Sociological paper---Philadelphia crime paper
fyaqkhe19t512 sources/A Second Chance and the Right to Vote - The New York Times.pdf
http://nyti.ms/1q64zJW
The Opinion Pages | E D I T O R I A L | L O C K E D O U T O F S O C I E T Y
A Second Chance and the Right to Vote By THE EDITORIAL BOARD MAY 7, 2016
Republican legislators in Virginia are threatening to sue Gov. Terry McAuliffe to block his executive order restoring voting rights to more than 200,000 residents who have completed their felony sentences. The lawmakers have no good legal case, and worse, such a suit would be affirming Virginia’s racist history.
Virginia is one of just four states — along with Iowa, Florida and Kentucky — that continue to impose a lifetime voting ban on people convicted of felonies. In recent years, both Democratic and Republican governors have worked to lift this burden, either by streamlining the application process for individuals or trying to restore rights to specific classes of people, like those convicted of nonviolent felonies.
Mr. McAuliffe took a bolder and more just step last month by restoring those rights to all people with felony convictions. Republican lawmakers say this action “overstepped the bounds of his authority and the constitutional limits on executive powers.”
They fail to point to any provision in the state’s Constitution or laws to support this claim, because there isn’t one. Virginia’s Constitution explicitly empowers the governor “to remove political disabilities consequent upon conviction” for felonies. It places no qualifications or limitations on that power.
The executive power of clemency is generally very broad, so state constitutions are clear when they intend to restrict it — say, by requiring an independent board to sign off on a governor’s decision. Virginia’s Constitution places some limits on the exercise of other forms of clemency, such as granting pardons or canceling fines, but
none on restoring the right to vote.
Mr. McAuliffe’s order is also an important moral step, too long delayed. Virginia’s voting ban, like most of the others that collectively disenfranchise about six million Americans, is a 19th-century relic rooted in racism — a direct reaction to the passage of the 15th Amendment, which guaranteed African-Americans the right to vote.
Politicians in Virginia were blunt about their motivation. In 1902, when Virginia’s voting ban was expanded at the state’s constitutional convention, Carter Glass, a state senator, said its purpose was to “eliminate the darkey as a political factor in this state in less than five years, so that in no single county of the Commonwealth will there be the least concern felt for the complete supremacy of the white race in the affairs of government.”
Before Mr. McAuliffe’s order, one in five black Virginians was permanently barred from voting because of a past felony conviction. Aside from its profound racial disparity, the lifetime voting ban has served to marginalize people who have already paid their debt to society.
Virginia’s ban is an embarrassment to the state and to the country. Rather than fighting to preserve it or grumbling about Mr. McAuliffe’s alleged political motivations, Republican lawmakers should be working to win the votes of the hundreds of thousands of Virginians who have been unfairly barred from exercising democracy’s most fundamental right.
Follow The New York Times Opinion section on Facebook and Twitter (@NYTopinion), and sign up for the Opinion Today newsletter.
A version of this editorial appears in print on May 8, 2016, on page SR10 of the New York edition with the headline: A Second Chance and the Right to Vote.
© 2016 The New York Times Company
12 sources/Anderson et al.pdf
Clubbing Masculinities and Crime: A Qualitative Study of Philadelphia Nightclub Scenes
Tammy Anderson,1 Kevin Daly,1 and Laura Rapp1
Abstract
The purpose of our article is to explore the relationship between masculinities and crime within the hip-hop (HH) and electronic dance music (EDM) nightclub scenes in Philadelphia. Given extant theory and research showing gender is a situated performance, the social context of the nightclub setting offers an important opportunity to contribute to the ever-growing masculinities and crime literature because it is an understudied setting populated by atypical offenders. Direct observation of 33 club events and interviews with 24 male clubbers yielded three important patterns: (a) Men with consistently high masculinities (hypermasculine types) reported the most frequent involvement in nightclub crime, (b) men with consistently low masculinity scores reported the least involvement, and (c) men with variable masculinity scores put on a more hypermasculine identity while clubbing, leading them to engage in nightclub crime. Contextual factors, such as excessive alcohol use, heightened sexuality, competitiveness, and commercialism, explain this more nuanced relationship between masculinity and crime.
Keywords
masculinity; nightclubs; ethnography; crime; assault
Urban nightclubs are an important and understudied location in criminological research. Victimization data for 2005 indicate that a roughly 22% of violent victimiza- tions took place during leisure activities, including those at bars and nightclubs (Bureau of Justice Statistics, 2006). Recent research has found that bars and night- clubs are significantly more prone to physical assault (Graham, Osgood, Wells, & Stockwell, 2006; Graham & Wells, 2001; Graham, Wells, & Jelley, 2002; Hopkins, 2004; Lipton & Gruenewald, 2002), sexual assault and harassment (Buddie & Parks, 2003; Parks, Miller, Collins, & Zetes-Zanatta, 1998), as well as drug use and sale (Anderson, Kavanaugh, Bachman, & Harrison, 2007; Cohen, Gorr, & Singh, 2003).
Articles
Feminist Criminology 4(4) 302–332
© The Author(s) 2009 Reprints and permission: http://www. sagepub.com/journalsPermissions.nav
DOI: 10.1177/1557085109343676 http://fc.sagepub.com
1University of Delaware, Newark
at TEMPLE UNIV on June 5, 2016fcx.sagepub.comDownloaded from
Anderson et al. 303
Such settings are new hot spots of crime and likely challenge existing theory in explaining offending patterns. One reason for this is that nightclubs have unique cul- tural and organizational aspects that differ from standard criminal settings. Second, these contextual factors define norms, shape identity, and invite novel behaviors across divergent demographic groups, especially males and females, but also by race, ethnicity, and social class. To date, there has been little work exploring how gendered performances and situational aspects of crime interact to produce crime in such dis- tinctive and popular social settings.
The purpose of this article is to unpack the relationship between masculinity and crime by considering offending patterns in an understudied social context: urban nightclub events. More specifically, our article explores how gendered performances are situated in social contexts—electronic dance music (EDM) and hip-hop (HH) events—and emit criminal outcomes. Understanding offending in these unusual social contexts follows from simultaneously considering both the situational aspects of gender and crime. By investigating the masculinity and crime relationship in urban nightclubs, we contribute to two main debates in criminology today. The first debate has to do with how fundamental forms of social organization, like gender, impact crime. The second pertains to the role of social context in shaping criminal outcomes. In short, by considering how masculinities materialize at nightclub events, we are able to offer new insights about how gender impacts crime across diverse social contexts. Before turning to our analysis, we review the literatures on masculinity and crime and situational crime at bars and nightclubs.
Masculinities and Crime Throughout its history, criminology has been slow to understand how fundamental forms of social stratification, for example, race, ethnicity, class, and gender, influence crime. Early attempts to address this shortcoming included disaggregating offenders into population subgroups to see whether major theories of crime explained offending patterns. Today, studies find criminological theories explain men’s or boy’s offending better than women’s or girls’ or vice versa (see Anderson, Miller & Ousey, 2008, for a review).
If theories of crime do not equally explain women and men’s offending, then per- haps there is something about being a boy or man, that is, masculine identification, or being a woman or girl, that is, feminine identification, that acts as a risk or protective factor of crime (Cohen & Harvey, 2006). This latter point calls attention to the role of gender identification in theories of criminality. Crime and masculinity scholars have taken up this issue, contending that masculinity, especially hypermascu- linity, drives male offending (e.g., Beirne & Messerschmidt, 2007; Connell & Messerschmidt, 2005; Messerschmidt, 2004; Mullins, 2006).
Over time, the growing literature on masculinity and crime has addressed the role of masculinity in terrorist acts (Hafez, 2007; Kaplan, 1978), intimate partner violence or sexual assault, prison misconduct (Lutz & Murphy, 1999; Man & Cronan, 2001),
at TEMPLE UNIV on June 5, 2016fcx.sagepub.comDownloaded from
304 Feminist Criminology 4(4)
and a range of other delinquent and criminal acts (e.g., drug dealing [Bourgois, 1996; Hutton, 2005], assault [Peralta & Cruz, 2006; Whitehead, 2005], vandalism [Harlan & McDowell, 1980; Messerschmidt, 1994]). Yet, how criminologists conceptualize mascu- linity remains underdeveloped (Cohen & Harvey, 2006), and few studies have explored the masculinity and crime relationship in leisure settings, including urban nightclubs.1
An understanding of the relationship between masculinity and crime in urban nightclubs can begin with a few words about feminist theory and gender. According to Kennelly, Merz, and Lorber (2001), gender is a social institution vitally connected to the family, law, economy, and culture. Within these institutions, gender performances occur and are either reinforced through norms or challenged by gender troublemakers (Lorber, 2000). Gender norms are enforced through “informal sanctions of gender- inappropriate behavior by peers and by formal punishment or threat of punishment by those in authority should behavior deviate too far from socially imposed standards for women and men” (Lorber, 1994, p. 18).
Although a dualistic masculine/feminine idea of gender is propagated in mainstream culture, research shows neither can be universally defined. Instead, masculinity and fem- ininity are shaped by the social processes and, importantly, the social context within which they exist. In fact, sociologists have shown that gender is socially constructed over time (Brickell, 2005, 2006; Copes & Hochstetler, 2003), making it unique to and fluid between certain social contexts and divergent situations. This is what Butler (1999) is referred to when she called gender a performance.
One particular gender performance that has been consistently linked to crime is hypermasculinity. Hypermasculinity describes men who adhere to amplified expres- sions of traditional gender roles (Kreiger & Dumka, 2006; Mosher & Tomkins, 1988; Murnen & Byrne, 1991). Hypermasculinity is linked to antisocial behavior, the exces- sive use of drugs and alcohol, and to the beliefs that violence is manly and danger is exciting (Kreiger & Dumbka, 2006; Mosher & Tomkins, 1988). Other characteristics attributed to hypermasculine men are having unsympathetic sexual attitudes toward women, suppressing emotions labeled as weak, such as sadness and fear, the domina- tion of others, and the approval of sexual aggression. Overall, hypermasculinity is constructed through negative and antisocial behaviors and attitudes (Kreiger & Dumka, 2006).
Gender performances can be influenced by the social setting, the form of interaction, and the dimension and appearance of those performing the act (Messerschmidt, 2004; Mullins, 2006). From this viewpoint, masculinity might produce crime because certain situations and environments call for hypermasculine performances, identities, and behaviors (West & Zimmerman, 1987). In this sense then, typologies of offenders become difficult to identify as anyone might be at risk for offending as gender is per- formed and relative to time and place (Copes & Hochstetler, 2003; Tomsen, 1997).
Fusing these ideas with the situational approach to crime has led scholars to sur- mise that masculine identities acquire meaning from the locations where individuals interact and fluctuate over the course of time (Connell & Messerschmidt, 2005). Masculinity is sensitive to social context and unique to situations (Schwalbe & Wolkomir, 2001) as locations (geographical and symbolic) and historical moments
at TEMPLE UNIV on June 5, 2016fcx.sagepub.comDownloaded from
Anderson et al. 305
have unique actors, norms, and constraints. Men adhere to socially constructed gender norms such as being competitive and unafraid to take risks, displaying these traits in settings where they are valued. When people lack socially acceptable ways of present- ing their masculinity, they may turn to crime for an opportunity to assert their masculinity.2 The assertion of masculinity, barring traditional outlets, has been linked to theft (Copes & Hochstetler, 2003), punking/bullying (Phillips, 2007), child homi- cide (Alder & Polk, 1996) and bar violence (Tomsen, 1997).
The discovery of masculinity as situational and temporally dependent makes it harder to theorize the influence of gender on crime. Rather than relying on a fixed measure of masculine traits, research would need to construct a sort of calculus that fused environmental triggers with masculine identity markers. By considering charac- teristics of social contexts, our analysis moves toward such a fusion.
Social Context and Crime To date, two broad criminological traditions (social disorganization theory and routine activities theory) have speculated about environmental or contextual influences on crime. However, neither considered gender as a sort of mediator between contextual factors and crime. To begin, social disorganization theory showed that crime patterns varied across neighborhoods, a type of social setting. Researchers (e.g., Shaw & McKay, 1942) found that neighborhoods with high crime rates also shared high levels of residential instability, ethnic heterogeneity, and poverty. This more structural approach forced criminology to consider that explanations of crime were sensitive to environmental (socioeconomic and cultural) conditions. Yet, it offered nothing at the individual level or about gender. In other words, although rates of offending might be vulnerable to forces unique to social settings, would individual offending be suscep- tible to such situational or contextual factors as well? If so, how should they be conceptualized and what role might gender play?
More recently, theories on crime situations have been advanced by the examination of crime hot spots—that is, social locations that are particularly prone to instances of crime and victimization (Sherman, 1995; Weisburd, 2002). Particularly attractive locations for crime were those where motivated offenders (e.g., drug dealers), suitable targets (e.g., persons under the influence of alcohol and drugs), and a lack of effective guardians (e.g., insufficient social control in and around such venues) converged in time and space (Cohen & Felson, 1979). Routine activities theory has been used to explain crime related to drug and other illegal marketing in urban neighborhoods as well as with alcohol-related phys- ical and sexual assault in bars and nightclubs (e.g., Roncek & Meier, 1991), a point which is highly relevant to our study of clubbing masculinities and crime. Yet, no work in this tradition has conceptualized gender traits or identities as influencing motivation, shaping the desirability of targets, or affecting the capability of guardians. Routine activities research has more often studied gender as an outcome, such as the victimization of a particular group of men or women to likely follow from its three main independent vari- ables. It has less often utilized gender to explain criminal outcomes.
at TEMPLE UNIV on June 5, 2016fcx.sagepub.comDownloaded from
306 Feminist Criminology 4(4)
This relative inattention to gender, social context, and crime by criminologists opens the door to our research objective. Furthermore, the nightclub setting and our methodological approach provide us with a useful strategy to explore the relationship between masculinities, social context, and nightclub crime.
Nightclubs as hot spots of crime. Today, cities such as New York, Philadelphia, and Washington D.C. are entertainment sites with thriving nightclub-centered leisure economies, where youth and young adults interact. Nationally, at least two nightclub- based subcultures have emerged in the urban corporate entertainment industry: the hip-hop (HH) scene and the EDM scene. Both cater to youth and young adults from diverse racial and ethnic backgrounds (Bennett, 2001). Currently, media accounts sug- gest, however, that such nightclub events might be hot spots of criminal activity. Reports from New York (Berkey-Gerard, 2001) cited death, overdose, violence, and murder among club drug users at nightclubs. In addition, Holmberg (2001) claimed that drug using and selling takes place at most nightclubs and restaurants catering to young adults in major U.S. cities.
Nightclubs housing HH and EDM events have come under scrutiny by local, state, and federal authorities as being noisy, socially disruptive breeding grounds for drug use and sales, and as sites conducive to violent crimes such as sexual and physical assault (Johnson, 2001; Mosler, 2001; U.S. Senate Subcommittee on Juvenile Justice, 1994; Valdez, 2002). As such, activities occurring at nightclub events have the potential to impact both local economies and the criminal justice system. Although scholars have begun to investigate such phenomena more recently, generally speaking, empirical investigation of the urban nightclub economy has been scant.
Masculinity and nightclubs. A few points about masculinity are especially relevant for the nightclub setting. First, to reiterate an observation from feminist theory above, masculinity needs an institutional context (Connell, 1993; Connell & Messerschmidt, 2005; Demetriou, 2001) in which to be realized. Leisure establish- ments, such as bars and nightclubs, are yet other important institutions for the performance of masculinity, especially among young adults. Gender performances in club settings and out of them, therefore, construct “masculinity rather than merely reflect its preexistence” (Brickell, 2005, p. 32).
Second, sexuality is a routine component of performed masculinity (Connell, 1993; Connell & Messerschmidt, 2005; Donaldson, 1993). Divergent norms and expectations for public sexuality characterize nightclub scenes and the events that comprise them (Anderson, 2009). Because nightclub events are often highly sexualized (Anderson et al., 2007), they are pertinent venues in studying masculinity and crime. Recently, Anderson et al. (2007) noted that although the HH and EDM scenes showcased differ- ent cultural traits, the more defining aspect of the clubbing experience, with respect to behavior and consequence, was if the event was commercial or underground. Commercial events are those with an audience preference for popular music (anything on commercial-sponsored radio), an atmosphere of elitism and clearly defined cliques, an overemphasis on social (scene) status and style, loose hierarchal status ordering, and
at TEMPLE UNIV on June 5, 2016fcx.sagepub.comDownloaded from
Anderson et al. 307
highly sexualized interactional styles featuring hook-up objectives (Kavanaugh & Anderson, 2008). On the contrary, underground events are warm and friendly, prioritize the importance of music and intelligent discourse, and celebrate uniqueness, diversity, and respect. Such contexts have relaxed and casual atmospheres that encourage people to be themselves. Commercial events, both HH and EDM, called for hypermasculine traits and prioritized sexuality, although underground events showcased more muted masculinities and deemphasized sexual motives and actions (Anderson et al., 2007; Kavanaugh & Anderson, 2008).
Research indicates that gender-role norms, such as male honor and masculine status, reinforce both criminal behavior and substance use, particularly with respect to alcohol use and aggression at bars and nightclubs (Graham et al., 2002, 2006; Graham & Wells, 2001, 2003; Norstrom, 1998; Polk, 1999). That is, young men view both violence and excessive substance use as proof of masculinity, and engaging in these acts constitutes normal behavior (Gorman & White, 1995; White & Gorman, 2000). This proclivity toward violence can be strengthened by involvement with criminal others (Akers, 1998; Warr, 2002), which can further normalize such behavior. This research underscores the importance of investigating the masculinity and crime rela- tionship across diverse social contexts.
Method The data for this article were drawn from a multimethod ethnography examining the alcohol, drug, and crime relationship in the nightclub industry of Philadelphia, Pennsylvania. The parent study featured (a) in-depth interviews with 51 diverse par- ticipants in the city’s EDM and HH nightclub scenes, and (b) direct observation of 29 diverse nightclub events. Nightclub events are parties (i.e., one-offs, weeklies, or monthlies), branded and otherwise, that take place at nightclubs. For the present arti- cle, we utilize in-depth interviews from only the 24 male respondents because they were (a) the only ones to report committing the four types of crime we investigate here, and (b) they consistently tied such actions to the masculinity-related demands of certain club contexts. Female respondents reported possessing drugs at nightclub events, a type of criminal offense, but most often reported being victims of crime. Plus, we found no relationship between performed feminine and/or masculine behav- iors by women, which would allow us to include them in a study of masculinity and crime. We chose not to include their testimony about men’s masculinity at club events because it would not bear directly on any given male respondent in our study.
A brief glance at Philadelphia. Philadelphia is the largest city in Pennsylvania and the fifth largest in the United States (U.S. Census Bureau, 2000). It has been called a “black and white” city, comprised mostly of native born African Americans and White Americans. The 2000 U.S. Census substantiates this characterization. For example, Philadelphia’s more than 1.5 million3 residents are 45% White, 43% African American, 4.4% Asian, and about 7% from other racial groups. Roughly 9% are also from
at TEMPLE UNIV on June 5, 2016fcx.sagepub.comDownloaded from
308 Feminist Criminology 4(4)
Hispanic backgrounds. Still, Philly’s Hispanics and Asian American populations have increased over the past 20 years and continue to accelerate.
Musically, HH and rhythm and blues have had a stronghold on Philly for quite some time. New York City’s East Coast HH sound had a significant impact on Philly— something aided by I95, the northeast corridor expressway. Philly is also home to HH icons like DJ Jazzy Jeff and Will Smith, who helped pioneer the sound and scene in Philly. Although the electronic scene also had its Philly pioneers (e.g., Josh Wink and Dieselboy), it has struggled compared to vibrant scenes in New York, Los Angeles, and San Francisco (Anderson, 2009).
Sampling strategy. An ethnographic mapping/maximum variation sampling approach was used to recruit members of each nightclub scene (Strauss & Corbin, 1990; Watters & Biernacki, 1989). Respondents were recruited in a variety of ways. Recruitment began at a local Philadelphia record store, a small independent venue specializing in EDM. Early on, two store staff members were hired as key informants to assist in participant recruitment. Although efficient, relying on this recruitment strategy alone could have introduced bias into our study as early contacts were situated within the same social networks. We ran into some of this bias, as our sample overrepresents people with long- standing ties to the scenes we studied and who were also linked to the record-store employees. Live recruitment during direct observation was an alternative recruitment strategy, one that permitted us quick access to unrelated respondents. It helped reduce selection bias by discovering new or opening up networks of respondents. Live recruit- ment at nightclub events produced some respondents, but they were fewer than predicted. To address this, two more key informants were hired: (a) an Asian male DJ working at the record store and (b) a White female HH enthusiast who worked at another record store frequented almost exclusively by HH fans.
The sampling strategy was designed to recruit comparable groups of people in each scene. The original goal was to sample equal numbers of respondents across the major race groups, with respect to how they were represented in each scene. For difficulty in recruiting populations (e.g., Hispanics), the research team adopted a more targeted strategy, approaching specific race group members at events once saturation was reached for other racial categories.
Respondent interviews. Face-to-face, in-depth interviews were the primary source of information. Each participant was paid a US$25 honorarium for the interview. All participants were promised strict confidentiality, and all participant names are pseud- onyms. All interviews were tape-recorded by mutual consent. The interview guide included structured and open-ended questions about the respondent’s background, living situation and lifestyle, involvement and commitment to the EDM and HH scenes, nightclub culture and interaction therein, and experiences with drugs, criminal activity, and victimization. Interviews lasted an average of 2 hr each, with a range of 1.5 hr to 3 hr.
We made a concerted effort to assess how interviewer effects might have impacted our research process and findings. We assigned interviewers to respondents at random and
at TEMPLE UNIV on June 5, 2016fcx.sagepub.comDownloaded from
Anderson et al. 309
later compared transcripts with same-sex versus opposite-sex interviewer–respondent combinations. We were unable to do this by race as the interviewers were all White. Our analysis of the transcripts showed longer interviews between gender-matched interviewers and respondents but few substantive differences in content. Age discrepancies likely played a larger role. Graduate student interviewers (one male and one female) obtained, in general, more detail on deviant events than did those conducted by the older professor. Respondents seemed reluctant to brag about deviant behavior with the professor as she shared less identity with the respondents than did the graduate students. Of the 24 men included in this study, the graduate students interviewed 18 and the professor 7.
The respondent pool is multiracial, with several Blacks (n = 9), Whites (n = 8), Hispanics (n = 3), and Asians (n = 4). Overall, our sample reflected Philadelphia’s racial and ethnic composition. The mean age was 25.7, with an age range of 18 to 32. Of the 24 men interviewed, 12 were primarily active in the EDM scene and 12 in the HH scene, although several participated in both. We assigned them to a primary scene based on the types of nightclub events they had recently attended. Tables 1 and 2 pro- vide a demographic breakdown of the respondents by scene type.
An examination of Tables 1 and 2 indicate that the respondents are in their mid-20s. For the most part, respondents were situated in the lower- to upper-middle classes. Level of employment spanned from low- to midlevel service positions (e.g., bartender, retail, and file clerk) to white-collar positions (e.g., accountant, IT support, research, engineer, advertising/marketing, and middle management).
Direct observation. The purpose of the direct observation component of the study was to obtain information on the organizational (social and physical) structure of the club events and to document how the environment shaped gender performances, for example, masculinity and criminal outcomes.
The designated time period of observation was from April 2005 to mid-December 2006. Events were attended roughly every week, for a total of 29. We alternated between weeknight (Sunday-Thursday) and weekend (Friday, Saturday) events. In many cases, the events we attended were nominated by interview participants or key informants. The time of observation was generally from 11 p.m. to 2 a.m. Some large- scale commercial venues had extended hours alcohol licensing, so certain events did not end until between 3 and 6 AM. These events were observed for longer periods of time. The mean time of observation was 3.5 hr (Table 3).
All direct observations were documented in a pocket-sized field journal at the event, and a second, more interpretive version of these field notes was recorded the following day on a computer. This allowed for elaboration on observation experiences shortly after they took place. When writing field notes, the researchers included descriptive (straight- forward documenting of behaviors, patterns, and trends; structural features and venue layout; etc.) as well as inferential (perceived motivations and emotions of the participants engaging in behavior) information. The recording of inferential information was important in describing the nightclub setting in as much detail as possible (Wolfinger, 2002).
at TEMPLE UNIV on June 5, 2016fcx.sagepub.comDownloaded from
310 Feminist Criminology 4(4)
Data management and analysis. All interview transcripts and direct observation notes were by coded research staff using the qualitative software program ATLAS.ti, using an open-coding method. Every individual participant and every transcript or field note in the database was identified with the month and year of collection. Similarly, every question in the open-ended interview schedule was identified with an open-coding method, and specific codes were then assigned to the response. Textual
Table 1. EDM Respondents’ Demographic Characteristic
Race/sex Total Mean age Mean income
No. completed high school
No. completed
some college
No. completed BA or MA
Job prestige scorea
White men 5 26.8 US$29,600 2 3 1 49.04 Black men 3 26.7 US$32,500 1 2 0 33.76 Asian men 2 29.5 US$57,500 0 2 0 45.11 Hispanic men 2 24 US$20,500 1 1 0 34.4 Totals/means 12 25 US$29,700 8 11 2 38.06
Note: EDM = electronic dance music. a. Job prestige scores were assigned to respondents’ primary or full-time jobs. The prestige scores were taken from the 1989 General Social Survey (GSS). GSS respondents were asked to rate 110 different oc- cupations on a scale of 1 to 9. These scores were converted, using a formula, so that the prestige scores would have a logical range from 0 (lowest) to 100 (highest).
Table 2. Hip-Hop Respondents’ Demographic Characteristics
Race/sex Total Mean age Mean income
No. completed high school
No. completed
some college
No. completed BA or MA
Job prestige score
White men 3 26.3 US$23,000 1 2 0 42.68 Black men 6 26 US$28,000 1 4 2 37.42 Asian men 2 26.5 US$60,000 1 0 1 62.75 Hispanic men 1 29 US$50,000 0 1 0 53.99 Totals/means 12 26.3 US$33,140 5 13 12 52.16
Table 3. Direct Observation of Nightclub Events
Event type # Attended Mean hours observed
Commercial EDM 10 4.5 Commercial HH 5 3 Underground EDM 9 3.5 Underground HH 5 3 Totals 29 3.5
Note: EDM = electronic dance music; HH = Hip Hop.
at TEMPLE UNIV on June 5, 2016fcx.sagepub.comDownloaded from
Anderson et al. 311
material about given themes (such as masculinity, drug use, and criminal behaviors) were able to be quickly accessed at any point.
Members of the research team coded the interview transcripts and direct observa- tion notes. Intercoder reliability was high. This was due to the researchers meeting frequently to touch base throughout data collection and analysis to ensure that emerg- ing codes and concepts were similar (Lincoln & Guba, 1985). This strategy ensured further reliability during the actual coding stage of the analysis, and any discrepancies or issues were able to be reconciled quickly.
Our analysis revealed three central points in the life course where men discussed masculinity: youth, early adulthood outside the nightclub context, and early adulthood inside the clubbing context. After identifying these masculinity concepts, we coded respondents as low, medium, and high on each based on their narratives surrounding the concepts’ indicators. High designations refer to the most hypermasculine traits, whereas medium and low classifications are more muted in this respect. More infor- mation about this can be found below. This strategy allows us to mark the fluidity of masculinities among the respondents and how they are related to crime. Therefore, our analytic strategy emanates from grounded theory principles, such that our concepts emerge from the narratives provided by the respondents.
Limitations: Validity and reliability. Qualitative research studies typically encounter a few major limitations. The first pertains to the validity and reliability of testimonial informa- tion obtained from interview data. Most commonly, respondents are susceptible to recall error when recounting past experiences. Misrepresentation of the truth due to social desirability effects is also a concern. Although research indicates that people are usually truthful when interviewed about illicit activities if provided with con- fidentiality (Stephens, 1991), reporting victimization experiences can present some challenges.
We encountered some of these obstacles during the study, and securing reliable infor- mation about crime and victimization was sometimes challenging. During the interviews, some respondents were reluctant to disclose the extent of their illicit behavior, due in part to being tape recorded or concerns about confidentiality. Many of these problems were quelled when we matched interviewers and respondents on demographic characteristics, such as age and sex (Daily & Claus, 2001; Wilson, Brown, Mejia, & Lavori, 2002). In our study, however, researcher–interviewer demographic matching was not possible for all interviews. In other cases, we tried to establish shared identity with the respondents and disclose personal information, when appropriate, to both make respondents more com- fortable about sharing sensitive information and to equalize the power differential between interviewers and respondents (Holstein & Gubrium, 2003).
The second limitation pertains to the generalizability of the research findings. Ideally, numerous interviews and observations of a representative sample need to be performed to generalize findings with any degree of confidence. This is especially difficult when looking at particular demographic groups across two different popula- tions (EDM and HH nightclub attendees), in a city as large as Philadelphia. It is also worth noting that generalizability is regarded somewhat differently in qualitative
at TEMPLE UNIV on June 5, 2016fcx.sagepub.comDownloaded from
312 Feminist Criminology 4(4)
research. It is viewed almost as a given that in qualitative research designs, generaliz- ability is necessarily sacrificed for a greater depth of understanding and a nuanced level of detail not typically attainable in quantitative research designs (Van Den Berg, 2005).
Masculinity in Youth (MIY) During the interviews, the respondents described themselves at different points in time, including their gender identities. The first picture respondents gave us was in adoles- cence. They defined their gender identities around three aspects: trouble in school, focused attention on girls, and involvement in sports activities. Although there was a wide range among the respondents on these dimensions (see Table 4), one thing was clear: All believed these three items were core to defining who a typical boy was, even if they did not exude them at the time. From these patterns, we derived the concept of masculinity in youth. We use it in our analysis as a comparison point to illustrate how masculinities are fluid across time and social setting.
Trouble in school. The respondents described traditional masculine behaviors during their childhood consistent with current literature, such as competitiveness, aggressive- ness, dominance, and independence. As indicated above, research has established a link between these traits and crime (Cullen, Golden, & Cullen, 1979). The problem behaviors respondents discussed included being in trouble at school, for example, in and out of school suspension, skipping class, purposefully disrupting the classroom setting, and speaking disrespectfully to teachers and principals. Beijing, a 28-year-old White man, noted,
I would be constantly getting into fights and stuff like that. So I constantly would be getting suspended.
The second dimension of trouble behaviors is engaging in illegal activities. The illegal activities included the purchase and use of fake identification to buy alcohol and to attend nightclubs and bars. Other illegal activities involved engaging in vandalism, such as graffiti, shoplifting and then selling the stolen items, and forgery. Munich, a 25-year-old White man, told us, “I was a huge shoplifter like scam artist shoplifter like doing jobs.”
Focus on girls. Respondent’s also identified focusing attention on girls and women as a part of their masculine identities. The respondents told us that this was often achieved by catcalling women and girls, making sexual advances, or dancing with them. Expressing such sentiments maintain the masculine identity of the respondent. Brussels, a 21-year-old Hispanic man, boasted that when he was a teenager he would “just start flirting” with a girl until he “got her phone number.” This quote demon- strates how embedded sexuality is to the masculine construction (Connell, 1993).
Throughout their early years, men are encouraged, directly and indirectly, to adopt a socially constructed view of manhood. The social construction of masculinity
at TEMPLE UNIV on June 5, 2016fcx.sagepub.comDownloaded from
Anderson et al. 313
includes a set of cultural beliefs that prescribe what men ought to be like, such as physically strong, dependable, powerful, independent, and sexually potent (Connell, 1993). Men enjoy the social and psychological resources generated by group cohesion and dramaturgical teamwork. Britain, a 28-year-old Asian man, states that he has “a tight-knit core group of people that are my people and everybody else is different.” Santiago, a 31-year-old Puerto Rican man, described a main component of his youth was “hooking up with chicks” and “having parties” with his friends. In addition, men who engage in this behavior are rewarded publicly by their friends, even if they do not achieve the goal of ‘hooking up’ with a woman (Grazian, 2007).
Sports and competition. The third major theme of masculinity the respondents iden- tified was engaging in masculine interests, such as sports. Feder et al. (2007) claimed that traditional masculine socialization heightens the risk of boys engaging in acts of
Table 4. Comparison of Masculinity Concepts, Clubbing Environment, and Criminal Activity
Masculinity in youth
Masculinity in present
Masculinity in clubs
Type club: Com/
Uga (%) Assault Vandalism Drug
dealing Theft
Britain High High High 67/33 X X X Santiago High High High 75/25 X Brussels Medium High High 71/29 X X Managua Medium High Medium 78/22 X X Barbados Medium Medium Medium 33/66 X X Beijing High Medium High 50/50 X X X Munich High Low Medium 70/30 X Montreal Medium High Low 73/27 X Baghdad Low Medium High 50/50 X X X Rangoon High Low High 60/40 X X Galway Medium Low Medium 0/100 X Prague Medium Medium High 71/29 X X Melbourne Low Medium High 38/62 X X Helsinki High High High 33/66 X Pittsburgh Low Medium High 71/29 X Warsaw Medium High Medium 37/63 Geneva Medium Low Low 50/50 Helena Medium High Medium 40/60 Zurich Medium Low High 50/50 Berlin Low Low High 78/22 X Denver Low Medium Medium 42/58 Havana Low Low Low 50/50 Sydney Low High Medium 70/30 Toronto Low Medium Medium 40/60
a. Com = Commercial Club; Ug = Underground Clubs. Cell numbers are percentages of the respon- dents’ attendance at commercial and underground clubs in the past 5 years. The first number refers to commercial clubs and the second number to underground clubs.
at TEMPLE UNIV on June 5, 2016fcx.sagepub.comDownloaded from
314 Feminist Criminology 4(4)
violence because gender roles prevent them from learning proper ways and vocabu- lary to express their emotions. This limits their reactions, which leads to an increased likelihood of violence. Previous research has noted that boys who excel at sports receive admiration while simultaneously fulfilling their gender roles. Male youth are revered by their peer group for their adventurous behaviors, physical prowess, and intellectual behavior (Pope & Englar-Carlson, 2001).
Most of the respondents engaged in various organized sporting activities, such as basketball, wrestling, soccer, karate, skateboarding, and baseball, throughout their youth. Britain4 stated that he “played football, [he] was a wide receiver.” Managua, a 24-year-old White man, also recalled playing sports in his childhood. Santiago dis- cusses the role of sports during his youth:
I was an athlete, I was captain, the freshman team, before you get to JV, and the senior teams. I was the captain and all I wanted to do was fucking play football.
Boys who succeeded at sports received admiration from their peers and fulfilled appropriate gender identity (Pope & Englar-Carlson, 2001).
Once constructed, we coded the respondents as low, medium, and high on the MIY concept based on whether they reported displaying the three indicators and the nature of those traits. Respondents who were low on all three dimensions were coded low on MIY. Those reporting two of the three dimensions were coded as medium, and those reporting all three were classified as high on MIY. Coding also took into account variations in the three dimensions. For example, playing football (evidence of the sporting-activity dimension) was considered higher on the MIY concept than gymnastics. From this, we coded six of the respondents as high on MIY, 10 as medium, and eight as low. The break- down of respondents on MIY is depicted in Table 4.
Masculinity in the Present (MIP) We gleaned information about respondents’ hobbies and activities in the present from their narratives in the same fashion as we did in the past for the MIY concept. These activities were similar to the three dimensions of the MIY concept as both are testi- mony about the kinds of things the respondents were active doing during a particular time period (youth and adulthood).
