Digital displacement: The spatialities of contentious politics in China's digital territory

Abstract

This paper conceptualises digital displacement as both a way through which the digital, dynamic and fragile spatialities of contentious politics can be examined and as a geographic critique of censorship. Digital displacement, understood here as the act of removing users from the digital places and spaces they wish to remain in and use, often through the act of deletion, is conceptualised through the digital displacement of two contentious political groups that attempted to contest the forced eviction of migrants from Beijing in 2017; hashtag focused #BeijingSurgery# and instant messaging group using BeijingTogether. Explored through participant observation, interviews and playful digital exploration, this paper examines the spatialities that made multiple digital displacements possible and the activist spatialities that emerged during and post-displacement. In exploring this, I develop a flexible vocabulary around digital place, space, scale, territory and mobility to analyse the practices of digital displacement, to understand the socio-spatial positionality of activists involved in digitally centred contentious politics and to contextualise their territorial positionality within Chinese digital territory and global digital territories. Through the examination of #BeijingSurgery# and BeijingTogether, the article highlights: the importance of digital territorial positionality for both activists and the digital places and spaces used for contentious politics; that within systems of digital spatial governance, deletion and displacement can be effective strategies of repressive governance with wide-ranging displacement effects; and that while digital displacement is not necessarily the ending point of contentious politics, the re-production of activist spatialities is more difficult when the authority being protested against governs the digital territory used for protest.

Abstract

This paper conceptualises digital displacement as: a way through which the digital, dynamic and fragile spatialities of contentious politics can be examined; a way to understand China’s digital territorialization project; and as a geographic critique of censorship. Digital displacement is conceptualised through the ethnographic research conducted in Beijing and examines efforts to contest the forced eviction of migrants from Beijing in 2017.

1 INTRODUCTION

The spatialities of contentious politics¹ evolved during the 2010s as a new protest cycle emerged, a cycle making use of occupation and the normalised use of digital spaces for political contestation (Routledge, 2015; Tarrow, 1993; Tilly & Tarrow, 2015; Yang, 2014). In this cycle, political contestation does not take place in physical or digital spaces, but across both simultaneously (Castells, 2015). During this period the practices of contentious politics in digital spaces have also evolved, with the use of hashtags dominating the early 2010s and instant messaging (IM) groups becoming popular in the late 2010s (Clark, 2016; Freelon et al., 2016; Walker, 2020). But one essential moment of protest cycles is rarely discussed in the literature on contentious politics; the moment when political contestation ends. When the end of political contestation is discussed, the users of the digital spaces of spatially hybrid contentious politics are often absent. These absences suggest that the digital spaces used for political contestation exist in perpetuity, perhaps unused and static, yet that is rarely the case. To remedy these absences, in this paper I focus on, and theorise from, China, a country where the digital spaces of contentious politics are subject to regular and often rapid deletion.

For those engaged in contentious politics in China, during the 2010s the deletion of politically contentious digital spaces became a commonly encountered phenomenon, with #MeToo#, #IAmGay# and #RefusePeriodShame#² just some of the deleted Weibo³ hashtags (Liao, 2019; Yue Yang, 2022; Zeng, 2020). Keeping these deletions in mind, in this article I develop the concept of digital displacement to understand the political and spatial effects of deletion, with a particular interest in how the deletion of politically contentious digital spaces affects those engaged in contentious politics and the causes they fight for. In conceptualising digital displacement – understood as the forced removal of users from digital spaces and places, such as IM groups and hashtags – I makes contributions to geographic understandings of displacement and contentious politics. Digital displacement extends displacement to include new spatialities and contexts, to consider the political and human effects of displacement from digital spaces.

In arguing for digital displacement as a conceptual tool to understand deletion, I am also arguing that displacement is a useful analytical lens through which political contestation can be understood. When examining the multiple spatialities of contentious politics (Leitner et al., 2008) it is important to consider the role of digital socio-spatial relations – including the effect of digital place, space, territory and (forced) mobility – on contentious politics. Examining contentious politics through displacement effects helps make sense of why, for instance, only some contentious political networks re-group and/or re-territorialise their projects after being displaced, and why those engaged in political contestation may eschew occupations or public spectacle (Arampatzi, 2017). Throughout the article I will show that displacement is a technique of governance used to control and end contentious politics, and that it is a technique particularly common in contexts where governments are producing sovereign digital territories (Fang, 2018).

In conceptualising digital displacement, I also make a geographic critique of how the term censorship is used. In academic literature on repressive governance in China, censorship has been used to describe a wide range of human activity, including actions taking place in both newsrooms and news feeds (Repnikova, 2017). But, if censorship is concerned with restricting ‘public expression of or public access to information by [the] authority’ (Roberts, 2018, p. 37), then how should the deletion of hashtags that are linguistically difficult for broader publics to understand, or the deletion of instant messaging groups used for semi-private assembly and activist planning, be understood? By understanding these events through digital displacement rather than censorship, I illustrate how the intended effect of deletion can be less about the control of information and more about the end of assembly (King et al., 2013).

Following the introduction, I reflect on the research methods used and the positionality of the researcher. Existing literature on contentious politics is then examined, with the digital socio-spatial contexts of political contestation highlighted. Following this, digital displacement is defined, and the spatial logics which underpin the concept are explained. Focusing on territory, I note how the territorialisation of digital space enables greater governance of digital space users while also highlighting the role of US digital territory in contentious politics, a territory where displacement is rare. Shifting the analytical lens away from US digital territory, I begin a theorisation of digital displacement from China, arguing that digital socio-spatial relations emerging from China help explain the uneven geography of digital displacement. Using ethnographic data to analyse two instances of anti-eviction political contestation in Beijing, I highlight the different forms that digital displacement can take, and the different activist spatialities that emerge in the wake of displacements. The article concludes with thoughts on the current spatial logics underpinning China's digital territory and how they are exploited to practise contentious politics.

