Abstract
This article attends to the idea of disconnection as a way of theorising people’s lived experience of social networking sites. Enrolling and extending a disconnective practice lens, we suggest that the disconnective strategies of suspension and prevention are operational necessities for those we might see as the users and owners of sites such as Facebook. Indeed, our work demonstrates that disconnection in these contexts need not be associated only with modes of resistance and departure, but can also act as socioeconomic lubricant.
Introduction
Facebook is now 10. One might be forgiven for thinking that it has got this far where others have not because of its strong connective affordances, associated displays of network externalities and resultant creation of path dependencies. This is one version of the story, and a very strong one at that. In this article, we introduce data that complicate this story. Almost the antithesis of Facebook, disconnection, is also shown to be an important actor. This article enrols and extends a theorisation of appropriation situated in this antithesis – one that puts disconnection and disconnective practice centre stage.
Since debates about the implications of the Internet for the creation and maintenance of community began to dominate new media studies in the mid-1990s, much research into digitally mediated networks has emphasised the potentials and problems of connectivity (Castells, 1996; Jones, 1995; Ludlow, 1996; Mayer-Schönberger, 2011; Quan-Haase and Collins, 2008; Rheingold, 1994; Wellman and Gulia, 1999). From within and among these discussions, various conceptions of networks, including networked individualism (Wellman, 2001), networked sociality (Wittel, 2001), networked collectivism (Baym, 2007) and networked publics (boyd, 2008; Ito, 2007) have emerged.
This tendency to understand the make-up and functioning of the networks that constitute various online communities through an often implicit, lens of connectivity is pervasive in research regarding social networking sites (SNSs). Connection and connectivity are key areas of emphasis in both our definitions of SNSs and in our understandings of how these technologies are used. In terms of definitions for example, boyd and Ellison (2007) have defined SNSs as
web-based services that allow individuals to (1) construct a public or semi-public profile within a bounded system, (2) articulate a list of other users with whom they share a connection, and (3) view and traverse their list of connections and those made by others within the system. (p. 2)
As with the descriptions of SNSs provided in the work of Lenhart and Madden (2007), Miller (2008) and Schwarz (2010), who each emphasises the importance of connection in these spaces, here boyd and Ellison were signalling the public display of connections as the most crucial, or defining, component of SNSs. Their updated version of this definition maintains the same emphasis – see Ellison and boyd (2013: 158). We have also seen ideas put forward in which connectivity is positioned and presented as a key element of social media logic (Van Dijck and Poell, 2013) and as having material and metaphorical importance in social media culture (Van Dijck, 2013a). Influential texts such as Baym’s Personal Connections in the Digital Age (Baym, 2010) and Van Dijck’s Culture of Connectivity (Van Dijck, 2013b) are also embedded with the idea of connection. These works of course attend to nuanced understandings of digitally mediated sociality, and often engage with ideas of appropriation. By appropriation, we refer, in the social shaping of technology (SST) sense, to the ongoing work that different groups put into their engagements with sociotechnical arrangements, particularly after the formal design stage. Taking this position, it is understood that sociotechnical arrangements are indeterminate, complex, heterogeneous and mutable. However, while we have such nuanced understandings of SNSs, it is important not to underestimate how connection and connectivity have become significant areas of emphasis in our definitions of them and in our understandings of how these ‘sociologies of associations’ (Latour, 2005) come into existence.
However, scholars in the area, including those cited above, have been clear to articulate that while SNSs, and the networked publics they are commonly associated with, might encourage a particular line of appropriation (boyd, 2008, 2011), or attempt to set the tone for use (Papacharissi, 2009), users may not experience them as the designer envisioned (Baym, 2010; Cassidy, 2013; Griffiths and Light, 2008; Light and McGrath, 2010; Van Dijck, 2013b; Van Dijck and Poell, 2013). Yet, although the SNS literatures articulate points of resistance and appropriation, this is usually, though not always, situated within discourses of connection and connectivity. To be clear, we do not think that giving emphasis to connection and connectivity is bad, inaccurate or wrong – it is wholly necessary. How could we argue otherwise? Both authors have and continue to research sites of connection and connectivity. We also do not see research that is concerned with connection and connectivity as automatically technologically, or for that matter, socially or culturally deterministic. What we are arguing for is an additional, complementary lens on SNSs that, in SST terms, explicitly adds symmetry to our understandings of such phenomena. Attempting to create some semblance of symmetry requires us to consider the role of disconnection as an active part of our engagements with SNSs. However, to date, and while many acknowledge the futility of technological determinism and the unexpected appropriation of SNSs, most research in this area does not make disconnection the major focus of investigation.