In addition to narrative accounts in adulthood about hobbies and activities, we uti- lized other information to inform our masculinity concept in the present (MIP). The interview guide queried a revised version of the classic Twenty Statements Test derived by symbolic interactionists (Kuhn & McPartland, 1954; Spitzer, Stratton, Fitzgerald, & Mach, 1966). It asks respondents to provide 20 statements to the ques- tion “Who am I?” We asked for 5 during the interviews. The respondents’ statements provided a present-day (i.e., time of the interview) description of the self. From them, we gleaned more information about masculine identification.5 We did not ask respon- dents to describe their past selves (e.g., during adolescence) with similar statements because such accounts would be highly vulnerable to problems related to retrospective recall.
at TEMPLE UNIV on June 5, 2016fcx.sagepub.comDownloaded from
Anderson et al. 315
In general, respondents’ adult gender identity classifications and their reported hobbies/activities matched. Men whose identity statements were coded as high (indi- cating hypermasculinity) also reported currently engaging in quite masculine activities and hobbies. Consider an exchange between an interviewer and Sydney, a 25-year-old Black man, who scored high on MIP (see Table 4). His recollection of a skateboarding incident is consistent with a hypermasculine narrative.
Interviewer: What hobbies do you have? Sydney: Skateboarding, making beats on alternative machines. I’m into PlayStation too. Interviewer: How long have you been doing skateboarding? Sydney: 10 years. Interviewer: That looks like a skateboard injury. It looks like you ran your arm against the pavement. Sydney: It was hilarious. I turned to go to the Wawa and there is a little brick, and I watched my friend jump over the brick onto the sidewalk. I’m like, guess I’ll do that too, but I’m tired so I’m just going to roll my top wheel up and roll over it. Didn’t happen. And I fell and skid a little bit. It was fun. I laughed it off.
According to Table 4, 9 of the 24 respondents were high on the masculinity in adulthood concept, whereas 8 were medium, and 7 were low. The majority (16 of 24) of the respondents reported a different masculinity classification in adulthood than they did in youth. More of these changes in masculinity over time were of increases (11 of 16) than decreases (5). In other words, of those respondents who reported changes in their masculine identification over time, the most common pattern was an increase toward a hypermasculine variety. Clearly, our data show that masculinity was quite fluid over time.
Masculinity in Nightclubs (MIC) Even though men are socialized from youth to adhere to traditional masculine traits, such as aggressiveness, competitiveness, independence, and sexual virility (Connell, 1993), norms for and displays of masculinity differ by social setting and specific situations (Alder & Polk, 1996; Copes, 2003). For example, competitive dancing and MC-ing are acceptable ways of displaying ones’ masculinity in a night- club (LaBosky, 2001; Majors & Billson, 1992).
We constructed a clubbing masculinities concept (MIC) in a fashion similar to our MIY concept, using the three dimensions of girl-chasing behavior, club competitive- ness, and excessive alcohol use (an adult form of problem behavior and noted marker of masculinity). Respondents were considered high on MIC if they reported possess- ing two or more of these dimensions within the clubbing context. Those who reported only one were considered medium, whereas those reporting none were considered low. Each of these three dimensions is also supported by past research. For example,
at TEMPLE UNIV on June 5, 2016fcx.sagepub.comDownloaded from
316 Feminist Criminology 4(4)
the girl-chasing dimension was found significant in Grazian’s work, which also stud- ied Philadelphia’s nightlife. He found men were encouraged by their same-sex peers to adopt a manhood that was physically strong, independent, and sexually potent while interacting in the nightclub setting. These collective masculine performances reinforced conceptions of masculinity. Santiago illustrates the girl-hunting dimen- sion of MIC by claiming,
Girls love me. You know what I mean? I have friends, guys that are wallflowers. They are cool dudes and all but you know, their approach to women was more like “yo, yo, how you doing.” No dog. That doesn’t work. I take a more direct approach and dance with them. Then they [guys] would see me and I would be dancing with like three or four girls, you know.
Other criteria for fulfilling the MIC concept included men who objectified women in the interview; some stated that one of their main objectives while clubbing was to meet women or voiced their dislike and frustration of being in a club with mostly men and limiting their opportunity to meet women. Rangoon, a 31-year-old Hispanic man, told us,
I don’t like clubs that are too one thing; too Black, too White, too many guys. You know what I mean? The worst kind of club is the sausage festival, where it’s just all dudes. I don’t like that shit. It’s too much trouble and not enough girls.
Club competitiveness is another important dimension of the MIC concept, referring to expertise in dancing, MC-ing, and DJ-ing. Competitive dance, like sports, allows indi- viduals to express themselves creatively while displaying physical prowess (Majors & Billson, 1992). Central to the activities of competitive dance and MC-ing is vying for others respect by trying to one up or out do them. For example, two styles of dance were prevalent among men in the nightclub setting and are viewed as expressions of masculin- ity are break-dancing and moshing. Break-dancers assert masculinity by attempting to outperform other men via dangerous acrobatic moves, invading the others’ personal space, emphasizing one’s virility, and issuing challenges and taunts to competitors (LaBoskey, 2001). Moshing, on the other hand, carves out exclusive male spheres of violently colliding bodies, errant elbows, and stage diving, all of which work to empha- size the masculine qualities of toughness and risk taking (Krenske & McKay, 2000; Mullaney, 2007). men who punk (Philips, 2007) other men while dancing achieve hyper- masculinity at club events. Competitive dancing and MC-ing, therefore, allow individuals to assert their masculinity by displaying their physical skills and by upstaging other men. Britain told us,
To be a serious dancer, and by serious dancer, I mean someone who dances in the circle, someone that when they start dancing, people gather around them and watch them dance. You have to be hard. I have been in battles with people and
at TEMPLE UNIV on June 5, 2016fcx.sagepub.comDownloaded from
Anderson et al. 317
all of a sudden they will punch you in the head and keep dancing and expect you to keep dancing.
Pittsburgh, a 28-year-old Black man, added,
The majority of guys don’t dance and that’s how you get the girls because the girls are looking to dance. I can go up to them and dance with them and flirt with them and stuff and I always have a thing, like “yea I can take their girl.”
Another example of the club competition of the MIC concept is DJ-ing. DJ’s who took a paternalistic approach to their DJ-ing, expressing a desire to show people what good music was, were included as fulfilling the club competition aspect of the MIC concept. This is because these DJs voiced their superiority and authority of musical knowledge as compared to others and their desire to impress it on them. For example, Toronto—a 29-year-old Middle Eastern man—claimed, “Naturally, I take this big, helping stand. I am going to save the world and show the world good hip hop.”
The final dimension of the MIC concept is overconsumption of alcohol as an appropriate sign of manhood. Men often believe that the amount of alcohol, as well as the type of alcohol they drink, is a reflection of their masculine status (Tomsen, 1997). Therefore, men who can hold their liquor without showing signs of intoxica- tion and drink manly drinks are viewed as more masculine than “two beer queers” and those who drink “girly” or fruity drinks (Tomsen, 1997). Cognizant of this alco- hol dimension of masculinity, Santiago defended his dislike of hard alcohol by bragging about the number of beers he can consume. This worked to reestablish his masculinity.
Like my friends they like to drink Hennessey straight. I don’t drink Hennessey straight because it’s rough for me, but I like to mix it with some pineapple. And they call me a pussy about that but I drink Heinekens—alright put it like this. You know, in the evening I can drink 12 beers and you know, a couple shots, and I’m all good.
Men who boasted about the number of drinks they consumed while clubbing and those who nonchalantly mentioned consuming mass quantities of alcohol while clubbing but only getting a “slight buzz” were considered high on the alcohol consumption component of the MIC concept. For example, Berlin, a 23-year-old Black man, told us,
I mean its like, I don’t want to say its really hard for me to get shit faced, but I have been working in the bar for so long and you get tired. I can do four or five shots and still drink beers and I will be somewhat tipsy.
at TEMPLE UNIV on June 5, 2016fcx.sagepub.comDownloaded from
318 Feminist Criminology 4(4)
Once the dimensions of the MIC concept were identified, respondents were coded accordingly. Those who reported none of the three dimensions were considered low on the MIC. Respondents with one of these three dimensions were considered medium on MIC. Respondents with two or more of these dimensions were coded as high on MIC. This resulted in 4 respondents as low on MIC, 9 as medium, and 13 as high (see Table 4).
Nightclub Crime The respondents reported engaging in a number of crimes while in the nightclub set- ting, ranging from various forms of drug use and selling to assault with a deadly weapon. Respondents were asked about their involvement in illegal activities (inside and outside of the clubbing context) during the past 5 years. Participating in a criminal activity in the past 5 years inside the club or in the immediate area surrounding the club was coded as nightclub crime. Also coded as nightclub crime were criminal activ- ities in conjunction with clubbing, such as taking drugs on the way to the club or fighting someone after leaving the club.
Four main categories of crime emerged: assault, vandalism, drug dealing, and theft. A total of 16 respondents reported committing at least one of the four types of crime at a nightclub event within the past 5 years. Three (Britain, Bejing, and Baghdad) reported committing three of the four types of crime; however, most of these 16 respondents committed at least two. Vandalism was reported by 9 of the respondents, whereas both assault and drug dealing were reported by 8 each. Theft was the least- often crime committed by the respondents in our pool.
Respondents reported committing these crimes inside and outside of the club, but in close proximity of it. In other words, the overwhelming majority of crimes reported during the interviews were related to clubbing activities. For example, patterns of assault illustrate how masculine performance in the nightclub setting can lead to crime. Santiago talks about how competition over women and the territoriality of men can lead to violence in the club.
I was at a, 2 years ago, I was at a bar in the northeast. Predominantly White neighborhood, me and my friends go in who are predominantly Puerto Rican, Black, but we knew the DJ. It was all good. These White guys were wallflowers. The fucking young girls, they were dancing with us. They were into us. We are not from the neighborhood, and they were hot little fucking chicks. They were drunk and everybody was cool. Then, then one of the White dudes started some shit.
Interviewer: So you think it was racially motivated. Santiago: Yea, he spit into his face and called him a spic. Interviewer: Did he do that to you? Santiago: No, not to me but that carries over and then there’s a brawl.
at TEMPLE UNIV on June 5, 2016fcx.sagepub.comDownloaded from
Anderson et al. 319
Nightclub-related violence outside the club is illustrated by Brussels’ (21-year-old Hispanic man) tale of distributing his own brand of justice to a group of men he saw braking his car window when he was leaving a nightclub.
I never flipped out before, but it had to happen. When I saw these kids do it, I ran to my car to see the damage and automatically flipped out and got in my car. I told everybody to walk around the corner and follow these kids for a block. As soon as their backs were turned, I turned the corner, and followed them. It’s not funny, but I followed these kids and I run them over, all 15 of them, and beat the shit out of them.
An exchange with Britain and an interviewer illustrates numerous types of offending (e.g., theft and drug selling) unique to the nightclub setting, all of which invoke hypermasculine norms and identity traits.
Britain: From 1996 to 1999, if you were getting ecstasy in Philadelphia clubs, you were getting it from me or you were getting it from the guy I was working for. The setup was we were in a club and I sell the drugs. Then, the bouncers come and take it and they sell it back to me at a discounted rate. Interviewer: So you would take other peoples drugs from them? Britain: I would have other people take drugs that I had just sold to, and then I would buy it back off of them. I would sell a pill for 25, buy it back for 10. Like I said, we had our little system of what was going on. Once, there was this guy selling [ecstasy] and he came in and we took his stuff, and we were like you can’t do this here. Finally, they [Britain’s partners] grabbed him, put him in the back room, and they did him well, I mean they fucked him up.
The Perceived Environment and Masculinity Table 4 shows that men with most actively involvement in club-based criminal activi- ties were also those classified as high or medium on each of our masculinity concepts: MIY, MIP, and MIC. Guys like Santiago, Britain, Brussels, Barbados, and Helsinki illustrate this pattern. Alternatively, we found that men classified lowest on these two concepts reported little club-based offending. Geneva, Denver, and Havana fit this pattern. These results support past research on crime and masculinity, that hypermas- culine men are more heavily involved in crime (Beirne & Messerschmidt, 2007; Messerschmidt, 2004; Mullins, 2006).
Yet, there remains a larger group of men who constitute an important exception to this pattern, which past research cannot explain. It has to do with men like Beijing, Munich, Baghdad, Galway, Prague, Melbourne, and Pittsburgh, who reported discrep- ancies between their positions on MIP and MIC, and club-based criminal offending. Why are they involved in crime? Does their criminal involvement have anything to do with their masculinity being in flux in certain environments? Table 4 shows that each
at TEMPLE UNIV on June 5, 2016fcx.sagepub.comDownloaded from
320 Feminist Criminology 4(4)
of these respondents reported increasing their masculinity toward the hyper variety while interacting at nightclubs. Is it possible that certain individuals may be influ- enced by environmental cues or contextual factors in reshaping their identities to match the situations in which they interact and, in doing so, find themselves more likely to engage in crime?
To investigate this possibility, we triangulate our interview and direct-observation data to define respondents’ perceived environments and how those perceptions may impact clubbing masculinities and crime. Our analysis uncovered three aspects of the perceived environment (our proxy for social context) that played a role in the connec- tion between masculinity and nightclub crime. They include (a) a belief that clubbing norms were about getting drunk and hooking up, (b) a belief that clubbing was a com- petition that featured status and character contests, and (c) a preference for commercial parties.
Clubbing norms are about getting drunk and hooking up. To begin, respondents with high masculinities (increased for the club setting) told us that the most important norms of the clubbing environment had to do with getting intoxicated on alcohol and hooking up (being sexually active, both mildly and seriously) with women. These norms shaped their perceptions of the clubbing environment. Prague’s, 27-year-old White man, comments convey this perspective:
To a lot of people, I know that is the event. For some people, smoking a blunt6 before you go in the club, taking a break and going outside to smoke a blunt during the party . . . It’s like, they gotta be really high the whole time. Other people, they have to be drinking heavily all night or they think they miss out.
Rangoon added,
I think that from my experiences it’s very rare to go anywhere and talk to a sober clear-headed mind; I don’t think you will ever find that at a club these days, just a sober person that hasn’t done anything all night; it’s very rare. It’s very rare that I’m out somewhere and I’m not on something.
Alternatively, respondents low our masculinity concepts did not define getting loaded and hooking up as dominating norms of the clubbing experience. To them, the most important norms had to do with musical appreciation and bonding with friends. According to Geneva, a 22-year-old White man, “People are just drinking. They are there for the music as far as I can tell.”
Clubbing is about competition, status, character contests. Perceiving the clubbing envi- ronment, or the activity of clubbing itself, as fundamentally a competition waged through status differentials (aesthetic and hedonistic) and character contests was a second dimension of the perceived environment that may help to explain why mascu- linities were escalated for the club setting. Respondents with high (increase)
at TEMPLE UNIV on June 5, 2016fcx.sagepub.comDownloaded from
Anderson et al. 321
masculinities perceived the clubbing environment to be a tense competitive place filled with heavy drug and alcohol norms and lots of status jockeying, where people must compete for desired outcomes. On the contrary, men low on masculinity con- cepts did not perceive clubbing in a similar fashion. To an extent, they too viewed clubbing as status oriented, but the status that matters to them is anchored in musical knowledge and loyalty, not in hedonistic activities. Geneva told us,
I like to have authentic music experiences. What I like about electronic music is the actual tones and the creative things people are doing with it. When I hear that, I want to absorb into it.
Although many of our respondents thought women and men had different patterns of drinking, respondents higher on the MIC concept often interpreted these drinking differences in a sexist fashion (see Adams & Fuller, 2006; Iwamoto, 2003; McCall, 1995; Payne, 2006; Stephens & Few, 2007). More specifically, these men often equated the ability to drink more alcohol with social worth and privileged men. Rangoon provides an example.
I think most girls when they go out, they drink but they can’t handle it. I mean I don’t know about you, but most of the time there’s always at least two or three girls that are being walked out like this, and there’s a guy too but for the most part it seems like most chicks don’t handle their alcohol very well, most girls are lightweight in that they can’t handle [their alcohol].
Respondents told us that insult, status claims, inappropriate attention to a girlfriend or boyfriend, and unwanted stares and glances often tipped off physical assaults. At times, fights are ignited by unintended personal affronts. Other times, the fights seem provoked and based on serious indignities. Most respondents believe that alcohol is often the tipping point for the assault, as it is believed to help provoke a fight that might not happen with abstinence. Havana—a 28-year-old Black man—recounted,
It was at a club where we were, the DJ . . . to dance music, but then they did like a set of heavy metal music for an hour, so this you know big skinhead guy came out and was dancing, pushing people around. This big skinhead guy was kind of, I guess, a tough guy. He bumped to my friend Eric and you know my friend Eric is White too, but he said something to him and it ended up . . . I didn’t see it happened because I just walked down to a bar, and when I came back he was gone and I didn’t know what happened with him because we came together. He called me, he was in a hospital. He needed stitches because the guy hit him in his head. They went to a court and it was like the whole thing. The other guy was arrested.
at TEMPLE UNIV on June 5, 2016fcx.sagepub.comDownloaded from
322 Feminist Criminology 4(4)
Interviewer: Was alcohol involved? Havana: Oh, sure. I know my friend was drinking and actually I was drinking too because at that point it was late at night. That is usually when people get a little crazy.
Melbourne—a 27-year-old Asian man—explained,
A lot of them [fights] happened in more of the mainstream events, so anytime I’d go to an old city, with my old-city friends, or if I go to an Asian party with my Asian friends, there is always going to be some kind of altercation. You have your stereotypical frat boy kind of guys in old city who stink and who think it’s okay to make fun of Asian guys or something like that. I’m not that type of person who will back down from stuff like that, especially when I have been drinking. So there is just a lot of face to face, a lot of pushing, things like that.
Favoring commercial parties. Another common pattern we found among respondents high on masculinity concepts was a preference for commercial parties as opposed to underground ones. Anderson et al. (2007) defined commercial parties as those where the vibe is centered on sexuality, intoxication, mainstream music, and a commercial aesthetic. After observing an event with a commercial vibe, an ethnographer wrote in his journal:
The dancing was sexual, but more party oriented sexual—there were several areas of the dance floor where the sexual nature of the dancing seemed more serious—and the conversation in the bathroom confirms that hooking up is an objective.
A female ethnographer described another commercial event as follows:
The environment felt masculine to me. There was a very male type of gaze. I felt men controlled the tone of the party and I did not feel this way at the mash-up. Males seized visible power. Women’s power came from their hanging out with female friends and from decisions about dancing liaisons with males. Males seemed to control physical space of venue; they also seemed to be more there to pick up females than to hang with buddies.
Underground parties showcase a much different vibe, one anchored in friendship, solidarity around music, being causal, and true to oneself. Managua spoke about this vibe.
The vibe at CLUB E is very friendly, open, and it feels kind of like you could go in there and not know anyone, and feel like you know everybody, mainly because security is so nice; they don’t look down on you or think they are tough or any- thing like that. The bartenders are very friendly, very alert, or at least they try to be.
at TEMPLE UNIV on June 5, 2016fcx.sagepub.comDownloaded from
Anderson et al. 323
With minor exception (see discussion below), Table 4 shows respondents higher on the MIP and MIC concepts more often attended commercial parties. Take for example, Santiago. He told us he attended commercial parties nearly three to one and made a connection between its vibe and minor assaults at such events:
What they are doing to you in a club environment, they are giving you booze and they are putting on music for you to shake your ass to. They are basically implying sex. You know what I mean? Sex, Fun!
Even Montreal—a 31-year-old Asian whose masculinity decreased7 while clubbing— told us,
When I promoted an Asian night, it was half trance and half of it was hip hop. And during the trance scene, there was no problem no fights, but as soon as we started playing the hardcore hip hop—what I would call the “thug” music, the kind where you are talking about jacking somebody up or killing somebody—a fight would break out. You don’t hear in the news about somebody shooting somebody else in front of a trance club. It usually involves a hip hop night club.
As the testimony above indicates, respondents tied a party’s vibe to certain kinds of deviant behavior. Specifically, the vibe at commercial HH and EDM parties was linked to excessive alcohol use, sexual harassment and assault, and physical altercations.
Discussion Our objective was to explore the relationship between masculinity and crime at con- temporary nightclub events, a new hot spot for crime. By considering how masculinities materialized in diverse nightclub events, we are able to offer new insights about how the masculinity and crime relationship holds across time and space. Furthermore, our findings enable us to inform criminology about how fundamental forms of social organization (e.g., gender) and characteristics of social context impact crime.
Our research found criminal activity among a middle-class adult male population, one that receives little scholarly attention. Their offending was largely situated to a unique social context (commercialized EDM and HH nightclub events) and was driven by masculinity issues situated to the context within which they interacted. We found that men who were consistently high on hypermasculine identity traits and those who adjusted their masculinities toward hyperversions for the clubbing context were more often involved in nightclub crime than other respondents. Specifically, respon- dents with consistently high masculinities over time (high or medium on all three masculinity concepts) reported more nightclub crime than those with more muted masculinities.
at TEMPLE UNIV on June 5, 2016fcx.sagepub.comDownloaded from
324 Feminist Criminology 4(4)
An especially noteworthy finding of our study was the increase in one’s hypermas- culine identification in the nightclub setting. Regarding this fluidity, respondents whose identities became more hypermasculine in the clubbing context were likely to report engaging in nightclub crime, whereas respondents whose masculine identities became less hypermasculine in the club were not involved in nightclub crime. This shores up support for the gender-as-performance perspective and identifies nightclubs as an important environment where gender is negotiated, with often negative conse- quences. The exceptions to this pattern are Managua, Montreal, and Zurich. Although all three of these cases reported attending commercial parties more often or equally as often during the past 5 years, they admitted to trending toward underground events recently so as to avoid some of the risks they experience in the past. This brings us to an important finding about social context.
As we noted above, the vibe of commercial and underground events have features that align differently with masculinity. To reiterate, commercial events boast a hyper- masculine, sexual and status-oriented vibe and aesthetic. Underground events are more benign in this respect, defined instead by friendship, solidarity, and music appre- ciation. Given this, we expected to find that respondents with high masculine identities would prefer commercial events and those with low masculine identities would prefer underground events. Our analysis provided support for this contention.
There is an important environmental matter that must be acknowledged in this explanation. All but one of the respondents (Melbourne) with high masculinities reported attending more commercial club events than underground ones. Those with medium masculinities (Barbados, Denver, and Toronto) reported attending under- ground events more so, indicating that such club attendance might impede involvement in crime. In fact, Denver and Toronto both reported no nightclub crime. Barbados is the exception to this pattern.
To sum up, respondents with high masculinities are most likely to be involved in nightclub crime. Many respondents adjust their masculinities upward (i.e., increased hypermasculinity) when interacting in the nightclub setting. They do this because they define clubbing as a status-oriented, highly sexualized, and hedonistic endeavor, which is something they must toughen or macho-up to navigate. This gender perfor- mance often leads them to crime, which they otherwise might not have committed. In fact, none of the respondents who adopted more hypermasculine identities in the nightclub setting reported any other serious offending outside of the clubbing context. This was not the case for respondents with consistently high masculinities inside and outside the club setting. Several of them had reported some serious incidents outside of the clubbing content.
Above, we noted that masculinity could result in crime because certain situations and environments called for hypermasculine performances, identities, and behaviors (West & Zimmerman, 1987). This made establishing typologies of offenders difficult to establish because many could be at risk for offending as gender performances were relative to time and place (Copes & Hochstetler, 2003; Tomsen, 1997). Our research
at TEMPLE UNIV on June 5, 2016fcx.sagepub.comDownloaded from
Anderson et al. 325
has uncovered certain environmental aspects that are especially likely to call up hyper- masculine identification and, subsequently, increase the likelihood of crime. Such environmental features are even likely to do so from those we may not suspect, such as men with more muted masculinities outside the nightclub setting. These environ- mental factors include contextual norms about excessive alcohol consumption, heightened sexuality, competition, and commercialism. If masculinity is a situated performance, then, our study enables researchers and policy makers to identify what types of environments or where the most problematic masculinities, that is, hyper varieties, are most likely to be invoked, for these settings are more likely to be hot spots of criminal activity. Rather than relying on a fixed measure of masculine traits, research would need to construct a sort of calculus that fused environmental triggers with masculine identity markers. By identifying certain environmental characteristics especially likely to both cue up hypermasculine identification and increase crime, our analysis has taken a step toward such a fusion.
What results is a sort of quagmire for criminology in theorizing crime or construct- ing offender profiles for use in future research, policy, and interventions. Individuals have fluid masculinities and seem to construct them with respect to certain aspects in their environments. These findings indicate that future criminological research would be well served to utilize a more nuanced gender concept in explaining crime. The relationship between masculinity and crime will likely be moderated or influenced by aspects of one’s environment. It is important, then, that future research moves toward a new calculus of identity–social context or, if you will, micro–macro link in explain- ing crime.
Given this, it is important that practitioners recognize and policy implications reflect the complexity of masculinity and crime. This research demonstrates that there is not one specific type or demonstration of masculinity; therefore, policy recommendations per- taining to men and crime should evaluate what type of masculinity the individual adheres to. Doing so might result in more effective intervention techniques that could lead to a reduction in criminal activity. This includes paying attention to both how club owners and staff promote and manage clubs and the parties they host in addition to how people per- ceive the environment and its demands on identity.
Declaration of Conflicting Interests
The authors declared that they had no conflicts of interests with respect to their authorship or the publication of this article.
Funding
The first author received funding from the National Institute of Justice that helped support the present paper. Anderson, Tammy L. (Principal Investigator). 2005. “Exploring the Drugs/ Crime Connection within the Electronic Dance and Hip Hop Nightclub Scenes,” National Institute of Justice. U.S. Department of Justice.
at TEMPLE UNIV on June 5, 2016fcx.sagepub.comDownloaded from
326 Feminist Criminology 4(4)
Notes
1. There is a notable British literature on masculinities and crime within the nightclub setting (e.g., Hall & Winlow, 2005; Hall, Winlow & Ancrum, 2005; Winlow & Hall, 2006). How- ever, their focus is more on how state-level actors, for example, the police, work with the nightclub industry to counter crime problems. At the individual level, Hall and colleagues’ pseudopacification process focuses on how consumer-based hedonism, created by British culture, drives nightclub crime. Their work does not directly assess how gender perfor- mances at nightclub events impact criminal outcomes among clubbers.
2. This is illustrated in Winlow and Hall (2006) work on how changing economic conditions in northern England led to a transfiguration of shop-floor masculinities to a more entrepre- neurial criminal masculinity in the night-time economy surrounding pubs and clubs.
3. The 2005 population estimate from the Census Bureau is 1.4 million residents in the city proper. We use this number to calculate crime rates in Table 4.
4. We insert the age, race/ethnicity, and gender after the first mention of a respondent and do not do so on subsequent mentions to facilitate the flow of the text.
5. Over the years, scholars from diverse disciplines have weighed in on the traits and char- acteristics that define masculinity and femininity. Three social-psychological tools have been widely used since their introduction in the 1970s and1980s: the Bem Sex Role In- ventory (Bem, 1974), Chafetz’s (1977) seven domains of masculinity, and the masculinity and femininity scale of the Minnesota Multiphasic Personality Inventory–2 (see Graham, 2005). Each scale aligns adjectives (which represent traits) with masculinity and feminin- ity. A perusal of the adjectives shows considerable overlap between them. For example, all three scales use words like self-reliant, independent, assertive, forceful, ambitious, athletic, strong, intellectual, and so on. to define masculinity. We used these adjectives to help code masculinity in the present (MIP). If four or five of the respondents’ identity statements included masculine adjectives, we coded them as high on MIP. If two or three statements included masculine adjectives, we coded them as medium. Finally, if one or fewer of the respondents’ identity statements mentioned masculine adjectives, we coded them as low on MIP. The inclusion of these identity statements into the MIP concept is a departure from the MIY and MIC concepts. However, each of the three masculinity concepts is grounded in the respondents’ narratives about their behaviors and experiences. By incorporating an additional self-identity element into the MIP concept, we enhance the theoretical validity of our concepts, a fundamental goal of grounded theory analyses in qualitative research projects (Corbin & Strauss, 2007). In addition to these identity statements, our hobbies and adult-activities questions provide consistent information about gender identification in adulthood.
6. A blunt is a hollowed out cigar shell filled with marijuana. 7. Montreal is a unique case due to his long-standing professional ties to the clubbing industry
that make him extremely cautious about his club behavior. This might explain the decrease in his masculinity while at clubs. He knows to keep it toned down now so that his livelihood can thrive.
at TEMPLE UNIV on June 5, 2016fcx.sagepub.comDownloaded from
Anderson et al. 327
References
Adams, T. M., & Fuller, D. B. (2006). The words have changed but the ideology remains the same: Misogynistic lyrics in rap music. Journal of Black Studies, 36, 938-957.
Akers, R. L. (1998). Social learning and social structure: A general theory of crime and devi- ance. Boston: Northeastern University Press.
Alder, C. M., & Polk, K. (1996). Masculinity and child homicide. British Journal of Criminol- ogy, 36, 396-411.
Anderson, T. L. (2009). Rave culture: The alteration and decline of a music scene. Philadelphia, PA: Temple University Press.
Anderson, T. L., Kavanaugh, P. R., Bachman, R., & Harrison, L. D. (2007). Exploring the drugs–crime connection within the electronic dance and hip-hop nightclub scenes, final report. Washington, DC: U.S. Department of Justice.
Anderson, T. Miller, S., & Ousey, G. (2008). Criminology in contemporary society. Manuscript submitted for publication.
Beirne, P., & Messerschmidt, J. (2007). Criminology (4th ed.). Oxford, UK: Oxford University Press.
Bem, S. L. (1974). The measurement of psychological androgyny. Journal of Consulting and Clinical Psychology, 42, 155-162.
Bennett, A. (2001). Cultures of popular music. Buckingham, UK: Open U Press. Berkey-Gerard, M. (2001, May 7). Neighbors and Nightclubs. Gotham Gazette. Retrieved from
http://www.gothamgazette.com/iotw/disco/. Bourgois, P. (1996). In search of masculinity: Violence, respect and sexuality among Puerto
Rican crack dealers in East Harlem. British Journal of Criminology, 36, 412. Brickell, C. (2005). Masculinities, performativity, and subversion: A sociological reappraisal.
Men and Masculinities, 8, 24-43. Brickell, C. (2006). The sociological construction of gender and sexuality. Sociological Review,
54, 87-113. Buddie, A. M., & Parks, K. A. (2003). The role of the bar context and social behaviors on
women’s risk for aggression. Journal of Interpersonal Violence, 18, 1378-1393. Bureau of Justice Statistics. (2006). Crime characteristics, summary findings 2005: Place of
occurrence. Washington, DC: U.S. Department of Justice. Butler, J. (1999). Gender trouble: Feminism and the subversion of identity. New York: Rout-
ledge Press. Chafetz, J. S. (1977). Masculine/feminine or human? An overview of the sociology of sex roles.
Itasca, IL: F. E. Peacock. Cohen, A. K. (1995). Delinquent boys. New York: Free Press. Cohen, J., Gorr, W., & Singh, P. (2003). Estimating intervention effects in varying risk settings:
Do police raids reduce illegal drug dealing at nuisance bars? Criminology, 41, 257-292. Cohen, J. W. & Harvey, P. J. (2006). Misconceptions of gender: Sex, masculinity, and the mea-
surement of crime, Journal of Men’s Studies, 14, 223-233. Cohen, L. E., & Felson, M. (1979). Social change and crime rate trends: A routine activity
approach. American Sociological Review, 44, 588-608.
at TEMPLE UNIV on June 5, 2016fcx.sagepub.comDownloaded from
328 Feminist Criminology 4(4)
Connell, R. W. (1993). The big picture: Masculinities in recent world history. Theory and Soci- ety, 22, 597-623.
Connell, R. W., & Messerschmidt, J. (2005). Hegemonic masculinity: Rethinking the concept. Gender & Society, 19, 829-859.
Copes, H. (2003). Streetlife and the rewards of auto theft. Deviant Behavior, 24, 309-332. Copes, H., & Hochstetler, A. (2003). Situational construction of masculinity among male street
thieves. Journal of Contemporary Ethnography, 32, 279-304. Corbin, J., & Strauss, A. (2007). Basics of qualitative research: Techniques and procedures for
developing grounded theory. Newbury Park, CA: Sage. Cullen, F. T., Golden K. M., & Cullen, J. B. (1979). Sex and delinquency: A partial test of the
masculinity hypothesis. Criminology, 17, 301-310. Daily, R., & Claus, R. E. (2001). The relationship between interviewer characteristics and phys-
ical and sexual abuse disclosures among substance users: A multilevel analysis. Journal of Drug Issues, 31, 867-888.
Demetriou, D. (2001). Connell’s concept of hegemonic masculinity: A critique. Theory and Society, 30, 337-361.
Donaldson, M. (1993). What is hegemonic masculinity? Theory and Society, 22, 643-657. Feder, J., Levant, R. F., & Dean, J. (2007). Boys and violence: A gender-informed analysis.
Professional Psychology: Research and Practice, 38, 385-391. Gorman, D. M., & White, R. H. (1995). You can choose your friends, but do they choose your
crime? Implications of differential association theories for crime prevention policy. In H. Barlow (Eds.), Criminology and public policy: Putting theory to work. Boulder, CO: West- view.
Graham, J. R. (2005). MMPI-2: Assessing personality and psychopathology. Oxford, UK: Oxford University Press.
Graham, K., Osgood, D. W., Wells, S., & Stockwell, T. (2006). To what extent is intoxication associated with aggression in bars? A multilevel analysis. Journal of Studies on Alcohol, 67, 382-390.
Graham, K., & Wells, S. (2001). Aggression among young adults in the social context of the bar. Addiction Research and Theory, 9, 193-219.
Graham, K., & Wells, S. (2003). “Somebody’s gonna get their head kicked in tonight!” Aggres- sion among young males in bars—A question of values? British Journal of Criminology, 43, 546-566.
Graham, K., Wells, S., & Jelley, J. (2002). The social context of physical aggression among adults. Journal of Interpersonal Violence, 17, 64-83.
Grazian, D. (2007). The girl hunt: Urban nightlife and the performance of masculinity as collec- tive behavior. Symbolic Interaction, 30, 221-243.
Grazian, D. (2008). One the make: The hustle of urban nightlife. Chicago: University of Chi- cago press.
Hafez, M. M. (2007). Martyrdom mythology in Iraq: How jihadists frame suicide terrorism in videos and biographies. Terrorism and Political Violence, 19, 95-115.
Hall, S., & S. Winlow. (2005). Night-time leisure and violence in the breakdown of the pseudo- pacification process. Probation Journal, 52, 376-389.
at TEMPLE UNIV on June 5, 2016fcx.sagepub.comDownloaded from
Anderson et al. 329
Hall, S., Winlow, S., & Ancrum, C. (2005). Radgies, gangstas, and mugs: Imaginary criminal identities in the twilight of the pseudo-pacification process. Social Justice, 32(1), 100-112.
Harlan, J. P., & McDowell, C. P. (1980). Vindictive vandalism and the schools—Some theoreti- cal considerations. Journal of Police Science and Administration, 8, 399-405.
Holmberg, M. (2001, June 6). US VA: Nightclubs in alphabet soup with ABC cracking down on X. Richmond Times-Dispatch (p. B1).
Holstein, J., & Gubrium, J. F. (2003). The life course. In Larry T. Reynolds & Nancy J. Herman- Kinney (Eds.), Handbook of symbolic interactionism (pp. 835-855). Walnut Creek, CA: AltaMira.
Hopkins, S. J. (2004). Let the drag race begin: The rewards of becoming a queen. Journal of Homosexuality, 46, 135-149.
Hutton, F. (2005). Risky business: Gender, drug dealing and risk. Addiction Research & Theory, 13, 545-554.
Iwamoto, D. (2003). Tupac Shakur: Understanding the identity formation of hyper-masculinity of a popular hip-hop artist. Black Scholar, 33(2), 44-49.
