PREVENTION: ENVIRONMENTAL AND TECHNOLOGICAL STRATEGIES

For many social scientists the physical environment is simply the space in which activity occurs. Others see this arena as a factor in affecting the behavior that it contains, but in so doing assign the physical context widely different levels of causal importance. Some take a direct, environmental determinist position, and argue that like any other variable, the built environment actively affects attitudes, conditions, and behaviors. In contrast, others argue that although environmental variables may be correlates of behavior, they do not directly affect it. This perspective sees ethnicity, income, life cycle, lifestyle, and other sociocultural factors as the ultimate generators of action, and a voluntary or forced affinity for specific types of physical environments as well. Thus, this selection model accepts that there is a strong association between environment and behavior, but suggests that assigning causality to it is spurious. Finally, a compromise position acknowledges causal importance in the architectural context, but stops short of determinism. This Darwinian model portrays physical settings as varying in the extent to which they permit or prevent behavior but locates the motivation for it elsewhere (see Michelson). All three of these positions can be found in explanations of criminal and delinquent behavior, and all have different implications for preventing it.

Environmental determinism and crime

Associations produced by post-factum analyses of rates of crime and delinquency and various aspects of the built environment, whether cities, neighborhoods, or types of dwelling, can almost always be explained by a selection argument. However, it is also possible that the physical environment causes those who inhabit it to feel and behave differently. Environmental sociologists and psychologists such as John Calhoun report a causal link between population density and social pathology, including sexual assault and extreme violence, among rats in laboratory experiments. Many social psychologists take the same position when they describe the impact of the urban environment on people, arguing that high population density can overload nervous systems, producing aggression or withdrawal (fight or flight) as a response. Debate on the extent to which population density affects criminal or other behavior continues (see Gove), but there is little doubt that density can affect humans. The real question is how high density has to go to produce specific results, especially criminal violence.

A more cognitive version of environmental determinism can be found in the argument that the physical setting provides "cues" to which actors will respond with either conforming or deviant behavior (see, for example, Brantingham and Brantingham, 1981). Graffiti, broken windows, and other signs of low surveillance in urban areas may elicit criminal behavior and stimulate out-migration of law-abiding residents as well. Philip Zimbardo's classic experiment, where a team of researchers timed how long it took for an unguarded automobile to be stripped on a New York City street, exemplifies this perspective. The study concluded that the car represents an opportunity as well as an indicator of social disorganization, motivating even normally conforming passers-by to engage in criminal behavior.

Earlier in the twentieth century, urban ecologists at the University of Chicago suggested that it is indeed social disorganization that links the urban environment to crime and delinquency. Louis Wirth, for example, argued that the size, density, and social heterogeneity of urban populations makes the city infertile ground for community, social organization, or control. This erodes traditional norms, and the freedom, anonymity, and hyperstimulation promote individual withdrawal as a defensive reaction, with bystander apathy as a consequence. (This urban indifference not only allows offenders to get away with pleasurable but illegal activities, but may even inspire individuals to engage in flamboyant nonconforming behavior in an effort to assert their individualism and be noticed.)

The Chicago School is not without its critics, and in other times and places urbanization is a negative not a positive correlate of crime (Gillis, 1989). However, recent cross-sectional research in North America shows that contemporary urban areas are indeed higher in rates of most types of crime than are rural regions. Further, urbanites are also more tolerant of differences and nonconformity, draw a sharper distinction between friends and strangers, and treat the latter with greater indifference than do their rural counterparts (see Gillis, 1995). Whether these variables actually intervene between the physical environment and crime is uncertain. Disorganization theory has been reformulated in terms of network analysis, and supported empirically with data from Great Britain (Sampson and Groves). However, this research does not support the optimism of the early 1960s that clearing slums and improving the physical environment will actually combat crime as well as urban blight. This cannot happen when criminogenic demographic, economic, and cultural conditions continue (see Wilson).

