A ccording to the pundits, the polls, and the politicians, violent crime is now America's number one problem. If the problem were properly defined and the lessons of past efforts were fully absorbed, this could be an opportunity to set national crime policy on a positive course. Instead, it is a dangerous moment. Intuition is driving the country toward desperate and ineffectual responses that will drive up prison costs, divert tax dollars from other vital purposes, and leave the public as insecure and dissatisfied as ever.
The pressures pushing federal and state politicians to vie for the distinction of being toughest on crime do not come only from apprehensive voters and the tabloid press. Some of the leading organs of elite opinion, notably the Wall Street Journal, have celebrated gut-level, impulsive reactions. In one Journal column ("Crime Solution: Lock 'em Up"), Ben J. Wattenberg writes that criminologists don't know what works. What works is what everyone intuitively knows: "A thug in prison cannot shoot your sister." In another Journal column ("The People Want Revenge"), the conservative intellectual Paul Johnson argues that government is failing ordinary people by ignoring their retributive wishes. Ordinary people, he writes, want neither to understand criminals nor to reform them. "They want them punished as severely and cheaply as possible."
Johnson is partly right and mostly wrong. Ordinary people want more than anything to walk the streets safely and to protect their families and their homes. Intuitively, like Wattenberg, many believe that more prisons and longer sentences offer safety along with punishment. But, especially in dealing with crime, intuition isn't always a sound basis for judgment.
The United States already has the highest rate of imprisonment of any major nation. The prisons have expanded enormously in recent years in part because of get-tough measures sending low-level drug offenders to jail. Intuitions were wrong: the available evidence does not suggest that imprisoning those offenders has made the public safer.
The current symbol of the intuitive lock-'em-up response is "three strikes and you're out"--life sentences for criminals convicted of three violent or serious felonies. The catchy slogan appears to have mesmerized politicians from one coast to the other and across party lines. Three-strikes fever began in the fall of 1993 in the wake of the intense media coverage of the abduction and murder of a 12-year-old California girl, Polly Klaas, who was the victim, according to police, of a criminal with a long and violent record. California's Republican Governor Pete Wilson took up the call for three strikes, and on March 7 the California legislature overwhelmingly approved the proposal. Even New York Governor Mario Cuomo endorsed a three strikes measure. The U.S. Senate has passed a crime bill that adopts three strikes as well as a major expansion of the federal role in financing state prisons and stiffening state sentencing policy. In his 1994 State of the Union address, President Clinton singled out the Senate legislation and three strikes for praise.
But will three strikes work? Teenagers and young men in their twenties commit the vast majority of violent offenses. The National Youth Survey, conducted by Colorado criminologist Delbert S. Elliott, found that serious violent offenses (aggravated assault, rape, and robbery involving some injury or weapon) peak at age 17. The rate is half as much at age 24 and declines significantly as offenders mature into their thirties.
If we impose life sentences on serious violent offenders on their third conviction--after they have served two sentences--we will generally do so in the twilight of their criminal careers. Three-strikes laws will eventually fill our prisons with geriatric offenders, whose care will be increasingly expensive when their propensities to commit crime are at the lowest.
Take the case of "Albert," described in the New York Times not long ago by Mimi Silbert, president of the Delancey Street Foundation in San Francisco. At age 10, Albert was the youngest member of a barrio gang. By the time he was sent to San Quentin at the age of 19, he had committed 27 armed robberies and fathered two children. Now 36, he is a plumber and substitute teacher who has for years been crime-free, drug-free, and violence-free. According to Silbert, the Delancey Street program has turned around the lives of more than 10,000 Alberts in the past 23 years.
To imprison the Alberts of the world for life makes sense if the purpose is retribution. But if life imprisonment is supposed to increase public safety, we will be disappointed with the results. To achieve that purpose, we need to focus on preventing violent crimes committed by high-risk youths. That is where the real problem lies.
The best that can be said of some three-strikes proposals is that they would be drawn so narrowly that they would have little effect. The impact depends on which felonies count as strikes. Richard H. Girgenti, director of the New York State Division of Criminal Justice Services, says that the measure supported by Governor Cuomo would affect only 300 people a year and be coupled with the release of nonviolent prisoners. President Clinton is also supporting a version of three strikes that is more narrowly drawn than California's. Proposals like California's, however, will result in incarcerating thousands of convicts into middle and old age.