Disconnection is implied in a variety of studies. For example, in her study of online fandom, Baym (2007) points to the use of SNSs in fragmented ways across multiple sites which helpfully points, implicitly, to notions of disconnection. Crawford (2009), in her work on disciplines of listening, engages not only with discourses of connection through listening, but also disconnection via her attention to acts of not listening. Cassidy’s (2013) work on participatory reluctance – the engagement with SNSs reluctantly – also nuances our understandings of SNS use and non-use. So too does Lange’s (2007) theorisation of publicly private and privately public participation strategies. Papacharissi’s (2010, 2011) work regarding private spheres of interaction, particularly her commentary on the creation of privée spaces and our abilities in respect of redactional acumen also attends implicitly to the need for disconnected participation. However, while all this work touches on disconnection, it is fair to say it is not the focus of attention throughout.
Where matters of disconnection enter into considerations of SNSs directly, it can be broached with respect to issues of non-use, and framed within discourses of digital inclusion and the digital divide (Hargittai, 2007, 2012), in terms of a general lack of interest in SNS activity (Portwood-Stacer, 2013; Tufekcki, 2008), digital abstention (Mayer-Schönberger, 2011), in relation to matrices of shades of use in comparison to heavy use (Hargittai and Hsieh, 2011), and, of course, as related to death (Marwick and Ellison, 2012). Perhaps the most relevant research to our thinking here is that concerning the modes and practices associated with digital suicide, trolling, lurking and leaving Facebook (Karppi, 2011, 2013, 2014). Here, we are in agreement with Karppi that disconnection is a necessity in SNS and social media environments more generally. Enrolling Karppi’s findings particularly in relation to account termination, we seek to explain how users and Facebook enrol disconnection before the point of account termination is reached. We also point to how disconnective strategies of prevention are drawn upon by Facebook as users attempt to engage with the account termination process. We show how Facebook deploys disconnection to stabilise SNS arrangements so that any threat to the company’s existence through mass exodus is averted. Therefore, while disconnection may threaten Facebook’s existence, it also acts as socioeconomic lubricant and secures it.
Interrogating disconnection using the SST
SST perspectives emphasise technological appropriation as something involving a range of human and non-human actors. SST critiques and transcends both socially and technologically deterministic accounts of appropriation (Latour, 2005; Sørensen, 2002). An SST reading of Facebook would therefore show that even though Facebook developers may design things in particular ways, or that, although Mark Zuckerberg publicly advocates particular SNS ideals (such as connectivity), there is no guarantee that these will always come into being or unfold in the ways they were expected to. Social shaping approaches attempt to illuminate the way that technologies are configured throughout appropriation along with various actors in different social groups (Bijker and Law, 1994). SST views recognise the ongoing work that is required to make things operate beyond the design room. However, the role of designers is not ruled out of our experiences of appropriation. Designers ‘inscribe’ their visions of the future ‘world’ into a given set of arrangements, for example, notions of how the technology should be used and in what contexts (Akrich, 1992). The design and use of technologies are therefore not linked in a linear and simple way. As Roharcher (2005a) states, ‘What we generally observe is an iterative process of the co-construction of technologies by designers and various groups of users and other actors’ (p. 11). It is necessary to signal that these ‘other actors’ may well include those that we might see as non-human in the Actor–Network Theory (ANT) tradition (Latour, 2005). For our purposes, mediators such as physical surroundings and objects alongside material effects of interfaces and functions of SNSs are important considerations.
