The Helpful Bystander: Current Evidence from CCTV-Captured Public Conflicts

The Helpful Bystander: Current Evidence from CCTV-Captured Public Conflicts

Lasse Suonperä Liebst, Richard Philpot, Poul Poder, & Marie Rosenkrantz Lindegaard

Walking through the city centre late at night, many of us may have witnessed examples of public space aggression in which an individual is assaulted. Seeing such incidents, many questions may flash through one’s mind: Should I intervene to help? Would I be helped by others if assaulted? Such questions have traditionally been difficult for researchers to study. Simply asking individuals whether they would intervene in such situations may not provide a realistic picture of how they would actually behave in a mid-violent event. Furthermore, experimental studies provide limited insights regarding how people would act in such dangerous situations, particularly given the ethical and practical difficulties of simulating actual violent emergencies.

The recent advances and coverage of video surveillance technology provides new avenues for researchers to examine how individuals actually behave during violent public events. At the forefront of these developments is an international collaboration between researchers from Denmark, the Netherlands, and the United Kingdom, comprising of sociologists, anthropologists, criminologists, and psychologists. Utilizing video data, this team have begun to challenge the established view within the social sciences that bystanders are largely apathetic and passive to the emergency unfolding in front of them. In the following article, we outline the results of these collaborative efforts and emphasize the importance of studying real-life behaviour in social scientific research.

Intervention is the norm in public space conflicts
An entrenched assumption of the social sciences is that when people are surrounded by strangers in public spaces, they tend to keep themselves to themselves. Such “norms of non-involvement” and “civil inattention” would suggest that in an emergency or moment of disruption, people would ignore and avoid the situation rather than act. This view is mirrored in the influential “bystander effect” theory, which suggests that individuals are less likely to intervene when in a crowd with others. The assessment of individuals is arguably “why should I intervene, when others are available to do so”. Within the last decade, however, experimental work has challenged the idea that bystanders are passive in dangerous situations, in which it is clear that the victim does require help. This work suggests that when it really matters, bystanders actually do intervene. Further, the presence of others offers a welcome support network rather than an excuse for non-intervention (for a research overview, see Liebst et al. 2019).

More recently – leaving the laboratory behind – studies have begun sampling real-life incidents of public aggression to see if witnesses to assaults do intervene. An assessment of 500 police case files on public assaults showed that around three-out-of-four incidences contained at least one intervening bystander who, for example, blocked contact between the antagonists or dragged the perpetrator away (Heinskou & Liebst, 2017). Critically, this high intervention rate may be underestimated, since police case files do not include events that did not receive police attention because of successful bystander intervention.

To get a more representative picture, Philpot, Liebst, Levine, Bernasco, and Lindegaard (2019) used CCTV video data – comprising of aggressive events both known and unknown to the police – to examine how common it is that at least someone intervenes to help. This study, which utilized data from the Netherlands, South Africa, and the United Kingdom, found that at least one bystander (but typically three or four) carried out helping interventions in nine-out-of-ten of the events. Moreover, similar rates of helping behaviour were observed across three different countries. Taken together, these data offer compelling evidence that bystander intervention is the norm in public space conflicts.

With a little help from my friends
With the recent shift towards a realisation that bystanders will intervene to help, the next question is what motivates individuals to do so. Here, researchers have examined different explanations of helping. These explanations include the number of individuals present at an emergency (i.e., the bystander effect), potential gender differences in the helping likelihood (e.g., males with greater physical strength may be more likely to intervene), and whether particular social relations to the perpetrator and victims promote helping. Using Danish CCTV data of public assaults in Copenhagen, Liebst and colleagues (2019) compared these three competing predictors. All three factors were shown to predict intervention, but the social relationship ties between the bystander and a person involved in the conflict played the largest role in explaining intervention. This suggests that friends help friends and that groups “self-police” individuals who behave aggressively.

Are bystander interventions helpful?
Unfortunately, little evidence has addressed this pertinent question. One reason for this relates to the fact that the literature until recently has assumed that bystanders are largely passive. One exception – a study relying on victimisation survey data – suggests that bystanders are perceived as neither harmful nor helpful in half of incidences, helpful in around one third, while only harmful in approximately one-tenth of situations (Hart & Miethe, 2008). While bystander intervention does not always placate the conflict, this research indicates that it is more likely to be de-escalatory than escalatory. Although important, these results are limited by the fact that they are derived from self-report accounts, which are known to provide an inaccurate picture of real-life human interactions.

