Skip to main content
Category: Effects
Type: Social norm / order-maintenance thesis
Origin: James Q. Wilson & George L. Kelling, The Atlantic, March 1982
Also known as: Broken Windows Effect, 破窗效应, Order-Maintenance Policing Thesis
Quick Answer — Broken Windows Theory argues that small, visible signs of disorder—an unrepaired window, graffiti, litter—tell people that rules are weakly enforced, so further disorder and serious crime become more likely. Wilson and Kelling popularized the idea in 1982; later field experiments show disorder can spread across norms, while large-scale crime drops remain heavily debated. Use it as a cue to fix early signals of neglect—not as a blank check for zero-tolerance policing.

​
What is Broken Windows Theory?

Broken Windows Theory is the claim that physical and social disorder, left unchecked, signals that “nobody cares,” which weakens informal social control and raises the odds of further rule-breaking and serious crime.
If a window in a building is broken and is left unrepaired, all the rest of the windows will soon be broken.
Wilson and Kelling’s famous metaphor is less about glass than about normative cues. One unrepaired break invites the next because vandals and residents both update beliefs about guardianship: who notices, who intervenes, and what is tolerated. The pathway can be direct (opportunistic offenders read low risk) or indirect (fearful residents withdraw, so bystander-style guardianship collapses). Related ideas such as bandwagon and herding describe how people copy what seems socially licensed; Broken Windows Theory applies that logic to public order.

​
Broken Windows Theory in 3 Depths

  • Beginner: Treat small messes as messages. An ignored scratch on a shared bike, an unanswered chat flame war, or trash left in a stairwell can teach others that “this place allows it.”
  • Practitioner: Close the loop fast—repair, clean, clarify the rule, and show someone is watching in a fair way. Speed of fix often matters more than severity of punishment.
  • Advanced: Separate the micro-mechanism (disorder cues spread rule-breaking) from the macro-claim (cracking down on minor offenses will cut homicide). The first has experimental support; the second is contested and can produce cobra-effect-style harms if incentives reward tickets over trust.

​
Origin

In March 1982, political scientist James Q. Wilson and criminologist George L. Kelling published “Broken Windows” in The Atlantic Monthly. They argued that disorder and crime are linked in a developmental sequence: untended disorder frightens residents, erodes informal control, and opens space for more serious offending. They drew on police foot-patrol studies and on Philip Zimbardo’s 1969 abandoned-car demonstrations—one car in the Bronx was stripped within minutes; a Palo Alto car sat untouched for more than a week until Zimbardo smashed it, after which destruction spread within hours. The thesis reshaped 1990s policing debates, especially in New York City, where Transit Police chief and later NYPD commissioner William Bratton and Mayor Rudy Giuliani pursued aggressive “quality of life” enforcement of minor offenses. Supporters credited order-maintenance policing for steep crime declines; critics argued national trends, changing drug markets, demography, and felony enforcement explain more than broken windows alone. In 2006, Bernard Harcourt and Jens Ludwig reanalyzed New York data and the Moving to Opportunity experiment and reported no support for a simple first-order disorder→crime link. In 2008, Kees Keizer, Siegwart Lindenberg, and Linda Steg published field experiments in Science showing that observed disorder can raise other norm violations—evidence for spreading disorder that does not by itself prove every policing package works.

​
Key Points

Broken Windows Theory is most useful when you treat disorder as a signal system—not as a slogan for punishing every petty offense.
1

Disorder is a public cue about enforcement

Litter, graffiti, fare evasion, or unanswered abuse tell observers whether rules are real. People need not be “criminal types”; many simply update: if nobody fixes this, my next small violation is cheap.
2

Fear drives withdrawal, and withdrawal removes guardians

Residents who feel disorder rising often avoid public spaces and neighbors. That withdrawal shrinks informal social control—the everyday noticing and gentle correction that prevent escalation—similar to how diffusion of responsibility fuels the bystander effect.
3

Cross-norm contagion is the micro engine

Keizer and colleagues’ Groningen experiments found that seeing one norm broken made people more likely to break another: graffiti nearly doubled littering rates, and messy cues raised envelope theft. Contagion across norms is the cleanest empirical core of the theory.
4

Policy packages are not the same as the metaphor

“Fix the window” can mean rapid repair and community presence—or mass misdemeanor arrests. Outcomes, legitimacy, and side effects differ. When enforcement metrics become the goal, cobra-effect dynamics and over-policing risk appear.

