THE HFG REVIEW OF RESEARCH (Vol. 2, No. 1, Spring 1997)

Crime is Down? Don't Confuse Us with the Facts
Karen Colvard

"HOMICIDES PLUNGE" The New York Times of June 2, 1997 reported new FBI statistics on the continuing rapid decrease in serious crime in the United States for the fifth year in a row. This downward trend surprised many when it began in 1992 - lots of us discounted it as a statistical blip. Now rates for violent crime and property crime in the U.S. are the lowest since crime rates began to rise in the late 1960s in tandem with dislocating changes in our inner cities. However there are two problems with this good news: criminologists can't explain why it is happening and ordinary citizens don't seem to believe that it's true.

In March 1997 America Online's national news discussion board, referring to the dramatic drop in violent crime being documented nationwide, asked its readers, "Do you feel safer?" Almost nobody did. A typical response:

I am sick and tired of hearing statistics being flouted about crime is down, etc. etc. Who exactly computes this impossible venture? No I don't feel safe....what we really need is police protection, victim rights...more laws to protect the innocent. Aren't we too taken up with the criminal's rights, race, etc.? Until they (whoever they are) stop spending money on false surveys about a serious problem such as the actual crime problems that are getting worse, take that money and put more police on the streets and burn half of those stupid laws that protect the obvious criminal. It's a disgrace that our country can't get this together.1

The public holds firm opinions about crime and justice, and the increasingly punitive policies being offered by elected (or would-be elected) policy makers--more prisons, stricter sentencing, an expanded death penalty, and harsh responses to offenses involving any type of drug--are meant to allay the fears of those who believe that America is a dangerous place, beset by random violence which could strike anyone, anywhere, fueled by an epidemic of drug addiction and related crime. This has led Norwegian criminologist Nils Christie to label American penal practices "democratic crime control," and he doesn't mean that as a compliment--he sees in it a coercive state controlling a disenfranchised minority who are perceived as a "criminal class" by the mainstream majority of voters, who fear crime committed by these others against them.2 A recent State Supreme Court ruling in California is typical of this: it allows police to prevent people they suspect of involvement in gang activities from any form of association--"standing, sitting, walking, driving, gathering, or appearing anywhere in public view," the court order said. They can also be arrested and imprisoned for making loud noises, climbing trees, or carrying beepers, marking pens, or marbles.3 Los Angeles County District Attorney Gil Garcetti told the New York Times in February 1997, "It gives us a valuable tool in our arsenal against gang activity. It permits us to move before crimes are committed."

The lack of high-level debate over such strategies was evident in the most recent presidential election campaign. Bill Clinton's reiterated pride in his Crime Bill's additional "100,000 police on the streets" was criticized: "It's been the most careful political calculation, with absolutely sublime indifference to the real nature of the problem.... [Violent crime] is a problem that is concentrated within very clearly defined geographic boundaries. And the President is going to spread cops into every suburb of the country."4 But not by his opponent Bob Dole, who in his campaign doggedly vowed "to attack the root cause of crime--criminals, violent criminals."5

However, the lack of sensible leadership at the highest levels of the government is matched by the bewilderment of those whose calling is to understand patterns of crime and violence. The mostly-liberal criminologists who firmly believe that restrictive laws and harsh punishments contribute to rises in crime, not decreases, have offered very little useful analysis of why violent crime is abating or advice about how to encourage further reductions. "This is a humbling time for all crime analysts. It is a puzzlement," says Princeton criminal justice professor John J. DiIulio, Jr., who is himself an equally persuasive proponent of preventive, faith-based social interventions and strong punitive state responses to criminality.6

Part of the problem may be that the public doesn't really know what it thinks it knows. In 1992 Canadian criminologist Julian Roberts reviewed studies by himself and others which revealed the public's extremely limited knowledge about crime and punishment.7 The Canadian public tended to overestimate victimization rates for violent crime, the number of violent crimes, the increase in crime, and recidivism rates, and not by small errors. For example, while some 13-17% of first-time offenders in Canada are subsequently convicted of another crime, 4 out of 5 of the survey respondents offered guesses ranging between 60 and 100%. They were also ignorant of penalties for particular crimes, which they estimated to be much lower than they were in fact. This poses a problem not only to the political realities of legislating responses to crime but also to deterrence theory. If people--including, presumably, some who might be contemplating committing a crime--think penalties for crime are slight, is anybody likely to be deterred by threat of punishment?

