As the United States ebbs closer to its next election season, with the anticipated risk of protests becoming violent, the swath of campus disturbances across the United States in April 2024 prompts these questions:
- What can be done to help police agencies improve their crowd management response?
- What might success look like for U.S. policing in this highly challenging and often controversial sphere of operational activity?
The Challenges of Policing Protest Crowds
According to a nongovernmental research organization, the Armed Conflict Location and Event Data project (or ACLED), the number of pro-Palestine demonstrations in the United States nearly tripled in April 2024 compared with March 2024. ACLED argues that this spike appears to have been sparked by events at Colombia University on April 18 and culminated in a large-scale confrontation between police and protestors at the University of California, Los Angeles (UCLA) on April 30. The ACLED data are also interesting because they highlight that, of the 553 campus protests identified between April 18 and May 3, 97 percent were actually overwhelmingly peaceful, with no associated violence or property damage.1
The conflicts and policing operations surrounding this wave of campus protests made headline news across the world, some even invoking memories of the tragedies of the Vietnam War protests at Kent State (Ohio) and Jackson State (Mississippi) in the 1970s. They were also a powerful reminder of the crisis of 2020. Sparked by the murder of George Floyd in Minneapolis, Minnesota, those demonstrations also rapidly spread , with several escalating into some of the most serious and embedded disturbances in U.S. history. For example, the city of Portland, Oregon, was confronted by 175 days of consecutive rioting. Yet, ACLED data are interesting once again, as they show that, of the 2,400 demonstrations that constituted the wave, less than 220 involved any form of violence. In other words, 93 percent of the 2020 Black Lives Matter (BLM) protests remained entirely peaceful.2
What this glimpse at the data exemplifies is how evidence often contradicts assumptions about crowd psychology and behavior. Shaped by headline-grabbing events, filtered through the lens of partisan 24/7 television news coverage and increasingly amplified via social media, one’s vision of protests can very easily become distorted. A belief dominates that protest crowds are volatile and dangerous with a tendency toward chaos and disorder. In fact, as the data make clear, the opposite is true. The general pattern is that the last two major waves of protests that have swept across the United States were actually overwhelmingly peaceful. It is in just a small minority of cases that confrontations developed.
Learning from Critical Incidents
There are very few truisms in evidence-based public order policing, but there is at least one: Reflection and learning nearly always follow some kind of critical incident that has, by definition, already brought the legitimacy and effectiveness of the police into question.
There is a noticeable pattern internationally that, periodically, crowd events occur that escalate into critical incidents. These have been stadium disasters or near misses, major riots, or fatalities. On each occasion, the critical incidents have created a shockwave so powerful it has forced statutory reflection by external stakeholders upon contemporary policing practices and policies. This has often taken the form of inquiries, which invariably have led to a series of recommendations. In the United States, in 2020, these also took the form of court injunctions rendering illegal what had otherwise been standard police tactics, such as the use of explosive munitions and gas. These processes then force the door of change to open, and sometimes evidence-based reforms emerge.
While the recent campus protests did not yet reach that formal scrutiny threshold in the United States, the Charlottesville, Virginia, disturbances of 2017, several of the BLM conflicts of 2020, and the invasion of Capitol Hill on January 6, 2021, all arguably did, and change is consequently occurring. This is nothing new, of course. In fact, perhaps the first time that this kind of critical incident and statutory scrutiny process happened in modern U.S. crowd policing history was in response to the disturbances that rocked communities in the late 1960s and early 1970s. The wave of riots and deaths of protestors ultimately led to two important and influential inquiries, the Kerner Commission in 1967–1968, closely followed by the U.S. President’s Commission on Campus Unrest or the so-called Scranton Commission in 1970.