2 RESEARCH CONTEXT

This article makes use of data compiled during two years of ethnography in the physical and digital geographies of Beijing and China. The long-term ethnographic work initially focused on how translocal migrants attempted to achieve stillness, the active practice of resisting forced mobility and staying in place (Morris, 2021), in Beijing, and included long-term participant observation in food stalls, restaurants and IM groups. Central to my methodological practices during fieldwork were three of Pink et al.’s (2015) principles of digital ethnography: multiplicity, non-digital-centric-ness and openness. This is because hybrid contentious politics is often an open event that requires the examination of a multitude of non-digitally-centred relations. Furthermore, I practised a scepticism of the spectacular and an awareness of the mundane (Leszczynski, 2020a; Pain, 2014) while understanding my fields as vibrant, material and voluminous landscapes made of bytes and bricks (Bennett, 2009; Billé, 2020; Dourish, 2017; Hall, 2020).

Unfortunately, spectacular acts of violence became mundane, and the mass displacement of translocal migrants and those living on the urban fringe inspired grassroots networks to engage in political contestation (Morris, 2022). These networks used hashtags, IM groups and cloud storage as key sites in their contentious political practice. I engaged in participant observation within publicly advertised Weixin⁴ groups that I was invited into, and I negotiated continued access with administrators. While in the groups I took part in their associated projects, including participation in data-gathering field trips with other group members. I treated these sites as part of a fluid, contentious-political field, noting how people moved their focus between physical and digital spaces, becoming simultaneously situated in multiple spatialities. To engage in safe research practice, I consulted scholarship on surveillance and censorship to ensure I was not inadvertently bringing harm to the groups (Ruan et al., 2020). This care continues in my writing, where I make use of pseudonyms and avoid long quotes. While these IM groups were openly advertised, I paid attention to public/private boundaries that existed within them. I was open (not loudmouthed) about my status as a researcher, and my publicly displayed Weixin name included my details, while my biography included project details and university affiliations.

In this article I discuss displacement, a subject often associated with the violent removal of people from their homes, yet the displacement described here has different consequences and affective dimensions. This article continues attempts to enrich scholarly understandings of displacement by examining displacement as an inter-disciplinary subject (Pasquetti & Sanyal, 2020; Roast et al., 2022). In discussing digital displacement, it is important not to erase the multiple violences associated with displacement from physical sites, but that does not mean digital displacement and deletion are not violent. Displacement in China ‘is one of many spatial-temporal events in the lives of migrants’ (Morris, 2021, p. 938), and for some the loss of an IM group or hashtag may be more distressing and cause greater long-term harm than the loss of a temporary home (Liao, 2019). When discussing digital displacement, it is important to remember displacement's violent history and to contextualise the term, but just because the spatialities of violence involved in digital displacements are different, they are no less real.

Researching digital displacement meant dealing with unexpected issues, notably, the instantaneous deletion of field sites. Deletion made it difficult to find out what happened in a field site, who it happened to and when it happened. Unlike physical displacement, digital displacement leaves few visible identifiers, and, as digital displacement often takes place simultaneously with censorship, textual and aesthetic representations of deleted things may be added to censorial blacklists and become difficult to search for (Ruan et al., 2020). Ethnography in the physical and digital geographies of Beijing enabled me to be in the right place at the right time, and, through contacts made in the field, information on contentious politically activity was shared to me. This led to interviews, referenced throughout this article, that, for safety, took place several months after political contestation had ended, either in encrypted digital spaces or places chosen in consultation with the interviewee (Longhurst, 2010). During interviews, which were uploaded to encrypted non-Chinese digital storage as soon as was possible following interview, records of deletion and displacement were shared to me as we engaged in application walkthroughs (Light et al., 2018). These records enabled me to work backwards, using temporal, textual and visual data to see what could be scavenged on Weibo. In the end, Weibo's own search function was a powerful tool. Through playful exploration, particularly playful use of search functions, I noticed conspicuous absences and patterns in how Weibo displays information, displaying 20 results a page from a database that included, but did not display, censored content, meaning many pages had less than 20 results. I followed these conspicuous absences to track the temporalities of deletion, displacement and censorship. Through URL manipulation of error pages of deleted content, I was often able to proceed through databases to a point where posts became visible again. These methods shed light on the scale of erasure, and I used these ‘fragments’ (McFarlane, 2019) of deletion alongside ethnographic and interview data to understand the spatialities of digital displacement.

As a white male from Europe, I was a clear outsider, and this influenced my research in many ways, including clunky linguistic skills, limited cultural knowledge, obvious foreign appearance and the complex power relations that this produced. During fieldwork I was brought inside networks and IM groups by those who knew me as an ‘outsider’, and I took the decision to trust their judgement in bringing me ‘inside’. While within the project I attempted to act ethically and safely, including by reflecting on my knowledge of governance practices in China and listening to advice from project members. As I was taking part in project practices, not just observing, I regularly reflected on the politics of my non-Chinese visibility, and, when conducting group fieldwork for these projects, I stayed in the background where possible, with hat and face covering on, to avoid my non-Chinese status raising suspicion amongst those my interlocutors hoped to help.