Roberts believes that part of this misinformation comes from news media reports, which tend to highlight violent crimes in decontextualized, oversimplified reporting. He cites a 1990 study in the U.S. which showed that 30% of news stories about crime featured homicide, which accounted that year for only .02% of reported crimes.

George Gerbner, the retired dean of the Annenberg School of Communications at the University of Pennsylvania, has shown that heavy television users, in comparison to moderate viewers, are particularly inaccurate in their perceptions of the threat of violent crime. He has spent his career demonstrating that television, in both news and entertainment programs, has fostered a "mean world" view of human relationships. "Growing up in a violence-laden culture breeds aggressiveness in some and desensitization, insecurity, mistrust, and anger in most," he says. "Punitive and vindictive action against dark forces in a mean world is made to look appealing, especially when presented as quick, decisive, and enhancing our sense of control and security."8 He thinks "facts" about crime garnered from TV and attitudes learned from how television characters, even--especially--heroes, use violence are important motivators for the public's support of capital punishment and coercive forms of crime control.

Roberts also reviewed experimental evidence from social psychology that people tend to be over-influenced by a single case, which can go on to determine their view of a complex problem. As with the advertisements featuring recidivist rapist and murderer Willie Horton in the Bush-Dukakis presidential contest, public opinion can be moved by a single example. Studies also show that people tend to hold opinions about criminal justice issues with a high degree of confidence, that we are most resistant to changing our most confident opinions, and that we tend to give most credit to evidence which confirms what we already think.

Misinformation and misplaced concerns are implicit even in the arena of criminal justice scholarship: A 1994 call for proposals from the National Science Foundation for a $12 million "Violence Research Consortium" gave three justifications for further research: high levels of crime, random violence, and fear of crime. In fact--and the U.S. government should have known--violent crime was already on a downward swing; "hot-spots" in poor neighborhoods in American inner cities were the quite predictable sites of violence and troubled young men the most likely perpetrators and victims; and widespread fear of crime might not have been much of a problem if the public understood these well-documented facts. Only then would rational public policies become appealing to most voters.9

The popular assessment that America's violent culture makes us the world leader in violent crime fuels the fear of crime which so worries the National Science Foundation and conceals within a cultural argument a simple instrumental explanation. The truth is that while U.S. rates of interpersonal crime such as assault are more or less equal to those in European and Asian countries which keep comparable statistics, our rates of homicide--even after recent reductions--are still some ten times as high. And this is because interpersonal crime in America is much more likely to be accompanied by the use of a gun, which is more likely than other weapons to lead to a death. Clinton administration advances in gun regulation, although they have been weak and partial, may account for some part of the recorded reductions in homicide, and further attention to this problem is one clear direction in which politicians and criminologists could find common ground, if resistance on the part of some segments of the public could be overcome.

"Disarming Youth," in this issue, discusses the contribution of gun availability to the dramatic rise of youth homicide in the 1980s and reviews some programs designed to get guns out of the hands of teenagers. Youth crime has also been decreasing (but only since 1992), and some analysts have explained the overall crime drop simply by noting that teenagers commit proportionately more crimes than adults and that we have fewer teenagers today than in the 1980s. They forecast another upswing when the children of the late-reproducing baby boom reach their teenage years. However, the rate of youth crime has decreased, not only the number of crimes committed by teenagers, and it is possible that teenagers of the future will continue to commit crimes at lower rates than 1980s teenagers, for whatever reasons.