The Kerner Commission was important because it challenged the widely shared belief that such disturbances can be understood merely as the result of the activity of so-called troublemakers hijacking crowds and agitating disorder. Instead, the commission ruled out such explanations.3 The Scranton Commission was significant because it recognized the important contributing role of poor interoperability and unnecessarily aggressive crowd policing. It recommended that police agencies improve their training and coordination. As such, these inquiries underpinned the emergence of Field Force Formations (FFF), a development that is the closest thing the United States currently has to national guidance and doctrine for public order policing—a doctrine that is already nearly 50 years old and, judging by contemporary research, theory, and practices, increasingly outdated as a unitary tactical response.4
A Postcard from Europe
During the last five decades since the Scranton Commission, there have been several statutory inquiries in the European context that have led to significant evidence-based reforms to crowd policing practices. Perhaps the most significant of these was in 2001, following a series of protests surrounding a European Union (EU) summit in the southern Swedish city of Gothenburg, which escalated into widespread rioting that continued for several days. A subsequent inquiry, led by a former Swedish prime minister, exposed poor levels of underlying police coordination and a lack of dialogue between police and protestors as primary contributing factors. Subsequently, the Swedish Police developed a new nationally coordinated strategic and tactical model called the Special Police Tactic. This was supported by a training framework informed by the latest scientific theory and evidence from crowd psychology, all designed to professionalize staff, improve interoperability, and enhance delivery. (The Scranton Commission would certainly have been impressed.) In parallel, the Swedish police also developed the world’s first Dialogue Police Unit, which operated as an integral part of their public order response. The role of dialogue police personnel was not enforcement but to operate with very high levels of discretion to problem-solve, promote police legitimacy, support de-escalation, and facilitate two-way communication between protestors and police. The new model was implemented in 2005, to immediately powerful effect, placing the Swedish Police at the forefront of evidence-based innovation in protest crowd management.5
In the United Kingdom (UK), much of the learning in this field of operational practice has also developed in the wake of disaster. In 1985, 39 people were killed as a result of crowd violence involving fans of the English soccer club Liverpool F.C., at the Heysel stadium in Belgium. Just four years later, poor police crowd management practices led to the death of 97 Liverpool fans during a crowd crush incident at Hillsborough Stadium, in the English city of Sheffield. Subsequent inquiries into these disasters have led to major reforms to public order policing across Europe, not least of which has been the development of coordinated guidance and accredited standards of training that must be possessed by all police officers in the UK before they can take on a significant command role in any public order operation, be that sporting or protest related.6 This contrasts heavily with many police agencies in the United States who still deploy incident commanders and operations chiefs who routinely have no training-based accreditation and sometimes little to no background experience. If history is anything to go by, such a lack of preparedness is inviting negative outcomes. As the Scranton Commission noted, professionalization is still urgently required.
The Value of Crowd Psychology
The lessons learned in Sweden and the UK coalesced in 2009, following another critical incident where a member of the public was killed as the result of the forceful police dispersal of a large demonstration surrounding a G20 summit in London, England. The subsequent inquiry by the UK’s main police oversight body made multiple recommendations for updating statutory guidance and training for crowd policing.7 Central to this was a call to modernize police commanders’ understanding of the latest science on crowd psychology. In short, a problem was identified: Despite having achieved a standardized national curriculum and accreditation framework, public order command training in the UK was still reliant on outdated ideas about mob psychology that had been rejected from science but were still being perpetuated across UK policing. It was recognized that this divergence from science was leading to poor decision-making by public order commanders who were misunderstanding the dynamics of the crowds that they were trying to manage. Consequently, this was increasing the risk that the police would inadvertently act disproportionately or miss opportunities to de-escalate.8
Stemming from 19th century France, these now-outdated psychological perspectives continue to create a distorted image of crowd psychology as inherently dangerous, irrational, volatile, and easily exploited by agitators. Despite evidence to the contrary, it is presumed ordinary people lose self-control in crowds as a function of anonymity. And as a consequence, they are open to mindless exploitation by a nefarious minority capable of whipping up a mob into a so-called frenzy. If this theory were valid, then it follows that the majority of crowd events would be violent and chaotic. As noted above, even a cursory glance at the evidence shows this concept is unsustainable, which is why this theoretical perspective has been almost entirely rejected by the scientific community.9 Nonetheless, such ideas still remain a mainstay of policing doctrine in the United States. This is perhaps why public order policing is often referred to as crowd control, measured by its capacity to suppress the assumed inherent disorderly tendencies of crowds through the ability to react with overwhelming force.
It is an unfortunate truth that research has demonstrated that viewing the dynamics of crowds in this way is not just wrong, it’s counterproductive. This is because it leads police agencies to respond to crowds in ways that can inadvertently escalate tension, most significantly through the use of indiscriminate force. This is not necessarily through the fault of the police or something that can always be avoided, so much as it is an outcome of a flawed understanding, which promotes blindness to the powerful role that social psychological dynamics actually play in shaping crowd behavior.