Regressing to the Mean
Before Governor Wilson signed the most draconian of the three-strikes bill introduced in the legislature, district attorneys across the state assailed the measure, arguing that it would clog courts, cost too much money, and result in disproportionate sentences for nonviolent offenders. So potent is the political crime panic in California that the pleas of the prosecutors were rebuffed.
The prospect in California is ominous. Even without three-strikes legislation, California is already the nation's biggest jailer, with one out of eight American prisoners occupying its cells. During the past 16 years, its prison population has grown 600 percent, while violent crime in the state has increased 40 percent. As Franklin E. Zimring and Gordon Hawkins demonstrate in a recent issue of the British Journal of Criminology, correctional growth in California was "in a class by itself" during the 1980s. The three next largest state prison systems (New York, Texas, and Florida) experienced half the growth of California, and western European systems about a quarter.
To pay for a five-fold increase in the corrections budget since 1980, Californians have had to sacrifice other services. Education especially has suffered. Ten years ago, California devoted 14 percent of its state budget to higher education and 4 percent to prisons. Today it devotes 9 percent to both.
The balance is now expected to shift sharply in favor of prisons. To pay for three strikes, California expects to spend $10.5 billion by the year 2001. The California Department of Corrections has estimated that three strikes will require the state to add 20 more prisons to the existing 28 and the 12 already on the drawing board. By 2001, there will be 109,000 more prisoners behind bars serving life sentences. A total of 275,621 more people are expected to be imprisoned over the next 30 years--the equivalent of building an electric fence around the city of Anaheim. By the year 2027 the cost of housing extra inmates is projected to hit $5.7 billion a year.
But will California be better off in 2027--indeed, will it have less crime--if it has 20 more prisons for aging offenders instead of 20 more college campuses for the young?
Of course, Wilson and other politicians are worrying about the next elections, not the next century. By the time the twice-convicted get out of prison, commit a third major offense, and are convicted and sentenced to life terms, Wilson and the others supporting three strikes will be out--that is, out of office, leaving future generations a legacy of an ineffectual and costly crime policy. To avoid that result, political leaders need to stop trying to out-tough one another and start trying to out-reason each other.
The Limits of Intuition
H.L.A. Hart, the noted legal philosopher, once observed that the Enlightenment made the form and severity of punishment "a matter to be thought about, to be reasoned about, and argued, and not merely a matter to be left to feelings and sentiment." Those aspirations ought still to be our guide.
The current push to enact three strikes proposals is reminiscent of the movement in the 1970s to enact mandatory sentencing laws, another effort to get tough, reduce judicial discretion, and appease the public furies. But mandatory sentencing has not yielded any discernible reduction in crime. Indeed, the result has been mainly to shift discretionary decision-making upstream in the criminal justice system since the laws have continued to allow great latitude in bringing charges and plea bargaining.
Ironically, mandatory sentencing allowed the serial freedom of Richard Allen Davis, the accused murderer of Polly Klaas. Before 1977, California had a system of indeterminate prison sentencing for felony offenders. For such felonies as second-degree murder, robbery, rape, and kidnapping, a convict might receive a sentence of 1 to 25 years, or even one year to life. The objective was to tailor sentences to behavior, to confine the most dangerous convicts longer, and to provide incentives for self-improvement. However, in 1977, declaring that the goal of imprisonment was punishment rather than rehabilitation, the state adopted supposedly tougher mandatory sentences. Richard Allen Davis benefited from two mandated sentence reductions, despite the prescient pre-sentencing report of a county probation officer who warned of Davis's "accelerating potential for violence" after his second major conviction. Under indeterminate sentencing, someone with Davis's personality and criminal history would likely have been imprisoned far longer than the mandated six years for his first set of offenses.
Most criminologists and policy analysts do not support the reliance on expanded prisons and the rigidities of habitual offender laws. Some, like David Rothman, have apologized for their naivet‚ joining the movement to establish determinate sentencing in the 1970s and now recognize that it has been a failure.
Others, like John J. DiIulio, Jr., take a harder line, although the hardness of DiIulio's line seems to depend on his forum. In a January 1994 column appearing-- where else?--in the Wall Street Journal, DiIulio supported the superficially toughest provisions of the Senate crime bill. (The Journal's headline writers called the column "Let 'em Rot," a title that DiIulio later protested, though his own text was scarcely less draconian.) But writing in The American Prospect in the fall of 1990 ("Getting Prisons Straight") and with Anne Morrison Piehl in the fall of 1991 for the Brookings Review, DiIulio's message was more tempered.
