The 'Peaks and Troughs' of Societal Violence: Revisiting the Actions of Turkish and Kurdish Shopkeepers During the 2011 London Riots

Introduction

1.1 The summer of 2011 saw riots in a number of cities in England, which were sparked by the police shooting of young black man Mark Duggan on Thursday 4 August, in Tottenham north London^[1]. However, what started as a protest at his killing on 6 August turned into an expression of more general disaffection at economic inequalities, as well as vandalism, violence and opportunistic looting (Allen et al. 2013; Millington 2011). According to police and government reports, disorder started in the borough of Haringey where Duggan was shot, before spreading to other areas of London the next day (HMIC 2011; HOC 2011; Metropolitan Police 2012)^[2]. By 8 August, rioting also occurred elsewhere in England and continued until the following day. Five members of the public were killed, thousands of commercial premises were attacked, while arson and looting was widespread (HOC 2011: 3).

1.2 There has been a significant amount of analysis of the riots since they occurred, which has sought to explain the causes, the motivations of participants and the police response (see for example: Allen et al 2013; Baudains et al. 2013; Dillon and Fanning 2012; Moxon 2011; Solomos 2011). In recent years, there has also been an increased interest in developing micro-sociological theories of violence that move away from criminological perspectives towards a situational and micro-interactional approach, placing violent practices at the centre of the analysis (Bakonyi & Bliesemann de Guevara 2012; Collins 2008). However, the privileging of the violent moment or the peak of the scene distorts reality by dismissing contexts and connections (Doná 2013), and sidelining the un-dramatic, intermittent and spread out character of violence and 'eliding the empty spaces and drawn-out times' (Collins 2008: 414). In this article, rather than focusing on the centres of unrest and the major protagonists – the rioters, the police, the victims – we look at a small incident of contained violence which occurred during the riots in Dalston, East London, when a group of Turkish and Kurdish shopkeepers chased off potential rioters to contain and avert violence rather than ignite it^[3].

1.3 Dalston is located a few miles south of Tottenham, where Mark Duggan was murdered, and has a similarly diverse population and history of social deprivation. The area is a historic site of migration, characterised by Afro-Caribbean and African migration for much of the second half of the 20^th century, it later became home to Turks and Kurds from Turkey and Cyprus. The Turkish and Kurdish community of Hackney is estimated at 6% of the borough's population (London Borough of Hackney 2013: 7). While estimates of Turkish-speaking residents in Hackney, which includes Turkish Cypriots as well as second and third generation Turkish speakers, are between 20,000 and 30,000 (Erdemir & Vasta 2007). More recently, Dalston has undergone a degree of rapid gentrification, leading to increasing economic disparity. Dalston now includes people in both the 'least deprived' and 'top 3% most deprived nationally' in the 2010 Index of Multiple of Deprivation (London Borough of Hackney 2013: 17).

1.4 Based on a small scale-empirical study, this article aims to contribute to an understanding of the micro-sociology of violence by looking not only at the 'peaks' of violence, but also at the 'troughs' of inactivity or violence averted during the 2011 riots. It also explores the ways in which the 'peaks' and 'troughs' of violent unrest in Dalston were represented and, as a result, how they reverberated to impact upon the flattening of violence elsewhere during the riots. Just as the general sociological literature on riots contains a great deal of analysis of city level data but few micro level studies (Rosenfeld 1997), so research on the English riots lacks a micro-level analysis, of specific localities, of flashpoints and critical incidents, and moments of escalation and deflation. This article adopts an interactional and layered approach to the riots, to include those who operate on the spatial-temporal peripheries of violence, but nonetheless contribute to the overall drama and help to define its character.

Analysing the 2011 riots

2.1 Before looking at the Dalston case-study of violence averted, we want to place this article in the context of previous analyses of the 2011 riots and how the 'peaks' and 'trough' approach differs. The past half century of research on riots has focused too much on the 'push' factors that lead to violence and not enough on the responses of the state, the psychological factors that encourage people to riot, and the micro-logic of target choice within these larger events (Wilkinson 2009). Analyses of the riots that took place in August 2011 in England have similarly focused more on push factors and less on the micro-dynamics of the events. The riots were variously described as the result of structural inequalities, austerity and the cuts (Allen et al. 2013; Millington 2011), consumer culture (Bauman 2011; Casey 2013; Moxon 2011), police 'stop and search' policies (Dillon & Fanning 2012) and a deficit in community cohesion and participation (Dillon & Fanning 2012). The Riots Communities and Victims Panel set up by the Government, which was criticized for not considering structural issues (Bridges 2012), concluded by blaming the breakdown of families and lack of character in the young for their involvement in the riots (RCVP 2012).

