After riots: Toward a research agenda on the long-term effects of urban unrest

ABSTRACT Urban societies are experiencing intensified social frictions which have manifested in a growing occurrence of riots. Urban riots are often treated as ephemeral outbursts which create only short-term or immediate responses by the state or residents. However, this paper aims to move beyond such present-centrism by proposing a research agenda for studying their contextually longer-term legacies. Particular consideration is given to three thematic areas: (a) policy responses, (b) temporalities and urban memory, and (c) feelings and subjectivities. Drawing upon empirical illustrations concerning riots in London (2011) and Stockholm (2013), the paper uncovers variations within and across the two cities, such as differing state and private sector interventions; reshaping of civil society; and how residents variously resist, adapt to, and promote change. The paper thus reflects upon how riots’ afterlives in different urban contexts are constructed, mobilized and contested. Such variations point to the importance of the situated study of urban riots’ diverse legacies.


Introduction
In one of few major studies of the long-term impacts of urban riots, 1 the sociologist James Button (1978, pp. 3-4) noted: [T]he causes, precipitating events, and participants of [the 1960s riots in the U.S.] have been thoroughly studied over the past several years. Yet what is remarkable about this extensive analysis is the almost complete neglect of the political effects or consequences of these pervasive disorders. . . . For many, such conflict has been viewed as . . . the antithesis of normal political participation. This view has effectively precluded rigorous attempts by social scientists to explore fully the results of the 1960s urban upheavals.
In this paper, we argue that the same remains true today, and outline a research agenda for addressing this lacuna. Existing research shows a wealth of insight into the structural causes, immediate triggers, and in-situ dynamics of riots, but a substantial knowledge gap exists in our understanding of what happens in the months and years afterward. This "present-centrism," concerning the immediate dimensions of riots, overshadows the potentially significant after-effects, legacies and impacts for the communities affected. Therefore, in this paper, we use two recent eruptions of urban rioting-London in 2011 and Stockholm in 2013-as a lens through which to develop a more robust and systematic research agenda for understanding the multiple long-term urban legacies of riots. Such an agenda, we argue, needs to bring together formerly-disparate literatures in order to grasp their full range and complexity.
The growing urbanization of the world's population is arguably leading to an "urban age" which brings with it a range of intensifying effects, including, as Dikeç (2017) argues, urban rage. The proliferation of both mass and individual forms of urban violence in the Global North leads to a sense of urgency with regard to the need for reflection on the role and significance of popular violence in this urbanizing world. With urban communities still feeling the uneven after-effects of the financial crisis of [2007][2008], and the longer-term unevenness of neoliberalization, scholars have argued that riots are an increasingly prominent occurrence (e.g., Dikeç, 2017;Mayer et al., 2016;Moran & Waddington, 2016). Dominant regimes in the Global North are producing deepening levels of inequality and marginalization, while neoliberal governance at the urban scale prioritizes competition and entrepreneurialism, seeking to remake cities according to (arguably unattainable) ideals of the "world-class city" (e.g., Belina & Helms, 2003). In this context, segregation, gentrification and social polarization have intensified (Hedin et al., 2012;Simpson, 2004).
However, "actually existing neoliberalism" (Brenner & Theodore, 2002) varies geographically, producing a multitude of locally-specific conditions, struggles, and cultures as populations seek to resist, adapt to, or capitalize on the particular variation of neoliberalization that they face (e.g., Borén & Young, 2013;Cumbers et al., 2016). This, for some, represents a diversification of the urban itself; a fragmentation of the singular "cityness" that is commonly assumed (Brenner & Schmid, 2015). As such, manifestations of popular agency in this context are diverse, as shifting urban regimes "transform both the environment . . . in which [social] movements operate, and the movements themselves" (Mayer, 2009, p. 362), doing so in diverse ways according to local, regional and/or national politicaleconomic conditions. In recent years, this confluence of processes has therefore resulted not only in highly politicized movements such as Occupy! or Black Lives Matter but also a wide array of spontaneous or unplanned eruptions of popular expression that sometimes manifest as, or feature, riots.
In this paper, we push the boundaries of existing research on urban riots and uprisings by presenting an agenda for analyzing the diverse long-term effects of such events. Positioning riots as part of a continuum of political expression, rather than a temporary rupture from "real" politics, allows scholars to see them clearly as events with material, policy, and psychosocial consequences. First, we review the literature on contemporary urban riots, noting how the after-effects have been decidedly under-explored, especially in comparison to research on other disruptive events such as natural disasters and wars. Taking inspiration from this latter research, however, sheds light on three key dimensions of riots that can form an analytical framework for studying their legacies: policies, temporalities, and feelings. These are not aspiring to be an exhaustive list of important issues, rather they represent themes that we find missing in urban critical scholarship in relation to the afterlife of riots and are pertinent to understand the wider dynamics of socially-polarized and neoliberalized cities. Following this section, we briefly introduce two cases-London (2011) and Stockholm (2013)-that provide short empirical illustrations for the three identified themes and for illuminating the contextually-specific effects that riots can have on the longterm social and material landscape of urban communities. The three themes then form a series of sections that outline relevant literatures and key issues for future research on contemporary riots. The concluding note reflects upon how the three dimensions intersect and identifies possibilities for future work.

Present-centrism in the literature? Causes and dynamics of riots
Given the turbulence and uncertainties of the last decade or so in the Global North, it is not surprising that riots have become increasingly common. However, the long-term significance of riots remains less clear, with most scholarship seeking only to explain underlying grievances and triggers. These factors tend to orbit class (Scambler & Scambler, 2011), ethnic inequalities (Amin, 2003), and problematic urban governance (Dikeç, 2007(Dikeç, , 2017. A substantial literature has also grappled with the dynamics of riots themselves and how to model or predict their occurrences (e.g., Holdo & Bengtsson, 2020;Loch, 2009;Newburn, 2016;Till, 2013), which also highlight a range of other dynamics beyond the immediate event-space of the riot. This wider literature identifies important dynamics but ultimately prioritizes the immediate temporal field before, during, or after riots, thus often missing the persistent significance of riots for urban communities.
