Hot Spots of Crime and Criminal Careers of Places PDF

HOT SPOTS OF CRIME AND CRIMINAL CAREERS OF PLACES by Lawrence W. Sherman University of Maryland, College Park Abstract: The explanation of crime has been preoccupied with individuals and communities as units of analysis...

HOT SPOTS OF CRIME AND CRIMINAL CAREERS OF PLACES by Lawrence W. Sherman University of Maryland, College Park Abstract: The explanation of crime has been preoccupied with individuals and communities as units of analysis. Recent work on offender decision making (Cornish and Clarke, 1986), situations (Clarke, 1983, 1992), envi- ronments (Brantingham and Brantingham 1981, 1993), routine activities (Cohen and Felson, 1979; Felson, 1994), and the spatial organization of drug dealing in the U.S. suggest a new unit of analysis: places. Crime is concen- trated heavily in a Jew "hot spots" of crime (Sherman et aL 1989). The concentration of crime among repeat places is more intensive than it is among repeat offenders (Spelman and Eck, 1989). The components of this concen- tration are analogous to the components of the criminal careers of persons: onset, desistance, continuance, specialization, and desistance. The theoret- ical explanationfor variance in these components is also stronger at the level of places than it is for individuals. These facts suggest a need for rethinking theories of crime, as well as a new approach to theorizing about crime for public policy. On Valentine's Day 1989, in the capital of the U. S., 14 people were hit by gunshots. Four of them were shot on one block of Drake Place, S.E. One of the wounded, age 26, died; another, a 15-year-old girl, was found in her apartment with 11 firearms, including a machine gun, a shotgun and five bullet-proof vests (Horwitz and Goldberg, 1989). This one small block of Drake Place and its adjacent public housing complex had wit- nessed five murders in 1988. That was bad enough. But in the first seven weeks of 1989, Drake Place was the site of 4 murders and 14 bullet wouridings (Horwitz and Wheeler, 1989). Drake Place was a "hot spot" of crime. It was so hot that the police said they stayed away from it as much as possible, unless they got a call (Horwitz and Goldberg, 1989). It was so hot that every night after dark, one officer claimed there were gun shots all night long (Horwitz and Address correspondence to: Lawrence Sherman, Department of Criminology and Criminal Justice. Suite 2220 LeFrak Hall, University of Maryland, College Park, MD 20742. 35 36 Lawrence W. Sherman Goldberg, 1989). It was so hot that after the St. Valentine's Day Massacre, the Washington, D.C. police assigned a special patrol car to guard the block 24 hours a day (Horwitz and Wheeler, 1989). Drake Place may have been one of the hottest spots of crime in the U.S. in 1989. Most hot spots never have any shootings, let alone murders. But all hot spots, as defined here, are small places in which the occurrence of crime is so frequent that it is highly predictable, at least over a one year period. Within this definition, the phenomenon of hot spots appears to be wide- spread in the U.S. and elsewhere. CONCENTRATION OF CRIME IN FEW PLACES In Minneapolis, for example, an analysis of 323,000 calls to the police in 1986 found that a small number of hot spots produced most of the crime in the city (Sherman et al., 1989). Only 3% of the places produced 50% of the calls to which the police were dispatched. This concentration was even greater for the predatory crimes of robbery, criminal sexual conduct and auto theft: only 5% of the 115,000 street addresses and intersections in the city produced 100% of the calls for those, usually stranger-perpetrated, offenses. One cause of that concentration, of course, is the small number of those crimes relative to the large number of places. Even without any repeat locations, for example, all of the robberies could only have occurred at 3.6% of all places. But the fact is that with repeat occurrences, they occurred at only 2.2% of all places—a 40% reduction from the hypothetical number of places if there were no repeat locations. Domestic violence is even more concentrated by place of occurrence than robbery. While 21% of the places in Minneapolis could have had a domestic disturbance call without repeats, only 8.6% actually did—a 59% reduction. The concentration of crime in a few hot places seems even greater when it is compared to the concentration of crime among individuals (Spelman and Eck, 1989). Wolfgang and his colleagues" (1972) examination of a 1945 Philadelphia cohort found that 18% of the individuals produced over 50% of the arrests, compared to the 3% of places producing 50% of calls in Minneapolis. If we disregard the differences between the cities and the possible effect of 17 more years of observation in Philadelphia than Minneapolis, it is striking that the concentration of crime among places is six times greater than it is among persons. This comparison raises profound questions for both criminological theory and crime control policy. For if future crime is six times more predictable by the address of the occurrence than by the identity of the Hot Spots of Crime and Criminal Careers 37 offender, why aren't we doing more about it? Why aren't we thinking more about wheredunit, rather than just whodunit? THEORY Explanatory Focus As Nettler (1978) points out, criminological theory has been dominated by two units of analysis: individuals and communities. The two major questions for these theories are, first, why this person and not that one committed a crime, and, second, why there is more crime of type X in this society than that one, or in this society now than before? Empirical