2.2 Solomos notes that while the riots were not race riots per se, the initial spark to the disturbances was the police shooting of a young black man, which triggered debate about the historic problematic relationship between the police and the black community (Solomos 2011). Similarly the visible involvement of young Afro-Caribbean men in the riots in North London led to the racialisation of the riots by some commentators, while the involvement of white youths in other areas led to a deracialised reading of the unrest (Murji and Neal 2011). The rioters were indeed racialised and classed in popular discourses, as reports viewed the 2011 unrest as the uprising of an underclass, with a racial profile (Rogers 2013; Tyler 2013). These discourses are important to note, as in this article we consider the actions of members of a community that has been similarly racialised and classed.

2.3 A spatial analysis of the spreading of the riots across London (Baudains et al. 2013; Till 2012) showed concentration and contagion patterns that demonstrate peaks and troughs of violence. The riots appeared to have clustered spatially in specific areas and to have spread as a result of interactions between different actors, especially the rioters and the police. When violence escalated, an affected area continued to experience unrest and violence spread to other areas (we identify this as a peak). Alternatively, as a result of police action, rioters were displaced from one area to another and violence was re-located (Baudains et al. 2013), or social actors contributed to the avoidance of violence (what we describe as a trough).

2.4 Moxon's account of the temporal features of the riots referred to three constitutive moments: the initial, the acquisitive and the nihilistic (Moxon 2011). The shooting of Mark Duggan by a Metropolitan Police officer and the protest that followed triggered the initial moment. It was followed by the acquisitive moment when looting took over, and it ended in a nihilistic moment of general disorder. Linking critical events to the riots, Waddington (2012) uses the concept of 'flashpoints' to refer to the incidents that precede, instigate and lead to the escalation of violence. The flashpoint of the riots, he argues, was not the shooting per se (a peak) but rather the reaction of the police to the protests of the local community, in particular the fact that no senior officers were present to receive protest marchers and ameliorate their concerns (a trough).

2.5 The riots were significant in the history of violent disorder in England, with 1,984 people appearing in court for involvement in the riots and almost 4,000 arrested for 5,100 individual offences, mostly acquisitive crimes, according to official and media reports (HMIC 2011: 17; HOC 2011: 8; The Guardian/LSE 2011). As a result, sociological analyses of the riots focused on the most visible actors, the rioters (Reicher & Stott 2011) and the police (Baudains et al. 2013; Denef et al. 2013; Gorringe & Rosie 2011), and to a much lesser degree on other social actors like victims, 'vigilantes' and bystanders. Yet in the macro-narrative of the English riots of 2011, most members of society were bystanders and the majority of citizens did not riot, even though many were involved as spectators, protestors or commentators. Some joined political protests at the shooting of Duggan, but left when rioting began. Others watched the events unfold on the streets as they protected their property or tried to intervene. The vast majority, however, stayed at home and watched the riots on TV and online, blogging, tweeting and texting.

2.6 The interest shown in the use of social media during the riots is another example of the focus on the peaks of exceptional behaviour, rather than the troughs of violence averted. Rioters' use of Twitter to communicate was seen as a new way of organising which added another dimension to the violence (House of Commons 2011; HMIC 2011; Tonkin et al., 2012). The Police's response to the riots was also influenced by their use of new social media, with Twitter being used as instrumental (to give orders) or expressive as a communication tool (to involve the public) (Denef et al. 2013). However, the peripheral actors who commented on the action as the riots were taking place, or organized clean-up events and collective responses afterwards have been less explored as they represent the troughs of activity on social media rather than the exceptional peaks of violence in process. The role played by social media in the 2011 riots is interesting, not only because of its potential to assist in the organization of protest or violence, but also because it allowed real-time commentary on the riots, including events in Dalston.

