King Mob: Perceptions, Prescriptions and Presumptions About the Policing of England's Riots

by Hugo Gorringe and Michael Rosie
University of Edinburgh

Sociological Research Online, 16 (4) 17

Received: 28 Sep 2011     Accepted: 21 Nov 2011    Published: 30 Nov 2011


As journalists and academics, politicians and other commentators struggled to make sense of the social unrest across England, they reached for theoretical understandings of the crowd that have long since been discredited. The powerful imagery of the madding crowd has always been a popular trope with journalists, but what concerned us was the way in which even sociological commentators echoed such ideas. This paper, therefore, draws on our past research, informal interviews with senior police officers and media accounts to offer an analysis of the riots, how they were policed, and contemporary understandings of crowd behaviour. In so doing we question whether current understandings of collective behaviour, deriving from socio-political expressions of anger or protest, are equipped to make sense of the English riots. Similarly, we ask whether police public order tactics need to change. We conclude that the residual attachment to myths of the madding crowd continues to hamper the search for flexible, graded and legitimate means of managing social unrest.

Keywords: Riots; Crowd Theory; Public Order Policing; Negotiated Management

Introduction: A Madding Crowd of Commentators

1.1 If anything was more ‘spontaneous, contagious and uncontrollable’ than the social unrest in many of the larger cities in England in August 2011, it was the rash of responses, explanations and solutions it prompted as journalists, academics, politicians and other commentators struggled to make sense of what was happening. There were useful insights in the rolling judgements on and explanations of the disorder, and the best ways to deal with the riots. More common, however, was ill-considered or half-baked theory. Accounts related more to the needs and demands of the 24-7 news cycle than to empirical data (Burns 2011). This also suited some political agendas, with London’s Conservative mayor, Boris Johnson, dismissing ‘economic and sociological justifications’ for the disorder. In response Brewer and Wollman (2011), on behalf of the British Sociological Association (BSA), called for perspective, noted that numerous sociologists have worked in relevant fields of study and emphasised the urgency of ‘real understanding’ rather than ‘easy condemnations of these disturbing events’.

1.2 So far, so good. However, they continued: ‘One loses sight of the fact that nine out of 10 local residents aren't rioting, that nine out of 10 who are rioting aren't local to the area, and that nine out of 10 of these non-locals aren't doing it to commit crime’. If the evidence that underpins this particular ‘real understanding’ was obscure, there was even less basis for the subsequent assertion that ‘crowds are irrational. Crowds don't have motives – that's far too calculating and rational. Crowd behaviour is dynamic in unpredictable ways, and reason and motive disappear when crowds move unpredictably’ (Brewer and Wollman 2011). We have worked on crowd dynamics (especially in a protest context) and we concede that the first statement may well contain kernels of truth, but we are deeply troubled by this emphasis on the ‘irrationality of the crowd’, the madness of the mob.

1.3 There is a long tradition of crowd research being harnessed by the state to justify forceful policing. ‘Violent police tactics were developed’, as Schweingruber (2000: 383) recounts, ‘not just in response to actual demonstrator’s behaviour but also to the presumed behaviours and characteristics of protestors as defined by mob sociology’. Casting crowds as ‘irrational’ as Reicher et al. (2004: 558) note, ‘suggests ways of dealing with crowds that will fail to resolve conflict and can often increase the level of violence’. Contrary to the myth of the madding crowd, most socio-historical accounts of riots and social disturbances suggest that the perpetrators are often well integrated into local settings and have specific concerns and grievances (Rudé 1995; Thompson 1991; Marx 1970). Whilst rioting may appear as irrational to outside observers, riotous actions have meaning for the actors, and only by trying to understand those meanings can we begin to address them (Marx 1970; Stott & Reicher 1998).

1.4 Whilst the 1981 riots and the subsequent Scarman Report obviously provide some context within which the recent unrest needs to be situated, public order policing in the UK has faced particular and intense scrutiny in the wake of the protests surrounding the 2009 meeting of G20 (the ‘Group of 20’ finance ministers and bank governors from leading economies) in London. The death of a passer-by caught up in the policing operation resulted in a major rethink of public-order policing (HMCIC 2009a and b) and an updated Manual of Guidance on Keeping the Peace (ACPO 2010). This new version of police guidelines insists that: ‘planning for public order and public safety events should never start from the premise that crowds are inherently irrational or dangerous’ (ACPO 2010: 87). It is not often that we find ourselves more in agreement with Association of Chief Police Officers than the BSA. A widespread media view was that post-G20 reforms have left police emasculated and unable to deal robustly with mass disorder. We draw on relevant secondary literature, informal interviews with public order commanders, and our own past research to engage with two main questions: Firstly, have dominant understandings of collective violence been overly inflected by work on political riots? Secondly, why was the initial police response seen as ‘timid’ and, what could (and perhaps should) the police have done differently?