criminology began with Quetelet's (1842) observations of more crime in some parts of France than in others. It continued that tradition with Merton's (1968) anomie theory of societies, and Shaw and McKay's (1942) ecological perspective on neighborhoods. Throughout, investigations of crime patterns have relied on theories of human motivation. Even in the macro-level approaches, the ultimate target of explanation was the individual. Whatever social forces may affect crime rates, they do so in ways that cause more individuals to offend. The more explicit theories of individuals, such as differential association, labeling and control theory, also focused on this binary question of why the world is divided into two parts: the criminal and the non-criminal. Two developments have shaken this tradition in the past decade. One is the increasing evidence of variance among criminals in how often they offend, for how many years and with what level of seriousness. As Blumstein and his colleagues (1986) have pointed out, the crime rate of any society or community is not determined simply by the percentage of persons in that society who are committing crimes. Rather, the crime rate is the joint product of the number of offenders and the frequency with which they offend. The second development shaking the binary tradition in criminological theory is the rise of empirical work on how and where crimes happen. This work has taken various units of analysis, few of them individualistic. Ronald Clarke (1983) led a decade of research at the U.K. Home Office on "situational crime prevention," showing how small changes in situational opportunities could greatly affect the volume of crime. He and his colleagues' observations that a mandatory helmet law for motorcyclists greatly reduced motorcycle theft, for example, illustrates how crime rates can change independent of factors explaining individual criminality. Op- portunistic theft declined because would-be thieves carried no helmets 38 Lawrence W. Sherman with them. Altering the crime situation has been shown to reduce: post office robberies (Ekblom, 1987); prostitution (Matthews, 1992); thefts from shoppers (Poyner and Webb, 1992); library book theft (Scherdin, 1992); theft of and from cars (Poyner, 1992); fare evasion on a subway system (Clarke, 1993b); as well as other crimes (Clarke, 1992, 1993a, 1994). Further, there are striking differences in murder rates by type of place. Murder rates for Texas gas station workers, for example, are three times higher than for all other retail clerks (Davis, 1987). Finally, there is increasing evidence showing that some locations have many repeat vic- timizations while other places have no crime events (Pierce et al., 1986; Forrester et al., 1988; Sherman et al., 1989; Polvi et al., 1990). A theoretical approach to both situational and spatial differences in predatory crime was suggested ten years ago in Cohen and Felson's (1979) work on routine activities. This research pointed out a crime triangle, much like the three elements of fire. On one side of the triangle is the target (a person in personal crimes and an object in property offenses). Suitable targets are like fuel: you can't commit crime without them. Another side of the triangle is the offender. Motivated offenders, of course, get most of the blame for crime, just as heat gets much of the credit for fire. But in both cases, it's really all teamwork. The third side of the triangle is the place. Unless the offender gets together with the target at the same place and time, no crime can take place. Places are analogous to oxygen: unless heat and fuel are brought together with oxygen, combustion cannot take place. Originally, Cohen and Felson (1979) considered only one crime sup- pressor, guardianship. Guardianship protects targets from offenders when the offender and target are in the same place. Just as removing fuel—protecting it from the influence of heat—prevents fires, keeping offenders from targets suppresses crime. The Secret Service protection around the U.S. White House is an extraordinary example of guardianship. Felson (1986) later added another potential crime suppressor. By integrating routine activity theory with control theory (Hirschi, 1969), Felson was able to show that there are people in offenders' lives—parents, relatives, spouses, teachers and coaches, for example—who, when pres- ent, will prevent the offender from deviating. Felson called these people "handlers." They control crime just as removing heat controls crime, one can think of them as cooling offenders. Finally, Eck (1994) suggests that there may be a third class of crime controllers. While guardians act on targets and handlers act on offenders, the third group acts on places. The people who manage places—store clerks, life guards, park rangers, airline attendants, and countless oth- ers—also control crime by regulating the behavior of place users (see also Hot Spots of Crime and Criminal Careers 39 Eck and Weisburd, Felson, and Eck in this volume). Effective place management either allows offenders and targets to coexist at the same place without crimes, or keeps potential offenders out of places. The triangle suggests that a predatory crime requires much more than a criminally inclined individual. It suggests that victims and offenders' guardianship must converge in space and time in the absence of control- lers—guardians, handlers and managers—in order for crime to occur (Eck, 1994 and Felson in this volume). A change in the density or character of any of the sides of the triangle or controllers can change the risk of any crime happening. Though Cohen and Felson (1979) originally used the routine