Title: Perspectives on Crime and Justice: 1998-1999 Lecture Series. Series: Research Forum Author: Mark A.R. Kleiman, Felton Earls, Sissela Bok, and James B. Jacobs Published: NIJ, November 1999 Subject: Crime prevention, white-collar crime, and family strengthening 73 pages 181,000 bytes ------------------------------- Figures, charts, forms, and tables are not included in this ASCII plain-text file. To view this document in its entirety, download the Adobe Acrobat graphic file available from this Web site or order a print copy from NCJRS at 800-851-3420 (877-712-9279 for TTY users). ------------------------------- U.S. Department of Justice Office of Justice Programs National Institute of Justice Perspectives on Crime and Justice: 1998-1999 Lecture Series Research Forum Volume III November 1999 National Institute of Justice ------------------------------- U.S. Department of Justice Office of Justice Programs 810 Seventh Street N.W. Washington, DC 20531 Janet Reno Attorney General Raymond C. Fisher Associate Attorney General Laurie Robinson Assistant Attorney General Noel Brennan Deputy Assistant Attorney General Jeremy Travis Director, National Institute of Justice ------------------------------- Office of Justice Programs World Wide Web Site http://www.ojp.usdoj.gov National Institute of Justice World Wide Web Site http://www.ojp.usdoj.gov/nij ------------------------------- To find out more information about the National Institute of Justice, please contact: National Criminal Justice Reference Service P.O. Box 6000 Rockville, MD 20849-6000 800-851-3420 e-mail: askncjrs@ncjrs.org To obtain an electronic version of this document, access the NIJ Web site (http://www.ojp.usdoj.gov/nij). If you have any questions, call or e-mail NCJRS. ------------------------------- National Institute of Justice Perspectives on Crime and Justice: 1998-1999 Lecture Series Mark A.R. Kleiman Felton Earls Sissela Bok James B. Jacobs November 1999 NCJ 178244 Jeremy Travis Director Amy Solomon Program Monitor The Professional Conference Series of the National Institute of Justice supports a variety of live, researcher-practitioner exchanges, such as conferences, workshops, planning and development meetings, and similar support to the criminal justice field. The Research Forum publication series was designed to share information about this and other forums with a larger audience. Opinions or points of view expressed in this document are those of the authors and do not necessarily reflect the official position of the U.S. Department of Justice. ------------------------------- The National Institute of Justice is a component of the Office of Justice Programs, which also includes the Bureau of Justice Assistance, the Bureau of Justice Statistics, the Office of Juvenile Justice and Delinquency Prevention, and the Office for Victims of Crime. ------------------------------- Preface As a Nation, we are very fortunate to have a robust academic community that thinks, researches, writes, teaches, and talks about the issues of crime and justice. Members of that academic community--including legal scholars, sociologists, economists, philosophers, organizational theorists, anthropologists, historians, psychologists, and a small army of criminologists--look at these issues in profoundly different ways. As a Nation we also are blessed with a national character trait that allows us to be open to many new ideas about the policies that we should adopt as we respond to the challenges of crime and justice. The ideas "in the air" today represent a stunning array of different and sometimes divergent policies and issues: truth in sentencing, restorative justice, waiver of juvenile offenders to adult court, problem-solving policing, zero tolerance, coerced abstinence, community prosecution, community notification, concealed weapons detection technology, right-to-carry statutes, home nurse visitation programs, community policing, hair testing, mandatory arrest, electronic bracelets, assessment centers, and reparative sentencing boards, to name a few. It is understandably hard to keep track of all of these new ideas, and it is harder still to think about designing, funding, and implementing rigorous evaluations of these new initiatives--particularly because so much of today's exciting innovation is happening at the local level and seldom rises to national attention. One way we can take a broader, deeper, and more critical view of our society's responses to crime and justice is to ask the national treasure that is our academic community to provide a policy perspective on the big issues, to help us keep the big picture in focus. To that end, we have asked some of the country's most prominent academics to put aside for the moment their latest research project, step outside the classroom, take a break from the latest book, and come to Washington and share their thoughts with a policymaking audience. Because we videotape and publish these lectures, we also intend to reach a national and, indeed, international audience as well. This lecture series was established 3 years ago to challenge conventional wisdom, stimulate lively discussion, and bring new perspectives to the policy debates that occur in our capital city and in other communities around the country on the challenges of crime and justice. Our speakers this year have fulfilled these goals. Mark A.R. Kleiman provoked us to reexamine theories of deterrence and the role of probation, parole, and police agencies in carrying out our deterrence policies. Felton Earls reminded us of the importance of fathers in our society and the paucity of the understanding of the functions of fathers. Sissela Bok addressed the timely and troubling topic of media violence and painted a picture of a society that sacrifices its social bonds through exposure to violence in the media. Finally, James B. Jacobs tackled a topic that lies at the core of our democratic institutions: the ability of government to resist the taint of corruption. He asked the difficult question, what price do we pay for our efforts to control corruption? We are pleased to present the lectures and commentaries in this volume and look forward to similarly lively presentations and discussions in the year to come. Jeremy Travis Director National Institute of Justice ------------------------------- Contents Preface Lecture 1: Getting Deterrence Right: Applying Tipping Models and Behavioral Economics to the Problems of Crime Control Mark A.R. Kleiman Lecture 2: Men and Fathers in the Community Felton Earls, M.D. Lecture 3: Violence, Free Speech, and the Media Sissela Bok, Ph.D. Lecture 4: Dilemmas of Corruption Control James B. Jacobs, J.D., Ph.D. ------------------------------- Getting Deterrence Right: Applying Tipping Models and Behavioral Economics to the Problems of Crime Control Presentation by Mark A.R. Kleiman Professor of Policy Studies University of California at Los Angeles December 3, 1998 Washington, D.C. My theme is crime control and how to do it better. I hope to convey the excitement I feel: we are, I believe, in the process of learning how to control crime much more effectively and with much less attendant suffering than has been within our reach up until now. Targeted Zero Tolerance It is well known that New York's subways and streets are much more civil and orderly now than they were 5 years ago and that serious crime there has fallen dramatically. Those efforts were inspired by, and seem to bear out, the "broken windows theory" proposed by James Q. Wilson and George L. Kelling, which forms much of the conceptual basis for "community policing."[1] But how is it that reducing graffiti reduces mugging? I have a story to tell about that, and a story that justifies the choice made by Bill Bratton in New York to employ what might be called "focused" or "targeted" zero tolerance rather than the usual meaning of zero tolerance: a policy of making arrests for every observed violation of the law. It is less well known that in Boston and Minneapolis, rates of youth homicide have fallen spectacularly as a result of a coordinated enforcement intervention targeted at gangs of violent juveniles--an intervention aimed, not at breaking them up, but at getting them to stop killing one another. The Boston pattern is consistent with falling homicide rates in the northeastern cities that were hit early by the crack epidemic, though the decline is still pretty dramatic. (See exhibit 1.) But Minneapolis is one of the cities in the Upper Midwest where crack arrived late and homicide rates have been soaring. There is no external reason why the Minneapolis homicide rate should have declined so much. (See exhibit 2.) How did these programs work? The details varied, but the concepts were identical (they were both invented by Harvard University's David Kennedy).[2] When its project started, Boston had an estimated 65 youth gangs with 1,300 active members. Deadly violence among young people was highly localized in gang-ridden neighborhoods and occurred largely along stable intergang dispute lines. Given the atmosphere, many participants in gang activity, and others around them, became afraid that not showing a willingness to respond to personal affront with deadly violence would be taken as a sign of weakness, and make them targets for violence by others. Investigators knew which gangs were shooting, but for all the usual reasons cases were difficult to make, and in any event would involve only the individual shooter. Shooting increased gang prestige and thus the prestige of the shooter within the gang. The Operation Cease-Fire team assembled police (gang, juvenile, drug), probation officers (adult and juvenile), prosecutors (State and Federal), DEA, ATF, State police, truant officers, and anyone else with potential authority over the young people involved. The gangs were brought in, in units, and given the following message: We know who you are. We know what you are doing, and most of it is illegal, from playing truant to selling crack. We cannot clamp down on everyone at once, but we can clamp down on any group we choose to. If we crack down on your group, you cannot live your (mostly illegal) life. There are new rules in force. The rules are: no deadly violence. The consequence of breaking the rules is concentrated enforcement attention on the whole group responsible until the whole group says "uncle" and turns in its guns. When it became necessary, the task force conducted a crackdown, publicized it, and protected members of the target gang after they gave up their guns. Using undercover drug buys, the task force then conducted an old-style attack on the most violent group, got them all sent away for long sentences, and publicized that. The results were spectacular and lasting. Starting from these two examples, I would like to explore the logic of the new crime control and suggest how employing it more consistently could greatly abate what remains, in my view, a society-threatening social problem and the single greatest barrier to improving economic opportunity in urban neighborhoods of concentrated poverty. Many things control crime--there is not much doubt that early childhood interventions, if we can learn to do them right, could have a big impact. But today I want to concentrate on the operations of what is loosely called the criminal justice system: police, courts, and corrections agencies. Rethinking Incapacitation The classic theory of crime control talks about three crime-control effects of enforcement and punishment: incapacitation, deterrence, and rehabilitation. Incapacitation is the easiest to understand: If we lock up people who commit more than their share of crimes, the population remaining free will include fewer high-rate offenders. Since prison cells are expensive, incapacitation makes sense only if we can lock up people who, on average, commit lots of crimes. There is a lot of evidence that we in fact do so, that the average offender in prison would be causing damage worth much more than $25,000 per year if he were running around free. But that is true only because a relatively small number--about 10 percent--of those behind bars commit astounding numbers of offenses.[3] The famous RAND Inmate Survey revealed that the median prisoner had committed 17 crimes in the year before incarceration, but the average offender had committed almost 200.[4] Offending is widespread, but persistent, serious offending is highly concentrated. As in so many cases, almost everything depends on the right-hand tail of the distribution. This is consistent with a famous generalization called "Pareto's Law": 20 percent of any population accounts for 80 percent of any activity. Pareto's Law holds true for more things than you might expect. The survey findings led Peter Greenwood of RAND, among others, to suggest that we should rethink sentencing policy to concentrate on high-rate offenders: a policy he called "selective incapacitation." Selective incapacitation turns out to be harder than it looks: an offender's criminal history (rap sheet) is a pretty poor indicator of his actual behavior, since most crimes never lead to an arrest.[5] Moreover, since criminals, like basketball players, tend to leave the game in their early 30s, basing sentences on the number of felony convictions--as in "three strikes"--tends to fill the prisons with people who would otherwise be retired muggers. Deterrence: Understanding Offenders' Decisionmaking Incapacitation is a pretty mechanical process. Deterrence, by contrast, depends on changing the behavior of potential offenders by changing their perception of the likely consequences of offending. The key to deterrent crime control lies in understanding the decision processes of (actual and potential) offenders. Since, as noted, serious, persistent offending is highly concentrated, it is the behavior of serious, persistent offenders--and of those who might become such offenders--that we most need to understand. There are two extreme competitors here: models of purely rational choice--offenders comparing the benefits of offending to its costs--and models of utterly irrational social determination. I want to offer a middle way: a model that assumes that offenders seek their own well-being but that also pays attention to factors that might cause them to deviate from normative rationality as economists understand it. If offenders tend to lapse from rationality in specific ways, and if we continue to pass laws and make policies as if offenders were perfectly rational, we will wind up inflicting more punishment than we need to for less crime control benefit than we could have. That is the "behavioral economics" of the title: offenders are more likely than most people to be both reckless, in a way described by "prospect theory," and irrationally present oriented. The argument is simple: people who calculate rationally and who react strongly to small probabilities of horrible consequences some time in the distant future do not become persistent offenders. James Q. Wilson and Richard J. Herrnstein have reviewed the data showing that persistent offenders tend to be, in psychological terms, reckless, impulsive, and present oriented.