A large part of the riots literature concerns their framing within popular, policy, and media discourses. For example, a number of texts concerning the 2014 unrest in Ferguson, Missouri, have shown how the media shape and drive often highly racialized discourses with regards to what are deemed "legitimate" and "illegitimate" forms of political expression (Blackstone et al., 2017;Mills, 2017), as well as the diverse ways in which social media was mobilized by supporters and opponents of the protesters (Bonilla & Rosa, 2015). These studies parallel similar engagements with the discourses and media coverage around the English riots in 2011 (e.g., Baker, 2012;Bennett, 2013;Phoenix & Phoenix, 2012). Importantly, as Baker (2012) notes, their mediation through social media and mobile technologies is a distinctive new dimension of riots in the contemporary period, echoed by studies examining movements such as the Arab Spring (e.g., Fuchs, 2012).
Many studies emphasize the intersections of race and class in constituting riots through urban inequalities and concentrations of deprivation, albeit "without immediately resorting to [explicitly] political claims" (Body-Gendrot, 2012, p. 6;cf. Brink Pinto & Ericsson, 2019). Classed and racialized patterns of segregation and ghettoization, for example, have been observed as a key factor in the instigation of riots, not only in Stockholm (Malmberg et al., 2013) and London (Phillips et al., 2013) but also elsewhere (e.g., Body-Gendrot, 2012;Gershenson & Hayes, 2018;Olzak et al., 1996;Young et al., 2014). With the majority of rioters tending to be young men, gender also looms large in the structural dimensions of riots. Ashe (2014) has noted how conservative commentators laid blame on the family unit in the English riots, using a lack of a male "authority figure" to explain the perceived failures of single mothers to control their unruly sons. Such discourse helped politicians to avert the popular gaze away from structural inequalities exacerbated by austerity, and toward a pathologizing explanation based on individual dysfunction (cf. Alexander, 2004).
Uneven processes of neoliberalization, globalization, and austerity governance often loom large in studies of riots in the period since the 2007-2008 financial crisis (see Mayer et al., 2016). Gilroy (2013) reprimands London's rioters for their extensive looting, accusing the rioters of embodying neoliberal subjectivity through "money-free shopping and pseudoinsurrection" (p. 556). Conversely, others point to the subtle neoliberal governmentality that manipulated riot-stricken communities through notions of "resilience" (Rogers, 2013): with rioters positioned by authorities as somehow outside the imagined community, consensus among the non-rioting majority could be easily manufactured for the purpose of maintaining social peace amidst the forced scarcity of harsh austerity policies. In Stockholm, scholars identify the neoliberalization and deregulation of this "welfare-state city" as a key factor in what Schierup et al. (2014, p. 9) call "the production of a marginalised reserve army of labour," who were capable, willing, and had little to lose from rioting (cf. Sernhede et al., 2016).
Cutting through much of the literature is a concern with the relations between the state and its subjects. From the molding of the passive neoliberal public (Wallace, 2014) to the technicalities of the criminal system (Finchett-Maddock, 2012), riots expose various forms of disconnect within the perceived stability in the Global North of the "social contract" between state and citizenry (Dikeç, 2007). Riots also show how cities such as Stockholm and London-popularly understood as inclusive, secure, and welcoming-have deep fissures, and that riots can shatter their carefully-crafted place branding in a world of competitive urbanisms (Bhattacharyya, 2013). Indeed, the failures of urban planners in particular to fully engage with communities at risk of displacement in these competitive cities is well documented, including where riots have taken place (e.g., Hansson et al., 2013). This has led to calls for scholars to reposition riots as potential "catalysts for change" (Phillips et al., 2013, p. 8;cf. Sokhi-Bulley, 2015). By foregrounding marginalized groups' voices and agency, it may be possible to investigate the extent to which riots may indeed catalyze change.
We therefore see a rich literature in urban studies and beyond engaging with the societal contexts, triggers, and in-situ dynamics of urban riots, alongside their immediate popular, policy and media reception. This is important work, yet what is less explicit is a sensitivity to the lasting consequences of riots. Riots, as with any other form of political expression, are usually intended to enact some kind of change, even if the changes desired are neither specific nor explicitly articulated or political in a traditional sense. Our concern in this paper is that this research gap is substantial and that it ultimately means scholars fail to grasp the full significance of riots. Some studies have engaged with the immediate planning (Dillon & Fanning, 2012) and policy responses (Newburn et al., 2018;Rich, 1993), but not only do these focus primarily on institutional actors but also they focus on immediate responses. Other studies consider the longer-term effects of riots but tend to do so as a side-note within a different story (e.g., Panton & Walters, 2018).
There have been efforts to study the long-term quantitative economic effects of riots (e.g., Collins & Margo, 2007), which provide important context but tell us very little about everyday life in cities and especially specific neighborhoods. In contrast, Frost and Phillips (2012) engagement with the afterlives of the 1981 Toxteth riots in Liverpool is distinctive, using the legacies of 1981 as a way of shedding light on 2011's riots. By demonstrating the political significance of 1981 and its enduring memory, the authors undermine politicians' narratives of 2011, which echoed similar narratives in 1981 that characterized the riots as simply "wanton destruction." Moreover, the far older sociological work of Button (1978) confirmed the political significance of riots long before this particular historical period of urban insurgency. In this text, he identified impacts on popular attitudes (e.g., toward violence, education, crime), policing, housing, and local/national policies in other fields, as well as how different cities and communities responded differently to ostensibly the same phenomena.