activities of individuals to explain societal crime rates, more recently routine activity theory has been used to explain differences in crime rates across places. Routine activities theory can go well beyond the binary focus of traditional criminology. Rather than just explaining why crime happens at some places and not others, routine activities may be able to account for the complex variation in the criminal careers of places. Dimensions of Criminal Careers Just how do criminal careers vary? There are at least six dimensions of the criminal careers of both individuals and places: 1. date of onset of any criminality; 2. the odds of recurrence of further crime, given one, two, or more prior crimes; 3. rate of frequency of crimes; 4. total career length, or number of years from first to last crime (sometimes described as the date of "desistence"); 5. average intermittency, or time period between any crimes at all; and 6. types of crimes committed, and the amount of specialization in certain types of crime (sometimes analyzed statistically as crime-switching). These basic dimensions of criminal careers of individuals have only recently been subjected to empirical analysis; for the criminal careers of places the first three are just being explored, and the last three are basically unknown. At the theoretical level, however, the situation is reversed: we may have a better idea of how to account for these variations in the criminal careers of places than we do in those of individuals. In fact, it is striking just how little theory there is to account for any of these career dimensions of 40 Lawrence W. Sherman individuals. Routine activities, on the other hand, generally suggests how to account for these variations in the criminal careers of places. 1. Onset The onset of crime in a geographic place can begin any time after its earliest human habitation, or its "birth" as a social space. More usefully, perhaps, it can begin after it takes on a distinctive character as a certain kind of place: a tavern, a church, a parking lot, a school. Consider this question: why would a place in which there had never been a robbery in seven years suddenly have one? Pure chance, perhaps. But perhaps also because of some changes in the routines of offenders, victims, guardians, handlers or managers. In November 1987, two blocks from the White House, a woman was robbed at gunpoint in an office building elevator. The woman was my wife, although that's not why I mention it. I mention it because in the seven years she worked in that relatively small building she had never heard of any crime in the elevators. The career criminal who robbed her had apparently never hit that building before. What explains the onset? Routine activities would suggest one or more possible changes, one of which actually occurred. Since its construction in the 1930s, the building's manual elevators had been served by elevator operators, 24 hours a day. In the spring of 1987, the elevators had been converted to automatic operation, thus changing place management. The elevator operators were laid off, and the elevators were left unattended. This argument is consistent with reports that burglary is less prevalent in apartment buildings with doormen than without them (Reppetto, 1974; Waller and Okihiro, 1978). It is also possible, of course, that the onset of robbery in this building was due to chance. 2. Recurrence In the 15 months since my wife's robbery, there have been no more reported crimes in that elevator, and no more robberies in that building. That is just what the Minneapolis data on recurrence would predict. An address with one robbery has only a 28% chance of having a second one within the remainder of that year. Even two robberies only bring the odds of recurrence up to 40%. But once a Minneapolis address has had three robberies, it is more likely than not to experience recurring robberies that year—a 58% chance of three robberies, and a 90% chance of 15 robberies. Hot Spots of Crime and Criminal Careers 41 There were ten addresses in 1986 that apparently had 15 or more robbery calls. How can recurring robberies be explained? Routine activities would suggest places with highly suitable victims, such as cash businesses open 24 hours a day. It would also suggest poor guardianship, such as one-clerk establishments. And it would suggest places with weak or absent man- agement. This type of explanation is not restricted to commercial places. Recent evidence about repeat residential burglaries in Canada (Polvi, et al., 1990) and in England (Farrell and Pease, 1993) suggest that poor guardianship of valuable targets may also help explain break-ins of apartments and homes. 3. Frequency The frequency of crime is very similar to the likelihood of recurrence of a specific crime type. We can use the term "frequency," however, to describe the number of times a year that any kind of crime occurs in a given place. These numbers can be enormous. Up to 810 calls to police in one year were recorded from a single address in Minneapolis—a large discount store, with many shoplifters arrested and 11 robberies reported in the parking lot. As routine activities theory predicts, places with larger populations appear more likely to have higher crime frequencies. The larger the population, the greater the chances of victims and offenders converging in space and time—which is why one analyst claims that the rate of crime in rural areas is actually higher than in cities, on a crimes-per-encounter basis. But population is only the starting point for routine activities, as a comparison of two places suggests. One is a Dallas hotel I analyzed as a defense witness in a lawsuit over