[6] To control the behavior of such people, we need consistent and immediate sanctions, even if the sanctions themselves are relatively mild. What we have now is a system of severe sanctions applied very sporadically and a long time after the offense. Calculating the Expected Value of Punishment Now that is not to say that deterrence always fails; in gross terms and for most people, it works pretty well. The simplest way to think about a deterrent threat when it is only occasionally carried out is in terms of what economists call "expected value." An expected value is a probability-weighted average. (Remember the typical American family of the 1950s with 2.7 kids? That is an expected value: all the kids divided by all the families was 2.7.) If 9 times out of 10 a mugging results in no punishment (because the mugger never gets caught), but the 10th time it results in 5 years in prison, then the expected-value punishment is 5 years times 1 chance in 10, or half a year. By looking at the total number of crimes of a given type and the total number of people in prison for that crime, we can easily calculate the expected-value punishment. Exhibit 3 is a graph of the expected value of punishment for burglary over time. The punishment rate fell sharply during the early 1960s due in part to the closing of big, old prisons in many States, and the rate of burglary started to climb explosively, which of course then pushed the punishment per crime down even further. The explosion of prison building in the mid-1970s was closely followed by a peak, and then a steady decline, in burglary. This is not nearly enough to show cause and effect, but it is about what a simple deterrence model would predict. On the other hand, the expected punishment price of a burglary never went below about 5 days, and it is now nearly 20 days. That does not seem like much of a penalty for breaking into your home, but compared to the average "take" from a burglary, it is really quite a lot. Given the low prices fences pay for stolen goods, a typical residential burglary nets the criminal perhaps $200. Even at 5 days per burglary, that amounted to $40 per day served behind bars. At today's higher punishment price, the effective take is closer to $10 per day served. Muggers get an even worse deal. That illustrates an important general principle of criminology: "The wages of sin are well below the legal minimum." Any purely rational-actor theory of crime would have a hard time explaining why people continue to commit such unrewarding misdeeds. The illicit-drug markets, where the reward-to-punishment ratio tends to be more favorable, still do not behave as economic rational-actor theories would predict. I have to plead guilty here: A dozen years ago, Peter Reuter and I came up with a model called "risks and prices," which asserted that drug prices ought to reflect punishment risks. Back then, the problem with the model was that the price of cocaine seemed unsustainably high compared with the punishment risk. Now, the puzzle is just the opposite. Even compared to the growth in cocaine shipments, the growth in cocaine punishment has been astounding. As a result, the punishment price per ton of cocaine has soared. Yet cocaine prices, rather than rising, have continued to fall. (See exhibit 4.) Something is wrong here. Of course, no offender faces the expected value of punishment, any more than any family has 2.7 kids. So one problem may be that randomized threats do not work as well as reliable threats, even if the long-term average is the same. But the standard theory of rational behavior predicts just the opposite. We expect rational people to be risk-averse; that is why they buy insurance policies. A curve representing how much someone's well-being rises as that person's income or wealth grows--what economists call a "utility curve"--turns out to rise less and less steeply: $100 means less to Bill Gates than to someone living on a minimum-wage job. (See exhibit 5.) If people's behavior matched this graph, the small threat of a big punishment would work wonders. This has led economists modeling principal/agent relationships to suggest controlling agent behavior by an infinitesimally small risk of an enormous punishment. It might work for corporate executives; I doubt it works as well for burglars. There is a competing theory, not of rational action--I do not think there is any doubt that rationality implies risk aversion--but of actual behavior. It is called "prospect theory," and is credited to Amos Tversky and Daniel Kahneman.[7] According to the theory, in many situations actual behavior is risk averse in gains but risk seeking in losses. Given a choice between 1 week in Hawaii for sure and an even chance at 2 weeks in Hawaii, most people will choose the sure thing, just as risk-aversion theory would suggest. But once we start dealing in losses from the current position--if the gamble is between 1 week in jail and an even chance at 2 weeks in jail--many people will act as if they are risk seeking and take the gamble rather than settling for a sure loss. So if offenders tend to behave according to prospect theory, an occasional big punishment will not deter them as well as reliable small punishments with the same expected value. That is part of the story about why the current prison-building boom has not reduced crime as much as we might have expected: we are overrelying on severity and underplaying certainty. This does not mean there are no benefits from building prisons: they do a fine job at deterring those who are risk averse, and they lock up a lot of people who would otherwise be out committing a lot of crimes. But it does suggest that the returns from further increasing severity will be diminishing. Estimating Present Versus Future Gains and Losses There is a second possible reason why persistent offenders persist in the face of high expected-value punishments. Punishments arrive after crimes, in many cases months after. By definition, the final 25 years of a 30-year sentence start 5 years from now. Even perfectly rational people value future gains or losses less than the same-sized losses arriving right away. After all, if you have money now, you can put it in the bank (or pay off your mortgage) and either way earn, or avoid paying, interest. Economists have a well-worked-out theory of how a rational person would trade off present versus future gains and losses. It goes under the name of discounting. The formula is one of geometric decay. It has the property of depending only on the distance in time between the earlier and later benefit or cost, and not at all on when the decision is made. Someone choosing between receiving $100 1 year from now and $110 2 years from now should make the same decision, whether the decision is to be made now or a year from now. The decision depends only on that person's personal discount rate, which in turn depends on the rest of that person's financial picture, particularly the rates at which that person can borrow or lend. The result is that the future is devalued compared to the present, but only at the rate of a few percentages per year. Again, there is an alternative theory, not of rational action but of actual behavior. It is well verified in the psychology and behavioral-economics laboratories, on species ranging from pigeons to sophomores. It has a different, hyperbolic, formula, which leads to a much more rapid fall-off over time. It also depends crucially on when the decision is made.[8] Imagine offering a 5-year-old one marshmallow now or two marshmallows tomorrow. Only a pretty tough 5-year-old would be able to wait. Now offer the same child a marshmallow in a week or two marshmallows in 8 days. It is the same choice: wait a day and double your marshmallow. But the result is likely to be quite different. Anyone who has ever stayed up all night preparing a paper, a proposal, a presentation, or a lecture has demonstrated hyperbolic discounting, or "temporal myopia" as it is known in the literature (though it has been suggested that "temporal false perspective" would be a more accurate label). If offenders are even more temporally myopic than the rest of us, the difference between a punishment now and a punishment 6 months from now may be, in Mark Twain's phrase, "the difference between the lightning and the lightning bug." One of the worst offenders caught in the Boston crackdown remained stone-faced through his arraignment on a charge under which he faced a minimum of 15 years in prison. But he went ballistic when the judge, in granting him bail, imposed a 6 p.m. curfew. Why? Because the sentence was some indefinite time in the future; the curfew was that night, and that night he had things to do. Misinformation About Punishment Risk A third theory is that deterrence works imperfectly because offenders misestimate deterrent threats. Offenders frequently find themselves quite ignorant of the actual threats they face, especially under obscure punishment-enhancement laws. That suggests it is important to communicate the threat. Jeremy Bentham wrote that the two most important characteristics of a deterrent punishment were certainty and celerity.[9] Prospect theory and temporal myopia suggest why. To Bentham's two C's I would add a third: communication. Communication, in this context, is not merely verbal. Actions also communicate, proverbially better than words. Offenders, like other people but probably more so, are likely to believe their own experience more than what they read or what others tell them. In judging the likely consequences of lawbreaking, they will refer largely to what has happened to them. For most offenses, no one is even arrested. Most arrests do not lead to convictions. And most convictions, especially for the minor, nonviolent offenses by juveniles that form the beginning of most criminal careers, do not lead to any real sanction. The insistence on very long sentences for identified persistent, serious offenders, combined with budget realities, means that the typical punishment for the first few detected offenses is probation: that is, under current conditions, nothing. That sends a very bad signal about the consequences of lawbreaking. It makes sense only if our goal is to develop career criminals. How is an offender supposed to judge which "last chance to go straight" is really his last? He is likely to keep testing the system until it lands on him hard, which is bad news for him and us alike. So a first implication of this line of thinking--which says that offenders tend to be reckless, short sighted, and ill informed rather than cool calculators of likely consequences--is that every detected nontrivial violation of law ought to lead to some nontrivial deprivation of liberty. That cannot mean prison or jail--we do not have enough cells--so by definition it must mean probation. The trick is making probation mean something. That is essential for another reason as well. Since serious, persistent offenders are eventually caught, most of them are (or should be) on probation or parole if not actually behind bars. This gives us the capacity to supervise their behavior in much more detail than is the case for those whose liberty is unconditional. But the current probation and parole systems rely on the same sort of low-probability, deferred, but occasionally severe sanctions used by the rest of the system. Community Corrections Community corrections is the potential key to crime control. To make that potential real requires that community corrections officers have three things: --The technology to monitor behavior that occurs outside their presence. --The authority to impose rapid and predictable sanctions. --The resources (budget and staff) to acquire and use the technology and administer the authority. For example, if every cocaine-using burglar on probation faced two drug tests a week, with 2 days' confinement immediately for every missed or "dirty" test, we could eliminate about 40 percent of the total market for cocaine and substantially reduce the amount of time that population spends behind bars. The drug test is a way to monitor, in 5 minutes and for $5, one slice of the offender's behavior over the past 72 hours.[10] Another potentially important technology of community corrections is electronic position monitoring. Day reporting, and the creation of data links to other institutions that come in contact with probationers, can also increase the effective capacity of a probation officer to observe probationer behavior. The point here is not the technical details, but the principle of establishing behavioral standards and enforcing them with swift and consistent sanctions. Giving the community corrections system that new technology and authority will require more resources: a "caseload" of 300 implies that very little effective supervision can occur. To make this work, we should be prepared to reallocate money away from prisons and jails--which with about one-fifth of the total corrections population account for nearly 90 percent of corrections spending--and toward probation and parole. But just adding more bodies without adding technology and authority contributes little to crime control. Current probation supervision in Los Angeles, for example, is effectively nothing, and twice, or even four times, nothing is still nothing. Pouring resources into a fundamentally misconceived process just increases its futility. In short, we need to apply to crime control the common-sense principles that apply to child raising and puppy training. Enforcement "Swamping" So much for the behavioral economics of crime control. Now for the "tipping" part. Even for perfectly rational offenders, the rational amount of offending by any one person depends in part on how many other people are committing the same offense in the same area. Thus, decisions to offend are interdependent decisions. If punishment capacity is scarce, punishment-per-crime tends to fall as the rate of offending rises, as happened during the 1960s. Therefore, some people who would not offend when offense rates are low will offend if offense rates are high, because having lots of "competition" for enforcement attention reduces their individual chances of being caught and punished. That is the logical structure of a riot. It is described by a class of models invented by Thomas Schelling and known by his label of "tipping" models because of the way that results "tip" from one extreme to another after some threshold is crossed.[11] I call the version of tipping applicable to crime "enforcement swamping": once violations are high enough to swamp the capacity of the enforcement system, increases in offense rates will feed on themselves.