A second, and related, point of relevance here is the nuanced conceptualisation of use that is embedded within SST. Inherent within SST is the idea of symmetry and thus, while one might emphasise action through the inclusion of concepts such as relevant social groups, interpretive flexibility (the development of different meanings for ‘the same’ sets of arrangements), stabilisation (the firming of meanings) and closure (the seeming finalisation of meanings – which can however be jolted open again), such conceptualisations demand an alternate reading in terms of what does not happen and who is not included. There is a need to attend to what have been termed non-relevant social groups (Oudshoorn and Pinch, 2005a), non-users (Wyatt, 2005: 76; Wyatt et al., 2002) and partial use (Miles and Thomas, 1995). SST requires not just a focus on binaries of use and non-use, it encourages us to focus upon the nuances within and across these typical categories presented in the literature.
The ideas in this article, methodologically and conceptually informed by SST, have emerged from empirical data collected via individual qualitative interviews with 26 people in the United Kingdom. The research participants ranged in age, gender, sexuality, race and socioeconomic background. Occupations ranged from those retired on the grounds of ill health, cleaning staff, train drivers through to school teachers, politicians and those working in information technology. The interviews focused on how participants engaged with SNSs, but in particular the themes of questioning related to the navigation of multiple spaces of everyday life. Participants were asked about how they understood SNSs, their use of devices and software, modes of connection made, their thoughts on ethics and law, public space usage and usage as related to health, work and play.
Disconnecting with Facebook
This article enrols and extends Light’s (2014) theory of disconnective practice. This work suggests that to understand disconnection, one can take account of geographies of disconnection, disconnectors, modes of disconnection, disconnective power and an ethics of disconnection. Geographies of Disconnection refer to the situated nature of disconnective practice and the need to attend to the site or sites in question. This might involve disconnecting with a SNS by not joining it in the first place, leaving it or suspending participation with it for some time. Disconnection may also occur within a single SNS in relation to the people, functions or features of a given space; between a SNS and another SNS, web site or application offering or attempting connection and between SNSs. Spaces of the physical world may also feed into considerations about disconnective practice. Disconnectors are those human and non-human actors that engage with us and engage us in disconnective practice. Disconnection Modes refer to the varying natures of automated and manual disconnective activity we undertake with SNS. Importantly, these modes can be enacted with and by those actors we might see as non-human. Disconnective Power rests on Lukes’ (1974, 2005) three-dimensional view of power and raises questions regarding the extent to which disconnection is based on the following: the exercise of obvious and explicit power (one-dimensional power), the confining of the scope of decision-making to issues deemed to be relatively safe (two-dimensional power) and creating mechanisms that aim to prohibit the public airing of conflict or is exercised in ways that those subject to it comply without realising they are doing so (three-dimensional power). Finally, the Ethics of Connection seeks to sensitise us to the potential role of morality in disconnective acts.
Disconnective practice, when thought of in this way, is a collection of lenses that allow us to understand who or what is involved in disconnection, where it occurs and how it is enacted. It also, crucially, sheds light on why such practices might be enrolled. Although Light (2014) suggests that privacy is perhaps the most obvious reason for engaging in disconnective practice, he also outlines other possibilities such as notions of respect, relevance, time, media overload, fun, a lack of anonymity and a desire for anonymity, and whether we understand our audiences, imagined or otherwise. It is also argued that disconnection makes it possible for us to live with connectivity. Moreover, disconnective practice is shown to allow a series of options commonly associated with SNSs to be translated into conditions that we do not necessarily have to choose between. For example, we can choose to be efficient in the management of our relationships, by having all our connections in one place if we wish, alongside being able to attempt to manage such a diverse audience. It is also possible to attempt to strengthen connections by selectively aiming to control our presence through, for example, disconnection via tiering (Dena, 2008), and the deployment of redactional acumen (Papacharissi, 2010). Finally, it is argued that disconnective practice can offer cover, the ability to craft what we might interpret as safe spaces in the so-called ‘real name web’. Crucially, Light (2014) demonstrates that disconnection is not merely about resistance to surveillance, it can also be about adding value to our experiences.