Addressing the question of bystander helpfulness with CCTV video data, Levine, Best and Taylor (2011) assessed the de-escalatory and escalatory actions of bystanders across 42 public assaults. They found that the vast majority of bystander actions were to calm the conflict rather than perpetuate it. Furthermore, their findings suggest that that one recipe of successful bystander intervention is a setup in which multiple bystanders collaborate to resolve the conflict. Finally, based on our experience of working with hundreds of real-life video recordings of violence, it is our clear impression that persistent de-escalatory bystander actions, with only few exceptions, are helpful in placating conflicts

Is it dangerous for me if I help?
With evidence now suggesting that intervention is helpful from the perspective of the victim, the next question is whether such behaviour is potentially dangerous for the bystander. In recent times, it has been difficult for crime prevention councils to recommend that members of the public intervene in violence, because of the unknown risks involved. In addressing this knowledge gap, Liebst, Heinskou, and Ejbye-Ernst (2018) used video footage of public assaults to ascertain the likelihood that a bystander was victimised after intervening to help. The overall victimisation rate for de-escalatory interveners was quite low, at just over one-in-ten. This study also confirmed the importance of group relationship ties in street violence, with intervening strangers being less likely to receive aggression than those who knew a victim. This suggests that when you intervene as a group partisan, you may also qualify as a target of aggression.

Implications and recommendations from these findings
Traditionally, bystanders to public aggression were conceptualised as passive individuals who remain inactive while a victim needs help. The recent work described above suggests a shift of this narrative away from “passivity” to “activity”. This reframing has important implications for the public’s perception of bystanders as a resource for crime preventive efforts. Given that we now know that bystander intervention is common, the question is not whether we should encourage people to intervene, as people are already actively doing so. Rather, the already active bystanders should be given advice on how best to intervene both effectively and safely.

The still accumulating evidence suggests that collaborative bystander efforts are advisable and that strangers have a privileged position of being able to intervene in ongoing violence with the lowest risk of victimisation. Furthermore, in contrast to previous assumptions, groups may be beneficial for public safety. This is because group members are likely to “self-police” friends who are acting aggressively. Taken together, the evidence alleviates the widespread fear that if attacked in the street no one will act. On the contrary, crowded public places offer a “safety in numbers” that should be appreciated by both the public and law enforcement agencies.

Hart, T. C., & Miethe, T. D. (2008). Exploring bystander presence and intervention in nonfatal violent victimization: When does helping really help? Violence and Victims, 23(5), 637-651.
Heinskou, M. B., & Liebst, L. S. (2017). Gadevold. En sociologisk kortlægning af vold i byen. København: Djøf Forlag.
Levine, M., Taylor, P. J., & Best, R. (2011). Third parties, violence, and conflict resolution: The role of group size and collective action in the microregulation of violence. Psychological Science, 22(3), 406-412.
Liebst, L. S., Heinskou, M. B., & Ejbye-Ernst, P. (2018). On the actual risk of bystander intervention: A statistical study based on naturally occurring violent emergencies. Journal of Research in Crime and Delinquency, 55(1), 27-50.
Liebst, L. S., Philpot, R., Bernasco, W., Dausel, K. L.; Ejbye-Ernst, P., Nicolaisen, M. H., Lindegaard, M. R. (2019). Social relations and presence of others affect bystander intervention : Evidence from violent incidents captured on CCTV. Aggressive Behavior.
Philpot, R., Liebst, L. S., Levine, M., Bernasco, W., & Lindegaard, M. R. (2019). American Psychologist.


Lasse Suonperä Liebst is Associate Professor at the Department of Sociology, University of Copenhagen. His research interests include micro-sociological theory, interpersonal violence, bystander helping behavior, and the development of quantitative methods for the study of face-to-face interaction dynamics. Marie Rosenkrantz Lindegaard is Senior Researcher at the Netherlands Institute for the Study of Crime and Law Enforcement and Associate Professor in the Department of Sociology, University of Copenhagen. Her research interests include interactional aspects of offending, victimization and guardianship, street culture, observational methods, and urban ethnography in South Africa. Richard Philpot is a Research Fellow at the Department of Psychology, Lancaster University. His research focuses on bystander helping and anti-social behavior in public emergencies. Poul Poder is Associate Professor at the department of Sociology, University of Copenhagen. His research interest covers sociological theory, sociology of emotions, conflict, violence, social suffering and positive organizational scholarship.

Image: In this intervention example, drawn from an actual video recording, a bystander steps in and blocks contact between two conflicting parties (illustration by Nor Voldum-Clausen).