Perhaps the field in which the long-term effects of riots have been given the most attention is social and political history. In this literature, historical riots are presented as responses to, and occasionally drivers of, the development of societies in the past. These works draw on archival material to piece together developmental dynamics in cities such as mediaeval enclosures of common land (Liddy, 2015), colonial cities and decolonization (Warren, 2001), and race relations (Kusmer & Trotter, 2009). The politics of rioting is also highlighted in some historical texts, which outline how the political cultures and political economies of cities shape and are shaped by popular action (e.g., Clover, 2016;Smith & Mitchell, 2018;Sernhede et al., 2016;Brink Pinto & Ericsson, 2019; see also Dikeç & Swyngedouw, 2017). With a gaze that spans decades or more, such work tells valuable and detailed stories of how popular violence is woven into the development of urban societies over time, and warrants caution against the present-centrism and the narrow temporal frames that dominate much of the literature examining contemporary riots. However, a frequent emphasis on macro-scale and top-down governmental changes in this historical literature also prompts us to pay closer attention to the everyday experiences and their lived effects of contemporary riots in specific urban communities.
While the employment of an extended temporal frame is, for obvious reasons, difficult to apply to present-day riots, analyses of these events, we propose, nonetheless ought to move beyond a concern with their immediate effects (typically those lasting from weeks to months). Where possible, studies need to explore transformations emerging over a number of years after the events, which is what this paper refers to as the "long-term" effects of riots in contrast to the timeframe of immediate responses and reactions. Such a temporal frame has at least two strengths: (1) it allows sufficient time for policies to be proposed, enacted, implemented, and widely experienced by those affected; (2) it is temporally close enough to the events for a large number of individuals close to the riots (e.g., residents, policymakers, civil society organizations, and researchers) to remain in the area and have sufficient memory of life both before and after the riots. This framing of "long-term" therefore is contextual in relation to the immediacy of initial responses to riots (short-term) and lived experiences over a significant period of time. In the next section, drawing from literature across geography, planning and the social sciences, we propose an agenda for research into the long-term effects of urban riots.

Learning from natural disasters
In this section, we draw on relevant literature from elsewhere to construct a more robust research framework to enhance scholars' understanding of the after-effects of riots. One body of literature that has taken great care in studying the long-term effects of disruptive urban incidents has been from the field of natural hazards and disaster recovery. This subject is distinct from riots because natural disasters are usually not instigated by human activity (even though the nature and extent of their damage is certainly shaped by human (in)actions). This means that a large part of the literature concerns what happens after them.
The material impacts of hurricanes, floods, earthquakes, and other natural disasters are undeniable, yet they also create wider social, economic and political effects that are often unevenly felt according to class, gender, ethnic, and urban-rural differences (e.g., Bolin & Kurtz, 2018;Cutter et al., 2016;Enarson, 2012;Masozera et al., 2007). These intersecting dimensions of the impacts of disasters, perhaps unsurprisingly, parallel closely those identified as causes of riots. This literature also highlights the role of psychological and psychosocial impacts on individuals and communities, emphasizing the way that trauma, emotions, and affects can shape and impact on not only people's sense of belonging and community but also more "material" recovery processes related to issues such as infrastructure and health (e.g., Cox & Perry, 2011;Lindell & Prater, 2003).
Nevertheless, much of the disaster recovery literature emphasizes the work of institutional actors and focus less on how others-individuals and communities-seek to "get by" and rebuild their lives during the years following disasters. Taking a specifically urban lens, the impacts of Hurricane Katrina (and Rita) on New Orleans in 2005 stands out as anomalous in this regard. In New Orleans, a number of key dimensions are apparent. First is the racial and class inequalities that were exacerbated as a result of the disaster, and not sufficiently addressed during the city's recovery and reconstruction (e.g., Masozera et al., 2007;Seltzer & Nobles, 2017). Another key issue is the criminalization of black communities-particularly young black men-both in the immediate aftermath of the hurricane and the years that followed (e.g., Berger, 2009). This criminalization helped to create discursive, policymaking and physical space for city planners and private developers to engage in redevelopment schemes that have (not unlike London, as we shall see) caused displacement of majority-black working-class communities (Gladstone & Préau, 2008) and the social cleansing of central areas for touristification (Fox Gotham, 2017). As a result of touristification in particular, and the accelerated commodification of the creole and African American musical traditions of New Orleans, memory became a central battleground between popular and institutional narratives of the city's unique identity and heritage (e.g., Thomas, 2009).
This ballad of New Orleans is particularly severe, but it raises a number of important themes that also run through the wider literature on urban responses to disasters: changes to policy agendas and implementation, the long (and still ongoing) timescale of Katrina's legacy, and changes to lived experiences of the city. It has been rightly argued that riots and natural disasters cannot be easily compared empirically or conceptually, since they are distinct at "the individual, organizational and community level" (Quarantelli, 1993, p. 67). Nevertheless, returning to the literature on urban riots, the few studies that have engaged with the legacies of riots certainly address similar themes. While there are undoubtedly other dynamics that have relevance, taking inspiration from a large and wellestablished body of scholarship on aftermaths of disasters is a firm basis for nurturing new research directions. These themes are also important aspects of understanding the role of riots for processes of political subjectification of people and places in the city, thereby indicating further significance for the study of such insurgencies. This paper therefore proposes an agenda for researching the long-term legacies of urban riots based around these three intersecting dimensions, which we refer to as: policies, temporalities, and feelings. This does not preclude other dimensions that we do not explicitly explore here, such as gender, materiality, or other spatialities; rather, the themes we identify should be seen as a starting point for wider debates and literatures to emerge. Following a brief introduction to two empirical case studies in Stockholm and London, we address each theme in turn, illustrated by observations from our empirical research.