inadequate security. The hotel looked much like the worst hot spots in Minneapolis. It had a high raw frequency of predatory crime, with 1,245 crime reports over a two-year period, 41 of them for violent crimes against persons. Yet the hotel covered 48 acres, with an estimated mean daily population of 3,000 guests, employees and visitors. The per capita robbery rate turned out to be 76% lower than the per capita robbery rate for the entire city (Sherman 1989).2 One possible cause for the "coolness" of this apparently hot spot was the much higher level of guardianship at the hotel. The ratio of patrol officers and private security officers to population was three times higher at the hotel than in the city as a whole. The density of patrol (private security) presence per acre was 63 times higher at the hotel than patrol 42 Lawrence W. Sherman (public police) presence city-wide. As Felson (1987) points out, growing inequality of security is characteristic of the modern metropolis. In contrast, the most violent bar on the Minneapolis hot-spot list had both the highest raw frequency of predatory crime and very high per capita rates of crime. With 25 robberies in one year—far more than the 3,000 person Dallas hotel—the bar had an estimated daily population of no more than 300. The bar had a robbery call rate of 83 per 1,000 persons—seven times higher than the call rate of 12 per 1,000 for the city's entire 1986 estimated population of 362,000. With 81 assaults, the per capita assault rate at the bar was 270 per 1,000, or more than one assault for every four persons in the bar over the year. Such an environment can reasonably be labeled as a dangerous place, in which individuals face substantially higher personal risks of criminal victimization than in the "average" place. The bar's crime frequency rate can be explained by high concentrations of drug dealers and prostitutes, as well as many suitably intoxicated or drug-influenced targets, and poor tavern management. The bartender and one waitress were subsequently charged with drug dealing, which sug- gests they had little interest in controlling illegal behavior at the place. 4. Career Length and Desistance That same bar had a very long career at the center of a vice block. Long before our 1986 study, the bar was implicated in a 1974 study of assaults around bars in Minneapolis (Fishbine and Joelson, 1978). Over a two year-period, 461 assaults had been reported within a half mile (.6) radius of the bar. The evidence suggests that the bar had been a hot spot of crime ever since its "birth." The criminal careers of people are often terminated after late adoles- cence, for reasons that are poorly understood. But the criminal careers of places show no such pattern. They seem more likely to be terminated by death—either from natural causes or capital punishment. The hot-spot bar in Minneapolis, for example, died a natural death in 1988 when it was torn down for urban renewal. Its resurrection in another location, however, was prevented by active police lobbying. Police have sought the death penalty for hot-spot buildings. For example, the famous New Briarfield apartments in the Newport News, VA demonstration of problem-oriented policing (Eck and Spelman, 1987) was torn down on police request after several years of police efforts to reduce its very high burglary rate. The criminal careers of hot spots can even end with vigilante "lynch- ings." Two Detroit men were recently acquitted of an arson they readily admitted. They burned down a crack house on their block, an abandoned Hot Spots of Crime and Criminal Careers 43 building that had generated frequent shootings and large crowds on their formerly peaceful residential street. The jury apparently thought the destruction of the property an appropriate form of punishment, or perhaps self-defense (Wilkerson, 1988). Other hot-spot careers are ended by incapacitation rather than capital punishment. To address the problems created by one persistent drug selling place, St. Louis police brought the condition of the property to the attention of the finance company holding the mortgage. "(T)he finance company foreclosed the mortgage, took possession and boarded up the property" (Hope 1994:24). Federal and local drug law has encouraged growing seizure of properties where crime occurs (or a discussion of place incapacitation, see Green in this volume). These seizures, however, raise the question of whether the criminal career of any place will ever end permanently, as long as the physical structure remains. 5. Intermittency Once the renovated drug apartments are turned over to the homeless, for example, they may become the scene of domestic violence, child abuse or even further drug dealing. They could become hot spots with entirely new populations, or they could go years without any reported offenses. Our Minneapolis analysis found many examples of high frequency levels of calls to police followed by months without further calls. What explains the intermittency of crime in places? The criminal habits of the occupants constitute one possible reason. But there are other explanations for intermittency, both external and internal to specific places. Eck (1994) claims that the economic difficulties of place owners makes it hard for them to control behavior in apartments. Owners of small apartment buildings in economically depressed areas have so little return that they cannot afford to evict tenants involved in drug dealing. In his article in this volume, Eck asserts that poor economic conditions in neighborhoods make places receptive to people engaged in illicit retail sales of all types. This is analogous to impaired immune responses