Selection

The idea that rates of crime and delinquency are concentrated in some spatial configurations is widely accepted, especially at the macro level (see Gillis, 1996). However, whether these associations between environment and behavior are causal, or the product of selection, is less clear. For example, cities depend on migration from rural areas for their survival and growth because of the relative infertility of urban populations. Since rural migrants to cities are not randomly selected, people in some types of circumstances are more likely to migrate than are those in others. It is possible that people who leave rural areas for the bright lights of the city are also less bound by family and other local ties, more inclined to take risks, and more likely to become involved in crime than are those who stay put. The effect of this selection would be to lower crime rates in rural areas while simultaneously increasing them in urban areas, producing a positive correlation between urbanization and rates of crime. In fact, cities or neighborhoods may even attract excitement seekers or criminals from rural areas who travel to cities only to engage in crime (Gibbs and Erickson).

The attraction to large urban areas of thrill-seekers and people in search of the unusual is easy to understand. Cities are not simply large versions of small towns. They are different. Small, dispersed populations, such as those in rural areas, villages, and towns cannot support markets for highly specialized goods and services. In contrast, the large, diverse populations of cities can harbor a wide range of atypical and criminal interests that merge as deviant subcultures. This may explain why cross-sectional analyses show that the rate of criminal activity in the U.S. not only varies between rural and urban areas, but directly with the current size of metropolitan populations as well (Fischer).

Moreover, once areas become known for the specialized goods and services which they provide, both short- and long-run migration patterns are likely to be affected, further enhancing the reputation of the area. Even Calhoun's rats who were able to nest in low-density sectors frequently ventured into the high-density areas of the site, apparently for excitement. This is consistent with the idea that specific environments not only operate as entertainment areas, but as "deviance service centers." Although they can be as large as states (e.g., Nevada, when it first legalized gambling, prostitution, and quick divorces), or metropolitan areas such as Atlantic City, deviance service centers are most often a section of a city, a peripheral neighborhood, or a suburb. One of the most famous of these centers was Southwark, directly across the Thames from London. In Shakespeare's time (the turn of the seventeenth century) he and his Globe Theatre were situated in Southwark, providing popular adult entertainment. For additional amusement, an abundance of drinking establishments, hotels, brothels, and prostitutes accommodated patrons who were unable or unwilling to make it back to London before curfew. Jails, including the original "Clink," housed more disruptive revellers, and Winchester Cathedral was not only available for repentance, but for licensing prostitutes. Thus, without being inside the physical, political, or moral boundaries of the city, Southwark provided both legitimate and illegitimate entertainment for the people of Elizabethan London (Gillis, 1995).

The out-migration of conformists can also distribute crime and delinquency differentially in urban areas. Some suburban neighborhoods are not only exclusive to those who can afford to live in them but, as "gated communities" with private security, even restrict access to residents and authorized visitors. Even entire suburban communities may be restricted, as "edge cities" near but off limits to the residents of core areas. On the other side, there are neighborhoods and even specific cities that have been not only abandoned by employable residents, but by industry and legitimate economic opportunities as well (Wilson). In this respect, the extent to which gated communities and edge cities enjoy low rates of crime is directly related to increasing crime rates in inner cities.

The policy implications here are both social and legal. Disorganized areas of inner cities require assistance at both structural and cultural levels to escape the downward spiral in which they are caught. Legitimate economic opportunities and the motivation and ability to take advantage of them are lacking. If employment opportunities exist, they are likely to be illegal, and this may or may not be viewed as a problem in neighboring jurisdictions. With respect to deviance service centers, activities that are considered deviant or criminal when committed in one section of a jurisdiction need not be prosecuted as such in others. Elizabethan London, contemporary Amsterdam, and other cities containing deviance service centers represent efforts to contain, rather than constrain, behavior that is regarded as undesirable in the wider population. This is accomplished by permitting bounded zones of the urban environment to contain outlets for both providers and consumers of disreputable activity. Whether life is better or worse for those within the service sectors is both an empirical and ethical question.

Opportunities, constraints, and defensible space

Interaction models specify conditional relationships, or scope limiters, whereby other variables soften the determinism of an independent variable. In the case of population density, for example, individuals can be seen as more or less equipped by their personalities or the capital provided by their cultures to cope successfully with crowding (Gillis, Richard, and Hagan). This could account for the inconsistent results reported by more deterministic additive equations examining the relationship between population density and criminal behavior.