[12] Imagine a parking lot designated "Shoppers Only: No Commuter Parking." It has 1,000 parking spots. Parking there is free for shoppers; commuters, if caught, face a $15 fine. A single attendant is in charge of locating and ticketing commuter vehicles; he has the capacity to write up to 100 tickets per day. A nearby lot offers all-day parking for $7.50. If all commuters who park illegally are ticketed, they would be better off paying for legal parking. On the other hand, if fewer than half are ticketed, then it is actually cheaper (on an expected-value basis) to park in the free lot and take your chances. So the least-cost solution to the parking problem from an individual commuter's viewpoint depends on how many of the other commuters decide to take their chances in the free lot. If more than 200 commuters risk it, they will all save money on average. Now consider how the situation might develop over time. If very few commuters park illegally, those who do will quickly learn that it is more expensive that way. That will drive the number of illegally parked cars toward zero. On the other hand, if many commuters park illegally, that will turn out to be the cheaper option, and the number will tend to grow. Eventually, the entire lot may fill with commuters, leaving no spaces for shoppers; that too is a stable, self-reinforcing condition. What could the parking authorities do once the high-violation equilibrium had established itself? One option is a temporary boost in ticket-writing capacity, on the model of a standard police "crackdown." In the example, if there is enough capacity to write more than 500 tickets per day, parking illegally will never be the least-average-cost option, no matter how many commuters try it, because even with 1,000 violators the probability of a ticket will be more than one-half. A sanction probability of 0.5 is the "crossover" or "tipping" point of the system--values below that tend toward zero, while values above that tend toward one. Once the burst of enforcement starts to drive down the number of violations, enforcement capacity can be reduced. Once the number of violators is safely below 200 per day, even a single ticket writer will be enough to keep legal parking the cheaper option. An alternative strategy would be to rely on a self-fulfilling bluff. Put up big signs that read, "As of Monday, all illegally parked vehicles will be ticketed." If enough people believe the signs, the threat will be carried out. If it works, the bluff alone will be sufficient. Or the threat could be added to an actual boost in enforcement as a way of speeding the transition from the high-violation to the low-violation stable state. Creeping Zero Tolerance Even if additional enforcement capacity is unavailable, all is not lost. If one-tenth of the lot is designated a "zero tolerance" area where all illegally parked vehicles will be ticketed, that threat can be carried out even with the original level of resources and 100-percent violations. The result will be that commuters learn not to park in the "zero tolerance" spaces. As the number of violators in that area shrinks toward zero, the zone of total enforcement can be gradually expanded without exhausting enforcement capacity. Eventually, unless the commuters can organize themselves for a campaign of massive resistance, the entire lot can be moved from a high-violation stable state to a low-violation stable state, without ever expanding the number of ticket writers. (Violators could be targeted on any other characteristic that does not include more than 10 percent of them in any one category: last digit of the license plate number, for example.) This creeping zero-tolerance model has four important implications: First, a modest increase in enforcement after an increase in offending may have little effect, especially if the increase is temporary or sporadic. The task is to change the entire climate of expectations that can make high offense rates self sustaining. That is likely to require large and sustained, rather than small and temporary, enforcement interventions. Second, since expectations matter, communicating the message that enforcement has increased is essential to success even if the offenders as individuals are perfectly rational, and even more important if some of them are not. (As noted, a sign that reads, "All misparked cars will be ticketed" will be true if enough people believe it.) This sort of advertising also will increase the value of any increase in actual enforcement. David Kennedy has proposed an innovative, communication-based approach to breaking up open retail drug markets. His approach seems likely to work at much lower cost than the standard long-term, concentrated crackdown typified by the "Pressure Point" operation that broke up the heroin bazaar on Manhattan's Lower East Side in the early 1980s, and perhaps even better than the more complex market disruption strategies pioneered by Tampa's QUAD.[13] Assume there is a big open market in a neighborhood called Vial City, and that we have decided to break up the market on D-Day, 30 days from now. Start now by leafletting the area with fliers reading, "We have decided to shut this market down as of [D-Day]. Do not buy or sell drugs here after that date, or we promise you will be arrested." Then have the police approach known buyers and sellers personally, conveying the same message. Note the license plate numbers of nonlocal buyers, and send postcards to the registered owners saying, "As of [D-Day], the drug bazaar in Vial City will be closed. As a reminder, automobiles found with drugs in them are subject to seizure and forfeiture." At D minus 14 days, start visiting the mothers and girlfriends of the known sellers. Tell them, "Ms. Smith, please make sure Johnny is not selling in Vial City on [D-Day]. If he does, he is sure to get busted." Keep up the leafletting and make another round of personal contacts starting at D minus 7 days. Put up a billboard if the budget will support it. On D-Day, send out some additional uniformed and undercover police officers, with instructions not to "sweep the streets," but merely to make arrests if they observe dealing activity. If the publicity has worked, they will be profoundly bored. Only a few very sleepy or skeptical buyers and sellers will show up, and as soon as the arrests start they are likely to make themselves scarce. The next day, the market will be as dead as vaudeville, and the courts will be less, rather than more, burdened with Vial City crack-distribution cases than before the crackdown. This is not a solution to the drug abuse problem; most (though not all) buyers will find other sources of supply once Vial City is closed. But it is a solution to the problem the market has imposed on Vial City's residents. (This is more than a thought-experiment; someone ought to actually try it.) Third, equal opportunity enforcement is sometimes the wrong rule. If you cannot get past the crossover point for everybody, get past it for somebody and work outward. Anything that creates a subset of offenders or offenses for whom enforcement is not swamped will eventually clean that subset of offending; then you slowly expand the "clean" subset until it includes everything. That is how the New York City subway system solved its graffiti problem. It designated--and clearly marked--10 percent of its cars as "clean cars" and committed that they would never roll until they were clean of graffiti. At that point, spraying those cars became unrewarding because no one would see the tags. At only 10 percent, the system could afford the commitment; if a few cars had to stand idle, that was no disaster. The Transit Authority then slowly expanded the "clean" fleet until it included all the cars. Another form of concentrated enforcement is to concentrate it, or threaten to concentrate it, on the most egregious offenders. That sets up a competition among offenders (or groups) to avoid enforcement attention. That is part of the story of Operation Cease-Fire in Boston. Fourth, anything that signals high offense rates will therefore serve as an invitation to offend; anything that signals low offense rates will serve as a deterrent to offending. That is one reason "broken windows" strategies can work. The analysis above suggests some new principles of crime control: --Ensure certainty and celerity before severity. --Communicate threats: Do not just strike the rock; talk to the rock. --Impose some nontrivial punishment for each conviction. --Add technology, authority, and resources to make community corrections effective punishment. If we cannot learn to punish people and control their behavior when we are not paying for their room and board, we are out of luck. --Target zero tolerance for specific "broken-windows" behaviors keyed to specific neighborhoods, to try to move them to low-offending equilibria. --Target zero tolerance for any offending by individuals and groups whose serious offenses earn them special attention. --Make offenders compete to avoid enforcement attention. Crime will always be with us. But the crime wave that swept the United States starting about 1960 and Western Europe starting about 10 years later does not have to be the wave of the future. We can do better. We should. Question-and-Answer Session Christine Mayman, Martinsburg Police Department, West Virginia: I agree with almost everything you said. However, in this time of inundation of new theories and new ideas, how do you convince policymakers, who are inundated with lots of varying recommendations over the years, that this brand of policy change is one they want to do, especially when it means not only changing what they do, but changing how they view things? M.K.: There are various kinds of academics who pay attention to crime. I am a policy analyst by trade; my job is trying to figure out what a sensible policy should be. The politics of crime is not my specialty. That said, policymakers are under the pressure of popular discontent in dealing with crime. One of the problems with adding resources to probation and parole is that people can (and have) run for reelection as Governor with an elective platform saying, "I built x number of prisons in my tenure; more than all the other Governors in the history of the State." Substitute "hired probation officers" for "built prisons" and you have a much less exciting campaign pitch. Corrections is a politically vulnerable area. The thing that is coming to bear now on the other side is that, in State after State, Governors are discovering that they can have bigger prison systems or a working university system, but not both. When I started thinking about these matters, building prisons seemed to be thoroughly justified. By calculation the other day, I saw that if you look at the Federal Bureau of Prisons budget and calculated the fraction of that that is for drug offenders, it turns out that we pay more money every year to keep last year's drug dealers in prison than we give to DEA to catch next year's drug dealers. That is probably a bad allocation of funds. Gary Albrecht, U.S. Capitol Police, Washington, D.C.: I absolutely agree that certainty and swiftness are critical to deterrence. I am very skeptical, however, that probation or parole can have any deterrent effect whatsoever. I've never run into a criminal who was very afraid of probation. I would suggest that pretrial incarceration, "disappearing immediately" the day they get arrested, has much more deterrent effect, both on the individual and on his colleagues who observe what happens. M.K.: Both halves of that comment are truly important. First, what you say is right about probation; it is known as "taking a walk." It is accepted as zero punishment. We have to change that. We simply do not have enough cells to hold everybody who needs some punishment, so we have to learn how to make probation real punishment. Probationers need to know that their behavior will be closely monitored in various ways and that any crossing of a clear set of lines is going to mean consequences. The folks conducting drug testing on probationers should say to their clients, "If you think you're going to be dirty, bring your toothbrush." I agree that probation as it stands is not real punishment. But if we do not figure out a way to make it real punishment, then we simply cannot get the job done. Someone looking at my day-per-offense analysis might have said that the President's program to add 100,000 police officers was not going to do anything very useful because we're already arresting a lot more people than we can lock up. But of course, arrest is also very unpleasant, and it happens right now. The fact is that a large fraction of the effective deterrence we are handing out is from arrest. But if it turns out arrest has no consequence past the arrest, then it starts to lose effect. There are some fascinating research results if you look at juveniles, most of whom, sometime in their growing up, do something against the law. Some get caught; some do not. If you look at juveniles in terms of their self-reported misbehavior, the ones who got caught go on to much more serious criminal careers. It looks as if being arrested causes more crime in the future. Some criminologists explain this in terms of "labeling," but there is a simpler, more parsimonious theory: Once you get arrested as a juvenile and see how little happens to you after you get arrested, then you have been reinforced in your offending. The kid who breaks a law and does not get caught still thinks that if he broke the law and got caught, it would be bad. The one who got caught and knows that nothing bad resulted may go on to be a serious offender. We need to look at what can really be done to juveniles that they really do not like (and that we are willing to do), remembering that two-thirds of the serious juvenile offenders (having done things that, if done by an adult, would have been felonies) never get arrested for any adult serious crime. Moving the punishment closer to the crime is essential, whether it is pretrial detention or a kind of pretrial release that is a serious means of control. Christian Warren, Public Defender Service, Washington, D.C.: I really was struck by your point that three-quarters of the population under justice supervision is on probation and parole, but that we spend 90 percent on prisons. I think that is where we have to start because most of the time my clients on probation will tell me things like, "Well I went to community service but nobody was there. I reported and then went home." The next thing they say is "I didn't do it." I don't think there are mechanisms in place that can change values. The other thing you said, that there was not a message being sent to offenders in terms of what the price is: I would really disagree with that. Most of my clients have fathers, uncles, brothers, who are doing 25 or 20 years in Lorton [correctional facility in Virginia] locked up. I do not know how much closer they could be to it. I don't know what the answer is, but most people who are multiple repeat offenders have relatives supervised by the criminal justice system. M.K.: I think people who get to be serious offenders take more seriously what is happening to themselves than what they hear has happened to somebody else. But you are right; it is a difficult question to see what they think the risk really is. Insofar as they think it is going to happen to them at some indefinite time in the future, that may not be powerful for them. You mentioned community service; that is a term I hate. "Community service" is joining the parent-teachers association. We are really talking about punitive labor. It can be a very effective nonincarcerative punishment if we learn how to do it. The key is getting community organizations to start developing projects for which they need labor and then make the deal that they will contribute the supervisory work in return for the labor. If we can do that, we can have an expandable capacity for serious but nonincarcerative punishment. Such programs can also be truly restorative if the beneficiary of the labor is the community that sustained the original injury. Marie Ragghianti, U.S. Parole Commission, Washington, D.C.: I'm wondering why you haven't mentioned graduated sanctions. It seems that the concept of graduated sanctions is one way to deliver the kind of sanction that you are talking about with both swiftness and certainty. It is also a way of making probation and parole effective. M.K.: I agree entirely. I think that we need graduated sanctions programs for probation and parole. And I think they ought to be formulaic, not discretionary, to put all the control in the hands of the offender. One of the things that Adele Harrell reported in her evaluation of the D.C. Drug Court's sanctions track is that it changed offenders' subjective locus of control. Usually, offenders talk about what happened to them in terms of somebody else's agency: "The judge did this to me, the cop did that, the probation officer did the other . . . ." This is what psychologists call having an "external locus of control." It is reinforced by discretion. Everybody in the system wants discretion, but discretion means to the offender that he has to accept everybody else's choices. In her report from the D.C. Drug Court, Dr. Harrell found that offenders started to change and say that they got sanctioned "because they used drugs." There is a California Drug Court judge who makes a little speech to people entering his program, "From now on, I have no control over what happens to you. The prosecutor, the probation officer, the defense lawyer--none of them has any control over what happens to you. You control what happens to you." What we need is formulaic sanctions, whether graduated