In the following sections, we draw on data from this project in support of further theorising disconnection. Here, there are two points to note. First, acknowledging the complexity, heterogeneity and mutability of user-developer characteristics and relations (Ferneley and Light, 2008; Friedman and Cornford, 1989; Oudshoorn and Pinch, 2005b; Roharcher, 2005b), we have constructed two broad relevant social groups – users and Facebook. Users are the general population using Facebook who flexibly interpret it as something they engage with to create and maintain connections with others. Facebook is configured as an institutional entity that flexibly interprets the space it owns as a site of commercial interest. Of course, matters are much more complicated than these two groups and forms of interpretive flexibility we set out here, but they are helpful for the theoretical proposition we are making. Second, we have analytically separated the data between those strategies that could be seen as generally concerned with the desire to prevent connection, and the desire to suspend connection. Both of these strategies are, initially at least, presented as, within the hands of the user. However, we will go on to posit they can be in the hands of Facebook too.
Following the interpretation of these data in terms of the prevention and suspension of connection, we put forward an analysis that positions these strategies with regard to their value in stalling and halting Facebook account termination. Ultimately, we render disconnection as socioeconomic lubricant, arguing that strategies of prevention and suspension can directly and indirectly interweave with, and augment, Facebook’s commercial need to keep users connected within the space.
Disconnective strategies for the prevention of connection
In this study, disconnective practices aimed at delaying or preventing the establishment of connections were spoken about in terms of ‘not-doing’ something, as knowingly avoiding the connective attempts of SNSs. There were users who spoke of geographies of disconnection such as not ‘liking’ business-oriented Facebook pages (unless the business concerned belonged to a friend) in order to avoid their profiles becoming connected with the commercial imperatives of Facebook. There were users who would delay or not process friend requests (by letting them sit unanswered), for instance, until they had determined whether the person requesting their friendship would become part of their social network. Ian noted, for instance, that ignoring friendship requests from colleagues more senior than him at work helped him to maintain a level of comfortable disconnection between his work life and social activities. ‘I’m very particular about which colleagues I am friends with because I treat Facebook as a social forum’, he said,
For me it’s not a professional thing … I’m not friends with anyone on Facebook that is senior to me at work … and if they ever try to add me as a friend I try to ignore them, because I don’t want to be connected with them in a kind of social sense … I don’t want them to know what I get up to on the weekends, basically. It’s none of their business.
Here, a one-dimensional view of disconnective power was enrolled. In addition to not wanting his personal and professional lives to connect, Jason’s further motivation for avoiding these linkages was his prior experience of finding it difficult to undo connections established in Facebook with a network associated with a past employer:
There was a network on Facebook for the [theatre] and I joined it once and then Facebook closed down networks and I realised there was no way to remove it, and so for years I had works at the [theatre] on my Facebook and there was no way to remove it. Then eventually, after about 2 years, Facebook rebuilt the functionality so that you could get out of it again but anyone who joined a network, for quite a while on Facebook, was permanently stuck with it on their profile, you couldn’t get rid of it. So I wouldn’t do things like that again because, obviously, it just connected me too much, it was too much of a link between me and my private social networking and my workplace.
For Jason, engaging in a geography of disconnection was therefore a way to pre-emptively manage his presence within the space.
Disconnective practices built around these kinds of avoidance techniques were also discussed in relation to SNS-based game-play, private-messaging and check-in functionalities. Interview participants reported, for instance, avoiding games that required or encouraged user connection (such as Farmville) in favour of those that could be played solo (such as Candy Crush Saga). Such avoidance was achieved by engaging manual modes of disconnective practice. According to interviewees, such games were chosen because they did not want to disrupt or impede the flow of group-based games for other users – for example, through non-participation at busy times in their lives. As Daphne pointed out, playing only single-user games, ‘that way it doesn’t matter if I don’t go on [Facebook] for a week or two’. Users also noted that, since the introduction of time and date stamps on private Facebook messages (which indicate when a message has been read), they tended not to access such messages until they were ready to engage with the person who sent it to them, reading only the text visible to them without fully opening the message. Inversely, by choosing to post material to SNSs in ways that would be readable only to specific sections of their networks, participants occasionally aimed to prevent other people from being able to access and connect with certain of their messages and material. One interviewee, Steve, said that he had purposely aired his frustrations about a work situation on Facebook in Polish to prevent his colleagues from seeing him complain about it. This was only marginally successful though as he went on to explain,
I complained in Polish and obviously nobody at work speaks Polish. So a mate of mine picked it up straight away and I explained what this was and what that was, and we started talking on Facebook and I found out that on that day, because I had left early, people at work were actually looking up what I’d written on Google Translate just to work out what I was complaining about. So you need to be careful.