Newman's Defensible Space (1972) presented an interaction model applied to the impact of the architectural environment on crime, from an environmental opportunity/constraint perspective. Newman suggested that specific aspects of urban design create unsupervised locations by inhibiting or preventing surveillance and control in neighborhoods, and even within apartment buildings. High walls around buildings, underground parking areas, long corridors, and secluded spaces such as stairwells provide opportunity for deviant or criminal activity. This is because these locations are concealed from formal policing on streets. At the same time, since this space lies outside their own dwelling units, residents do not feel obliged to maintain surveillance on these areas either. Distance from home is inversely related to willingness to intervene (Gillis and Hagan, 1990). As with the perspective developed by the Chicago School, then, Newman's argument accounts for the location of criminal activity, and suggests that structural barriers (in this case architectural) can block the local social ties that are community, permitting increased rates of crime and delinquency. In this respect, Pruitt-Igoe, the infamous public housing project in St. Louis, exemplifies how not to design.

The scale of neighborhoods, projects, and buildings may also contribute to the degree of disorganization and rate of mayhem. Pruitt-Igoe contained a concentration of disadvantaged people. This may have also provided a critical mass of young, criminally disposed males, who were not only able to act as individual offenders, but who coalesced as gangs, creating criminal subcultures with a program of recruitment. In line with this, juvenile residents of high-rises report greater use of illicit drugs than do their counterparts in low-density housing (Gillis and Hagan, 1982). In any case, since the debacle and destruction of Pruitt-Igoe, many communities try to avoid large-scale public housing projects, and follow a program of dispersal instead of concentration.

Interestingly, Newman attaches great importance to the presence of graffiti as an indicator of lapsed control and portentous of future crime. Declines in crime in cities such as New York during the late 1990s are typically explained by civic officials as the result of acting against minor as well as major offenses, including graffiti. This suggests that environmental cues may elicit illegal activity, as noted earlier, but the actual causal sequence here is uncertain. It is also noteworthy that the interaction model can be interpreted as a statement that a causal sequence will not hold under all conditions. Thus, increasing the lighting on streets will not reduce mugging any more than a vault will reduce bank robberies if no offenders are present in the population. Similarly, changing traffic flow by establishing a one-way street will not reduce prostitution in areas where there are neither prostitutes nor johns. On the other side, if Pruitt-Igoe and its surroundings had contained older women instead of young, truly disadvantaged males, the project's crime rate would have been low. The interaction argument is that environmental design can discourage informal social control, and that this will in turn result in increased criminal activity, but only in populations with a propensity to offend in the first place. Thus, criminogenic architecture and planning will have no effect on the crime rate of a population of conformists. However, in a population of risk takers, a crime prevention strategy based on environmental design and increased informal surveillance would make a difference, especially in urban areas. In rural areas and small towns, traditional occupations for husbands and wives revolve around the home, where surveillance is maintained and passersby noticed. In contrast, occupations and routine cosmopolitan activities in modern urban areas frequently draw adults away from home, leaving an unattended store of valuable goods, which may attract burglars (Cohen and Felson), even from the hinterland. In this way, specific areas may produce criminal opportunity more than offender personalities, and by so doing inflate the rate of crime in those environments.

On a broad level, commercial neighborhoods containing too few residences must depend solely on policing for after-hours surveillance. The planning implication here is to avoid desertion after 6 p.m. by zoning for a mix of residential and commercial land use.