or not. That means convincing people, particularly probation officers and judges, that their job is not clinical. One of the conclusions coming out of the D.C. Drug Court experience was that "changing addict behavior is easy; changing judges' behavior is hard." George Kanuck, Center for Substance Abuse Treatment, U.S. Department of Health and Human Services, Washington, D.C.: More than just drug testing, what is the role of meaningful drug treatment for these people? This is a continuum problem. We have to do something more than just test them. M.K.: That is a question I hope CSAT will research. Imagine a seriously drug-involved offender. There are a number of things we could do. We could do nothing; we could offer treatment if he wants it; we could mandate treatment; or we could say, "You will be tested twice a week and every time you are dirty you are spending 48 hours in jail." We could also impose testing and sanctions and, on top of that, offer or mandate treatment. Which one of those boxes is going to be the most effective and most cost-effective? No one knows. The answer probably varies from offender type to offender type. Here is what I propose. Everybody who is convicted of a crime that has a high drug-use rate associated with it is to be tested at arrest; if tested positive, that person should be subject, as part of a probation sentence or postincarceration-supervised release, to twice-a-week drug testing. Many members of that population will be able to control their own drug use under that sort of pressure, usually after they fail a couple of times. Eventually, they will stop. But there is a group (I do not know how big; it probably varies from offender type to offender type) who will just keep failing. That group obviously needs treatment. Having failed several times, they are also likely to have noticed that they need treatment. If you talk to a treatment professional, what is the first step in helping someone into recovery? Breaking through denial. Testing and sanctions is a wonderful way to break through denial and to create a therapeutic alliance. It is not, "You have to come to my treatment program or I'll tell your probation officer on you," but "You have to get clean or that probation officer is going to put you in jail. So let's work together on getting you clean." Testing and sanctions will work for a large portion of the population. I bet testing and sanctions alone will outperform treatment alone. Maybe it won't outperform treatment that is delivered by really great providers and that clients go to all the time, but that treatment is not in the real world. Testing and sanctions ought to be compared with the treatment we are actually capable of delivering and the attendance rates we are actually going to get. But treatment is going to cost a multiple of what testing and sanctions cost. Testing and sanctions, with referral to treatment for people who keep failing, has to be the right solution. And it is the solution everyone will be pushed to if you have a working testing-and-sanctions program, because testing-and-sanctions programs mean that continued drug use by that population is expensive for the criminal justice system. Let me try and state two views on this issue. One view is that we know something about drug abuse. We know it is a treatable disorder, and a very large fraction of persistent offenders have this disorder. If they stopped having that disorder, they would be better off and so would we. We should ensure that everybody who enters the criminal justice system gets an evaluation of his or her treatment needs. And we ought to prepare a treatment plan for that person, require compliance, and make the treatment resources available. That is the view embodied in the "Breaking the Cycle" program. Here is what I find problematic about that. We do not have the treatment resources. If we took every possible treatment slot and assigned it to the criminal justice system, we still would not have the resources. And the capacity of the system to make people go to treatment is not nearly as great as we would like it to be. Requiring treatment first risks bankrupting the treatment system and the capacity to do the needs assessments. Imagine 5 million people are coming into probation every year and needing 5 million drug treatment needs assessments. Who is going to conduct them? I'm worried that we will be able to implement that sort of Cadillac program for only about 10 percent of the population; everybody else will have to wait in line. What we can do on a mass scale is testing and sanctions. That's cheap and efficient, and it is adequate for many drug-using offenders. Since we know how to do that, let's do that for everybody and let the people who need treatment identify themselves by failing. Then we can make and execute treatment plans for that hard-core population rather than trying to do it for everyone. In terms of crime control, I love drug treatment; making drug treatment more available to offenders who want it is probably one of the most important things we can do to control crime. But if we let our need to reduce drug consumption among offenders wait for Congress to allocate enough treatment money, we are going to wait a long time. Jim Gibson, U.S. Sentencing Commission, Washington, D.C.: We don't have time to go into details with you about targeting crime hot spots. But I'm wondering about the connection between that idea and your need for swift punishment. If we can figure out where these hot spots are, would that be enough to formulate an effective response that would be swift and certain punishment? M.K.: I think you could look at a process like the one that Bill Bratton put into effect in New York. When there is a situation where crime is overwhelming enforcement, instead of trying to control everything at once, try to control something. Pick an area; pick an offender group; pick an offense type. Say, "We are going to try to drive this pretty close to zero." Then, once you have done that, spread it out. That's the way they went after the graffiti problem in the New York subways. They didn't try to control all the cars at once. They controlled 10 percent of the cars. Then when they convinced the graffiti artists that there was no point in painting those cars because they were never going to leave the station dirty, then they made it 20 percent, and so on. The trick is to get enough concentration so that you can move the offending rate and transform the group dynamics. In fisheries management and in forest management, there is an idea of "maximum sustainable catch" or "maximum sustainable yield." You try to figure out how many trees you can take or fish you can catch without depleting the population. Too much of our law enforcement looks like we are trying to achieve the "maximum sustainable catch" of arrests. What we want to do is "over-fish" or extinguish kinds of behaviors so that arrest counts go down. The goal is maximum control with a minimum number of arrests, and that is done by concentrating. You know those old movies in which the Texas Ranger is facing an angry mob and he has only one bullet left in his gun. How does he control the entire mob with one bullet? He says, "The first guy who takes two steps forward, I'm goin' to kill!" If people believe that, he does not have to shoot anyone. That ranger ought to be our icon. Notes 1. Wilson, James Q., and George L. Kelling, "Broken Windows: The Police and Neighborhood Safety," The Atlantic Monthly 249 (3) (March 1982): 29-38; Wilson, James Q., and George L. Kelling, "Making Neighborhoods Safe," The Atlantic Monthly 263 (2) (February 1989): 46- 52; Perspectives on Policing (series), published jointly by the Program in Criminal Justice, Harvard University, and the National Institute of Justice; Goldstein, Herman, Problem Oriented Policing, New York: McGraw-Hill: 1990; Skogan, Wesley G., Disorder and Decline: Crime and the Spiral of Decay in American Neighborhoods, Berkeley: University of California Press, 1992. 2. Kennedy, David M., "Pulling Levers: Getting Deterrence Right," National Institute of Justice Journal 236 (July 1998): 2-8. 3. Kleiman, Mark A.R., Kerry D. Smith, Richard A. Rogers, and David P. Cavanagh, Imprisonment-To-Offense Ratios, Boston: BOTEC Analysis Corporation, 1988. 4. Chaiken, Jan M., and Marcia R. Chaiken, Varieties of Criminal Behavior, Santa Monica: RAND Corporation, 1982. 5. Greenwood, Peter W., Selective Incapacitation, Santa Monica: RAND Corporation, 1982. 6. Wilson, James Q., and Richard J. Herrnstein, Crime and Human Nature, New York: Simon and Schuster, 1985. 7. Tversky, Amos, and Daniel Kahneman, "The Framing of Decisions and the Psychology of Choice," Science 211 (January 1981): 453. 8. Ainslie, George, "Specious Reward: A Behavioural Theory of Impulsiveness and Impulse Control," Psychological Bulletin, 82 (1975): 463-496. 9. Bentham, Jeremy, An Introduction to the Principles of Morals and Legislation, Oxford: Clarendon Press, 1996, Chapter 14, paragraphs 18-19. 10. Kleiman, Mark A.R., "Coerced Abstinence: A Neopaternalist Drug Policy Initiative," in The New Paternalism, ed., Lawrence M. Mead, Washington, DC: Brookings Institution Press, 1997: 209-210. 11. Schelling, Thomas C., Micromotives and Macrobehavior, New York: W.W. Norton and Company, 1978. 12. Kleiman, Mark A.R., "Enforcement Swamping: A Positive-Feedback Mechanism in Rates of Illicit Activity," Mathematical and Computer Modeling 17 (2) (January 1993): 65-75. 13. Kennedy, David M., and Mark Moore, "Closing the Market: Controlling the Drug Trade in Tampa, Florida," Kennedy School of Government: Case Studies in Public Policy and Management, Cambridge: Harvard University, January 1993. ------------------------------- Men and Fathers in the Community Presentation by Felton Earls, M.D. Professor of Child Psychiatry Harvard Medical School February 4, 1999 Washington, D.C. As a physician and social scientist, I have been interested in the idea that fathers matter in some specific and crucial way to children and to the wider society since I began research in the late 1960s.[1] The interest was spurred by a belief I developed in medical school that the introduction of the contraceptive pill would mark the greatest achievement of the 20th century--a development more revolutionary than atomic energy or space exploration. Despite the profound implications of understanding how men generally and fathers particularly are adapting to changes brought on by the reproductive and political emancipation of women, this issue has remained the least informed and murkiest of all my interests. I am driven by the conviction that social science research can increase understanding of our social world and eventually become a basis for the promotion of justice and security. But social science is young, immature in the best sense of the word, and aspiring, to be sure. So being in a policy arena such as this is akin to being on a skateboard: exciting, daring, and requiring extraordinary balance and constant attention to perform well. As a 10-year grantee of the National Institute of Justice (NIJ), I am heartened to witness NIJ's effort to address the basic issues of human development in relation to criminal justice. I have three aims in this presentation. The first is to characterize the status of the American father in relation to the family and the larger society. The second is to place fatherhood in a wider context of manhood. Finally, the third is to discuss both these issues as they relate to the Project on Human Development in Chicago Neighborhoods, a large-scale interdisciplinary study of youth development in an urban community. This project is concerned with two pertinent problems: charting the developmental pathways of males and females from birth to young adulthood and examining the mechanisms by which communities guide or deflect these pathways toward healthy or pathological behaviors. It is this second concern that most acutely occupies the project today; that is, how community influences are conceptualized, how compelling they are alone and in combination with other factors, and what clues suggest efficient, ethical, and potent strategies for strengthening communities. Fathers in Time and Space Beyond being fertilized as an egg and having a biologically and socially competent mother, one might naively ask, What is the specific significance of having a father? Or to put the question more generally, Is there any reason to believe that men play a crucial role in the rearing of children? Admittedly this is a crude question but one a contemporary social scientist should be asking. It is instructive to begin with a broad view of American family life as it has evolved over the past century. Donald Hernandez describes two distinctive demographic transitions during this period.[2] The first occurred gradually as the population shifted from farms to cities. With less dependence on a large number of children, families in these urban communities became smaller. Correlated with this change were a greater emphasis on schooling and an increase in the educational attainment of parents and their children. This transition was completed by the 1940s. New changes were set in motion after the postwar economic boom. With industrialization on the decline and new economic challenges facing society, the forces that drove families to cities were redirected; families began to look elsewhere and discovered the suburbs. These same factors, combined with continued emancipation, led women to join the labor force in greater numbers than ever before. While it is not possible to say definitively why these changes occurred, two phenomena quickly resulted and show no signs of reversing: an increase in breakdown of families resulting in growing numbers of mother-headed households and an increase in child poverty. The latter is a direct consequence of fathers contributing less and less to their children's welfare. The traditional two-parent family structure has dramatically declined during the past four decades. Currently, only 12 percent of American children live in families with a father who works full time and a mother who does not work outside the home; 26 percent live in families with a father working full time and a mother working outside the home part time; and 16 percent live in families in which both the mother and the father work full time. If these two-parent structures are "ideal," nearly half of America's children live in "less than ideal" family configurations. Because I am focusing on both family-level and community-level phenomena in relation to fathering, it is worth noting that indicators of declining civic engagement may well be coupled with the decline in traditional family structure. Decline of Male Authority The following passage from Alan Ehrenhalt's The Lost City captures in a poignant way the significance of male authority figures at the neighborhood level:[3] If it was football season, Father Lynch didn't want anyone to leave the church before he could pressure them to attend the upcoming game. . . . The Catholic elementary schools played on Sunday afternoons in the 1950s and for Father Lynch football was merely an extension of the religious service. Virtually all the boys went out for football, even ones without a trace of athletic ability. Nearly all the eighth grade girls were cheerleaders. He was a sincere believer in the doctrine that "football is a new crusade; it kills prejudice and it stimulates faith." These are the strong values and potent words of a patriarch. But the religious orientation of the priest was reflected in the secular sectors of this neighborhood as well, by teachers, police, and local businessmen. St. Nicholas of Tolentine, which is depicted in The Lost City, was a complex, enveloping institution in the neighborhood. In exchange for membership it provided a moral and social anchor for life. It dispensed rules