The Brookings article reviews the debate over the cost-effectiveness of prisons. Imprisonment costs between $20,000 to $50,000 per prisoner per year. But is that price worth the benefit of limiting the crimes that could have been committed by prisoners if they were on the street? "Based on existing statistical evidence," DiIulio and Piehl, "the relationship between crime rates and imprisonment is ambiguous." This is hardly a mandate for "letting 'em rot." DiIulio and Piehl recognize that the certainty of punishment is more effective than the length and that "even if we find that `prison pays' at the margin, it would not mean that every convicted criminal deserves prison; it would not mean that it is cost effective to imprison every convicted felon." I agree and so do most crimi- nologists. Does DiIulio read DiIulio?
The Rise of Imprisonment
Two trends are responsible for the increase in imprisonment. First, the courts are imposing longer sentences for such nonviolent felonies as larceny, theft, and motor vehicle theft. In 1992 these accounted for 65.9 percent of crime in America, according to the Federal Bureau of Investigation's Uniform Crime Reports.
Second, drugs have become the driving force of crime. More than half of all violent offenders are under the influence of alcohol or drugs (most often alcohol) when they commit their crimes. The National Institute of Justice has shown that in 23 American cities, the percentage of arrested and booked males testing positive for any of ten illegal drugs ranged from a low of 48 percent in Omaha to 79 percent in Philadelphia. The median cities, Fort Lauderdale and Miami, checked out at 62 percent.
There has been an explosion of arrests and convictions and increasingly longer sentences for possessing and selling drugs. A Justice Department study, completed last summer but withheld from the public until February this year, found that of the 90,000 federal prison inmates, one-fifth are low-level drug offenders with no current or prior violence or previous prison time. They are jamming the prisons.
The federal prison population, through mandated and determinate sentences, has tripled in the past decade. Under current policy, it will rise by 50 percent by the century's turn, with drug offenders accounting for 60 percent of the additional prisoners. Three-strikes legislation will doubtless solidify our already singular position as the top jailer of the civilized world.
The Fear Factor
The lock-'em-up approach plays to people's fear of crime, which is rising, while actual crime rates are stabilizing or declining. This is by no means to argue that fear of crime is unjustified. Crime has risen enormously in the United States in the last quarter-century, but it is no more serious in 1994 than it was in 1991. The FBI's crime index declined 4 percent from 1991 to 1992.
In California, a legislative report released in January indicates that the overall crime rate per 100,000 people declined slightly from 1991 to 1992, dropping from 3,503.3 to 3,491.5. Violent crimes--homicide, forcible rape, robbery, and aggravated assault--rose slightly, from 1,079.8 to 1,103.9. Early figures for 1993 show a small decline.
On the other coast, New York City reported a slight decline in homicides, 1,960 in 1993, compared with 1,995 in 1992, and they are clustered in 12 of the city's 75 police districts, places like East New York and the South Bronx. "On the east side of Manhattan," writes Matthew Purdy in the New York Times, "in the neighborhood of United Nations diplomats and quiet streets of exclusive apartments, the gunfire might as well be in a distant city."
So why, when crime rates are flat, has crime become America's number one problem in the polls? Part of the answer is that fear of crime rises with publicity, especially on television. Polly Klaas's murder, the killing of tourists in Florida, the roadside murder of the father of former basketball star Michael Jordan, and the killing of commuters on a Long Island Railroad train sent a scary message to the majority of Americans who do not reside in the inner cities. The message seemed to be that random violence is everywhere and you are no longer safe--not in your suburban home, commuter train, or automobile--and the police and the courts cannot or will not protect you.
A recent and as yet unpublished study by Zimring and Hawkins argues that Ameri- ca's problem is not crime per se but random violence. They compare Los Angeles and Sydney, Australia. Both cities have a population of 3.6 million, and both are multicultural (although Sydney is less so). Crime in Sydney is a serious annoyance but not a major threat. My wife and I, like other tourists, walked through Sydney at night last spring with no fear of being assaulted.
Sydney's crime pattern explains the difference. Its burglary rate is actually 10 percent higher than L.A.'s, and its theft rate is 73 percent of L.A.'s. But its robbery and homicide rates are strikingly lower, with only 12.5 percent of L.A.'s robbery rate and only 7.3 percent of L.A.'s homicide rate.
Americans and Australians don't like any kind of crime, but most auto thefts and many burglaries are annoying rather than terrifying. It is random violent crime, like a shooting in a fast-food restaurant, that is driving fear.