What is evident in the interviewees’ discussions of these preventative varieties of disconnective practice is that participants were often using various forms of disconnection – from games, from interaction, from professional networks and so on – as tools to help them maintain these very same connections. For Ian, ignoring friend requests on Facebook from staff at his workplace more senior than him, afforded him the kind of relationship, or connection, with those senior figures that he felt comfortable with.
Similar motivations underpinned participants’ discussions of disconnective practices such as not tagging friends in places and photos where they may not wish to be seen and where automatic connections were not engaged. Katherine, for example, described how she had not tagged her friends or uploaded pictures of them on Facebook out of what might be described as her perceived duty of care towards them; she engaged in an ethics of disconnection. Similarly, another interviewee noted that he did not connect his dating site account with his social networking profile to protect the connection he had with his 12-year-old nephew and other family members he was linked with in that space. This was the case even though the site in question offered to make the connection for him. Indeed, in a number of cases, interviewees discussed the use of disconnective practices aimed at preventing connection – among various people and technologies – as highly valuable tools allowing them to protect, preserve and enhance their relationships. Nina, for instance, indicated that she maintained geographies of disconnection by keeping her Twitter, and Spotify accounts, separate from Facebook so as not to bombard her friends with excessive and/or unnecessary information. ‘[I]t’s annoying to double post things, I think, to other people’, she said, before later adding that ‘It’s also really annoying for the people to have to see what music I’m interested in all the time’. Nina was acutely aware of the interpretive flexibility of Facebook and that others would not necessarily read it as a music sharing space.
Preventing people in one’s Facebook network from seeing what music you listen to on Spotify might also be read as a disconnective practice aimed at maintaining an element of privacy. Nina also noted how she sometimes created private Spotify sessions where ‘you can listen to all the cheesy music you want without anyone judging you’. Through this activity, we can see how Nina enrolled a two-dimensional form of disconnective power. She determined the limits of her perceived audiences’ engagement with assessments of her musical tastes. This desire for an element of privacy also underpinned users’ discussions about preventing people in their social networks from seeing what articles they read, what games they played or did not play and what other activities they might engage in on the Internet (such as Internet dating, watching pornography or reading about conspiracy theories). For these people, disconnective practices aimed at limiting the visibility of available connections to these activities sometimes involved recontextualisation work and the establishment of backchannels. Although backchannels are typically established with sites to facilitate connection among a group of people, they also allow for disconnective practice, and can be read as geographies of disconnection, because they function as spaces removed or unconnected from the broader, more ‘public’ social network. As one of the participants in this study discussed in relation to lesbian, gay, bisexual and transgender (LGBT)-oriented backchannels, such geographies allow an engagement with ‘mainstream’ SNSs while simultaneously keeping certain information about oneself, activities or interests, private. The other way this feat was achieved by users in this study was by eschewing the connective affordances of SNSs – such as automated Facebook ‘like’ buttons embedded in other contexts. For example, one user discussed copying and pasting a URL from the gay men’s SNS ‘Fitlads’ into Facebook. This was done to share an article with his father, via his Facebook account, without having to invite his farther to Fitlads.
