a/justice· 14 October 2026 · 8 min read

What stops violence before it starts?

Art: Omxus Press

In July 2019, in Texas, a grandmother answered the phone. Her grandson said he had bought an assault rifle, planned to shoot up his surroundings, then die in a hail of police bullets. She could hear the gun click as he spoke. She stayed calm.

You can keep lopping branches. Or you can look at the roots.

A phone call

She told him she understood his pain, and she convinced him to get medical help. He had an AK-47, 17 magazines and several knives in a hotel room in Lubbock. Instead of headlines about another mass shooting, there was a firearms charge and a statement from the US Attorney praising a grandmother "who saved lives by interrupting this plot" (Global News, 2019; Borum & Rowe, 2021).

He'd planned his own death along with everyone else's: he could see no other way out. His grandmother showed him another one. Threat assessment teams know the pattern. The best warning before an attack comes from family, friends and colleagues, and what stops them is not knowing whom to tell, or fearing what happens if they do (Borum & Rowe, 2021).

We talk about violence as if it were weather, or as if it lived in bad people. Mostly it's the end of an escalation, and escalations have exits. The question is who's standing near one.

Violence has a moment before it

Violence often unfolds as a closing funnel of options. Someone cornered escalates, like an animal with its back to a wall. Negotiators know to build a "golden bridge" for an opponent to retreat across. And exhausted people, whose self-control has run down, answer provocation with more aggression than rested ones (DeWall et al., 2007). Stepping in early widens the funnel: pause, get help, back down, all before the only path left can't be undone.

Criminologists name the person who opens the exit. Routine activity theory says crime needs a motivated offender and a suitable target to meet in the absence of a "capable guardian", and a guardian's presence alone can prevent it (Cohen & Felson, 1979). Even one friend or passer-by says: you're seen. And being seen gives someone a reason to stop, and a way to do it without losing face.

Strangers help more than we think

Real life turns out kinder than the bystander effect suggests. Researchers coded 219 public conflicts caught on CCTV in Amsterdam, Lancaster and Cape Town. In 91% of them at least one bystander intervened, and the more bystanders present, the more likely someone helped (Philpot et al., 2020). The interventions were often simple: stepping between two people, a gesture, a calming touch. One person stepping in often brings others. We've spent fifty years telling each other that bystanders do nothing. The footage says otherwise.

But help depends on what people think they are seeing. In a classic experiment a woman being attacked shouted either "I don't know you!" or "I don't know why I ever married you!". Bystanders intervened 65% of the time for the stranger and 19% for the "husband" (Shotland & Straw, 1976). People hang back from what they read as a private matter, which is exactly where intervention is most needed. Protection orders get violated in roughly two cases in five (Spitzberg, 2002). Paper only goes so far. A woman's safety can come down to whether the neighbours think it's their business.

When a street tips toward helping

Each person has a threshold: the number of others who must act before they will (Granovetter, 1978). If everyone's threshold is high, no one acts. If a few people with low thresholds move first, they trigger others and the behaviour cascades. In 343 Chicago neighbourhoods, collective efficacy, the shared trust among neighbours plus a willingness to intervene for the common good, predicted lower violence and mediated much of the effect of concentrated disadvantage (Sampson et al., 1997).

Paying neighbours to step in

Ameena Matthews grew up in Englewood, on Chicago's South Side, the daughter of the gang leader Jeff Fort, and spent years as an enforcer in a drug ring. By the time the film The Interrupters caught up with her, she was throwing herself into the middle of a street fight outside her office and talking people down (Chicago Magazine, 2011; Frontline, 2012). An outsider couldn't have done it. She'd lived the life, and the street knew it.

Cure Violence, the programme behind that work, turns stepping in into a job. It hires people from the community, often once incarcerated or in gangs, to find brewing conflicts and mediate before anyone fires (Butts et al., 2015). A John Jay College evaluation compared two New York areas with matched comparison areas. Residents' stated willingness to use violence in serious disputes fell 33% in the program areas against 12% in the comparisons. Gun injuries fell 50% in East New York against 5%; shooting victimisations fell 63% in the South Bronx against 17% (Delgado et al., 2017). The City Council's analysis of all 28 precincts that received a program between 2012 and 2021 found shootings fell about 17% in the first year relative to precincts without one (Nuñez et al., 2024).

Run the same design over the city's public record, 29,744 shooting victims across 76 precincts from 2006 to 2024, and it holds (New York Police Department, 2025). Program precincts had far more shootings to start with. In the five years before each programme began they tracked the others closely. Once it began, shooting victims fell about 13% relative to other precincts, and the gap opened in the first year. When those start years get handed to random precincts 5,000 times, only 1.1% of draws match it.

Who gets helped, and how to make it easier

Willingness to intervene falls with social distance, moral ambiguity and personal risk. A scoping review of bystander intervention in intimate partner violence found one study in which participants were 70% less likely to intervene for an acquaintance than for a family member (Kuskoff & Parsell, 2023). The flip side is a cruel inequity: many victims suffer in silence because neighbours see a "personal matter", when the same people would break down a door for their sister. Yet a meta-analysis of 105 independent effect sizes found the bystander effect shrinks in dangerous emergencies, when the danger is clear (Fischer et al., 2011). Risk can be divided: if several people intervene together, the danger to each falls.