Johnson, M. (2001). Successful rave operations. Law and Order, 49, 184-188. Kaplan, A. (1978). The psychodynamics of terrorism. Terrorism: An International Journal, 1,
237-254. Kavanaugh, P. R., & Anderson, T. L. (2008). Solidarity and drug use in the electronic dance
music scene. Sociological Quarterly, 49, 189-208. Kennelly, I., Merz, S. N. & Lorber, J. (2001). What is gender? American Sociological Review,
66, 598-605. Kreiger, T. S., & Dumka, L. E. (2006). The relationship between hypergender, gender, and psy-
chological adjustment. Sex Roles, 54, 777-785. Krenske, L., & McKay, J. (2000). “Hard and heavy”: Gender and power in a heavy metal music
subculture. Gender, Place, and Culture, 7, 287-304. Kuhn, M. H., & McPartland, T. S. (1954). An empirical investigation of self-attitudes. American
Sociological Review, 19, 68-76. LaBoskey, S. (2001). Getting off: Portrayals of masculinity in hip hop dance in film. Dance
Research Journal, 33(2): 112-120. Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic inquiry. Beverly Hills, CA: Sage. Lipton, R., & Gruenewald, P. (2002). The spatial dynamics of violence and alcohol outlets.
Journal of Studies on Alcohol, 63, 187-195. Lorber, J. (1994). “Night to his day”: The social construction of gender. In Judith Lorber (Ed.),
Paradoxes of gender (Chapter 1, pp. 13-36). New Haven, CT: Yale University Press. Lorber, J. (2000). Using gender to undo gender: A feminist degendering movement. Feminist
Theory, 1(1), 79-95. Lutz, F. E., & Murphy, D. W. (1999). Ultramasculine prison environments and inmates’ adjust-
ment: It’s time to move beyond the boys will. Justice Quarterly, 16, 709-733. Majors, R., & Billson, J. M. (1992). Cool pose: The dilemmas of Black manhood in America.
New York: Lexington Books. Man, D. C., & Cronan, J. P. (2001). Forecasting sexual abuse in prison: The prison subculture
of masculinity as a backdrop for “deliberate indifference”. Journal of Criminal Law and Criminology, 92, 127-185.
at TEMPLE UNIV on June 5, 2016fcx.sagepub.comDownloaded from
330 Feminist Criminology 4(4)
McCall, N. (1995). Gangsta rap promotes violence in the Black community. In C. Wekesser (Eds.), Violence in the media (pp. 159-162). San Diego, CA: Greenhaven Press.
Messerschmidt, J. W. (1994). Schooling, masculinities, and youth crimes by White boys. In T. Newburn & E. A. Stanko (Eds.), Just boys doing business? Men, masculinities and crime (pp. 81-99). London: Routledge.
Messerschmidt, J. W. (2004). Flesh and blood: Adolescent gender diversity and violence. Oxford, UK: Rowman & Littlefield.
Mosher, D. L., & Tomkins, S. S. (1988). Scripting the macho man: Hypermasculinity, socializa- tion and enculturation. Journal of Sex Research, 25, 60-84.
Mosler, D. (2001). Club drugs. Law Enforcement Quarterly, 30(2), 5-10. Mullaney, J. L. (2007). “Unity admirable but not necessarily heeded”: Going rates and gender
boundaries in the straight edge hardcore music scene. Gender & Society, 21(3), 384-408. Mullins, C. W. (2006). Holding your square: Masculinities, streetlife and violence. Devon, UK:
Willan Press. Murnen, S. K., & Byrne, D. (1991). Hyperfemininity: Measurement and initial validation of
construct. Journal of Sex Research, 28, 479-489. Norstrom, T. (1998). Effects on criminal violence of different beverage types and private and
public drinking. Addiction, 93, 689-699. Parks, K. A., Miller, B. A., Collins, R. L., & Zetes-Zanatta, L. (1998). Women’s descriptions of
drinking in bars: Reasons and risks. Sex Roles, 38, 701-717. Payne, Y. A. (2006). “A gangster and a gentleman”: How street life-oriented, U.S.-born African
men negotiate issues of survival in relation to their masculinity. Men and Masculinities, 8, 288-297.
Peralta R. L., & Cruz, J. M. (2006). Conferring meaning onto alcohol-related violence: An analysis of alcohol use and gender in a sample of college youth. Journal of Men’s Studies, 14, 109-125.
Phillips, D. A. (2007). Punking and bullying: Strategies in middle school, high school and beyond. Journal of Interpersonal Violence, 22, 158-178.
Polk, K. (1999). Males and honor contest violence. Homicide Studies, 3(1), 6-29. Pope, M., & Englar-Carlson, M. (2001). Fathers and sons: The relationship between violence
and masculinity. Family Journal, 9, 367-374. Roncek, D. W., & Meier, P. A. (1991). Bars, blocks and crimes revisited: Linking the theory of
routine activities to the empiricism of “Hot Spots.” Criminology, 29, 725-753. Schwalbe, M., & Wolkomir, M. (2001). The masculine self as problem and resource in inter-
view studies of men. Men and Masculinities, 4, 90-103. Shaw, H. D., & McKay, C. R. (1942). Juvenile delinquency in urban areas. Chicago: University
of Chicago Press. Sherman, L. W. (1995). Public regulation of private crime prevention. Annals of the American
Academy of Political and Social Science, 539, 102-113. Spitzer, S. P., Stratton, J. R., Fitzgerald, J. D., & Mach, B. K. (1966). The self concept: Test
equivalence and perceived validity. Sociological Quarterly, 7, 265-280. Stephens, D. P., & Few, A. L. (2007). The effects of images of African American women in hip
hop on early adolescents’ attitudes toward physical attractiveness and interpersonal relation- ships. Sex Roles: A Journal of Research, 56, 251-264.
at TEMPLE UNIV on June 5, 2016fcx.sagepub.comDownloaded from
Anderson et al. 331
Stephens, R. C. (1991). The street addict role: A theory of heroin addiction. Albany: State Uni- versity of New York at Albany Press.
Strauss, A., & Corbin, J. (1990). Basics of qualitative research: Grounded theory procedures and techniques. Newbury Park, CA: Sage.
Tomsen, S. (1997). A top night: Social protest, masculinity and the culture of drinking violence. British Journal of Criminology, 37, 90-102.
U.S. Bureau of the Census. 2005 Population Estimates. http://www.census.gov/main/www/ cen2000.html
U. S. Senate Subcommittee on Juvenile Justice. (1994). Shaping our responses to violent and demeaning imagery in popular music. Washington, DC: U.S. Government Printing Office.
Valdez, A. (2002). Club drugs. Police: The Law Enforcement Magazine, 26(4), 74-77. Van den Berg, H. (2005). Reanalyzing qualitative interviews from different angles: The risk
of decontextualization and other problems of sharing qualitative data. Forum Qualitative Sozialforschung/Forum: Qualitative Social Research, 6(1).
Warr, M. (2002). Comparisons in crime: The social aspects of criminal conduct. New York: Cambridge University Press.
Watters, J. K., & Biernacki, P. (1989). Targeted sampling: Options for the study of hidden popu- lations. Social Problems, 36, 416-430.
Weisburd, D. (2002). From criminals to criminal contexts: Reorienting crime prevention research and policy. In E. Waring & D. Weisburd (Eds.), Crime and social organization (pp. 197-216). New Brunswick, NJ: Transaction.
West, C., & Zimmerman, D. H. (1987). Doing gender. Gender & Society, 1, 125-151. White, H. R., & Gorman, D. (2000). Dynamics of the drug–crime relationship. In G. La Free
(Eds.), The nature of crime: Continuity and change (pp. 151-218). Washington, DC: U.S. Department of Justice.
Whitehead, A. (2005). Man to man violence: How masculinity may work as a dynamic risk fac- tor. Howard Journal of Criminal Justice, 44, 411-422.
Wilson, S. R., Brown, N. L., Mejia, C., & Lavori, P. W. (2002). Effects of interviewer char- acteristics on reported sexual behavior of California Latino couples. Hispanic Journal of Behavioral Sciences, 24(1), 38-62.
Winlow, S., & Hall, S. (2006). Violent night: Urban leisure and contemporary culture. Oxford, United Kingdom: Berg.
Wolfinger, N. H. (2002). On writing fieldnotes: Collection strategies and background expectan- cies. Qualitative Research, 2(1), 85-95.
Bios
Tammy Anderson is an Associate Professor in the University of Delaware’s Department of Sociology and Criminal Justice. She has published many articles and an edited volume (Neither Villain nor Victim, Rutgers University Press 2008) on drug abuse, identity, gender, race, and stigma. Her new book Rave Culture: the Alteration and Decline of a Philadelphia Music Scene (Temple University Press 2009) is an innovative and comparative ethnographic study about the social, cultural, and economic forces that alter youth-based music scenes. Her work has been sponsored by the National Science Foundation and the National Institute of Justice. For more information, please visit www.udel.edu/soc/tammya.
at TEMPLE UNIV on June 5, 2016fcx.sagepub.comDownloaded from
332 Feminist Criminology 4(4)
Kevin Daly is a doctoral student in Sociology at the University of Delaware and a research assistant at the University of Delaware’s Center for Drugs and Alcohol Studies. His research focuses are on deviance, law and society, and the criminal justice system. Laura A. Rapp is pursuing her doctorate in Sociology at the University of Delaware. Her main area of interest is exploring the interconnections of gender and race on sexual violence. Her research includes men convicted of sexual offenses, masculinity, sexual violence, and inequal- ity/stratification issues.
at TEMPLE UNIV on June 5, 2016fcx.sagepub.comDownloaded from
12 sources/Bartkowski.pdf
See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/228802346
Love Thy Neighbor? Moral Communities, Civic Engagement, and Juvenile Homicide in Rural Areas
Article in Social Forces · March 2004
Impact Factor: 1.29 · DOI: 10.1353/sof.2004.0044
CITATIONS
92
READS
60
2 authors, including:
John P. Bartkowski
University of Texas at San A…
70 PUBLICATIONS 1,644 CITATIONS
SEE PROFILE
Available from: John P. Bartkowski
Retrieved on: 19 June 2016
Love Thy Neighbor? / 1001
© The University of North Carolina Press Social Forces, March 2004, 82(3):1001-1035
* Partial funding for this project was provided by grant number SES 0237968 from the National Science Foundation to the first author. Funding was also provided to both authors by grant number 4 D1A RH 00005-01-01 from the Office of Rural Health Policy of the Department of Health and Human Services through the Rural Health Safety and Security Institute, Social Science Research Center, Mississippi State University, and through a grant from the Criss Fund at Mississippi State University. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the National Science Foundation, the Office of Rural Health Policy, or the Criss Fund. A previous draft of this article was presented at the 53rd annual meeting of the American Society of Criminology in Atlanta, Georgia, November 2001. We thank J. Michelle Estis for help with data set construction and anonymous Social Forces reviewers for helpful comments on a prior draft of the manuscript. Direct correspondence to Matthew R. Lee, Department of Sociology, Anthropology, and Social Work, Mississippi State University, P.O. Box C, Mississippi State, MS 39762. E-mail [email protected].
Love Thy Neighbor? Moral Communities, Civic Engagement, and Juvenile Homicide in Rural Areas*
MATTHEW R. LEE, Mississippi State University JOHN P. BARTKOWSKI, Mississippi State University
Abstract
While juvenile homicide garnered a tremendous amount of attention from the general public, the media, and policymakers around 1990, macro-level research examining intercommunity variations in juvenile homicide is generally sparse. In addition, most studies addressing this topic focus on urban areas, neglecting the equally important issue of juvenile homicide in rural communities. This analysis extends prior research by investigating the structural sources of variation in rural juvenile homicide rates and by examining the influence of religion on this phenomenon. Informing our analyses with theoretical insights drawn from the moral communities and civil society literatures, we investigate the protective effects of civically engaged religious denominations on juvenile family, acquaintance, and stranger homicides in rural counties. For comparative purposes, we also perform parallel analyses on a sample of urban areas. The empirical analyses of county-level data using negative binomial regression estimation techniques indicate that the presence of civically engaged religious adherents is inversely associated with juvenile homicide in rural areas (net of the effects of a range of control variables), but that this protective effect is primarily confined to juvenile family homicides. In contrast, the measure of civically engaged denominations has no effect on juvenile homicide in urban areas. We conclude with a discussion of the theoretical importance of these findings and directions for future research.
1002 / Social Forces 82:3, March 2004
In the late 1980s and early 1990s, the phenomenon of juvenile homicide garnered national attention, claiming cover stories in prominent newspapers and popular periodicals throughout the U.S. (see, e.g., Fox & Pierce 1994; Gibbs & Grace 1994). Two factors were largely responsible for catapulting this issue to the forefront of American life. First, urban juvenile homicide rates rose dramatically during this period (see Blumstein 1995; Blumstein & Wallman 2000; Fox 1996; Fox & Zawitz 2001). Second, juveniles appeared to be engaging in school-based violence with greater frequency (see D. Anderson 1998; Holmes & Holmes 2001).
A great deal of speculation surrounded this issue, with some observers attributing the increase to the transformation of American youth from good kids to “super-predators” with a penchant for extremely violent behavior (Bennett, DiIulio & Walters 1996). Other investigators highlighted the presence of concentrated socioeconomic disadvantage and the frustration it generated among youth (E. Anderson 1998). Still others pointed to the proliferation of drug markets as the driving force behind increased youth violence (Blumstein 1995). Although these and other explanations for the surge in juvenile homicide rates are compelling, the scholarly literature continues to be marked by a lack of empirical research addressing cross-community variations in juvenile homicide. In addition, most research in this tradition focuses on explaining community-level variation in urban juvenile homicide rates, thereby neglecting the equally interesting and important phenomenon of rural juvenile homicide. This propensity to focus on urban areas can be partially attributed to the fact that macro-level theories have widely been considered urban theories (see Osgood & Chambers 2000) and also results from the empirical reality that juvenile homicide rates on average are higher in urban than in rural areas (see Fox & Zawitz 2001).
Despite these facts, we see no compelling reason to dismiss the need for structural analyses of rural juvenile violence out of hand. Like urban areas, rural communities exhibit tremendous variation in rates of crime. And as Lee and Ousey (2001:582) note, “by neglecting rural settings, researchers have ignored important data that may yield new insight into the factors that explain crime rate variations across diverse geographic communities.” Moreover, the differences in rates of crime between rural and urban areas are not as great as commonly believed, and the absolute difference has actually been declining over time (Weisheit & Donnermeyer 2000). Considering these issues, the first goal of this study is to provide a macro-level analysis of rural juvenile homicide.
In addition, we also hope to advance the literature by examining the role of religious organizations in creating moral communities. The one study of which we are aware that investigates the structural covariates of rural juvenile violence fails to take into account the theoretical tradition in sociology sug- gesting that moral communities and civically engaged religious denominations may provide protective effects against juvenile violence (Osgood & Chambers
Love Thy Neighbor? / 1003
2000). Yet sociologists have long studied the effects of religious organizations on the communities in which they are situated. One of the most formidable theories on this topic is the religious ecology hypothesis — otherwise known as the moral communities thesis (Chadwick & Top 1993; Cochran & Akers 1989; Junger & Polder 1993; Regnerus 2000; Stark, Doyle & Kent 1980; Stark, Kent & Doyle 1982; Welch, Tittle & Petee 1991; Woodrum & Hoban 1992). Durkheimian in origin (see Stark 1996; Wolf 1970), this thesis posits that reli- gious institutions create a moral ecology fostering community integration and social control while discouraging deviance and criminal activity. We extend the moral communities hypothesis by arguing that the benefits of religious involve- ment are especially derived from the presence of civically engaged denomi- nations.1 Specifically, when a substantial proportion of a community’s popu- lation adheres to civically engaged religious institutions, horizontal social net- works may be strengthened, normative consensus on acceptable and unaccept- able behaviors may be elevated, interpersonal trust may be enhanced, and the community’s ability to express and pursue collective goals may be bolstered.
Furthermore, given the profamily character of religious organizations (e.g., Abbott, Berry & Meredith 1990; Bartkowski 2001; Christiano 2000; Eberly 1999; Gay, Ellison & Powers 1996; Hertel & Hughes 1987; Pankhurst & Houseknecht 2000; Wilcox 2002), the theoretical model we delineate below suggests faith-based forms of civic engagement may be particularly effective for retarding juvenile family violence. Thus, protections afforded to those outside the nexus of family relationships are anticipated to be less robust. While an evolving body of research has begun documenting the benefits of moral communities or civically engaged religious denominations for other aspects of community life such as population stability (Irwin, Tolbert & Lyson 1999), residential integration (Irwin et al. 2002), and socioeconomic well-being (Tolbert, Lyson & Irwin 1998), no studies to date have investigated the possible beneficial effects of a civically engaged population base on juvenile violence at the ecological level.
Finally, we take a cue from research on the centrality of religious institu- tions to rural civic life by anticipating an especially robust deterrent effect of civically engaged denominations on juvenile homicide in nonmetropolitan areas. Religious institutions have shown themselves to be crucial in creating and strengthening the civic infrastructure of nonmetropolitan communities (Bartkowski & Regis 2002, 2003; Goreham 2002; King, Elder & Whitbeck 1997; Parisi et al. 2002). In fact, it is not an overstatement to say that religious enti- ties such as congregations, ministerial programs, and interfaith alliances are the key institutional conduit through which rural communities in America gen- erate their most formidable civic bonds. Research has shown that religious participation is a particularly critical resource in southern nonmetropolitan communities for marginalized groups such as blacks, for whom rural churches
1004 / Social Forces 82:3, March 2004
are considered a “semi-involuntary” social institution (Ellison & Sherkat 1995; Hunt & Hunt 2000; Sherkat & Cunningham 1998). Thus, scholars now recog- nize that rural religious communities provide an important springboard for the cultivation of social ties and the facilitation of civic engagement.
More germane to our focus on juvenile homicide, it is worth noting that religious institutions play a central role in the civic participation of youth in rural areas. Although there is not a great deal of evidence on this score, that which is available points to the importance of rural religious institutions in the lives of youth. Survey data have revealed that youngsters raised in rural areas are more supportive of religious values and more involved in religious institutions, including church-based youth groups (King, Elder & Whitbeck 1997). And, in their recent book on the social standing of youth in the rural Midwest, Elder and Conger (2000) provide interview data which underscores the profound attachment that rural teens exhibit toward youth ministry programs. Thus, religion seems to exercise an especially important influence on the lives of rural youth. On the whole, the lack of research attention to rural juvenile homicide rates and the failure of analysts to explore more fully the links between religiously based civic engagement and crime rates signals a substantial void in the research literature. Among its other aims, our analysis is designed to redress these omissions.
Background on Juvenile Homicide
Social scientists are acutely aware of the fact that from the mid-1980s until the early 1990s, homicide rates in the U.S. skyrocketed.2 One of the most noteworthy aspects of this trend is the increase in homicidal offending perpetrated by certain segments of the population, specifically juveniles. For example, the homicide rate for juveniles 14 to 17 years old changed from 7.0 to 19.1 per 100,000 people between 1985 and 1994, a 173% increase (Fox 1996). This same age group experienced a 40.5% increase in their rate of arrest for homicide between 1989 and 1994 (Fox 1996). All the available evidence indicates that juvenile homicide rates reached an all-time high around 1990, even as adult homicide rates were generally declining (Fox 1996; Rosenfeld 2000).
Although longitudinal trends in juvenile homicide are interesting in their own right, national trends mask a great deal of community-level variation in juvenile homicide rates. Some communities experience extremely high rates of juvenile homicide, while others witness none at all (see Ousey & Augustine 2001). Such variations characterize urban as well as rural areas. And while some attention has been directed toward explaining cross-sectional variation in urban juvenile homicide rates (see Ousey & Augustine 2001; Sampson 1987; Shihadeh & Steffensmeier 1994), we are aware of only one study that examines the correlates of cross-sectional variation in rural juvenile homicide rates (Osgood
Love Thy Neighbor? / 1005
& Chambers 2000). Yet there are several compelling reasons to devote research attention to this issue.
Consider first the fact that several substantially rural states have some of the highest juvenile homicide rates in the nation. In 1990 the homicide offending rates per 100,000 juveniles age 14–17 were 18.79 in Georgia, 42.11 in Louisiana, 19.39 in Tennessee, 31.48 in Texas, and 24.28 in Virginia. These rates were observed in a year when the national rate for this age group was 16.2. Consider next, however, that several other rural states had extremely low rates of juvenile homicide, including Idaho (1.65), North Dakota (0.00), and Wyoming (3.71).3 This tremendous variation clearly underscores the need for further research on the structural sources of rural juvenile homicide.
Consider next that juvenile homicide in rural areas is likely to endure as a social problem. While rural areas in the U.S. are home to roughly a quarter of the population, nonmetropolitan communities have been experiencing population growth. Approximately 74% of nonmetropolitan counties experienced population growth between 1990 and 2000, resulting in a net gain of 5.6 million people (Johnson & Beale 2001). Moreover, evidence is accumulating that drug production, distribution, and use are emerging in rural areas (Donnermeyer 1992; Herz 2000). Given the longstanding concern over population change and crime (Barnett & Mencken 2000; Lee, Martinez & Rosenfeld 2001; Shaw & McKay 1942) and the more recent concerns expressed over drugs and crime (discussed below), it is likely that juvenile crime will continue to be a problem in rural areas.
Theoretical Models
CONCENTRATED DISADVANTAGE
The bulk of research on juvenile homicide has focused on two primary explanations: concentrated disadvantage and drug markets. A great deal of research in the last two decades has examined the links between structurally embedded disadvantage and rates of homicide. From a social disorganization perspective, socioeconomic disadvantage may undermine community social control (Bursik 1988; Sampson & Groves 1989; Shaw & McKay 1942), while from a strain perspective it may generate frustration and anomie (Blau & Blau 1982; Merton 1938; Williams & Flewelling 1988). Early research tended to focus on the association between absolute or relative deprivation (poverty or income inequality, respectively) and crime, but the publication of Wilson’s (1987) work on the concentration of multiple forms of disadvantage has led to the routine use of multidimensional disadvantage indices. Encouraged by the landmark work of Land, McCall, and Cohen (1990), which demonstrated that much of the divergent findings of prior aggregate research on homicide could be accounted for by
1006 / Social Forces 82:3, March 2004
methodological problems like multicollinearity, scholars now routinely employ multidimensional disadvantage indices consisting of several indicators of disadvantage (such as poverty, unemployment, and female-headed households). The general idea behind this research is that the effects of high levels of poverty or unemployment may be exacerbated by the presence of other forms of disadvantage. Results for research employing this tactic are very consistent, with most studies reporting that socioeconomic disadvantage indices and homicide rates are positively associated across macro-level units of analysis (see Baller et al. 2001; Land, McCall & Cohen 1990; Messner & Golden 1992; Ousey 1999; Parker & McCall 1999; Rosenfeld, Messner & Baumer 2001).
Although almost all the research in this tradition is based on urban units of analysis, the available evidence (though somewhat mixed) suggests that socioeconomic disadvantage and homicide are positively associated in rural areas as well. For example, Kposowa and colleagues (Kposowa & Breault 1993; Kposowa, Breault & Harrison 1995) report that poverty and homicide are positively related across rural counties. In contrast, Petee and Kowalski (1993) find no association between the percentage of low-income households in the population and violent crime, and Osgood and Chambers (2000) report no association between poverty and juvenile homicide in rural counties. Recent studies using the now conventional multidimensional disadvantage indices discussed above, however, have tended to detect the expected positive association between concentrated disadvantage and homicide outside metropolitan areas (Barnett & Mencken 2000; Lee, Maume & Ousey 2003; Lee & Ousey 2001).
DRUG MARKETS
The second leading explanation employed to account for cross-sectional variation in juvenile homicide rates explores the influence of open-air drug markets (Blumstein 1995). In the mid-1980s crack — a highly addictive and cheap cocaine derivative — was introduced to American streets. Demand for this drug skyrocketed, leading to its distribution on street corners at a level far exceeding that of previous drug epidemics such as heroin. Concomitant with the increasing use of crack cocaine was the inflation of juvenile homicide rates. Although many observers asserted unequivocally that crack and violence were causally connected, several distinct explanations for the crack–homicide link emerged (see Goldstein 1985). The most relevant structural explanation is the systemic violence model (Goldstein 1985). The systemic model argues that in the competitive world of illegal crack markets, where the police cannot be relied on to resolve disputes, violence is often used as a means of regulating competition, ensuring product quality, and protecting the business and turf of drug dealers. The level of violence in drug markets may be augmented by the fact that they are often staffed by young people with few ties to conventional institutions of social mobility and social control, who have ready access to
Love Thy Neighbor? / 1007
handguns, and who lack fundamental dispute resolution skills and impulse controls. The growing body of research examining intercity differences in the presence of drug markets and levels of lethal violence for the most part provides support for the systemic model (Baumer 1994; Baumer et al. 1998; Cork 1999; Ousey & Lee 2002).
The drug market explanation for juvenile violence is interesting and appears to hold scientific merit, but all the studies (both empirical and theoretical) testing this model share a common limitation: their inductively generated expectations are derived from observations on urban drug markets. Although Weisheit and Donnermeyer (2000:320) report that “nonmetropolitan 12th graders in 1995 had similar rates for past-year use of inhalants and powder cocaine, and slightly higher rates for crack cocaine, stimulants, barbiturates, and tranquilizers,” the nature and extent of rural drug markets essentially remains a black box for the time being. Yet we consider it fair to suggest that where there are drug markets in rural areas, they deviate substantially from those normally found in large urban areas. The distinctiveness of rural drug markets stems from the ecological layout of rural communities, which are not likely to be conducive to the street-level dealing widely observed in urban locales, the posited source of much urban violence. Hence, while we expect a positive association between concentrated disadvantage and juvenile homicide in rural areas, we expect a rather weak, if not nonexistent, association between drug market indicators and juvenile homicide in rural communities.
The Neglected Model: Moral Communities, Civic Engagement, and Crime
Although useful, the concentrated disadvantage and drug market explanations for juvenile violence have been applied to the exclusion of another potentially powerful explanation for cross-sectional community-level variation in violence — namely, the moral communities thesis. This thesis asserts that communities characterized by a proreligious climate will experience fewer crime problems. By introducing this framework, we hope to broaden the perspective of scholars beyond a narrow focus on community deficits (such as family breakdown, economic underdevelopment, and drug markets) to the repository of cultural resources that communities may use to their collective benefit (in this case, those provided by religious organizations; Lee & Bartkowski 2004). Such a shift in focus is well overdue because sociologists of religion are themselves engaged in a debate about the extent to which religious values effectively deter people, especially young people, from engaging in illicit and deviant activities.
Evidence on the inverse relationship between religious activity and juvenile delinquency is mixed (for recent reviews, see Baier & Wright 2001; Donahue & Benson 1995; Johnson et al. 2000; Johnson et al. 2001; Regnerus 2000; Stark 1996).
1008 / Social Forces 82:3, March 2004
Most germane to the focus of our study, there remains some disagreement about the means and extent to which religion exerts a deterrent effect on juvenile crime, with some scholars debating whether protective effects of religion exist at the ecological or contextual level (e.g., Chadwick & Top 1993; Cochran & Akers 1989; Elifson, Petersen & Hadaway 1983; Higgins & Albrecht 1977; Regnerus 2000; Stark 1984, 1996; Stark, Kent & Doyle 1982). To complicate matters further, several studies have shown that religion exerts a deterrent effect primarily against those types of action that are strongly condemned by faith communities (e.g., antiascetic behaviors such as substance abuse and premarital sex among teens; Burkett & White 1974; Cochran & Akers 1989 — see Baier & Wright 2001 and Johnson et al. 2000 for reviews). Yet even evidence in this line of inquiry is conflicted, in that some studies point to broadly protective effects of religiosity and others suggest effects that are more specific to the type of behavior in question (e.g., criminal act versus status offense; see Cochran 1988; Cochran & Akers 1989).
Given these debates, authors of current reviews and empirical studies have urged scholars to pay careful attention to measurement issues — i.e., the conceptualization of religion and the specification of particular types of delinquency — as the key in clarifying the nature of this complex relationship (Johnson et al. 2000; Sloane & Potvin 1986; Stark 1996). Stark (1996) has suggested that debates in the research literature result from faulty efforts at operationalizing social context and religious ecology. He argues convincingly for spatial context — namely, five broad regions of the U.S. — as the primary means by which moral communities deter criminal behavior (see also Stark 1984). To this end, Stark encourages sociologists to “discard the assumption that religion is primarily an individual trait . . . and substitute the assumption that religion is, first and foremost a social structure . . . a group property . . . [namely,] the proportion of persons in a given ecological setting who are actively religious” (Stark 1996:164, emphasis in original; see also Stark 1984). Drawing largely on survey data from the Study of High School and Beyond, Stark (1996) finds striking regional variations in high school seniors’ self-reports of having trouble with the law. The “hellfire effect,” as Stark has long called this protective influence (see Hirschi & Stark 1969), does not hold in irreligious climates (the Pacific West), is robust in highly religious parts of the country (the South), and strikes a middle range in modestly religious locales (the Mountain region). Stark notes that these results lend strong confirmation to his definition of religion as a “group property.” In this way, then, religion seems to serve as the institutional framework around which moral communities are sustained.
Yet there are several issues left unaddressed by this pioneering effort. First, despite the strength of his findings, Stark (1996:168) concedes that it is “impossible to demonstrate empirically . . . why the relationship between religion and delinquency waxes and wanes” (emphasis in original). In other words, the specific
Love Thy Neighbor? / 1009
mechanism through which religious communities deter delinquent behavior remains unexplored. In this sense, Stark’s study is more exploratory than explanatory.
Second, while Stark argues strongly for an ecological approach to studying religion, his use of survey data to do so leaves much to be desired. As is well known by demographers and others who regularly conduct ecological analyses, survey data on “having trouble with the law” collected from a select sample of the population — in the case of Stark’s study, high school seniors — do not readily lend themselves to community-level analyses. If religious organizations in fact promote moral communities, then researchers should use the community itself as the unit of analysis and explanation (see Bartkowski, Howell & Lai 2002; Lee & Bartkowski 2004). Clearly, the data most appropriate for studying moral communities are ecological in nature, rather than individual-level responses drawn from a sample of survey respondents.
Third, Stark himself concedes that his operationalization of social context as five regions of the U.S. is crude and begs for refinement. Here again, if scholars are to learn about religiously based moral communities, then great care needs to be taken concerning the boundaries and character of such communities. Given subregional social and cultural variations in the South as well as other parts of the country, it is difficult to speak of any region in the U.S. as a homogeneous community. Since communities in the U.S. are typically defined by more local boundaries, there is much to be gained from defining moral communities more narrowly.
Finally, in its current form, the moral communities thesis, while speaking directly to juvenile crime, is incapable of addressing another important issue — namely, identifying the types of persons that juveniles are most likely to victimize. From a theoretical standpoint, it is desirable to delineate as clearly as possible the scope conditions under which the theoretical perspective is most applicable. Given that heterogeneity in the victim/offender relationship is evident among juvenile crimes, subtle yet important variations in the explanatory power of the moral communities thesis across victim type may be expected. Although prior research generally has not extended the moral communities thesis to the issue of juvenile offenders and whom they victimize, much might be gained from doing so. Note also that this goal of theoretical precision is consonant with a broader trend in macrocriminological research, where analysts are increasingly moving toward more specificity in both their theoretical frameworks and their outcomes of interest (see Jacobs & Wood 1999; Parker 2001; Parker & McCall 1999; Steffensmeier & Haynie 2000a, 2000b).
The civil society literature provides theoretical tools for refining the moral communities thesis and addressing the shortcomings evident in prior research. Civil society theorists highlight the role of civic engagement and its correlate, social capital, in promoting a vibrant public sphere and fostering integrative social bonds. Recent scholarship has highlighted the immense benefit derived from civic
1010 / Social Forces 82:3, March 2004
engagement and high stocks of social capital for Americans at large and U.S. youth in particular (see Baron, Field & Schuller 2000; Lin 2001; Portes 1998; Putnam 2000). Within this literature, social capital is broadly defined as integration within social networks that promote norms of reciprocity, trust, and an ethic of civic engagement. Thus, networks, norms, and trust are considered by many leading scholars to be the three defining features of social capital. When working in concert, this triad promotes social integration, effective action, and the pursuit of the collective good (Putnam 2000).
On the heels of these theoretical advances, a nascent literature within the sociology of religion has begun to examine the ways in which religious organi- zations promote civic engagement and generate robust forms of social capital (e.g., Ammerman 1997; Bartkowski & Regis 2003; Cnaan 2002; Lee & Bartkowski 2004; Wuthnow 1999; see Sherkat & Ellison 1999 for a review). Religious congregations and faith-based organizations have a long history of facilitating dense social ties in local communities via the worship services, youth groups, special events, and social ministry programs they offer to disadvantaged citizens. Moreover, religious organizations often imbue their social norms and networks with theological dictates and scriptural edicts. The organizing prin- ciples and ethical frameworks that govern religious culture — ideologies of faith and deity, as well as moral codes rooted in sacred texts and communal traditions — are markedly distinctive from those found in secular voluntary associations (e.g., Bartkowski 2001; Bartkowski & Regis 2003; Becker 1999). Yet many researchers have noted that religious involvement promotes secular vol- unteering and engagement with the broader civil society as well (Becker & Dhingra 2001; Greeley 1997; Skocpol 2000; Smidt 1999; Wuthnow 1999).
One element of the civic engagement thesis advanced by Tolbert and colleagues (1998) and Putnam (2000) is that civically engaged religious denominations — those whose members are actively involved in other community organizations and affairs apart from the core religious activities — provide benefits to communities (see also Ammerman 1997; Wuthnow 1999). Broad-based religious commitment strengthens local social networks, elevates interpersonal trust, and helps to cultivate common community-level norms and values. While there has been much speculation to the effect that communities possessing large stocks of social capital have lower crime rates, empirical evidence on this point is just beginning to emerge (Rosenfeld, Messner & Baumer 2001), and we are aware of no published empirical studies assessing the relationships between civically engaged religious denominations and crime. This shortage of research is largely attributable to the fact that the theoretical mechanisms by which civic engagement would explain crime have rarely been explicitly articulated. Yet both the systemic social disorganization perspective and the community-level strain (institutional anomie) perspective suggest that these factors should be associated with lower crime rates.
Love Thy Neighbor? / 1011
From a systemic social organization perspective (Bursik & Grasmick 1993; Kasarda & Janowitz 1974; Sampson 1988; Sampson & Groves 1989), religious communities may foster horizontal social networks among community members. Social disorganization theory in sociology has long relied on the notion that strong social networks enable communities to stave off criminal behavior among their residents because the community members are able to formulate collective goals and cooperatively work toward realizing them (see Bellair 1997, 2000; Bursik 1988; Sampson & Groves 1989; Sampson, Raudenbush & Earls 1997). This normative social control argument assumes that civically engaged populations should function with shared normative understandings of acceptable and unacceptable behavior. Variation across communities in their crime rates may then partially be a function of variation in the level of civic engagement exhibited by their members.
From a community-level strain or institutional anomie perspective (Agnew 1999; Messner & Rosenfeld 1997), overwhelming cultural pressures to achieve financial success or status at any cost may be attenuated or mediated by ties to noneconomic civically engaged institutions (see also Chamlin & Cochran 1995; Messner & Rosenfeld 1997; Piquero & Piquero 1998; Savolainen 2000). These positive social outcomes may be traced to the way in which civic engagement promotes an ethic of helping others through volunteering and associational participation instead of taking on an extreme individualistic orientation. The presence of civically engaged religious denominations may then attenuate or moderate criminogenic pressures induced by an overwhelming cultural emphasis on the primacy of economic success at any cost. Put another way, widespread commitment to civic engagement may alleviate some of the materialist pressures engendered by a strong emphasis on economic success.