Understanding Crowd Dynamics

2.1 Until the late 1960s the basic tenets of ‘mob sociology’ held sway amongst police commanders. The core of this approach is the belief ‘that all crowds can transform into law-breaking mobs’ (Schweingruber, 2000: 373). Influential authors focused on the emotional character of crowds and the supposed emergence of a ‘group mind’: within the crowd the individual’s “conscious personality vanishes” (Le Bon 1895:2). Crowds were seen as herd-like, suggestible and spontaneous. This understanding of (especially riotous) crowds began to be challenged in the 1960s. Smelser (1963) drew on empirical work to argue that crowds were far from being irrational or senseless. Despite this, Smelser regarded public disorder as an attempt to ‘short-circuit’ the political system rather than a rational and considered attempt to achieve social change. Although acknowledging that ‘political powers, through repression and unresponsiveness, are directly involved “in the origins and the content of the ‘outburst’ itself”’ (1970: 51), his major work underplayed the interactive nature of collective behaviour. Downplaying the exacerbation (or indeed instigation) of collective violence by police action, Smelser saw uncompromising police intervention as key to restoring order in riot situations (1963: 267; Waddington & King 2005: 501).

2.2 Waddington (2007: 10) described a gradual shift in UK policing styles from the 1960s and 1970s. This moved away from ‘escalated force’ models – where police ramped up interventions ahead of the ‘riot curve’ – towards dialogue, under-enforcement and negotiation. These paralleled growing recognition that police themselves were important actors in situations of disorder. Following the riots of the early 1980s, unrest was widely seen as a rational response to socio-economic dislocation and aggressive policing (Waddington 1996: 118). The Scarman Report (1981) urged police to invest more in community liaison and in communicating with their public. Waddington and King’s (2009) comparative study of riots in France and the UK suggested that disorder typically occurred in segregated, deprived and stigmatised localities, where young people felt without any ‘future stake in society’. Here riots were sparked off by specific incidents which appeared to confirm police hostility or indifference.

2.3 In short, the current critically informed consensus on mass disorder is that it is often highly patterned and selective, that rioters are more representative of their communities than a depiction as ‘mindless criminals’ suggests, and that police are crucial players in the build up to, outbreak and management of public disorder (Marx 1970; Currie and Skolnick 1970; Reicher et al. 2004; Waddington and King 2005; Wilkinson 2009). The problem is, as Marx (1970: 23) astutely noted, that in correcting flawed depictions of rioters as frenzied and irrational, there is a danger of swinging too far in the opposite direction. Much recent work, for example, focuses on riots associated with social movements and political protests. In the US the focus has been on ‘race riots’; in the UK on violence associated with Luddites, Chartists, and Black and Minority Ethnic youth; in France on unrest in the banlieues. All of these highlight socio-economic deprivation and political injustice. Such a focus has shaped beliefs about the purpose or origin of mass violence. Marx cautioned against the tendency to ‘see all violent outbursts as “rational”, “intrinsically political”, and “instrumental and purposive”’ (1970: 23), and pointed towards what he called ‘issueless riots’.

In such riots, I do not mean to imply that behaviour is necessarily, on the average, any more or less rational, emotional, or destructive than in non-crowd circumstances, nor that it may not be individually instrumental. Such riots are issueless in the sense that a critique of the social order and the belief that violence will help bring about needed social change are relatively unimportant as motivating factors.

The English Riots: Timid policing and tactics?

3.1 As the dust settled on England’s riots, however, the academic consensus on coping with public disorder was shelved as leading politicians were quick to blame an individual criminality born of welfare dependency and ‘entitlement’. Furthermore, in a throw-back to Smelser’s ideas, many in government also blamed the police for delay and timidity, and for having stood back as rioters looted and burned shops. Prime Minister David Cameron complained that ‘there were simply far too few police deployed on to our streets and the tactics they were using weren't working. Initially, the police treated the situation too much as a public order issue - rather than essentially one of crime’.[1] Some police sources suggested a reluctance to ‘go in harder’ since they were wary of being accused of using too much force.[2]

