Illusion of Order: The False Promise
of Broken Windows Policing
By Bernard Harcourt
Harvard University Press
288 pages, $35
On March 16, 2000, a team of three undercover narcotics officers
working Midtown Manhattan held out for one more bust, hoping to
round out the day's arrests at 10. For the setup, one of the officers
played the user. He approached a black man standing outside a cocktail
lounge: "You know where I can find some weed?" Offended at being
taken for a dealer, the man told the undercover cop to get lost.
Angry words escalated to flying fists. Another officer rushed in,
gun drawn. A moment later, the man was dead.
The officer who shot unarmed Patrick Dorismond belonged to the
NYPD's recently hatched Operation Condor, which aggressively pursues
misdemeanor drug offenders. (The unit shares its name with another
notorious undercover operation--the U.S.-sanctioned hunt for South
American dissidents during the '70s and '80s.) New York's Condor
is but the newest incarnation of an "order maintenance" approach
to policing that has dominated New York for almost a decade. Order
maintenance--as old a concept as policing itself--was revived nearly
20 years ago by the much-lauded "broken windows" theory. In The
Atlantic Monthly, criminologists James Q. Wilson and George
L. Kelling set out their common-sense thesis that disorderly streets
communicate a lack of control, emboldening criminals, distressing
law abiders and plunging the community down a "spiral of urban decay."
Broken windows is perhaps the most seductive criminal justice theory
to emerge in
decades. But in
Illusion of Order: The False Promise of Broken Windows Policing,
legal scholar Bernard Harcourt meticulously scrutinizes and evaluates
the theory that has inspired, among others, the "quality-of-life"
campaign in New York, loitering ordinances in Chicago, a juvenile
snitching initiative in Charleston and the youth curfews observed
in hundreds of cities throughout the United States.
From the '70s onward, policymakers faced with social disillusionment
and dwindling public resources increasingly abandoned the root-cause
approach to crime prevention. By the late '80s, their cities plagued
by delinquency and decrepitude, public policymakers seized upon
the broken windows theory and its promise of a cure-all for crime.
Yet there is no evidence that disorder actually causes crime, Harcourt
argues, and policing it has proven at odds with democratic principles
of civil liberty.
The broken windows theory also marks a turn toward incorporating
social meaning and norms into the classical model of criminology.
According to "social norm" proponents, the "social meaning" of order
dissuades would-be criminals by signaling that the community is
in control and crime ill-tolerated. By dispersing loiterers, painting
over graffiti and mending the proverbial broken windows, cities
can control serious crimes such as homicide and rape.
But police officers are not maintenance men or social workers.
Order-maintenance policing may in fact cut crime, Harcourt says,
but not through the mechanism of social influence. "It is somewhat
jarring to uncover what appears to be a straightforward policy of
aggressive misdemeanor arrests masquerading as a neighborhood beautification
program or as an innocent phenomenon of social influence," he writes.
Harcourt notes that Kelling himself has said that broken windows
reduces crime "at least in part because restoring order puts police
in contact with persons who carry weapons and who commit serious
Under the guise of order maintenance, then, police cast a wider
net for illegal weapons and drugs, ensnaring those with outstanding
warrants and enlisting more informants in the process. Harcourt
cites former New York Police Commissioner William Bratton's telling
description of early subway sweeps as a "bonanza." According to
Bratton, "Every arrest was like opening a box of Cracker Jack. What
kind of toy am I going to get? Got a gun? Got a knife? Got a warrant?"
Although this approach has led to the capture of some real criminals,
a vast number of innocent people, such as Patrick Dorismond, are
unavoidably enmeshed in the sweeps, stop-and-frisks and buy-and-busts
that one NYPD officer referred to as "fishing expeditions." The
New York State Attorney General has found that 7 to 10 people are
stopped before the NYPD nets one arrest.
Though some have attempted to distinguish between broken windows
and zero-tolerance--indiscriminate arrests for minor offenses--Harcourt
shows that there is little distinction. From the start, order-maintenance
policing in New York City and elsewhere has emphasized sweeps, arrests
Unsurprisingly, these methods have swelled courts, jails and probation
officers' caseloads. In New York, the mean streets have been painted
blue with more than 40,000 officers, in rarefied units like Street
Crimes and Condor, who puff up arrest statistics by busting panhandlers,
prostitutes, peddlers, drunks, junkies and low-level dealers. After
Dorismond's death, it was reported that 75 percent of Condor arrests
were misdemeanors, and that misdemeanor narcotics arrests had increased
by 68 percent over the previous year. In Chicago, under anti-gang
and anti-narcotics loitering ordinances, police have arrested tens
of thousands of people for just standing around.
When police exercise discretion--true zero-tolerance is not really
feasible--it is often at the expense of people of color, as numerous
studies of racial profiling show. Indeed, broken windows criminalizes
large swaths of the population--including who Wilson and Kelling
call the merely "obstreperous or unpredictable" --making it
easier for the rest of us to condone treatment of others that we
would not accept for ourselves.
Historically, disorder was annoying, but not criminal. John Stuart
Mill's famous "harm principle" holds that "the only purpose for
which power can be rightfully exercised over any member of a civilized
community, against his will, is to prevent harm to others." The
broken windows theory has undermined this principle by asserting
the harm of disorder. At the expense of civil liberties, Harcourt
writes, the theory enforces "aesthetic preferences" rather than
Returning to Michel Foucault's Discipline and Punish, Harcourt
relates this development to an earlier shift in the law from the
criminal act to the criminal soul. Broken windows, he says, "focuses
on the presence of the disorderly rather than on the criminal act.
It judges the disorderly not simply by giving the individual a criminal
record, and not simply by convicting the person, but by turning
the individual into someone who needs to be policed and surveyed,
relocated and controlled."
Worse yet, Harcourt demonstrates through painstaking statistical
analysis that the broken windows theory is just not supported by
the data. The most frequently cited studies "proving" the theory
at best establish a tenuous connection between minor disorder and
serious crime--specifically, robbery. At worst, researchers have
misrepresented their own data; studies show absolutely no causal
relationship between disorder and crime. The most careful data show,
by contrast, that both crime and disorder likely have deeper roots
in structural disadvantage, disfranchisement and a lack of community
Harcourt's analysis shows that New York's crime wave had already
begun to ebb before the quality-of-life initiative, and that there
are plenty of other likely factors for the decrease--fewer males
aged 18 to 24, more felons in prison for longer, an improved economy,
less crack cocaine, and new computerized crime-tracking systems.
Moreover, Harcourt notes that cities such as San Diego, which reduced
misdemeanor arrests, also experienced sharp declines in street crime
during the '90s.
If it is doubtful that broken windows has been the mechanism behind
diminishing crime, it is clear that by targeting the "disorderly,"
the theory has fueled aggressive police practices that bear most
heavily on the dark, young and poor. While Manhattan's main drags
may seem cleaner and safer than they were a decade ago, off to the
side, in the neighborhoods and prisons and precinct houses, lurks
a larger disorder in the form of racial profiling, police brutality,
increased detention, expanded surveillance and the criminalization
of a greater share of our citizenry. Among the many casualties are
innocents such as the 26-year-old Dorismond, a security guard and
father of two young girls. "What we are left with today is a system
of severe punishments for major offenders and severe treatment for
minor offenders and ordinary citizens, especially minorities," Harcourt
writes. "We are left with the worst of both worlds."
Megan Costello, a freelance writer based in New York
City, can be reached at [email protected]