My ‘insider-outsider’ status evolved during fieldwork, and I became acutely aware of the various, changing positions I held after being invited to share my thoughts into project practice with project members (Zhao, 2017). At this moment I was brought ‘inside’ as a discussant and considered simultaneously inside and outside as a participant observer. During this time, I started to understand a further status that influenced my research: the uncompromised outsider. As a foreign researcher outside of the Chinese professional and political system, I was seen as uncompromised in relation to fears surrounding state-employed spies infiltrating political projects. This status was confirmed during interviews, but it was not a magic bullet, and I was initially denied access to certain spaces due to fears I was an unwanted foreign journalist who could cause trouble. My non-local nationality and non-local institutional statuses aided my entry, but I continued to reflect on this access, to make sure I did not act like an unwanted journalist. Complicating insider-outsider statuses further, the vast majority of interlocutors were themselves semi-’outsiders’ in the context of their activism: while many were migrants to Beijing, they also had different class and cultural positions to the evicted migrants they sought to aid and many had never engaged in this form of contentious political practice before.⁵ This positionality affected their practice (discussed below) but it meant that I had a lot in common – including class position, cultural capital, educational attainment and technology skills – with my interlocutors, commonalities that made my participation in the projects easier. Thus, my insider-outsider status – non-local in many comprehensible and seemingly safe ways – was never static, was regularly reflected on and was something that has aided in the collection of this data and the critical discussion of these subjects.

It is also important to acknowledge a broader positionality that affects my analysis and writing (Rose, 1997). I write this article from the position of someone who has spent years in Chinese digital territory but who is also deeply embedded in anglophone digital territories: digitally transnational. From this position, it seems that many digital phenomena causing concern in the anglophone world have been under discussion for years in China, such as fears over so-called ‘fake news’. This positionality has led me to theorise digital socio-spatial relations from China, rather than just using China as a case study. Theorising digital socio-spatial relations from China has been made easier by the Chinese state's own use of a spatial vocabulary in describing digital socio-spatial relations, a vocabulary which I build on and add to (Fang, 2018).

3 DISPLACING CONTENTIOUS POLITICAL EVENTS

To help conceptualise digital displacement and the displacement of contentious political events, it is useful to first note changes in what Tarrow (1993) describes as cycles of protest, as it is in these new cycles that displacement has emerged as an important (often repressive) technique of governance (Tejerina et al., 2013). While occupation as a form of protest is not new, during the 2010s the occupation of public and quasi-public spaces became a prominent feature of contentious politics (Arampatzi, 2017; Mitchell, 2012; Routledge, 2015). During this period, squares and plazas became ‘public spheres where people could not only share alternatives, if not counter-hegemonic discourse, information, viewpoints, and ideas, but also where they could develop a sense of community’ (Tejerina et al., 2013, p. 382). Alongside occupation, another mode of contentious politics has emerged: the use of digital sites, including hashtags and IM groups (Freelon et al., 2016). These digital spaces hold many parallels with physical spaces of contentious politics, becoming sites ‘of collective world-making’ (Vasudevan, 2015, p. 318) and ‘transformative empathy’ (Rodino-Colocino, 2018, p. 97). While the use of digital spaces in contentious politics is not a new trend – digital spaces helped the Zapatista movement mobilise ‘a net of supporters’ to reshape ‘the context of the uprising and spawn new connections outside the Internet’ (Froehling, 2013, p. 165) – during the 2010s there was an important change from the use of the websites, as contentious political artefacts holding manifestos, to the use of digital spaces of sociality, where multi-stakeholder contentious politics took place in real time.

Recent geographic conceptualisations of political contestation treat it as having polymorphic ‘socio-spatial relations’ where multiple spatialities are ‘mutually constitutive and relationally intertwined’ (Jessop et al., 2008, p. 389), including place, scale, network, territory and mobility (Leitner et al., 2008). Yet in recent studies examining the multiple spatialities of contentious politics, such as Halvorsen's (2017) analysis of Occupy London's spatial dialectics, networks are centred while the spatialities in which the digital parts of these networks are (re)produced become absent. This makes sense when examining, for instance, the listserv – an exchange connecting a network and ‘the primary mode of digital networking and communication in the global justice movements’ (Juris, 2012, p. 266) – but fails to account for the role of hashtags in movements such as Occupy London. Juris, discussing Occupy Boston, shows that digital spaces were central to continuous multi-site mobility, enabling Occupy Boston to harness support from around the world in ways similar to other spatially hybrid movements (Castells, 2015). Digital spaces have been used for organisation globally, including during the Arab Spring (Gerbaudo, 2012) and during tribal nationalism in Indonesia (Lim, 2017), while they also were used to ‘address layers of injustices’ (Duarte, 2017, p. 7) during the multi-sited Rio Yaqui water rights movement. Importantly for this article, after these movements seemed to end, the digital places used often persist; some continue to be used as places of organisation, others become carriers of symbolic significance (Recuero et al., 2015), and their persistence offers latent potential for the re-territorialisation of political contestation (Arampatzi, 2017).

The scholarship on political contestation regularly contains detailed analysis of how political contestation begins, but rarely offers the same analysis to the multiple endings of political contestation (Lim, 2018; Tarrow, 1993). This is despite numerous examples of violent and complex endings to hybrid political contestation that demand further attention (Castells, 2015; Gerbaudo, 2012; Lee, 2009). A recent example of complex endings is Hong Kong's 2019/20 anti-Extradition Law Amendment Bill (anti-ELAB) movement, a movement during which protesters made use of persistent digital sites of organisation even after protests taking place in physical sites across the city ended (Ting, 2020). During this movement, those engaged in contentious action – initially marches over occupations, as 2014’s occupations had been displaced (Lam, 2015) – used the protest philosophies of ‘be water’ and ‘blossom everywhere’ (Keck, 2019). Being water enabled protesters to withdraw from violent situations, allowing return at different space-times, blossoming everywhere and territorialising the movement not just in Hong Kong's centre but in neighbourhoods across the city, leading to widespread disruption. Here, the persistent presence of protesters in IM groups aided mobility through Hong Kong as anti-ELAB protesters attempted to occupy, or contest what they saw as the occupation of, the entire city of Hong Kong (Lee et al., 2021). During this movement, contentious politics included numerous spatial–temporal moments, including within digital spaces, problematising when and where political contestation began and ended. Due to the use of digital spaces, a thorough understanding of Hong Kong's anti-ELAB movement would require analysis of Telegram's ‘geography of secrecy’ (Cobham et al., 2015), of the geo-fenced spatialities of forum, LIHKG, and the role that fears over the end to freedom of digital mobility may have influenced protest tactics (Lee et al., 2021), particularly the occupation of the Chinese University of Hong Kong, home to the Hong Kong Internet Exchange. During the anti-ELAB movement, digital spaces played an important role in the multiple spatialities of contentious politics, not just as conduits of information but as sites of organisation and places to be protected.