3.2 As we have argued elsewhere (see Stott et al. 2010), there are claims that HMCIC recommendations have led to overly ‘soft’ policing. We drew on contacts accumulated in nearly a decade of research on public order policing to carry out informal interviews with three senior public order commanders and one front-line officer. Given the sensitivities of the recent riots we have chosen not to name any of our respondents in this piece. The public order commanders we spoke to flatly rejected claims of ‘soft policing’. When asked if the policing in London had been ‘too timid’, one commander’s immediate response was: ‘no, [that’s] absolute bullshit’. One Scottish-based Inspector who had been deployed in London pointed out that he was given authority to turn out his officers in full ‘Code 1’ protective gear and to adopt whatever tactics he deemed appropriate. He received ‘a very clear and explicit steer to ‘nip things in the bud’ through dynamic intervention’. Secondly, the intense scrutiny of public order tactics has focused attention too narrowly on protest policing. Another respondent ruefully reflected that the latest version of policing guidelines had ‘lost the disorder element’ noting that: ‘The emphasis is on Keeping the Peace rather than on containing disorder’. Hoggett and Stott (2010) found that public order training and tactics tend revolve around arrest, containment and dispersal and may be ill-equipped to deal with either an emphasis on protest facilitation or with widespread and ‘unfocussed’ disorder.

3.3 A former senior officer believed that too few officers were on duty in Tottenham and that ‘if the officers [present] had acted rather than standing back - we saw the pictures on the television of police officers standing back and allowing people to go looting - then I don't think we would have had the copycat violence [elsewhere]’.[3] One of our interviewees reflected on the differences entailed in policing protest on the one hand, and riot on the other: ‘Tottenham began as a protest and was policed as a protest … the Met [Metropolitan Police Service] were caught on the hop and didn’t expect that kind of immediate escalation’. He noted how aggressive policing at the protest event would have been counter-productive. Another respondent noted, ‘the spontaneity and technology [i.e. communications available to rioters] meant that the disorder was hard to predict and control. There were no leaders to speak to. With no-one to speak to the only recourse was numbers’. This, however, is precisely what the police in Tottenham lacked.

3.4 All interviewees spoke of the fatigue brought on by constant fire-fighting, and noted a lack of numbers on the ground. One respondent highlighted the ‘health and safety aspect: without numbers on the ground, intervention is ineffective at best, dangerous at worst’. All our interviewees agreed with ACPO President Hugh Orde that assets such as water-cannon and rubber-bullets would have been of limited value – merely moving people on to less protected targets – and come at great cost. Not least amongst the costs would be with regards to police legitimacy, already fragile (at best) in many of the affected areas. Such tactics are ineffective at policing what Zygmunt Bauman characterised as ‘riots of defective and disqualified consumers’.[4] As PAJ Waddington, an advocate of paramilitary policing puts it: ‘Might is not right when it comes to the suppression of rioting. The police may “win the battle” but lose the war if their actions are widely perceived as excessive’ (1996: 122 original emphasis). Once disorder breaks out the only effective tactic, without creating problems of legitimacy downstream, is the deployment of very large numbers of officers.

3.5 In the event, flooding London with officers from across the UK put a lid on the riots. Existing police resources and tactics appeared to be very sorely tested by rioters/looters marked by their numbers and their mobility, but flooding an area with officers is not sustainable and so other tactics need to be considered. ‘Facilitation’ is irrelevant where individual and crowd objectives are theft and arson rather than political demands, so do we need to tear up the rule book (again)? Whilst claims of police timidity might suggest as much, such an analysis ignores the social contexts of the unrest. Emphasis on frenzied criminality forecloses analysis of the underlying beliefs, feelings and moral justifications for action. Stepping back from the ‘madness of the moment’ allows us to place the disorder in context and think through alternative modes of police intervention.

The Riots in Context

4.1 We should first recognise that there were multiple riots with varying causes. Events in Tottenham evidenced ongoing tensions between the police and some sections within the local community, partly relating to race, partly relating to anti-gang operations. Tottenham has a tense history of riot – but it would be a mistake to think that the tensions here were mirrored elsewhere. Neither Ealing nor Gloucester, for example, matches this context. In some areas we saw deliberate targeting of ‘designer’ shops; in others the police were the explicit target. We thus find a very varied mixture of ‘historic’ community tensions; a general rebelliousness amongst ‘disaffected’ young people; and criminal opportunism. If we must identify a singular cause, it is the conjunction of three things: disaffection with economic prospects; serious levels of mistrust between young people and police; and a realisation that when faced with large and mobile numbers of rioters, the police are often powerless to stop looting and arson.

4.2 These may not be necessary conditions, and they are certainly not sufficient: many areas of economic and social deprivation saw no disorder. This raises serious questions about ‘mindless, copycat criminality’, and highlights a need to ask why unrest occurred in some areas rather than others. This is not just an issue of policing. Our respondents were intensely annoyed by the focus on policing when they saw the root causes of the disorder as long-standing social problems. Whilst accepting this there are, perhaps questions to be asked about policing in advance of the events. Sharp and Atherton’s (2007: 760) study of policing in the Midlands found that ‘young people from black and other ethnic minority groups … lack confidence and trust in the police service as a whole’.