The micro-sociological analysis of violence

3.1 Our analysis falls within a wider micro-sociological analysis of violence which attends to the contexts and pathways that do or do not lead to violence. We agree with Collins that rather than being the normal outcome, the enactment of violence is unusual, difficult and often incompetent, to the extent that most situations of conflict do not lead to violence but rather to mild forms of verbal aggression, retreat and face saving rituals (Collins 2008). Whether violence is expressed or contained 'depends on a series of conditions or turning-points that shape the tension and fear in particular directions, reorganizing the emotions as an interactional process involving everyone present: the antagonists, audience, and even ostensibly disengaged bystanders' (Collins 2008: 8). While Collins's analysis is persuasive, according to Dolan & Connolly his work does not explain the temporal and spatial context surrounding micro-interactional manifestations of violence (Dolan & Connolly 2014) and does not take into account the reverberations that occur during social unrest. As a result, an analysis of violence that does not incorporate surrounding events, actors and processes is subject to limitations.

3.2 Firstly, by looking at violent actions themselves, violence is explained simply as an escalation of tension, reflecting a bias towards 'extraordinariness'. This privileges visible, extreme and dramatic accounts of the peaks of violence, which occur rarely, at the expense of the much more common troughs or pathways leading to the reduction, containment or avoidance of violence. Flashpoints, or the incidents that precede, instigate and lead to violence (Waddington 2012), link peaks of violent behaviour to other significant contextualizing events that may also include the troughs of non-violence. Yet flashpoints are used to explain the rise in tension leading to a peak when violence is accomplished, but not the pathways that reduce, avert or contain it. If violence can be understood as a set of pathways around confrontational tension and fear, then both peaks and troughs can be seen to shape these pathways.

3.3 Secondly, various social agents occupy the spatial/temporal context of violence, yet there is a 'criminological bias' that explains violence through the actions of the aggressors and the notion of violent propensities and predispositions (Cooney 2009). While the contributions of others present in the space of violence is sometimes acknowledged, their actions are rarely analysed in depth and their perspectives on accomplished or dampened violence are rarely sought. Darley and Latané's (1970) social psychological micro-level analysis of bystanders' responses to emergencies, including episodes of violence, offers a classic example of scholarship on the behaviors of those who are neither victims nor perpetrators of violence, but whose actions can affect the course of events as they engage in decision-making processes about whether to act or not. What they do not do is go further to explain how individuals transition between roles during episodes of violence or potential violence.

Methodology

4.1 This paper is based on a small-scale empirical study conducted by the two researchers in Dalston, East London, in late 2012/early 2013. We used blended methodologies, such as semi-structured interviews and a thematic analysis of social media. We interviewed six shopkeepers/shop workers in the area, who were selected using purposeful and snowball sampling. All participants were guaranteed anonymity and pseudonyms have been used. We asked interviewees to relate their account of the events of the 8 August, eliciting information about the 'facts' of the incident, as well as their motivations for acting (or not acting), their views on how they had been represented and the ongoing impact (if any) of their intervention. We also conducted informal conversations with other shopkeepers in the area about what they remembered about the night, to provide supportive data. In addition, we conducted an interview with a video journalist for The Guardian who witnessed and filmed the events in Dalston on 8 August^[4].

4.2 We also conducted a search of Twitter using the keywords 'riot', 'Dalston', 'Turkish', 'Kurdish' and 'shopkeeper' within the timeframe of 8-10 August 2011. The search generated approximately 400 tweets, which were coded and subjected to a thematic analysis. (Some of these were retweets and others tweeted links to online videos or articles.) In addition, blogs and online videos which mentioned the action of the shopkeepers in Dalston were examined, as well as traditional media reports of the incident.

4.3 A number of challenges were faced during the research. Due to the sensitive nature of the subject matter and the time that had elapsed since the riots, it was difficult to find willing participants for the study. In some instances, personnel had changed in shops that had featured in news reports and events were not remembered as clearly as they would have been immediately after the riots. There was also a reluctance to discuss the events and a desire to play down their significance in comparison to other events during the riots. For example, restaurant owner Salih told us: 'In this area it was not a problem. The problem was Tottenham High Road and also in Hackney on Mare Street, or maybe Haringey.'