There is also strong reason to expect widespread civic engagement to affect subtypes of crime in different ways. Specifically, there are theoretical reasons to expect that homicide rates based on relational distance between victims and offenders (i.e., family, acquaintance, and stranger homicides) will have different levels of association with any measure of civic engagement. As Coleman (1988:100–101) notes, social capital (a latent construct, of which civic engagement is one dimension — see Rosenfeld, Messner & Baumer 2001) exists “in the relations among persons.” Hence, people who are not connected can accumulate no social capital because there is no relation among them (i.e., a diffuse network without closure). Further, Putnam (1993:3) argues that “social capital typically consists in ties, norms, and trust transferable from one setting to another” and concludes that “those who have social capital tend to accumulate more — them as has, gets.” In short, because there are not networks, norms, and trust among people who do not know each other, they should derive no benefit from living in a civically engaged community, whereas those who do know each other should derive benefit from it.
1012 / Social Forces 82:3, March 2004
It is important to point out that we are not making the ecological fallacy here (see Robinson 1950). Our analytical interpretation does not assume that homicide offenders are civically engaged or that their victims are civically engaged. To the contrary, we argue that communities exhibiting high levels of civic engagement will have lower average rates of juvenile violence committed against family members because the normative social control processes central to our perspective operate through social networks and cannot produce benefits outside the network structure. Hence, we would expect community levels of civic engagement to have a depressing effect on the volume of crime among people who know each other, because on average said communities will have strong networks, norms, and trust among those having relational associations but not among those lacking such associations.
Summary and Expectations
Taken together, a civic engagement perspective on the moral communities thesis is consonant with and significantly extends at least two main theoretical perspectives in macrosociology. Moreover, considering the potential for a civic engagement perspective to explain some subtypes of crime better than others, we expect that widespread participation in civically engaged religious denominations should provide particularly strong protective effects against juvenile family homicides, as opposed to juvenile acquaintance or stranger homicides. Hence the following empirical analyses test the working hypothesis that juvenile homicide rates will be lower in rural communities having a larger proportion of their population base associated with civically engaged denominations. Further, given the institutional linkages between religion and family, we anticipate that this protective effect will be particularly salient for juvenile family homicide rates.
Data and Measures
In order to test our main expectation that rural juvenile homicide rates will be lower where a larger proportion of the population adheres to civically engaged religious denominations, we employ counties as our unit of analysis. Counties are reasonable proxies for rural “communities” because their population size is similar to that of urban neighborhoods (see Osgood & Chambers 2000). In addition, counties are widely used in macro-level criminological research (see Baller et al. 2001; Kposowa & Breault 1993; Lee & Ousey 2001; Osgood 2000; Osgood & Chambers 2000) and the data necessary to test our hypothesized link between faith-based civic engagement and juvenile homicide is available for counties but not for other levels of analysis such as central cities. In addition, Land, McCall, and Cohen (1990) and Parker, McCall, and Land (1999) assert that the selection of macro-level units of analysis in aggregate research is
Love Thy Neighbor? / 1013
essentially arbitrary because there tends to be convergence in findings across studies employing various units of analysis. Because we are particularly interested in rural juvenile homicide rates, we base our selection criterion on population size and select those counties with a population less than or equal to 20,000 people.4 For comparative purposes, we also examine a sample of urban counties, employing the commonly used criterion in the literature of counties having a total population base of 100,000 or more people. All measures described below are derived from data sources around the year 1990.
DEPENDENT VARIABLES
We measure juvenile homicide with data drawn from the Supplementary Homicide Reports offender file (Fox 2000). Because we employ various disaggregation techniques (i.e., by age and type of relationship), we follow convention in the literature and select only those cases having a single offender and a single victim (Peterson & Krivo 1993; Williams & Flewelling 1988). To be consistent with other studies, we operationalize juvenile homicide as the pooled (summed) number of juvenile homicides from 1990 to 1992 committed by offenders under the age of 18 (see Osgood 2000; Osgood & Chambers 2000; Rosenfeld, Messner & Baumer 2001). In addition, we identify the nature of the relationship between juvenile offenders and their victims and partition them into juvenile family, juvenile acquaintance, and juvenile stranger homicides. Limitations of these data have been discussed at some length elsewhere (see Maxfield 1989; Reidel 1999). Nevertheless, they remain widely used by researchers, and to our knowledge they are the only data available providing a reasonable degree of national coverage on juvenile homicide offenders and whom they victimize.
KEY EXPLANATORY VARIABLE
Our key explanatory variable is a measure of the proportion of the population adhering to civically engaged denominations. Data used to construct this measure are drawn from county-level measures found in the Census of Churches produced by the Glenmary Research Center (Association of Statisticians of American Religious Bodies 1992), which is featured online in the American Religion Data Archive <www.arda.tm>. This data set provides detailed information on more than 100 denominations, the number of churches and adherents. We use the county-level measures found in this data set. We use the same method as Tolbert, Lyson, and Irwin (1998) to identify civically engaged denominations and express the number of adherents to these denominations per capita as our measure (see also Lee & Bartkowski 2004). Briefly, Tolbert, Lyson, and Irwin (1998) use data from the General Social Survey to estimate
1014 / Social Forces 82:3, March 2004
the level of civic engagement exhibited by members of various religious denominations. As they describe,
we take data on individuals’ denominational affiliation and number of voluntary associations and other group memberships. In the GSS, the aggregate number of memberships item sums affiliations with groups such as fraternal organizations, service clubs, labor unions, sports clubs or teams, hobby or garden clubs, and professional or trade associations. . . . We identify as civically engaged those denominations (or other forms or religious organizations) whose adherents in our GSS sample reported an above-average number of voluntary association membership (i.e., > 1.58). (p. 424)5
CONTROL VARIABLES
By necessity we employ a number of control variables that prior research has identified as potentially important covariates of homicide rates. First is a disadvantage index, constructed as the average of the standardized scores of the percentage in poverty, the percentage unemployed, the percentage over age 25 with less than a high school education, the percentage of the population that is black, the percentage of female-headed households, and the degree of spatial concentration of the county’s poor residents.6 In addition, measures of the size of the population at risk (the juvenile population), the percentage of adults divorced, and the proportion of the population in the 15–29 age range (the most crime-prone age group) are constructed. Because a good deal of research has been conducted on the unique aspects of conservative Protestant culture and family life (Ellison 1991; Ellison, Bartkowski & Segal 1996a, 1996b; Ellison & Sherkat 1993), we control for the possibly confounding effects of a large conservative Protestant population in the county with a measure of the proportion adhering to a conservative Protestant denomination (also derived from the Glenmary data).7 Moreover, taking into account the literature suggesting a link between the presence of drug markets and lethal violence discussed above, we construct a measure of juvenile arrests for possession of cocaine and opiates as the average rate for the three-year period 1990– 92.8 We also account for potential regional effects identified in some research with an indicator variable coded 1 for counties in the South and 0 otherwise. Finally, there is the possibility that any observed association between faith-based civic engagement and crime may be due to a nonreciprocal relationship. That is, high rates of juvenile homicide may actually undermine the degree of faith-based civic engagement, in much the same way that crime is often cited as a cause of depopulation and community instability more generally (Liska & Bellair 1995; Morenoff & Sampson 1997). To try to address this possibility, we control for the relevant lagged (1980) juvenile homicide rates, constructed in the same fashion as the dependent variables. The final samples consist of 1,440 rural counties and 449 urban counties.
Love Thy Neighbor? / 1015
Analytical Method
Because homicide is a rare event in rural areas, and is actually also rare relative to population size in most urban areas, our data exhibit extremely skewed distributions and subsequently violate the assumptions of ordinary least squares (OLS) regression. Current convention in the literature dealing with rare event data is to employ one of a variety of Poisson class estimators, the choice depending on the nature of the data (Osgood 2000). In the presence of events that are rare relative to population size, a Poisson estimator provides several advantages over traditional OLS estimation. First, in light of low crime counts and a small population base, such data are typically not normally distributed, violating a crucial assumption of OLS estimation. Second, as Osgood (2000) notes, crime rates computed with a small population base as the denominator rapidly become imprecise and unstable as the size of the denominator decreases. Third, such data are usually heteroskedastic because the error variance decreases as the population size of the units of analysis increases. Poisson estimation provides a means of overcoming these problems.
There are two additional issues related to our estimation technique. First, the Poisson estimator assumes that the mean and variance of the outcome of interest are equal. When this assumption is violated, the variable is said to be overdispersed, and a modification to the estimator is required that allows the introduction of an error term. The negative binomial estimator is widely employed to deal with this problem. Our data exhibit this quality, and so the negative binomial estimator is the model we employ here. Second, while Poisson class estimators typically are used to model counts of rare events, they can also be used to model rates of rare events by specifying the logged size of the population at risk as an offset variable, which constrains the population size for all counties to 1. Taking these two issues into account, we therefore employ the negative binomial estimator to predict the logged rate of juvenile homicide across U.S. counties.9
Results
We begin our analysis with a description of our key variables of interest, the juvenile homicide rates and the faith-based civic engagement indicator, which are presented in Table 1.10 In the rural sample, the mean juvenile homicide rate is roughly 3.1 homicides per 100,000 juveniles, although as the standard deviation of 14.31 suggests, there is tremendous variability across counties in this measure, underscoring our assertion that rural communities differ widely in the level of lethal violence committed by juveniles. In addition, juvenile acquaintance homicide is more frequent than the other relationship-specific types in rural communities, although all exhibit a good deal of variation, as is evident in their standard deviations. As expected, the urban juvenile homicide
1016 / Social Forces 82:3, March 2004
TABLE 1: Descriptive Statistics
1,440 Rural Counties 449 Urban Counties
Mean S.D. Mean S.D.
Dependent variables
Juvenile-family homicide rate/100,000 .81 6.82 .97 1.65 Juvenile-acquaintance homicide rate/100,000 1.73 10.83 4.73 7.03 Juvenile-stranger homicide rate/100,000 .28 3.65 1.56 2.95 Total juvenile homicide rate/100,000 3.11 14.31 7.98 10.63
Key explanatory variable Civically engaged denominations .17 .13 .12 .08
Control variables Poverty rate .19 .08 .12 .05 Unemployment rate .07 .04 .06 .02 High school dropouts .33 .11 .22 .07 Proportion black .07 .15 .11 .12 Female-headed households .05 .03 .06 .02 Poverty concentration .22 .09 .22 .09 Juvenile population (ln)a 2,863.16 1,533.94 105,417.06 177,508.5 Proportion divorced .07 .02 .08 .02 Proportion age 15–29 .19 .04 .24 .04 South .43 — .35 — Drug arrest rate (ln)a 1.65 12.70 23.03 38.05 Conservative Protestant .48 .23 .30 .16 Juvenile-family homicide rate, 1980 1.13 8.37 1.14 1.99 Juvenile-acquaintance homicide rate, 1980 1.78 10.98 2.92 3.96 Juvenile-stranger homicide rate, 1980 .68 7.30 1.07 2.33
Total juvenile homicide rate, 1980 3.62 15.70 5.63 7.37
a Variable transformed to natural logarithm in multivariate analysis because of high skewness (see text).
rate is much higher, and the pattern observed in rural areas — that acquain- tance homicides are the most frequent — also holds true in urban areas. Our key measure of moral communities, the proportion adhering to civically en- gaged denominations, indicates that on average 17% of the population in ru- ral areas can be considered civically engaged, with a standard deviation of 13%. In contrast, the prevalence of faith-based civic engagement in urban areas is markedly lower, and there is on average less variation across communities, as indicated by the standard deviation.
Turning to the multivariate analysis, model 1 in Table 2 presents the results from the baseline negative binomial model predicting the rural juvenile homicide rate with the vector of control variables. As expected, the presence of overlapping forms of socioeconomic disadvantage captured in our disadvantage index is associated with higher juvenile homicide rates. The coefficient for this variable indicates that a one standard deviation increase (.79) in the level of disadvantage results in a 34.9% increase in the juvenile homicide
Love Thy Neighbor? / 1017
rate, a relatively substantial effect.11 It is interesting that none of the other coefficients in the model attains statistical significance. Model 2 of Table 2 extends this specification with the introduction of our main indicator of interest, the faith-based civic engagement variable (i.e., the proportion of the population affiliated with a civically engaged denomination). As expected, the faith-based civic engagement variable exhibits a statistically significant negative effect on juvenile homicide rates. Like the effect of the disadvantage index in model 1, the effect of faith-based civic engagement on juvenile homicide is relatively sharp, with a one standard deviation increase in adherence to civically engaged religious denominations (13%) translating into a 28.7% decrease in the juvenile homicide rate. In common terms, rural counties with a larger proportion of their population adhering to civically engaged religious denominations experience lower juvenile homicide rates. We also note that with the introduction of the measure of civically engaged denominations, the coefficient for the disadvantage index declines by about 12.4%, and this translates into a standardized effect size of 29.9% (as opposed to 34.9% in the previous model).
Models 3 and 4 in Table 2 report the same specifications for the sample of urban counties. There are several noticeable differences between these results and those for the rural sample. As model 3 indicates, juvenile homicide rates are higher in urban counties experiencing higher levels of disadvantage, a larger juvenile population, a higher divorce rate, a larger proportion of the population in the crime-prone age group, and where there is more drug market activity. It is notable that urban juvenile homicide rates are also associated with the lagged juvenile homicide rates. And as model 4 indicates, our measure of faith- based civic engagement exhibits no association with the urban juvenile homicide rate. As we noted above, this finding is not surprising, especially in light of the fact that it is clear from these models that concentrated socioeconomic disadvantage and drug activity play important roles in explaining cross-sectional variation in the prevalence of juvenile homicide in urban areas. It is also notable from these models that model fit is much higher for the urban sample than for the rural sample.
The basic specification reported above generally supports our expectation; however, researchers are increasingly recognizing the importance of moving toward specificity in measuring homicide rates (see Parker 2001). While the operationalization of juvenile homicide in Table 2 does capture what we intend, there are subcategories of juvenile homicide as well, and as discussed above, our reading of the literature suggests that the effects of faith-based civic engagement on them may not be constant. Indeed, as we argue above, there is reason to believe that religiously based civic engagement may have particularly strong protective effects with respect to family homicides because of the paramount importance that religious groups in the U.S. place on family cohesion and the socialization of youngsters (e.g., Abbott, Berry & Meredith 1990; Bartkowski 2001; Bartkowski
1018 / Social Forces 82:3, March 2004
& Wilcox 2000; Bartkowski & Xu 2000; Christiano 2000; Eberly 1999; Pankhurst & Houseknecht 2000; Wilcox 2002; Wilcox & Bartkowski 1999). Moreover, because people who are situated outside closed social networks cannot share norms and trust (Coleman 1988), they should derive considerably less benefit from residing in a community characterized by a preponderance of civically engaged denominations. To test this expectation, Table 3 presents the results of our models predicting juvenile family, acquaintance, and stranger homicide rates in rural counties with our main explanatory variable of interest, civically engaged denominations, and our vector of control variables.
The results of these models are rather stark and straightforward. In model 1 it is apparent that the measure of civically engaged denominations has the expected effect on juvenile family homicide rates. The statistically significant
TABLE 2: Negative Binomial Regression Models Predicting Total Juvenile Homicide Rates, Urban and Rural Counties*
Rural Counties Urban Counties
Model 1 Model 2 Model 3 Model 4
Civically engaged denominations — –2.602* — .123 (1.242) (.593)
Disadvantage index .379* .332* .927** .929** (.152) (.152) (.083) (.084)
Juvenile population size (ln) .083 .111 .226** .226** (.246) (.249) (.063) (.063)
Divorce rate 6.379 4.670 .107** .107** (6.582) (6.503) (.027) (.028)
South .206 .284 .049 .045 (.311) (.309) (.109) (.111)
Age 15–29 1.903 .830 .029* .029* (3.206) (3.156) (.013) (.013)
Drug arrest rate (ln) .092 .062 .112** .112** (.131) (.130) (.031) (.031)
Conservative Protestant .048 –.281 .417 .415 (.565) (.573) (.343) (.343)
Lagged juvenile homicide rate .112 .114 .198** .197** (.099) (.089) (.059) (.059)
�2 b (overdispersion parameter) 40.19** 39.08* 991.99** 991.72**
Normed maximum likelihood R2 .036 .044 .542 .542 Likelihood ratio test 21.43** 26.45** 348.08** 348.12
a Standard errors in parenthesses. b Likelihood Ratio test of � =0 (i.e., no overdispersion). Significant value indicates data overdispersed
and negative binomial model is appropriate.
* p < .05 ** p < .01
Love Thy Neighbor? / 1019
negative coefficient indicates that a one standard deviation increase in the proportion adhering to civically engaged religious denominations is associated with 57.20% decrease in the juvenile family homicide rate. In contrast to model 1, the only variables achieving significance in model 2 predicting juvenile acquaintance homicide are the disadvantage index and the divorce rate, whereas in model 3 predicting stranger homicides, none of the variables exhibits an effect significantly different from zero, and the likelihood ratio test statistic, as would be expected, is insignificant. In short, these models provide another layer of support for our argument, while at the same time highlighting the importance of disaggregating homicides along important dimensions.
Table 4 reports the results of parallel analyses on the urban sample. As expected, the measure of civically engaged religious denominations has no association with any of the disaggregated homicide rates in urban areas.
TABLE 3: Negative Binomial Regression Models Predicting Juvenile Family, Acquaintance, and Stranger Homicide Rates, 1,440 Rural Counties*
Model 1 Model 2 Model 3
Family Acquaintance Stranger
Civically engaged denominations –6.528* –1.899 1.095 (2.962) (1.520) (3.613)
Disadvantage index .154 .444* .835 (.327) (.188) (.502)
Juvenile population size (ln) .227 .196 1.608 (.486) (.323) (1.124)
Divorce rate –15.943 19.352* –21.773 (13.003) (8.489) (24.357)
South 1.109 .085 .788 (.603) (.404) (1.155)
Age 15–29 –15.650 2.320 –5.365 (10.290) (3.680) (15.097)
Drug arrest rate (ln) –4.552 .104 .174 (567.220) (.154) (.327)
Conservative Protestant .315 –1.070 1.415 (1.038) (.760) (2.169)
Lagged juvenile homicide rate –.015 .024 –3.926 (.325) (.182) (322.412)
�2 b (overdispersion parameter) 5.78** 17.87** 14.28** Normed maximum likelihood R2 .066 .058 .109
Likelihood ratio test 16.90* 24.28** 14.45
a Standard errors in parenthesses. b Likelihood Ratio test of � = 0 (i.e., no overdispersion). Significant value indicates data overdispersed
and negative binomial model is appropriate.
* p < .05 ** p < .01
1020 / Social Forces 82:3, March 2004
However, several other important variables perform as expected. For example, concentrated disadvantage clearly plays an important role in driving all forms of juvenile homicide in urban areas. Similarly, the measure of drug market activity is positively and significantly associated with all forms of juvenile homicide. Taken together, the effects of the disadvantage index and the drug market measure appear to provide support for the two main theoretical explanations of urban crime that have surfaced in recent years. It is also notable that increases in the rate of divorce in urban areas contributes to higher juvenile homicide rates.
It is important to probe the sensitivity of our results to alternative model specifications, especially since the effect of our main indicator of interest, civically engaged religious denominations, varies across type of juvenile homicide. For the models reported above, a number of diagnostic measures
TABLE 4: Negative Binomial Regression Models Predicting Juvenile Family, Acquaintance, and Stranger Homicide Rates, 449 Urban Counties
Model 1 Model 2 Model 3 Family Acquaintance Stranger
Civically engaged denominations .138 .422 –.319 (.874) (.722) (1.011)
Disadvantage index .497** .923** .976** (.093) (.099) (.125)
Juvenile population size (ln) .033 .180* .386** (.069) (.076) (.102)
Divorce rate 12.224** 9.978** 12.265** (3.798) (3.369) (4.366)
South –.134 –.017 –.006 (.156) (.137) (.187)
Age 15–29 3.250 3.929* .678 (2.053) (1.611) (2.420)
Drug arrest rate (ln) .157** .154** .135** (.056) (.038) (.052)
Conservative Protestant 1.608** .534 –.085 (.519) (.426) (.572)
Juvenile homicide rate .031 .216** .067 (.107) (.073) (.119)
�2 b (overdispersion parameter) 8.21** 854.53** 320.34**
Normed maximum likelihood R2 .220 .448 .338
Likelihood ratio test 97.67** 262.93** 172.73**
* p < .05 ** p < .01
Love Thy Neighbor? / 1021
were implemented, primarily to probe for multicollinearity and omitted variable bias. First, we reexamined the models reported above using an OLS estimator to secure variance inflation factors (VIFs) and tolerance estimates. The results of these models indicate no signs of multicollinearity, as all VIFs were below 2.0, well below the usual criterion of 4.0 employed in macro-level research and the conservative criterion of 2.5 suggested by Allison (1999). A series of auxiliary models were also estimated to ensure that the observed effect of faith-based civic engagement on juvenile family homicide was not due to sample composition and that it was not accounted for by the introduction of other potentially important covariates into the model. The results from the auxiliary models are presented in Table 5.
For example, in model 1 we report the coefficient and standard error for our civic engagement measure derived from a full model (reported in Table 3) with a 10% random sample deletion. This change in our sample has little effect on the size of the coefficient or the standard error. In model 2, we follow Osgood and Chambers (2000) and introduce a dummy variable coded 1 for rural counties adjacent to urban counties, and 0 otherwise.12 The introduction of this variable could capture spatial autocorrelation or detect a spatial spillover effect wherein criminal activity or criminal norms and values filter over to the countryside. Despite the theoretical integrity of this measure, it does not wipe out the main effect of the faith-based civic engagement variable when predicting juvenile family homicides.
In addition, some research indicates that crime rates are lower where welfare expenditures are higher or where social altruism is higher in the form of giving to the United Way (Chamlin & Cochran 1997; DeFronzo 1997; Hannon & DeFronzo 1998). To address the potential of this variable to also account for the effects of faith-based civic engagement (because civic engagement may actually be higher where there is a social climate of concern for the welfare of fellow citizens; Cnaan 2002), we included a measure of public assistance and report the effect of civically engaged denominations in the presence of this measure in model 3. Again, however, this potentially key variable failed to account for the negative relationship between civically engaged religious denominations and rural juvenile family homicide. Then, given a longstanding tradition in sociology concerned with the effects of population turnover on crime, we reestimated the model introducing this measure, but as model 4 demonstrates, this has little effect on the relationship between civically engaged denominations and juvenile family homicide. In models 5 and 6 we take into account the local religious climate more generally by including variables measuring the proportion of the local population adhering to churches of any kind and a measure of access to churches (number of churches per 1,000), but these too have little effect on our substantive outcome of interest.13 Finally, in models 7 and 8 we introduce measures of secular forms of civic engagement, average voter turnout for the 1988 and 1992 presidential
1022 / Social Forces 82:3, March 2004
elections and a measure of the number of social and civic organizations per capita in the county in 1990.14 However, neither variable accounts for the observed relationship between our main measure of civically engaged denominations and juvenile family homicide. In sum, our results indicating that juvenile family homicide in particular is lower where local populations are more invested in civically engaged denominations appear to be rather robust.
Discussion and Conclusions
This analysis was designed to address the intersection of several significant voids in prior macro-level research on homicide. First, although all indicators suggest juvenile violence was a major social problem around 1990, few empirical studies have examined the structural antecedents of juvenile violence. Second, while a sizeable body of research on aggregate rates of violence has been conducted employing urban areas as units of analysis, the literature on rural violence is comparatively slim. Third, despite what we consider a high level of theoretical integrity, our reading of the literature suggests that no studies to date have provided an adequate empirical test of the moral communities thesis with respect to homicide, and only one study of which we are aware (Rosenfeld, Messner & Baumer 2001) has attempted to link the related concept of civic engagement from the social capital literature to macro-level homicide rates. Finally, we are aware of no direct theoretical or empirical attempts to specify a relationship
TABLE 5: Coefficients and Standard Errors for Civically Engaged Denominations Variable Predicting Juvenile Family Homicide under Alternative Model Specifications, 1,440 Rural Counties
Coefficient Standard Error
Model 1: 10% random sample deletion –6.653* 3.005
Model 2: controlling for adjacency indicator –6.402* 2.973
Model 3: controlling for public assistance –6.358* 2.955
Model 4: controlling for population turnover –6.557* 2.978
Model 5: controlling for percent adhering –6.725* 3.164
Model 6: controlling for churches per 1,000 –8.124* 3.447
Model 7: controlling for voter turnout –6.686* 3.022
Model 8: controlling for social and civic organizations –6.681* 3.000
* p < .05
Love Thy Neighbor? / 1023
between religiously based civic engagement and relationship-specific homicide rates.15
Our theoretical contribution to the research literature lies in more clearly delineating the nature of moral communities and identifying the specific mechanisms through which they may serve to depress crime rates. Drawing on the social capital and civic engagement literature, our framework posits that the effect of religious adherence is transmitted through its cultivation of social activities outside the core religious realm. The classification scheme developed by other analysts (e.g., Tolbert, Lyson & Irwin 1998) identifies certain religious denominations as more civically engaged than others. Correspondingly, we employed this measure with the expectation that communities having a substantial population adhering to such denominations would have lower juvenile crime rates, not solely because of a “hellfire effect” (the notion that religions condemn antisocial behavior), but because of the strengthening of the local social fabric engendered by faith-based civic engagement.
Our conceptualization of moral communities is broadly consistent with what Athens (1998) has termed “civil communities.” Civil communities are characterized by norms of nonviolence and the rule of law. The dominant values in civil communities place nonviolent persons in positions of power. Moreover, disputes in these communities are settled through nonviolent, juridical means. Athens contrasts civil communities with what he calls malignant communities. In malignant communities, violent persons occupy positions of dominance and illegal conduct sometimes confers status. When disputes arise in malignant communities, they are settled through violent confrontation — through the “tooth and claw” exercise of power rather than the rule of law. According to Athens, turbulent communities are characterized by a moderate degree of violence and occupy a middle ground between the poles of civil communities on the one hand and malignant communities on the other. In this typology, future research might compare the ecological role of religion in civil, turbulent, and malignant communities. It is quite possible that civil communities use religion (particularly civically engaged religious denominations) as a public resource through which they create an ecology of nonviolence and enforce a moral code of law-abiding conduct. It is also likely that religion is a largely absent civic resource in malignant communities and is only marginally present in turbulent communities. Such a perspective would dovetail nicely with the Durkheimian notion of religion as form of collective conscience.
Yet our study also suggests that overly sanguine perspectives that equate religion with a “sacred canopy” of protection from violence are flawed. The social capital literature clearly suggests that civic engagement is beneficial for those embedded in social networks. But such is not the case for those outside these networks. Consistent with social capital theory, we find that the resources engendered by faith-based civic engagement retard the volume of crime
1024 / Social Forces 82:3, March 2004
occurring between family members but fail to do so for people who are unfamiliar with one another. Where juvenile homicide is concerned, the prosocial character of religious ties are most pronounced within kin networks and lose considerable force outside them. Thus, civically engaged religious denominations provide a circumscribed “umbrella” of protection against some forms of violence, rather than a thoroughgoing “canopy” of protection of violence writ large.
Of course, it is important not to reduce our theoretical argument to the individual level. Our conceptual argument does not suggest that individual juveniles who are involved in religious activities will be less likely to kill a family member. Rather, our perspective expects communities with a large proportion of the population involved in civically engaged religious denominations to have a lower volume of family homicides committed by juveniles. This expectation is based on the premise that the normative social control and trust engendered by faith-based civic engagement should be particularly strong for those with close interpersonal relations but weaker as relational distance increases. It may also be a product of the specific linkages between religion and family, which are, after all, often cast as complementary social institutions. Given these premises, we expected faith-based civic engagement to have a less robust dampening influence on acquaintance and stranger homicide rates than on family homicide.
Our empirical analysis of rural juvenile homicide rates disaggregated by victim/ offender relationship reveals that our measure of faith-based civic engagement has a negative relationship with juvenile family homicide rates in particular. Further, this relationship appears to be rather robust, as it is stable in the face of a series of well-established covariates of homicide. In addition, our extensive sensitivity analyses, including a 10% random sample deletion and the introduction of other potentially relevant variables (including secular forms of civic engagement), indicate that the results are relatively invariant to alternative model specifications. Hence, these data indicate that rural juvenile family homicide rates are on average lower where a greater proportion of the population adheres to civically engaged religious denominations.
Still, why would these same deterrent effects not be manifested in urban areas? In explaining this disparity, we point to research that has underscored the centrality of religious institutions to rural civic life (Bartkowski & Regis 2003; Ellison & Sherkat 1995; Parisi et al. 2002), particularly for youth in nonmetropolitan communities (King, Elder & Whitbeck 1997). Religious organizations are a most critical civic institution in rural areas, where the local culture and public life are strongly informed by religious conviction and by the plethora of religious organizations in such areas. By contrast, network density in urban areas may dilute the influence (and, hence, the protective effects) of religious participation by providing an array of alternative avenues for civic engagement that are quite different in character from faith-based civic engagement. Although our data
Love Thy Neighbor? / 1025
limitations make such statements speculative, urban civic alternatives might include workplace associations and political activist groups for adults, as well as peer networks cultivated in schools, secular clubs, and urban sports leagues for youth. This is not to say that such secular alternatives are wholly absent in rural areas. But urban network density means that nonreligious forms of civic engagement are likely to be more widely utilized in metropolitan locales.
While the macro-level literature on age-disaggregated crime rates remains in an early developmental phase, the study of age-graded crime in rural areas is literally in its infancy. This degree of development is partially attributable to conceptual issues surrounding the definition of what exactly constitutes a rural locale. As we noted above, our selection criterion was intended to tap into rurality at a fundamental level while allowing for variation along its primary dimension — population size. However, other analysts may wish to consider exploring more thoroughly whether results obtained from empirical models hold across other selection criteria. Augmenting this issue is the problem of specifying meaningful age-specific theoretical models. Although some in the past have successfully delineated meaningful age-specific relationships between theoretical variables and crime rates (see Allan & Steffensmeier 1989), criminological theory in general is still evolving to reach the point where it can accommodate age-graded specifications. However, if our results are any indication, future scholarship in this area is warranted. Finally, we would encourage scholars to examine the extent to which a preponderance of civically engaged denominations in local communities might deter other forms of violent juvenile crime and nonviolent forms of adolescent delinquency as well.
The fact that we focused in part on marrying the moral communities and civic engagement literatures to explicate more thoroughly the role of faith- based civic engagement on crime at the community level should not detract from the insight that civic engagement exhibits both a secular and religious character (see Rosenfeld, Messner & Baumer 2001 for measurement of the secular character). Although our key findings held when controlling for two such secular measures (voter turnout and the number of social and civic organizations), social researchers are well advised to pursue more sophisticated measures of both faith-based and secular forms of civic engagement. In anticipation of the achievement of that goal, our study clearly underscores the significant, though circumscribed, role of religious communities in promoting a crime-deterrent civic infrastructure for American youth.
Notes
1. Following Tolbert, Lyson, and Irwin (1998), we view as civically engaged those religious denominations whose members exhibit high levels of participation in voluntary associations and other noneconomic groups (such as sports clubs). It is worth noting that the moral communities thesis underscores the prosocial character of religion on
1026 / Social Forces 82:3, March 2004
the social ecology of a locale. The converse of a moral community may best be described as what Athens (1998) calls a “malignant community” — those in which criminal elements occupy high-status positions and disputes are commonly settled though the use of violence. The closest approximation of moral communities in Athens’ typology are what he calls “civil communities.” (What Athens calls “turbulent communities” are situated in the middle of this continuum.) We revisit the relationship between religion and these communities types in the concluding section of our study.
2. Indeed, several complete volumes are now available on the problem of juvenile violence and the general topic of the crime rate increase and decrease. See Blumstein and Wallman (2000), Tonry and Moore (1998), and the 1998 summer issue of the Journal of Criminal Law and Criminology.
3. These figures are derived from data available at <www.ojp.usdoj.gov/bjs/ dtdata.htm#crime>
4. There is currently no standard criterion for identifying rural ecological areas (see Weisheit & Donnermeyer 2000). Although our selection criterion is somewhat arbitrary, we suggest that focusing on counties with populations less than or equal to 20,000 people helps us address the challenge put forth by Weisheit and Donnermeyer (2000:312): to “capture the essence of rural while also appreciating wide variations among rural areas.”
5. This method identifies the following denominations as civically engaged (listed alphabetically): African Methodist Episcopal Zion, American Baptist, Church of Christ, Congregational Christian, Disciples of Christ, Episcopal, Jewish, Latter-Day Saints, Lutheran, Methodist, Presbyterian, Unitarian (see Tolbert, Lyson & Irwin 1998).
6. The spatial concentration measure is formulated using a class-based P* index of poverty concentration derived from block-group-level data. See Bell (1954) for the computation. The block-group-level data are drawn from Summary Tape File 3A of the 1990 U.S. Census, while the other measures are drawn from Summary Tape File 3C of the 1990 U.S. Census.
An unrotated principal components analysis of these measures indicates that they load on a single dimension for both the rural and the urban samples. For the rural sample, the component matrix loadings range from .923 to .646, with an eigenvalue of 3.877 explaining 63.616% of the variance. For the urban sample, the matrix loadings range from .852 to .417, with an eigenvalue of 3.908, which explains 65.139% of the variance.
7. Adhering to convention in the religion literature, we identify the following denominations as Conservative Protestant: Baptist, Missouri Synod Lutheran/Evangelical Lutheran, Assembly/-lies of God, Christian and Missionary Alliance, Christian Reformed Church, Christian Scientist, Church of God, Church of God in Christ, Church of the Nazarene, Church of Christ, Community Churches, Evangelical Covenant, Evangelical Free, Full Gospel Fellowship, Foursquare Gospel, Jehovah’s Witness, Mennonite, Pentecostal/Pietist/Holiness, Seventh-Day Adventists, Independent Fundamentalist.
8. These data are drawn from county-level UCR arrest files. We acknowledge that this measure is less than perfect because it conflates cocaine and opiates, but it is currently the best proxy measure for drug markets available for a large sample of rural counties. See Ousey and Lee (2002) and Rosenfeld and Decker (1999).
Love Thy Neighbor? / 1027
9. For more on Poisson and negative binomial modeling, see Agresti (1996), Land, McCall, and Nagin (1996), and Long (1997). For recent examples in macro-level research, see Lee, Martinez, and Rosenfeld (2001), Lee and Ousey (2001), Osgood (2000), and Osgood and Chambers (2000).
10. So as not to confuse the reader, for our descriptive statistics in Table 1 we provide the juvenile homicide rates per 100,000 juveniles.
11. In order to interpret the negative binomial coefficients as standardized coefficients, the standard deviation of the independent variable of interest is multiplied by the raw coefficient, the product is then exponentiated, 1 is subtracted from this, and the result is multiplied by 100. The formal formula for this equation takes the form
( )[ ]{ }exp( * ) 1 * 100k ksβ − , where sk is the standard deviation of the independent variable, and �
k is the raw coefficient from the negative binomial equation. In addition,
the normed maximum likelihood R2 statistic provides an adjusted estimate of the reduction in the log-likelihood between intercept only and full models (see Long 1997). Finally, the likelihood ratio test statistic, when statistically significant, indicates that the independent variables account for more variation in the juvenile homicide rate than would be expected simply by chance.
12. This coding scheme comes from the rural–urban continuum codes (see Butler & Beale 1994).
13. These measures are derived from the same data source as the civic engagement measure.
14. The voter turnout data are derived from the USA Counties CD-ROM, and the measure of social and civic organizations are derived from the 1990 County Business Patterns data set from the U.S. census.
15. Still, it is worth noting that Rosenfeld, Messner, and Baumer (2001:301) pose the following question at the conclusion of their study: “Given the nature of the linkages between social capital and intervening causal processes, is social capital more relevant to the explanation of selected forms of homicide (e.g., stranger homicide) in comparison with other forms (e.g., intimate partner homicide)?”
References
Abbott, Douglas A., Margaret Berry, and William H. Meredith. 1990. “Religious Belief and Practice: A Potential Asset in Helping Families.” Family Relations 39:443–48.
Agnew, Robert. 1999. “A General Strain Theory of Community Differences in Crime Rates.” Journal of Research in Crime and Delinquency 36:123–55.