“The policing model created an environment where the crowds effectively policed themselves”
It is now understood that collective action in crowds is made possible through participants sharing a meaningful social identity. These identities are a form of group level psychology that create mutual subjective meanings among crowd participants in ways that form and constrain collective action. For example, where group members’ identities are defined by beliefs and values that reject violence, then those promoting confrontation tend not to be influential, as they are seen by other crowd participants as counter-normative outsiders. However, these social identities are not fixed, because they are intimately linked to the social context in which they arise. As such, they can be highly dynamic in both form (i.e., what the crowd participants believe is appropriate and inappropriate to do) and content (i.e., who is and is not considered by crowd participants as “one of us”).
Robust research findings have shown that identity change during crowd events can occur when coercive police action is experienced as indiscriminate and illegitimate by crowd participants. These kinds of intergroup interactions inadvertently provoke situations where identity content alters in such a way that confrontation against the police comes to be seen by many crowd participants as appropriate (e.g., as legitimate resistance and the reassertion of First Amendment rights). This is primarily because it is the police who are understood by those participants to be behaving illegitimately. In addition, the boundaries of identity also shift—such that those involved are now not merely people focused on the initial issue motivating the protests but those also interested in asserting perceived rights. With this shift, those ideologically oriented toward confrontation are more likely to be seen as sharing a common identity, as one of “us,” rather than as previously marginalized “outsiders.”10
A New Tool in the Toolkit
While it is a difficult lesson to administer, there is substantial evidence that this social psychological process, revolving around the dynamics of identity, power, and legitimacy, is at the heart of why many crowd events transition into violence and spread from one location to another.11 But this theoretical approach should not be naively dismissed on the mistaken assumption that it is just a bland academic criticism of policing. Rather, it should be understood for what it has become—a very powerful tool for policing, because, in the words of Kurt Lewin, a famous social psychologist, “There is nothing more practical than good theory.”12 Indeed, this new crowd psychology has been helping police agencies across the world to manage crowd events more efficiently and effectively for over two decades, most commonly in Europe but more recently within the United States.
One of the best examples in Europe to date was the policing of a major international soccer tournament in Portugal in 2004. Plagued by a history of major rioting at previous competitions, the Portuguese authorities utilized the new crowd theory to design an evidence-based policing model for the month-long tournament. The flexibility and effectiveness of their graded tactical approach meant the police managed to respond more proportionately to emerging tensions and avoid inadvertently igniting the crowd dynamics that would have otherwise escalated the crowds into riots. A major scientific study of the tournament demonstrated that the policing model generated widespread perceptions of the legitimacy of policing among fans attending the tournament. The social context that approach created promoted high levels of positive affiliation between fans and police, which empowered behavioral norms of self-regulation among the fans. In other words, the policing model created an environment where the crowds effectively policed themselves.13
“Led to increasing self-regulation within crowds in ways that have dramatically reduced the need for arrests or any police use of force”
There were no major incidents of disorder in any of the match cities, an achievement unheard of since the origins of the tournament in the early 1980s. Indeed, it is interesting that two consecutive nights of rioting related to the tournament did occur, but these were outside of the host cities in an area controlled by Portugal’s second police force, which did not adopt this new science-led approach.14 As a result, the new science and its associated policing model was adopted by the EU and Europe’s main soccer regulating authority, the Union of European Football Associations (UEFA) as the basis for their safety and security policy and guidance to date.15
A Dialogue-Led Approach
At first, the evidence-based graded policing approach inspired by social psychology was largely confined to Europe and had not yet been tested in the United States. The crisis of 2020 changed that. In the midst of the inquiry into what went wrong, the social psychology approach came to the attention of the Seattle, Washington, Office of the Inspector General. Through this, a relationship was developed with the Seattle Police Department (SPD). Via existing collaborations with respect to their Bicycle Rapid Response Teams (BRRTs), the network soon grew to involve the Columbus, Ohio, Division of Police (CPD). Shortly afterward, the U.S. Federal Protection Services (FPS) became involved as well as the Portland, Oregon, Police Bureau. Over the past three years, these agencies have been investing in the science and adapting the associated policing model to their own jurisdictions.