Findings

The riots in Dalston – a brief moment in time

5.1 Dalston is demographically similar and geographically close to Tottenham, where protests at Mark Duggan's death sparked the riots. Yet no large scale looting or disorder occurred there and only minor damage to shops and other property was reported. Indeed, Dalston is not mentioned in official police or government reports on the riots (HMIC 2011; HOC 2011; Metropolitan Police 2012). A number of incidents did occur in Dalston on 8 August 2011, which include a bus being set on fire on Shacklewell Lane (The Guardian 8/8/2011), objects being thrown at the fast-food outlet Nandos on Crossway, and minor looting at Kingsland shopping centre (Evening Standard 8/8/2011). Yet these episodes of violence represent minor peaks amidst the troughs that better represent the days before and after 8 August, as Dalston remained at the periphery of the spatial contagion of violence that spread further east and south (The Guardian/LSE 2011).

5.2 However, there is an inconsistency between official narratives of the riots (such as the police reports) and unofficial narratives (such as Twitter activity). The minor violent episodes were absent from official reports of the London riots because they were not seen to be of sufficient strength, intensity and frequency for Dalston to be considered a place of rioting, yet they nevertheless represent peaks of localized violence. However, the gathering of Turkish shopkeepers to avert violence was seen as significant and noteworthy elsewhere, attracting the attention of social, national and international media, out of proportion to the scale of the event (see for example, AlJazeera 10/8/2011; Daily Mail 9/8/2011; Hurriyet Daily News 10/8/2011; The Telegraph 10/08/2011a). A popular Tweet on the night said: 'Reports of heroic scenes on #Dalston High Street. Turkish families lining the streets to oppose #riots. What great Londoners.' In spite of the apparent disparity, both the official (government and police) reports and the media representation of the events in Dalston implicitly construct of violence through its peaks. While official reports focused on incidents assessed as sufficiently serious to qualify as peaks of violence, the media repackaged examples of violence averted (such as the action of the Dalston shopkeepers) as peaks, thereby eliminating the troughs from the wider narrative of social unrest.

5.3 Contrary to newspaper reports and social media, our research showed that the activity in Dalston on 8 August 2011 consisted mostly of waiting rather than action, as shopkeepers gathered outside their shops or stood on the intersection between Kingsland High Street and Crossway. Shopkeepers were alerted to the potential for unrest in Dalston by texts from friends and relatives, through social media and by observing events of the previous day. As shopworker Ersoy recalled: 'They were communicating on the phones and maybe social media as well. Ok they gonna arrive in Dalston in two hours, or they going this way or they going that way. So they actually organize, gather too.' When young people, perceived to be rioters, tried to advance down the main thoroughfare Kingsland Road, they were quickly chased away by the shopkeepers. Media reports referred to 'hundreds' of 'Turkish' men being involved (Daily Mail 9/8/2011; Hackney Gazette 10/8/2011). The importance placed on the events by the shopkeepers themselves, however, suggests that they were small scale and relatively brief. Guardian video journalist John Dokomos, who filmed the moment when the shopkeepers chased away potential rioters (Dokomos 9/8/2011), confirmed that the action itself was: 'a relative pinprick of time […] from when I got there to when the whole thing was done and dusted, that kind of charge down the street […] was about 5 or 6 minutes'.

5.4 The brevity of this escalation shows that the containment of violence in Dalston is best seen as waiting, of troughs in anticipation of a violent peak. Mehmet describes sleeping in his shop overnight in fear of violence: 'We stayed here for 24 hours, we stayed from the morning, but they didn't do anything'. Yet, the waiting appears to have been saturated with emotions and thoughts of possible violence, what Dokomos described as: 'A pregnant pause […] like a very heavy raincloud when you know something's going to come.' The peak in terms of the activity in Dalston occurred when the waiting culminated in a chase, which was fuelled by adrenalin and fear. However, in terms of the riots as a whole, the episode is better understood as a trough, an example of violence averted and contained rather than violence escalated and achieved.

Reverberations and the flattening of violence

5.5 Three types of reverberations were in evidence during the events in Dalston, which demonstrate the importance of taking into account the temporal, spatial and social context of social unrest (Dolan & Connolly 2014). Firstly, anticipatory reverberations can be seen in the way that the shopkeepers demonstrated their awareness of events elsewhere in London and, therefore, anticipated rioting in Dalston. Secondly, experiential reverberations were revealed as the shopkeepers drew on their own experience of political and social unrest in Turkey to assess the situation in Dalston and make decisions about how to respond. Finally, representational reverberations occurred when the reporting of the action of the shopkeepers raised public awareness of the potential for community resistance and may have contributed to the containment of violence elsewhere.

