Broken Windows Theory

What Is the Broken Windows Theory?

Broken windows theory proposes that visible signs of disorder, such as litter or public drinking, encourage more serious crime and antisocial behaviour.

In short, if small problems are ignored, bigger ones tend to follow.

Wilson and Kelling argued that unattended disorder sends a social signal: no one is in charge, and rule-breaking goes unpunished.

The Broken Windows Chain Reaction

1. Disorder sends a signal: Broken windows, graffiti, or loitering suggest that “no one is in charge.”

2. Residents withdraw: Law-abiding people start to feel unsafe and spend less time in public spaces.

3. Informal control breaks down: With fewer “eyes on the street,” community monitoring weakens.

4. Crime increases: The area becomes more attractive to offenders, leading to serious crime and further decay.

In this way, small disorders like vandalism and public drinking are seen not as harmless misdemeanors, but as early warning signs of deeper social breakdown.

Source: Hinkle, J. C., & Weisburd, D. (2008). The irony of broken windows policing: A micro-place study of the relationship between disorder, focused police crackdowns and fear of crime. Journal of Criminal Justice, 36(6), 503-512.

What Counts as “Disorder”?

BWT distinguishes between two main types of visible disorder:

Physical Disorder

At the heart of BWT is the idea that visible disorder communicates meaning. It acts like a signal that tells residents, visitors, and even potential offenders something about the community.

This refers to the visible decay of the physical environment, such as:

• Abandoned or run-down buildings
• Broken windows
• Graffiti and vandalism
• Litter, rubbish, or abandoned cars

When small acts of disorder go unchecked, it sends the message that no one cares or no one is enforcing the rules.

For potential offenders, this can make crime seem less risky, since they assume there’s a low chance of being caught.

For residents, it can signal a loss of community pride or authority, creating a sense of uncertainty and vulnerability.

Social Disorder

Disorder doesn’t just affect how people see their surroundings. It also shapes how they behave.

When people perceive that others ignore social rules, they may begin to mirror that behaviour.

Social disorder refers to behaviours that make a place feel unsafe or uncontrolled, such as:

• Loitering

• Public drinking or drug use

• Prostitution

• Panhandling or harassment

• Loud noise or public arguments

Both disorders send messages about the social norms of a community. An orderly, well-maintained area signals that rule-breaking isn’t tolerated. A neglected one suggests that people can get away with more.

Why Environment Matters

Academics explain the broken windows theory using three main ideas:

1. Social norms and conformity: People adjust their behaviour to fit what seems “normal” for an area. If disorder is visible, they may relax their own standards.

2. Routine monitoring: In safe, orderly places, both residents and businesses act as informal guardians, what Jane Jacobs called having “eyes on the street.”

3. Social signaling: Visible neglect tells potential offenders that the risk of getting caught is low, while a clean, cared-for environment communicates control and vigilance.

This means that the environment itself acts as a form of communication, shaping people’s sense of safety and influencing behaviour.

Testing the Theory: Keizer, Lindenberg and Steg (2008)

The clearest direct test of whether disorder actually spreads bad behavior came from field experiments in Groningen, the Netherlands, published in Science as “The spreading of disorder” (Keizer, Lindenberg, & Steg, 2008).

• Aim: To test whether visible disorder in one domain increases rule-breaking in a completely unrelated domain, not just more of the same disorder.
• Method: Researchers attached an advertising flyer to bicycles parked in an alley with no bin nearby, so a returning cyclist had to carry it away or drop it. In the disorder condition, the alley wall was covered in graffiti next to a “no graffiti” sign; in the order condition, the wall was clean.
• Results: Littering roughly doubled in the graffiti condition compared to the clean condition, even though graffiti and littering are unrelated rule violations. Further experiments found the same pattern for other norms, including a higher rate of petty theft from an envelope with a visible banknote in disordered settings.
• Conclusion: Keizer and colleagues called this a cross-norm inhibition effect: seeing one norm broken weakens people’s commitment to rules in general. It is the closest the evidence comes to a direct causal mechanism for broken windows, though it shows short-term contagion of minor rule-breaking, not the original claim that disorder drives serious, predatory crime.

Examples of Broken Windows Policing

1969: Philip Zimbardo’s Abandoned-Car Experiment in the Bronx and Palo Alto

In 1969, Stanford psychologist Philip Zimbardo ran a social experiment with two abandoned cars. Both had their license plates removed and their hoods left up, but he placed them in very different neighborhoods.

The first was a predominantly poor, high-crime neighborhood in the Bronx, and the second was a fairly affluent area of Palo Alto, California. He then observed two very different outcomes.

In the Bronx, the car was vandalized within ten minutes.

A family first removed the radiator and battery, and within 24 hours, every valuable part had been stripped.