Agresti, Alan. 1996. An Introduction to Categorical Data Analysis. Wiley.
Allan, Emilie Andersen, and Darrell Steffensmeier. 1989. “Youth, Underemployment, and Property Crime: Differential Effects of Job Availability and Job Quality on Juvenile and Young Adult Arrest Rates.” American Sociological Review 54:107–23.
Allison, Paul D. 1999. Multiple Regression in Practice. Pine Forge Press.
Ammerman, Nancy T. 1997. Congregation and Community. Rutgers University Press.
1028 / Social Forces 82:3, March 2004
Anderson, David C. 1998. “Curriculum, Culture, and Community: The Challenge of School Violence.” Crime and Justice: A Review of Research 24:317–64.
Anderson, Elijah. 1998. “The Social Ecology of Youth Violence.” Crime and Justice: A Review of Research 24:65–104.
Association of Statisticians of American Religious Bodies. 1992. Churches and Church Membership in the United States, 1990 [computer file]. University of Connecticut, Roper Center. <www.arda.tm>
Athens, Lonnie. 1998. “Dominance, Ghettos, and Violent Crime.” Sociological Quarterly 39:673– 91.
Baier, Colin J., and Bradley R.E. Wright. 2001. “‘If You Love Me, Keep My Commandments’: A Meta-Analysis of the Effects of Religion on Crime.” Journal of Research in Crime and Delinquency 38:3–21.
Baller, Robert D., Luc Anselin, Steven F. Messner, Glenn Deane, and Darnell F. Hawkins. 2001. “Structural Covariates of U.S. County Homicide Rates: Incorporating Spatial Effects.” Criminology 39:561–90.
Barnett, Cynthia, and F. Carson Mencken. 2000. “Population Change and the Contextual Nature of Crime in Nonmetropolitan Counties.” Paper presented at the annual meeting of the Rural Sociological Society, Washington, DC.
Baron, Stephen, John Field, and Tom Schuller. 2000. Social Capital: Critical Perspectives. Oxford University Press.
Bartkowski, John P. 2001. Remaking the Godly Marriage: Gender Negotiation in Evangelical Families. Rutgers University Press.
Bartkowski, John P., Frank M. Howell, and Shu-Chuan Lai. 2002. “Spatial Variations in Church Burnings: The Social Ecology of Victimized Communities in the South.” Rural Sociology 67:578-602.
Bartkowski, John P., and Helen A. Regis. 2002. “The Promise and Peril of Charitable Choice: Religion, Poverty Relief, and Welfare Reform in the South.” Southern Rural Sociology 18:222–58.
———. 2003. Charitable Choices: Religion, Race, and Poverty in the Post-Welfare Era. New York University Press.
Bartkowski, John P., and W. Bradford Wilcox. 2000. “Conservative Protestant Child Discipline: The Case of Parental Yelling.” Social Forces 79:265–90.
Bartkowski, John P., and Xiaohe Xu. 2000. “Distant Patriarchs or Expressive Dads? The Discourse and Practice of Fathering in Conservative Protestant Families.” Sociological Quarterly 41:465–85.
Baumer, Eric. 1994. “Poverty, Crack, and Crime: A Cross-City Analysis.” Journal of Research in Crime and Delinquency 31:311–27.
Baumer, Eric, Janet L. Lauritsen, Richard Rosenfeld, and Richard Wright. 1998. “The Influence of Crack Cocaine on Robbery, Burglary, and Homicide Rates: A Cross-City, Longitudinal Analysis.” Journal of Research in Crime and Delinquency 35:316–40.
Becker, Penny Edgell. 1999. Congregations in Conflict: Cultural Models of Local Religious Life. Cambridge University Press.
Becker, Penny Edgell, and Pawan H. Dhingra. 2001. “Religious Involvement and Volunteering: Implications for Civil Society.” Sociology of Religion 62:315–35.
Love Thy Neighbor? / 1029
Bell, Wendell. 1954. “A Probability Model for the Measurement of Ecological Segregation.” Social Forces 32:357–64.
Bellair, Paul E. 1997. “Social Interaction and Community Crime: Examining the Importance of Neighbor Networks.” Criminology 35:677–703.
———. 2000. “Informal Surveillance and Street Crime: A Complex Relationship.” Criminology 38:137–69.
Bennett, William J., John J. DiIulio Jr., and John P. Walters. 1996. Body Count — Moral Poverty . . . and How to Win America’s War against Crime and Drugs. Simon & Schuster.
Blau, Judith R., and Peter M. Blau. 1982. “The Cost of Inequality: Metropolitan Structure and Violent Crime.” American Sociological Review 47:114–29.
Blumstein, Alfred. 1995. “Youth Violence, Guns, and the Illicit Drug Industry.” Journal of Criminal Law and Criminology 86:10–36.
Blumstein, Alfred, and Joel Wallman (eds.). 2000. The Crime Drop in America. Cambridge University Press.
Burkett, Steven R., and Melvin White. 1974. “Hellfire and Delinquency: Another Look.” Journal for the Scientific Study of Religion 13:455–62.
Bursik, Robert J. 1988. “Social Disorganization and Theories of Crime and Delinquency: Problems and Prospects.” Criminology 26:519–51.
Bursik, Robert J., and Harold G. Grasmick. 1993. Neighborhoods and Crime: The Dimensions of Effective Community Control. Lexington Books.
Butler, Margaret A., and Calvin L. Beale. 1994. Rural–Urban Continuum Codes for Metro and Nonmetro Counties, 1993. Staff Report No. 9425. U.S. Department of Agriculture.
Census of Population and Housing. 1990. Summary Tape File 3 on CD–ROM [machine- readable data files]. Washington, DC: Bureau of the Census [producer and distributor].
Chadwick, Bruce A., and Brent L. Top. 1993. “Religiosity and Delinquency among LDS Adolescents.” Journal for the Scientific Study of Religion 32:51–67.
Chamlin, Mitchell B., and John K. Cochran. 1995. “Assessing Messner and Rosenfeld’s Institutional Anomie Theory: A Partial Test.” Criminology 33:411-29.
———. 1997. “Social Altruism and Crime.” Criminology 35:203–27.
Christiano, Kevin J. 2000. “Religion and the Family in Modern American Culture.” Pp. 43–78 in Family, Religion, and Social Change in Diverse Societies, edited by Sharon K. Houseknecht and Jerry G. Pankhurst. Oxford University Press.
Cnaan, Ram A. 2002. The Invisible Caring Hand: American Congregations and the Provision of Welfare. New York University Press.
Cochran, John K. 1988. “The Effect of Religiosity on Secular and Ascetic Deviance.” Sociological Focus 21:293–306.
Cochran, John K., and Ronald L. Akers. 1989. “Beyond Hellfire: An Exploration of the Variable Effects of Religiosity on Adolescent Marijuana and Alcohol Use.” Journal of Research in Crime and Delinquency 26:198–225.
Coleman, James S. 1988. “Social Capital in the Creation of Human Capital.” American Journal of Sociology 94:95–120.
1030 / Social Forces 82:3, March 2004
Cork, Daniel. 1999. “Examining Space–Time Interaction in City-Level Homicide Data: Crack Markets and the Diffusion of Guns among Youth.” Journal of Quantitative Criminology 15:379–406.
DeFronzo, James. 1997. “Welfare and Homicide.” Journal of Research in Crime and Delinquency 34:395–406.
Donahue, Michael J., and Peter L. Benson. 1995. “Religion and the Well-Being of Adolescents.” Journal of Social Issues 51:145–60.
Donnermeyer, Joseph F. 1992. “The Use of Alcohol, Marijuana, and Hard Drugs by Rural Adolescents: A Review of Recent Research.” Pp. 31–76 in Drug Use in Rural American Communities, edited by Ruth W. Edwards. Hawthorne Press.
Eberly, Don E. (ed.). 1999. The Faith Factor in Fatherhood: Renewing the Sacred Vocation of Fathering. Lexington Books.
Elder, Glen H. Jr., and Rand D. Conger. 2000. Children of the Land: Adversity and Success in Rural America. University of Chicago Press.
Elifson, Kirk W., David M. Petersen, and C. Kirk Hadaway. 1983. “Religiosity and Delinquency: A Contextual Analysis.” Criminology 21:505–27.
Ellison, Christopher G. 1991. “An Eye for an Eye? A Note on the Southern Subculture of Violence Thesis.” Social Forces 69:1223–39.
Ellison, Christopher G., John P. Bartkowski, and Michelle L. Segal. 1996a. “Conservative Protestantism and the Parental Use of Corporal Punishment.” Social Forces 74:1003–28.
———. 1996b. “Do Conservative Protestant Parents Spank More Often? Further Evidence from the National Survey of Families and Households.” Social Science Quarterly 77:663–73.
Ellison, Christopher G., and Darren E. Sherkat. 1993. “Conservative Protestantism and Support for Corporal Punishment.” American Sociological Review 58:131–44.
———. 1995. “The ‘Semi-Involuntary Institution’ Revisited: Regional Variations in Church Participation among Black Americans.” Social Forces 73:1415–37.
Fox, James Alan. 1996. Trends in Juvenile Violence: A Report to the United States Attorney General on Current and Future Rates of Juvenile Offending. Prepared for the Bureau of Justice Statistics. U.S. Department of Justice.
Fox, James Alan. 2000. Uniform Crime Reports [United States]: Supplementary Homicide Reports, 1976-1998 [Computer File]. ICPSR version. Boston, Mass.: Northeastern University, College of Criminal Justice [Producer], Ann Arbor, Mich. ICPSR [distributor].
Fox, James Alan, and Glenn Pierce. 1994. “American Killers Are Getting Younger.” USA Today Magazine, January 24–26.
Fox, James Alan, and Marianne W. Zawitz. 2001. Homicide Trends in the United States. Bureau of Justice Statistics, U.S. Department of Justice <www.ojp.usdoj.gov/bjs>.
Gay, Dav id A., Christopher G. Ellison, and Daniel A. Powers. 1996. “In Search of Denominational Subcultures: Religious Affiliation and ‘Pro-Family’ Issues Revisited.” Review of Religious Research 38:3–17.
Gibbs, Nancy R., and Julie Grace. 1994. “Murder in Miniature.” Time, Sept. 19, 54–60.
Goldstein, Paul J. 1985. “The Drugs/Violence Nexus: A Tripartite Conceptual Framework.” Journal of Drug Issues 14:493–506.
Love Thy Neighbor? / 1031
Goreham, Gary A. 2002. “Community and Religion in a Post-Modern Global World: A Review Essay on Anabaptist Religious Groups.” Rural Sociology 67:299–305.
Greeley, Andrew. 1997. “The Other Civic America: Religion and Social Capital.” American Prospect 32:68–73.
Hannon, Lance, and James DeFronzo. 1998. “The Truly Disadvantaged, Public Assistance, and Crime.” Social Problems 45:383–92.
Hertel, Bradley R., and Michael Hughes. 1987. “Religious Affiliation, Attendance, and Support for ‘Pro-Family’ Issues in the United States.” Social Forces 65:858–82.
Herz, Denise C. 2000. “Drugs in the Heartland: Methamphetamine Use in Rural Nebraska.” National Institute of Justice: Research in Brief.
Higgins, Paul C., and Gary L. Albrecht. 1977. “Hellfire and Delinquency Revisited.” Social Forces 55:952–58.
Hirschi, Travis, and Rodney Stark. 1969. “Hellfire and Delinquency.” Social Problems 17:202– 13.
Holmes, Ronald M., and Stephen T. Holmes. 2001. Mass Murder in the United States. Prentice- Hall.
Hunt, Matthew O., and Larry L. Hunt. 2000. “Regional Religions? Extending the ‘Semi- Involuntary’ Thesis of African-American Religious Participation.” Sociological Forum 15:569–94.
Irwin, Michael, Troy Blanchard, Charles Tolbert, Thomas Lyson, and Alfred Nucci. 2002. “Dwelling Together: Community Civic Structure and Residential Integration.” Paper presented at the annual meetings of the Southwestern Social Science Association, New Orleans, LA.
Irwin, Michael, Charles Tolbert, and Thomas Lyson. 1999. “There’s No Place like Home: Nonmigration and Civic Engagement.” Environment and Planning A 31:2223–38.
Jacobs, David, and Katherine Wood. 1999. “Interracial Conflict and Interracial Homicide: Do Political and Economic Rivalries Explain White Killing of Blacks and Black Killing of Whites?” American Journal of Sociology 105:157–90.
Johnson, Byron R., Spencer De Li, David B. Larson, and Michael McCullough. 2000. “A Systematic Review of the Religiosity and Delinquency Literature: A Research Note.” Journal of Contemporary Criminal Justice 16:32–52.
Johnson, Byron R., Sung Joon Jang, David B. Larson, and Spencer De Li. 2001. “Does Adolescent Religious Commitment Matter? A Reexamination of the Effects of Religiosity on Delinquency.” Journal of Research in Crime and Delinquency 38:22–44.
Johnson, Kenneth M., and Calvin L. Beale. 2001. “The Rural Rebound: Recent Nonmetropolitan Demographic Trends in the United States.” [http://www.luc.edu/depts/sociology/johnson/ p99webn.html].
Junger, Marianne, and Wim Polder. 1993. “Religiosity, Religious Climate, and Delinquency among Ethnic Groups in the Netherlands.” British Journal of Criminology 33:416–35.
Kasarda, John D., and Morris Janowitz. 1974. “Community Attachment in Mass Society.” American Sociological Review 39:328–39.
1032 / Social Forces 82:3, March 2004
King, Valarie, Glen H. Elder Jr., and Les B. Whitbeck. 1997. “Religious Involvement among Rural Youth: An Ecological and Life-Course Perspective.” Journal of Research on Adolescence 7:431–56.
Kposowa, Augustine J., and Kevin D. Breault. 1993. “Reassessing the Structural Covariates of U.S. Homicide Rates: A County Level Study.” Sociological Focus 26:27–46.
Kposowa, Augustine, Kevin Breault, and Beatrice Harrison. 1995. “Reassessing the Structural Covariates of Violent and Property Crimes in the USA: A County Level Analysis.” British Journal of Sociology 46:79–105.
Land, Kenneth C., Patricia L. McCall, and Lawrence E. Cohen. 1990. “Structural Covariates of Homicide Rates: Are There Any Invariances across Time and Social Space?” American Journal of Sociology 95:922–63.
Land, Kenneth C., Patricia L. McCall, and Daniel S. Nagin. 1996. “A Comparison of Poisson, Negative Binomial, and Semiparametric Mixed Poisson Regression Models with Empirical Applications to Criminal Careers Data.” Sociological Methods and Research 24:387–442.
Lee, Matthew R., and John P. Bartkowski. 2004. “Civic Participation, Regional Subcultures, and Violence: The Differential Effects of Secular and Religious Participation on Adult and Juvenile Homicide.” Homicide Studies 7:1–35.
Lee, Matthew R., Michael O. Maume, and Graham C. Ousey. 2003. “Social Isolation and Lethal Violence across the Metro/NonMetro Divide: The Effects of Socioeconomic Disadvantage and Poverty Concentration on Homicide.” Rural Sociology 68:107–31.
Lee, Matthew T., Ramiro Martinez Jr., and Richard Rosenfeld. 2001. “Does Immigration Increase Homicide? Negative Evidence from Three Border Cities.” Sociological Quarterly 42:559–80.
Lee, Matthew R., and Graham C. Ousey. 2001. “Size Matters: Examining the Link between Small Manufacturing, Socioeconomic Deprivation, and Crime Rates in Non-metropolitan Communities.” Sociological Quarterly 42:581–602.
Lin, Nan. 2001. Social Capital: A Theory of Social Structure and Action. Cambridge University Press.
Liska, Allen E., and Paul E. Bellair. 1995. “Violent Crime Rates and Racial Composition: Convergence over Time.” American Journal of Sociology 101:578–610.
Long, J. Scott. 1997. Regression Models for Categorical and Limited Dependent Variables. Sage Publications.
Maxfield, Michael G. 1989. “Circumstances in Supplementary Homicide Reports: Variety and Validity.” Criminology 27:671–95.
Merton, Robert K. 1938. “Social Structure and Anomie.” American Sociological Review 3:672– 82.
Messner, Steven F., and Reid M. Golden. 1992. “Racial Inequality and Racially Disaggregated Homicide Rates: An Assessment of Alternative Theoretical Explanations.” Criminology 30:421–45.
Messner, Steven F., and Richard Rosenfeld. 1997. Crime and the American Dream, 2d ed. Wadsworth.
Morenoff, Jeffrey D., and Robert J. Sampson. 1997. “Violent Crime and the Spatial Dynamics of Neighborhood Transition: Chicago, 1970–1990.” Social Forces 76:31–64.
Love Thy Neighbor? / 1033
Osgood, D. Wayne. 2000. “Poisson-Based Regression Analysis of Aggregate Crime Rates.” Journal of Quantitative Criminology 16:21–43.
Osgood, D. Wayne, and Jeff M. Chambers. 2000. “Social Disorganization outside the Metropolis: An Analysis of Rural Youth Violence.” Criminology 38:81–115.
Ousey, Graham C. 1999. “Homicide, Structural Factors, and the Racial Invariance Assumption.” Criminology 37:405–26.
Ousey, Graham C., and Michelle Campbell Augustine. 2001. “Young Guns: Examining Alternative Explanations of Juvenile Firearm Homicide Rates.” Criminology 39:933–63.
Ousey, Graham C., and Matthew R. Lee. 2002. “Examining the Conditional Nature of the Illicit Drug Market–Homicide Relationship: A Partial Test of the Theory of Contingent Causation.” Criminology 40:73–102.
Pankhurst, Jerry G., and Sharon K. Houseknecht. 2000. “Introduction: The Religion–Family Linkage and Social Change — A Neglected Area of Study.” Pp. 1–42 in Family, Religion, and Social Change in Diverse Societies, edited by Sharon K. Houseknecht and Jerry G. Pankhurst. Oxford University Press.
Parisi, Domenico, Diane K. McLaughlin, Michael Taquino, Steven Michael Grice, and Neil R. White. 2002. “TANF/Welfare Client Decline and Community Context in the Rural South, 1997–2000.” Southern Rural Sociology 18:154–87.
Parker, Karen F. 2001. “A Move toward Specificity: Examining the Similarities in Race and RelationshipSpecific Homicide Rates.” Journal of Quantitative Criminology 17:89–110.
Parker, Karen F., and Patricia L. McCall. 1999. “Structural Conditions and Racial Homicide Patterns: A Look at the Multiple Disadvantages in Urban Areas.” Criminology 37:447–77.
Parker, Karen F., Patricia L. McCall, and Kenneth C. Land. 1999. “Determining Social-Structural Predictors of Homicide: Units of Analysis and Related Methodological Concerns.” Pp. 107– 24 in Homicide: A Sourcebook of Social Research, edited by M. Dwayne Smith and Margaret A. Zahn. Sage Publications.
Petee, Thomas, and Gregory Kowalski. 1993. “Modeling Rural Violent Crime Rates: A Test of Social Disorganization Theory.” Sociological Focus 26:87–89.
Peterson, Ruth D., and Lauren Krivo. 1993. “Racial Segregation and Black Urban Homicide.” Social Forces 71:1001–26.
Piquero, Alex, and Nicole Leeper Piquero. 1998. “On Testing Institutional Anomie Theory with Varying Specifications.” Studies on Crime and Crime Prevention 7:61–84.
Portes, Alejandro. 1998. “Social Capital: Its Origins and Applications in Modern Sociology.” Annual Review of Sociology 24:1–24.
Putnam, Robert D. 1993. “The Prosperous Community: Social Capital and Public Life.” The American Prospect 4. http://www.prospect.org/print/V4/13/putnamr.html.
———. 2000. Bowling Alone: The Collapse and Revival of American Community. Simon & Schuster.
Regnerus, Mark D. 2000. “Moral Communities and Adolescent Delinquency: Subcultural Aspects of Social Disorganization.” Paper presented at the annual meetings of the Association for the Sociology of Religion, Washington, DC.
Reidel, Marc. 1999. “Sources of Homicide Data: A Review and Comparison.” Pp. 75–95 in Homicide: A Sourcebook of Social Research, edited by M. Dwayne Smith and Margaret A. Zahn. Sage Publications.
1034 / Social Forces 82:3, March 2004
Robinson, W.S. 1950. “Ecological Correlations and the Behavior of Individuals.” American Sociological Review 16:351–57.
Rosenfeld, Richard. 2000. “Patterns in Adult Homicide: 1980–1995.” Pp. 130–63 in The Crime Drop in America, edited by Alfred Blumstein and Joel Wallman. Cambridge University Press.
Rosenfeld, Richard, and Scott H. Decker. 1999. “Are Arrest Statistics a Valid Measure of Illicit Drug Use? The Relationship between Criminal Justice and Public Health Indicators of Cocaine, Heroin, and Marijuana Use.” Justice Quarterly 16:685–99.
Rosenfeld, Richard, Steven F. Messner, and Eric P. Baumer. 2001. “Social Capital and Homicide.” Social Forces 80:283–309.
Sampson, Robert J. 1987. “Urban Black Violence: The Effect of Male Joblessness and Family Disruption.” American Journal of Sociology 93:348–82.
———. 1988. “Local Friendship Ties and Community Attachment in Mass Society: A Multilevel Systemic Model.” American Sociological Review 53:766–79.
Sampson, Robert J., and W. Byron Groves. 1989. “Community Structure and Crime: Testing Social Disorganization Theory.” American Journal of Sociology 94:774–802.
Sampson, Robert J., Stephen W. Raudenbush, and Felton Earls. 1997. “Neighborhoods and Violent Crime: A Multilevel Study of Collective Efficacy.” Science 277:918–24.
Savolainen, Jukka. 2000. “Inequality, Welfare State, and Homicide: Further Support for the Institutional Anomie Theory.” Criminology 38:1021–42.
Skocpol, Theda. 2000. “Religion, Civil Society, and Social Provision in the U.S.” Pp. 21–50 in Who Will Provide? The Changing Role of Religion in American Social Welfare, edited by Mary Jo Bane, Brent Coffin, and Ronald Thiemann. Westview Press.
Shaw, Clifford R., and Henry D. McKay. 1942. Juvenile Delinquency and Urban Areas. University of Chicago Press.
Sherkat, Darren E., and Shannon A. Cunningham. 1998. “Extending the Semi-Involuntary Institution: Regional Differences and Social Constraints on Private Religious Consumption among African Americans.” Journal for the Scientific Study of Religion 37:383–96.
Sherkat, Darren E., and Christopher G. Ellison. 1999. “Recent Developments and Current Controversies in the Sociology of Religion.” Annual Review of Sociology 25:363–94.
Shihadeh, Edward S., and Darrell Steffensmeier. 1994. “Economic Inequality, Family Disruption, and Urban Black Violence: Cities As Units of Stratification and Social Control.” Social Forces 73:729–51.
Sloane, Douglas M., and Raymond H. Potvin. 1986. “Religion and Delinquency: Cutting through the Maze.” Social Forces 65:87–105.
Smidt, Corwin. 1999. “Religion and Civic Engagement: A Comparative Analysis.” Annals of the American Academy of Political and Social Science 565:176–92.
Stark, Rodney. 1984. “Religion and Conformity: Reaffirming a Sociology of Religion.” Sociological Analysis 45:273–82.
———. 1996. “Religion As Context: Hellfire and Delinquency One More Time.” Sociology of Religion 57:163–73.
Stark, Rodney, Daniel P. Doyle, and Lori Kent. 1980. “Rediscovering Moral Communities: Church Membership and Crime.” Pp. 43–50 in Understanding Crime, edited by Travis Hirschi and Michael Gottfredson. Sage Publications.
Love Thy Neighbor? / 1035
Stark, Rodney, Lori Kent, and Daniel P. Doyle. 1982. “Religion and Delinquency: The Ecology of a ‘Lost’ Relationship.” Journal of Research in Crime and Delinquency 19:4–24.
Steffensmeier, Darrell, and Dana L. Haynie. 2000a. “Gender, Structural Disadvantage, and Urban Crime: Do Macrosocial Variables Also Explain Female Offending Rates?” Criminology 38:403–38.
———. 2000b. “The Structural Sources of Urban Female Violence in the United States: A Macrosocial Gender-Disaggregated Analysis of Adult and Juvenile Homicide Offending Rates.” Homicide Studies 4:107–34.
Tolbert, Charles M., Thomas A. Lyson, and Michael D. Irwin. 1998. “Local Capitalism, Civic Engagement, and Socioeconomic Well-Being.” Social Forces 77:401–29.
Tonry, Michael, and Mark H. Moore (eds.). 1998. Youth Violence, Crime, and Justice: A Review of Research, vol. 24. University of Chicago Press.
Weisheit, Ralph A., and Joseph F. Donnermeyer. 2000. “Change and Continuity in Crime in Rural America.” Pp. 309–57 in Criminal Justice 2000, vol. 1, The Nature of Crime — Continuity and Change, edited by Gary LaFree. National Institute of Justice.
Welch, Michael R., Charles R. Tittle, and Thomas Petee. 1991. “Religion and Deviance among Adult Catholics: A Test of the ‘Moral Communities’ Hypothesis.” Journal for the Scientific Study of Religion 30:159–72.
Wilcox, W. Bradford. 2002. “Religion, Convention, and Paternal Involvement.” Journal of Marriage and Family 64:780–92.
Wilcox, W. Bradford, and John P. Bartkowski. 1999. “The Evangelical Family Paradox: Conservative Rhetoric, Progressive Practice.” Responsive Community 9:34–39.
Williams, Kirk R., and Robert L. Flewelling. 1988. “The Social Production of Criminal Homicide: A Comparative Study of Disaggregated Rates in American Cities.” American Sociological Review 53:421–31.
Wilson, William Julius. 1987. The Truly Disadvantaged: The Inner City, the Underclass, and Public Policy. University of Chicago Press.
Wolf, C.P. 1970. “The Durkheim Thesis: Occupational Groups and Moral Integration.” Journal for the Scientific Study of Religion 9:17–32.
Woodrum, Eric, and Thomas Hoban. 1992. “Support for Prayer in School and Creationism.” Sociological Analysis 53:309–21.
Wuthnow, Robert. 1999. “Mobilizing Civic Engagement: The Changing Impact of Religious Involvement.” Pp. 331–63 in Civic Engagement in American Democracy, edited by Theda Skocpol and Morris P. Fiorina. Brookings Institution Press.
12 sources/Braga et al.pdf
The Concentration and Stability of Gun Violence at Micro Places in Boston, 1980–2008 Author(s): Anthony A. Braga, Andrew V. Papachristos and David M. Hureau Source: Journal of Quantitative Criminology, Vol. 26, No. 1, Special Issue on Empirical Evidence on the Relevance of Place in Criminology (March 2010), pp. 33-53 Published by: Springer Stable URL: http://www.jstor.org/stable/23367576 Accessed: 19-06-2016 04:51 UTC
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact [email protected].
Springer is collaborating with JSTOR to digitize, preserve and extend access to Journal of Quantitative Criminology
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:51:16 UTC All use subject to http://about.jstor.org/terms
J Quant Criminol (2010) 26:33-53 DOI 10.1007/s 10940-009-9082-x
ORIGINAL PAPER
The Concentration and Stability of Gun Violence at Micro Places in Boston, 1980-2008
Anthony A. Braga • Andrew V. Papachristos • David M. Hureau
Published online: 31 December 2009
© Springer Science+Business Media, LLC 2009
Abstract Boston, like many other major U.S. cities, experienced an epidemic of gun violence during the late 1980s and early 1990s that was followed by a sudden large downturn in gun violence in the mid 1990s. The gun violence drop continued until the early part of the new millennium. Recent advances in criminological research suggest that there is significant clustering of crime in micro places, or "hot spots," that generate a disproportionate amount of criminal events in a city. In this paper, we use growth curve regression models to uncover distinctive developmental trends in gun assault incidents at street segments and intersections in Boston over a 29-year period. We find that Boston gun violence is intensely concentrated at a small number of street segments and intersections rather than spread evenly across the urban landscape between 1980 and 2008. Gun violence trends at these high-activity micro places follow two general trajectories: stable concen trations of gun assaults incidents over time and volatile concentrations of gun assault incidents over time. Micro places with volatile trajectories represent less than 3% of street segments and intersections, generate more than half of all gun violence incidents, and seem to be the primary drivers of overall gun violence trends in Boston. Our findings suggest that the urban gun violence epidemic, and sudden downturn in urban gun violence in the late 1990s, may be best understood by examining highly volatile micro-level trends at a relatively small number of places in urban environments.
Keywords Guns ■ Gun violence • Hot spots • Epidemic
A. A. Braga (El) • A. V. Papachristos • D. M. Hureau Harvard University, Cambridge, MA, USA e-mail: [email protected]; [email protected]
A. A. Braga University of California, Berkeley, CA, USA
A. V. Papachristos University of Massachusetts, Amherst, MA, USA
Ô Springer
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:51:16 UTC All use subject to http://about.jstor.org/terms
34 J Quant Criminol (2010) 26:33-53
Introduction
Beginning in the late 1980s and continuing through the early 1990s, the United States experienced a dramatic increase in firearms violence reaching its zenith in 1993, with 17,075 homicides committed with firearms (Blumstein 1995; Cook and Laub 1998, 2002).1 This dramatic increase was followed by a puzzling decrease. By 2000, gun homicide had decreased by 40% to 10,203 incidents and remained relatively low with 10,661 gun homicides in 2004. Criminologists and public policy analysts have examined a wide range of factors that may have been associated with this drop, including innovative policing strategies, a strong economy, higher imprisonment rates, stronger gun control, and stabi lizing street-level drug markets (Blumstein and Wallman 2000). In recent years, however, some observers have expressed concern of a resurgence of urban gun violence that was developing nationwide (e.g., Police Executive Research Forum 2006). For example, in 2005 the U.S. Bureau of Justice Statistics reported that the number of gun homicides had increased by 6% to 11,346. Fortunately, this increase seemed to be short lived as the Federal Bureau of Investigation recently estimated that the number of gun homicides once again declined to 10,086 in 2007.2
Some research equated the epidemic of firearms violence that spanned the late 1980s and early 1990s with a "flood in a canyon" as it was intensely concentrated in disad vantaged inner-city areas and among young minority males, who were often gang-involved and well known to the criminal justice system (Cook and Laub 2002; Braga 2003). What is more, criminological evidence on the concentration of crime in a small number of highly active micro places suggests that a few "hot spot" locations in disadvantaged urban neighborhoods may be primarily responsible for overall city wide gun violence trends (see, e.g., Sherman et al. 1989; Weisburd et al. 2004). Unfortunately, few scientific inquiries have examined the spatial distribution of gun violence during these dramatic shifts, and when they have, they have typically not looked at units of analysis smaller than the Census tract or block group. If citywide gun violence epidemics can be best understood in terms of large changes at a few micro places, these findings would suggest that an array of violence prevention programs involving the deployment of criminal justice, social service, and community-based resources should be similarly concentrated rather than diffused across larger urban areas.
In this paper, we use growth curve regression models to uncover distinctive develop mental trends in gun assault incidents at street segments and intersections in Boston between 1980 and 2008. The main goal of this paper is to examine the ways in which crime patterns in micro places influence the vanguard of crime trends in the City of Boston during the epidemic and post-epidemic periods. In support of micro-places research, our findings suggest that an extremely small percentage of micro places that exhibit relatively stable crime trajectories are responsible for the majority of gun violence trends in Boston during this time period. In fact, only 5% of street segments and intersections in Boston are responsible for 74% of serious gun assault incidents even when controlling for prior levels of gun violence and existing linear and nonlinear trends. These highly active places account for the bulk of the increase in gun violence during epidemic years and the decrease in gun violence during crime drop years.
1 Unless otherwise noted, the homicide data in this paragraph were acquired from the U.S. Bureau of Justice Statistics (accessed May 20, 2009) http://www.ojp.usdoj.gov/bjs/homicide/tables/weaponstab.htm.
2 http://www.fbi.gov/ucr/cius2007/offenses/expanded_information/data/shrtable_07.html (accessed May 20, 2009).
Springer
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:51:16 UTC All use subject to http://about.jstor.org/terms
J Quant Criminol (2010) 26:33-53 35
The first two sections of this paper briefly summarize the existing literature on crime and place and the nature of the gun violence epidemic between the late 1980s and early 1990s. These sections pay particular attention to the relevant dimensions of gun vio lence in Boston. The next three sections describe the data collection methodology, detail the statistical models used to analyze the data, and present the results of the quantitative analyses. The conclusions drawn from the research findings are discussed in the final section.
The Concentration and Stability of Crime in Micro Places
The observation that the distribution of crime varies within and between neighborhoods has existed for some time (see, e.g., Shaw and McKay 1942). However, due to limited analytical capacities, little empirical research has examined this variance beyond the community or neighborhood level of analysis with the U.S. Census tract or block group serving as the most common units of analysis. With the advent of powerful computer systems and software packages in the late 1980s, analysts began to further hone their focus on even smaller geographic units of analysis. Two well-known cross-sectional studies found that some 5% of city addresses generated over 50% of citizen emergency calls for service to the police in Boston (Pierce et al. 1988) and Minneapolis (Sherman et al. 1989). Even within high-crime neighborhoods, these studies found that crime clusters at a few discrete "hot spot" micro places, leaving blocks of areas within neighborhoods relatively crime-free. Put another way, not every block or corner in a high crime neighborhood experiences high levels of crime. Rather, certain blocks or address experience high levels of crime, while others are relatively crime free. Further, research by Taylor and Gott fredson (1986) revealed conclusive evidence that links this spatial variation to the physical and social characteristics of particular blocks and multiple dwellings within a neighborhood.
More recently, a research team from the University of Maryland analyzed crime incidents at the level of street segments in Seattle over a 14 year period and found that, year to year, about 50% of the crime was concentrated in approximately 4.5% of street segments (Weisburd et al. 2004). Of course, the concentration of crime year to year does not preclude the possibility that each year different crime hot spots would develop, or that hot spots in 1 year would naturally become cool the next. For this reason, the Maryland research team also examined the developmental trends of crime at street segments in Seattle over the 14 year period (Weisburd et al. 2004). Using semi parametric, group-based trajectory procedures (i.e., TRAJ models, see Nagin 1999), the approximately 30,000 street segments in Seattle were grouped into trajectories with similar developmental trends over time. These analyses revealed that there was a high degree of stability of crime at micro places over time. In other words, crime remained concentrated in a small number of micro places in Seattle rather than spread across the city over time. Weisburd et al. (2004) also found that a relatively small proportion of places belonged to groups with steeply rising and or declining trajectories and that these places were primarily responsible for overall crime trends in Seattle between 1989 and 2002.
Several other studies have come to similar conclusions about the stability of crime at specific micro places over time. Spelman (1995) analyzed calls-for-service at high schools, housing projects, subway stations, and parks in Boston, and found that the risks at these public places remained fairly constant over time. Taylor (1999) also reports evidence of a
<£) Springer
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:51:16 UTC All use subject to http://about.jstor.org/terms
36 J Quant Criminol (2010) 26:33-53
high degree of stability of crime at place over time, examining crime and fear of crime at ninety street blocks in Baltimore, Maryland using a panel design with data collected in 1981 and 1994 (see Robinson et al. 2003; Taylor 2001). Data included not only official crime statistics, but also measures of citizen perceptions of crime and observations of physical conditions at the sites. Although Taylor and his colleagues observed significant deterioration in physical conditions at the blocks studied, they found that neither fear of crime nor crime showed significant or consistent differences across the two time periods.