4.3 Our interviews supported this finding. Reflecting on the experience of policing in London, for instance, one respondent felt that there was no gradual escalation: violence, he reported, ‘goes from zero to five - full on - in an instant’. Correspondingly the policing response was also said to go from turning a blind-eye to relatively minor offences to in-your-face policing in an instant. This was said to be indicative of wider police-community relations – deep mistrust on both sides - across inner-London. We are still some considerable distance from Scarman’s 1981 call for ‘responsible communications’ between the Metropolitan Police and much of their public. The police have worked hard to implement these recommendations, but all too often community liaison means talking to community representatives who do not always represent the community. One constant issue voiced by community activists, rioters and looters was antipathy towards and distrust of the police (see Stephenson 2011). The critical importance – and difficulty - of building trust is emphasised by Reicher et al. (2004: 561) who point towards the deep-seated resentment that often fuels riots and note that ‘groups have collective memories which can sometimes go back well beyond the experience or even the lifetime of any individual member’.

4.4 Whilst much of the rioting has been described as ‘a mass free for all shopping spree’ (Guardian 2011), the immediate trigger for the riots came in a community where tensions between police and local youths were exceptionally tense. Issues of trust and legitimacy are widely seen as helping us understand the causes of the unrest, but are largely absent from analyses of how police should react to it. Much current police tactics are predicated on the actions of rational actors who are prepared to talk. It is all too easy to conclude from this that the principles of facilitation and dialogue that emerged from critique of G20 policing need to be supplemented with more aggressive police measures or equipment to cope with riotous groups.

Conclusion: Far from the madding crowd?

5.1 We have previously (Rosie and Gorringe 2009) suggested that where police public order training is based on worst-case scenarios it risks not simply preparing officers for trouble but causing them to anticipate it. Clearly police do need to prepare for the worst, building trust in kit, tactics, and colleagues. Equally, however, it is ever more apparent that training needs to develop the ‘softer’ public order skills and tactics emphasised in the HMCIC ‘policing by consent’ model. For a start, there is evidence that facilitating the peaceful intentions of crowds, avoiding the disproportionate and indiscriminate use of force, and adopting a ‘graded’ tactical approach are effective (Stott, 2011). More significantly, however, Reicher et al. (2004: 559) argue that ‘Effective public order strategies can transform relationships for the better. Indeed such strategies may be one of the best ways in which one can begin to turn around a spiral of negative inter-relationships with marginalised groups in society.’.

5.2 In March 2011, South Yorkshire Police deployed liaison and dialogue as their key strategy in policing the Liberal Democrat spring conference in Sheffield. We observed liaison officers establish rapport and legitimacy where none had existed before. Using ‘soft’ policing skills and innovative forms of social media engagement (McSeveny and Waddington 2011) they succeeded in reaching out to people who would not normally speak to, let alone work with, the police. (SYP 2011; Theobald 2011). The number of people signing up to follow the police on Twitter during the riots suggests that proactive policing not only promotes police legitimacy before disorder, but can be a useful tool during it too. The sooner we dispense with notions of king mob, as peddled by government ministers and by leading sociologists, the sooner we escape the recurrent search for counter-productive and expensive technological solutions to riotous crowds like water-cannon and rubber-bullets. Perhaps then we could embrace – and resource - innovative and proactive policing that might help contain, and even prevent, outbreaks of mass violent disorder.


1 <>.

2 <>.

3 <>.

4 <>.


ACPO (2010) Manual of Guidance on Keeping the Peace. Specialist Operations Centre, Wyboston: On behalf of the Association of Chief Police Officers and ACPO in Scotland by the National Policing Improvement Agency.

BREWER, J and Wollman, H (2011) ‘Sociologists’ offer to unravel the riots’, Letter to The Guardian, 11 August 2011 (Accessed 09 September 2011): <>.

BURNS, T (2011) ‘Pseudoscience and the London Riots: Folk Psychology Run Amok’, Scientific American Guest Blog August 16, 2011: < > (Accessed 09 September 2011).

CURRIE, E & Skolnick, J (1970) ‘A critical note on conceptions of collective behaviour’, Annals of the American Academy of Political and Social Science. Vol. 391: pp. 34-45. [doi:://]

GUARDIAN (2011) ‘UK Riots Aftermath – Friday 12 August’, The Guardian Online: (Accessed 16 September 2011): <>.