to live by or to rebel against, clear rules with no grey zone. You either accepted Father Lynch or despised him. When Ehrenhalt returned to this same neighborhood in the early 1990s, this kindly, intimidating, but wholly committed authority figure had been lost. There was no replacement for Father Lynch. The neighborhood retained many of its physical features and sense of place, but to the writer it appeared valueless and socially adrift. This was reflected not just in the visibility of the church but with attitudes that characterized the schools, the police, and local entrepreneurs. This is not an uncommon experience for many Americans of my generation. When I read this account of a Chicago neighborhood in the 1950s, I reminisced about my own neighborhood in New Orleans. Our patriarch was not a Catholic priest, but Oscar Hansberry, a committed professor of music and the most authoritarian teacher imaginable. His mission in life was to tame young children by teaching them music and making them members of a band in which they would learn to keep time and be punctual. They also would learn the importance of interdependence and loyalty and to accept authority as the inevitable price a child has to pay to achieve anything worthwhile. The male authority figure in the community and the father as head of the family both became relics in the decades following the 1950s. The American family could not meet massive new obligations within this patriarchal system. Indeed, social scientists learned that terms such as patriarchal and matriarchal were hypothetical categories that no longer fit the existing profile of American families. By the 1960s the family was well along the path of becoming egalitarian, and with this change marriage had to become a partnership. Psychology backed up this spirit of democratic marriages and bilateral child rearing. Russell Lynes, the social critic,[4] put it succinctly and in terms that were undeniably emasculating: "The man, once known as the head of the family is now partner in the family firm, part-time man, part-time mother and part-time maid." These words carry such strong negative connotations that they must have made many people at the time wonder whether this phenomenon could ever be adaptive. We now know the answer: Yes, it is adaptive. Yet the words have not lost much of their chastising intonation. With the loss of authoritarianism and patriarchy (however mythical the term), it may be argued that we have lost not only the ability to enforce standards of conduct but also any clear sense of what standards we would wish to enforce. In most areas of American life we have enshrined choice, and all choices are supposed to be treated equally. Surfing television channels or the Internet have become the contemporary metaphors for the values that regulate life, that provide whatever degree of order and predictability in daily affairs we have. To quote Ehrenhalt again,[5] This generation will come to adulthood in the early years of the next century with an entirely different set of childhood and adolescent memories from the ones their parents absorbed. They will remember being bombarded with choices, and the ideology of choice as a good in itself; living in transient neighborhoods and broken and recombinant families where no arrangement could be treated as permanent; having parents who feared to impose rules because rules might stifle their freedom and individuality. In place of authoritarianism, as personified by Father Lynch and Professor Hansberry, we have drifted into a kind of egalitarian obesity. Our children simply have too much choice. Let me be clear that this pessimistic opinion represents a white, working- and middle-class perspective. Further, it is not the speculation of a researcher, but of a social observer. Can scientific evidence back it up? Before turning to research, however, I must introduce a contrasting perspective on manhood and fatherhood, that of the African-American male, because it is sufficiently distinct in historical, social, and economic terms to command special attention. My Mother Who Fathered Me--Female-Conditioned Socialization of Boys It is a challenge for two parents to raise a child in a way that does not undermine their curiosity or their constant urge to explore, experiment, and test limits. The average single parent has an impossible task. Being careful not to risk stereotyping, I am deeply worried about the plight of a large proportion of African-American families, those that live in poverty and that at least functionally fit the category of single-parent, female-headed household. This group of families has never done well economically in American society, and their plight today is as serious and bleak as ever. The phrase, "My mother who fathered me," comes from the title of a West Indian novel that describes a boy growing up in a female-dominated household and community (a so-called matriarchal society). Many of the circumstances that define the relationship between men and women, patterns of decisionmaking and working to achieve a sense of economic security, appear to be the catastrophic and persisting legacy of the past. The striking correlation between the proportion of female-headed households in poverty and community violence represents a deeply rooted and inescapable reality of much of contemporary urban life. Geoffrey Canada, in Reaching Up for Manhood,[6] recalls his father in these terms: He was simply a man who had helped to create me and seemed to take no special interest or pride in that fact. In many ways our family was luckier than most who had no fathers because we had a strong mother who did everything she could to make up for the fact that our father was not around and not interested in his children. Yet even though I came to accept that my father would never be there for me, would never see me play varsity football, or basketball, never see me graduate from high school or college, I couldn't help but wonder what it felt like to have a father in your life. In another passage, Canada captures exactly the research question that has occupied me for so many years:[7] And while some of my friends had fathers, we couldn't figure out if in the end their families were that much better off than the families of those of us who didn't. The fathers on our block were spoken of in hushed tones. You never saw them outside teaching their children to roller-skate or play football. When my friends talked about their fathers it was usually in the context of discipline. Fathers, we were told, hit harder than mothers. They were hard characters who demanded quiet when they came home from work, and quick and complete obedience to their every demand, and who expected mother and child to act respectful and even fearful. If it is the private persona of fathers that matters in the creation of family life, it is the public persona of men acting in fatherly ways that seems important in repairing or building strengths at the community level. The public image of fathers has taken a serious beating, with terms such as "deadbeat dads," "throwaway dads," and "underground fathers" becoming commonplace. Men may be more marginalized in relation to their families than ever before in American history. And this circumstance is not a temporary separation caused by going away to war or working at a remote location. The terms used today specify a father who borders on being virtually useless. How has this happened so quickly? Does research tell us why we are in this state of affairs, how deep it is, and how out of control it is? What Research Has Shown About Fathers In providing an overview of research on fathers, it is important to underscore the reality that social science rarely rises above the norms and conventions of the historical period in which it is conducted. Interest in fathering was not reflected in child psychology until the decade following World War II. This first generation of questions involved the effects on children of absent fathers brought on by the war. This body of work produced important findings, but overall the effects were thought to be largely transient and indirect (that is, filtered through the mother-child relationship). It is sobering to look back at the era of the 1950s to witness this as the period in which traditional family structure began its steady decline. The issue of absent fathers and children's development spawned work on the impact of the death of a father as the most extreme example of loss. The basic question was: Does the death of a father leave emotional scars equivalent to those left by the death of a mother? The answer was a qualified yes. The loss of a father appears to have a greater effect on adolescents than on younger children, but frequent remarriage and family reconstitution following paternal death produce their own separate effects, making it difficult to say what effect death alone has on behavioral and emotional functioning. As I completed my training in pediatrics and child psychiatry in the 1970s, a third generation of research was taking root. On the basis of previous decades of work indicating that the absence or loss of a father was important, new questions were asked about the specific contributions made by a present and involved father. This agenda began by addressing the sanctity of attachment theory, a psychoanalytical concept positing that a secure affiliation with the mother in a single and dominant relationship was a requirement of good mental health. Research challenged and eventually undermined this notion. Children not only become emotionally bonded to multiple figures including fathers, but their relationships with fathers also possess a quality that distinguishes it from the relationship with mothers. Involvement with fathers appears to promote independence and stimulates academic success over and beyond that of involvement with mothers.[8] Although it has not been demonstrated, to the best of my knowledge, that father involvement operates as a specific protective factor against the development of criminal behavior, this can be inferred from the literature. It is difficult for the researcher to demonstrate the importance of fathers during the same period in which society appears to turn its back against them. One important area of research that relates directly to my major interest unfortunately never materialized over the past four decades. It concerns the role fathers play beyond the family in fostering a climate for healthy development of the child within the community. In large part this is due to the overwhelming focus on the mother relationship and family influences over community influences in child development research. Various efforts were made to approach groups of at-risk fathers in a manner similar to that adopted for at-risk mothers. This approach typically includes forming a therapeutic or support group to educate parents about child development and provide emotional support and concrete advice. In practice, this approach generally does not work for men, yet I can find only a single paper in the vast literature of child development that presents some explanation of why it does not. While setting up a fathers' group parallel with an ongoing mothers' group in a disadvantaged section of Chicago, Samuel Tuck discovered that men wanted a more active role in changing the community climate for children than could be realized from participating in support groups.[9] This group of fathers started by planning a series of activities, such as street fairs and field trips, to stimulate academic success in their preschool children. But this level of action was soon superseded by a more ambitious program aimed at transforming the neighborhood environment and gaining greater community control. At the time Tuck's report was written, the fathers had operated an ice cream truck during the summer months and were establishing a supermarket and a much-needed cooperative credit union. They had made an important transition from increasing the academic skills of their nursery school children to creating a healthful community environment for all children. When I read Tuck's article 25 years ago, I was convinced that it contained strong clues for designing effective child development programs and research. Until quite recently, the significance of these ideas has been dormant. Child researchers continue to design much of their science around the perspectives and activities of mothers, and programs of all types continue to treat fathers as marginal. Toward a New Research Agenda and the Chicago Study Summing up a 1994 report covering the past 30 years of research on fathering, the National Research Council (NRC) identified four limitations of the existing research, two of which are relevant to this presentation.[10] One limitation is the failure by researchers to understand fathering as a negotiated matter. "Men do not just assume the role of father," says the NRC, "they negotiate it, either in partnership or in conflict with their children's mothers." The report urges researchers to examine how this process of fathering is established in families with, for example, a disabled child, where virtually nothing is known about the father's roles but where the impact of paternal involvement may be particularly discernible. The second limitation relates to the community context of fathering. This concern arose while discussing the role of older men in inner-city communities but was expanded into questions about how community norms of responsible fathering are established, communicated, sustained, and transmitted from one generation to the next. These issues are especially acute for children raised in father-absent households for whom community norms and nonparental male role models are likely to be important. But where are these role models? They are rare in elementary schools and even rarer in preschool and early child care settings. How fathers view their role in the family, how they behave privately and publicly, and the nature of their formal and informal civic contributions within various social settings are richly deserving of study. The Chicago project provides at least four areas in which to study the relationship of men and fathers in response to large-scale community transformations: --The devolution of public housing. --The acculturation of Latinos, primarily rural Mexicans, to the lifestyle of a large city. --The impact of welfare reform on both fathers and mothers. --The marginalization of a substantial proportion of African-American men in the criminal justice system. Each of these conditions is important to research because of the demographic characteristics that are pivotal to forecasting what our cities will be like over the next generation. Cities represent the leading wave of population change that is reflected in "minorities" becoming the statistical majority in some sectors of American society. Another reality, already discussed, is the change in family structure that results in men no longer playing the role of dominant breadwinner or authority figure. The adjustment and modeling of boys and men to their negotiated manhood in the community and their negotiated fatherhood in the family are centrally important to how these changes ultimately support or undermine the stability of American families and communities. The Chicago study engages two research strategies to address these problems. The first is to longitudinally chart the extent and nature of men's involvement with children as fathers, other relatives, teachers, and coaches and its impact on children's well-being. Social class and ethnic variations are of great importance in this analysis. The second strategy is to study the attitudes of boys and men toward fertility and child rearing and examine these as aspects of a changing male ideology. Terms such as "hypermasculinity" (indicating antifemale attitudes used to justify