The Spatial Nature of the Gun Violence Epidemic
Although the direct causes of the gun violence epidemic in the late 1980s and early 1990s remain somewhat elusive, a concrete story emerged that seems to fit with experience and research evidence. Moore and Tonry's (1998) synthesis of key events provides a useful framework and is quickly summarized here. Building off of the work of Wilson (1987, 1996), Moore and Tonry recounted how the deindustrialization of the economy during late seventies and early eighties fundamentally changed the structural factors conditions in inner-city minority communities. The social and economic structure of many urban neighborhoods collapsed under a variety of social and economic pressures as employment opportunities, business, and the middle class fled the inner-city. A subsequent cascade of negative conditions decimated inner city neighborhoods, including the disruption of families, increased mass incarceration, and heightened social and economic isolation. Gangs continued to flourish as youth responded to geographic and structural isolation by turning to these groups in search of affiliation, security, and, in some cities, new economic opportunities (see, e.g., Hagedorn 1988).
Moore and Tonry's (1998) review then suggests that an epidemic of crack cocaine hit many of these troubled communities during the mid to late eighties (see also Blumstein 1995). Some existing youth gangs and other non-gang involved youth participated in street-level drug markets and armed themselves with guns to protect themselves and resolve business disputes. The arming of youth participating in street drug sales produced both dangerous conditions on the street and a cultural style that encouraged other youth to acquire guns in response. A large supply of available guns made it possible for other youth to acquire guns out of self-protection, style, and status concerns. The widespread arming of youth in these disadvantaged neighborhoods made everyday conflicts much more lethal.
This account suggests strong spatial dimensions to the spread of gun violence in U.S. cities over the course of the late 1980s and early 1990s. Since many homicides, whether gang-related or not, are retaliatory in nature (Block 1977; Wolfgang 1958), homicides may themselves instigate a sequence of events that leads to further violence in a spatially channeled way. As such, a homicide in one neighborhood may spark a retaliatory killing in a nearby neighborhood. Most homicides occur among persons who are known to each other (Reiss and Roth 1993) and these networks of associations can follow geographic vectors across and within neighborhoods and specific places (Papachristos 2009). As such, several studies attempted to understand the nature of the gun violence epidemic by examining the geographic distribution of gun violence over time. These analyses generally reported strong spatial associations between homicide concentrations and the spatial dis tribution of poor, black neighborhoods that experience gang, drug, and gun problems (Cohen and Tita 1999; Rosenfeld et al. 1999). These analyses also sought to determine whether gun violence diffused across urban landscapes over time.
â Springer
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:51:16 UTC All use subject to http://about.jstor.org/terms
J Quant Criminol (2010) 26:33-53 37
In Pittsburgh, Cohen and Tita (1999) reported that spatial diffusion of increasing homicide rates across neighboring U.S. Census tracts was evident only during the year of peak growth in total homicides, when high local rates of youth-gang homicides were followed by significant increases in neighboring youth-nongang rates. Otherwise, Cohen and Tita (1999) reported that increases in both youth-gang and youth-nongang homicides generally occurred simultaneously in non-neighboring Census tracts. Using Census block groups as the unit of analysis, Rosenfeld et al. (1999) provided some evidence that gang motivated homicides in St. Louis spread in a contagious manner and speculated that gang membership may be the mechanism by which such events spread. A separate analysis of the distribution of homicides in 78 counties in and around St. Louis reported strong evidence of the highly localized nature of spatial dependencies in homicides over time at the county-level and some modest evidence of spatial diffusion of homicide in particular urban counties, and presented findings that affluent and rural areas serve as barriers against the spread homicides across counties (Messner et al. 1999).
More recently, Griffiths and Chavez (2004) merged Exploratory Spatial Data Analysis (ESDA) and TRAJ models to identify total, street gun, and other weapon homicide tra jectories across 831 Census tracts in Chicago between 1980 and 1995. Griffiths and Chavez (2004) reported a weapon substitution effect in violent neighborhoods (i.e., Census tracts) that are proximate to one another, a defensive diffusion effect of exclusively street gun specific homicide increases in neighborhoods bordering the most violent areas, and a spatial decay effect of temporal homicide trends in which the most violent areas are buffered from the least violent by tracts experiencing mid-range levels of homicide over time. The Census tracts associated with the largest increases in street gun homicide rates were characterized as areas associated with high-levels of gun violence, drug market activity, and street gang activity. Other studies have revealed that gang wars over drug markets in Chicago were prevalent between 1987 and 1994 and concentrated in a small number of hot spot locations (Block and Block 1993; Block et al. 1996).
The Trajectory and Nature of Gun Violence in Boston, 1980-2008
Like many American cities during the late 1980s and early 1990s, Boston suffered an epidemic of gun violence that had its roots in the rapid spread of street-level crack-cocaine markets (Kennedy et al. 1996; Braga 2003). Measured as a homicide problem, Boston experienced a dramatic increase in the number of fatal gun shot wound victims. During the "pre-epidemic" years of 1980 through 1988, Boston averaged approximately 40 gun homicides per year. The number of gun homicides increased to 57 victims in 1989 and peaked at 86 victims in 1990. While gun homicide subsequently decreased from the peak year, the yearly number of victims remained high between 1991 and 1995 as Boston averaged nearly 62 gun homicides per year.3 In 1996, the number of gun homicides dropped steeply to 38 victims and, in 1999, Boston experienced only 19 gun homicides.4
3 After street crack-cocaine markets stabilized, drug-related violence decreased in Boston. Unfortunately, serious gun violence had become "decoupled" from the crack trade. Guns were used by Boston youth to settle disputes that were once dealt with by fists, sticks, and knives (Kennedy et al. 1996; Braga 2003).
4 An interagency problem-oriented policing intervention, which tightly focused criminal justice attention on a small number of chronically offending gang-involved youth, was associated with the significant reduction in youth homicide and non-fatal gun violence when it was operational between 1996 and 2000 (Braga et al., 2001). The implementation and impact of the Operation Ceasefire intervention has been extensively doc umented elsewhere (see, e.g., Kennedy et al. 1996, 2001; Braga et al. 2008) and is not the subject of this paper.
<£) Springer
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:51:16 UTC All use subject to http://about.jstor.org/terms
38 J Quant Criminol (2010) 26:33-53
Unfortunately, beginning in 2001, the number of gun homicides steadily grew to a second peak of 55 victims in 2006. In 2007 and 2008, the number of gun homicides decreased modestly to 52 and 49 victims, respectively.
The two periodical increases in Boston gun violence have been characterized as highly concentrated among a small number of people and in a small number of places. Problem analysis research conducted in the mid-1990s (Kennedy et al. 1996) and in the mid-2000s (Braga et al. 2008) describe gun homicide as being driven by approximately 1% of the city youth aged 15-24 who participated in ongoing gang conflicts governed by disrespect and status concerns and who were very well known to the criminal justice system. These accounts also pointed to the strong geographic concentration of serious gun violence. Kennedy et al. (1997) revealed that gang turf covered only 3.6% of Boston's 48 square miles but experienced 24% of gun assaults and 27% of youth homicide in 1994. More recently, Braga et al. (2008) reported that gun violence hot spots covered only 5.1% of Boston's geography but generated nearly 53% of fatal and non-fatal shootings. Braga et al. (2008) also noted that these gun violence hot spots were largely the same places that experienced the bulk of gun violence during the epidemic years of the late 1980s and early 1990s. Unfortunately, these cross-sectional studies did not attempt any longitudinal anal ysis of gun violence at specific places over time to determine whether the same locations were indeed generating a bulk of citywide gun violence trends.
While research in Boston and other U.S. cities has examined various dimensions of the
crime epidemic and the subsequent drop (e.g., the role of gangs, guns, and drugs), very little attention has been directed towards the spatial analysis of these epidemics. When spatial analysis has been done, it has been at the city, county, or Census tract-level which takes for granted the role of micro places in these epidemics. Indeed, the existing spatial diffusion analyses have not seriously considered the existence of micro-level variation of gun violence at particular street corners and street blocks within larger areal units. In this paper, we hope to address this gap in the literature by examining how the growth and decline of serious gun violence in Boston are influenced by different micro-crime trajec tories at street segments and intersections. As such, we examine the salience of micro-level units of analysis in understanding citywide gun violence trends rather than conducting analyses of spatial diffusion processes.
Data and Unit of Analysis
In this study, we measure serious gun violence by using computerized records of Boston Police Department official reports of Assault and Battery by Means of a Deadly Weapon— Firearm (ABDW—Firearm) incidents between January 1, 1980 and December 31, 2008. Incident reports are generated in the Boston Police Department by detectives or police officers after an initial response to a request for police service. These data were used to cast a wider net in examining the spatial distribution of gun violence in Boston and to increase the stability of our estimates through their larger yearly numbers over time. In the State of Massachusetts, ABDW—Firearm incidents essentially represent shooting events where guns were fired and victims were physically wounded by the fired bullets.5 Boston experienced 7,602 ABDW-Firearm incidents over the 29-year study period. As Fig. 1 shows, ABDW-Firearm incidents followed essentially the same trajectory as gun homicide in Boston between 1980 and 2008 (gun homicide counts were multiplied by five to show
5 See Massachusetts General Laws, Chap. 265, Sect. 15A.
^ Springer
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:51:16 UTC All use subject to http://about.jstor.org/terms
J Quant Criminol (2010) 26:33-53 39
— — Gun Homicide X 5 ——♦— ABDW-Firearm |
Flg. 1 Gun homicide and ABDW-firearm incident trends in Boston, 1980-2008. N = 7,602 incidents over 29 years
the trend on the same graph). ABDW-Firearm incidents were general stable during the early 1980s, climbed to a peak of 464 incidents in 1990, fell dramatically to a low of 110 incidents in 1997, and then rose again to a second peak of 311 incidents in 2006. These Boston-specific trends are representative of national-level trends in serious gun violence (see Cook and Laub 2002).
It is well known that police incident data, such as the FBI's Uniform Crime Reports, have shortcomings. For instance, crime incident data are biased by the absence of crimes not reported by citizens to the police and by police decisions not to record all crimes reported by citizens (see Black 1970). Although incident reports have flaws, careful analyses of these data can yield useful insights on crime (Schneider and Wiersema 1990). Moreover, official police incident data are widely used for assessing trends and patterns of gun crime (Blumstein 1995; Cook and Laub 2002) and the evaluation of gun violence reduction programs (see, e.g., Sherman and Rogan 1995; McGarrell et al. 2001; Cohen and Ludwig 2003).
The geographic units of interest for our study are micro places, defined as street seg ments and intersections, in Boston, Massachusetts. Street segments, sometimes referred to as street block faces, were defined as "the two block faces on both sides of a street between
two intersections" (Weisburd et al. 2004, p. 290). Drawing on the influential work of David Weisburd et al. (2004), we selected the street segment because it allowed a unit large enough to avoid unnecessary coding errors associated with smaller units such as addresses (Klinger and Bridges 1997; Weisburd and Green 1994), and small enough to avoid aggregation that might hide specific micro-level place trends. Street block faces have also been recognized as useful units of analysis for micro places that capture regularly recurring rhythms of social activity within the small physical boundary of a street segment (Hunter and Baumer 1982; Taylor et al. 1984; Weisburd et al. 2004).
Intersections, often called street corners, were defined as locations where two or more
streets crossed. Intersections were included in this analysis for practical and substantive
â Springer
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:51:16 UTC All use subject to http://about.jstor.org/terms
40 J Quant Criminol (2010) 26:33-53
reasons. When crimes occur at an intersection, police often record the location on the incident report as the intersection of two streets (e.g., "Massachusetts Avenue & Tremont Street") rather than assigning a specific address (e.g., "1010 Massachusetts Avenue") on a street segment. Rather than excluding events that were recorded at intersections, we decided to include intersections as a unit of analysis in the study. Previous studies excluded crime incidents at intersections for technical reasons, such as concerns over assigning these events to adjoining street segments (see Weisburd et al. 2004, p. 291). Substantively, many sociological inquiries have found that some inner-city residents meet, socialize, and sometimes live out significant portions of their daily lives on street corners (e.g., Liebow 1967; Whyte 1943). Street corners can also be important activity hubs within gang turf areas (Kennedy et al. 1997; Tita et al. 2003) and serve as locations for illicit enterprises such as street-level drag markets (Rengert et al. 2005; Weisburd and Green 1994).
To create a database suitable for executing longitudinal analyses of serious gun violence trends at micro places, we first created a database of records for each street segment and intersection in Boston. We then geocoded all ABDW-Firearm incidents to specific addresses and intersections so yearly counts could be tabulated for each individual street segments and intersections over the study time period. Using ArcGIS 9.3 SP1 and SQL Server 2000 software, we created a database with a record for each of the N — 18,155 street segments and N = 10,375 intersections in Boston. Next, 7,359 ABDW-Firearm incidents were successfully geocoded to a specific street address or intersection (96.8% of 7,602 ABDW-Firearm incidents). Incident reports with a location that could not be geo coded to a specific street segment or intersection (e.g., "Boston Common" or "Franklin Park") were not included in our analysis. 79.1% of ABDW-Firearm incidents were mat ched to a specific address on a street segment (5,823 of 7,359) and 21.9% of ABDW Firearm incidents were matched to a specific intersection (1,536 of 7,359). Geocoded incidents were aggregated to specific street segments and intersections and, for each of these street units, tallied into yearly counts over the 29-year period.
Analysis
Distribution of ABDW-Firearm Incidents at Street Segments and Intersections in Boston
For analytic purposes, street segments and intersections were treated as a single unit of analysis called a "street unit" (N = 28,530). When aggregated, only 3,294 "street units" experienced at least one ABDW-Firearm incident over the 29-year period (11.5% of 28,530). Put another way, the vast majority of street segments and intersections in Boston (88.5%) never experienced an ABDW-Firearm incident between 1980 and 2008. Fig. 2 presents the yearly counts of street units in Boston with at least one ABDW-Firearm incident between 1980 and 2008. The yearly number of street units that experience at least one ABDW-Firearm incident followed the same trajectory as the total yearly counts with 229 street units in 1980 (0.80% of 28,530), a peak of 369 street units in 1990 (1.29% of 28,530), a low of 103 street units in 1997 (0.36% of 28,530), and a second peak of 257 street units in 2006 (0.90% of 28,530). On average, during the study time period, less than 1% of street units (0.78% of 28,530) experienced at least one ABDW-Firearm incident in a given year.
Table 1 presents the distribution of ABDW-Firearm incidents at street units in Boston between 1980 and 2008. While the vast majority of street units in Boston never
£) Springer
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:51:16 UTC All use subject to http://about.jstor.org/terms
J Quant Criminol (2010) 26:33-53 41
J ^ ^-f J? J? / J? J? & ^ ^ ^ ^ & J" / ■!? ,<# / / ^ ^
Fig. 2 Combined segments and intersections that had at least one ABDW-firearm incident in Boston, 1980-2008. 3,294 "street units" had at least one ABDW-firearm incident during this time period (11.5% of 28,530)
Table 1 Distribution of ABDW-firearm incidents at street units in Boston, 1980-2008
N of incidents N of street % of Cum % Sum of % of Cum %
per street unit units street units street units incidents incidents incident
10 or more 65 0.23 0.23 1,032 14.02 14.02
5-9 269 0.94 1.17 1,730 23.51 37.53
2-4 1,037 3.63 4.80 2,674 36.34 73.87
1 1,923 6.74 11.54 1,923 26.13 100.0
0 25,246 88.46 100.0 0 0.00 100.0
Total 28,530 7,359
experienced a single incident, another 6.74% of street units experienced only one ABDW Firearm incident over the 29-year study period. This table also reveals that particular micro places suffer a vastly disproportionate amount of serious gun violence relative to other micro places in the city; only 65 street units generated 1,032 ABDW-Firearm incidents in Boston between 1980 and 2008. These highly active street units represented only 0.23% of the street segments and intersections in Boston but accounted for 14.02% of ABDW Firearm incidents between 1980 and 2008. Street units with five or more ABDW-Firearm
incidents represented only 1.17% of the street segments and intersections in Boston but accounted for 37.53% of ABDW-Firearm incidents between 1980 and 2008. Street units
with two or more ABDW-Firearm incidents represented only 4.8% of the street segments and intersection in Boston but generated 73.87% of the ABDW-Firearm incidents over the 29-year study time period.
Springer
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:51:16 UTC All use subject to http://about.jstor.org/terms
42 J Quant Criminol (2010) 26:33-53
This striking spatial concentration of gun violence in micro-places faces one important statistical challenge—to some extent, such a concentration may be an artifact of natural clustering of a small number of incidents among a large geographic space. The fact that each year, on average, there are fewer than 254 ABDW-Firearm incidents among nearly 28,530 street units suggests that even a purely random distribution might produce the observed clustering. To address this issue, we constructed a negative binomial distribution of the expected number of street segments that would have between zero and "5 or more" incidents for each 5-year period in the data.6 Then, we conducted a simple chi-square goodness of fit test to assess the extent to which the observed concentration significantly differs from the randomly generated negative binomial distributions. The results, presented in Table 5 of the Appendix to this paper, clearly demonstrate that in each time period there is significantly more clustering in the observed distribution than in the expected distri bution; this is especially true among the "5 or more category." This finding provides considerable support for the fact that the observed distribution is not merely an artifact of natural clustering.
Growth Curve Regression Models
As Table 1 suggests, yearly counts of ABDW-Firearm incidents at street segments and intersections in Boston were distributed in the form of rare event counts. There are well
documented problems associated with treating event count variables, which are discrete, as continuous realizations of a normal data generating process (King 1989). As such, methods such as standard mean difference tests and ordinary least squares regression that assume population normality of the dependent variable should not be used to analyze count data (Gardner et al. 1995). Rather, Poisson and negative binomial regression models are generally used to estimate models of the event counts (Long 1997).
In this analysis, we use a variation of a multilevel negative binomial regression model in order to analyze the change in ABDW-Firearm trends at micro places in Boston over the observation period. More specifically, we developed individual growth curve models to estimate street unit changes in ABDW-Firearm incidents over the observation period (Gelman 2005; Singer and Willet 2003). Here we use a longitudinal negative binomial model where we predict within unit variation at level 1 and between unit variation at level 2 using level 1 intercepts and slopes as outcomes. In non-technical terms, we are interested in accurately analyzing the overall serious gun violence trend of each of the street units during the observation period. Each street unit is also allowed to have its own slope and intercept in order to model different starting levels of serious gun violence as well as different rates of change.7 Formally, as shown in Eq. 1, the model is specified as:
pr(r*== r(2^+\)(rb;) (rh;) (1) where v„ is the count for the rth observation in the /th group. In the random effects model S-, is allowed to vary randomly across groups; namely, we assume that (1/(1 + <5) ~ Beta (r, s)). Equation 2 shows the joint probability of the counts for the (th group is:
6 Similar results were produced using a Poisson distribution and 3-year intervals.
7 Fixed-effects negative binomial regression models yield essentially the same results as the findings presented here. The results of the fixed-effects models are available upon request from the authors.
£} Springer
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:51:16 UTC All use subject to http://about.jstor.org/terms
J Quant Criminol (2010) 26:33-53 43
Pr(iö = y/i. ■ • Yb,, = yini |X,) = = y»\ Xit,ôi)(ôj)dôi
r(r + s)r(r + i ja)r(£+yn) tt +yu) r(r)r(5)r(r + 5 + YZi 4 + E"li yu) fi rft,)r(y# +1)
For Xi = (jtn,..., xini) and were/is the probability density function for <5;. As Eq. 3 shows, the resulting loglikehood is:
where Xit = cxp(xitß + offset,•,) and w; is the weight for the i'th group (Hausman et al. 1984).
Alternatively, previous studies have also employed latent class or semiparametric group based approaches (such as TRAJ models, e.g., Nagin 1999, 2005; Nagin and Land 1993) to predict "groups" of trajectories. These models use an approach similar to factor analysis which reduces large amounts of data into smaller theoretical groups. In so doing, each unit takes the slope (or coefficient) of the entire "grouping." Although TRAJ models have been used mainly to explain the trajectories of criminal careers, a few studies have applied them to place-based crime data (Griffiths and Chavez 2004; Weisburd et al. 2004). Recently, there has been much debate comparing these group-based approaches and growth curve models such as ones we employ here (e.g., Eggleston et al. 2004; Nagin 2004). While the present study does not wish to directly engage this debate, we wish to recognize this as a viable alternative approach to ours with potentially interesting results.
That said, we employ the longitudinal negative binomial models for two main reasons. First and foremost, our primary research interest is not to group street units into specific groups or classes, but rather to assess how the vanguard of ABDW-Firearm trends in Boston are driven by the specific ABDW-Firearm trends of individual geographic units. In other words, growth curve models allow a full characterization of the temporal sequence under consideration for each unit over the whole time period (see also Kubrin and Herting 2003). Thus, rather than assign individual street units to groups, we wish to assess individual slopes over the time period. Our analysis finds that once accurate slopes are obtained for each unit, dividing units into quartiles of slopes is an accurate way of portraying data visually and empirically without losing the actual value assigned to individual units.
Our second reason for using these models is an empirical one: there are large numbers of street units that have only one observation point. This suggests that a sizeable portion of all street units have no "trend" per se, but simply experience an isolated event. For this reason, we limited our main analysis to Boston street units with more than one event (N = 1,371; see Table 1). It is significant to note, however, that the findings of our models presented below remain robust even when including single-event-only street units.8
(3)
+ £ {In r(4 + yit) - In T(Ait) - In r(y;, + 1)}
8 When the single-event-only street units are included in our growth curve regression models, the results vary only slightly in magnitude. These results are available upon request from the authors.
ö Springer
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:51:16 UTC All use subject to http://about.jstor.org/terms
44 J Quant Criminol (2010) 26:33-53
Table 2 Summary statistics for key variables in growth curve regression models
Variable Mean SD Range
All street units model, N = 38,388
Number of ABDW-firearm incidents per year
Lagged N of ABDW-firearm incidents per year
Street unit type (0 = intersection, 1 = segment)
Street segments only model, N = 30,240
Length of street segment (Meters)
Number of ABDW-firearm incidents per year
Lagged N of ABDW-firearm incidents per year
Intersections only model, N = 8,148
Number of ABDW-firearm incidents per year
Lagged N of ABDW-firearm incidents per year
0.136 0.429 0-8
0.135 0.425 0-8
0.788 0.408 0-1
130.26 85.58 3.33-641.59
0.141 0.441 0-8
0.139 0.435 0-8
0.121 0.387 0-6
0.122 0.386 0-6
To capture linear and non-linear trends, the final growth curve regression models included Time, Time , and Time as covariates
Specification of the Growth Curve Regression Models in this Analysis
To ensure that there were no substantive differences in ABDW-Firearm trends at Boston
street segments and intersections, we estimated three separate growth curve regression models for intersections only, street segments only, and combined intersections and street segments (i.e., street units described above). For each model, the number of observations was calculated by multiplying the number of street units by the number of years. For instance, in the combined model, there were 38,388 observations derived from 1,371 street units over 28 years.9
Table 2 presents the summary statistics for key variables included in our final growth curve regression models. Our dependent variable is the number of ABDW-Firearm inci dents reported in each street unit per year. Consistent with prior criminological research which tends to show that past levels of violence are significant predictors of current levels of violence (e.g., Sampson et al. 1997) we included a covariate for the lagged (t—1) number of ABDW-Firearm incidents for each street unit. Since longer street segments are at an elevated risk of experiencing an ABDW-Firearm incident, we used ArcGIS 9.3 SP1 spatial analysis tools to calculate the length in meters of each street segment included in the analysis. Since intersections are represented by a point on the map, length could not be calculated for the intersections included in this analysis. However, in our combined model, we included a dichotomous dummy variable indicating whether the street unit was a street segment (1) or an intersection (0). Finally, to account for linear and nonlinear yearly trends in the dependent variable, we included a series of Time, Time2, and Time3 covariates. The base Time variable was measured as the simple linear additive progression for each year over the course of the 29-year observation period.
9 As discussed below, we lagged the number of ABDW-Firearm incidents for each street unit by 1 year. To calculate this variable, the time series loses the first year of data. Therefore, our final models analyzed 28 years of data rather than 29 years of data.
â Springer
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:51:16 UTC All use subject to http://about.jstor.org/terms
J Quant Criminol (2010) 26:33-53 45
Table 3 Results of growth curve regression models
Street segments only (1) Intersections only (2) Combined (3)
Time -0.0737*** -0.0839*** -0.0753***
(0.0053) (0.011) (0.0048) Time2 -0.000819*** -0.00346*** -0.00128***
(0.00030) (0.00065) (0.00027) Time3 0.000518*** 0.000657*** 0.000540***
(0.000042) (0.000089) (0.000038)
Length of street segment 0.00136***
(0.00022)
Lagged N of ABDW—Firearm 0.336*** 0.221*** 0.332*** (0.024) (0.071) (0.023)
Unit type 0.116***
(1 = street segment, 0 = intersection) (0.044) Constant 146.1*** 166.9*** 149.6***
(10.6) (21.7) (9.51) BIC 25,162.02 6,204.987 31,367.8
Wald Chi-square 502.16 92.78 540.46
Log likelihood -12,539.741 -3,070.9744 -15,641.677
Observations (units x years) 30,240 8,148 38,388
Number of "street units" 1,080 291 1,371
Standard errors in parentheses
*** p < 0.01, ** p< 0.05, * p< 0.1
As Eq. 4 shows, our final full model with all street units takes the following form:
Yij = a,- + Bu(TIME) + B2,(TIME)2 + B3,(TIME)3 + B4,(Lagged ABDW) + B5,-(UnitType) + eit (4)
where Yis the annual counts of ABDW-Firearm incidents at each street unit assuming the negative binomial dispersion discussed above.
Results
Table 3 presents the results of the three growth curve regression models examining trends in the observed count of incidents per street unit each year. Model (1) presents results of the equation for only the street segments, Model (2) presents the results for only the intersections, and Model (3) presents the results for the combined street units (street segments and intersections). The overall results were consistent across models; this sug gests that micro-level serious gun violence trends were similar at street segments and intersections in Boston over the 29-year study time period. In the three models, the neg ative and statistically significant (p < 0.01) coefficients on the Time and Time3 covariates suggest that ABDW-Firearm incidents as a whole declined among both street segments and intersections in Boston between 1980 and 2008. The positive and statistically significant (p < 0.01) coefficient on the Time2 covariate captures the increases in ABDW-Firearm incidents leading to the peaks in 1990 and 2006.
The positive and statistically significant (p < 0.01) coefficient for the lagged number of ABDW-Firearm incidents suggests that street units with higher ABDW-Firearm incidents continue to have higher levels of serious gun violence as time progresses. In Model (1), the
•ö Springer
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:51:16 UTC All use subject to http://about.jstor.org/terms
46 J Quant Criminol (2010) 26:33-53
coefficient for the length of street segment is positive and statistically significant (p < 0.01). This suggests that longer street segments experience higher numbers of ABDW-Firearm incidents relative to shorter street segments. Model (3) reveals that, when street segments and intersections are combined, street segments tend to experience higher ABDW-Firearm incident numbers as compared to intersections (p < 0.01).
Thus far, the growth curve regression analysis simply modeled each street unit's trend over the study time period; post-estimation visual analysis was then used to compare the trends of individual units to the overall citywide trend in serious gun violence. The graphs in Fig. 3 divide Boston street units with more than one ABDW-Firearm incident into quartiles of the predicted linear slope and intercept from the growth curve regression models presented in Table 3. For illustration purposes, we present the mean slope of all Boston street units in that quartile. At face value, there seems to be two basic patterns in these four graphs. Street units in Groups 1 and 4 follow a volatile trend that is very similar to the overall trend in ABDW-Firearm incidents in Boston between 1980 and 2008. While
there are some modest peaks and valleys, street units in Groups 2 and 3 are generally stable over the same time period. This suggests that Boston has essentially two types of highly active street units: those that have volatile gun violence concentrations over time and those that have stable gun violence concentrations over time.
Table 4 presents the distribution of ABDW-Firearm Incidents in Boston between 1980 and 2008 among street units in quartile groups defined by growth curve regression models. Overall, the street units in the four groups accounted for 4.8% of the street units in Boston and 73.9% of the ABDW-Firearm incidents in Boston over the 1980-2008 time period. As the graphs in Fig. 2 suggest, street units in Groups 1 and 4 followed a more volatile trajectory while street units in Groups 2 and 3 exhibited a more stable trajectory. Following these general patterns, these street units were collapsed into two larger groups. Street units
Z o_
s
H
I
8 -
o -
1980 1990 2000 20101980 1990 2000 2010
year
Fig. 3 Mean slopes of street units in quartiles
â Springer
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:51:16 UTC All use subject to http://about.jstor.org/terms
J Quant Criminol (2010) 26:33-53 47
Table 4 Distribution of ABDW-firearm incidents in Boston, 1980-2008, among street units in quartile groups defined by growth curve regression models
Group N of street units % of 28,530 Sum of incidents % of 7,359 incidents street units
1 571 2.0 1,519 20.6
2 201 0.7 413 5.6
3 332 1.2 1,157 15.7
4 267 0.9 2,347 31.9
Total 1,371 4.8 5,436 73.9
Stable (Groups 2, 3) 533 1.9 1,570 21.3
Volatile (Groups 1, 4) 838 2.9 3,866 52.5
One incident only 1,923 6.7 1,923 26.1
with a "Stable" trajectory accounted for 1.9% of the street units in Boston and 21.3% of the ABDW-Firearm incidents in Boston over the 1980-2008 time period. Street units with a "Volatile" trajectory accounted for 2.9% of the street units in Boston and 52.5% of the ABDW-Firearm incidents in Boston over the 1980-2008 time period. As discussed earlier, street units that experienced only one incident during the entire 1980-2008 time period accounted for 6.7% of the 28,530 street units in Boston and 26.1% of the 7,359 total ABDW-Firearm incidents.
Fig. 4 presents the group yearly counts of ABDW-Firearm incidents in Boston over the study time period. This graph reveals that the street units in the "Volatile" concentration group are responsible for the largest share of the peaks and valleys in gun violence over time in Boston that correspond closely with the gun violence epidemic of the late 1980s and
Fig. 4 Group yearly counts of ABDW-firearm incidents in Boston, 1980-2008
£} Springer
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:51:16 UTC All use subject to http://about.jstor.org/terms
48 J Quant Criminol (2010) 26:33-53
early 1990s, the gun violence drop of the mid to late 1990s, and the post-2004 resurgence in gun violence. The street units in the "Stable" concentration group account for a smaller share of the yearly ABDW-Firearm incidents and exhibit a similar but much more modest trajectory when compared to the "Volatile" group. Street units in the one incident only group follow a generally stable trajectory that is congruent with overall citywide trends in gun violence but without the steep upturns and downturns.
Fig. 5 presents a Boston map that reveals the geography of the street segments and intersections in the "Volatile" and "Stable" trajectory groups. These high-gun violence activity street units tend to cluster along major thoroughfares, like Blue Hill Avenue and Washington Street, that run through Roxbury, Dorchester, and Mattapan. Groups of these high-gun violence activity street units cluster in well-known gun violence hot spot areas such as the Lenox Street/Lower Roxbury area, Heath and Academy housing projects, Egleston Square, Dudley Square, Orchard Park housing project, Grove Hall, Franklin Hill housing project, Franklin Field housing projects, and Morton and Norfolk Streets neigh borhood (see Braga et al. 2008).
Conclusion
Our analyses suggest that city-level gun violence trends may best be understood by the analyses of trends at a very small number of micro places, such as street segments and intersections, rather than analyses of trends at larger areal units such as neighborhoods, arbitrarily-defined policing districts, or Census tracts. These levels of aggregation may obscure important place-based dynamics that vary within larger geographic boundaries. As Fig. 5 suggests, a longitudinal spatial analysis of gun violence trends at larger spatial units in Boston Police Department's District D-4, covering the Back Bay, Fenway, South End, and Lower Roxbury neighborhoods, would miss important micro-level variations associ ated with particular street segments and intersections that surround and include historical gang turfs and rivalries in the Lenox Street, Villa Victoria, and Castle Square housing projects (located in Fig. 5 in the D-4 areas with the dense concentrations of high-activity gun violence street units).
Our analyses also suggest that the "flood in a canyon" characterization of the gun violence epidemic of the late 1980s and early 1990s may actually be an understatement. Defining the at-risk population as including young, minority males living in disadvantaged neighborhoods is not refined enough to capture the extreme concentration of gun violence in urban environments. Urban gun violence trends may be best understood as generated by a very small number of high-risk individuals who participate in high-risk social networks and perpetrate their shootings at a very small number of high-risk micro places. In 2006, about 1% of Boston's youth ages 15-24 participated in gangs and these gangs accounted for 50% of total homicides, 77% of youth homicides, and 70% of fatal and non-fatal shootings in Boston (Braga et al. 2008). These findings are consistent with previous research on the high concentration of gun violence among a small number of gang involved individuals during the early to mid-1990s (Kennedy et al. 1997). In this analysis, almost 89% of Boston street segments and intersections never experienced a single ABDW-firearm incident between 1980 and 2008. Some 6% of street segments and inter sections experienced a single ABDW-firearm incident during this same time period. Boston gun violence trends were largely generated by repeated incidents at less than 5% of its street segments and intersections; the gun violence epidemic and sudden downturn was
£} Springer
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:51:16 UTC All use subject to http://about.jstor.org/terms
J Quant Criminol (2010) 26:33-53 49
Fig. 5 The spatial distribution of micro places with stable and volatile concentrations of serious gun violence in Boston
almost completely driven by trends at about 3% of the city's micro places that exhibited volatile concentrations of serious gun violence over time.
Many analyses of the spread of lethal gun violence during the epidemic of the late 1980s and early 1990s sought to determine whether homicide spatially diffused across
<£) Springer
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:51:16 UTC All use subject to http://about.jstor.org/terms
50 J Quant Criminol (2010) 26:33-53
communities and, if it did, to articulate the mechanism(s) associated with the spread over time. As Blumstein and Cohen (2002, p. 8) suggest,
It was hypothesized that, similar to the role of a mosquito in transmitting malaria, guns serve as a vector of the homicide epidemic. Presumably the presence of guns is transmitted from individuals directly involved in crack markets or youth gangs, and the neighborhoods in which these enterprises are located, to other non-participating youths. Those others would likely be peers from the same neighborhood or adjoining neighborhoods, but could also be physically more remote because social networks are not necessarily confined geographically.
To the extent that gun violence does diffuse across urban landscapes, our analyses suggest that spatial diffusion would be limited to very few locations within so-called violent neighborhoods. Our analyses suggest that gun violence upswings and downturns are largely concentrated at a small number of gun violence hot spots that intensify and diminish over time. It is possible that gun violence trends at these places follow trajectories that are consistent with a spatial diffusion process (e.g., a suddenly "hot" street comer that drives up gun violence levels on surrounding street corners over the course of several months). In future analyses, we will examine spatial and temporal diffusion of assaultive gun violence across street segments and intersections in Boston. In particular, we will search for evidence of contagion of gun violence among adjacent street segments and intersections.
Finally, these findings strongly support the perspective that a city's portfolio of gun violence prevention programs should include interventions that are explicitly place-based; that is, certain prevention efforts should be focused in very specific locations rather than diffused across larger neighborhoods. For instance, there is a growing body of research that suggests hot spots policing is effective in preventing crime (Braga 2001 ; Skogan and Frydl 2004; Weisburd and Eck 2004). Hot spots policing programs have also been shown to produce crime prevention benefits when focused on places with high level of violent gun crimes (Cohen and Ludwig 2003; McGarrell et al. 2001; Sherman and Rogan 1995). Police executives should explicitly deploy officers to these locations with the charge of enhancing their visibility, increasing contacts with potential offenders, and engaging community problem solving techniques to understand the underlying conditions that give rise to these violent places.
Social service and opportunity provision programs should also be oriented towards par ticular street corners and blocks that generate high levels of gun violence. For instance, street outreach workers can be deployed in these areas to work with gang-involved and criminally active youth who are at an elevated risk of shooting someone or being shot themselves (Kennedy et al. 1996; Skogan et al. 2008). It is obviously important to consider addressing the social networks and relations among groups that drive violent behavior that is manifested in repeated gun assaults at particular places. If gun violence can be reduced at a small number of micro places in the city, city wide gun violence rates will be positively impacted.