Social reaction and offensible space

Disorganization theory and the opportunity-constraint argument suggest that the built environment can reduce community control and inflate crime rates. Directly opposed to this, a social reaction perspective alerts us to the possibility that variation in official rates of crime and delinquency may measure different patterns of enforcement and recording of offenses more than in patterns of criminal behavior. In the case of the physical environment, the likelihood of detection, apprehension, and recording of offenses may be higher in some locations than in others. Rather than seeing cities as places to commit crime with impunity, then, the reaction perspective suggests that because of their high density and more extensive surveillance, urban areas may actually increase the likelihood of criminal activity being detected and apprehended than in the countryside. Although many urban neighborhoods may look disorganized, especially to middle-class social scientists, the communities which they contain may be alive and well, with strong patterns of surveillance and intervention. Moreover, police are more likely to patrol high-density areas and detect any offenses that occur within them. Juveniles in high-density locations in urban areas corroborate this by reporting higher levels of police contact than do adolescents in low-density neighborhoods (Hagan, Gillis, and Chan). This may also occur on the macro level. Historically, the ratio of police to population is higher in urban than in rural areas (Gillis, 1989). Further, bureaucratized urban police forces are more likely to process and record the offenses that they detect than are their rural counterparts. Between increased informal surveillance of residents and formal control of urban policing, then, higher official rates of crime in cities may reflect heightened bureaucracy more than higher rates of criminal activity. Thus, a feedback model may best capture the relationship between crime and space.

A. R. Gillis

See also Crime Causation: Sociological Theories; Ecology of Crime; Guns, Regulation of; Police: Private Police and Industrial Security.

BIBLIOGRAPHY

Brantingham, Paul J., and Brantingham, Patricia, eds. Environmental Criminology. Beverly Hills, Calif.: Sage, 1981.

Calhoun, John B. "Population Density and Social Pathology." Scientific American 206 (1963): 139–148.

Cohen, Lawrence E., and Felson, Marcus. "Social Change and Crime Rate Trends: A Routine Activities Approach." American Sociological Review 44 (1979): 588–608.

Cohen, Lawrence E., and Machalek, Richard. "A General Theory of Expropriative Crime."American Journal of Sociology 94 (1988): 465–501.

Fischer, Claude S. The Urban Experience, 2d ed. New York: Harcourt, Brace, Jovanovich, 1984.

Gibbs, Jack, and Erickson, Maynard. "Crime Rates of American Cities in an Ecological Context." American Journal of Sociology 82 (1976): 605–620.

Gillis, A. R. "Crime and State Surveillance in Nineteenth-Century France." American Journal of Sociology 95, no. 2 (1989): 307–341.

——. "Urbanization." In New Society. Edited by R. J. Brym. Toronto: Harcourt Brace, 1995. Pages 13–40

——. "Urbanization, Crime, and Historical Context." In Criminological Controversies. Edited by J. Hagan, A. R. Gillis, and D. Brown-field. Boulder, Colo.: Westview Press, 1996. Pages 47–74.

Gillis, A. R., and Hagan, John. "Density, Delinquency and Design: Formal and Informal Control and the Residential Environment." Criminology 19, no. 4 (1982): 514–529.

——. "Delinquent Samaritans: Social Conflict, Network Structure, and the Willingness to Intervene." Journal of Research in Crime and Delinquency 27, no. 1 (1990): 30–51.

Gillis, A. R.; Richard, Madeline A.; and Hagan, John. "Ethnic Susceptibility to Crowding: An Empirical Analysis." Environment and Behavior 18, no. 6 (1986): 683–706.

Gove, Walter. "Review of Household Crowding and Its Consequences by Edwards et al." American Journal of Sociology 101, no. 4 (1996): 1137–1138.

Hagan, John; Gillis, A. R.; and Chan, Janet. "Explaining Official Delinquency." The Sociological Quarterly 19, no. 3 (1978): 386–398.

Michelson, William. Man and His Urban Environment, 2d ed. Reading, Mass.: Addison-Wesley, 1976.

Newman, Oscar. Defensible Space. New York: Macmillan, 1972.

Newman, Oscar, and Franck, Karen A. Factors Influencing Crime and Instability in Urban Housing Projects. Washington, D.C.: U.S. National Institute of Justice, 1980.

Sampson, Robert J., and Groves, W. Byron. "Community Structure and Crime: Testing Social Disorganization Theory." American Journal of Sociology 94 (1988): 774–802.

Wilson, William J. The Truly Disadvantaged: The Inner City, the Underclass, and Public Policy. Chicago: University of Chicago Press, 1987.

Wirth, Louis. "Urbanism as a Way of Life." American Journal of Sociology 44 (1938): 3–24.

Zimbardo, Phillip. "A Field Experiment in Auto Shaping." In Vandalism. Edited by Colin Ward. London: Architectural Press, 1973.