a lack of involvement and caretaking behaviors) have crept into the research literature without a great deal of evidence for its occurrence. We are primarily interested in the extent to which these behaviors are shaped and sustained by community expectations and values. Producing results such as these takes some time. In the current phase of the study, we are chiefly concerned with the level at which men participate in the study and the accuracy of their responses. We examine community knowledge and residents' perceptions of the social organization of their neighborhoods. This work has shown that social cohesion among adults and their willingness to intervene in the lives of children are correlated with lower rates of violence. The sample was identified by randomly selecting households from each community and requesting that an adult household member participate in an interview averaging 45 minutes and addressing issues of safety, security, and quality of life for children. All neighborhoods in Chicago were stratified according to the race, ethnicity, and economic status of residents. Much to our surprise, 41 percent of respondents were male. We are surprised because, when conducting research on children, we have become conditioned to look for mothers to participate in our interviews. But we also are surprised by how little variation there is among neighborhoods in the proportion of men involved in the study. This figure ranged from 33 percent in low-income African-American areas to more than 50 percent in some working- and middle-class neighborhoods. What this means is that men are at home to answer the door, they agree to participate, and they provide responses that are similar to and as valid as those provided by women. Although this does not jibe with the prevailing societal view of men and fathers, it is not a radical finding. We have examined our data for evidence of gender differences in perception of neighborhood characteristics and find very few to report. Scores on our measure of collective efficacy (social cohesion and informal control), the major focus of our work thus far, are exactly the same for men and women. The few differences are: 1) men participate in neighborhood organizations at a higher rate than do women and 2) women perceive greater levels of danger and disorder than do men. The broad similarities are encouraging and compel us to look more closely at how men and fathers are involved in the four family- and community-level processes previously mentioned (public housing devolution, welfare reform, immigration of Latinos, and overrepresentation of African-Americans in the criminal justice system). What's in Store for the Next Decade? I would like to conclude by envisioning the types of events that may well focus our attention during the next decade on the role of men and fathers in creating a higher level of security and a higher standard of living in our cities. First, the character and function of a social movement may give greater definition and clarity to the position of men in American society. This movement has been given license by several writers and activists, prominent among them Robert Bly in Iron John,[11] the Good Fatherhood Program of David Blankenhorn, the Father to Father Campaign established by Vice President Al Gore, and the Rheedlen Centers established by Geoffrey Canada. One manifestation of this movement is the impetus and meaning it gave to events such as the Million Man March and the Promise Keepers movement. As poorly understood mass events on the American scene, they each have a continuing presence at the local level. The failure to document and understand this presence is as much a shortcoming of public policy and the media as it is of social science research. Major new research centers at our universities, such as the National Center on Fathers and Families at the University of Pennsylvania, will stimulate new work in this area. This center has set a research agenda that includes the following seven propositions to guide research and practice over the next decade:[12] --Fathers care. --Fathers' presence matters. --Coparenting must be bargained. --Impregnation carries a deep and enduring responsibility. --Joblessness undermines fathers assuming responsibility. --Government policies regulating family life act as barriers to achieving responsible fathering. --The intergenerational transfer of values, beliefs, and behaviors is a powerful force in shaping good parenting. These hypotheses reflect much of what I have discussed in terms of manhood and fatherhood, but in their comprehensiveness and depth they still stop short of helping to establish the community norms stressed in this presentation. Increasing men's engagement in the lives of children in their neighborhoods can contribute to the decline in violence now documented to be taking place in many cities around the country. This can be seen in Boston with groups such as the Ten Point Coalition, a group of African-American ministers who seriously monitor and care for at-risk youth. This image stands in marked contrast to the depiction of African-American men in Geoffrey Canada's community. Indeed, it is a celebration of human potential that Canada is now committed to training boys to become the good father he never had. A second area to watch and study is the expansion and intensification of the professionalization of child rearing. Child care has increasingly become a commercial market, and most parents have difficulty finding acceptable arrangements for their children. This means that mothers and fathers are becoming less critical determinants of their children's success or failure in life. Their achievements are increasingly related to the indirect influence of the quality and stability of child care and school and after-school programs, when available. The decline in family structure witnessed during the past 50 years may be followed by a decline in the significance of parents over the next few decades. We are witnessing a particularly ugly rendition of this breakdown of families in some parts of Eastern Europe and Russia with their reliance on large and impersonal institutional care for young children.[13] I hate to imagine that any approximation of this nightmare could happen in the United States. The Roles of Minority Fathers Two issues relate to minority populations. Much of the future of urban communities in this country is contingent on how Latinos become acculturated. I anticipate that some of the most important findings to emerge from the Chicago study are related to this issue. An essential aspect of the acculturation process involves adaptation by men and boys. Latinos' ideology about masculinity and their involvement with children and communities in relation to American values and lifestyles represent perhaps the most significant current social transformation in our society. Much research is reactive in nature. Researchers have not kept up with the pace and direction of changes taking place in society. A new social norm evolves or a new policy is introduced, and researchers then aim to discover the reason. Two examples had a major impact on African-American fathers. First, current welfare reform legislation might make fathers even more distant from their children. Requiring these mostly poor and underemployed men to make child payments could easily backfire by fostering increased resentment toward their children's mothers and the government. Second, as the number of incarcerated minority males has risen dramatically, so has the stigma of worthlessness and debilitation that is being attached to ex-convicts. Not only does this have a negative effect on individual families; given the numbers and the selectivity of the phenomenon, it must exert a negative impact on communities as well. How well prepared are we for the reentry of these men into their communities and families? Might they be returning less well prepared for the roles and obligations of manhood and fatherhood than they were before entering the criminal justice system? One of my keenest and most immediate aspirations is that our research forcefully addresses these questions. The continued support of governmental and nongovernmental research agencies may help create a scientific platform from which to anticipate changes in our social world and better understand the distinctive contributions both men and women, both fathers and mothers, make to justice and security. Levels of crime and violence are symptomatic of the way families and communities function and how they are changing. A more balanced research agenda, one that pays as much attention to men's as to women's roles in society, is needed to advance our understanding of how to create a secure and prosperous community. The relatively hidden male must be more sharply revealed, and the relatively overexposed female must be diminished in this endeavor. Social science research represents a powerful vehicle to achieve this goal. Question-and-Answer Session Devon Brown, District of Columbia Department of Corrections, Washington, D.C.: My question relates to how differences in the perception of masculinity relate to crime prevention. I think it's pretty well established in the correctional field that one crime seems to be race-based, and that is child molestation. Child molestation is not usually an African-American crime. Is this because the community believes in the idea of hypermasculinity and child molestation would be the worst thing to happen? F.E.: This is a tough question. There are differences in racial experience relating to specific kinds of crime in the family. A certain attitude might guard against committing child sexual abuse. The social class distribution of child sexual abuse differs from child physical abuse. Child physical abuse is heavily related to poverty; sexual abuse occurs more equally across all income groups. As a child psychiatrist in Boston, I once looked for sample cases of matricide among black families and was unable to find one. This kind of homicide is one area in which there may be an important difference from white families and perhaps it relates to black families being more matriarchal. Donald Murray, National Association of Counties, Washington, D.C.: What differences have you seen, as a result of recent welfare reform, in the mother's use of time, considering the importance of children's early years? F.E.: In Chicago, I have heard the phrase "the clock is ticking." Many women turn to informal child care arrangements when they enter the labor market. There has been rather murky reporting on that type of child care, with the assumption that a center is the best arrangement, then family day care, and next an informal child care arrangement. I'm not sure about this assumption. Many women in Illinois work with organizations like United Parcel Service or nursing homes and must work shifts for which no child care is available. What kind of arrangements are these women making? Professionalization of child care is a concern. I would not want to see here in the United States what I saw in the Romanian orphanages. Christopher Stone, Vera Institute of Justice, New York, New York: I want to ask, in relation to men and fathers, how much is defined beyond the individual male role? By that I mean social policies, behavioral influences, and involvement of the community, etc. F.E.: The meaning of biology is being redefined in our time. Last year, 20,000 babies were born from in vitro fertilizations. Biology should speak of a social commitment, but there is no compelling reason that it will. We are seeing an uncoupling of the implications of social commitment in biology. As an example of commitment in communities and families, I intentionally chose a Catholic priest in my example, a "father" who has no children. Men who care about children have an important role in communities. Most families and communities have both cooperation and conflict. Children need to be able to witness and learn from the process of going in and out of circumstances of cooperation and conflict. This is a process of learning from social experience. Lauren Hampton, Charles Mott Foundation, Flint, Michigan: Have you looked at collected data on workforce participation, on a day-to-day basis, in the neighborhoods of your project? F.E.: Yes. It takes more time to get a clear picture; people make some very complex arrangements. We have spent much more time asking about labor market items than I expected. People often have more than one job, and their jobs change. Transportation can be a very big issue. Where people live and where the jobs are differ a lot. The transportation system is frequently opposite to their needs. They spend a lot of time waiting on bus routes, etc. This is one area that particularly shows the need for a support system for men returning from incarceration. Mentally ill people who are coming out of prison may simply be told to go to a certain mental health center. The center could have a long waiting list. Prerelease programs should look at employment and other support. There are no clear health standards or standards on community support. People are released with serious problems, ranging from asthma to substance abuse and mental illness. The effects of these problems have not been clearly documented. David Anderson, National Victim Center, Arlington, Virginia: There has been a lot of criticism about social service programs that "do to" people rather than "do with" people. What are your impressions of John McKnight's asset-building or capacity-building approach? F.E.: Measures need to be bottom-up as well as top-down. Robert Sampson, one of my colleagues, is looking at structures of leadership in the community. Leaders may come from business, the police, the schools, or the citizens, and there will be much variation in their views. The leadership structure may be horizontal across sectors, or vertical and truncated with a strong father/authority figure and not many other links. An important question is sustainability of effort. Many communities do work. To repair or help communities, we need better intervention strategies. Jean O'Neil, National Crime Prevention Council, Washington, D.C.: One piece of information you are giving to communities is that men are at home; they answer the door and are concerned about children. We should clear up the idea that these men are not valuable, direct some help to them, and support fathers to be a valuable part of the community. Charles Sullivan, Citizens United for Rehabilitation of Errants (CURE), Washington, D.C.: I'd like to hear about incarcerated fathers. We are going to have 2 million incarcerated by the year 2000 and 94 percent of those are men, of whom 84 percent are also fathers. About 50 percent are under the age of 25. What can we do so they can connect with their children? We will have to bring the children to the prisons and make prisons more "child-friendly" to keep those fathers close to their children. F.E.: Prisons are one of the more invisible institutions in our society. We have quadrupled the prison population. Some look at these issues as human rights violations. We have to make the system more visible and publicly accountable for the quality of life men and women have in these institutions. Only then can they be made "child-friendly." Notes 1. Earls, Felton, "The Fathers (Not the Mothers): Their Importance and Influence with Infants and Young Children," Psychiatry 39 (1976): 209-226. 2. Hernandez, Donald J., "Child Development and the Social Demography of Childhood," Child Development 68 (1997): 149-169. 3. Ehrenhalt, Alan, The Lost City: The Forgotten Virtues of Community in America, New York: Basic Books, 1995: 124-125. 4. Ibid., 233. 5. Ibid., 281. 6. Canada, Geoffrey, Reaching Up for Manhood: Transforming the Lives of Boys in America, Boston: Beacon Press, 1998: 47. 7. Ibid., 37. 