Urban riots in context: London and Stockholm
London has both driven and experienced a long process of neoliberalization since the late 1970s. After decades of urban policy characterized by social-democratic principles and attempts to reduce urban and regional disparities, urban development strategies increasingly prioritized the private sector. With the lifting of redistributive regional policies, London experienced a great boom in speculative finance and real estate capital, accompanied by a marked rise in white-collar businesses and leisure services (Allmendinger, 2016). International investments in London's real estate in a context of deregulation-fueled price inflation in the housing market and contributed to a process of "super-gentrification" that further reinforced high real estate value (Deverteuil & Manley, 2017). At the same time, the provision of social housing declined in London and the UK more generally. These enduring trends have contributed to a complex and fragmented urban sociospatial structure (Wills, 2012). Successive governments have responded to these disparities by targeting disadvantaged neighborhoods for regeneration. This has often been carried out by contracting private developers and frequently resulted in gentrification and the displacement of working-class communities, in what some have called "social cleansing." Thus, market-led urban development, the co-existence of affluence and deprivation as well as high levels of ethnic segregation characterize contemporary London.
The social tensions emanating from the growing disparities within London provided the context for the 2011 riots, aggravated by everyday practices of violence by the police toward residents in black and ethnic minority neighborhoods. The shooting of a young black man by the police on August 4, 2011, in the deprived neighborhood of Tottenham in north London was the spark that ignited the riots. The violent unrest rapidly spread to other boroughs in London and to other cities, leading to substantial property damage and mass arrests (Lewis et al., 2011;Metropolitan Police, 2012). Riot participants were mainly young individuals from ethnic minorities and working-class neighborhoods (Lewis et al., 2011). These neighborhoods had suffered the effects of housing privatization and deregulation, of substantial cuts to education and youth services as well as precaritization in the urban labor market (e.g., Slater, 2016). Residents' and rioters' accounts refer to these longer-term processes of neoliberalization and more recent austerity measures underlying the riots. Our research in London focused on Tottenham, an ethnically diverse and deprived but gentrifying area where the riots began, and Croydon, an outer-city borough with profound spatial and social divisions along intersections of race and class.
Like London, Stockholm has also undergone neoliberalization, albeit later. Since the early 1990s, governance in Stockholm has changed from upholding a supposed "social democratic utopia" (P. Hall, 1999) toward a distinctive neoliberal agenda. Privatization of housing stock (see e.g., Stahre, 2014), together with a long-lasting socio-spatial segregation and skewed patterns of urban investment, have contributed to a growing social polarization and to socioeconomic inequality between different areas. Today's multicultural neighborhoods, built in the 1960s and 1970s as part of the 'million homes program,' have been characterized as disadvantaged or 'vulnerable' in diverse respects, such as income, health, life expectancy, levels of employment and education, school performance and drop-outs (Stockholms stad, 2015). Official neglect and disinvestment (for example, cuts in youth services and other public services), together with privatization programs, have negatively impacted these neighborhoods, and a high level of youth unemployment as well as school drop-outs have created the conditions for the growth of criminal gangs. As in London, regular police harassment bred resentment as well as distrust for state authorities among resident youth.
Shortly after a middle-aged resident in Husby, one of these disadvantaged multicultural neighborhoods, was shot dead by the police in May 2013, the already tense situation became explosive. The 2013 riots took place on an unprecedented scale: they lasted for five nights, involved large numbers of youth and spread rapidly to other parts of Stockholm and to other major cities in the country. Importantly, the affected areas were largely disadvantaged multicultural residential areas (Lindell et al., 2019), pointing to the importance of underlying economic and political conditions. Residents' narratives of the events referred to not enjoying the same rights as those of the majority society, and feelings of exclusion related to governmental neglect and disinvestment; to a democratic deficit in the ways in which these areas were governed; to racism in police practices; and to the unjust image that the media offers of these neighborhoods (De los Reyes et al., 2014).
Despite many similarities between London and Stockholm in terms of contextual factors, trajectories and triggers, strategies to govern affected neighborhoods after the riots have differed considerably between the two cities, as explained below. The empirical material that we draw from is based on semi-structured interviews and focus groups with local residents, activists, nonprofit organizations and government officials in each city. This is combined with documentary analysis of policies, planning documents and official reports during the period from shortly before the riots in each city to 2019. The remainder of the paper draws together relevant literature around the three key themes identified above, alongside illustrative empirical observations from this research.

Interrogating policy responses
As policy at all scales throughout much of the Western urban world in the last decades has developed as a retraction or "roll-back" (Peck & Tickell, 2002) of policy measures aiming at a general condition of welfare for urban inhabitants, one overarching policy response or "rollout" is the securitization, or "scanscape" (Davis, 1990) of cities. The extent to which this is done directly in response to riots is, however, unclear since most securitization emerges independently of unrest. Nevertheless, Dear (2000) connects the evolving urban landscape of social polarization directly to the riots in Los Angeles in 1992. The richer "citadel" population are protected from the criminal, rioting and looting poor not only by the roll-out of "bunker architecture," related processes of planning, gentrification and urban regeneration, and new types of institutions (e.g., general "work fare" policy measures) but also (welfarist) social policy which was never totally eliminated. However, what remains is too little, too "projectified" and too "fragmented" to matter preemptively. Neoliberalized urban social policy measures intervene-like the fire brigade-when things have gone too far, rather than on an idea that encompasses all places and citizens within the municipal jurisdiction.
Policy responses to riots seem to follow this logic, whereby new strategies are often deployed and directed toward neighborhoods that have been the epicenter of violent insurrections. Such strategies may introduce new actors into these areas, create new partnerships and collaborations, or generate new ways of governing these riot-prone neighborhoods, as was done in the war on poverty following the 1960s' race riots in the U.S., and in many places over the world in later eruptions of urban unrest. In trying to make hard physical regeneration interact with social, soft development it is hoped that places would become better for living. There is, however, as Shaw and Butler (2020, p. 97) point out, a "slippage" if urban renewal also means social and community renewal-in short, "is it the place or the population that is to be regenerated, and by what means?" Despite the importance of such interventions for understanding how these areas change after riots, they are nevertheless seldom considered in the riot literature, given its general present-centrism. There is also a limited understanding of how contextual factors shape such responses, and thus of the diversity of strategies employed, of underlying rationalities and of outcomes across places. Our research from London and Stockholm uncovers some of this diversity.