​
Applications

Use Broken Windows Theory to design early repair and clear norms—especially where shared spaces and bandwagon dynamics amplify small cues.

Shared housing and campuses

Fix lobby lights, broken doors, and trash within hours, not weeks. Post who owns the fix and when it will happen. Visible stewardship beats angry posters after the hallway already looks abandoned.

Product and community moderation

Ship a fast first response to spam, harassment, or broken UI. Early, consistent moderation teaches norms; late, selective enforcement teaches that rules are theater—and herding follows the visible behavior.

Teams and workplaces

Close small process breaches (missed standups, ignored deadlines, rude Slack pings) with fair, predictable follow-up. Culture is mostly what gets repaired without drama.

Cities and public space

Prioritize lighting, trash pickup, and rapid graffiti removal as service—not only as arrest volume. Pair cleanup with resident voice so order feels protective rather than hostile.

​
Case Study

In 2008, University of Groningen researchers Kees Keizer, Siegwart Lindenberg, and Linda Steg tested whether disorder spreads across norms in real public settings in Groningen, Netherlands. In one alley where bicycles were parked, a clear “no graffiti” sign was posted. When walls were clean, 33% of people retrieving bikes littered a flyer left on the handlebar. When the same alley’s walls were covered with graffiti, littering rose to 69%—roughly double—despite the same flyer, weather, and time-of-day controls across conditions (about 77 participants per condition). In a later mailbox test, an envelope with a visible €5 note was left protruding from a slot. When the box and surroundings were clean, about 13% of passers-by stole it. With graffiti on the mailbox or litter on the ground, theft roughly doubled to about 25–27%. Across six field experiments, the authors argued that seeing one rule broken weakens inhibition against breaking others—experimental support for the spreading-disorder half of Broken Windows Theory. The boundary note matters for policy. These studies measure short-run norm contagion in everyday settings; they do not prove that every zero-tolerance arrest program will cut homicide. New York City’s 1990–1999 drop in homicide (often cited near 73%) happened alongside many changes—policing tactics, economics, drug markets—so attributing the whole decline to broken windows alone overclaims the metaphor.

​
Boundaries and Failure Modes

Broken Windows Theory fails when the metaphor is treated as a complete crime theory or as moral license for harshness. Boundary 1 — Disorder cues ≠ proven homicide lever. Experimental work supports cross-norm contagion; citywide causal claims linking misdemeanor crackdowns to serious violent crime remain disputed (notably Harcourt & Ludwig, 2006). National crime trends often moved in parallel. Boundary 2 — “Disorder” is socially defined. Behaviors labeled disorderly can track poverty, race, homelessness, or youthful presence more than imminent violence. Without clear, narrowly defined harms, order-maintenance becomes taste enforcement. Common misuse — Zero tolerance without repair capacity. Ticketing every minor offense while leaving broken infrastructure unrepaired teaches fear of police, not care of place. Metrics chase can produce cobra-effect gaming and erode legitimacy—the opposite of informal social control the theory needs.

​
Common Misconceptions

The phrase is famous; the evidence boundaries are easier to miss.
Crime fell sharply in New York in the 1990s, and order-maintenance tactics were part of the story for some analysts. Other cities also saw drops without the same package, and careful reanalyses question a simple disorder→crime chain. Treat the NYC narrative as contested history, not settled proof.
The original metaphor emphasizes repair and presence—“fix the window”—more than maximum penalty. Fast cleanup and fair norms often restore cues better than stacking citations that residents experience as harassment.
The same cue logic appears in classrooms, open-source repos, family chores, and online forums. Wherever shared spaces broadcast “nobody minds,” small neglect becomes the default—whether or not felony crime is involved.
These pages help separate norm contagion, social withdrawal, and policy blowback.

Bandwagon Effect

Explains how visible majority behavior licenses imitation—close to how disorder cues spread.

Bystander Effect

Shows why fear and diffusion of responsibility shrink everyday guardianship.

Herding Effect

Describes following others’ moves when signals are ambiguous—disorder is one such signal.

Cobra Effect

Warns how well-meant enforcement incentives can backfire when metrics become the goal.

Butterfly Effect

Reminds that small initial differences can cascade—useful analogy, different mechanism than norms.

Normalcy Bias

Helps explain why people under-react to early warning signs until damage is obvious.

​
One-Line Takeaway

Fix the first broken window quickly and fairly—treat early disorder as a signal to restore care, not as a license to punish everything that looks messy.