4 DIGITAL DISPLACEMENT

Unsurprisingly, research on activist spatialities suggests that the spatial practices of contentious politics are anything but simple (Routledge, 2015). Activist spatialities – the borders of an occupation, the mobilities involved in a nationwide caravan movement and attempts to produce translocal networks of solidarity – are dynamically produced, active across traditional territorial borders and complex in how they come about as assemblages that both reflect and change the spaces they form (Leitner et al., 2008; McFarlane, 2011). But, while the impact of hybrid contentious politics has been explored and questioned (Arampatzi, 2017; Gerbaudo, 2012), the dynamic and complex spatialities of the digital spaces where these hybrid movements take place still remains underexamined. Castells (2015, p. 63) highlights the effects of internet shutdowns in Egypt, showing how routes around this repressive governance formed, but geographic research on repressive internet governance in other contexts remains lacking. One result is that a variety of spatially dynamic practices that do more than just inhibit flows of information are being described as ‘censorship’. In the shadow of censorship, many spatially oriented questions remain underexamined, unanswered or unasked, including: What happens when people are forced out of digital sites of political contestation by authorities hoping to bring an end to contentious politics? How are hybrid contentious politics affected by repressive governance in the territory housing servers and employees? How does the dispersal from digital spaces of those engaged in contentious politics affect the re-territorialisation of political projects?

Geographic literature on displacement offers insights into how people are forcefully removed from places, including during multi-scalar displacement events, such as evictions. Through violence and legal regimes, people may be forced away from spaces they desire to remain in (Brickell et al., 2017; Butcher & Dickens, 2016), with these practices of displacement resembling those used by state and non-state actors to end political contestation. While displacement has multiple spatialities, place and scale are common themes within the literature. As place-based events, during displacements, place users may be forcefully removed from places or see the places they use destroyed (Porteous & Smith, 2001; Shao, 2013). As multi-scalar events, displacement from any single site may lead to displacement across multiple scales, including mass-displacement and affective displacement from neighbourhoods (Andreas & Zhan, 2016; Crossa, 2013; Gonzalez & Dawson, 2018). Displacement is also described as a temporally diverse event with uncertain outcomes (Nixon, 2011; Shin, 2014), and displacements can be rapid and spectacular or occur stealthily over years (Asante & Helbrecht, 2020; Kern, 2016; Pain, 2019). As multiple temporalities can be involved, displacement should not be understood ‘on the basis of onetime snapshots of change’ (Elliott-Cooper et al., 2020, p. 501).

Building on these understandings of displacement, I conceptualise digital displacement as the practice of removing users from the digital spaces they wish to remain in and use. This conceptualisation of digital displacement builds on the understanding that digital relations are socio-spatial relations, not just in their infrastructural qualities, but also in how they are governed and used in everyday life (Ash, 2010; Crang et al., 2007). Underpinning this conceptualisation is the position that space is the ‘sphere of the possibility of the existence of multiplicity, [it] is never finished; never closed’ (Massey, 1999, p. 2). One potential articulation of space, as socio-spatial constructs produced by humans that take many forms, is digital space, and it is from digital spaces (and places) that space users are displaced. In this conceptualisation, coded spaces made of bits and bytes are considered a part of volumetric and materially vibrant space (Billé, 2020; Dourish, 2017; Hall, 2020). While taking part in (digital) socio-spatial relations, a user maybe within a ‘digital place’ (a Weibo hashtag) which is part of a ‘digital space’ (Weibo) and embedded within a ‘digital territory’ (China's). This user is a hybrid (de Souza e Silva, 2006), in that through ‘constant’ (Wajcman & Rose, 2011) and ‘continuous’ (Wilson, 2014) connectivity, they are always simultaneously present in physical and digital spaces, whether they wish to be present or not, and are represented in many ways, including through accounts, avatars, cookies and national identification numbers. The user may be displaced in numerous ways, with deletion of digital places a common form of displacement, while through the blocking of Internet Protocol (IP) addresses, digital spaces may be made inaccessible, displacing the previously connected user (Griffiths, 2019). Networks play an obvious role in digital socio-spatial relations, with ‘networked publics’ being ‘simultaneously a space and a collection of people’ (boyd, 2011, p. 41).

In this conceptualisation of digital displacement, the relationship between user and space is centred. Centring the spatial here means considering how the removal of people from digital space affects more than the circulation of information, it affects how sociality occurs, the forms that political contestation can take and what information is produced. Centring the spatial aspects of networked publics means remembering, as Jessop et al. (2008) and Leitner et al. (2008) remind us, that networks are in mutually constitutive relations with place, territory, scale and mobility. As space is the sphere of possibility, the displacement of user from space does more than stop the flow of information, it attempts to narrow the realm of possibility across all forms of social practice. By conceptualising digital displacement, I am not ignoring networks; instead, I am further enquiring into the effects of digital spaces on networks. For instance, what is the effect of deleting an IM group where a politically contentious network is situated? How does deletion affect the capacity for politically contentious groups to re-territorialise their projects? Furthermore, does it matter that the deleted IM group was on Weixin, a digital space territorialised in China, rather than Telegram? In short, I question if the spatial context in which a network starts, ends or is forced into mobility can be ignored.