HER MAJESTY’S CHIEF INSPECTOR OF CONSTABULAry (2009a) Adapting to Protest. London: HMIC: < >

HER MAJESTY’S CHIEF INSPECTOR OF CONSTABULARY (2009b) Adapting to Protest: Nurturing the British Model of Policing. London: HMIC: < 091125.pdf>.

HOGGETT, J & Stott, C (2010) ‘The role of crowd theory in determining the use of force in public order policing’, Policing & Society 20(2): pp232-236.

LE BON, G (1895) The Crowd: A Study of the Popular Mind, Mineola: Dover Publications (2002).

MARX, G (1970) ‘Issueless Riots’. Annals of the American Academy of Political and Social Science. Vol. 391: pp. 21-33.

MCSEVENY, K & Waddington, D (2011) ‘Up Close and Personal – The Interplay between Information Technology and Human Agency in the Policing of the 2011 Sheffield Anti-Lib Dem Protest’, in B. Akhgar & S. Yates (eds). Intelligence Management: Knowledge Driven Frameworks for Combating Terrorism and Organised Crime, Part 3, pp199-212. London: Springer-Verlag:

REICHER, S; Stott, C; Cronin, P & Adang, O (2004)An integrated approach to crowd psychology and public order policing’. Policing: An International Journal of Police Strategies and Management 27(4), pp.558-572.

ROSIE, M & Gorringe, H (2009) ‘What a Difference a Death Makes: Protest, Policing and the Press at the G20’, Sociological Research Online, 14(5).

RUDÉ, G (1995) The Crowd in History. London: Serif: 1995(1964).

SCARMAN, L (1981) The Brixton Disorders, 10th-12th April, 1981. London: HMSO.

SCHWEINGRUBER, D (2000) ‘Mob Sociology and Escalated Force: Sociology’s Contribution to Repressive Police Tactics’, The Sociological Quarterly 41(3): pp371-389.

SHARP, D & Atherton, S (2007) ‘To Serve and Protect? The experiences of policing in the community of young people from Black and other Ethnic Minority groups’, British Journal of Criminology 47(5): pp746-763. [doi:://]

SMELSER, N (1963) Theory of Collective Behavior. New York: Free Press.

SMELSER, N (1970) ‘Two critics in search of a Bias: A response to Currie and Skolnick’, Annals of the American Academy of Political and Social Science 391: pp46-55. [doi:://]

STEPHENSON, W (2011) ‘The Riots: How they Began’, BBC Radio 4 – The Report. First Broadcast 25 August 2011 (Accessed 27 September 2011): <>.

STOTT, C (2011) ‘Crowd Dynamics and Public Order Policing’, in T. Madensen & J. Knutsson (eds) Preventing Crowd Violence, pp25-46. Boulder: Lynne Rienner.

STOTT, C; Gorringe, H and Rosie, M (2010) ‘HMIC goes to Millbank: Public order policing following student disorder’, Police Professional 232, 25 November 2010.

STOTT, C and Reicher, S (1998) ‘How conflict escalates: the inter-group dynamics of collective football crowd ‘violence’’, Sociology 32(1): pp353-377.

SOUTH YORKSHIRE POLICE (2011) ‘Social media helps police and police authority inform and reassure region’, South Yorkshire Police Website 19 August 2011 (Accessed 16 September 2011): <>.

THEOBALD, J (2011) ‘Social media helps Police & Police Authority inform and reassure region’, HR Media News August 2011: <> (Accessed 16 September 2011).

THOMPSON, E. P (1991) Customs in Common. Harmondsworth: Penguin.

WADDINGTON, D (2007) Policing Public Disorder: Theory and Practice. Cullompton: Willan Publishing.

WADDINGTON, D and King, M (2005) ‘The Disorderly Crowd: From Classical Psychological Reductionism to Socio-Contextual Theory – The Impact on Public Order Policing Strategies’, The Howard Journal 44(5), pp.490-503.

WADDINGTON, D and King, M (2009) ‘Identifying Common Causes of UK and French Riots Occurring Since the 1980s’, The Howard Journal 48(3): pp245-256. [doi:://]

WADDINGTON, P. A. J. (1996) ‘Public order policing: citizenship and moral ambiguity’, in F. Leishman, B. Loveday and S. Savage (eds) Core Issues in Policing, pp114-130. London: Longman.

WILKINSON, S (2009) ‘Riots’, Annual Review of Political Science 12: pp329-43. [doi:://]

UniS: University of Surrey logo University of Stirling logo British Sociological Association logo Sage Publications logo Electronic Libraries Programme logo Epress logo