Is Crime Contagious?

Experiments vindicate the broken windows theory of how disorder spreads


New York City was a mess in the 1980s. I know because I lived in the East Village for a good stretch of that decade. Fortunately, my apartment building was on the marijuana block (the crack block was two over), so things were relatively mellow. My building's first story was covered in graffiti. Bums (I mean homeless men) often curled up on the floor of the foyer to sleep when it got cold. A shanty town sheltering 100 homeless people established itself in nearby Tompkins Square Park. All of the cars on my block had signs in their windows reading, "Don't Bother. Radio Already Stolen."

Nearly every subway car was decorated with graffiti. (There was fun graffiti. Artist Keith Haring used chalk to draw his own urban hieroglyphics on the black paper used to cover up old advertisements in subway stations. If only I had managed to peel one off and take it home.) Of course, drivers in New York also had windshield cleaning services offered by legions of squeegee men. And we attended poetry slams in Alphabet City bars like the appropriately named Safety in Numbers. The crime rate was legendary, inspiring many New Yorkers to celebrate "subway vigilante" Bernhard Goetz for shooting four men he said were trying to rob him. Nevertheless, I loved New York (and still do).

But what explains the disorder in Gotham and other American cities? In 1982, two social scientists, George L. Kelling at Rutgers University and James Q. Wilson at Harvard University, proposed the "broken windows theory" to explain both how disorder spreads and how it is sustained. In The Atlantic Monthly, the two asserted that "disorder and crime are usually inextricably linked, in a kind of developmental sequence. Social psychologists and police officers tend to agree that if a window in a building is broken and is left unrepaired, all the rest of the windows will soon be broken….one unrepaired broken window is a signal that no one cares, and so breaking more windows costs nothing." Before I moved to the East Village, for instance, I was sleeping on a friend's couch in the north Bronx. Everyday, I took the train past miles of abandoned apartment buildings where the city had replaced thousands of broken windows with plywood painted to look like real windows. Some even featured silhouettes of people and potted plants. They fooled no one.

The idea behind the broken windows theory is that if people look around and see other people violating norms, they will tend to violate them as well. In the 1980s and 1990s, city governments and police departments stepped up their enforcement measures against petty crimes, such as painting graffiti, panhandling, littering, and subway fare jumping. The hope was that by minimizing public disorder, the police would help communities create crime-deterrent environments. Most of the evidence for the value of this kind of policing is based on studies of what happened to crime rates once police began to crack down on incivilities. In recent years, some analyses have questioned the broken windows theory as a strategy for effective policing.

Now, a new study (additional online info here) published in Science provides some strong experimental backing for the broken windows theory. Dutch researchers from the University of Groningen, led by social scientist Kees Keizer. conducted six experiments to see if signs of disorder would encourage people to engage in norm violation themselves. The short answer: Yes.

In the first study, the setting was an alley in Groningen near a shopping district that is commonly used to park bicycles. A prominent sign in the alley prohibits graffiti. The researchers used rubber bands to attach flyers to the handlebars of each bike wishing shoppers a happy holiday from a non-existent sportswear store. The researchers monitored what the bikers did with the flyers when the wall in the alley was free of graffiti and when it was covered with it. The result: only 33 percent littered when the alley was graffiti free whereas 69 percent did when graffiti was present.

In a second study, the researchers set up a temporary fence closing off the main entrance to a car park. But they left a 20-inch gap in the fence with two signs posted in the immediate vicinity—one sign forbade locking bicycles to the fence and the other prohibited the use of the closed entrance and directed people to another entrance about 200 yards away. When four bikes were parked but not locked to the fence, only 27 percent of people stepped through the gap to go to their cars. When the bikes were locked to the fence, 82 percent walked through the prohibited gap.

In a third study, the researchers set up a situation in which a grocery store posted a sign in a parking garage asking people to please return their shopping carts to the store. They then put the same holiday greeting flyers as in the first experiment under the windshield wipers of cars. When no shopping carts were in the garage, 30 percent littered and when shopping carts were present, 58 percent littered.

Do only visual cues trigger disorder? In the Netherlands, very strict laws forbid setting off fireworks in the weeks before New Year's Eve. So the researchers lit fire crackers out of view of people picking up their bikes from a busy parking shed near a train station. Once again, the bikes had those flyers attached to them. When no fire crackers were set off, 52 percent littered, but 80 percent did when they heard the bangs.

In the final two studies, the researchers wanted to see if signs of disorder would induce people to steal. They left an addressed envelope with a transparent window hanging out of a mail box with a five Euro note clearly visible inside. In one case, the mail box was graffiti free and the area around it was clear of litter. Only 13 percent of people stole the envelope. When the mailbox was covered with graffiti, 27 percent stole the money. When the area around the mailbox was littered, 25 percent took the envelope. The researchers report, "We found that when people observe that others violated a certain social norm or legitimate rule, they are more likely to violate even other norms or rules, which causes disorder to spread."

As a fairly frequent visitor to New York, I can attest that much of the city has been transformed in the past two decades. My old block in the East Village is now graffiti free and lined with trees, shops, and restaurants. How much credit to give to policing based on the insights of the broken window theory for lower crime rates is controversial, but this new study shows that the theory deserves some. As the Dutch researchers conclude: "There is a clear message for policy makers and police officers: Early disorder diagnosis and intervention are of vital importance when fighting the spread of disorder."

Ronald Bailey is reason's science correspondent. His book Liberation Biology: The Scientific and Moral Case for the Biotech Revolution is now available from Prometheus Books.