The state territorialisation of, and development of sovereignty over, digital space plays an important role in digital displacement, as practised in China. Möllers (2021, p. 114) argues that the territorialisation of digital spaces occurs through ‘territorialization projects’, moments when ‘states mobilize scientists and engineers in order to transform globally distributed information infrastructure into bounded national territory.’ This is ‘territory as a political technology’ (Elden, 2017, p. 206), with servers and data centres offering political materiality as digital terrain (Elden, 2021). Territorialisation projects extend the volumetric boundaries of territory to digital space, and, in many cases, this extends the reach of the state. The production of digital territory is the foundation for a range of activities, including the bordering of digital space, governance over companies that control digital spaces, legal regimes that make users and their actions legible through surveillance, and the production and extraction of data for state projects (Byler, 2022; Gillespie, 2018). Territorialisation also accelerates, and further enables, the development of state sovereignty over digital spaces (Fang, 2018; Pohle & Thiel, 2020).

Digital territorialisation is not just an armchair academic debate, and the most significant discussions of digital socio-spatial relations have been happening outside of academia. Since 2010 there has been an ongoing digital socio-spatial discourse in Chinese policy (Creemers, 2020), and both domestically and internationally China has attempted to normalise ‘cyber sovereignty’ (Belarus et al., 2019). Domestic Chinese law increasingly demands the territorialisation of (critical) data, services and digital spaces in China's physical territory. Away from China, the allegedly free internet has been described as colonised by the US, particularly in the realms of data extraction, surveillance and infrastructure (Couldry & Mejias, 2019; Kwet, 2019). One response has been European Union territorialisation projects, including the General Data Protection Regulation (GDPR) (Floridi, 2021; Tikkinen-Piri et al., 2018). In India, fearing both US and Chinese influence, Chinese applications have been banned and WhatsApp threatened with ban (Dhillon, 2020; Kamra & Williams, 2021). Even the US, under Donald Trump, moved down a path of territorialisation by threatening a digital border to exclude Chinese companies (Trump, 2019), a significant departure from the anti-digital sovereignty position traditionally taken by US policy-makers (Couture & Toupin, 2019). US avoidance of the digital territory/sovereignty debate and US advocacy of a Barlowian-style (Barlow, 2016) free internet is not surprising as US-based institutions benefit the most from the status quo. At present, there is a greater than average chance that an internet user anywhere in the world is using a space, service or hardware infrastructure operated by a US company (de Seta, 2021).

In this spatialised view of digital relations, where, then, does digital displacement occur? So far, this article has suggested that the displacement of contentious political events is possible, but that digital sites of contentious political action seem to persist. This is a geography of contentious politics that is simultaneously situated in hybrid physical and digital sites, a contentious politics in which ‘once the movement has extended its reach from the space of flows to the space of places, it is too late to stop it’ (Castells, 2015, p. 67). In this situation, territory seems insignificant as people around the world flow to spaces such as Twitter and Telegram. Explored through the prism of anglophone and Global North political contestation, digital displacement seems relatively absent, as wherever the body of the protester is in physical space, the digital site they use is either ensconced in the digital territory of the US, operating under US political norms, or, outside of the territorial control of the authorities being protested against. Furthermore, as the operator of the digital space of protest profits from these interactions, they have few incentives to practise digital displacement. First, it is important not to homogenise and view with rose-tinted glasses the US experience. American companies have been found to regularly delete activist content, deletions that inhibited the circulation of information related to murdered indigenous women and contentious activity in Colombia and Palestine (Gebeily, 2021). Elsewhere, in Cambodia, close links between local Facebook staff and government representatives have allegedly ‘streamlined’ the processes that delete and displace political content and groups (Rajagopalan, 2018). More recently, the displacement of Parler users⁶ became an example of mass-digital displacement in the US (Floridi, 2021). These cases suggest that digital displacement is as an emerging practice in US digital territory, but that when those displaced reside outside of US physical territory, their displacements may become less visible.

Shifting the analytical lens away from US digital territory and towards Chinese digital territory makes visible regular acts of digital displacement (King et al., 2013; Liao, 2019; Lin & Yang, 2019; Roberts, 2018). The deletion of hashtags is one common method of digital displacement, a mode of governing through digital space that ‘affects yet decentres’ (Morris, 2022 2022NaN, p. 2) the body and the user's account. In attempts to re-territorialise activist projects after displacement, protesters make use of semantically similar terms, homonyms, English translations and transliterations to create new spaces of protest (Meng, 2011).⁷ Understood through the logic of censorship, where the control of information is centred (Roberts, 2018), the deletion of synonyms and homonyms that only make sense to those already within movements, which are difficult to find unless one has pre-knowledge of where to look and that have limited information-sharing potential due to their complex linguistic formation, does little to affect the flow of information. But as places where solidarity, action and world-making can occur, places in which others are encouraged to share their stories of inequality or practise the changes they wish to see in the world, displacement becomes a powerful tool in stopping nascent movements growing. In China, this makes hashtag activism and digitally centred politics a tale of displacement and deletion, not one of long-term struggle in persistent digital place.