Violent crime, as I suggested earlier, is chiefly the work of young men between the ages of 15 and 24. The magnitude of teenage male involvement in violent crime is frightening. "At the peak age (17)," Delbert Elliott writes, "36 percent of African-American (black) males and 25 percent of non-Hispanic (white) males report one or more serious violent offenses." Nor are young women free of violence. One in five African-American females and one in ten white females report having committed a serious violent offense.
Blacks are more likely than whites to continue their violence into their adult years. Elliott considers this finding to be an important insight into the high arrest and incarceration rates of young adult black males. As teenagers, black and white males are roughly comparable in their disposition to violence. "Yet," Elliott writes, "once involved in a lifestyle that includes serious forms of violence, theft, and substance use, persons from disadvantaged families and neighborhoods find it very difficult to escape. They have fewer opportunities for conventional adult roles, and they are more deeply embedded in and dependent upon the gangs and the illicit economy that flourish in their neighborhoods."
The key to reformation, Elliott argues, is the capacity to make the transition into conventional adult work and family roles. His data show that those who successfully make the change "give up their involvement in violence." Confinement in what will surely be overcrowded prisons can scarcely facilitate that transition, while community-based programs like Delancey Street have proven successful.
Just as violent crime is concentrated among the young, so is drug use. Drug treatment must be a key feature of crime prevention both in prisons and outside. There is some good news here. In early 1994, President Clinton and a half-dozen cabinet members visited a Maryland prison that boasts a model drug-treatment program to announce a national drug strategy that sharply increases spending for drug treatment and rehabilitation. Although the major share of the anti-drug budget, 59 percent, is still allocated to law enforcement, the change is in the right direction. A number of jurisdictions across the country have developed promising court-ordered rehabilitation programs that seem to be succeeding in reducing both drug use and the criminality of drug-using offenders.
Drugs are one area where get-tough policies to disrupt supply have been a signal failure, both internationally and domestically. Interdiction and efforts to suppress drug agriculture and manufacture within such countries as Peru and Columbia have run up against what I have called "the Darwinian Trafficker Dilemma." Such efforts undercut the marginally efficient traffickers, while the fittest--the most efficient, the best organized, the most ruthless, the most corrupting of police and judges--survive. Cocaine prices, the best measure of success or failure, dropped precipitously in the late 1980s. They have recovered somewhat, but likely more from monopolistic pricing than government interference.
Domestically, get-tough intuitions have inspired us to threaten drug kingpins with long prison terms or death. Partly, we wish to punish and incapacitate them, but mostly we wish to deter others from following in their felonious paths. Unfortunately, such policies are undermined by the "Felix Mitchell Dilemma," which I named in honor of the West Coast's once notorious kingpin, who received a life sentence in the 1980s, albeit a short one since he was murdered in federal prison. Mitchell's sentence and early demise did not deter drug sellers in the Bay Area. On the contrary, drug sales continued and, with Mitchell's monopolistic pricing eliminated, competition reduced the price of crack. The main effect of Mitchell's imprisonment was to destabilize the market, lower drug prices, and increase violence as rival gang members challenged each other for market share. Drug-related drive-by shootings, street homicides, and felonious assaults increased.
Recently, two of Mitchell's successors, Timothy Bluitt and Marvin Johnson, were arrested and sent to prison. So will peace finally come to the streets? "When a guy like Bluitt goes down, someone takes his place and gets an even bigger slice of the pie," an anonymous federal agent told the San Francisco Chronicle this past January. "The whole process is about consolidating turf and power."
Youngsters who sell drugs in Oakland, Denver, Detroit, South Central Los Angeles, Atlanta, and New York are part of generations who have learned to see crime as economic opportunity. This does not excuse their behavior, but it does intensify our need to break the cycle of poverty, abuse, and violence that dominates their lives. Prisons do not deter criminals partly because the Mitchells and Bluitts do not rationally calculate choices with the same points of reference that legislators employ. Drug dealers already face the death penalty on the streets.
History reminds us that gang violence is not novel, but it has not always been so lethal. The benchmark sociological study of the urban gang is Frederick Thrasher's research on 1,313 Chicago gangs published in 1927. The disorder and violence of these gangs appalled Thrasher, who observed that they were beyond the ordinary controls of police and other social agencies. He described gang youth, of which only 7.2 percent were "Negro," as "lawless, godless, wild." Why didn't more of them kill each other? They fought with fists and knives, not assault weapons.