Soon after, random acts of destruction began: the windows were smashed, the seats were torn, and children started using the wrecked vehicle as a playground.

Meanwhile, the car in Palo Alto sat untouched for more than a week. Then Zimbardo himself struck it with a sledgehammer, and only then did bystanders join in the destruction.

Zimbardo concluded that visible neglect invites further vandalism.

When something looks abandoned, people feel less constrained by social norms and more likely to act destructively.

1982: Kelling and Wilson’s Follow-Up Article

Thirteen years after Zimbardo’s study was published, political scientist James Q. Wilson and criminologist George Kelling published an article in The Atlantic that applied Zimbardo’s findings to entire communities.

The article grew out of a 1970s foot-patrol experiment in Newark, New Jersey, that Kelling had evaluated.

Returning officers to foot patrol did not reduce recorded crime, yet residents in those areas felt safer anyway.

Wilson and Kelling took that finding seriously.

Foot-patrol officers, they argued, were not mainly fighting crime. They were maintaining order: moving along rowdy teenagers and enforcing the informal rules residents cared about.

Crucially, the 1982 article was an argument, not an empirical demonstration. Wilson and Kelling offered a plausible mechanism and a memorable metaphor, but they never ran a controlled test showing that disorder actually causes crime.

Skogan’s 1990 Study and Its Later Reanalysis

Through the 1980s and 1990s, sociologist Wesley Skogan gave the theory its first empirical scaffolding. He surveyed and observed conditions across roughly forty U.S. neighborhoods.

Skogan reported a robust link between disorder and both fear of crime and later victimization, describing a self-reinforcing “spiral of decay” (Skogan, 1990). His work was widely cited as proof that disorder and crime were empirically connected.

Later reanalyses questioned that conclusion. Once researchers accounted for a handful of influential cases and confounding neighborhood characteristics, the disorder–crime link looked weaker.

Disorder and crime, critics noted, are often measured in overlapping ways and cluster in the same disadvantaged places, which builds on the older social disorganization theory of crime.

George Kelling and Catherine Coles drew these strands together in their 1996 book Fixing Broken Windows, which became the operational manual for the policing programs of the 1990s.

Early 1990s: Bratton and Giuliani’s implementation in NYC

The most famous application of the Broken Windows Theory took place in New York City during the mid-1990s.

• Policy adoption: City leaders introduced order-maintenance policing, also called zero-tolerance or quality-of-life policing, based on the idea that targeting small crimes would prevent bigger ones.

• Key figures: Mayor Rudy Giuliani and Police Commissioner William Bratton credited Wilson and Kelling’s 1982 essay, arguing the theory explained New York’s crime decline.

• How it worked: Police strictly enforced minor laws, including turnstile jumping, panhandling, public drinking, vandalism, loitering, and prostitution. Misdemeanor arrests rose from about 133,000 in 1993 to over 205,000 in 1996.

• Perceived success: Supporters credited the crackdown with New York’s historic drop in violent crime. A 2001 study by Kelling and Sousa attributed much of the decline to misdemeanor arrests.

• Debate: Critics argue the drop had other causes: the waning crack epidemic, economic growth, and mean reversion, where high-crime areas naturally regress toward average over time.

The Broken Windows Theory (BWT) and its policy counterpart, order-maintenance policing (OMP) or zero-tolerance policing, have shaped crime prevention strategies worldwide.

However, decades of research and critical analysis have raised major doubts about the theory’s core assumptions, its causal claims, and its real-world consequences.

Critics argue that BWT misinterprets the relationship between disorder and crime, fails empirically, and reinforces racial and class inequalities through over-policing.

Misinterpreting the Relationship Between Disorder and Crime

A central flaw of the Broken Windows Theory is its claim that visible disorder directly causes serious crime.

Critics suggest that this relationship is correlational, not causal, and that both disorder and crime are symptoms of deeper social conditions.

1. The Collective Efficacy Hypothesis

Sociologists Robert Sampson, Stephen Raudenbush, and Felton Earls (1997) proposed that a third factor, collective efficacy, better explains why some communities have higher crime rates.

Collective efficacy is the mutual trust among neighbors combined with their shared willingness to intervene and maintain public order.

Sampson and Raudenbush (1999) then tested this directly against broken windows theory using systematic social observation. Trained researchers coded thousands of videotaped street segments for disorder, and the disorder-crime link mostly vanished once collective efficacy was taken into account.

Neighborhoods with strong social cohesion tend to have both lower disorder and lower crime. This is not because one causes the other, but because residents collectively manage their own environment.

2. Empirical Evidence Against a Causal Link

A meta-analysis of 300 studies by O’Brien et al. (2019) found no direct causal connection between visible disorder and increased crime.