5 THE DIGITAL DISPLACEMENT OF #BEIJINGSURGERY#

To examine the multiple forms that digital displacement takes and digital displacements' multiple effects, I turn to the efforts of two groups that used digital spaces to contest the mass eviction of migrants from Beijing: #BeijingSurgery# and BeijingTogether. On 18 November 2017, a fire in the Beijing urban village of Xinjian killed 19 people, 17 of whom were officially classified as migrants, holding non-local hukou.⁸ This event was the first of many tragedies, and the day after the fire, Beijing Party Secretary, Cai Qi, announced a 40-day ‘Citywide Safety and Hidden Dangers’ campaign (Luo, 2017). Mass evictions followed, and tens of thousands of mostly low-income, translocal migrants working in Beijing were displaced – the so-called ‘low-end population’ (Rudolph, 2017). Understanding the evictions as politically motivated,⁹ many formal organisations considered relief efforts dangerous. Alongside this, reporting of the evictions was censored, in news rooms and digital spaces (Wade, 2017). These two gaps – of care and enquiry – were filled by informal groups, including the two groups analysed here: the anti-eviction and eviction mapping hashtag, #Beijing Surgery#, created by Pinger and Xiangyun, and the IM group, BeijingTogether, led by Mingyan.¹⁰

After visiting the rubble of Xinjian in the days following the fire, Pinger and Xiangyun discussed what they could do to help the displaced. Reflecting on eviction mapping projects seen during university studies, they concluded that a Weibo-centred hashtag project focused not on active contestation, but rather on data collection, may succeed. The original post that began #BeijingSurgery# contained only two things, the textual hashtag and a long-image.¹¹ This long-image contextualised recent events while also explaining the goals and practice of the project to readers: include eviction data in posts which are added to the hashtag and repost relevant posts into the hashtag. After the 40-day campaign finished, the long-image noted, a map would be released. The name, #BeijingSurgery#, had strong affective and narrative qualities (Yang, 2016), and, due to these qualities, the ambiguously named hashtag evaded keyword censorship and came into being. #BeijingSurgery# was a place for those affected by evictions to share data and make a map, but it became a place where stories were shared. #BeijingSurgery# was not just information; it was a place where knowledge was produced and a site where people could organise, criticise and store eviction data. Within 24 h, and after #BeijingSurgery#’s initial influencer-aided surge in popularity, the hashtag was deleted. Posts that took part in the hashtag were deleted or shadow banned (Are, 2021), and, while some data was saved by organisers, the majority of data was lost, users were displaced from #BeijingSurgery# and this nascent network was homeless. Where people had previously entered the hashtag through hyperlinks or searches and then explored and took part in the discussions that were taking place, this was now impossible. This was hashtag as a space of activism and discussion, and it was gone.

#BeijingSurgery#’s displacement was twofold. First, #BeijingSurgery# was deleted and those who used it were displaced from the hashtag. The idea that place, not person, was targeted is reinforced by the fact that no #BeijingSurgery# personal accounts were deleted from Weibo, just hashtags and posts. Second, those wishing to discuss #BeijingSurgery# were effectively displaced from Weibo at multiple scales, as keyword censorship would stop new posts and hashtags that used language or imagery related to #BeijingSurgery# from being posted. Thus, anyone wishing to re-territorialise or discuss #BeijingSurgery# would have to leave Weibo: a multi-scalar displacement effect forcing people from Weibo.¹² These deletions made Weibo a space where information could circulate but where the coming-together of people for political contestation was, by degrees, made more difficult.

#BeijingSurgery#’s deletion is tied to Weibo's territorialisation and the Chinese state's digital sovereignty. While #BeijingSurgery#’s deletion was likely done by Sina Weibo employees, research on similar forms of deletion suggests that deletion is practised to follow regulations and the political mood (Chen, Liu, et al., 2018; Chen, Mao, & Qiu, 2018). Political interpretation is necessary as powerful regulations, centred around the Cybersecurity Law, include references to ‘core socialist values’, and companies not governing in accordance to these values may suffer ‘corrective measures’, including ‘a temporary suspension of operations, a suspension of business for corrective measures, closing down of websites, revocation of relevant operations permits, or cancellation of business licenses’ (Creemers et al., 2017). For instance, due to ‘vulgar’ content, the large, ByteDance-owned joke-sharing space, Neihan Duanzi, was permanently shuttered (Ho, 2018). The development of state sovereignty over China's digital territory made #BeijingSurgery# politically problematic for Sina Weibo, leading to deletion and digital displacement.

But #BeijingSurgery#’s deletion was specific to Weibo, and, though the project was hashtag centred, a secondary Weixin administration IM group was accessible through a QR code embedded within the long-image. When scanned the QR code could take users to Weixin, and, aided by this QR code, a massive relocation from Weibo hashtag to Weixin IM group began. Relocation was not without problems, as, while QR codes offer the potential for multi-site mobility, after more than 99 people are in a Weixin group associated with a QR code, the code stops functioning: a design decision that does not stop information spreading but that does stop large groups of people gathering within digital places. But, if displacement from Weibo was done to end political contestation this had failed, and #BeijingSurgery# had successfully re-territorialised the project in Weixin after the initial displacement from Weibo. Within 72 h of displacement, #BeijingSurgery# included four main Weixin IM groups, several administrative groups, data-cleaning groups, a secondary administration group on Telegram, over 1500 people and data storage outside of China's digital territory, on Telegram and Google Drive. All data on evictions posted into the groups were automatically scraped by a bot¹³ and then saved outside of China's digital territory in multiple digital sites, with most of the labour done by those within China's physical territory. These territorially non-Chinese digital spaces were accessed through the Virtual Private Network (VPN), a technology that in this context becomes one of transnational mobility between digital territories. #BeijingSurgery#’s experience shows that digital displacement is an effective tool to end place-based political contestation, but, if the potential for deletion and displacement is considered, political contestation can continue in alternative digital spaces. For #BeijingSurgery#, displacement became one moment in an evolving protest cycle, and mobility to Weixin shows that displaced people and groups can have afterlives, even when displacement is almost instantaneous.