Disconnective strategies for the suspension of connection
As individuals’ engagements with people, companies, ideas, identities, digital devices and SNSs change, connections established to reflect and/or enable these relationships with SNSs often also change. This can involve the creation of disconnections where they did not previously exist or adding to an already figured suite of disconnective practice. Accordingly, the explicit removal or suspension of existing connection was a recurring practice mentioned by participants. Whereas disconnective practices aimed at preventing connection were often associated with avoiding or ‘not-doing’ something, those practices aimed at suspending or removing connections, often involved activities of disablement, undoing and removal. Common ways that interviewees reported pursuing and achieving such disconnections were the following: (1) through the permanent removal or suspension of their relationship with a SNS profile, (2) through temporarily pausing or suspending their engagements with SNSs, (3) by ‘unfriending’ or disestablishing online links with members of their networks and (4) through a range of activities aimed at altering the ways that SNSs, and people engaging with SNSs, record events, actions and associations, such as editing one’s Facebook timeline or untagging photographs.
Many of the activities participants engaged in to suspend or remove existing connections were also often motivated by the desire to aid or enhance other connections. Several participants noted, for instance, that they regularly took breaks from Facebook in order to focus upon other forms of sociality and socialisation, which then allowed them to bring new materials into those SNSs to help further fuel connectivity. This activity was also not just done on an individual basis. For example, one person discussed how they would go on group camping trips involving a ‘technology blackout’ with friends they shared SNS engagements with. When everyone returned from camping, following the shared enactment of disconnective power through the generation of a temporary geography of disconnection, they reported that the group would have the camping trip to reminisce about, back within Facebook. Unfriending someone, for connective enhancement purposes, was also prevalent. Several participants discussed how, through enacting disconnective power, going on a ‘deleting rampage’, ‘cleaning out’ friends, or ‘friend culling’ had improved their experiences of various SNSs. An important note here is that this is done with Facebook because the platform allows users to do it. Katherine, for example, demonstrates this point while discussing how she had effectively disconnected with some of her SNS contacts by simply delineating between her ‘friends’ and ‘acquaintances’ in these environments in order to strengthen relationships with her close friends:
I know a few people who went on a friends deleting rampage because they didn’t stay in touch with them. People were posting ‘oh I’ve just cleared all of my friends so if you’re still on the list, consider yourself lucky’ and stuff like that. I suppose there is a role that it plays in it and [that was] the reason why I divided my friends between friends and acquaintances, because they [‘friends’] are people that I want to have a connection to, in a way. [B]ut it’s not like [‘acquaintances’] were close friends, I wouldn’t want them to be able to follow everything.
While participants like Katherine noted that disconnection had improved her connections/relationships with SNSs and the various people within them, actively suspending SNS-based connections was, for other interviewees, also part of an approach to online security or an effort to regain control of automated posts to their account. For example, Bob, a cleaner, wary of having too many profiles and passwords closed his YouTube account because he did not see the need for having a profile on the site because he only wanted to watch videos and post them to Facebook.
As a user of multiple interconnected SNSs and mobile phone apps, Katie discussed how suspending established connections revolved around her undoing of an automated link that had established itself between the TripIt app and Facebook. Katie wanted to edit the way that Facebook and TripIt were narrating recent life events on her timeline:
Basically TripIt would notify LinkedIn and my professional Facebook when I was going to be away so people would know that I was away and stuff, which was really valuable. And then me and my partner split up in February, and months ago we had booked a holiday to Grenada in Spain a couple of weeks ago, and all of a sudden a couple of weeks ago I had people writing on my Facebook wall ‘oh I hope you’re having a great time, have a brilliant holiday’, and people tweeting me, and I was like ‘what?’ And then I realised that TripIt doesn’t know that my relationship has ended and it was carrying on feeding out this holiday booking saying I’m am just about to got to Grenada for 5 days, so that was a bit of a weird one actually … I got tripped up by TripIt basically …
Through Katie’s experience we can also see how TripIt and Facebook engaged with a mode of automatic disconnection as they generated an interpretation of Katie’s everyday life as part of a couple about to go on a romantic break, which was disconnected with her actual situation as a single person who would be at work. Katie had to intervene by suspending this automatic connection and by undertaking additional work to set the record straight.