in people, making them susceptible to opportunistic diseases. Other external reasons for intermittency could include changes in traffic flow from new bus routes or street engineering, a changing popu- lation in the surrounding neighborhood, or seasonal weather changes, especially in recreational areas like parks. These changes can affect the flow of targets as claimed by routine activity theory. But the changes may 44 Lawrence W. Sherman also affect the flow of offenders and alter their abilities to detect vulnerable targets (Brantingham and Brantingham, 1993). Internal reasons for intermittency are increasingly caused by changes in security measures. In 1979, singer Connie Francis won over $1 million dollars in a lawsuit against a hotel where she was raped. Her claim was that security measures were inadequate. In the past two decades, such lawsuits have been growing rapidly in many states. Losing such lawsuits, or the threat of them, has been a powerful inducement for many hotels and other businesses. One California hospital was sued over a nighttime shooting in its unguarded parking lot. Ever since then, there has been a guard stationed in a glass booth all night in the middle of the lot, which is now lit more brightly than daylight. For eight years after the shooting, there were no further violent crimes in that lot. 6. Crime Types The final question is whether certain places tend to specialize in certain types of crime. A perspective closely allied with routine activity theory is rational choice (Clarke and Felson, 1993). A rational choice perspective (Cornish and Clarke, 1986) asserts that particular situations structure offender choices. To the extent that place features enhance the ability of offenders to commit some crimes, but not any crime, in theory, places should display crime specialization. The evidence is still limited on this question. Weisburd, Maher and Sherman (1992) gave mixed results for places in Minneapolis. Disorder crimes were not highly correlated with serious predatory crimes, and burglaries of residences were negatively related to burglaries of busi- nesses. But domestic calls and residential burglaries were highly corre- lated, as were assaults and personal robberies. In Minneapolis, "flashers" (exposers) tend to specialize in parks. Cash businesses with high robbery rates also tend to have high shoplifting rates. The routine activities approach explains some of these patterns quite nicely. It may be obvious that domestic violence is concentrated in residences, but we should not sneeze at any strong correlation in this business. Nationally, the majority of robberies occur on public streets (Harlow, 1987). But in Minneapolis, 90% of robbery hot spots are found on a mere eight streets—the busiest boulevards in town, which also have the most bars (Linnell, 1988). At a very general level, patterns of concentration of suitable targets and motivated offenders at places with poor guardianship, absent handlers, Hot Spots of Crime and Criminal Careers 45 and inadequate management helps to explain onset, recurrence, fre- quency, career length, intermittency, and crime type. Future Research The challenge for future research is to elaborate the relationships among aspects of the criminal event triangle and crime controllers. How much change in guardianship, for example, is needed before crime fre- quency declines in a given place? How much, and what kind, of change in victim suitability increases the recurrence of robbery? And why do some parking lots experience more vandalism per car than others? Such questions may be answered through longitudinal cohort studies. But studying entire populations of places, as we have done in Minneapolis, may not be the most useful approach. Cohorts of certain kinds of places, such as taverns or convenience stores, might provide a more powerful design. By holding constant the basic function of a place, the design could address more of the specific reasons for variance in its criminal career. A study of 135 bars in the Vancouver, Canada area, for example, found an average rate of one violent act for every 13.5 hours of observation. But the bars varied widely, with up to three acts of violence per hour (Graham et al., 1980). Such enormous variance across taverns has been explained by four different approaches: patrons, management, behavior settings, and neighborhood. A patron hypothesis argues that high-crime bars are where high-crime people congregate. The amount of crime associated with each bar depends on the number of offenders in them (Fishbine and Joelson, 1978). A patron hypothesis is consistent with either increasing the number of targets or the number of offenders in the bars. A management hypothesis suggests that regardless of the neighbor- hood, management can influence who the patrons are and how they behave. Good management, this theory suggests, can control disorder and reduce the risks of crime on or about the premises (cf. Marsh, 1980). Eck (1994, and in this volume) makes this point regarding apartment build- ings. A behavior setting hypothesis (Barker 1968;Cavan 1966; Wicker 1979) takes all of these factors into account: the configuration of people, informal rules, space, time and objects associated with more or less violence. These three hypotheses are all derivable from routine activity. There is a forth hypothesis that is not. A neighborhood hypothesis suggests that bars in bad neighborhoods will generate more crime than bars in good neighborhoods, regardless of the criminality of the patrons. Patrons leaving the bars will be more likely to be attacked in bad neighborhoods 46 Lawrence W. Sherman (Skogan, 1986); otherwise peaceful persons