In the London case, redevelopment emerged as a major strategy for transforming the riotaffected neighborhoods of Tottenham and Croydon. Tottenham was a minor target for gentrification prior to the 2011 riots, but relatively low property prices were further deflated in the aftermath of the riots, which provided an opportunity for public and private developers to expand and accelerate redevelopment plans in the area. Regeneration, in the form of private residential developments, led to the large-scale demolition or financialization of social housing and displacement of long-term residents. In Croydon, regeneration efforts-though less extensive than in Tottenham given its location further from central London-have focused on the redevelopment of two shopping centers, oriented toward attracting middle class commuters. A public-private coalition, the Croydon Partnership, was created in 2013 to materialize these projects. These redevelopment projects represent partly an attempt to re-brand the area by remaking the retail landscape and their rationale makes explicit reference to the 2011 riots (Mayor of London, 2013). They have been criticized for their high cost, and for the level of economic and cultural exclusivity of the retail opportunities being developed.
In both neighborhoods, then, government agencies and corporate businesses used the riots as a justification to accelerate and intensify redevelopment processes. However, strategies and outcomes were also shaped by the different histories of these areas, their demographic composition, degrees of connectivity to central London and, not least, by different levels of grassroots political organization. While in Croydon such political organization is relatively weak, in Tottenham, residents strongly contested the redevelopment plans of the local council and private investors, and in one area proposed a subsequently successful community-led "counter-plan" for affordable housing, based on principles of participatory planning and self-management (St Ann's Redevelopment Trust, 2017).
In the Stockholm case, the dominant strategy came to focus on the securitization of the affected neighborhoods. The 2013 riots did however also generate reflection among political elites in the city. In the following years, Stockholm Municipality expressed goals of equality, cohesion and integration in key policy documents and created the Commission for a Socially Sustainable Stockholm to investigate and suggest changes to mitigate the marked spatial inequalities within the municipality (see Borén, 2019;Stockholms stad, 2015Stockholms stad, , 2017)-a work that would be discontinued after a rightward electoral shift in 2018. At the same time, since the 2013 riots, there has been an increase in various forms of violence in disadvantaged neighborhoods, related to the recruitment of youth into various criminal gangs and violent networks. Such violence and the eminent risk of new riots became conflated under a predominant narrative of insecurity, that represented these neighborhoods as dangerous "no-go zones." Security concerns linked to multicultural suburbs thus came to dominate national political debates, generating growing political pressure toward the adoption of "tougher measures" to curb crime and violent actions in the suburbs. These neighborhoods were ranked and categorized by the police (Polisen, 2017) into different levels of insecurity. Identified "vulnerable areas" were singled out and targeted for securitizing and strategic interventions, mainly in the form of increased police presence and the setting up of surveillance cameras.
But concerns with "safety" also become the focus of initiatives among a diverse set of actors. Major housing companies in these areas-for whom redevelopment is too risky an option, not least given the substantial opposition by residents against earlier regeneration attempts-expanded their 'safety'-enhancing activities, as a means of maintaining stability and property values in these neighborhoods. Local district authorities developed networks and collaborations with both housing companies and residents associations with a view to monitoring the risk of unrest. Yet, this strong focus on securitizing public spaces means that structural factors have received little attention. In fact, our studies in London and Stockholm suggest that the long-term systemic problems that were at the root of riots have remained un-addressed or worsened. Segregation in Stockholm is still increasing (Andersson, 2017) and un-employment figures remain high, particularly among certain migrant groups likely to inhabit the concerned neighborhoods. Furthermore, since the elections of 2018, labor and housing rights have experienced additional pressures.
In sum, different types of policy agendas came to dominate attempts to transform riot-stricken areas in London and Stockholm: redeveloping the built environment and securitizing public spaces, respectively. Despite a great amount of research focusing more generally on processes of urban redevelopment, gentrification and securitization of urban space, the ways in which these and other strategies are specifically deployed to reclaim, control or reform riot-prone neighborhoods are seldom considered. Research concerned with riots and their recurrence ought to develop an interest in the diverse strategies-material, discursive, institutional and relational-used to these ends and in their effects for the targeted neighborhoods and their residents. Studying variations across neighborhoods and cities will also improve our understanding of the role of contextual factors in shaping the nature of strategies adopted in different places.

Excavating urban memories and temporalities
In seeking to understand the legacies of anything, questions of temporality-of continuity and change over time-must take a central position. This involves attention to the pace, rhythm, and dynamics of change itself, including the varying forms of memory that are evoked (or forgotten) regarding how people and institutions understand places. In his theory of emergence, George Herbert Mead, perhaps the most prominent philosopher of temporality, noted how we act in the emergent event of the present through our relationship with the past and future (1959); in other words, making sense of our present conditions always takes place in relation to other temporal fields. Since continuity is often taken for granted in the practices of daily life, it is discontinuity that forms the foundation of our experience of time, and dramatic or sudden discontinuities (such as riots) especially so. However, as seen from within the concerned places, riots may merely signify "a magnification of the ordinary rather than an outburst of the extraordinary" (J. Till, 2013, p. 71), which further underlines the need of research containing a fine-tuned understanding of the differences between neighborhoods, and between the concerned neighborhoods and the city at large.
In the context of Tottenham, the 2011 riots created opportunities for collaboration between local government and private developers to accelerate gentrification in areas that had been earmarked for redevelopment long before the riots themselves. The subsequent struggles over the right to "stay put" within this traditionally left-wing working-class area were shaped largely through people's connections to place, drawing on memories of a long tradition of community resistance. The memory of previous riots in 1985 endured in the local community, and in 2011 long-time activists evoked the memory of the 1985 riot, which was also sparked by racist policing, disinvestment and deprivation. Remembering, in short, becomes part of resistance. As one community activist noted, Memories can live a very long time. Things don't just happen in a vacuum. History is not just a series of unconnected events. So, there is a long folk memory, whether it's class struggles, whether it's black struggles, whether it's particular campaigns . . . . It goes into the community.