Acknowledgments We would like to thank Commissioner Edward F. Davis, Superintendent Paul Fitz gerald, Carl Walter, and Richard Laird for their support of this research. We would also like to thank David Weisburd, Alex Piquero, James Lynch, and the participants at the Crime and Place Working Group special session on "The Empirical Evidence on the Relevance of Place in Criminology" held at George Mason University on April 22, 2009 for their helpful comments on an earlier version of this paper.
Appendix
See Table 5.
^ Springer
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:51:16 UTC All use subject to http://about.jstor.org/terms
J Quant Criminol (2010) 26:33-53 51
Table S Distribution of ABDW-firearm incidents in Boston by observed and expected number of street units each year, (a) 1980-1984, (b) 1985-1989, (c) 1990-1994, (d) 1995-1999, (e) 2000-2004, and (f) 2005-2008
1980-1984 1985-1989 1990-1994
Observed Expected Observed Expected Observed Expected
0 27,565 28,227 27,529 28,242 27,309 28,141
1 791 210 776 200 883 239
2 112 70 129 55 201 104
3 35 14 56 21 76 25
4 9 9 17 6 27 13
5+ 18 0 23 6 34 8
Chi-square 1.18e + 0.04, p = 0.000 6.5e + 04, p = 0.000 3.8e + 04, p = 0.000
1995-1999 2000-2004 2005-2008 Total
Observed Expected Observed Expected Observed Expected Observed Expected
0 27,872 28,339 27,904 28,341 27,753 28,287 5,236 27,250
1 528 132 506 120 572 164 1,923 717
2 86 47 87 46 129 53 598 291
3 23 7 23 16 45 19 278 143
4 16 5 5 4 16 5 161 63
5+ 5 0 5 3 15 2 334 66
Chi-square 1.8e + 04, p = 000 1.2e + 04, p = 0.000 1.9e + 04, p = 0.000 2.1e + 05, p = 0.000
References
Black D (1970) The production of crime rates. Am Sociol Rev 35:733-748 Block R (1977) Violent crime: environment, interaction, and death. Lexington books, Lexington Block CR, Block R (1993) Street gang crime in Chicago, research in brief. U.S. National Institute of Justice,
Washington, DC Block CR, Christakos A, Jacob A, Przybylski R (1996) Street gangs and crime: patterns and trends in
Chicago, research bulletin. Illinois Criminal Justice Information Authority, Chicago Blumstein A (1995) Youth violence, guns, and the illicit-drug industry. J Crim Law Criminol 86:10-36 Blumstein A, Cohen J (2002) Diffusion processes in homicide, final report submitted to the U.S. Department
of Justice, Carnegie Mellon University, Pittsburgh, PA Blumstein A, Wallman J (eds) (2000) The crime drop in America. Cambridge University Press, New York Braga AA (2001) The effects of hot spots policing on crime. Ann Am Acad Polit Soc Sei 578:104-125 Braga AA (2003) Serious youth gun offenders and the epidemic of youth violence in Boston. J Quant
Criminol 19:33-54
Braga AA, Kennedy DM, Waring EW, Piehl AM (2001) Problem-oriented policing, deterrence, and youth violence: an evaluation of Boston's Operation Ceasefire. J Res Crime Delinq 38:195-225
Braga AA, Hureau D, Winship C (2008) Losing faith? Police, black churches, and the resurgence of youth violence in Boston. Ohio St J Crim Law 6:141-172
Cohen J, Ludwig J (2003) Policing crime guns. In: Ludwig J, Cook PJ (eds) Evaluating gun policy: effects on crime and violence. Brookings Institution Press, Washington, DC, pp 217-239
Cohen J, Tita G (1999) Diffusion in homicide: exploring a general method for detecting spatial diffusion processes. J Quant Criminol 15:451-494
Cook P, Laub J (1998) The unprecedented epidemic in youth violence. In: Tonry M, Moore M (eds) Youth violence, vol 24. Crime and justice: a review of research. University of Chicago Press, Chicago, pp 27-64
â Springer
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:51:16 UTC All use subject to http://about.jstor.org/terms
52 J Quant Criminol (2010) 26:33-53
Cook P, Laub J (2002) After the epidemic: recent trends in youth violence in the United States. In: Tonry M (ed) Crime and justice: a review of research, vol 29. University of Chicago Press, Chicago, pp 1-38
Eggleston EP, Laub JH, Sampson RJ (2004) Methodological sensitivities to latent class analysis of long term criminal trajectories. J Quant Criminol 20:1-26
Gardner W, Mulvey EP, Shaw EC (1995) Regression analyses of counts and rates: Poisson, overdispersed Poisson, and negative binomial models. Psychol Bull 118:392-404
Gelman A (2005) Analysis of variance: why it is more important than ever. Ann Statist 33:1-53 Griffiths E, Chavez JM (2004) Communities, street guns, and homicide in Chicago, 1980-1995: merging
methods for examining homicide trends across space and time. Criminology 42:941-978 Hagedom JM (1988) People and folks: gangs, crime and the underclass in a Rustbelt City. Lake View Press,
Chicago Hausman J, Hall BH, Griliches Z (1984) Econometric models for count data with an application to the
patents-R&D relationship. Econometrica 52:909-938 Hunter AJ, Baumer TL (1982) Street traffic, social integration, and fear of crime. Sociol Inq 52:122-131 Kennedy DM, Piehl AM, Braga AA (1996) Youth violence in Boston: gun markets, serious youth offenders,
and a use-reduction strategy. Law Contemp Probl 59:147-196 Kennedy DM, Braga AA, Piehl AM (1997) The (un)known universe: mapping gangs and gang violence in
Boston. In: Weisburd D, McEwen JT (eds) Crime mapping and crime prevention. Criminal Justice Press, Monsey, pp 219-262
Kennedy DM, Braga AA, Piehl AM (2001) Developing and implementing Operation Ceasefire. In: Reducing gun violence: The Boston gun project's operation ceasefire, National Institute of Justice, U.S. Department of Justice, Washington, DC, pp 5-53
King G (1989) Event count models for international relations: generalizations and applications. Int Stud Q 33:123-147
Klinger D, Bridges G (1997) Measurement error in calls-for-service as an indicator of crime. Criminology 35:705-726
Kubrin CE, Herting JR (2003) Neighborhood correlates of homicide trends: an analysis using growth-curve modeling. Sociol Q 44:329-350
Liebow E (1967) Tally's comer: a study of Negro streetcorner men. Little, Brown, Boston Long JS (1997) Regression models for categorical and limited dependent variables, advanced quantitative
techniques in the social sciences, vol 7. Sage Publications, Thousand Oaks McGarrell EF, Chermak S, Weiss A, Wilson J (2001) Reducing firearms violence through directed police
patrol. Criminol Pub Pol 1:119-148 Messner SF, Anselin L, Baller RD, Hawkins DF, Deane G, Tolnay SE (1999) The spatial patterning of
county homicide rates: an application of exploratory data analysis. J Quant Criminol 15:423-450 Moore M, Tonry M (1998) Youth violence in America. In: Tonry M, Moore M (eds) Youth violence, vol 24.
Crime and justice: a review of research. University of Chicago Press, Chicago, pp 1-26 Nagin D (1999) Analyzing developmental trajectories: a semiparametric group-based approach. Pyschol
Meth 4:139-157
Nagin D (2004) Response to "methodological sensitivities to latent class analysis of long-term criminal trajectories". J Quant Criminol 20:26-37
Nagin D (2005) Group-based modeling of development over the life course. Harvard University Press, Cambridge
Nagin D, Land KC (1993) Age, criminal careers, and population heterogeneity: specification and estimation of a nonparametric, mixed Poisson model. Criminology 31:327-362
Papachristos A (2009) Murder by structure: the social structure of gang homicide. Am J Sociol 115:74-128 Pierce GL, Spaar S, Briggs LR (1988) The character of police work: strategic and tactical implications.
Center for Applied Social Research, Northeastern University, Boston Police Executive Research Forum (2006) A gathering storm—violent crime in America. Police Executive
Research Forum, Washington, DC Reiss AJ, Roth J (eds) (1993) Understanding and preventing violence. National Academy Press, Wash
ington, DC Rengert GF, Ratcliffe JH, Chakravorty S (2005) Policing illegal drug markets: geographic approaches to
crime reduction. Criminal Justice Press, Monsey Robinson JB, Lawton BA, Taylor RB, Perkins DD (2003) Multilevel longitudinal impacts of incivilities:
fear of crime, expected safety, and block satisfaction. J Quant Criminol 19:237-274 Rosenfeld R, Bray TM, Egley A (1999) Facilitating violence: a comparison of gang-motivated, gang
affiliated, and nongang youth homicides. J Quant Criminol 15:495-516 Sampson RJ, Raudenbush SW, Earls F (1997) Neighborhoods and violent crime: a multilevel study of
collective efficacy. Science 277:918-924
â Springer
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:51:16 UTC All use subject to http://about.jstor.org/terms
J Quant Criminol (2010) 26:33-53 53
Schneider VW, Wiersema B (1990) Limits and use of uniform crime reports. In: MacKenzie DL, Baunach PJ, Roberg RR (eds) Measuring crime. State University of New York Press, Albany, pp 21-48
Shaw CR, McKay HD (1942) Juvenile delinquency in urban areas. University of Chicago Press, Chicago Sherman LW, Rogan D (1995) Effects of gun seizures on gun violence: 'hot spots' patrol in Kansas city. Just
Q 12:755-782 Sherman LW, Gartin P, Buerger M (1989) Hot spots of predatory crime: routine activities and the crimi
nology of place. Criminology 27:27-55 Singer JD, Willet JB (2003) Applied longitudinal data analysis: modeling change and event occurrence.
Oxford University Press, New York Skogan W, Frydl K (eds) (2004) Fairness and effectiveness in policing: the evidence, committee to review
research on police policy and practices. National Academies Press, Washington, DC Skogan W, Hartnett S, Bump N, DuBois J (2008) Executive summary evaluation of ceasefire-Chicago, final
report to the U.S. National Institute of Justice, Northwestern University, Institute for Policy Research, Chicago
Spelman W (1995) Criminal careers of public places. In: Eck JE, Weisburd DL (eds) Crime and place. Criminal Justice Press, Monsey, pp 115-144
Taylor RB (1999) Crime, grime, fear and decline: a longitudinal look, research in brief. U.S. National Institute of Justice, Washington, DC
Taylor RB (2001) Breaking away from broken windows: Baltimore neighborhoods and the nationwide fight against crime. Westview Press, Boulder
Taylor RB, Gottfredson SD (1986) Environmental design, crime, and prevention: an examination of community dynamics. In: Reiss AJ, Tonry M (eds) Communities and crime, vol 8. Crime and justice: a review of research. University of Chicago Press, Chicago, pp 387-416
Taylor RB, Gottfredson SD, Brower S (1984) Block crime and fear: defensible space, local social ties, and territorial functioning. J Res Crime Delinq 21:303-331
Tita G, Riley KJ, Greenwood P (2003) From Boston to Boyle Heights: the process and prospects of a "pulling levers" strategy in a Los Angeles barrio. In: Decker SH (ed) Policing gangs and youth violence. Wadsworth, Belmont, pp 102-130
Weisburd DL, Eck JE (2004) What can police do to reduce crime, disorder, and fear? Ann Am Acad Polit Soc Sei 593:42-65
Weisburd DL, Green (Mazerolle) L (1994) Defining the street-level drug market. In: MacKenzie DL, Uchida CD (eds) Drugs and crime: evaluating public policy initiatives. Sage Publications, Thousand Oaks, pp 61-76
Weisburd DL, Bushway S, Lum C, Yang S-M (2004) Trajectories of crime at places: a longitudinal study of street segments in the city of Seattle. Criminology 42:283-321
Whyte WF (1943) Street corner society: the social structure of an Italian slum. University of Chicago Press, Chicago
Wilson WJ (1987) The truly disadvantaged. University of Chicago Press, Chicago Wilson WJ (1996) When work disappears. Knopf, New York Wolfgang M (1958) Patterns in criminal homicide. University of Pennsylvania Press, Philadelphia
Ô Springer
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:51:16 UTC All use subject to http://about.jstor.org/terms
- Contents
- p. [33]
- p. 34
- p. 35
- p. 36
- p. 37
- p. 38
- p. 39
- p. 40
- p. 41
- p. 42
- p. 43
- p. 44
- p. 45
- p. 46
- p. 47
- p. 48
- p. 49
- p. 50
- p. 51
- p. 52
- p. 53
- Issue Table of Contents
- Journal of Quantitative Criminology, Vol. 26, No. 1 (March 2010) pp. 1-163
- Front Matter
- Editors' Introduction: Empirical Evidence on the Relevance of Place in Criminology [pp. 1-6]
- Is it Important to Examine Crime Trends at a Local "Micro" Level?: A Longitudinal Analysis of Street to Street Variability in Crime Trajectories [pp. 7-32]
- þÿ�T�h�e� �C�o�n�c�e�n�t�r�a�t�i�o�n� �a�n�d� �S�t�a�b�i�l�i�t�y� �o�f� �G�u�n� �V�i�o�l�e�n�c�e� �a�t� �M�i�c�r�o� �P�l�a�c�e�s� �i�n� �B�o�s�t�o�n�,� �1�9�8�0���2�0�0�8� �[�p�p�.� �3�3�-�5�3�]
- Activity Fields and the Dynamics of Crime: Advancing Knowledge About the Role of the Environment in Crime Causation [pp. 55-87]
- Permeability and Burglary Risk: Are Cul-de-Sacs Safer? [pp. 89-111]
- Modeling Micro-Level Crime Location Choice: Application of the Discrete Choice Framework to Crime at Places [pp. 113-138]
- þÿ�A�s�s�e�s�s�i�n�g� �t�h�e� �S�p�a�t�i�a�l���T�e�m�p�o�r�a�l� �R�e�l�a�t�i�o�n�s�h�i�p� �B�e�t�w�e�e�n� �D�i�s�o�r�d�e�r� �a�n�d� �V�i�o�l�e�n�c�e� �[�p�p�.� �1�3�9�-�1�6�3�]
12 sources/Brewer and Heitzeg.pdf
http://abs.sagepub.com
American Behavioral Scientist
DOI: 10.1177/0002764207307745 2008; 51; 625 American Behavioral Scientist
Rose M. Brewer and Nancy A. Heitzeg Complex
Color-Blind Racism, and the Political Economy of the Prison Industrial The Racialization of Crime and Punishment: Criminal Justice,
http://abs.sagepub.com/cgi/content/abstract/51/5/625 The online version of this article can be found at:
Published by:
http://www.sagepublications.com
can be found at:American Behavioral Scientist Additional services and information for
http://abs.sagepub.com/cgi/alerts Email Alerts:
http://abs.sagepub.com/subscriptions Subscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.com/journalsPermissions.navPermissions:
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. by on January 5, 2008 http://abs.sagepub.comDownloaded from
625
American Behavioral Scientist Volume 51 Number 5
January 2008 625-644 © 2008 Sage Publications
10.1177/0002764207307745 http://abs.sagepub.com
hosted at http://online.sagepub.com
The Racialization of Crime and Punishment Criminal Justice, Color-Blind Racism, and the Political Economy of the Prison Industrial Complex Rose M. Brewer University of Minnesota at Twin Cities Nancy A. Heitzeg College of St. Catherine, St. Paul, Minnesota
The current explosion in criminalization and incarceration is unprecedented in size, scope, and negative consequences—both direct and collateral—for communities of color. These macro systems exist in relationality to the micro dynamics of living in the midst of police scrutiny, economic marginalization, and political disenfranchisement. Critical race theory is a guide for pedagogy and praxis in exploring the racist and clas- sist foundations of current micro and macro injustices. Using Supreme Court opinions and the voices of political prisoner/prisoners of conscience as evidence of the dominant text and the dissent, this article explores the following issues: the roots of U.S. law, criminal justice, and mass imprisonment in classism and racism; the political economy of the criminal justice system and the prison industrial complex; the intersectionality of injustices rooted in micro and macro systems; and the role of prisoners of conscience/ political prisoners in inspiring resistance to micro and macro injustice.
Keywords: prison industrial complex; color-blind racism; critical race theory; racism and the law; racism and the criminal justice system
The post–civil rights era explosion in criminalization and incarceration is funda-mentally a project in racialization and macro injustice. It is, too, a project deeply connected to political economic changes in advanced capitalism. Multinational glob- alization in search of cheaper and cheaper labor and profit maximization is part and parcel of the growth of the prison industrial complex. The ideological underpinnings of racialization and the political economy of inequality are at the core of this dis- cussion. It is the latest in a historically uninterrupted series of legal and political machinations designed to enforce White supremacy with its economic and social benefits both in and with the law; “all domination is, in the last instance, maintained through social control strategies” (Bonilla-Silva, 2001, p. 103). As movements for abolition and civil rights end the institutions of slavery, lynching, and legalized
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. by on January 5, 2008 http://abs.sagepub.comDownloaded from
626 American Behavioral Scientist
segregation, new and more indirect mechanisms for perpetuating systemic racism and its economic underpinnings have emerged. In this era of color-blind racism, there has been a corresponding shift from de jure racism codified explicitly into the law and legal systems to a de facto racism where people of color, especially African Americans, are subject to unequal protection of the laws, excessive surveillance, extreme segregation, and neo–slave labor via incarceration, all in the name of crime control. It is the current manifestation of the legal legacy of the racialized transfor- mations of plantations into prisons, of Slave Codes into Black Codes, of lynching into state-sponsored executions. The “imputation of crime to color” (Douglass, cited in Foner, 1955, p. 379) continues to the present as racial profiling and culminates ultimately in the new plantation—in the prison industrial complex.
As Coates (2004) rightly observes,
the reliance on social justice rather than civil justice is of critical importance. Civil jus- tice, with its explicit links to the rights and privileges of citizenship, does not (as Chief Justice Taney clearly argued), apply to those outside the pale of societal membership. Civil justice, also reliant upon the laws of the day, is controlled and contrived to pro- tect the powerful, often at the expense of the weak. (p. 848)
Indeed, if so-called civil justice is at the foundation of the persistent White supremacist economic and political structure, then it is futile to call for only macro- level political and legal remedies. It is, we will argue, this very civil justice that has been used in the service of a series of racial projects. It has been used to enslave, to segregate, to mete out unequal punishments for comparable crimes on the basis of race. In the past, it has been used to explicitly reify via the law the essentialist White supremacist paradigm. At present, civil justice has been at the center of legal claims of color-blindness, forwarding the notion that if race is no longer the basis for legal- ized discrimination, then it is no longer relevant to the law at all. It is civil justice that currently claims that when explicit racial discrimination is removed from the language of the law, it is magically removed from any societal impact and any sub- sequent legal remedy.
Social justice requires that the role of civil justice in racialization be made trans- parent. This requires social justice projects that emanate from the micro level, from stories and struggle. Using the theory and methods of Critical Race Theory (CRT), this article will attempt to begin this work. CRT proceeds from the premise that racial privilege and related oppression are deeply rooted in both our history and our law, thus making racism a “normal and ingrained feature of our landscape” (Delgado & Stefancic, 2000, p. xvii). CRT acknowledges the myriad ways in which the legal constructions of race have produced and reproduced systemic economic, political, and social advantages for Whites. Challenges to racism require micro-level efforts to expose the deep structures of racism first made possible by the legal benefits of what Harris (1993) called “whiteness as property.” “Whiteness” produced both tangible and intangible value to those who possessed it:
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. by on January 5, 2008 http://abs.sagepub.comDownloaded from
The concept of whiteness was premised on white supremacy rather than mere differ- ence. “White” was defined and constructed in ways that increased its value by rein- forcing its exclusivity. Indeed, just as whiteness as property embraced the right to exclude, whiteness as a theoretical construct evolved for the very purpose of racial exclusion. Thus, the concept of whiteness is built on both exclusion and racial subju- gation. This fact was particularly evident during the period of the most rigid racial exclusion, as whiteness signified racial privilege and took the form of status property. (Harris, 1993, p. 116)
Removing White supremacy from the law did not, of course, erase its property benefits, nor did a shift to color-blindness in the law eradicate racism. CRT offers a critique of civil rights legal reforms by noting that they failed to fundamentally chal- lenge racial inequality. As Bell (2000) noted, “the subordination of blacks seems to reassure whites of an unspoken, but no less certain, property right in their whiteness” (p. 7). In the post–civil rights era, this subordination continues via color-blind legal mechanisms, particularly criminal justice.
(It should be noted that whereas all communities of color suffer from racism in general and its manifestation in criminal justice in particular, “Black” has been the literal and figurative counterpart of “White.” Anti-Black racism is arguably at the very foundation of White supremacy [Bonilla-Silva, 2001, 2006; Feagin, 2000]. For this reason, in combination with the excessive overrepresentation of African Americans in the criminal justice system and the prison industrial complex, our analysis will largely focus on the ways in which the law has been a tool for the oppression of African Americans.)
CRT also offers the use of narratives and context to surface these deep structures. We adopt these methods here, relying on the narratives and counternarratives of both judicial opinions and political prisoners/prisoners of conscience, referred to by James (2003) as “imprisoned intellectuals.” The dominant story and the dissent reveal the deep roots of current practices and the extent to which methods of legally enforcing White supremacy merely shift and change shape over time. These com- peting narratives provide a micro-level foundation for exposing and challenging the systemic injustice that has persisted over centuries. The words of those empowered to interpret the racial meanings of the Constitution and the words of those oppressed by these very same decisions will illuminate the law and its application as a consis- tent, albeit subtly changing, project in racialization.
Current Situation of Criminal Injustice
There is no dispute as to the extent of the dramatic escalation in criminalization and incarceration in the United States that has occurred during the past 35 years. Much of this increase can be traced to the war on drugs and the rise of mandatory minimum sentences for drug crimes and some other felonies. More than 47 million
Brewer, Heitzeg / Prison Industrial Complex 627
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. by on January 5, 2008 http://abs.sagepub.comDownloaded from
Americans (or 25% of the adult population) have state or federal criminal records. An estimated 13 million Americans—6% of the adult population—are either cur- rently serving a sentence for a felony conviction or have been convicted of a felony in the past (Mauer & Chesney-Lind, 2002, p. 51). Approximately 7 million adults are currently under some sort of correctional supervision in the United States. More than 4 million are under some sort of community correctional supervision such as probation and parole; another 2 million are incarcerated in prisons and jails. More than 3,500 of these are awaiting execution, some for federal crimes and most for cap- ital offenses in 1 of the 38 states that still allow for capital punishment. For every 100,000 Americans, there are 699 in prison—this is the highest incarceration rate in the world (Bureau of Justice Statistics, 2004).
There is also no dispute that the poor and people of color, particularly African Americans, are dramatically overrepresented in these statistics at every phase of the criminal justice system. The overwhelming majority of those in prisons and jails were unemployed or employed in the minimum wage service sector at the time of their commitment offense (Bureau of Justice Statistics, 2004). More than three quar- ters of a million Black men are now behind bars, and 2 million are under some form of correctional supervision. One in every 8 Black men between the ages of 25 and 34 is in prison or jail. One in 3 African American men and 1 in 10 Latinos between the ages of 20 and 29 are under some sort of correctional supervision (Mauer & Chesney-Lind, 2002). Approximately 50% of all prisoners are Black, 30% are White, and 17% Hispanic. Whereas the adult male prison population has tripled in the past 20 years, the number of women incarcerated has increased tenfold during the same time span. Women represent the fastest growing sector of the prison popu- lation. More than 90,000 prisoners are women, and they are overwhelmingly women of color. African American women are 3 times more likely than Latinas and 6 times more likely than White women to be in prison. More than 60% of women who are in prison are serving time for nonviolent offense, especially for drugs (Bureau of Justice Statistics, 2004).
And there is no dispute as to the devastating impact of these policies and prac- tices on communities of color. In addition to the direct impact of mass criminaliza- tion and incarceration, there is a plethora of what Mauer and Chesney-Lind (2002) refer to as “invisible punishments.” These additional collateral consequences further decimate communities of color politically, economically, and socially. The current expansion of criminalization and mass incarceration is accompanied by legislation that further limits the political and economic opportunities of convicted felons and former inmates. Felony disenfranchisement is permanent in 14 states. Forty-eight states do not permit prison inmates to vote, 32 states disenfranchise felons on parole, and 28 states prohibit probationers from voting. Nationally, 40 million felons are dis- enfranchised; 2% of the nation on average cannot vote as a result of a felony con- viction. Of African American males, 13% are disenfranchised; in 7 states, 1 in 4 are permanently barred form voting. In Florida alone, nearly one third of all Black men
628 American Behavioral Scientist
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. by on January 5, 2008 http://abs.sagepub.comDownloaded from
are permanently disenfranchised (Mauer & Chesney-Lind, 2002). Twenty-five states bar felons from ever holding public office, 33 states place a lifetime ban on gun own- ership for convicted felons, and all states require driver’s license suspension for con- victed drug felons. States have also increased the occupational bans for convicted felons, prohibiting them from teaching, child care work, related work with children, or law enforcement. This is accompanied by eased access to criminal records, an increase in all employers’ checking criminal backgrounds, and new technology, which facilitates quick checks. Research indicates that the explosion in incarceration is negatively correlated with Black male employment rates (Travis, 2002). Drug felons are permanently barred from receiving public assistance such as Temporary Assistance for Needy Families, Medicaid, food stamps, or Supplemental Security Income. Drug use, possession, or sales are the only offenses other than welfare fraud that result in a ban on federal assistance. The welfare fraud ban is limited to 10 years. Probation or parole violations also result in the temporary suspension of federal assistance. Drug felons are also permanently prohibited from receiving federal finan- cial aid for education. Those convicted of drug felonies “or violent criminal activity or other criminal activity which would adversely affect the health, safety, or right to peaceful enjoyment of the premises by others” (Rubenstein & Mukamal, 2002) are permanently barred from public housing or Section 8. A growing number of private rental properties also screen for convicted felons. More than 20,000 persons each year are denied federal housing assistance due to a felony conviction (Rubenstein & Mukamal, 2002; Travis, 2002). A felony conviction by anyone in the household is grounds for eviction from public housing. Recent legislation also creates barriers for families and has particularly devastating consequences for women. Certain con- victed felons are prevented from being approved as adoptive or foster parents. Congress has accelerated the termination of parental rights for children who have been in foster care for 15 of the most recent 22 months. Nineteen states regard felony conviction as grounds for parental termination; 29 states identify felony conviction as grounds for divorce (Chesney-Lind, 2002; Ritchie, 2002). And finally, the condi- tions of incarceration contribute to physical illnesses (e.g., Hepatitis B and C, HIV/AIDS, tuberculosis, general lack of adequate medical care), injuries (e.g., phys- ical and sexual assaults from correctional officers and other inmates), and mental disorders that continue to plague former inmates, families, and their communities on release (Fellner, 2004).
The reasons for this unprecedented explosion in criminalization and incarcera- tion, however, are in dispute. The rhetoric of color-blind racism would have us believe that this situation is the unfortunate result of disproportionate Black and Latino participation in crime. These so-called “racial realists” (Brown et al., 2005) argue that racism is over, successfully eradicated by civil rights legislation, and that if racial inequality persists, it is “the problem of the people who fail to take respon- sibility for their own lives” (Brown et al., 2005, p. vii). This adherence to the ideol- ogy of color-blindness (a co-optation and subversion of the dream of Dr. King)
Brewer, Heitzeg / Prison Industrial Complex 629
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. by on January 5, 2008 http://abs.sagepub.comDownloaded from
pervades conservative political and intellectual discourse, the corporate media, and the minds of the public. Racism is widely held to be an individual problem, rather than structural and systemic, an integral feature of what Bonilla-Silva (2001) referred to as “racialized social systems” (p. 57). From this vantage point, the issue then is crime, not race, and certainly not racism.
On the contrary, many contest this color-blind interpretation of contemporary racial arrangements as well as its specific application to criminal justice (Bonilla- Silva, 2001, 2006; Brown et al., 2005; Feagin, 2000; Mauer & Chesney-Lind, 2002; Walker, Spohn, & DeLone, 2004). The scale, scope, and extremes of negative con- sequences—both direct and collateral—for communities of color are new, especially for women,1 but the role of criminal justice in policing, prosecuting, imprisoning, and executing people of color has deep historical roots. What is not new is the racist and classist economic and political agenda that is foundational. The paradigms shift from essentialist to color-blind and the practices of oppression are refined and renamed, but the resulting inequality remains much the same. The law and its atten- dant machinery were, and still are, enforcers of both White supremacy and capital- ist interests.
The Past Is the Present
It is well established that our Constitution was written with a narrowly construed view of citizenship that at the time included only White, property-holding men. This property included both wives and children, but the most lucrative property of all— indeed that property that made any economic survival, let alone prosperity, possible— was slaves. By the time of the Constitutional Convention of 1787, the racial lines defining slave and free had already been rigidly drawn—White was free, and Black was slave. The Three Fifths Clause, the restriction on future bans of the slave trade, and limits on the possibility of emancipation through escape were all clear indications of the significance of slavery to the founders. Any doubt as to the centrality of White supremacy was erased a few decades later in the case of Scott v. Sandford (1857), where a majority of the Supreme Court denied the citizenship claims of Dred Scott and went further to declare that the Missouri Compromise requirement of balance between free and slave states in the expanding United States was a violation of the due process rights of slaveholders. Referring to the legal status of African Americans, Justice Taney’s opinion for the majority makes it painfully clear:
They are not included, and were not intended to be included, under the word “citizens” in the Constitution, and can therefore claim none of the rights and privileges which that instrument provides for and secures to citizens of the United States. On the contrary, they were at that time considered as a subordinate and inferior class of beings, who had been subjugated by the dominant race, and, whether emancipated or not, yet remained
630 American Behavioral Scientist
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. by on January 5, 2008 http://abs.sagepub.comDownloaded from
subject to their authority, and had no rights or privileges but such as those who held the power and the Government might choose to grant them.
The growing abolition movement could not overcome this legal bar with debates, written appeals, or legislative action. The economic and political interests of the slave states were too dependent on the rising trade in slaves and cotton. It took the armed resistance of slaves and radical abolitionists—of Vesey and Prosser, of Tubman, Turner, and finally Brown—to push the question into conflict. Frederick Douglass (1881, cited in Zinn, 2004, pp. 18-19) observed,
If John Brown did not end the war that ended slavery, he did at least begin the war that ended slavery. . . . Until this blow was struck, the prospect for freedom was dim, shadowy and uncertain. The irrepressible conflict was one of words, votes and com- promises. When John Brown stretched forth his arm, the sky was cleared. The time for compromises was gone.
The abolition of slavery did not result in the abolition of essentialist racism in the law; it merely called for new methods of legally upholding the property interests of Whiteness. In the presence of now freed Black labor, the vote was now offered to unpropertied White men, and, as Du Bois and others have argued, Whiteness played a central role in the reduction of class tensions. The “wages of whiteness” for the working class were material as well as social; “whiteness produced—and was repro- duced by the social advantage that accompanied it” (Harris, 1993, p. 116).
Postslavery, White supremacy in the law was accomplished by the introduction of a series of segregationist Jim Crow laws, a new model for an essentialist racial par- adigm that was now legitimated by so-called biology; the laws did not mandate that Blacks be accorded equality under the law because nature—not man, not power, not violence—has determined their degraded status (Harris, 1993, p. 118). The courts were complicit and explicit in their support for the purity and attendant property rights of Whiteness. This is made most dramatically clear in Plessy v. Ferguson (1896). In a challenge to the legalized segregation of public transportation in the state of Louisiana, Plessy argued that these laws denied him equality before the law. The majority disagreed and set forth the principle of separate but equal. Justice Brown wrote for the majority:
It is claimed by the plaintiff in error that, in a mixed community, the reputation of belonging to the dominant race, in this instance the white race, is “property,” in the same sense that a right of action or of inheritance is property. . . . We are unable to see how this statute deprives him of, or in any way affects his right to, such property. If he be a white man, and assigned to a colored coach, he may have his action for dam- ages against the company for being deprived of his so-called “property.” Upon the other hand, if he be a colored man, and be so assigned, he has been deprived of no property, since he is not lawfully entitled to the reputation of being a white man.
Brewer, Heitzeg / Prison Industrial Complex 631
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. by on January 5, 2008 http://abs.sagepub.comDownloaded from
The sole dissenter in Plessy sets up the juxtaposition of Jim Crow and color- blindness that frames the contemporary debate on race today. Justice Harlan, while acknowledging the reality of White supremacy, decries its support with the law:
The white race deems itself to be the dominant race in this country. And so it is, in pres- tige, in achievements, in education, in wealth, and in power. So, I doubt not, it will con- tinue to be for all time, if it remains true to its great heritage, and holds fast to the principles of constitutional liberty. But in view of the constitution, in the eye of the law, there is in this country no superior, dominant, ruling class of citizens. There is no caste here. Our constitution is color-blind, and neither knows nor tolerates classes among citizens. In respect of civil rights, all citizens are equal before the law.
The corollary to the enhanced promotion of Whiteness was—and still is—the ongoing devaluation of Blackness. The criminal justice system begins to play a new and crucial role here. Angela Y. Davis (2003), in Are Prisons Obsolete? traced the initial rise of the penitentiary system to the abolition of slavery; “in the immediate aftermath of slavery, the southern states hastened to develop a criminal justice system that could legally restrict the possibilities of freedom for the newly released slaves” (p. 29). There was a subsequent transformation of the Slave Codes into the Black Codes and the plantations into prisons. Laws were quickly passed that echoed the restrictions associated with slavery and criminalized a range of activities if the perpetrator was Black. The newly acquired 15th Amendment right to vote was cur- tailed by the tailoring of felony disenfranchisement laws to include crimes that were supposedly more frequently committed by Blacks (Human Rights Watch, 1998). And the libratory promise of the 13th Amendment—“Neither slavery nor involun- tary servitude shall exist in the United States”—contained a dangerous loophole: “except as a punishment for crime.” This allowed for the conversion of the old plan- tations to penitentiaries, and this, with the introduction of the convict lease system, permitted the South to continue to economically benefit from the unpaid labor of Blacks. As Davis (2003) noted, “the expansion of the convict lease system and the county chain gang meant that the antebellum criminal justice system, defined crim- inal justice largely as a means for controlling black labor” (p. 31).
After decades of resistance via legal challenges, grassroots organizing, boycotts, Letters from the Birmingham Jail, sit-ins, jail-ins, marches, and mass protest, legal- ized segregation began to come undone. Indeed, part of its undoing was the role that activists played in exposing the official and extralegal violence that had previously been cloaked in the legitimacy of the law. Emitt Till, Birmingham, Bloody Sunday, and more bared the lie. In the historic 1954 Brown v. Board of Education decision, Plessy was overturned; the essentialist racist paradigm was no longer codified in law. This was complete with the passage of the Civil Rights Act of 1964, the Voting Rights Act of 1965, and the passage of the 24th Amendment to the Constitution. Whereas there was once hope that the law itself could be pressed into the service of racial equality, those victories now seem bittersweet. Judge Robert L. Carter (1980),
632 American Behavioral Scientist
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. by on January 5, 2008 http://abs.sagepub.comDownloaded from
one of the attorneys who argued Brown, noted, “The fundamental vice was not legally enforced racial segregation itself; this was a mere by-product, a symptom of the greater and more pernicious disease—white supremacy” (pp. 23-24). Legally supported essentialist racism was about to be replaced with a more insidious counterpart—the paradigm of color-blindness.
