareness that virtual space must be produced by people who endure difficult working practices dismantles the graphic ‘detail’ that can be a provider of a sense of the real (Schwartz, 2006). P1 experiences virtual spaces in RDR2 as interconnected with the contested spaces of work, programming and alienation in which games are produced, empirically echoing Dyer-Witheford and de Peuter’s (2009) work on how the damaging nature of global capital is distilled in video game worlds. P1’s comments reflect on their right to play the game at all: we made this argument in our ‘lived space’ subsection, where participants repeatedly drew upon their transgressive imaginations (O’Neil and Seal, 2012) informed from exterior cultural experiences and reformed their playthrough accordingly – wanting to use their playtime to seduce women or find confrontation in taverns.

This echoes the findings of Möring and Leino (2016) that while games are often presented in a liberal fashion – especially games like RDR2 that front the rhetoric of agency and freedom critiqued by P1, they often do so through a neoliberal lens that perpetuates the same sorts of classed or gendered rhetoric and capitalistic alienation, or in this case, industrialisation, that are inherited from broader socioeconomic developments.

Third (3), we find that in focusing on realistic, wild vistas, RDR2 became an unsettling spatial experience. For example, P6 spent most of their gameplay stranded in between spaces after having accidentally shot their horse:

excuse me sir. No? Is the horse still running off? He’s not even dead. OK, maybe a stagecoach will be easier. Excuse me sir. (P6)

Politely asking passers-by if they could have a lift back to a space where a new horse could be purchased, P6 ended up walking for 18 minutes of their 30-minute gameplay. P4 acknowledged the game’s open spaces as something that ‘humans bottled up in the big city’ long for, but also acknowledged the ‘transition between the natural and the industrial’ as being a stark experience for players who have not had the opportunity to experience the wild in non-virtual life. The fact that you can ‘play for eight hours and nothing could happen’ (P14) was simultaneously a draw and a frustration for our participants. We made this argument in our ‘perceived space’ subsection, where we highlighted players’ drive for exploration and will to push the boundaries of the virtual space they inhabit.

Bonner (2018) refers to these liminal spaces between the wild and the city as ‘striated wilderness’, critiquing the very idea that there can be such a thing as conceived (or ‘programmed’) wilderness – perhaps instead we should be referring to the practice of wildness, where players are encouraged to dig out their dormant survival instincts and play as wild in a conceived environment.

‘When Marx was writing, [capitalism] grew because the world was pretty empty’ (Harvey, 2014), and our participants regularly reflected on this expansion of efficient railways into previously untamed wilderness, and electric companies installing overhead power cables as something that was not routinely presented in their lived space of the Spaghetti Western. They saw the open-world game experience as one of a simultaneous freedom to explore worlds unspoilt by the mechanisation of the city countered by feelings of exclusion where those vistas began to be encroached on. They have grown up in towns and cities and, as University students, currently reside in a city. They are not excluded from the urban landscape in their daily lives, but are able to recognise, play and embody feelings and behaviours associated with the harms of gentrification, urbanisation and oppressive competitional capitalism – the game allows them a social experience that is profoundly abstract from their own.

In this way, RDR2 reflects on and encourages player engagement with capitalism’s drive towards endless growth and expansion into previously untouched spaces and, in doing so, becomes a playable spatial re-enactment of the sorts of ‘hegemonic liberal and neoliberal market logics’ (Harvey, 2012: 3) that characterise modern urban spaces. It urges players to think on the sorts of ‘collective right’ (Harvey, 2012: 3) to occupy spaces, and act in spaces, that Lefebvre (1995 [1968]) speaks to in The Right to the City. It also impacted on their modes of play.

As Harvey (2012) details, ‘the question of what kind of city we want cannot be divorced from the question of what kind of people we want to be’ and therefore, what kinds of people are discouraged from existing in the city (p. 4). In these terms, we can better understand the frictions present in our data between the conceived and the perceived spaces, where the sorts of player that the game encourages are divorced from the sorts of play that are engaged in, and the game itself makes playable the sorts of idealistic and exclusionary spatio-politics that have come to define modern urban spaces. Lefebvre (1991) has referred to this tension as ‘contested space’, where the corners of his triad in their overlap are creating both tension and reflection among its users, but it is present too in The Right to the City in the context of those who are permitted to inhabit and those who are excluded.

Players’ inner-city play strategies tended towards the avoidance of conflict, towards purchasing and repeatedly vocalising the risk associated with transgression inside of the city and the need to head towards the countryside in order to engage in activities they would associate with a cowboy. In these behaviours, our participants acknowledged Harvey’s relationship between kinds of people and kinds of cities where the sorts of personalities they choose to present are altered by the process of urbanisation.

Players use open-world video games like RDR2 to indulge in the ‘contemporary urban experience with an aura of freedom and choice in the market’ (Harvey, 2012: 14), using their playtime to circumvent what they perceive to be the conceived intentions of programmers and developers, inside of a playable space that mirrors the major aspects of the modern, urban political economy. They claim no ‘shaping power over the process of urbanization’ (Harvey, 2012: 5) that would constitute a right to the virtual city, but they do embrace the right to the city as a form of urban resistance where they are cognisant of the relationship between citizenship and capitalism (Purcell, 2002), that it reshapes, and ‘affirm[s] [. . .] the right of users to make known their ideas on the space and time of their activities’ (Lefebvre, 1995 [1968]: 35) by acting against what they feel are the conceived intentions of the game’s developers.

Moving forward, what this research demonstrates is the importance of ‘doing’ spatial theory in the context of video games, of applying and testing theoretical frameworks in an effort towards better understanding the interrelationship between video games and lived experience. Here, we have highlighted the ways in which virtual and non-virtual spaces are deeply intertwined, and in doing so we have shown the value of using theory not just to underscore empirical work, but to enrich our interpretive and conceptual toolboxes and to frame ongoing debates about the importance and impact of interactive entertainment.