In Seattle, the collaboration led to the development of a new dialogue police unit, which the SPD refer to as the Public Order Outreach and Engagement Team (POET). To support this innovation, the SPD also commissioned a two-day training course on the underlying science, which was delivered to all SPD command staff in 2023.16 Correspondingly, the FPS incorporated the theory into their new training program. The Columbus Division of Police also recognized that they needed to move away from a crowd control philosophy toward a science-based approach. Consequently, they developed and began deploying a specialized unit of dialogue police in June of 2022. Initially this was just a few officers in blue tabards, deployed to establish and explain their role to protestors. But even these small ad-hoc deployments quickly produced positive outcomes in terms of de-escalating tensions, managing emergent problems, and avoiding circumstances where police use of force was required. Their Dialogue Team quickly grew to a unit of approximately 30 officers who all received specialized training in crowd science and are now regularly deployed as CPD’s primary tactical response to First Amendment activity throughout the city.17
Given these early successes, in late 2023, the U.S. Department of Justice’s Office of Community Oriented Policing Services, along with the City of Columbus agreed to fund a strategic partnership between the Ohio State University, Keele University in the UK, and the Columbus Division of Police.18 This led to the development of the ENABLE project, a methodology designed to develop and help implement an evidence-based approach to crowd management in the United States. The project has undertaken multiple observations to examine the evolution of the Columbus crowd policing response as well as to map its impact. Early findings from the project have provided some very promising results and exciting developments. The work has identified highly effective interventions by the Dialogue Team that have led to increasing self-regulation within crowds in ways that have dramatically reduced the need for arrests or any police use of force, while still facilitating the expression of protestors’ First Amendment rights. The ENABLE project has also allowed for exchange so that teams from different agencies have been able to formally observe and collaborate with each other during deployments to help advance knowledge and share good practices.
Dialogue officers are different than typical protest police personnel because they are discouraged from engaging in police enforcement action, focusing instead on facilitation of First Amendment activity. They do this by seeking to establish rapport and engaging in transparent and open dialogue with protestors, who are often initially highly suspicious of the police and refuse to openly engage in conversation with them. Regardless of this reticence, ENABLE research has shown that dialogue officers in both Columbus and Seattle have worked to build channels of effective communication with protestors through engagement, negotiation, and problem-solving. Over several months, these officers have been able to develop functioning bonds with protestors in ways that have bolstered and promoted self-regulation in the crowd.
These dialogue officers do not operate in isolation but are part of a graded tactical response. Consequently, they play a key role as liaisons or negotiators working between police commanders and protest groups. As well as negotiating solutions, dialogue officers provide enhanced situational awareness for commanders who, as a result, have been able to make more informed decisions about proportionate police responses. The two-way communication has helped build a “no surprises” approach supported by additional layers, which include low-level enforcement bicycle squads and arrest and control teams. Each of the units has a defined role within the overall mission. While the Dialogue Team engages with the crowd, bicycle squads help direct and protect the crowd if it becomes mobile, and enforcement teams are responsible for information-led targeted use of force and arrest actions if those responses become necessary.
The Way Forward for the United States
In January 2024, the Portland, Oregon, Bureau of Police organized and hosted the Western States Public Order Conference that was attended by over 300 delegates from police agencies across the United States and Canada. The ENABLE project was presented on the fourth day of the event, where discussion moved from reflections about 2020 to the way forward for U.S. crowd management policing. The ENABLE presentation focused on the work undertaken in Columbus, which ignited considerable interest from other agencies. Since then, training in dialogue policing has been delivered to more than 20 U.S. agencies.
Perhaps most notably in the United States, the Columbus Dialogue Team was recently invited by the Milwaukee, Wisconsin, Police Department (MPD) to provide mutual aid during the Republican National Convention in July 2024. Following strategic discussions with MPD’s operational command, the Dialogue Team was deployed as the primary tactic for managing the main protest on the first day of the event. The impact was immediate and profound. The Dialogue Team, integrated into a command structure designed to maximize their effectiveness, achieved remarkable success in promoting self-regulation among the crowd, enhancing situational awareness for command decisions and ensuring the proportional use of force only when necessary. These efforts collectively played a critical role in preventing conflict escalation, demonstrating the effectiveness of a dialogue-led approach in maintaining public order during high-profile events.19
There is a clear and growing interest in dialogue-led, evidence-based policing across the United States, supported by an expanding body of research showing its effectiveness in diverse contexts. As police agencies increasingly recognize the value of this approach, it is essential to implement these strategies appropriately to enhance public safety, build community trust, and promote police legitimacy. ENABLE’s team of researchers stand ready to assist in that endeavor.
To conclude, it is worth noting that, as of May 2024, police agencies had faced more than 130 lawsuits stemming from claims of police misconduct during the crisis of 2020, with settlements totalling nearly $150 million paid out to protesters, journalists, legal observers, and bystanders.20 These legal challenges highlight the urgent need for a transformative shift in crowd management strategies within the United States. The research-driven, dialogue-led policing model offers a proactive solution to reduce such costly litigation and conflict—while restoring public trust and enhancing police legitimacy.