8. Parke, Ross D., and Armin A. Brott, Throwaway Dads: The Myths and Barriers That Keep Men from Being the Fathers They Want To Be, New York: Houghton Mifflin, 1999: 3-13. 9. Tuck, Samuel, "Working with Black Fathers," American Journal of Orthopsychiatry 31 (1971): 465-472. 10. National Research Council, America's Fathers and Public Policy: Report of a Workshop from the Board of Children and Families, Washington, DC: National Academy Press, 1994: 38. 11. Bly, Robert, Iron John: A Book About Men, Reading, MA: Addison-Wesley, 1990. 12. Visit the Web site of the National Center on Fathers and Families at the University of Pennsylvania for additional information: http://fatherfamilylink.gse.upenn.edu/. 13. Human Rights Watch, Abandoned to the State: Cruelty and Neglect in Russian Orphanages, New York: Human Rights Watch, 1998. ------------------------------- Violence, Free Speech, and the Media Presentation by Sissela Bok, Ph.D. Distinguished Fellow Harvard Center for Population and Development Studies April 12, 1999 Washington, D.C. I want to consider how public perspectives on violent crime can skew and vitiate the administration and the exercise of justice and ask about the role that the media play in forming such perspectives. I shall speak of "perspectives" on violent crime both in the general sense of how we perceive and understand it and in the more particular sense, taken from optics, of how we can visualize what is three-dimensional when it is presented on two-dimensional surfaces such as paintings, printed pages, or TV screens. Using examples taken from media coverage of the "copycat" school shootings in recent years, I shall ask: What does it mean for our perspectives on violent crime that we should be more exposed than any previous generation and any other society to media violence so massively, graphically, and often alluringly conveyed from early childhood on? How can we best respond to the risks from such exposure--risks not only of higher levels of aggression but also of misguided policies resulting from biased, uninformed, or misinformed perspectives on violent crime? What options exist for policymaking with respect to media violence that do not move in the direction of government controls and censorship--something that would do violence of a different sort to our democratic traditions, including freedom of speech? And what might it mean for journalists and media executives to exercise their roles with respect to media violence more responsibly without compromising their duty to safeguard free speech?[1] In considering such options for individuals and communities, for public policy, and for the media, I will suggest areas where research continues to be needed and others where sensible precautionary choices are needed before all the evidence is in. Violence Brought Home by Television I want to begin with a quotation from the writer E.B. White about the future of television--words that he spoke, amazingly, already in 1938: I believe television is going to be the test of the modern world, and in this new opportunity to see beyond the range of our vision, we shall discover either a new and unbearable disturbance of the general peace or a saving radiance in the sky. We shall stand or fall by television--of that I am quite sure.[2] The ability to expand our perspective so extraordinarily--to see beyond the range of our vision (tele and vision) and to do so in our own homes, having at our fingertips the power to extend our vision into territories we never knew existed--this ability, which seemed so astounding at the time, is now so commonplace to all of us that we hardly realize what a change it has brought. Has it allowed us to discover either "a new and unbearable disturbance of the general peace" or "a saving radiance in the sky"? My view is that it has in fact done both: It has not only expanded our perspective in astounding ways far beyond the range of our vision but in the process often altered that perspective in ways that are so hard to fathom as to be at times paradoxical--and rarely more so than when it comes to violence. The idea that violence should be fun, alluring, glamorous has long seemed paradoxical to many, much like that of a "kindly rapist" or a "two-sided triangle," even as others have seen this idea as unproblematic and simply confirming the perennial human delight in mayhem. But until our time, it has never been possible for viewers to tune in for the enjoyment of violent programming in their own homes at all hours of the day and night; nor in the past have up-close graphic scenes of violence, often sexual in nature, become part of the experience of viewers of all ages and been marketed so relentlessly to young people. The sense of paradox is especially striking the younger the children acculturated to finding violence pleasurable. Neither children nor adults take to such enjoyment without being taught new perspectives, new ways of perceiving violence, whether by the media or in their communities. We have to ask, therefore, whether these changed perspectives on violence affect beliefs or conduct outside of the viewing experience: What do these new perspectives do to viewers as moral agents in their dealings with others? Four Phases of Moral Response "Perspective"--on crime as on any other human problem--concerns, in the most general sense of the term, how we perceive something. Such perception is the indispensable beginning of the four phases of moral response: perceiving a problem, thinking it over or deliberating about it, arriving at a choice, and carrying it out. Of course there are times when we need not go through these phases in a conscious way--as when, seeing a child about to fall into a manhole, we respond intuitively with no need to think through what is best to do. But when we confront a conflict that pulls us in different directions, the process of responding has the four distinguishable phases of perception, deliberation, choice, and implementation. True, the four phases of moral response can never be neatly disentangled in practice, nor do they always occur in the same sequence. Each of the four, moreover, can be skewed, short circuited, or altogether absent in a situation of moral conflict. In each phase, the temptation is strong to opt for premature closure: for bypassing efforts at the fullest perception and the fairest, most careful deliberation, and settling for uncritical acceptance of simplistic choices and clumsy, inappropriate forms of implementation. Of the four phases, perception is primary and utterly indispensable. To the extent that our perception of a problem is biased, fragmentary, or compartmentalized, the deliberation about alternative courses of response is vitiated from the outset. These deficiencies in turn affect our choices, including policy choices we arrive at collectively, as well as their implementation. The Screen as a Lens A more specific sense of the word "perspective" refers to optics and concerns how we can visualize objects and persons as three-dimensional on something as flat as a television, movie, or computer screen. The screen allows us to do so in unprecedented ways. It both mediates our experience of what we perceive and renders it more immediate, making it possible for us to cross the boundary between being near to and at a great distance from screen images. When it comes to violent crime, the screen is today the primary conduit for perception: it acts as a lens for all the different kinds of violence in society, whether "real" or not, whether intended as entertainment or not, and whether meant to be artistic or not. It is on screens that most Americans witness all forms of violence--in news programs, "infotainment," or entertainment violence, including interactive video games and virtual reality programs. For today's children in particular the screen has become the lens through which they learn about violence. Their perception of and perspective on violence is affected long before they have had the opportunity to consent to being acculturated to violence in this way. It is through the magnifying and often distorting power of this lens that the everyday life of America's children becomes filled with guns, drugs, family violence, gang warfare, kidnapings, and all that contributes to violence in our society. The shift to such an overwhelming role for the screen as conveyor of all the world's violence is surprisingly recent in most societies. In America, it is only in the past 10 years that viewers, including children, have had access to the present variety and amount of violent programming on their home screens. And it is only since the mid-1990s that video rentals and computer-based games have made it possible to view and take part in violent programming over and over again. Today's screen violence is often glamorized and meant not just to intrigue viewers but to invite them to relish the infliction of injury and suffering. Youngsters are familiar with films portraying the murder of classmates, parents, police, and teachers as thrilling. "First-person shooter" games reward players for shooting and eviscerating victims in gory and photorealistic detail. In the video game Carmageddon, players can choose from more than 25 vehicles with which to run down as many animals and people as possible: to "Waste contestants, pedestrians, and farm animals for points and credits!" The rules suggest that players try "to think of new and humorous ways of pulping pedestrians!" In the words of a reviewer for Wired magazine, "It's not whether you win or lose, it's how many innocent bystanders you smear down the sidewalk in the process."[3] Familiar and once-shocking figures from the early 1990s warned us that the average child leaving elementary school has watched 8,000 murders and more than 100,000 acts of violence. Now that violence in films, television, and interactive entertainment has reached new levels, those statistics have easily been surpassed, especially because technology makes it possible to replay video sequences at will on the computer or the VCR. Blurring the Line Between Fantasy and Reality The sense of paradox is heightened for many who have never been acculturated to experiencing violence as fun or exciting or glamorous in the first place: How do individuals get to the point of enjoying it so vividly, so often, and at such length in such realistic forms of entertainment? What has happened to their perspective? Is their enjoyment of violence always confined to their role as consumers of entertainment violence? The line between consuming violence and engaging in it in real life is not always easy to maintain. For young people who are disturbed, neglected, or abused, media violence contributes powerfully to blurring that line. They may play and replay video games that reward them for devising and executing strategies for torture, rape, and murder or model themselves on movie actors showing such violence as thrilling. If they also have access to semiautomatic weapons, the temptation to put similar plans into effect in their own lives may overwhelm them. Immaturity, anger, fierce enjoyment of media violence, and access to deadly weapons make for a combustible mix. The copycat school shootings of the past 2 years have focused attention on young people, sometimes children, who have crossed the line from imagined to real-life violence. The public's grief and anguish after each new school shooting is heightened by the thrill and lack of pity that the young killers evince in carrying out their scheme of mass murder. How could they not only delight in so many forms of violence on the screen and off but also manage to cross the line between real-life violence and fantasy, sometimes even finding incitement and instruction from media violence? Few imagine that media violence makes the young perpetrators of these shootings act out their aggressions. At issue is a convergence of factors, not all present in any one case: depression, drugs, anxiety, parental abuse or neglect, and access to lethal weapons and Internet instructions on constructing bombs. For the most disturbed and psychologically vulnerable young people, glamorized media violence contributes especially powerfully to blurring the line between fantasy and reality. Earlier, many representatives of the entertainment industry had dismissed all talk of such interaction between risk factors as baseless or cited conflicting studies about precise levels and forms of interaction, much as representatives of the tobacco industry have long insisted that links between smoking and lung cancer and other diseases have not been definitively proved. It now seems less far-fetched to journalists and members of the entertainment industry that a minority of already troubled youngsters, exposed to so much inducement to regard murder and torture as thrilling, could take the step to actual killing. The same turns out to be the case in countries such as Japan with its profusion of extremely graphic media violence but with strict gun control laws, where school killings according to some studies are becoming more common than in the United States.[4] Is There a Link Between Media Violence and Societal Violence? Of all the possible effects of entertainment violence, the one most on the public's mind is that it creates greater aggressiveness among viewers. Polls show that about 80 percent of Americans believe that media violence is linked to the high levels of homicide, gang violence, rape, child abuse, and other violence in our society.[5] This is also borne out by the vast majority of studies on the subject, which agree that media violence can have both short- and long-term debilitating effects in lowering barriers to aggression among some viewers.[6] There is no consensus, however, about the degree to which exposure to media violence contributes to societal violence, although all researchers reject the view of 21 percent of Americans in one 1995 poll who blamed television more than any other factor for teenage violence.[7] Even the more common lower estimates that correlate 5 to 10 percent of societal violence with television violence point to the infliction of a substantial amount of suffering and death in a society with levels of aggression as high as America's.[8] Americans have special reasons to inquire into any factor that may be contributing to societal violence. While we are in no sense a uniquely violent society, we need to ask about the many reasons our levels of violent crime are higher than in all other stable industrialized democracies. Our homicide rate would be higher still if we did not imprison more of our citizens than any society in the world, and if emergency medical care had not improved so greatly in recent decades that a larger proportion of shooting victims survives than in the past. And although our homicide rate has declined in the 1990s, the rates for suicide, rape, and murder involving children and adolescents in many regions have rarely followed suit. The sharp media focus on recent copycat school shootings is understandable, given that the perpetrators and most of the victims are so young and that schools are considered havens from the violence that permeates much of the rest of society. But too often, the very sharpness of the media focus also contributes to a narrowing of perspectives on the victims of violent crime. Children are often safer by far in schools than in their own families and communities, where vastly more common acts of brutality and homicide take place daily across the country. It is hardly surprising that many in the public wrongly imagine that school killings represent the main problem with youth violence, given the disproportionate media attention to such crimes compared with the more numerous homicides committed by and inflicted upon children in society at large. Censorship Versus Free Speech The continuing focus on school shootings has also brought new calls for censorship of the violent material available on television, in movies and interactive video games, and on the Internet. Staunch defenders of free speech oppose any controls whatsoever, regardless of any societal harm that violent material can be shown to inflict. In between these positions, a number of Americans sense an insuperable dilemma. To them, it seems impossible to protect society, and especially children, against media exploitation of violence and the glamorization that can raise levels of aggression, while safeguarding our traditions of free speech. They see the two goals as irreconcilable, with no way of "going between the horns" of the dilemma. In my book Mayhem: Violence as Public Entertainment I argue that censorship is even more dangerous than the ills it is designed to remedy. Censorship is not only unacceptable in its own right, as shown by the harm done to free speech and democracy in countries the world over by governments engaging in it, but it also will be increasingly difficult to impose as new technologies offer more and more ways to circumvent censors.