Preventing Violent Crime
If violent crime prevention is our strategic aim, we need to test tactics. We need to go beyond the Brady Bill and introduce a tight regulatory system on weapons and ammunition, and we need more research and analysis to figure out what control system would be most effective. Successful gun and ammunition control would do far more to stem the tide of life-threatening violence than expensive prisons with mandated sentences.
The Senate crime bill, however, promises to increase the nation's rate of imprisonment. Besides its three strikes provisions, the legislation incorporates Senator Robert Byrd's $3 billion regional prison proposal. If enacted, states can apply to house their prisoners in 10 regional prisons, each with a capacity of 2,500 inmates.
To qualify, states must adopt "truth in sentencing" laws mandating that offenders convicted of violent crimes serve "at least 85 percent of the sentence ordered," the current average served by federal offenders. They also must approve pretrial detention laws similar to those in the federal system. And the states must ensure that four categories of crime--murder, firearms offenses resulting in death or serious bodily injury, sex offenses broadly defined, and child abuse--are punished as severely as they are under federal law. In effect, the Senate crime bill federalizes sentencing policy.
According to H. Scott Wallace of the National Legal Aid and Defender's Association, the mandate will add about 12,000 prisoners to the average state's correctional population but will offer only about 3 percent of the space needed to house them.
The most costly provision of the Senate crime bill--$9 billion worth--is its proposal for 100,000 more police, a measure endorsed by the administration. Its potential value in reducing crime is unclear. We need more research on constructive policing, including community policing, which can be either an effective approach or merely a fashionable buzzword. We need to address the deficiencies of police culture revealed in the corruption uncovered by New York City's Mollen Commission and the excessive force revealed on the Rodney King beating videotape. More police may help in some places but not much in others. And they are very expensive.
A leading police researcher, David H. Bayley, has explained the ten-for-one rule of police visibility: ten cops must be hired to put one officer on the street. Only about two-thirds of police are uniformed patrol officers. They work three shifts, take vacation and sick leave, and require periodic retraining. Consequently, 100,000 new officers will mean only about 10,000 on the street for any one shift for the entire United States.
Even if we were to have more and better police, there is no guarantee they will deter crime. Criminologists have found no marginal effect on crime rates from putting more cops on the street. Indeed, Congress and the president need look no farther than down their own streets to discover that simply increasing police doesn't necessarily make the streets safe. Washington, D.C., boasts the highest police-per-resident ratio in the nation with one cop for every 150 civilians. It is also America's homicide capital.
We might get more bang for the patrolling buck by investing in para-police, or the police corps, or private police, rather than by paying for more fully sworn and expensive officers. Under the leadership of former Chief Raymond Davis, Santa Ana, California, had the most effective community-oriented policing department in the nation. Davis, who faced a weak police union, could innovate with community- and service-oriented civilians who wore blue uniforms but carried no guns--a new and cost-effective blue line.
The crime bill allocates approximately $3 billion for boot camps, another get-tough favorite. Criminologist Doris MacKenzie has found, contrary to intuition, no significant difference between camp graduates and former prison inmates in the rate at which they return to prison. Similarly, a General Accounting Office report concluded that there is no evidence that boot camps reduce recidivism.
If the public wants boot camps primarily for retribution, it doesn't matter whether they work. Under the Eighth Amendment's bar on cruel and unusual punishment, we're not permitted to impose corporal punishment with whips and clubs. In boot camps, however, we can require painful exercises and hard and demeaning labor to teach these miscreant youth a message of retribution. But if correctional boot camps are intended to resocialize youth and to prepare for them noncriminal civilian life, the camps are inadequate.
We need to experiment with boot camps plus--the "plus" including skills training, education, jobs, community reconstruction. Conservatives who stress moral revitalization and family values as an antidote to youth crime have the right idea. Yet they rarely, if ever, consider how important are the structural underpinnings--education, opportunity, employment, family functioning, community support--for developing such values.
Eventually, we are going to have to choose between our retributive urges and the possibilities of crime prevention. We cannot fool ourselves into thinking they are the same. The punishment meted out by criminal law is a blunt and largely ineffectual instrument of public protection. It deters some, it incapacitates others, and it does send a limited moral message. But if we want primarily to enhance public safety by preventing crime, we need to mistrust our intuitions and adopt strategies and tactics that have been researched, tested, and critically evaluated. In short, we need to embrace the values of the enlightenment over those of the dark ages.