will be more likely to challenge each other, due to a neighborhood's subculture of violence. Though not derivable from routine activity theory, a neighborhood hypothesis is not inconsistent with a routine activity approach. A neighborhood hypothesis may act with one or more of the other hypotheses. For example, bars in bad neighborhood may be more violent compared to bars in good neigh- borhoods, but a few of the bad-neighborhood bars may be exceptionally violent because of patrons, management or behavioral settings (see Eck, in this volume, for a description of how place and neighborhood theories may be integrated). Alternatively, the neighborhood hypothesis may be without any empirical support once these other hypotheses are taken into consideration. Though cross-sectional research can shed some light on the relative strengths of these hypotheses, longitudinal analysis will be more defini- tive. Longitudinal research is necessary, for example, to observe the effect of any changes in management while holding neighborhood factors con- stant. Even more useful would be longitudinal experiments, using random assignment to alter certain features of some violent bars and not others. POLICY The criminal careers of places are not only interesting, but are also very useful to policymakers. And the drug crisis has revived their long- dormant interest in hot spots. At one time, criminal justice officials were very sensitive to places. Baltimore judges used to know their neighborhoods well enough to track certain high-crime street corners. Once a corner got bad enough, they would "indict" the corner. Anyone found on the corner could be automat- ically arrested (Chaasnow 1989). While this practice was unconstitutional, other hot-spot statutes have been on the books for years. Laws against keeping a disorderly house, for example, were originally used to control brothels. Taverns have been highly regulated for centuries, since they have always been potential hot spots of crime. Certain crimes have had steeper penalties when committed in New York City's subways, or close to a school. What effect these statutes have had is unclear. But we have good prospects of finding out, for the focus on places solves a major problem of evaluation. The past two decades of policy have focused on large commu- nities. These so-called neighborhoods—as in "neighborhood" watch, or "community" policing—are not only too big to treat, they are also too large in size and too small in number to evaluate the effects of treatments. A policy focus on hot spots solves both problems. It provides manage- able, "bite-sized" challenges for police and other agencies. And it provides Hot Spots of Crime and Criminal Careers 47 enough of them to make controlled experiments feasible. A number of such experiments have been conducted or are underway. In Minneapolis we used places as the unit of analysis to examine the effect of patrolling hot spots (Sherman and Weisburd, 1995). We are completing a similar exper- iment in Kansas City to determine if drug raids reduce calls for service. In Jersey City, places have been used to find out if problem-oriented approaches reduce drug dealing (Weisburd and Green, 1995). Finally, in San Diego, Eck (1995) is analyzing the results of a randomized place experiment to determine the effects of inducing landlords to be better place managers. But the policy implications of hot spots are much broader. They go well beyond the police and sentencing policy. They suggest a role for many other agencies of government in the regulation and treatment of places. At a meeting with a local drug czar, for example, I suggested having the health department offer mobile drug treatment centers in combination with police crackdowns on open-air drug markets (Sherman, 1990). Public housing authorities have been increasing rigorous tenant screening, including criminal record checks as a way of controlling drug dealing (Weisel, 1990). And in one large privately owned apartment complex in Baltimore County, MD, the police worked with other local government agencies to obtain a U.S. Department of Housing and Urban Development grant to establish a child care, health and elderly services center for residents (Eck, 1995). The Current Limits of Policy There are no easy answers. Hot spots are often chronic and intractable. We cannot stop crime at every hot spot. We may not even succeed at most of them. A controlled experiment in problem solving at 500 repeat-call addresses in Minneapolis revealed how difficult it is to get other city agencies, and especially landlords, to do anything about their hot spots (Sherman et al., 1989). We have a long way to go in developing effective strategies. We also risk the dangers of excess. The Washington D.C. City Council considered a statute allowing evictions of persons who fail to tell police about apparent drug dealing in their buildings. Given the potential vengeance of drug dealers, such a policy seems grossly unfair. Police crackdowns on drug dealing hot spots have made indiscriminate arrests without probable cause (Cockburn 1988). Whether hot spot control can avoid such excesses is still unclear. Yet this much is clear. The criminal justice system in U.S. cities is swamped. It was already overwhelmed with cases before the recent 48 Lawrence W. Sherman onslaught of drug arrests. In many cities, many serious offenders go unpunished—not from liberalism, but from sheer exhaustion. The case for crime prevention has never been stronger. Prevention efforts focused on people as individuals might work, with enough time, money and commitment. But U.S. citizens seem to have little patience for it, at least with adults. Prevention