Criminologists hold firm opinions, too, and can resist new ideas just as firmly as do other members of the public. Many analysts probably to some degree deserve law professor Randall Kennedy's exasperated comment (speaking specifically about disagreements over race and crime): "Too many commentators make exaggerated claims for results they prefer, denying evidence and arguments that contradict, or at least complicate, the positions they espouse."10

In a 1995 review article of several good books taking positions against the explosive growth of incarceration in America, Francis T. Cullen admits that

a whole generation of criminologists were raised to mistrust state power to do good, to believe that "nothing works" to change offenders, and to embrace "doing justice" as a means of "doing less harm." This pessimistic narrative, which seeks to restrain abuse and not to accomplish good, remains plausible to many criminologists, but it sparks little response from the public and has largely lost its power to humanize corrections.11

Cullen goes on to argue that prisons can become effective rehabilitators of career criminals.12

The always quotable Professor DiIulio has said, "It seems that you need a Ph.D. in criminology to doubt the proposition that putting criminals in prison will keep down crime."13 He would argue, for example, that if an average street robber (like those reported on in "Creating the Illusion of Impending Death," in this issue) commits even ten crimes in a year, each year in prison prevents the victimization of at least ten targets. From the point of view of one of those people not held up at gunpoint, prison works. But can it be credited with the recent dramatic decreases in crime?

Over the past twenty-so years crime rates have fluctuated: up in the early seventies, down again in the early eighties, up again from 1985 until 1992 when the current dip began. Yet the state and federal prison population has risen steadily from just over 200,000 in 1973 to over one million today. The argument for more imprisonment founders on this paradox: when crime decreases, imprisonment has done it. But when crime increases, we need more prisons. The reasonable assessment is that imprisonment has indeed contributed something to crime control; however, the extremely high rate of imprisonment in the U.S. compared to other countries, particularly other democracies, is a symptom of a social problem equal in weight to our problem with criminal violence and deserving equal concern. When 800 per 100,000 of our male population and 4,350 per 100,000 of our black male population are incarcerated, accusations that the penitentiary is serving as a substitute for more rational and effective ways of giving meaning to the lives of the marginal and dysfunctional begin to make sense.

However, it is not the rate of incarceration that has attracted the most attention--and the greatest credit--for the current crime reduction. Instead news has been about a new style of policing in the streets and communities of our big cities, often called "community policing."

A 1982 article in the Atlantic Monthly by James Q. Wilson and George L. Kelling provided the inspiration for these efforts.14 They described a slippery slope of decline in any neighborhood which might begin with one broken window, which no-one fixes. Pretty soon another window gets broken, kids throw rocks at the rest of them, the neglected building gives the impression of a down-at-the heels neighborhood, fearful citizens avoid it, criminal citizens perhaps move in, and you have a dangerous slum. The sunny converse of this scenario is a neighborhood where small problems are attended to and never grow to be large ones.

Kelling went on to hands-on work as an advisor to William Bratton, who as police commissioner in Boston had initiated community policing there before moving to New York in 1990 to become the chief of the NYC Transit Police. The city's subway system had become very run-down, dirty, and disorderly and was being deserted by law-abiding riders, who had good reason to fear crime in its tunnels and trains. Bratton, following Kelling's theory, reasoned that if transit police reacted to petty crimes such as turnstile jumping, aggressive panhandling, and public drunkenness (the broken windows of the dismal subway experience), once these problems were fixed, the improved atmosphere of the system would encourage riders to return to the system and discourage the crimes given opportunity by disorderly conditions. And it worked. At the same time Police Commissioner Ray Kelly was doing the same sort of thing aboveground. Rudolph Giuliani was elected mayor in 1993 and replaced Kelly with Bratton and subsequently replaced Bratton with present police commissioner Howard Safir. Throughout this change of personnel, the basic ideas of community policing continued to be implemented, and then New York City led the nation's big cities in the impressive decrease in crime. William Bratton has gone on to advise other cities about community policing, and with success. Premier Tokyo Sexwale of Gauteng, Johannesburg's province in South Africa, recently announced in a New York talk that he had invited Bratton to Johannesburg to advise his government on sorely needed crime-prevention strategies there.