Confrontations against "domicide," the destruction of the sense of home caused by displacement and social cleansing (Pull & Richard, 2019), are therefore influenced not only by the material reshaping of urban spaces but also people's senses of belonging across generations. While disruptive events have been conceptualized, following Badiou (2007), as "ruptures [to] the established order of things" (Cloke et al., 2017, p. 70), they can also lead to a reaffirmation of that established order-even if, in this case, that order is a relatively rebellious one. Remembering is therefore inherently temporal and political: as Crinson (2005) argues, cities are particularly intensive sites for the formation and contestation of memories, since they are such dense agglomerations of human life.
Croydon council's long-standing reputation for inertia and a powerful, conservative bureaucracy contrasted starkly with the decisiveness of a strong and stable Labor Party-led council to accelerate gentrification in Tottenham following the riots. This meant that a major commercial redevelopment in the center of Croydon that had been in discussion for several years before 2011, and which was broadly (if critically) supported by many locals when formally approved in 2017, was delayed and then significantly scaled back in early 2020. Thus, temporality also highlights the "speed" of change that is again often associated with local institutional and political cultures. Various avoidable lapses in institutional memory within Croydon Council further reinforced the popular perception of an unwieldy and disorganized institutional culture, leading to exasperation among community groups and its own employees alike. In one case, the council replicated environmental assessments that had already been conducted 8 years previously, which they had forgotten about. It left one local government community worker "wondering how many times this happens, that the wheel just gets reinvented" through institutional failures.
Thus, the built environment-be it formal public projects or more subtle affinities to certain placesacts as an anchor for representing particular moments or events in people's lives, individually or collectively, as well as shaping relations between the local state and its subjects. Rose-Redwood et al. (2008), however, remind us to be mindful of how unequal access to commemorative practices can influence whose experiences and narratives are remembered, and whose are not: "[w]hat memories are ultimately made visible (or invisible) on the landscape do not simply emerge out of thin air. Rather, they result directly from people's commemorative decisions and actions as embedded within and constrained by particular sociospatial conditions" (p. 161).
While notions of "collective memory" have been accused of essentializing much more diverselyexperienced pasts into overly-homogeneous narratives (Feindt et al., 2014;Massey, 1991), it is clear that major events or emergencies such as riots can have powerful collective resonance in the years and decades that follow them. In this regard, Marion Hirsch (e.g., 1997) has developed the notion of "postmemory" to explain how those who were not of the generation to experience a particular thing can nonetheless be affected by it. However, there are multiple forms of interpretation that take place between generations in this process of passing down memories, such that the memories of later generations are shaped and filtered by those passing them down. Thus, postmemory is rendered "a generational structure of transmission embedded in multiple forms of mediation" (Hirsch, 2019, p. 172). For the youths of Tottenham, who never experienced the 1985 riots, the memories shared by parents, older siblings, and community leaders contributed very explicitly to their upbringing. One youth worker described how the 1985 riots live on in the present as "mythic." This mythical nature of riot memories allows for political and cultural traditions to be passed on through older generations' narratives, while also being shaped and mediated through place in terms of associations with both points in the urban landscape and the repeated "scents, sounds and textures" associated with particular places and events (Tolia-Kelly, 2004, p. 314). This type of mythic, place-based memory may partly also explain how urban unrest can often re-occur in the same places over time (cf. Olzak et al., 1996).
The 2013 Stockholm riots, while attaining unprecedented international media attention, were far from the first manifestation of urban rebellion in the Swedish context. Some percursors include the 'hunger riots' of 1917 that spread to a large number of Swedish cities and towns; the 1948 Stockholm Easter Riots when a confrontation escalated between youth and police in a central working-class area; and waves of uprising in the 1950s and 1960s (see Brink Pinto & Ericsson, 2019; Peterson et al., 2017;Sernhede et al., 2016)-some of which had significant policy effects. Despite commonalities, contemporary riots (since the 2000s) differ from earlier rebellions in that they do not primarily occur in central city areas but in suburban multicultural neighborhoods and thus have a racial dimension. Significant insurgencies have occurred since 2008 in Malmö, Gothenburg and Uppsala and contributed to the rise of new grassroots youth organizations in the suburbs of Stockholm and Gothenburg that explicitly challenged urban segregation, exclusion and racism (Ålund, 2014;Peterson et al., 2017;Stahre, 2014). Importantly, these (and some other) organizations in the suburbs have drawn upon "the traditional ethos inherent in early twentieth-century Swedish popular movements" and have collaborated with "old" movements, and suburban activists have even made reference to the 1917 "hunger riots" (Sernhede et al., 2016, p. 168). Thus, the 2013 riots, while apparently new to Stockholm's northern suburbs, did not take place in a vacuum-they were preceded (possibly inspired) by earlier rebellions elsewhere, occurred in the context of a contestive culture of resistance, itself informed by active memories of earlier movements and rebellions. This memory politics might be locally very important for the positioning of actors and for their capacity to mobilize support for future action.
The memories of the 2013 riots surface as highly diverse and contradictory. When asking urban officials at the municipal level if the memory of the riots is alive in some way in their practices, they generally answered negatively, even if they as private persons well remembered the events. This remained the case even for interviewees at the Office of Land Development (Exploateringskontoret) who relayed stories of how stones where thrown at them when working in the vulnerable suburbs. On the other hand, the top political leadership seems not to have forgotten the riots. A leading Stockholm politician stated that the memory of the riots "is obviously very strong . . . it is a distinct event in the recent history of the city that absolutely is fresh in memory and surfaces in discussions." Indeed, a heightened awareness that large scale riots can happen again has prompted some responses from public and private actors, associations and NGOs, hoping to prevent such a volatile future. The disruptive riots also seem to have widened local consciousness in local communities, but local memories of the riots and concerns with their future recurrence have partly become overshadowed by the intensified gang violence in recent years.