5.6 Attempts to see Dalston as a discrete space, both in physical and social terms underestimates the networks that exist ordinarily and were heightened during the riots. Those interviewed explained how they moved physically between different locations during the unrest, as well as moving between different subject positions. Hasan, who has a taxi firm in Dalston, explained how he went first to Tottenham to protest against Duggan's shooting, 'then when it started in Enfield went to demonstrate there. And then […] when we saw all this looting […] I didn't take a part in that'. Therefore, anticipatory reverberations were evident as events elsewhere influenced the decision of the shopkeepers to defend their shops. Hasan said, 'We knew the youngsters are gathering and they will come to Dalston.' Similarly, Mehmet got a call from a friend in Hackney to say 'They're going come to Dalston. This is what happened to our area, so it might happen in your area.' Likewise, friends and family from outside Dalston were phoned and asked to come and defend family businesses. Analysis of the riots has made reference to the way in which mobile phones and social media were used to organize rioting behaviour (Baker 2012; The Guardian/LSE 2011; Tonkin et al. 2012). However, the use of these technologies to resist or deflect violence has been largely ignored, yet was apparent in our research.

5.7 The social and political standpoint of the shopkeepers in Dalston also produced experiential reverberations, as they assessed how to respond. Some made an explicit transnational comparison between the London riots and their own experience of political protest, policing and violence in Turkey. Ersoy laughed as he made reference to demonstrations of 10,000 or 20,000 people in Turkey: 'We have lots of huge stuff, I'm taking about back home, so the riots [in London] aren't really big.' These reverberations demonstrate that the shopkeepers' decision to confront the rioters was not made in a vacuum, but was rather informed by concurrent events including peaks of violence in other locations, as well as prior experience of the political and social context locally, nationally and internationally. The action of shopkeepers in Dalston was informed by their own political beliefs (some of them joined in protests against the shooting of Duggan), their experience of social unrest in Turkey (which may have informed their decision to migrate to Britain), their knowledge of recent unrest in other parts of London (which they had either witnessed or been informed of by family and friends), as well as being motivated by their desire to protect their businesses.

5.8 A third type of reverberation occurred when the representation of the action of the shopkeepers in the media and in popular discourse, appeared to influence the behaviour of other communities during the unrest. For example, according to news reports, people gathered in Eltham south London on 9 August to show their resistance to the riots. Although, unlike the action in Dalston, this gathering appeared to have racist overtones in an area with a history of far right activism (The Telegraph 10/8/2011b). Guardian journalist Dokomos also filmed this event and expressed the belief that there was an explicit decision in Eltham to mimic the actions of the Turkish shopkeepers and others (Dokomos 10/08/2011). More poignantly, a day later in Birmingham, more than 300 Sikhs and Muslims held a vigil at the petrol station where three Asian men had been killed during the riots (BBC 11/8/2011). In spite of some who were present calling for a march to the homes of the attackers and others suggesting a protest march to the city centre, community leaders were able to persuade those present to disperse and go home. The potential for a flashpoint to be ignited and a peak of violence to occur was great, as feelings ran high and those present were at a tipping point. This is evidence of the ways in which groups and individuals negotiate the pathways around violence and of how a potential peak of violence can be flattened into a trough of diffused action. Those in Birmingham were informed by their prior knowledge of the tragic outcome of violence in the city, as well as what they had seen of the riots in the media and on the streets in previous days. However, the success of the Dalston shopkeepers in diffusing violence would also have been known to them, due to widespread reporting.

