Numbers are useful in politics, because they are more neutral than adjectival speech and because they express magnitude—that is, they can tell us not only that we confront a danger but also what the depth and direction of the danger are. The most important numbers in America deal with violence—not the occasional terrorist violence but the terror of everyday life as it is lived by millions of citizens today, and as it threatens to become for many more of us for the rest of this century and beyond.
During his campaign and since, President Bill Clinton has spoken of a sharp decline in the strength of the nation's police forces. In the 1960s the United States as a whole had 3.3 police officers for every violent crime reported per year. In 1993 it had 3.47 violent crimes reported for every police officer. In relation to the amount of violent crime, then, we have less than one tenth the effective police power of thirty years ago; or, in another formulation, each police officer today must deal with 11.45 times as many violent crimes as his predecessor of years gone by.
Title I of the 1994 crime bill intends to add 100,000 police officers nationally by the year 2000. (Most experts believe that far fewer new officers—perhaps 25,000—will actually be hired. For the purposes of this argument, though, let us assume the larger figure.) There are now some 554,000 officers serving on all state and local police forces; 100,000 more would be an increase of 18.4 percent. Rather than having 3.47 times as many violent crimes as police officers, we would have 2.94 times as many; or, each police officer would face not 11.45 times as many violent crimes as his predecessor but 9.7 times as many. All this assumes that the number of violent crimes will not increase over the next several years; if it does, the number of violent crimes relative to police officers will again increase.
If we wished to return to the ratio of police officers to violent crimes which gave many of us peace and security in the 1960s, we would have to add not 100,000 new police officers but about five million. When this number was mentioned to some Department of Justice staffers recently, they giggled; and it is understandable that the idea of such a national mobilization, such tremendous expenditures, should strike them as laughable. However, the American people are already paying out of their own pockets for an additional 1.5 million private police officers, to provide, at least in part, the protection that the public police are unable to furnish.
Private police guard office buildings, shopping malls, apartments. Businesses pay them to patrol certain downtown streets, such as those around New York's Grand Central Station and public library. And they patrol residential areas. Private patrol cars thread the streets of Los Angeles, and more than fifty applications are before the city council to close off streets so as to make those patrols more effective. Across the country much new housing is being built in gated communities, walled off and privately guarded. We are well on the way to having several million police officers, and the next decade will bring us much closer. If current trends continue, however, most of the new officers will be privately paid, available for the protection not of the citizenry as a whole—and certainly not of citizens living in the most violent ghettos and housing projects—but of the commercial and residential enclaves that can afford them. Between these enclaves there will be plenty of room to lose a country.
People hire police officers because they are afraid—above all of violence. Their fear is occasionally a source of puzzlement and mild disdain in the press, which cannot understand why so many Americans say that crime is the nation's most urgent problem and their own greatest fear. Indeed, all through 1993 official agencies claimed that crime was declining. The FBI said that violent crime in the first six months was down three percent overall, and down eight percent in the Northeast.
For crime to be down even eight percent would mean that a precinct that had had a hundred murders in 1992 had ninety-two in 1993. But nobody came around on New Year's Day of 1993 to give everyone's memory a rinse, obliterating the horrors of the previous year. The effect is not disjunctive but cumulative. By the end of 1993, ninety-two additional people had been murdered.
Many people can also remember years before 1992, in large cities and in small. In 1960, for example, six murders, four rapes, and sixteen robberies were reported in New Haven, Connecticut. In 1990 that city, with a population 14 percent smaller, had thirty-one murders, 168 rapes, and 1,784 robberies: robbery increased more than 100 times, or 10,000 percent, over thirty years. In this perspective a one-year decrease of seven percent would seem less than impressive.
New Haven is not unique. In Milwaukee in 1965 there were twenty-seven murders, thirty-three rapes, and 214 robberies, and in 1990, when the city was smaller, there were 165 murders, 598 rapes, and 4,472 robberies: robbery became twenty-one times as frequent in twenty-five years. New York City in 1951 had 244 murders; every year for more than a decade it has had nearly 2,000 murders. We experience the crime wave not as separate moments in time but as one long descending night. A loved one lost echoes in the heart for decades. Every working police officer knows the murder scene: the shocked family and neighbors, too numb yet to grieve; fear and desolation spreading to the street, the workplace, the school, the home, creating an invisible but indelible network of anguish and loss. We have experienced more than 20,000 such scenes every year for more than a decade, and few of them have been truly forgotten.
The memory of a mugging may fade but does not vanish. Nine percent of those responding to a recent poll in New York Newsday said that they had been mugged or assaulted in the past year. This suggests an annual total for the city of more than 600,000 muggings and assaults (remember also that many people, in poor neighborhoods especially, are assaulted more than once). That would be four times as many robberies and assaults as are reported to the police department. The Department of Justice says that not three quarters but only half of all violent crimes go unreported: it may be that many report as having happened "last year" an incident from more than a year ago.
Nevertheless, these are stunning numbers, especially when some other common crimes are added in. Eight percent of those polled (implying 560,000 New Yorkers) said their houses or apartments had been broken into; 22 percent (1,540,000) said their cars had been broken into. In all, 42 percent (nearly three million New Yorkers)said they had been the victims of crime in 1993. And, of course, about 2,000 were murdered. This is what it means to say that crime in 1993 was down eight percent.
In October of 1994 the Bureau of Justice Statistics reported that violent crime had not, after all, declined in 1993 but had risen by 5.6 percent.