So make it easy. Alerts that reach trusted people first. Bystander training as ordinary as first aid. Safe ways to report, because personal risk stops even the willing. And helpers who come to help: someone who comes to punish can close the exit they came to open. Domestic homicide reviews so often find that somebody noticed and did nothing, or tried and had no support.

What could be wrong

Much of the evidence is observational. The CCTV study shows how often people help, not what would have happened without them. Programs went to the precincts with the most shootings, and the flat pre-period argues against a coincidental decline but cannot rule out a change that arrived in the same year. Cure Violence works inside a small catchment, so a precinct-level count dilutes the effect, and each program ran alongside other city services, so the estimate is for the package, not for interrupters alone. New York is not Perth or Alice Springs: Australian evaluations of community interrupters and night patrols are still needed.

Neighbours, friends and bystanders don't replace the justice system. They're the first line of something kinder, and they get there first.

What you can do

A lot of life's worst hours happen with somebody nearby who didn't know what to do. You can be the person who does.

  1. If you see a fight or hear one through the wall, you don't have to be brave alone. Get someone with you. Call 000 if anyone's in danger. Sometimes "Are you okay?" is enough to break the moment.
  2. If someone tells you they might hurt themselves or others, stay with them, as that grandmother did. Stay calm, tell them you understand it hurts, and bring in help: 000 if it's now, Lifeline on 13 11 14 for them or for you.
  3. If it's a neighbour's house and you're not sure, 1800RESPECT (1800 737 732) will talk it through with you. You needn't be certain before you ask.
  4. Ask your council or state MP what funds community night patrols and interrupter programmes where you live, and how they're evaluated.

That grandmother isn't supposed to be a miracle. Stories like hers get told because they're rare. They needn't be.

A grandmother picked up the phone. That's what roots look like.

Sources

Global News. (2019, August). Grandmother stops mass shooting plot by taking grandson for treatment. https://globalnews.ca/news/5730352/grandma-stops-mass-shooting-texas
Borum, R., & Rowe, M. (2021). The importance of bystanders in threat assessment and management. In J. R. Meloy & J. Hoffmann (Eds.), International handbook of threat assessment (2nd ed.). Oxford University Press. https://doi.org/10.1093/med-psych/9780190940164.003.0024
Butts, J. A., Roman, C. G., Bostwick, L., & Porter, J. R. (2015). Cure Violence: A public health model to reduce gun violence. Annual Review of Public Health, 36, 39–53. https://doi.org/10.1146/annurev-publhealth-031914-122509
Chicago Magazine. (2011). Stars of The Interrupters: Ameena Matthews, Cobe Williams, and Eddie Bocanegra. https://chicagomag.com/?p=7566
Cohen, L. E., & Felson, M. (1979). Social change and crime rate trends: A routine activity approach. American Sociological Review, 44(4), 588–608. https://doi.org/10.2307/2094589
Delgado, S. A., Alsabahi, L., Wolff, K., Alexander, N., Cobar, P., & Butts, J. A. (2017). Demonstrating the effectiveness of the Cure Violence model in New York City [Evaluation report]. John Jay College of Criminal Justice.
DeWall, C. N., Baumeister, R. F., Stillman, T. F., & Gailliot, M. T. (2007). Violence restrained: Effects of self-regulation and its depletion on aggression. Journal of Experimental Social Psychology, 43(1), 62–76. https://doi.org/10.1016/j.jesp.2005.12.005
Fischer, P., Krueger, J. I., Greitemeyer, T., Vogrincic, C., Kastenmüller, A., Frey, D., Heene, M., Wicher, M., & Kainbacher, M. (2011). The bystander-effect: A meta-analytic review on bystander intervention in dangerous and non-dangerous emergencies. Psychological Bulletin, 137(4), 517–537. https://doi.org/10.1037/a0023304
Frontline. (2012). Meet the Interrupters. PBS. https://www.pbs.org/wgbh/frontline/article/meet-the-interrupters
Granovetter, M. (1978). Threshold models of collective behavior. American Journal of Sociology, 83(6), 1420–1443. https://doi.org/10.1086/226707
Kuskoff, E., & Parsell, C. (2023). Bystander intervention in intimate partner violence: A scoping review of experiences and outcomes. Trauma, Violence, & Abuse. https://doi.org/10.1177/15248380231195886
New York Police Department. (2025). NYPD shooting incident data (historic) [Data set]. NYC Open Data.
Nuñez, M., Avram, R., Koepcke, E. J., Martinez, R., & Moussawi, A. (2024). Cure Violence programs in NYC. New York City Council Data Team. https://council.nyc.gov/data/cure/
Philpot, R., Liebst, L. S., Levine, M., Bernasco, W., & Lindegaard, M. R. (2020). Would I be helped? Cross-national CCTV footage shows that intervention is the norm in public conflicts. American Psychologist, 75(1), 66–75. https://doi.org/10.1037/amp0000469
Sampson, R. J., Raudenbush, S. W., & Earls, F. (1997). Neighborhoods and violent crime: A multilevel study of collective efficacy. Science, 277(5328), 918–924. https://doi.org/10.1126/science.277.5328.918
Shotland, R. L., & Straw, M. K. (1976). Bystander response to an assault: When a man attacks a woman. Journal of Personality and Social Psychology, 34(5), 990–999. https://doi.org/10.1037/0022-3514.34.5.990
Spitzberg, B. H. (2002). The tactical topography of stalking victimization and management. Trauma, Violence, & Abuse, 3(4), 261–288. https://doi.org/10.1177/1524838002237330

Keep reading