The ENABLE approach, which fosters a partnership between police practitioners and academic researchers, serves as a model for effective and legitimate public order policing in 2024 and beyond. By focusing on de-escalation, open communication, and respect for First Amendment rights, this model helps police not only manage crowds more effectively but also prevent conflicts from escalating.
This proactive strategy is about more than just responding to crises; it’s about building community relationships and strengthening the legitimacy of police actions. With the 2026 FIFA World Cup also looming on the horizon, police chiefs should consider how this approach can benefit their agencies and their communities. Engaging with the ENABLE network provides access to cutting-edge research, specialized training, and opportunities for international collaboration, all aimed at enhancing public order policing.12 The project demonstrates how social psychology and policing can work together to set a new standard for crowd management in the United States that not only prioritizes safety and accountability but also fosters lasting police legitimacy and trust within the communities that both police and academic institutions are dedicated to serving. d
Notes:
1Bianca Ho and Kieran Doyle, “Pro-Palestine US Student Protests Nearly Triple in April,” ACLED Insight, May 2, 2024.
2Roudabeh Kishi and Sam Jones, “Demonstrations and Political Violence in America: New Data for Summer 2020,” ACLED Insight, September 3, 2020.
3National Advisory Commission on Civil Disorders, Report of The National Advisory Commission on Civil Disorders (U.S. Government Printing Office, 1968).
4President’s Commission on Campus Unrest, Report of the President’s Commission on Campus Unrest (U.S. Government Printing Office, 1970).
5Kęstutis Lančinskas, “The Riot That Changed European Policing,” Opinions & Reviews (EU blog), originally published in Evropeiska Pravda on September 21, 2017.
6UK College of Policing, “Public Order Public Safety,” webpage.
7Her Majesty’s Inspectorate of Constabulary, Adapting to Protest: Nurturing the British Model of Policing (London, UK: 2009).
8James Hoggett and Clifford Stott, “Post G20: The Challenge of Change, Implementing Evidence-based Public Order Policing,” Journal of Investigative Psychology and Offender Profiling 9, no. 2 (June 2012): 174–183.
9John Drury and Stephen Reicher, “Crowds and Collective Behaviour,” in Oxford Research Encyclopedia of Psychology (Oxford University Press, 2020), accessed via Sussex University Figshare.
10Clifford Stott, “Crowd Psychology and Public Order Policing: An Overview of Scientific Theory and Evidence” (presented to the HMIC Policing of Public Protest Review Team, University of Liverpool, UK, School of Psychology, September 2009).
11John Drury et al., “A Social Identity Model of Riot Diffusion: From Injustice to Empowerment in the 2011 London Riots,” European Journal of Social Psychology 50, no. 3 (April 2020): 646–661.
12Kurt Lewin, Field Theory in Social Science: Selected Theoretical Papers by Kurt Lewin, ed. Dorwin Cartright Lewin (Tavistock, 1964): 169.
13Clifford Stott et al., “Tackling Football Hooliganism: A Quantitative Study of Public Order, Policing and Crowd Psychology,” Psychology, Public Policy, and Law 14, no. 2 (2008): 115–141.
14Clifford Stott et al., “Variability in the Collective Behaviour of England Fans at Euro2004: ‘Hooliganism’, Public Order Policing and Social Change,” European Journal of Social Psychology 37 (2007): 75–100.
15EU Council Res. Notice no. 2016/C 444/01, EU Football Handbook (November 29, 2016).
16Hanna Scott, “Report on 2020 Protests Recommends New Training for SPD, Debunks ‘Mob Psychology,’” MyNorthwest News, May 20, 2020.
17Nora Igelnik, “Promoting Conversation: Columbus Police Dialogue Team Informs Protestors of Rights,” The Lantern, March 20, 2024.
18Award Number: 15JCOPS-23-GG-02066-PPSE.
19John Diedrich and Alison Dirr, “A ‘Police Dialogue Team’ from Ohio Is Facilitating Peaceful Protests during the RNC. Here’s How,” Milwaukee Journal Sentinel, July 15, 2024.
20Alice Speri, “$150m Paid in Police Misconduct Claims Shows Violent Response to 2020 Protests, Experts Say,” The Guardian, May 25, 2024.
21Ohio State University, “Strategic Command and Urban Disorder,” course webpage (John Glenn College of Public Affairs).
Please cite as
Clifford Stott, “Leveraging Crowd Psychology to Prevent Violence,” Police Chief Online, September 18, 2024.