Several years ago the Department of Justice estimated that 83 percent of all Americans would be victims of violent crime at least once in their lives. About a quarter would be victims of three or more violent crimes. We are progressing steadily toward the fulfillment of that prediction.
Our greatest fear is of violence from a nameless, faceless stranger. Officials have always reassured citizens by stating that the great majority of murders, at any rate, are committed by a relative or an acquaintance of the victim's; a 1993 Department of Justice report said the figure for 1988 was eight out of ten.
Unfortunately, that report described only murders in which the killer was known to prosecutors and an arrest was made. It did not mention that more and more killers remain unknown and at liberty after a full police investigation; every year the police make arrests in a smaller proportion of murder cases. In our largest cities the police now make arrests in fewer than three out of five murder cases. In other words, two out of every five killers are completely untouched by the law.
When a killing is a family tragedy, or takes place between friends or acquaintances, the police make an arrest virtually every time. When the police make an arrest, they say that the crime has been "cleared"; the percentage of crimes for which they make arrests is referred to as the clearance rate. Because murder has historically been a matter principally among families and friends, the homicide clearance rate in the past was often greater than 95 percent, even in the largest cities. As late as 1965 the national homicide clearance rate was 91 percent. However, as crime has spread and changed its character over the past generation, clearance rates have steadily dropped. In the past two years the national homicide clearance rate averaged 65.5 percent. The rate in the sixty-two largest cities is 60.5 percent. In the very largest cities—those with populations over a million—the rate is 58.3 percent.
The missing killers are almost certainly not family members, friends, or neighbors. Rather, they are overwhelmingly strangers to their victims, and their acts are called "stranger murders." Here is the true arithmetic: The 40 percent of killings in which city police departments are unable to identify and arrest perpetrators must overwhelmingly be counted as stranger murders; let us assume that 90 percent of them are committed by killers unknown to the victims. That number is equivalent to 36 percent of the total of all city murders. We know that of the 60 percent of killers the police do succeed in arresting, 20 percent have murdered strangers. That is, they have committed 12 percent of all murders. As best we can count, then, at least 48 percent of city murders are now being committed by killers who are not relatives or acquaintances of the victims.
This simple arithmetic has been available to the government and its experts for years. However, the first government document to acknowledge these facts was the FBI's annual report on crime in the United States for 1993, which was released last December. The FBI now estimates that 53 percent of all homicides are being committed by strangers. For more than two decades, as homicide clearance rates have plummeted, law enforcement agencies have continued to assure the public that four fifths of all killings are the result of personal passions. Thus were we counseled to fear our loved ones above all, to regard the family hearth as the most dangerous place. Now that falsehood has been unmasked:the FBI tells us that actually 12 percent of all homicides take place within the family. Ihave heard no public official anywhere in the United States say a word about any of this.
There is another important aspect to the arithmetic: the odds facing a robber or holdup man as he decides whether to let his victim live. Again, at least 48 percent of city homicides are stranger murders, but only 12 percent of city homicides result in arrest. That is, the odds that a holdup man who kills a stranger will be arrested appear to be one in four. The Department of Justice tells us that of all those who are arrested for murder, 73 percent will be convicted of some crime; and when convicted, the killers of strangers tend to get the heaviest penalties. Nevertheless, the cumulative chances of getting clean away with the murder of a stranger are greater than 80 percent. Street thugs may be smarter than they are usually given credit for being. They do not consult government reports, but they appear to know the facts. New York bodega workers have experienced an increasing incidence of holdups ending in murder even when they have offered no resistance. Killing eliminates the possibility of witness identification.
Murder is the most frightening crime, but is the least common. Much more frequent are robbery and assault. Robbery, the forcible taking of property from the person of the victim, is the crime most likely to be committed by a stranger; 75 percent of victims are robbed by strangers. Aggravated assault, the use of a weapon or other major force with the intention of causing serious bodily harm, is the most common violent crime; 58 percent of aggravated assaults are committed by strangers.
Attacks across racial lines are a special case of crimes by strangers. Most crimes, including 80 percent of violent crimes, are committed by persons of the same race as their victims. However, the experiences of blacks and whites diverge in some respects. In cases involving a lone offender, 56 percent of white and Hispanic robbery victims report that their assailant was white or Hispanic and 40 percent that he was black. When two or more robbers commit the crime, white and Hispanic victims 38 percent of the time report them to be white or Hispanic, 46 percent of the time black, and 10 percent of the time mixed. About eight percent of black victims, in contrast, are robbed by whites or Hispanics, and more than 85 percent by blacks, whether the offenders are alone or in groups. Blacks and whites are robbed equally—75 percent of the time—by strangers, but as these figures indicate, whites are far more likely to be robbed by strangers of a different race.
This result occurs because there are many more white people and many more white victims: 87 percent of all violent crimes are committed against whites and Hispanics. In robberies lone white offenders select white victims 96 percent of the time, and lone black offenders select white victims 62 percent of the time. White rapists select white victims 97 percent of the time; black rapists select white victims 48 percent of the time. Whites committing aggravated assault attack blacks in three percent of cases; blacks commit about half their assaults against whites.
When all violent crimes are taken together, 58 percent of white victims and 54 percent of black victims report that their assailant was a stranger. Citizens of all races who are fearful of random violence have good reason for their concern. Storekeepers, utility workers, police officers, and ordinary citizens out for a carton of milk or a family dinner are all increasingly at risk.