6 DISPLACING BEIJINGTOGETHER

#BeijingSurgery#’s displacement from Weibo and the subsequent re-territorialisation in Chinese digital territory through mobility to Weixin IM groups had unforeseen consequences, notably, changing from a horizontally structured hashtag model of organisation to a bordered group with a single ‘Group Leader’. Due to recent regulations on IM governance,¹⁴ this Group Leader was legally responsible for everything that happened in the group. Displacement from Weibo also changed how project leaders and many members viewed the threat of state violence. Deletion was now an eventuality, though it could be delayed through in-group moderation, moderation practised due to fear of displacement and a communal desire to continue the project. In one instance of in-group moderation, project leaders requested a stop to discussions about halting evictions in east Beijing's Pi village. After multiple requests to cease discussing Pi village, the QR code for another IM group, Tour Group, was shared into #BeijingSurgery#. Members of Tour Group said Pi village could be discussed there and, once again, people moved. As day turned into night, a small number of citizens converged on Pi village to critically observe the planned evictions, and, by the time I arrived in Pi village that night, the village government had announced a delay to evictions. Many in Tour Group believed their pressure, through the temporary territorialisation of Tour Group and other IM groups in Pi village, helped achieve this outcome. Members thought this spatial practice could be replicated, and with this in mind, within 24 h Tour Group had a new name, BeijingTogether, a new Group Leader, Mingyan, and a new motto: ‘Go to the scene!’ Over the next week the group grew to almost 500 members, with group members going to ‘scenes’ across Beijing. This practice was aided by smaller, neighbourhood-focused sub-groups that emerged from within BeijingTogether.

As an IM group, BeijingTogether's manifesto and group entry dissemination model evolved, with BeijingTogether using a website alongside a QR code. This website, inspired by multi-level marketing tactics, included a form through which people could enter their Weixin details to gain entry to BeijingTogether. This enabled BeijingTogether to work around Weixin's QR code entry, spreading an entry key that would continue to function. To understand who was in the BeijingTogether and #BeijingSurgery# Weixin groups, I employed IP-masked, anonymous surveys with data stored outside of China's digital territory. Of the nearly 200 respondents, the majority were born after 1985 and were either completing or had completed an undergraduate or postgraduate degree, and over 90% of members had, or were working towards, an undergraduate degree. While the vast majority of participants reported monthly wages of less than 10,000 RMB,¹⁵ roughly 43% reported monthly parental income of over 10,000 RMB. These statistics suggested a somewhat homogeneous group of educated young adults, but the groups were home to a variety of people from across the political spectrum, and as one influential interviewee involved in both #BeijingSurgery# and BeijingTogether noted, ‘those who would be sworn enemies elsewhere were there, leftists [and] righties.’ With this foundation, BeijingTogether territorialised a space of solidarity, debate, theorisation and action within China's digital territory, showing that solidarity and empathy were possible (Arampatzi, 2017; Routledge, 2015). While the overall aim of BeijingTogether was to use the spatial practices suggested in the group to affect Beijing's physical territory, according to discussions in the group, visits to eviction sites by members looking to offer help regularly achieved little. Project members cited a lack of trust towards their motives and an inability for project members to understand the needs of evicted migrants as reasons for the project's inability to affect ‘the scene’. BeijingTogether found that the initial place-making and territorialisation of the IM group within Chinese digital territory was relatively simple but moving the project ‘offline’ – temporarily territorialising the project in physical spaces through spatial practice – more difficult.

Almost 3 weeks after the Xinjian fire and with no warning, BeijingTogether was deleted and the members displaced. BeijingTogether was deleted not by Tencent but by Group Leader Mingyan. Mingyan's decision to delete BeijingTogether came after a group of university students, gathering data for student projects, were questioned by local police officers in east Beijing. Soon, members of like-minded digital projects residing outside of Beijing reported their own interrogations, interrogations including chatlogs from Weixin groups. By suddenly deleting BeijingTogether and displacing everyone, Mingyan was not controlling abstract flows of information, but attempting to remove people from a digital place that could bring danger through association. The (self-)displacement of BeijingTogether project members by Mingyan was an action that could not be reversed, nor was it a decision Mingyan desired to make. Instead, the territorial realities of BeijingTogether and its members influenced Mingyan's decision-making; with members bodily in China's physical territory and the IM group data in China's digital territory, authorities had access to both body and data. This differs from contentious politics where body and digital place of resistance are in separate territories.

Influenced by BeijingTogether's ‘socio-spatial positionality’ (Leitner et al., 2008), Mingyan followed ‘guerrilla’¹⁶ tactics used by long-term activists in China. In doing this, Mingyan chose to sacrifice an already territorialised foothold in Weixin – one that could be deleted at any moment – for a continued presence within Chinese digital and physical territory alongside a new network of associates. BeijingTogether's deletion also had wider displacement effects, as, while the deletion of BeijingTogether did not result in the deletion of BeijingTogether's smaller neighbourhood-focused sub-groups, the remaining sub-groups fell relatively silent post-deletion even though they were still accessible: an end to the protest cycle. Here we see once again that digital displacement can have unanticipated chain reactions, silencing activists across multiple spatialities. BeijingTogether's experience eventually resulted in smaller informal networks emerging, using the practice of digital flash mobbing to hold events in temporary IM groups, but while the temporary coming together of people enabled discussions to be had and different voices to be heard, it is difficult to know how sustainable this cycle of protest will be. There is a logic to Mingyan's actions, but it is difficult to know if, when employed in Chinese digital territory, the guerrilla spatialities and spatial practices espoused by long-term activists in China enable activists to side-step repressive state governance or instead aid state governance through self-displacement.

7 CONCLUSIONS

For both #BeijingSurgery# and BeijingTogether, displacement was not the end of political contestation, but it did make continuity difficult. BeijingTogether's experience shows that displacement can be an action one is coerced into through fear, that it can be rapid, affect hundreds of users and seem final. BeijingTogether's displacement also highlights the uncertainty of digital displacement, in that deletion arrived from an unexpected direction. More broadly, these data suggest that in Chinese digital territory one can produce but not easily reproduce places of contentious political activity, that digital space and place are fragile, and that state and corporate actors can control the reproduction of digital place and the re-territorialisation of contentious political projects if sovereignty over digital territory is achieved. The effect of reproduction being tightly governed is that, to attract users into a controlled system, the production of place is made relatively easy, offering opportunities for new activist spatialities to be produced. As governance of newly produced places and information is rarely instantaneous, contentious political networks can produce temporary places where moments of resistance take place. If they use visual, linguistic and narrative tactics to seem mundane – as employed by #BeijingSurgery# – the moment of resistance can be extended. Within a system of spatial governance, displacement will continue to be effective (Morris, 2022), but reliance on deletion, displacement and controlling the reproduction of space means opportunities for moments of creative resistance.