Disconnective strategies and preventing termination
So far in this article, we present data that sheds light on how users deploy disconnective practice to prevent and suspend different kinds of connection. What we do not discuss are the attempts of users to remove themselves from Facebook completely. Of course, using the word completely is problematic. As previous work that examines disconnection in relation to Facebook account termination articulates, leaving involves shades of disconnection, whereby users merely log out of their account, deactivate their account or delete their account permanently (Karppi, 2011, 2014). At the time of writing, only the latter strategy removes user data from Facebook. Moreover, as Karppi (2011) argues, in relation to artistic and playful attempts at engaging users in digital suicide with Facebook through his analysis of Seppukoo.com and Suicidemachine.org, such practices may not harm Facebook and instead may show alternate ways of using it.
When a user aims to leave Facebook, we see how disconnective strategies of suspension and prevention are enrolled – this time by Facebook. As Karppi (2011) notes, while it may be relatively easy for a user to log off from their account, Facebook makes it difficult for a user to deactivate or delete an account permanently. Facebook’s actions in this respect involve making it difficult to find the way for users to engage in account deactivation or deletion. Here, disconnective power is exercised in its most basic form. Facebook explicitly aims to halt a user exiting the space. Once users find the account deletion functionality, discourses of the loss of friendships, and the importance of user data, are enrolled by Facebook in a last-ditch attempt to have them remain in the space. Here, an automatic mode of disconnection is enrolled and disconnective power operates in a more complex fashion, involving elements of Lukes’ (1974, 2005) two-dimensional and three-dimensional power. Facebook attempts to keep a user’s decision-making within safe limits by having them take a ‘time out’ by logging off, or by suspending their account and leaving the data intact. This leaves open the opportunity for the user in question to return easily to the space and, crucially, it retains their data in Facebook’s databases. Elements of three-dimensional power come into play at this point too, whereby notions of friendship and data loss are enrolled. This exercise of two- and three-dimensional power is even evident if a user dies – perhaps the ultimate form of disconnective practice one might think is possible where SNSs are concerned. However, when someone dies, Facebook offers other users a variety of opportunities to connect with and through the deceased via the memorialisation of their profile and/or the creation of a new memorial page in its own right (Karppi, 2014; Marwick and Ellison, 2012). In effect, Facebook seeks to continue to generate commercially viable data from the interactions surrounding such events as it attempts to translate disconnection into connection.
Perhaps the most obvious point of engagement by Facebook in attempting to prevent termination is at the point when a user (or in the case of death, someone acting on behalf of the user) tries to close their account. However, we can also read elements of seemingly user-led disconnective strategies of prevention and suspension as acting in support of Facebook’s desire to avoid account termination. In terms of the prevention of connection, users talked of such activities as not liking, not friending, not connecting games/sites/apps, not accessing messages, not tagging and not sharing. In terms of the suspension of connection, users talked of such activities as defriending, logging off, removing timeline content, untagging content and disconnecting apps. In some ways, such activity could be read as resisting the vision Facebook has inscribed throughout its site. Such activities do not instantly correspond with the current tagline for the site: ‘Facebook helps you connect and share with people in your life’. However, one has to take a step back and reflect on what resistance is and what that means for a site such as Facebook. In general terms, when we think of resistance, we are thinking of a counter-power and, as such, resistant activities involve acts of doing and not doing in opposition to some other given power arrangement. Therefore, an easy reading of Facebook is that the disconnective strategies of prevention of connection and suspension of connection are acts of resistance. In effect, users are choosing not to do what Facebook wants them to; they are privileging disconnection over connection.
However, an alternate reading is that users are doing exactly what Facebook allows and, to some extent, wants them to. Despite Facebook’s desire and need for connection and global sharing, the site cannot operate solely on those terms. Facebook therefore is inscribed with disconnective levers and release valves that allow users to remain with it. For example, while Facebook would like users to like things in order to extract commercial benefit from the affective associations (Lovnik, 2011) they generate, and the corresponding ‘like economy’ (Gerlitz and Helmond, 2013), users can engage with the site without undertaking this activity. It would be naïve to assume that Facebook expects everyone to like everything and therefore the extent to which choosing not to like something as an act of resistance is subject to discussion. In allowing users to think that not liking things is an act of resistance, Facebook merely seeks to create the conditions by which conflict does not arise in the first place.