George Kelling, supported by the Manhattan Institute (which reliably sponsors serious and creative thinking about big-city problems), has written Fixing Broken Windows, with Catherine M. Coles, to tell the story of community policing in New York, and it makes clear the dramatic change in police culture necessary to facilitate the dramatic reductions in crime.15

Community policing demanded a real revision of how police think about and do their work. Especially in big cities, police forces had focused on adversarial "crime fighting" (the kind George Gerbner flinches to see on TV), concentrating on solving serious crimes after they were committed. Police thought of themselves as too busy doing the real dirty work for what they considered "social work" on neighborhood problems. Kelling says this crisis-response style of policing had an especially weak record in dealing with poor, mostly minority neighborhoods and that many patrolmen were out of touch with the neighborhoods they supposedly protected, scared even to get out of their patrol cars. People in the neighborhoods didn't feel much warmth toward the police, either.

Community policing got the patrolmen out of their cars and onto the streets. Foot patrol, Kelling asserts, led to "a broad mandate from those they policed." And in fact, it was the problematic poor neighborhoods which had suffered most from the dangers of disorder--drug business disputes, predatory neighbors, and menacing teenagers with nothing productive to do--and had most to gain from more effective crime control. A better understanding of community relations with the police right now in areas where community policing claims its successes would enable us better to evaluate those claims. Have things changed? Baltimore, Boston, San Francisco, and Seattle have followed New York in pursuing their own campaigns for safer streets, with good results. And police attention to small crimes has paid off for old-style crime solving too: many petty offenders have been found to be involved in more serious crimes, as in the dramatic case of multiple assault and murder resolved in New York last year by the handy arrest of the alleged perpetrator when he jumped a turnstile to avoid paying the $1.50 subway fare.

The true success of community policing depends on the judicious exercise of discretion by police who know their neighborhoods: who is challenged, who is arrested, who is merely warned, when toughness counts and when tenderness works better, are decisions which depend on police flexibility and negotiations with community residents. Kelling says the police leaders who pioneered community policing were concerned about potential abuse of citizens in such cases, particularly the powerless citizens living in the poorest, most dangerous neighborhoods. He gives a thorough review of how problems in this area derailed other attempts at neighborhood crime control. And it is reassuring that complaints about the police did not increase during the NYC Transit Authority's conversion to more aggressive, interventionist policing. However, since much of the narrative evidence of racial bias in the criminal justice system describes situations in the streets and interactions between patrolmen and street and youth culture, this remains an area of worry. Legal controls, outside watchdog agencies, and community awareness all can underscore the message to police that targeting individuals for harassment because of race, age, sex, or other group membership is simply illegal.

Despite a deplorable history, and discounting recent legal strategies of the "war on drugs" which may have been foreseeably racist,16 the criminal justice system in America is not systematically biased against minorities.17 Although this country must squarely face the tragedy of widely disparate white and black crime rates, our criminal justice system is healthy enough to deal with possible abuse of citizens when community policing succumbs to the racism of individuals.

But did community policing cause the reduction in crime? Critics point out that homicide dropped by 33% in one year in Los Angeles although that city has not attempted the change in police culture advocated by Kelling and other supporters of community policing. Crime in Minneapolis is up, despite community policing, and no one knows why. However, a lot of credit for the overall decrease in crime must go to this new style of involved, preventive police intervention. But there is still more to the story.

Although their effects are hard to evaluate, communities all over the U.S. have launched their own crime-prevention efforts. While dissension and sloppy execution mar the effectiveness of many of these, some of these interventions may work. Big Brothers/Big Sisters of America, a highly professional mentoring program which requires solid and sustained commitment from volunteer big siblings, was shown to produce substantial results for 10-16 year-old participants, most of whom would be considered at high risk for developing anti-social behavior, in an eighteen-month evaluation in 1995. Participants were almost one-third less likely to hit someone during the eighteen months of the study period than were the controls, subjects much like them who had not participated in the program. They were 27% less likely to begin using alcohol, and 46% less likely to begin using drugs during that time. Their school grades were unaffected, but they were much less likely to skip school or lie to their parents than control subjects. A steady relationship with a sincerely caring adult seems to work most directly to prevent self-destructive behaviors: because somebody cares.

Other people speculate that residents of inner-city neighborhoods--particularly those who might have fallen into criminality and self-destruction in the past decades--are simply tired of violence and wise to how it harms themselves and others. "Guns and violence are much less socially acceptable than they were a few years ago," Carl Bell, an influential psychiatrist who is chief executive officer of the Chicago Mental Health Council, told Fox Butterfield of the New York Times in June 1997.18 And Salahadeen Betts, a 20-year-old resident of Harlem, corroborated: "People are getting smarter.... It's no more doing things on a whim. Before, people would say, I want to sell drugs, because it was the cool thing to do. Now people are thinking and planning, they are more educated about guns and drugs."