efforts at the neighborhood level attract great support, in theory. But few Americans have enough time to really get involved in the crime problems of a large neighborhood. Where they do get involved is on their own block—especially if they start to perceive it as a hot spot. And that is the level at which government may be most effective, at least if the block is hot enough to justify a substantial investment of resources. Our own block was not. When we experienced a rash of vandalism and several muggings one summer, we traced the cause to a new disco in a Chinese restaurant on the next block. Complaints to police and the mayor's office produced no response. But then there was a stabbing in the disco. When our activist neighbor called the restaurant's insurance com- pany to let them know of the situation, they threatened to cancel the insurance. The disco closed the next week, and the criminal career of our block desisted, at least temporarily. The more serious—and well publicized—hot spots do receive govern- mental attention. That is what saved the citizens of Drake Place in Washington D.C., or at least reduced their high number of murders. That is what has already saved the children of Hanover Place and Mayfair Mansions, two other former District of Columbia hot spots that attracted heavy police presence. But not every hot spot can be a top priority. The modern triage of crime control has left many hot spots unattended. The challenge for policy, like that for theory, is to identify the most important aspects of routine activity theory in a wide range of places. But unlike theory, policy has to find the things that can be changed—easily, and without great expense. In the long run, crime may be no easier to control through places than through individuals. But most scientists, investors, and gamblers are attracted by predictability. As long as criminal places are far more predictable than criminal persons, that may be the where the public should bet its scarce crime control dollars. Acknowledgments: This paper is a revised version of the 1989 Gwynne Nettler Lecture, University of Alberta, Edmonton, Alberta, Canada. This Hot Spots of Crime and Criminal Careers 49 work was supported in part by grants 86-IJ-CX-0037 and 88-IJ-CX-0009 from the National Institute of Justice to the Crime Control Institute, Washington, D.C. Points of view or opinions expressed herein do not necessarily represent the official position of the U.S. Department of Justice. My thanks to Professor Leslie Kennedy for his great patience. NOTES 1. A. Reiss (1988) has pointed out that the crime rates are even more complicated: they reflect the number of active offenders times the number of crimes each offender commits, but only when reduced by the number of co-offenders involved in each crime. The majority of all U.S. robberies, for example, are committed by groups of two or more offenders (Harlow, 1987). A further complication is the number of victims or separate legal offenses involved in each criminal event. 2. The hotel won the suit. REFERENCES Barker, R. (1968). Ecological Psychology. Stanford, CA: Stanford University Press. Blumstein, A., J. Cohen, J.A. Roth and C A. Visher (eds.) (1986). Criminal Careers and "Career Criminals." Washington, DC: National Academy Press. Brantingham, P.J. and P.L. Brantingham (1981). Environmental Criminol- ogy. Beverly Hills, CA: Sage. (1993). "Environment, Routine, and Situation: Toward a Pattern Theory of Crime." In: R. V. Clarke and M. Felson (eds.). Routine Activity and Rational Choice. Advances in Criminological Theory, Vol. 5. New Brunswick, NJ: Transaction Books. Cavan, S. (1966). Liquor License: An Ethnography of Bar Behavior. Chicago, IL: Aldine. Chaasnow, H. (1989). Personal communication to the author from a former Maryland circuit court judge. Clarke, R.V. (1983). "Situational Crime Prevention: Its Theoretical Basis and Practical Scope." In: M. Tonry and N. Morris (eds.), Crime and Justice. Chicago, IL: University of Chicago Press. (ed.) (1992). Situational Crime Prevention: Successful Case Studies. Albany, NY: Harrow and Heston. (ed.) (1993a). Crime Prevention Studies, Vol. 1. Monsey, NY: Criminal Justice Press. 50 Lawrence W. Sherman (1993b). "Fare Evasion and Automatic Ticket Collection in the London Underground." In: R. V. Clarke (ed.), Crime Prevention Studies, Vol. 1. Monsey, NY: Criminal Justice Press. (ed.) (1994). Crime Prevention Studies, Vol. 2. Monsey, NY: Criminal Justice Press. and M. Felson (1993). "Introduction: Criminology, Routine Activity and Rational Choice." In: R. V. Clarke and M. Felson (eds.), Routine Activity and Rational Choice. Advances in Criminological Theory, Vol. 5. New Brunswick, NJ: Transaction Books. Cockburn, A. (1988). "Los Angeles Can't Win Drug War by Cracking the Whip." Wall Street Journal, May 5, (eastern edition). Cohen, L. E. and M. Felson (1979). "Social Change and Crime Rate Trends: A Routine Activity Approach." American Sociological Review 44:588- 605. Cornish, D. and R. Clarke (eds.) (1986). The Reasoning Criminal: Rational Choice Perspectives on Offending. New York, NY: Springer-Verlag. Davis, H. (1987). "Workplace Homicides of Texas Males." American Journal of Public Health 77:1290-1293. Eck, J.E. (1994). "Drug Markets and Drug Places: A Case-Control Study of the Spatial Structure of Illicit Drug Dealing." Doctoral dissertation, University of Maryland, College Park. (1995). Personal communication to the author. and W. Spelman (1987). Problem-Solving: Problem-Oriented Policing in Newport News. Washington, DC: Police Executive Research Forum. Ekblom, P. (1987). Preventing Robberies at Sub-Post Offices: An Evaluation of a Security Initiative. Home