Memories of the riots differ among groups as to what they signify, and, maybe more importantly, as to what they imply in terms of actions for local and urban politics. While local activists held and shared powerful memories through lived experience, local officials are the ones with most structural power to make that (hi)story heard outside of the community. However, without the officials inside the urban administrations knowing or understanding the history of the concerned places, eventual efforts can be lost in translation. The temporal dimension in tracing the legacies of riots is therefore far from linear or predictable, with complex processes of remembering and forgetting, alongside both continuities and changes. The ways that locally-embedded traditions and knowledge are transmitted between generations and localities are likewise far from clear, but there is certainly a connection between particular political or community traditions and the forms that social action takes in the present, both locally in the concerned neighborhoods and on the wider-scale levels of urban administration and policy. Thus, sensitivity to the temporal dynamics of meaning through which riots' legacies are made allows scholars to pay closer attention to the ways in which past and future are contested in urban planning and development.

Exploring feelings and subjectivities
"[O]ur sense of who and what we are is continually (re)shaped by how we feel" (Davidson & Milligan, 2004, p. 524). By using the term feelings, we are being deliberately ambiguous: individuals and communities feel emotions and affects as a result of traumatic events, but also "feel their way" through their aftermaths using a variety of tactics-improvising, adapting, and negotiating their post-event world through multiple practices of everyday life (De Certeau, 2011). As such, we use the term feelings to describe the diverse psychosocial effects of urban riots, which can be immediate but also outlive any material damage to the city, through shaping everyday patterns and habits of life. Much like memory, above, living in the (social and material) ruins of an event like a riot can have psychological impacts that affect individual and collective behaviors. The making and remaking of urban space is a processual one in which "people's knowledge of the environment undergoes continuous formation in the very course of their moving about in it" (Ingold, 2000, p. 230), forming what Anderson and Holden term "affective urbanism," by which they mean "an urbanism attentive to how various modalities of the more than/less than rational, including affects, emotions, and feelings, compose urban life" (Anderson & Holden, 2008, p. 144).
A small but growing literature in geography has begun exploring these psychosocial dimensions of urban life, using notions such as "chronic urban trauma" (Pain, 2019) and "wounded cities" (K. E. Till, 2012) to explain the ways in which both sudden ruptures and longer-term "slow violence" (Nixon, 2011) impact on people's understandings of themselves and their places. Equally, as interviewees from both case studies indicated, sometimes ruptures, however disruptive or traumatic, can help to expose injustices and inequalities that would otherwise have gone unnoticed by those in power. An economic development officer in Croydon stated that following the riots, "they [the council] were saying 'we love Croydon!'" but residents wanted to ask "but where were you before?" As such, where the trauma of slow urban processes such as managed decline or neglect has become so "hard-wired" (Pain, 2019, p. 394) into a place that it has become mundane and familiar, riots can have the power to render these processes visible to policymakers and communities.
This point, which also relates to urban memory, is illustrated well by a leading politician in Stockholm, who talks of the Husby riots as an "alarm clock." She says: It is always difficult to say when things started, but I feel very much that the riots in some way symbolize when, I don't know [but] from my perspective, the segregation or the inequalities had passed a limit. And other politicians would say . . . have other explanations. But in the discourse, the debate, absolutely, I still think that it is a tipping point in how one sees the city.
Urban trauma, then, is not necessarily a chronic condition. Nevertheless, it can be transmitted over time through spatial reference points and social networks to expand its impacts far beyond those who directly experienced it (Degloma, 2009). The long-term emotional afterlives of traumatic events take shape gradually and can change characteristics as they "move." Berlant (2011) argues that "trauma" may even be the wrong word to use, since it often (although not always) implies something exceptional and surprising; on the contrary, the slow violences of structural conditions that provide experiential frameworks for riots are usually far more ordinary. The economic context surrounding the London and Stockholm riots was a story of long-term, neoliberalization-driven economic hardship within working class and migrant communities, which, most clearly in London, had been exacerbated and accelerated by government-led austerity measures that stripped back welfare, social services and youth facilities further. As such, the collective feeling in such neighborhoods was already one of loss, struggle, and "getting by," long before and after the riots took place. The scarcity generated by austerity created particular affective atmospheres through which residents were experiencing the world and navigating their way through it (see S. M. Hall, 2019). Everyday practices of survival in the aftermath of the riots in London and Stockholm were therefore shaped by austerity's state-induced wounds on both the physical landscape and community infrastructure.
In Stockholm, for example, prior to the riots there was a long-term feeling of not being listened to in the concerned neighborhoods, not to mention overall feelings of racism and exclusion from mainstream society, which also generated a strong sense of place and a stout local community (De los Reyes et al., 2014). Nevertheless, interviews in Stockholm also show a sense of hopelessness and frustration among youth in vulnerable suburbs. As one social worker engaged in democracy development explains: So I try to back them up. And why I do it is because I hear, sometimes, the hopelessness connected to being raised in a socio-economically deprived area. Sometimes we have poetry evenings. Those youths that write poetry, this poetry they write, it is so much about, almost always about exclusion, injustices, and the everyday, the daily, the everyday that they run into.
As K. E. Till (2012) notes, using the metaphor of wounds implies a more complex temporality than a binary "before" and "after"; instead, wounds take time to heal and may become aggravated by infection or leave scars. Moreover, scarring is a process that brings with it a range of different emotions-not just sorrow, loss or pain but potentially also pride, courage and togetherness (Storm, 2014). Scarring can thus become part of the topology of a spatially varied affective urbanism (Anderson & Holden, 2008) which is intertwined with a number of other geographies and processes both within the concerned neighborhoods, but also outside of them. These places will for a long time to come be connected with the urban rage of the riots, giving them emotional meaning in the city (and the nation) at large.