Blurred boundaries and fluid roles

5.9 The analysis of violence and social unrest is often based on the assumption that individuals play fixed roles. For example, the large scale study into the riots by The Guardian/LSE (2011) separated out interviewees into the categories of 'rioters', 'victims' and 'police'. The reporting of the riots also made an often crude distinction between the perpetrators (the rioters), the victims (those who were attacked or whose property was damaged or stolen), the enforcers (the police), the bystanders (who watched but did nothing) and the heroes (those who intervened to prevent violence or damage). For example, Pauline Pearce, a black woman who confronted rioters in east London and questioned their motivation, was widely referred to as 'the Hackney Heroine' (see for example, The Mirror 14/8/2011). While rioters were typically described as 'huge gangs of masked youths' or 'mobs' who were able to 'swarm' and 'rampage' (Daily Mail 11/8/2011). However, in moments of chaos and unrest boundaries are more likely to be blurred, as individuals move between subject positions. For example, some of the shopkeepers initially joined protests against the shooting of Duggan and what they saw as 'police injustice', but left the demonstration as looting began and, as Ersoy explained, 'the colour of the riot changed'. When they heard that the riots might move to Hackney, they feared that instead of being political demonstrators or passive observers of the riots, they could instead become victims. This informed their decision to guard their shops and chase rioters away as they became active citizens or, according to some, 'vigilantes'.

5.10 The action taken by the shopkeepers was based on their decisions about what constituted moral or wise behaviour, as they decided whether to let violence erupt and observe it or whether to intervene to contain it. Some were reluctant to attack the rioters because they saw their argument as being with the police, while others were worried that direct intervention with rioters would be seen as unlawful. Kebab shop owner Mehmet saw a young man throwing a 'missile' at Nandos but chose not to act, other than to pick up the thrown object. He explained his dilemma: 'What could we do? We couldn't hold him. If we had, we could have found ourselves in trouble […] The police didn't do anything.' However, during an earlier incident when lives were in danger Mehmet intervened to extinguish a fire that was started by rioters on a bus and help passengers to safety. When asked why he acted he said: 'It is human instinct to want to help… I am against wickedness. What I noticed was the government was very slow to react to the situation.' His judgment about when it is moral and wise to intervene depends upon the perceived danger to others and the consequences for himself, while the non-intervention of the police in both cases is seen as contrary to the expected norms of behaviour. It is also important to note that not all of the shopkeepers saw themselves and the rioters as polar opposites. For example, Hasan spoke of warning his taxi drivers not to attack the rioters saying, 'this is the war between them and the police forces so don't do anything', making a distinction between those protesting against historic racial discrimination and those interested in acquisitive looting.

5.11 Another example of blurred boundaries can be seen in the way that the behaviour of the police during the riots was perceived (Gorringe and Rosie 2011). In situations of social unrest, the police are expected to play either a peacekeeping and containment role as they seek to dampen potential flashpoints and avert peaks of violence, or a more proactive restraint, arrest or even aggressive policing role when peaks of violence occur. As stated in the Metropolitan Police report on the riots: 'When confronted with a scene of serious disorder, public order officers are faced with four basic options; to isolate, contain, arrest or disperse the crowd.' (Metropolitan Police Service 2012: 118). However, shopkeepers perceived an absence or passivity of the police during the unrest. Hasan reported seeing police cars parked in front of Stoke Newington and Tottenham police station 'just protecting themselves'. Meanwhile, during the bus fire, Mehmet felt that, 'The police didn't come until much later; they only came after we had put the fire out. In my view the police were very slow to respond.' While Hasan felt that the police 'used Kurdish people as a fire extinguisher'. The police also appeared to be absent during the advance of rioters into Dalston and the subsequent chase by the shopkeepers down Kingsland High Road. Video evidence of the incident shows police vans driving down the high road at speed but not stopping (Dokomos 9.8.2011).

5.12 On Twitter, the police were also represented as passive, in contrast to the active role of the shopkeepers. The 'Turkish community' were presented as a de facto police force, who were 'doing a better job than the police getting rid of rioters' and 'doing what riot police can't, protecting their property'. Even though the intervention of the shopkeepers was relatively brief and small scale, on social media the shopkeepers were seen to stand in for the police, while the police were bystanders. For example, tweets stated: 'Seems the Turks in Dalston have done the job we pay our taxes for the police and army to do'; 'Who needs riot police when you have Turkish shop owners?' and 'Yup, you absolutely do not mess with the Turks in Dalston. They are legendary. They often come to the rescue when the police don't.' This tells us something about the way in which the police were perceived during the riots: on the one hand as brutal (they killed Mark Duggan) and, on the other hand, as ineffectual (a few shopkeepers could do a better job).