Office Crime Prevention Unit Paper No. 9. London, UK: Her Majesty's Stationery Office. Farrell, G. and K. Pease (1993). Once Bitten, Twice Bitten: Repeat Victimiza- tion and its Implications for Crime Prevention. Home Office Crime Prevention Unit Series No. 46. London, UK: Her Majesty's Stationery Office. Felson, M. (1986). "Linking Criminal Choices, Routine Activities, Informal Control, and Criminal Outcomes." In: D. Cornish and R. V. Clarke (eds.), The Reasoning Criminal: Rational Choice Perspectives on Offend- ing. New York. NY: Springer-Verlag. —— (1987). "Routine Activities and Crime Prevention in the Developing Metropolis." Criminology 25:911-932. (1994). Crime and Everyday Life: Insight and Implications for Society. Thousand Oaks, CA: Pine Forge Press. Fishbine, G.M. and M.R. Joelson (1978). Crime Impact Statements: A Strategy Suggested From The Study of Crime Around Bars. Minneapo- lis, MN: Minnesota Crime Prevention Center. Hot Spots of Crime and Criminal Careers Forrester, D.H., M.R. Chatterton and K. Pease (1988). The Kirkholt Burglary Prevention Demonstration Project Home Office Crime Prevention Unit Paper No. 13. London, UK: Her Majesty's Stationery Office. Graham, K., L. La Rocque, R. Yetman, T.J. Ross and E. Guistra (1980). "Aggression and Barroom Environments." Journal oJStudies on Alcohol 41:277-292. Harlow, C.W. (1987). Robbery Victims. Washington, DC: Bureau of Justice Statistics, Department of Justice. Hirschi, T. (1969). Causes of Delinquency. Berkeley, CA: University of California Press. Hope, T.J. (1994). "Problem-Oriented Policing and Drug Market Locations: Three Case Studies." In: R.V. Clarke (ed.), Crime Prevention Studies, Vol. 2. Monsey, NY: Criminal Justice Press. Horwitz, S. and J. Goldberg (1989). "13 Shot in District Within 24 Hours." Washington Post, February 16, p. 1. and L. Wheeler (1989). "Drug Market Intensifies Violence on Drake Place." Washington Post, February 17, p. D l. Linnell, D. (1988). 'The Geographic Distribution of Hot Spots of Robbery, Rape, and Auto Theft in Minneapolis." Masters thesis, University of Maryland. Marsh, P. (1980). "Violence at the Pub." New Society 52:210-212. Matthews, R. (1992). "Developing More Effective Strategies for Curbing Prostitution." In: R.V. Clarke (ed.), Situational Crime Prevention: Suc- cessful Case Studies. Albany, NY: Harrow and Heston. Merton, R.K. (1968). SocialTheory and Social Structure. New York, NY: Free Press. Nettler, G. (1978). Explaining Crime. 2d ed. New York, NY: McGraw Hill. Pierce, G.L., S. Spaar and L.R. Briggs (1986). The Character of Police Work: Strategic and Tactical Implications. Boston, MA: Center for Applied Social Research, Northeastern University. Polvi, N., T. Looman, C. Humphries and K. Pease (1990). "Repeat Break- and-Enter Victimizations: Time-course and Crime Prevention Oppor- tunity." Journal of Police Science and Administration 17:8-11. Poyner, B. (1992). "Situational Crime Prevention in Two Parking Facilities." In: R.V. Clarke (ed.), Situational Crime Prevention: Successful Case Studies. Albany, NY: Harrow and Heston. and B. Webb (1992). "Reducing Theft from Shopping Bags in City Center Markets." In: R.V. Clarke (ed.), Situational Crime Prevention: Successful Case Studies. Albany, NY: Harrow and Heston. Quetelet, A.J. (1842). A Treatise of Man. Gainesville, FL: Scholar's Facsimil- ies and Reprints (1969 ed.). 52 Lawrence W. Sherman Reiss, A.J., Jr. (1988). "Co-Offending and Criminal Careers." In: M. Tonry and N. Morris (eds.), Crime and Justice: A Review of Research, Vol. 10. Chicago, IL: University of Chicago Press. Reppetto, T.A. (1974). Residential Crime. Cambridge. MA: Ballinger. Scherdin, M.J. (1992). The Halo Effect: Psychological Deterrence of Elec- tronic Security Systems." In: R.V. Clarke (ed.), Situational Crime Prevention: Successjul Case Studies. Albany, NY: Harrow and Heston. Shaw, C.R. and H.D. McKay (1942). Juvenile Delinquency and Urban Areas. Chicago, IL: University of Chicago Press. Sherman, W.L. (1989). Violent Stranger Crime at a Large Hotel: A Case Study in Risk Assessment." Security Journal 1:40-46. Sherman, L.W., P.R. Gartin and M.E. Buerger (1989). "Hot Spots of Predatory Crime: Routine Activities and the Criminology of Place." Criminology 27:27-55. Sherman, L.W. and D. Weisburd (1995). "General Deterrent Effects of Police Patrol in Crime 'Hot Spots': A Randomized, Controlled Trial." Justice Quarterly (forthcoming). Skogan, W. (1986). "Fear of Crime and Neighborhood Change." In: A.J. Reiss, Jr., and M. Tonry (eds.), Communities and Crime. Crime and Justice: A Review of Research, Vol. 8. Chicago, IL: University of Chicago Press. Spelman, W. and J.E. Eck (1989). "Sitting Ducks, Ravenous Wolves, and Helping Hands: New Approaches to Urban Policing." Public Affairs Comment 35:1-9. Waller, I. and N. Okihiro (1978). Burglary: The Victim and the Public. Toronto, CAN: University of Toronto Press. Weisburd, D. and L. Green (1995). "Policing Drug Hotspots: The New Jersey City DMA Experiment." Justice Quarterly (forthcoming). L. Maher and L.W. Sherman (1992). "Contrasting Crime General and Crime Specific Theory: The Case of Hot Spots of Crime." In: F. Adler and W.S. Laufer (eds.), Advances in Criminological Theory, Vol. 4. New Brunswick, NJ: Transaction. Weisel, D.L. (1990). Tackling Drug Problems in Public Housing: A Guide for Police. Washington, DC: Police Executive Research Forum. Wicker, A.W. (1979). An Introduction to Ecological Psychology. Belmont, CA: Brooks-Cole. Wilkerson, I. (1988). "Crack House Fire: Justice or Vigilantism?" New York Times, October 22, p. 1. Wolfgang, M.E., R. Figlio and T. Sellin (1972). Delinquency in a Birth Cohort. Chicago, IL: University of Chicago Press.