Something is working. This should not be regarded as a blanket endorsement of all the "Stop the Violence" efforts and conflict resolution training programs, some of which can make educators and neighborhood activists complacent about their own interventions and use up resources without doing much good for those about whom we are concerned. (See John Devine's comments about conflict resolution training in violent high schools in "Violence: The Latest Curricular Specialty" in this issue.) However, the successes of intervention and education underscore the message that crime prevention should be a task shared by all of us, regardless of our personal risk of crime victimization.

A distinguished Presidential Commission on the Causes and Prevention of Violence wrote in the introduction to its weighty fourteen-volume report:

We believe that the twin objectives of the social order must be to make violence both unnecessary and unrewarding. To make violence unnecessary, our institutions must be capable of providing justice for all who live under them--of giving all a satisfactory stake in the normal life of the community and the nation. To make violence unrewarding, our institutions must be able to control violence when it occurs, and to do so firmly, fairly, and within the law.19

Wise advice. And sobering to realize that this was the commission empaneled, in a wave of sincere concern about violent crime in America, by President Lyndon Johnson in 1968, the year before our contemporary "epidemic of violence" is said to have begun, the year whose homicide rate we are so proud--if puzzled--to have achieved again.

Karen Colvard is Senior Program Officer at the Harry Frank Guggenheim Foundation.


1. Spelling and punctuation as in original.
2. Quoted in David Rothman, "The Crime of Punishment," New York Review of Books, February 17, 1994.
3. New York Times, February 1, 1997.
4. Philip B. Heymann, quoted in the New York Times, August 1, 1996.
5. New York Times, August 12, 1996.
6. Quoted in Jerome Skolnick, "Making Sense of the Crime Decline," Newsday, February 2, 1997.
7. Julian V. Roberts, "Public Opinion, Crime, and Criminal Justice," in Crime and Justice: A Review of Research, vol. 16. Chicago: University of Chicago Press.
8. Quoted in Scott Stossel, "The Man Who Counts the Killings," Atlantic Monthly, May 1997.
9. The research group which won the competition for these funds, which is directed by accomplished criminologist Alfred Blumstein at Carnegie-Mellon University, can be expected to contribute clear data-based understandings of crime and solid recommendations for policy, and--it is to be hoped--influence for the better the National Science Foundation's priorities for future criminal justice research.
10. Randall Kennedy, Race, Crime, and the Law (New York: Pantheon), p. xii.
11. "Assessing the Penal Harm Movement," Journal of Research in Crime and Delinquency 32 (August 1996): 338-358.
12. He cites F. T. Cullen, J. P. Wright, and B. K. Applegate, "Control in the Community: The Limitations of Reform," in Choosing Correctional Interventions that Work: Defining the Demand and Evaluating the Supply, ed. Alan T. Harland (Newbury Park, CA: Sage), to support his argument.
13. Quoted in "Violent and irrational--and that's just the policy," The Economist, June 8, 1996.
14. "The Police and Neighborhood Safety," Atlantic Monthly, March 1982, 29-38.
15. Fixing Broken Windows. New York: The Free Press, 1996.
16. See Michael Tonry, Malign Neglect: Race, Crime, and Punishment in America (New York: Oxford University Press, 1995) and Randall Kennedy, Race, Crime, and the Law (New York: Pantheon, 1997) for views on this question.
17. See R. J. Sampson and J. L. Lauritsen, "Racial and Ethnic Disparities in Crime and Criminal Justice in the United States," in Ethnicity, Crime, and Immigration, ed. M. Tonry (Chicago: University of Chicago Press, 1997) for a thorough and persuasive review of this evidence.
18. New York Times, June 8, 1997.
19. Introduction to Violence in America: Final Report of the National Commission on the Causes and Prevention of Violence, vol. 1, p. xxii, New York: Chelsea House, 1983.

| back |