In addition, the absence of a dislike function within Facebook contributes to such a power context where only positive affective associations are given priority within the formal functions of the interface. Not only does this seek to protect the brand identity of any products and services being marketed within the space, it provides Facebook with a positive data set on which to base their relationships with commercial entities and aims to keep Facebook a positive place for less selective users. While not all disconnective practices that we might categorise as concerned with the prevention or suspension of connection will have a direct commercial link in the same way as the like function, we argue that they are, at least indirectly, of some value to Facebook. Disconnection does not necessarily corrupt Facebook’s desires and need for user connections; conversely, it is integral to it. One way of reading this situation is that by allowing a façade of resistance and workarounds through the sanctioned enactment of a range of disconnective practices, Facebook engages in the exercise of three-dimensional power (Lukes, 1974, 2005). Facebook keeps users in their place and suppresses conflicts so that money can be made. To further reinforce this point, in contrast, one might think of how resistance that is not sanctioned is dealt with in Facebook, for example, whereby nude photographs are posted in closed groups (which are swiftly removed), or as Karppi has interrogated, practices of trolling (Karppi, 2013). As Karppi (2013) states, ‘Facebook’s business is connected to the ways it can produce valid data but trolls and the data they produce both directly and indirectly, through molar and molecular categories, are invalid for Facebook’ (p. 292). Disconnective practice then is allowed, but only on very particular socioeconomic terms.
Conclusion
At the time of writing, Facebook has been in existence with many of us for 10 years. During this time, it has become a key actor in defining our conceptions of what SNSs are and what they do, particularly, though not exclusively, in the western world. One way of interpreting Facebook is as emblematic of the increasing emphasis upon connectivity that has become prominent in commercial definitions of new media and which can have emphasis in studies concerned with SNSs. However, an alternate reading that starts from a position of disconnection is also possible.
Connectivity and its problems have long been the concern of scholars of new media. We have attempted to contribute to this work by theoretically foregrounding disconnection. Disconnection is clearly not a new phenomenon that has arisen as a mechanism for navigating SNSs. The word disconnection is a verb, and implies the removal or breaking of connection. It is also a state of existence. The practices of disconnection therefore relate to the maintenance of that state and the creation of that state. Although disconnection implies the breaking of a connection, and consequently that a connection has already been made, disconnection can exist in its own right, in relationship to connection as a mere possibility. Focussing on disconnection helps us to understand how states of disconnection come into being with SNSs and how they are maintained. This position therefore goes beyond generic prescriptions and is better regarded as a suite of activity involving a variety of actors.
Drawing upon prior work concerning disconnection during SNS use and analyses of disconnection regarding digital suicide and account termination practices in relation to SNSs, we offer an extension to thinking regarding disconnective practice. We conceptualise disconnection as something enacted prior to joining a SNS, during engagement and in relation to account termination. Moreover, we demonstrate how the disconnective strategies of the prevention of connection and the suspension of connection can be implicated in work that seeks to prevent account termination. Such strategies are integral to account termination avoidance before a user decides to leave Facebook, or dies with an account in place. Disconnective strategies of the suspension of connection are even enrolled and offered by Facebook during the account termination process as the site seeks to keep what it can of its users for commercial gain. While the interpretation of the data we provide here is based on users’ engagements with Facebook, we believe this framework has wider applicability for studies of other SNSs and other digital media sharing similar characteristics.
Our analysis also contributes to work regarding the commercial interests in SNSs, and digital media more generally, that has attended to the metrification, commercialisation and commodification of interactivity. Conceptualising disconnection as socioeconomic lubricant, we show how sanctioned resistance conditions created by Facebook are necessary to secure continued user participation and connection making. This sanctioned resistance can be indirectly and directly translated into economic value for Facebook and the businesses that engage with the platform. Disconnection, on these terms at least, augments rather than corrupts Facebook.
Footnotes
Acknowledgements
Ben Light would like to thank the participants of the interviews for giving their time upon which the ideas in this article rest. The authors would like to thank the anonymous reviewers, and the editors of this special issue, for their contribution to the intellectual development of the ideas in this article.
Funding
This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.