It is in this affective context where inhabitants of neighborhoods seek to recover from, or at least cope with, the legacies and aftermaths of ruptures such as riots. For example, technical ambiguities concerning insurance and bookkeeping were overlaid on a wider economic slump to severely limit the ability of shopkeepers to recover from the riots in Croydon, despite well-meaning but misdirected support from the council. The ability of communities to self-organize to support one another with dignity is therefore dependent as much on the strength of community connections as it is on the responses of those in power; indeed, on a number of registers, policy responses in our case studies often appeared more harmful than helpful. As such, institutional efforts to generate "resilience" may not always be the panacea that decision-makers hope them to be (MacKinnon & Driscoll Derickson, 2013). In Tottenham, it was quickly apparent that government-engineered efforts at blaming the riots on a lack of community cohesion and resilience were inaccurate when 3,000 residents marched under a banner demanding "Give our kids a future" a few days after the riots ended.
The diverse everyday feelings that follow riots can inform our understandings of how people and places adapt and respond to such traumatic incidents. In doing so, scholars can be more attentive to the intersections of emotional and other forms of change and how these map onto wider politicaleconomic contexts. In the final section, we draw together the three identified dimensions of riots' legacies.

Concluding points
In Button's (1978) seminal text on the long-term legacies of riots, he concludes by asking "when does the squeaky wheel get the grease?" (p. 174); in other words, when do disadvantaged groups not only receive recognition of their disadvantage but also substantive changes stemming from new, revised, or halted policies? While this is an important question to ask, we also contend that the "grease"-policy change-is only one small part of a much broader set of stories concerning the long-term effects of riots on the cities in which they erupt. The emotional and affective registers discussed in the previous section link to the policy dimensions that we have discussed for the analysis of riots' legacies and afterlives. The feelings that endure long after riots bring together the impacts of amended, reversed, or expanded policy frameworks on the sense of place by redevelopment or securitization, alongside the complex temporal fields that thread through and beyond riots. In many ways, these are subtle processes that are not always explicit or clearly tangible: unarticulated and emotional factors loom large in both institutions' and individuals' action or inaction following sudden disruptive events. Indeed, feelings of fear and anxiety that rippled through publics in both Stockholm and London also highlight impacts on the wider political context. Such anxieties prompted identifiable though diverse political responses, from securitization in Stockholm to exclusionary discourses that strengthened rationales for redevelopment in London. In some regards, the riots also contributed to (often white) majority anxieties in both countries about migration and unruly internal "others" (Bhattacharyya, 2013;cf. Hansson & Jansson, 2021), which have fed into wider policy shifts away from multiculturalism across Europe. Together, the three themes of policies, temporalities, and feelings can help make sense of the different dynamics that riots can produce, yet they also stand alone as separate focal points for research in their own right.
This pulling-together of diverse dynamics in urban life, and reconstructing multiple threads of continuity and change, requires conversations across multiple disciplinary fields of analysis. There is a number of methodological challenges that come with such an endeavor, such as how to represent lived and "official" forms of remembering and forgetting in relation to one another, and identifying particular modes of comparison to understand the similarities and differences of riots' legacies-not only between different cities but also, as our empirical materials show, between different neighborhoods of the same city. What constitutes an appropriate temporal range is also raised when discussing legacies, which we have argued relies on a contextually-specific appreciation of the event, and what exact questions are to be asked (and of whom). For this project, the focus on lived experiences and policy change lends itself to a temporal field that numbers several years, whereas a different framing around, say, economic change, might be more appropriately measured in a decade or more. The relativity of time to sociospatial context is therefore essential when designing research into legacies of disruptive events like riots.
We began this article by arguing that the long-term legacies and afterlives of contemporary urban riots are decidedly under-researched. While this is certainly true, it is also clear that there is a range of conceptual and analytical tools that can help urban scholars to understand the significance of contemporary riots and similar eruptions. In undertaking this task, it may be possible both to understand the diverse impacts of riots on cities and to extrapolate this beyond the specificities of riots themselves, toward other disruptive events. Indeed, by studying riots, we can unlock deeper understandings of how dynamics like violence, contestation, and trauma-often understood as reactions to urban development-can themselves reshape urban lives, policies, and landscapes, in multiple and contrasting ways.

Note
1. There is a wealth of words used to describe riots, and a complex politics of naming. Terms such as insurrection, uprising, or unrest have been used to challenge negative connotations of riots (as mindless, apolitical destruction), but we use the term riot as a generally intelligible catch-all term that encompasses the broad range of forms and motivations for mass popular violence in public spaces.

Disclosure statement
No financial interest or benefit has arisen from this research.

Funding
This work was supported by FORMAS, under grant No. 2015-01315; and by EU Horizon 2020, Grant Agreement No. 727097.

About the authors
Anthony Ince is Senior Lecturer in Human Geography at Cardiff University. My research is informed principally by anarchist political philosophy and focuses on broad-ranging issues around agency and the politics of everyday life. In particular, I have written extensively about the state and territoriality, the far right, and living with difference in a globalized world.
Thomas Borén is Associate Professor in Human Geography, Stockholm University. The focus in my research is on urban and cultural geography in Europe with a particular interest in the interaction between urban development strategies, cultural producers and urban decision makers. For more information on current projects and publications, see https:// www.su.se/english/profiles/boren Ilda Lindell is Associate Professor, Department of Human Geography, Stockholm University. My work addresses the politics of urban informality in Africa, with a focus on livelihoods. A more recent interest is on large-scale transport infrastructure projects in urban Africa and how they restructure opportunities in the city. For information about projects and publications, please visit the website of the Urban Africa group at the above department: https://humangeo.su.se/ urbanafrica ORCID Anthony Ince http://orcid.org/0000-0001-5279-0997 Thomas Borén http://orcid.org/0000-0001-6787-2936 Ilda Lindell http://orcid.org/0000-0003-1986-1671