5.13 Another example of how roles are fluid at time of violence can be seen in the way in which those who watched, tweeted or blogged on the riots from home, later took action once the unrest finished. In some cases, this took the form of a clean-up, as residents came together to clean the streets after the riots, some using the Twitter hashtag #riotwombles or calling themselves the 'broom army'. Some attention has been paid to the troughs of activity in the aftermath of the riots, as the 'mediated crowd' was mobilized through social networking in both virtual and geographic arenas (Baker 2012), with hashtags being used to respond and organise cleaning and prayers (Glasgow & Fink 2013). Elsewhere money was raised for shopkeepers whose premises had been damaged (Tonkin et al. 2012). The transition of these 'virtual bystanders' from observers to active participants in the clean-up phase demonstrates how social unrest can have a life beyond the peak of violence, impacting upon behaviour in the troughs that follow. It also shows how individuals are able to transition between roles during social unrest, so that 'victim', 'bystander' 'peacekeeper' or 'rioter' are not closed categories but are rather open to negotiation.

Re-imagining resistance: The glamour of peaks versus the banality of troughs

5.14 The action of the shopkeepers was almost universally portrayed positively, in the media (including social media), as well as by politicians and social commentators. However, it was the brief moment or peak of action in Dalston, rather than the prolonged troughs of waiting and watching, which were commented on. Many praised the perceived heroism of the shopkeepers, who were seen as the protectors of Dalston, of their community, their families and their property. Ellie Slee, a blogger for the Huffington Post and Dalston resident, wrote of the 'communal bravery' and 'courage' of the shopkeepers who acted when others did not and protected 'us', stating that the shopkeepers: 'came out in force to protect their patch, chasing away rioters with a force that the police simply did not have. In turn, they protected us – other residents of Dalston' (Slee 2011).

5.15 Although the representation of the shopkeepers was largely positive, it included elements of racial stereotyping. The 'Turkish men' were represented as father figures who could bring about safety and order in the lawless streets, suggesting an ethnicised über masculinity. One of the most commonly retweeted statements about the shopkeepers said: 'Things calmed down in Dalston – largely due to heroic mobs of Turkish men standing guard down high street'. The juxtaposition of the contradictory words 'heroic' and 'mob' suggest both bravery and a kind of lawless aggression, while 'standing guard' places the shopkeepers in a quasi-militaristic role. Although no physical violence occurred, the privileging of peaks of action on social media led to the shopkeepers being represented through the use of metaphors of war and violence. Many tweets described the shopkeepers as a victorious and righteous force, as the 'Turkish army' or 'militia' who were 'fighting the rioters', 'patrolling' and 'kicking the living shit out of hoodies'. This analogy was taken a step further by those who tweeted about shopkeepers being armed with 'baseball bats', 'sticks', 'doner kebab knives', 'machetes' and 'metal poles'. The general response to this perceived level of violence was one of approval that 'someone's doing something', as tweeters proclaimed 'that's how you fucking do it!'. This behaviour was seen as essentially 'Turkish' and 'male' and seemed to generate a palpable excitement among tweeters. However, in reality, eyewitness statements and video evidence make little reference to weapons being carried by shopkeepers or to violence being perpetrated against the rioters (Dokomos 9/8/2011). So, although these tweets are supportive of the shopkeepers, they also play into racial stereotypes of Turkish and Kurdish men as aggressive, macho, violent and lawless. The admiration of this perceived behaviour is akin to the admiration shown for fictional vigilantes or maverick police officers. The distance between the tweeters and the action which taking place in Dalston only emphasizes the filmic quality of the reportage on Twitter, as the glamour of perceived peaks took over from reality.

5.16 The resistance of the shopkeepers to the rioters and their containment of violence were seen to symbolise the actions of 'good' citizens, in contrast to the rioters who were portrayed as 'bad' citizens. This polarization was suggested by a popular tweet during the riots which said 'it's balaclava vs baklava^[5] in Dalston'. The shopkeepers were represented as heroes, exemplary citizens, the epitome of community and a symbol of hope at a troubled time – at the same time as being seen by some as 'vigilantes' (Murji and Neal 2011). Rioters on the other hand were seen as lawless, nihilistic, selfish anti-citizens (Tyler 2013). As Rao notes, there is a tendency 'for invidious and insidious distinctions between good "model minority" immigrants and the bad undeserving poor (immigrant or not)' (2011) – which can be seen in the way the Dalston shopkeepers were represented. Even the description of the shopkeepers as 'Turkish' or 'Kurdish' may play to this stereotype, as many of those involved were British citizens of Turkish or Kurdish origin. Therefore, the decision to stress their ethnicity in the media and other political and popular discourses privileges a particular aspect of their identity. In addition, the representation of the Dalston shopkeepers cannot be separated from the way in which the riots were radicalised and classed in general (Murji and Neal 2011; Solomos 2011).