[9] With the advent of the Internet, even the most coercive societies find it difficult to maintain strict controls over communication. There is a further reason censorship is not only the wrong way to go but also unlikely to have an effect on harm stemming from media violence. Censorship of violent material is primarily focused on the risks of additional aggression, as in the case of copycat murders and serial killings. But apart from whether or not someone heavily exposed to media mayhem turns out to be more aggressive in real life, the fear about the effects of such exposure goes deeper. It concerns injury to viewers themselves. Were censorship deployed to reduce the fearfulness and desensitization to which media violence also contributes, it would have to cast a far wider net. In the long run, it would fail even more spectacularly to achieve the desired blockage of all that might contribute to such effects. The late British philosopher and novelist Iris Murdoch has written that the damage done to the inner life of viewers "through the imposition of banal or pornographic or violent images by television is a considerable wound."[10] For her, what is at stake is damage to the spirit: to resilience and to the capacity to feel for others and to respect them. Lessened resilience and corresponding heightened fearfulness can lead to choices about self-protection that are more "risk-averse" than would otherwise be the case. Results of Increased Fear and Reduced Pity Studies show that the sense that threats abound in the outside world is common among TV viewers of all ages who watch a lot of television. Media expert George Gerbner describes this effect as the "mean world" syndrome.[11] He and his colleagues have found that viewers who watch television more than 3 hours a day are more likely than viewers exposed to 2 hours or less to feel at high risk of victimization from violence, to perceive their neighborhoods as unsafe, and to regard the world as "mean and gloomy." News coverage plays a strong role in generating such attitudes. You would not know, from watching local television news in many American communities, that crime rates have been going down in the 1990s. What the Rocky Mountain Media Watch calls the "Mayhem Index," or proportion of media time devoted to crime, disaster, war, and terrorism, has remained as high as ever--averaging around 40 percent in 1998 and more than 80 percent for some stations. At times, as with each new school shooting, the percentage skyrockets in a great many more locales, and television viewers see scenes of mayhem or grief incessantly replayed on their screens every day. The media attention to these crimes intensifies the fear of parents and school children about safety at school, even though a minute proportion of children killed are actually killed at school. We need to ask to what extent heavy exposure to media violence relates to the rising levels of depression that now affect a far greater proportion of persons of all ages, including children and young people, than in the past. There is as yet very little research on this score, but as clinical depression moves to the forefront of illnesses affecting people worldwide, it will be important to sort out what forms of psychological stress lie behind such changes. We know that depression among adolescents has increased tenfold since the 1950s and that suicide rates for adolescents have tripled in the past three decades. Clinical depression is sharply on the rise in America and throughout the world. In America more than 23 million people suffer from anxiety illnesses. Many factors clearly contribute to these conditions, but daily exposure to media violence cannot be ignored as one such factor. Even before we have adequate data concerning the influence of heavy exposure to media violence in the development and persistence of depression, it would seem reasonable for parents to try to reduce its impact before turning to psychotherapy and medication. Given the stress on such coverage in local news reports, questions affecting policy choices of great importance for communities and the Nation inevitably receive correspondingly short shrift. The media's disproportionate emphasis on violence and mayhem contributes to skewing the perspective of unsuspecting viewers. And their skewed perspective may in turn lead to faulty reasoning and deliberation about choices in their own lives and in their communities, as well as in their views about collective policymaking with respect to crime and justice. Many are influenced by such a perspective to entertain views on prevention and punishment and on the severity of punishment criminals deserve, views that are not fully thought through in light of existing evidence. This skewing of perspectives and the corresponding effects on attitudes toward crime and justice are increased still further by another commonly noted effect of exposure to media violence: that of numbing, or desensitization. Like fearfulness, desensitization affects a far larger proportion of the public than risks of increased aggression. The effects on policymaking can be severe when widespread fearfulness and desensitization interact: when many in the public have developed heightened risk-aversion for themselves and their families, alongside lowered concern for injustices done to others. By itself, desensitization in the face of terrifying experiences is an utterly natural form of self-shielding. From infancy on, our reflexes guard us against being overwhelmed by the sensory impressions and images that surround us. Had we no such protection, we would be exposed to far more stimulation than we can cope with even at the most primitive sensory basis: it would be like having no skin. Gradually we learn to shield our emotions more selectively from the full impact of scenes or knowledge that might otherwise be too painful. A degree of numbing is a natural response and need not be problematic, unless it involves suppressing any impulse to empathy--the crucial ability to feel with and for others and to respond to their suffering. This numbing is especially likely to occur once enjoyment of the violence begins. For example, to enjoy running down farm animals and pedestrians in the game Carmageddon, players have to suppress empathy and any sense of trying to understand what their victims might be going through and learn not only pitilessness but the vicarious pleasure of maiming and killing. Such learned pitilessness presents great risks from a moral point of view. Empathy and fellow-feeling form the very basis of morality and of the first phase of moral response--that of perception. Without some rudimentary perception of the needs and feelings of others, there can be no beginnings of felt responsibility toward them. Because empathy is so basic to the capacity for moral growth, I believe that its erosion is the most far-reaching of the risks from entertainment violence and one that not only parents but all of us have every reason to take seriously. To be sure, many different influences promote or delay such moral growth. The trouble is, we hardly have an oversupply of empathy and fellow feeling. As a result, it is worth asking about any influence at all that wears away at empathy. As empathy wears down, those most strongly attracted by media violence find that they need higher and higher levels of violence on the screen to provide the stimulation they seek. They have developed a tolerance for violence and suffering that they once found intolerable. Does this tolerance carry over into daily life as well? According to the American Psychological Association, "Viewing violence increases desensitization to violence, resulting in calloused attitudes toward violence directed at others and a decreased likelihood to take action on behalf of the victim when violence occurs (behavioral apathy)."[12] More research is needed to learn about the extent of such changes in perception. We know that it can affect persons of all ages suffering from posttraumatic stress disorder and that more callous attitudes have been found among young males exposed to films containing sexual violence. Many experience "compassion fatigue" with respect to news reports about worldwide suffering. Is there a link between the two effects of media violence--increased fearfulness and desensitization--so that people become more concerned about risks to themselves and family and friends, even as they grow less concerned about risks to outsiders, foreigners, and other people they cannot bring themselves to care about in the same way? If so, what is at issue is an ominous combination: learned fearfulness with respect to self and learned pitilessness with respect to others. This leads me to a more speculative set of questions: Can these learned responses also influence perspectives on crime and justice? To what extent are they related to America's place in the world as first in the proportion of people it imprisons and alone among advanced industrial democracies in accepting capital punishment as a legitimate exercise in retributive justice--even, in some States, when applied to minors and the retarded? And might there not be a relationship between learned tolerance of media violence and the acceptance, in the richest country in the world, of homelessness and hunger that less affluent societies do not tolerate? Former New York Governor Mario Cuomo has spoken of the paradox of finding such inequity in one of the richest societies in the world. He has pointed to a "second city" of Americans who have not benefited from this country's prosperity, where the glitter does not show--the many Americans who live in poverty and have no health insurance. The disconcerting question for him, he has explained, is why there is not more of a public reaction to this state of affairs. There are surely many factors that contribute to the lack of public response to all the human misery of that second city--misery that would cause an outcry in all other advanced democracies and generate more forceful public policies in response. But might the public tolerance of such inequity be strengthened by a relentless media focus on violence and mayhem of all kinds? Might learned fearfulness for ourselves and learned pitilessness toward others help to skew our perspectives on justice in this regard as well as when it comes to crime? Promoting Reasoned Debate and Solutions In writing about violence, free speech, and censorship, I have been struck by the role of unexamined and often skewed perspectives in blurring perception and obstructing reasoned debate. I am reminded of the remark attributed to Winston Churchill in 1939 likening the Soviet Union to "a riddle wrapped in a mystery inside an enigma." In the controversies about media violence, I have come to think that we confront a somewhat different configuration: a conceptual paradox over how to view violence, wrapped in a tangle of factual conflicts about the effects of media violence, locked between the horns of an apparently insuperable policy dilemma regarding free speech. Once we examine the reasons for that sense of paradox and work to resolve the factual conflicts, we shall find that the policy dilemma is far from insuperable. Without infringing on our traditions of free speech, it is possible to undertake responses at every level, beginning with what we do in our own homes, and linking such measures to communitywide and societywide efforts. The media have a central role to play in these efforts and in asking how to avoid helping people become accustomed to violence and losing the capacity for pity. As yet there is nothing like the interconnected efforts at all levels, including in the media, currently under way to cope with the health threats posed by smoking and drunk driving. We are only beginning to have the nationwide debate that we need on this score and that other societies, such as Canada and Norway, have encouraged in their antiviolence campaigns.[13] As we seek alternatives to the dilemma without sliding into passivity, we can learn from initiatives already under way here and abroad. Examples abound of what individuals, families, schools, communities, and governments can do without infringing upon free speech, much less emulating authoritarian societies. These steps range from initiatives at the personal level to societal policies and international agreements. They may involve presenting families with more choice in quality television programs; organizing efforts by consumers, including PTAs and other parent organizations; wide-ranging press inquiries; and collective actions by screenwriters, actors, health professionals, and many others. The 1990s have seen a sharp increase in studies of media violence and self-help literature for persons who want to break out of burdensome viewing habits. During this period, many citizen campaigns have been undertaken, along with consumer boycotts and publicity directed at the producers and commercial sponsors of violent programming. There have been numerous legislative proposals for regulations concerning ratings, filtering devices, and "watershed hours," before which violent programming is controlled. The most effective responses to media violence increasingly make imaginative use of the new media. Technologies such as browsers and filters make it possible for consumers to be selective and to shut out unwanted programming. Many domestic and international groups distribute videos and other audiovisual materials concerning media literacy programs and other responses to media violence; a number have Web sites and e-mail addresses to exchange views and provide information. Just as those who produce and market media violence have long operated on a global scale, so consumers and concerned organizations now have the benefit of worldwide dialogue and collaboration. E.B. White was right to point, 60 years ago, to the remarkable changes likely to occur once TV brought the ability to see beyond the range of our vision. But he might now wish to revise his prediction that we would stand or fall with television. He might envisage, not so much an either/or contrast as a dynamic relation between the serious risks and the extraordinary opportunities that modern media have made possible. It is up to all of us to strike a better balance between the two than the one that burdens our society at present. Violence is taught, promoted, and glamorized; it can be unlearned, resisted, and deglamorized. Doing so will require strong leadership at all levels and resistance to all that contributes to skewing and narrowing our perspectives on crime and justice. Question-and-Answer Session Bernard S. Arons, Center for Mental Health Services, U.S. Department of Health and Human Services, Washington, D.C.: I am the director of a Federal agency that