The experiences of #BeijingSurgery# and BeijingTogether paint a picture of digitally centred contested politics that differs from the Twitter and Instagram experiences of #Occupy and #BlackLivesMatter. To understand #BeijingSurgery# and BeijingTogether, it makes sense to look towards Admiralty, Tahrir Square and BlackLivesMatter street protests; political contestation where space users were forced from their sites of protest. My case studies also highlight a key difference in the spatialities of contentious politics between physical and digital sites, in that, while counter-movements could be made to reclaim and re-territorialise protest in the streets of Athens (Arampatzi, 2017), #BeijingSurgery# could not re-occupy their deleted hashtag. Similarly, the displaced members of BeijingTogether struggled to find each other through Weixin, as project members had been connected through an IM group they were now displaced from and not their IM accounts. In discussing deletion, it can be argued that these deleted places still exist somewhere, in a database controlled by operating companies or the Chinese state, but they exist only as artefacts: they are displaced activist spatialities existing as data artefacts hidden from their original users.

Finally, these accounts of digital displacement contribute to geographic understandings of displacement by exploring how resistance to displacement is, in certain spatial contexts, made impossible due to the speed of displacement and an overall control of space. Current accounts of displacement elaborate on the capacity for the (to be) displaced to resist their displacement, including through occupation and (im)mobility (Gillespie et al., 2018; Shin, 2014). Reflecting on this article's findings, I encourage scholars and activists to consider how unexpected and almost invisible forms of rapid displacement can be researched and/or resisted, and how place can be (re)produced when corporate actors, potentially aligned with state interests, have high levels of control over space. The frameworks outlined in this article also make it easier to track, compare and understand the relationship between governance in territorialised digital spaces constructed out of code and ‘hybrid spaces that are co-created, transduced, and augmented by the digital’ (Graham, 2020, p. 453). Thus, while this article theorises from China, this does not mean that its findings cannot help us understand other geographic contexts, including displacements within US digital territory or other physical territories. With platforms playing an increasingly significant role in everyday urban development around the world (Datta, 2020; Leszczynski, 2020b), it is important to examine if and how the modes of governance used within the digital spaces – a quick deletion here, and ‘accidental’ removal there, a blacklist that stops those actions being discussed everywhere – influence governance in other spatialities (McNeill, 2021). Ongoing research within digital geography helps understand ‘social control in the networked city’ (Iveson & Maalsen, 2019), but, in China at least, the companies that regularly delete and censor are also the ones creating the hybrid city. Examining how such companies govern political contestation in the digital spaces they control may help us understand how they will deal with political contestation in the hybrid urban spaces they co-produce.

ACKNOWLEDGEMENTS

The author would like to thank Laura Antona, Andrea Pia, Chenchen Zhang, Hanlun Li, Eden Palmer, Peter Gries, Yawei Zhao, Yueming Zhang, Claire Mercer and Hyun Bang Shin for commenting on this paper and the ideas presented in it at different stages.

Endnotes

¹ The term contentious politics stands for various forms of “concerted, counterhegemonic social and political action, in which differently positioned participants come together to challenge dominant systems of authority” (Leitner et al., 2008, p. 157), and here it includes social movements and activism.

² Chinese micro-blogging space Weibo uses hashtags on either side tagged text.

³ Weibo is the most popular micro-blogging service in China. It is Twitter-like in practice but has different spatialities.

⁴ Weixin is the most popular instant messaging service in China and has become an essential infrastructure in everyday personal and professional life. It is sometimes called WeChat, but WeChat is the international version of Weixin, and it has distinct differences in governance.

⁵ According to the surveys conducted.

⁶ Parler, a Twitter-like service used by those supporting or associated with Donald Trump's politics, went offline on 10 January 2021, with Amazon Web Services cancelling its hosting services. This occurred 4 days after its users played a role in the 2021 US Capitol attack. The service resumed in February, for over a month its users were temporarily displaced.

⁷ For instance, #RiceBunny# is an English translation of the deleted #MiTu#, which is a homophone of the already deleted #MeToo# (Yuan Yang, 2018).

⁸ For information on Household Registration, see Chan and Zhang (1999); Zhang (2018).

⁹ Related to national policy on population controls (Chen, Liu, et al., 2018; Chen, Mao, & Qiu, 2018).

¹⁰ All names are pseudonyms. All four are young adults with university educations, three females and one male. They had little or no experience in activist circles and had no activist affiliations. All were familiar with issues of activism and the politics involved and had politically active acquaintances.

¹¹ A long, vertically scrolling image often made through screenshots. It can be shared and acts as an article, working around the constraints of text posts that limit how text and imagery can be presented. Long-images also offer low-levels of protection from keyword censorship (Liu & Zhao, 2020).

¹² Numerous semantically similar hashtags to #BeijingSurgery# were deleted, including hashtags with typos, spelling mistakes and grammatical errors.

¹³ Designed by a project member.

¹⁴ See Internet Group Information Service Management Provisions (2017).

¹⁵ This is US$1547 a month, with 1 RMB equalling US$6.46.

¹⁶ Describe by Mingyan in this way.

Open Research

DATA AVAILABILITY STATEMENT

Data sharing is not applicable to this article as no new data were created or analyzed in this study.