5.17 The fascination with the peaks of action in Dalston, which in reality were brief moments, tells us about the way in which violence is perceived. This was made explicit by Dokomos whose short video about Dalston went viral, while he was seen as an expert witness and was invited to comment on the incident on TV and radio. He explained: 'You know how these things work, they're like, "OK the story of today is shopkeepers fighting back." … They were saying it was like vigilantes in action. They wanted stories of kebab knives… They wanted drama and I was almost trying to play it down.' His observation demonstrates the ways in which society understands violence through its peaks and flashpoints, even to the extent that minor incidents which flatten violence and contribute to the troughs of inaction surrounding violent episodes are repackaged as peaks. The brief moment when the shopkeepers chased rioters down the street was used to represent much more in the days after the riots than it had meant to the shopkeepers at the time.

Conclusion

6.1 This article demonstrates how a peaks and troughs analysis can help to theorise the micro-sociology of violence. We agree with Collins that the achievement of violence is often the exception rather than the rule and, therefore, propose that the troughs, when violence is contained or averted, need to be included in any analysis, alongside the peaks or flashpoints of violent action (Collins 2008: 414). In addition, in line with Dolan & Connolly's critique of Collins, we recognize the importance of context in understanding the reasons why violence does or does not occur (Dolan & Connolly 2014). Decisions to exhibit violent behaviour, to act to dissipate violence, or to stand by, are made in a wider social context that is informed by events in other times and spaces, where the peaks and troughs of violence led to an individual and collective understanding of violence and its consequences. While Waddington (2012) refers to the concept of 'flashpoints' to refer to the incidents that precede, instigate and lead to the escalation of violence, this paper explores the more nuanced 'pathways' that may lead to violence or its containment.

6.2 We identify three types of reverberations in this study – anticipatory, experiential and representational – to show how the socio-cultural, political, temporal and spatial contexts influence decisions which lead to an escalation or flattening of violence. By focusing on the micro level case study of Dalston during the 2011 riots, we are able to explore an instance of violence averted and place it within the larger narrative of the riots. Using a model of peaks and troughs that sees violence not just in the isolated incidents or flashpoints, but as a continuum along which the peaks of violence are informed by the troughs of non-violence – and vice-versa, we propose a transitional understanding of violence.

6.3 This paper also recognizes the attraction and the glamour of peaks of violence, identifying the ways in which troughs are ignored (for example, in the police and government reports of the riots), or are repackaged as peaks (for example, in the representation of the events in Dalston on social media) (HMIC 2011; HOC 2011; Metropolitan Police 2012). The fascination with peaks distorts the reality of events that are better described by their troughs. In addition, by demonstrating that the roles played by social actors in situations of violent unrest are not necessarily fixed but are rather fluid and subject to change, we add another layer to the reading of the riots. A transitional understanding of violence recognizes that individuals and groups may transition between the roles of victim, perpetrator, bystander and actor within a wider spatial-temporal social context. Just as there are peaks and troughs of violence, there may be peaks and troughs in the behaviour of an individual in the context of social violence, along a continuum from action to inaction, or violence and the prevention of violence.

Notes

¹ Mark Duggan was a 29-year-old black man shot dead by police in Tottenham, north London, on 4 August 2011 (HOC 2011: 5). The police believed him to be armed at the time of the shooting, while his family maintain that he was unarmed. An inquest declared his killing 'lawful' in January 2014.

² For an interactive map of incidents during the riots see: http://www.theguardian.com/news/datablog/interactive/2011/aug/09/uk-riots-incident-map.

³ Media reports, social media and political commentary during the riots often referred to shopkeepers in Dalston as Turkish. Many are actually Kurds from Turkey who would identify themselves as Kurdish rather than Turkish. Some may also be Turkish Cypriots from Cyprus. This paper refers to the shopkeepers collectively as Turkish and Kurdish.

⁴ John Dokomos: http://www.theguardian.com/profile/johndomokos.

⁵ Baklava is a sweet Turkish pastry.

References