The authors noted that many earlier studies relied on flawed methods, failing to separate perception from reality or cause from correlation.

Disorder, they concluded, does not automatically lead residents to commit more crimes or even feel significantly less safe.

Similarly, David Thacher (2003) argued that crime reductions in cities like New York during the 1990s were wrongly attributed to Broken Windows-style policing.

He and others (e.g., Metcalf, 2006; Sridhar, 2006) noted that the timing of falling crime coincided with other powerful social and economic shifts, including:

• The waning of the crack cocaine epidemic

• The Rockefeller drug law crackdowns

• New York City’s late-1990s economic boom

Further, Bernard Harcourt (2009) showed that many cities without Broken Windows policies also saw sharp crime declines, while others that adopted them did not.

A pivotal study by Harcourt and Ludwig (2006) examined the U.S. Department of Housing and Urban Development’s relocation program, which moved inner-city residents to more orderly neighborhoods.

Contrary to BWT predictions, these residents did not commit fewer crimes after relocation. This is strong evidence that environmental “order” alone does not reduce criminal behavior.

3. Contemporary Research

A 2024 systematic review and meta-analysis updated the evidence base (Braga, Welsh, & Schnell, 2024). It separated disorder-focused strategies into two kinds: community problem-solving efforts and aggressive order-maintenance arrests.

The community and problem-solving strategies produced modest but real reductions in crime. Aggressive arrest tactics showed little benefit and carried serious costs.

Newer studies also keep testing the theory outside the United States.

Research linking clandestine dumpsites to local crime patterns in Mexico City found associations broadly consistent with the disorder–crime link. These studies inherit the same causal-inference limits as the earlier U.S. research (Vilalta et al., 2020; Fondevila et al., 2023).

Misunderstanding Why Crimes Are Committed

Broken Windows Theory assumes that people are rational actors deterred by the fear of getting caught.

In more “orderly” neighborhoods, informal social control supposedly increases the likelihood of detection, discouraging crime.

However, critics argue this assumption is overly simplistic. Many individuals engage in crime due to structural or personal factors such as:

• Poverty and unemployment

• Social pressure and inequality

• Mental illness or addiction

• Homelessness and lack of resources

These causes are often unrelated to the perceived likelihood of arrest.

By focusing on surface-level signs of disorder rather than structural inequality, Broken Windows policing risks punishing the symptoms of poverty instead of addressing its causes.

Over-Policing and the Expansion of Police Power

The application of BWT has often led to aggressive enforcement of minor offenses, producing a cycle of over-policing and criminalization of marginalized groups.

1. Escalation of Arrests and Coercion

In New York City, quality-of-life policing led to a 50% rise in misdemeanor arrests between 1993 and 1996.

People were detained for minor offenses like fare evasion, loitering, or sleeping on benches – sometimes spending up to 60 hours in custody.

Complaints of police brutality and misconduct soared during this period (Civilian Complaint Review Board reports).

Even Wilson and Kelling acknowledged that maintaining order sometimes required “informal or extralegal steps”, quoting officers who admitted, “We kick ass.”

These tactics often inflicted deep harm on low-income communities while showing little proven benefit in reducing serious crime.

Racial and Class Bias

Perhaps the most powerful criticism of BWT concerns its racialized and class-based impact.

1. Racial Profiling and Perception Bias

By giving officers broad discretion to decide what counts as “disorderly,” BWT allows implicit biases to shape enforcement.

• Roberts (1998) argues that this discretion enables police to criminalize communities of color under the guise of maintaining order.

• In a landmark study, Sampson and Raudenbush (2004) found that when shown identical neighborhoods with equal levels of graffiti and litter, participants perceived more disorder in areas with more Black residents.

• This shows that racial stereotypes, not objective conditions, strongly influence perceptions of disorder, and by extension, where policing is concentrated.

The result is a cycle where minority neighborhoods are more heavily policed, leading to more arrests, reinforcing public perceptions of criminality, and justifying further surveillance.

2. Class Inequality and Criminalizing Poverty

BWT also disproportionately targets working-class and poor individuals.

Acts considered “disorderly” (e.g., sleeping in public, panhandling) are often survival behaviors, not crimes.

Yet they are punished more harshly in public spaces where visibility makes poverty criminal.

People without private spaces, especially the homeless, are thus more vulnerable to arrest and harassment.

Critics argue this transforms policing into a “war on poverty crimes” rather than a strategy for public safety.

Broken Windows to Stop-and-Frisk

Broken Windows policing has evolved into or justified more invasive practices, such as stop-and-frisk, brief, suspicion-based searches of individuals in public.

In 2008, New York police made 250,000 stops, but fewer than 0.1% led to the discovery of a firearm (Vedantam et al., 2016).

By 2011, police stopped 685,000 people, and nine out of ten were completely innocent (Dunn & Shames, 2020).

These figures highlight both the inefficiency of the method and its disproportionate targeting of young Black and Latino men.

Critics argue that Broken Windows provided the intellectual foundation for such intrusive practices. In their view, it expanded police discretion at the cost of civil rights.

Real-World Applications Beyond Policing

Broken windows thinking has spread well beyond street policing, wherever visible order is thought to shape behavior.

Urban Design and Greening

The theory converges with Jane Jacobs’s earlier idea that lively, well-used streets deter crime by concentrating informal surveillance, or “eyes on the street” (Jacobs, 1961).

One modern application reverses the theory’s usual logic: instead of punishing disorder, it removes it through greening. A difference-in-differences study in Philadelphia found that cleaning and planting vacant, rubbish-strewn lots reduced gun assaults and residents’ fear, and improved perceived safety (Branas et al., 2011).

Schools

Educators have imported the same framework. Visible disorder in a school, such as vandalism or unaddressed rule-breaking, is thought to erode students’ sense of safety and staff’s collective efficacy.

Plank, Bradshaw, and Young (2009) applied broken windows and related ideas to middle schools and found that physical and social disorder were linked to heightened student fear and weaker collective efficacy among staff.

Online Communities

The same logic underpins much online moderation. Unaddressed spam, abuse, or vandalized wiki pages are thought to signal that norms go unenforced and invite further degradation.

Moderators often remove low-level violations quickly to head off this “digital broken windows” spiral, extending the cross-norm signaling mechanism Keizer and colleagues demonstrated experimentally rather than a separately tested finding of its own.

Public Health

Because disorder is linked to fear, stress, and withdrawal from public space, broken-windows-style interventions overlap with public health goals. Efforts to improve neighborhood conditions, like better lighting and blight removal, target wellbeing and physical activity as much as crime (Branas et al., 2011).

What alternative crime prevention strategies exist?

Alternative crime prevention strategies take a very different approach from Broken Windows policing.

Instead of focusing on punishing minor disorder, these methods aim to tackle the root causes of crime, reduce opportunities for offending, and strengthen communities through prevention and rehabilitation rather than punishment.

1. Tackling Social and Economic Inequality

Many sociologists argue that crime grows out of social inequality, unemployment, and poor living conditions, not broken windows or graffiti.

Approaches inspired by theories like Conflict Theory and Social Ecology focus on fixing these deeper problems.

• Create Economic Opportunities: Offer good jobs and fair wages in struggling urban areas, so people have less reason to turn to crime out of desperation.

• Improve Neighborhood Conditions: Invest in affordable housing, clean public spaces, and safe infrastructure. Poverty, overcrowding, and neglected buildings can all contribute to crime.

• Community Greening: Fund local projects that hire residents to clean up abandoned spaces or turn empty lots into parks, building pride and community ownership.

2. Reducing Opportunities for Crime

Other strategies focus on making crime harder to commit by changing the environment or increasing supervision.

These ideas come from Routine Activity Theory and Situational Crime Prevention.

• Situational Crime Prevention (SCP): Design environments so crime is less tempting: add better lighting, lock doors, and improve surveillance to make crime riskier and harder to get away with.

• Problem-Oriented Policing (POP): Instead of reacting after crimes happen, police identify specific local problems and create tailored prevention plans with community input.

• Civil Remedies (Third-Party Policing): Use laws that make landlords, bar owners, or other third parties help prevent crime, such as nuisance abatement laws and alcohol server liability rules.

3. Strengthening Families and Communities

Crime is also linked to weak family bonds and lack of social support.

Strategies in this area focus on helping people build strong relationships and develop positive social behavior, starting early in life.

• Support Families: Parenting programs can help strengthen communication and discipline within families, reducing the risk of youth offending.

• Early Childhood Intervention: Provide education and support for children in high-risk households before problems escalate.

• Youth Programs: Offer after-school and community activities that give young people structure, connection, and purpose.

• Gender Socialization: Encourage boys and men to develop empathy, emotional regulation, and non-aggressive coping skills.

4. Reforming the Criminal Justice System

Finally, some approaches aim to transform the justice system itself by focusing on rehabilitation and reintegration instead of punishment.

• Rehabilitation and Diversion: Provide education, job training, and substance-abuse treatment to help offenders build better lives, steering low-level offenders into community support instead of prison.

• Community-Based Corrections: Use alternatives to incarceration, like probation, community service, or supervised programs, that keep people connected to their families and jobs.

• Restorative Justice: Bring together victims, offenders, and community members to repair harm and rebuild trust, helping offenders take responsibility and reintegrate into society.