Following the end of legalized racial discrimination, there was an especially con- certed effort to escalate the control of African Americans via the criminal justice system. Marable (1983) made this point: “White racists began to rely almost exclu- sively on the state apparatus to carry out the battle for white supremacy. . . . The criminal justice system became, in short, a modern instrument to perpetuate white hegemony” (pp. 120-121). These practices gain primacy during the post–civil rights years as the essentialist racist paradigm gives way to the new color-blind racism where race and racism are ostensibly absent from the law and all aspects of its enforcement. The criminal justice system provides a convenient vehicle for physically maintaining the old legally enforced color lines as African Americans are disproportionately policed, prosecuted, convicted, disenfranchised, and imprisoned. The criminal justice system and its culmination in the prison industrial complex also continues to guarantee the perpetual profits from the forced labor of inmates, now justifying their slavery as punishment for crime. Finally, the reliance on the criminal system provides the color-blind racist regime the perfect set of codes to describe racialized patterns of alleged crime and actual punishment without ever referring to race. As Davis (1998a) observed, “crime is one of the masquerades behind which ‘race’, with all its menacing ideological complexity, mobilizes old public ears and creates new ones” (p. 62). There is no discussion of race and racism; there is only public discourse about crime, criminals, gangs, and drug-infested neighborhoods. This color-bind conflagration of crime with race is, in addition, insidious in its dis- honesty and indirect effects; as Justice Powell, writing for the Supreme Court, noted, “discrimination within the judicial system is most pernicious because it is ‘a stimu- lant to that race prejudice which is an impediment to securing to [Black citizens] that equal justice which the law aims to secure to all others’“ (quoting Strauder v. West Virginia, 1880, in Batson v. Kentucky, 1986).
There were early warnings about the potentially devastating interconnections between race, crime, and the law in the era of late capitalism. The mid-20th-century criticism of the criminal justice system as foundationally racist initially emanated from the Black Power Movement’s critique of institutionalized racism and police bru- tality in communities of color. The writings of political prisoners (e.g., Angela Davis, Huey P. Newton, Assate Shakur) and prisoners of conscience (e.g., Malcolm X and George Jackson) brought racism and its intimate connection with the penitentiary to light. The 10 Point Program of the Black Panther Party began to make the connec- tions between capitalist exploitation of the Black community and the criminal justice system. Their demands provided the foundations for the contemporary critiques of the role of criminal justice in upholding both capitalism and White supremacy.
Brewer, Heitzeg / Prison Industrial Complex 633
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. by on January 5, 2008 http://abs.sagepub.comDownloaded from
We want an immediate end to police brutality and murder of black people. We want freedom for all black men held in federal, state, county and city prisons and jails. We want all black people when brought to trial to be tried in court by a jury of their peer group or people from their black communities, as defined by the Constitution of the United States. (Foner, 1970, pp. 3-5)
These warnings went unheeded; indeed, they were too often violently suppressed. The conflation of race and crime and the resultant rise of the criminal and prison industrial complexes did and does find support in judicial decisions that legitimate the central tenets of color-blind racism. The color-blind Constitution foreshadowed in Harlan’s dissent has now become the voice of the Supreme Court’s majority. Color- blindness as the new legal doctrine begins to emerge—despite judicial dissent—in cases involving affirmative action and other remedies to centuries of racial inequality. The Supreme Court adopts the color-blind model in Regents of the University of California v. Bakke (1978), where the ruling is in favor of a White student who claimed racial discrimination in his denial of admission to medical school. If the Constitution is to be color-blind, race can only be considered with strict scrutiny, even as a remedy for past discrimination. Justices Brennan and Marshall, in separate dis- sents, pointed out the flaws of this approach. Brennan observed,
Claims that law must be “color-blind” or that the datum of race is no longer relevant to public policy must be seen as aspiration rather than as description of reality . . . for reality rebukes us that race has too often been used by those who would stigmatize and oppress minorities. Yet we cannot . . . let color blindness become myopia which masks the reality that many “created equal” have been treated within our lifetimes as inferior both by the law and by their fellow citizens.
Justice Marshall’s dissent is even more prescient:
For it must be remembered that, during most of the past 200 years, the Constitution as interpreted by this Court did not prohibit the most ingenious and pervasive forms of dis- crimination against the Negro. Now, when a state acts to remedy the effects of that legacy of discrimination, I cannot believe that this same Constitution stands as a barrier.
As a series of subsequent cases from Bakke to Gratz v. Bollinger (2003) have shown, that same Constitution has indeed erected a legal barrier with claims of color-blindness (Bell, 2000; Brown et al., 2005).
Perhaps the most significant barrier to the pursuit of equality before the law comes in the case of McCleskey v. Kemp (1987). After a series of death penalty cases wherein the court decried racial discrimination in the application of the criminal laws’ ultimate penalty, it is here that the Supreme Court, in a 5 to 4 decision, clearly defined discrimination as individual, not institutionalized. Citing statistical evidence from the now famous Baldus study, McCleskey argued that the application of the death penalty in Georgia was fraught with racism. Defendants charged with killing
634 American Behavioral Scientist
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. by on January 5, 2008 http://abs.sagepub.comDownloaded from
White victims were more likely to receive the death penalty, and in fact, cases involving Black defendants and White victims were more likely to result in a sen- tence of death than cases involving any other racial combination. The majority did not dispute the statistical evidence but feared the consequences. If the court were to accept McCleskey’s claim, then the Equal Protection Clause of the 14th Amendment would apply to patterns of discrimination, to institutionalized racism and sexism, and to questions of structured inequality. These fears are expressed in Powell’s opin- ion for the majority:
First, McCleskey’s claim, taken to its logical conclusion, [481 U.S. 279, 315] throws into serious question the principles that underlie our entire criminal justice system. The Eighth Amendment is not limited in application to capital punishment, but applies to all penalties. Solem v. Helm, 463 U.S. 277, 289-290 (1983); see Rummel v. Estelle, 445 U.S. 263, 293 (1980) (POWELL, J., dissenting). Thus, if we accepted McCleskey’s claim that racial bias has impermissibly tainted the capital sentencing decision, we could soon be faced with similar claims as to other types of penalty. Moreover, the claim that his sentence [481 U.S. 279, 316] rests on the irrelevant factor of race easily could be extended to apply to claims based on unexplained discrepancies that correlate to membership in other minority groups, and [481 U.S. 279, 317] even to gender.
Justice Brennan’s impassioned dissent makes clear the implications of this decision:
Yet it has been scarcely a generation since this Court’s first decision striking down racial segregation, and barely two decades since the legislative prohibition of racial dis- crimination in major domains of national life. These have been honorable steps, but we cannot pretend that in three decades we have completely escaped the grip of a histori- cal legacy spanning centuries. Warren McCleskey’s evidence confronts us with the sub- tle and persistent influence of the past. His message is a disturbing one to a society that has formally repudiated racism, and a frustrating one to a Nation accustomed to regard- ing its destiny as the product of its own will. Nonetheless, we ignore him at our peril, for we remain imprisoned by the past as long as we deny its influence in the present.
Color-blind racism, with its call to ignore race and its treatment of any residual racism as individual and intentional, was now ensconced. Equal protection of the law was for individuals, not oppressed groups, and discrimination must be intentional and similarly individual. McCleskey closed off the last avenue for remedying structural inequality with the law and left us imprisoned by the past, imprisoned with the present.
Intersections: Criminal Injustice, Race, and Political Economy
The legal entrenchment of color-blind racism allowed White supremacist political and economic advantage to be pursued—unchecked by either law or public discourse— under the guise of criminal justice. Davis (1998b) noted,
Brewer, Heitzeg / Prison Industrial Complex 635
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. by on January 5, 2008 http://abs.sagepub.comDownloaded from
When the structural character of racism is ignored in discussions of crime and the ris- ing population of incarcerated people, the racial imbalance in jails and prisons is treated as a contingency. . . . The high proportion of black people in the criminal justice system is this normalized and neither the state nor the general public is required to talk or act on the meaning of this imbalance. . . . By relying on the alleged “race-blindness” of the law, black people are scrumptiously constructed as racial subjects, thus manipu- lated, exploited, and abused, while the structural persistence of racism—albeit in changed forms—is ignored. (p. 62)
As before, this newest political and legal construction of White supremacy is inti- mately interconnected with capitalist economic interests. The extreme racialization of criminal justice and the rise of the prison industrial complex are directly tied to the expansion of global economy, the decline of the industry and rise of the mini- mum wage service sector in the United States, and the growth of privatization of pubic services. The internationalization of the labor force and the turn to robotics, computers, and hi-tech are having a profound impact on labor in the United States and globally. The prison industrial complex is an expression and re-articulation of the political economy of late capitalism. The intense concentration and privatization of wealth in a few hands continues unchecked in this country. Indeed, the unparal- leled growth of corporate power is at the heart of the economic inequality African Americans and all working people are confronting.
Angela Davis (2003) again becomes important in interpreting the multiple inter- sections of race, economy, and the prison industrial complex. She traced the histor- ical links between current practices and the policies that emerged during the post–civil war era:
Vast amounts of black labor became increasingly available for use by private agents through the convict lease system and related systems such as debt peonage. This tran- sition set the historical stage for the easy acceptance of disproportionately black prison populations today. . . . We are approaching the proportion of black prisoners to white, during the era of the southern convict lease and country chain gang systems. Whether this human raw material is used for purposes of labor or for the consumption of com- modities provided by a rising number of corporations directly implicated in the prison industrial complex, it is clear that black bodies are considered dispensable within the “free world,” but as a source of profit in the prison world. (p. 95)
This quest for dispensable labor increasingly includes women of color who, in light of globalization, deindustrialization, and the dismantling of social services, are propelled by state economic interests into the slave labor markets of the prison industrial complex.
The prison industrial complex is not a conspiracy, but a confluence of special interests that include politicians who exploit crime to win votes, private companies that make millions by running or supplying prisons and small town officials who have turned to prisons as a method of economic development. (Silverstein, 1997)
636 American Behavioral Scientist
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. by on January 5, 2008 http://abs.sagepub.comDownloaded from
This complex now includes more than 3,300 jails, more than 1,500 state prisons, and 100 federal prisons in the United States. Nearly 300 of these are private prisons. More than 30 of these institutions are super-maximum facilities, not including the super-maximum units located in most other prisons. The prison industrial complex consumes vast amounts of tax dollars at the expense of education and other social programs. Each year, the United States spends more than $146 billion dollars on the criminal justice system, including police, the judiciary and court systems, and cor- rections. More than $50 billion of this is spent directly on corrections, with the majority of those expenditures going toward incarceration and executions—the two most expensive sentencing options (Bureau of Justice Statistics, 2004). The quest for profit has led to international U.S. expansion of the prison industrial complex in the United States. Both private companies and the U.S. military industrial complex rely on the global proliferation of both U.S. prisons and their internal practices at Basra, Abu Ghraib, Guantanamo Bay, and untold other locations.2
In essence, the prison industrial complex is a self-perpetuating machine where the vast profits (e.g., cheap labor, private and public supply and construction contracts, job creation, continued media profits from exaggerated crime reporting, and crime/punishment as entertainment) and perceived political benefits (e.g., reduced unemployment rates, “get tough on crime” and public safety rhetoric, funding increases for police, and criminal justice system agencies and professionals) lead to policies that are additionally designed to ensure an endless supply of “clients” for the criminal justice system (e.g., enhanced police presence in poor neighborhoods and communities of color; racial profiling; decreased funding for public education combined with zero-tolerance policies and increased rates of expulsion for students of color; increased rates of adult certification for juvenile offenders; mandatory min- imum and three-strikes sentencing; draconian conditions of incarceration and a reduction of prison services that contribute to the likelihood of recidivism; collateral consequences—such as felony disenfranchisement, prohibitions on welfare receipt, public housing, gun ownership, voting and political participation, and employment— that nearly guarantee continued participation in crime and return to the prison indus- trial complex following initial release). As Donzinger (1996) aptly noted,
Companies that service the criminal justice system need sufficient quantities of raw materials to guarantee long term growth in the criminal justice field, the raw material is prisoners. . . . The industry will do what it must to guarantee a steady supply. For the supply of prisoners to grow, criminal justice policies must insure a sufficient number of incarcerated Americans whether crime is rising or the incarceration is necessary. (p. 87)
In sum, Black workers, men and women, are at the center of this prison industrial process. They are used again as exploited labor and as consumers—of products pro- duced by prison labor. African Americans and other working people are less needed in the free labor market under current conditions of globalization. Highly exploited
Brewer, Heitzeg / Prison Industrial Complex 637
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. by on January 5, 2008 http://abs.sagepub.comDownloaded from
global workers match cheap prison labor. So the processes of deindustrialization and economic restructuring contribute to the process of accumulation for capital and the increasing immiseration of the Black poor, and this is true because many of the deci- sions are explicitly racial in form. Corporate actors choose to move out of Black communities on racial grounds (Brewer, l983). Thus, private prisons play a key role in the political economy of transnational capital. But so do public prisons. These prisons are equally tied to the corporate economy “and constitute an ever growing source of capitalist growth” (Davis, 2003, p. 96).
This exploitation of Black labor continues, made permissible, indeed possible, with the law. Although the names and legal legitimations have changed, there is little to distinguish the plantation from the penitentiary. Nevertheless, in the United States, Blacks have been a central political force in checking unabashed profit realization. Historically, this occurs through political struggle. We contend that it is only through organized political struggle and radical pedagogies for change that the current situ- ation will be transformed for social justice.
Transparency, Political Struggle, and Radical Pedagogies for Social Justice
The call for social justice is “an implicit call for solutions, a call for remedies, a call for action” (Coates, 2004, p. 850). As we have seen, the call for social justice cannot rely on civil justice or macro-level remedies alone; law has been the hand- maiden of what hooks (1992) has termed “the white supremacist capitalist patri- archy” in the ever-evolving political and economic exploitation of persons of color. To paraphrase Bell (1992), the 14th Amendment cannot save us. The call for social justice requires more.
As the latest project in racialization, criminal justice and the prison industrial complex have fundamentally racist and classist roots that must be exposed and abol- ished. Reform is insufficient; “there can be no compromise with capitalism. . . . There can be no compromise with racism, patriarchy, homophobia and imperialism” (Marable, 2002, p. 59). The work of justice must begin at the micro level; it must emerge from the grass roots. Drawing links between the movements to abolish slav- ery and segregation, Davis (2003) asked us to imagine the abolition of prisons and the creation of alternatives to mass incarceration with all its racist and classist corol- laries. Davis (1998b) identified three key dimensions of this work—public policy, community organizing, and academic research:
In order to be successful, this project must build bridges between academic work, leg- islative and other policy interventions, and grassroots campaigns calling, for example for the decriminalization of drugs and prostitution, and for the reversal of the present proliferation of prisons and jails. (pp. 71-72)
638 American Behavioral Scientist
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. by on January 5, 2008 http://abs.sagepub.comDownloaded from
Much of this work is in progress. Organizations such as The Sentencing Project (http://www.sentencingproject.org/), the Prison Moratorium Project (http://www.nomoreprisons.org), Critical Resistance (http://www.criticalresistance .org), Families Against Mandatory Minimum Sentencing, Amnesty International, Human Rights Watch, and the Prison Activist Resource Center (http://www .prisonactivist.org) have successfully linked a large and growing body of research with a critique of current practices and a call for legislative and policy change.
But this latest abolition movement faces a unique challenge. The paradigm of color-blind racism must be exposed before the deep connections between race, crime, and political economy become transparent. Hegemonic media coverage and misrepresentations about the reality of crime and criminal justice must be countered by multiple voices (Davis, 2003; Entman & Rojecki, 2000; Sussman, 2002). As long as the public course centers on crime—not race, class, or gendered racism—the true role of criminal justice and the prison industrial complex in preserving White supremacy in the context of advanced capitalism remains invisible. Davis (1998a) warns us,
The real human beings, designated by these numbers in a seemingly race neutral way, are deemed fetishtically exchangeable with the crimes they have committed. . . . The real impact of imprisonment on their lives need never be examined. The inevitable part played by the punishment industry in the reproduction of crime need never be dis- cussed. The dangerous and indeed fascist trend toward progressively greater numbers of hidden, incarcerated human populations is itself rendered invisible. (p. 63)
The true underpinnings of criminal justice and the prison industrial complex must become transparent. They must be surfaced by micro-level social justice projects. They must be surfaced via radical and relentless pedagogies of resistance; they must be surfaced in the stories, the narratives, of political prisoners and prisoners of con- science; they must be surfaced through the research, writing, and teaching of those whom Mumia Abu-Jamal (2005) called “radical intellectuals” and ultimately through the coalitions between the two that bridge the lines of difference between freedom and incarceration, as well as those of race, class, and gender.
As noted earlier, the writings of political prisoners and prisoners of conscience sounded the early warning about the role of the police, the courts, and prisons in eco- nomic and political repression of people of color. These works publicly clarified the extent to which there were political prisoners in the United States and served to raise the consciousness of what, in 1970, were the 200,000 mostly Black and Brown inmates in prisons and jails. Just as the writings of George Jackson, Assata Shakur, Huey P. Newton, and the early Angela Davis inspired an earlier generation of activists, so too do new voices rise in dissent from our prisons and jails. Leonard Peltier, Sanykia Shakur, Paul Wright’s Prison Legal News (http://www.prisonlegalnews .org), Marilyn Buck, and the prolific Mumia Abu-Jamal have given voice to the more
Brewer, Heitzeg / Prison Industrial Complex 639
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. by on January 5, 2008 http://abs.sagepub.comDownloaded from
than 2 million who are now incarcerated in increasingly harsh and isolated condi- tions. They have made the invisible horrors of the prison industrial complex visible and again sparked the call for resistance. They offer both an insider’s view and a deep critique of the law. Abu-Jamal (1995) wrote of Pennsylvania’s death row,
From daybreak to dusk black voices resound in exchanges of daily dramas that mark time in the dead zone. Echoes of Dred Scott ring in McCleskey’s opinion, again noting the paucity of black rights in the land of the free. Chief Justice Taney sits again, rein- carnate in the Rehnquist Court of the Modern Age. . . . One hundred and thirty three years after Scott, and still unequal in life, as in death. (pp. 92-93)
The writings of many political prisoners might have remained suppressed were it not for the efforts of scholars to bring them forward. This coalition between “organic and radical intellectuals” (Abu-Jamal, 2005) is crucial to the uncovering of the deep structural connections between race, political economy, and crime. The work of Angela Davis and Joy James is exemplary here. Their extensive writings on these matters and their careful attendance to connecting with those inside prison walls serve as a model for future work. In Imprisoned Intellectuals, James (2003) gave voice to the range of political prisoners and traced the common thread of resistance across generations, nationalities, racial/ethnic differences, genders, sexual orienta- tions, and political causes. She hopes that writing and reading will force a transfor- mative encounter “between those in the so-called free world seeking personal and collective freedoms and those in captivity seeking liberation from economic, mili- tary, racial/sexual systems” (James, 2003, p. 4).
The call to social justice, especially when addressing complex and cloaked sys- tems of racialization, requires critical and systematic documentation, the surfacing of deep political and economic structures, and bold confrontation. It requires the analytical tools and methods of multiple disciplines, as we have attempted to offer here. The dismantling of the White supremacist and capitalist machinery of criminal justice requires coalitions between intellectuals of all sorts. In the words of Mumia Abu-Jamal (2005),
Yet this world and life itself, is broader than the ivory towers of academia. Make external connections. Build bridges to the larger, nonacademic community. Build social, political and communal networks. . . . The word “radical” means from the roots—so, build roots! Touch base with real folks, and work for the only real source of liberty—life! (p. 179-184)
Ultimately, the realization of social justice will require still broader coalitions. Criminal justice and the prison industrial complex represent particular manifesta- tions of the entanglements of racialization, the law, and the global economy in late capitalism. Truly challenging this project in racialization calls for coalitions with those who are addressing different aspects of these foundation dilemmas. Audre Lorde (1984) reminded us that much of Western European history conditions us to
640 American Behavioral Scientist
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. by on January 5, 2008 http://abs.sagepub.comDownloaded from
see human differences in simplistic opposition to each other: dominant/subordinate, good/bad, up/down, superior/inferior. In a society where the good is defined in terms of profit rather than in terms of human need, there must always be some group of people who, through systematized oppression, can be made to feel surplus, to occupy the place of the dehumanized inferior. Within this society, that group is made up of Black and Third World people, working-class people, older people, women, gays/lesbians, and physically different and physically challenged people. Lorde went on to say,
Institutionalized rejection of difference is an absolute necessity in a profit economy which needs outsiders as surplus people. As members of such an economy, we have all been programmed to respond to the human differences among us with fear and loathing and to handle that difference in one of three ways: ignore it, and if that is not possible, copy it if we think it is dominant, or destroy it if we think is subordinate. But we have no patterns for relating across our human differences as equals. As a result, those differences have been misnamed and misused in the service of separation and confusion. (p. 115)
Most important, we must organize, continuing the legacy of struggle. We must come together across boundaries of national identity, gender, race, class, and ethnic- ity. We must work in alliance to realize the vision that another world is possible.
Notes
1. The extreme escalation in the female incarceration rate is new. In the United States, institutional- ization has been highly gendered, with prisons largely reserved for men and psychiatric facilities as the favored source of social control for “deviant women.” This is consistent with a historical pattern that has tended to “criminalize” men and “medicalize” women. Race and class, however, have always complicated this gendered division of punishment (see Ehrenreich & English, 1973). Women of color and women who are poor have always been more likely to be incarcerated than their White counterparts, and so it is unsur- prising that the current expansion of the women’s prison population has largely been at the expense of African Americans and Latinas. As the period of late capitalism is marked by a steady decimation of gov- ernmental funding in the areas of economics, education, and housing, these women increasingly become an important “disposable” source of profit for the prison industrial complex (Davis, 2003).
Whereas women’s prisons of the past used to be less draconian, this is no longer the case. Women, on average, serve longer sentences than men and are increasingly forced into harsh conditions of labor, including, at the extreme, female chain gangs. They are often housed in more makeshift conditions, located long distances from family and friends, denied routine medical care, and at great risk for sexual assault by the too often male correctional staff (Chesney-Lind, 2002; Davis, 2003).
The direct and collateral consequences of imprisonment for women are also consistent with historical patterns of exploitation and control that women of color have faced. From slavery to the present, women of color have faced sexual exploitation, state-sponsored family disruptions, and attempts to regulate repro- duction. Davis (2003) observed that the sexual assaults faced regularly by women in prison—in the forms of both attacks and routine practices such as body cavity searches—echo the sexualized punishments that accompanied slavery. In Killing the Black Body, critical legal scholar Dorothy Roberts (1997) docu- mented the long history of legal control over the reproduction of women of color, particularly African Americans. One of the contributing factors to the rise of the female prison population is the criminalization
Brewer, Heitzeg / Prison Industrial Complex 641
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. by on January 5, 2008 http://abs.sagepub.comDownloaded from
of reproduction for women of color who are poor or users of drugs. Women on welfare have been forced into the use of Depo and Norplant and offered economic incentives for having abortions. Pregnant women who have tested positive for substance use (usually crack cocaine) have been arrested and imprisoned for child abuse. These arrests and prosecutions have almost invariably involved Black women, even though they could equally apply to middle-class White women who continue to drink during pregnancies. Finally, the overwhelming majority of women in prison have children. The incarceration of women further deci- mates families and, in combination with new laws that speed the process of termination of parental rights, nearly guarantees a permanent loss of parental rights. In the broader sense, the prison industrial complex itself represents a new mechanism for controlling the reproduction of people of color. Just as it has replaced slavery as a new source of profit, it has also replaced the older program of suppressing Black reproduction via sterilization (Roberts, 1997) with the isolation and segregation of men and women behind prison walls.
2. The globalization of the prison industrial complex includes not only the proliferation of the struc- tures but abusive practices as well. The horrific abuses documented at Abu Ghraib are not an anomaly; they represent regular practices at U.S. prisons across the nation. Amnesty International and Human Rights Watch have documented decades-old patterns of human and civil rights abuses by local and fed- eral police/law enforcement officers as well as prison, jail, and Immigration and Naturalization Service detention officials. These include a variety of abusive police practices (e.g., racial profiling; excessive use of force—including the kicking and beating of restrained suspects with fists, batons, and flashlights; excessive use of dangerous chokeholds, hog-ties, and other restraints that have resulted in death; exces- sive use of tasers, stun guns, and chemical sprays; excessive use of deadly force—particularly in situa- tions involving car chases or unarmed Black male suspects; inappropriate use of strip searches; use of sexual abuse and torture in police precincts to extract confessions; and prison procedures [e.g., dangerous use of restraints—including four-point restraints, the “rail,” and the restraint chair—that have resulted in multiple deaths; the shackling of pregnant inmates; use of nudity, strip searches, and sexual humiliation and assault as a source of social control; failure to curtail sexual assaults on both male and female inmates by other inmates and guards; beatings by guards; denial of medical care or treatment; use of dogs, tasers, and chemical sprays; excessive use of super-max and isolation confinement; and denial of rights on reli- gious freedom, communications, and right to counsel]) (see Amnesty International, 1998; Eisner, 2004; Fellner, 2004; Human Rights Watch, 1998, 2000).
International nonprofit human rights organizations have consistently claimed that U.S. police and prison practices are in violation of several international treaties and protocols including, but not limited to, the Geneva Conventions, International Covenant on Civil and Political Rights, Convention Against Torture, International Convention on the Elimination of All Forms of Racial Discrimination, UN Universal Declaration of Human Rights, UN Body of Principles for the Protection of All Persons Under Any Form of Detention, UN Standard Minimum Rules for the Treatment of Prisoners, UN Code of Conduct for Law Enforcement Officials, and UN Basic Principles on the Use of Force and Firearms by Law Enforcement Officials. Human Rights Watch observed in a May 14, 2004, press release,
Perhaps if photos or videotapes of abuse in U.S. prisons were to circulate publicly, Americans would be galvanized to protest such treatment as they have the treatment of Iraqi prisoners. Absent such graphic and unavoidable evidence, it is all too likely that abuse will continue to be a part of many prison sentences.
Or perhaps they would continue to be publicly and legally ignored. In a chillingly accurate observation made in his dissent to the recent decision of Johnson v. California (2005), Justice Clarence Thomas noted, “The Constitution has always demanded less within the prison walls.”
642 American Behavioral Scientist
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. by on January 5, 2008 http://abs.sagepub.comDownloaded from
References
Abu-Jamal, M. (1995). Live from death row. New York: Seven Stories Press. Abu-Jamal, M. (2005). Intellectuals and the gallows. In J. James (Ed.), Imprisoned intellectuals:
America’s political prisoners write on life, liberation and rebellion (p. 179). New York: Rowman and Littlefield.
Amnesty International. (1998). United States of America: Rights for all. London: Author. Batson v. Kentucky, 476 U.S. 79, 87-88 (1986). Bell, D. (1992). Faces at the bottom of the well: The permanence of racism. New York: Basic Books. Bell, D. (2000). After we’re gone: Prudent speculations on America in a post-racial epoch. In R. Delgado
& J. Stefancic (Eds.), Critical race theory: The cutting edge (2nd ed., pp. 2-8). Philadelphia: Temple University Press.
Bonilla-Silva, E. (2001). White supremacy and racism in the post–civil rights era. Boulder, CO: Lynne Rienner. Bonilla-Silva, E. (2006). Racism without racist: Color-blind racism and the persistence of racial
inequality in the United States. New York: Rowman and Littlefield. Brewer, R. (1983). Black workers and corporate flight. Third World Socialists, 1, 9-13. Brown v. Board of Educ., 347 U.S. 483 (1954). Brown, M. K., Carnoy, M., Currie, E., Duster, T., Oppenheimer, D. B., Schultz, M. K., et al. (2005). White-
washing race: The myth of a color-blind society. Berkeley: University of California Press. Bureau of Justice Statistics. (2004). Sourcebook of criminal justice statistics. Washington, DC: Government
Printing Office. Carter, R. L. (1980). A reassessment of Brown v Board. In D. Bell (Ed.), Studies of Brown: New
perspectives on school de-segregation (PP. ??). New York: Teacher’s College Columbia University. Chesney-Lind, M. (2002). Imprisoning women: The unintended victims of mass incarceration. In M. Mauer
& M. Chesney-Lind (Eds.), Invisible punishment: The collateral consequences of mass imprisonment (pp. 79-94). New York: New Press.
Coates, R. (2004). If a tree falls in the wilderness: Reparations, academic silences, and social justice. Social Forces, 83(2), 841-864.
Davis, A. (1998a). Race and criminalization: Black Americans and the punishment industry. In J. James (Ed.), The Angela Y. Davis reader (pp. 61-73). New York: Blackwell.
Davis, A. (1998b). Racialized punishment and prison abolition. In J. James (Ed.), The Angela Y. Davis reader (pp. 96-110). New York: Blackwell.
Davis, A. (2003). Are prisons obsolete? New York: Seven Stories Press. Delgado, R., & Stefancic, J. (Eds.). (2000). Critical race theory: The cutting edge (2nd ed.). Philadelphia:
Temple University Press. Donzinger, S. (1996). The real war on crime: Report of the National Criminal Justice Commission. New York:
Perennial. Ehrenreich, B., & English, D. (1973). Complaints and disorders: The sexual politics of sickness. New York:
Feminist Press. Eisner, A. (2004). Gates of injustice: The crisis in America’s prisons. New York: Pearson. Entman, R. M., & Rojecki, A. (2000). The Black image White mind: Media and race in America. Chicago:
University of Chicago Press. Feagin, J. R. (2000). Racist America: Roots, realities, and future reparations. New York: Routledge. Fellner, James Esq. (2004) Prisoner Abuse: How Different are U.S. Prisons? New York: Human Rights
Watch. Foner, P. S. (Ed.). (1955). The life and writings of Frederick Douglass: Vol. 4. Reconstruction and after.
New York: International. Foner, P. S. (Ed.). (1970). The Black Panthers speak. Philadelphia: J. B. Lippincott. Gratz v. Bollinger, 02-516 (2003).
Brewer, Heitzeg / Prison Industrial Complex 643
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. by on January 5, 2008 http://abs.sagepub.comDownloaded from
Harris, C. (1993). Whiteness as property. In D. R. Roediger (Ed.), Black on White (pp. 103-118). New York: Schocken.
hooks, b. (1992). Black looks. Boston: South End. Human Rights Watch. (1998). Losing the vote: The impact of felony disenfranchisement laws in the U.S.
New York: Author. Human Rights Watch. (2000). Out of sight: Super-maximum security confinement in the United States.
NY: Author. Human Rights Watch. (2003). Ill-Equipped: U.S. Prisons and Offenders in the United States. NY: Author James, J. (Ed.). (2003). Imprisoned intellectuals: America’s political prisoners write on life, liberation
and rebellion. New York: Rowman and Littlefield. Johnson v. California, 03-636 543 U.S. 499 (2005). Lorde, A. (1984). Sister outsider: Essays and speeches. Freedom, CA: Crossing Press. Marable, M. (1983). How capitalism underdeveloped Black America. Boston: South End. Marable, M. (2002). Black radicalism and an economy of incarceration. In J. James (Ed.), States of
confinement: Policing, detention and prisons (p. 59). New York: Palgrave. Mauer, M., & Chesney-Lind, M. (Eds.). (2002). Invisible punishment: The collateral consequences of
mass imprisonment. New York: New Press. McCleskey v. Kemp, 482 U.S. 279 (1987). Plessy v. Ferguson, 163 U.S. 537 (1896). Regents of the University of California v. Bakke, 438 U.S. 265 (1978). Ritchie, B. (2002). The social impact of the mass incarceration of women. In M. Mauer & M. Chesney-
Lind (Eds.), Invisible punishment: The collateral consequences of mass imprisonment (pp. 136-149). New York: New Press.
Roberts, D. (1997). Killing the Black body: Race, reproduction and the meaning of liberty. New York: Vintage. Rubenstein, G., & Mukamal, D. (2002). Welfare and housing—Denial of benefits to drug offenders. In
M. Mauer & M. Chesney-Lind (Eds.), Invisible punishment: The collateral consequences of mass imprisonment (pp. 37-50). New York: New Press.
Scott v. Sandford, 60 U.S. 393 (1857). Silverstein, K. (1997, June). America’s private gulag. Prison Legal News. Sussman, P. Y. (2002). Media on prisons: Censorship and stereotypes. In M. Mauer & M. Chesney-Lind
(Eds.), Invisible punishment: The collateral consequences of mass imprisonment (pp. 258-278). New York: New Press.
Travis, J. (2002). Invisible punishment: An instrument of social exclusion. In M. Mauer & M. Chesney-Lind (Eds.), Invisible punishment: The collateral consequences of mass imprisonment (pp. 15-36). New York: New Press.
Walker, S., Spohn, C., & DeLone, M. (2004). The color of justice: Race, ethnicity and crime in America (3rd ed.). Belmont, CA: Wadsworth.
Zinn, H. (2004). Let the people speak. New York: Perennial.
Rose M. Brewer is the Morse Alumni Distinguished Teaching Professor of African American & African Studies at the University of Minnesota at Twin Cities. She has written extensively on Black families; race, class, and gender; and public policy and social change. Her most recent coauthored book is The Color of Wealth (2006, New Press).
Nancy A. Heitzeg is an associate professor of sociology and a codirector of the Critical Studies of Race/Ethnicity program at the College of St. Catherine, St. Paul, Minnesota. She is the author of Deviance: Rule-Makers and Rule-Breakers and has written and presented widely on race, class, gender, and social control; formal, medical, and extralegal color-blind racism; and social movements/social change.
644 American Behavioral Scientist
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. by on January 5, 2008 http://abs.sagepub.comDownloaded from
<< /ASCII85EncodePages false /AllowTransparency false /AutoPositionEPSFiles true /AutoRotatePages /None /Binding /Left /CalGrayProfile (Dot Gain 20%) /CalRGBProfile (sRGB IEC61966-2.1) /CalCMYKProfile (U.S. Web Coated \050SWOP\051 v2) /sRGBProfile (sRGB IEC61966-2.1) /CannotEmbedFontPolicy /Error /CompatibilityLevel 1.3 /CompressObjects /Off /CompressPages true /ConvertImagesToIndexed true /PassThroughJPEGImages true /CreateJDFFile false /CreateJobTicket false /DefaultRenderingIntent /Default /DetectBlends true /ColorConversionStrategy /LeaveColorUnchanged /DoThumbnails false /EmbedAllFonts true /EmbedJobOptions true /DSCReportingLevel 0 /EmitDSCWarnings false /EndPage -1 /ImageMemory 1048576 /LockDistillerParams false /MaxSubsetPct 100 /Optimize false /OPM 1 /ParseDSCComments true /ParseDSCCommentsForDocInfo true /PreserveCopyPage true /PreserveEPSInfo true /PreserveHalftoneInfo false /PreserveOPIComments false /PreserveOverprintSettings true /StartPage 1 /SubsetFonts true /TransferFunctionInfo /Apply /UCRandBGInfo /Remove /UsePrologue false /ColorSettingsFile () /AlwaysEmbed [ true /ACaslon-Bold /ACaslon-BoldItalic /ACaslon-Italic /ACaslon-Ornaments /ACaslon-Regular /ACaslon-Semibold /ACaslon-SemiboldItalic /AdobeCorpID-Acrobat /AdobeCorpID-Adobe /AdobeCorpID-Bullet /AdobeCorpID-MinionBd /AdobeCorpID-MinionBdIt /AdobeCorpID-MinionRg /AdobeCorpID-MinionRgIt /AdobeCorpID-MinionSb /AdobeCorpID-MinionSbIt /AdobeCorpID-MyriadBd /AdobeCorpID-MyriadBdIt /AdobeCorpID-MyriadBdScn /AdobeCorpID-MyriadBdScnIt /AdobeCorpID-MyriadBl /AdobeCorpID-MyriadBlIt /AdobeCorpID-MyriadLt /AdobeCorpID-MyriadLtIt /AdobeCorpID-MyriadPkg /AdobeCorpID-MyriadRg /AdobeCorpID-MyriadRgIt /AdobeCorpID-MyriadRgScn /AdobeCorpID-MyriadRgScnIt /AdobeCorpID-MyriadSb /AdobeCorpID-MyriadSbIt /AdobeCorpID-MyriadSbScn /AdobeCorpID-MyriadSbScnIt /AdobeCorpID-PScript /AGaramond-BoldScaps /AGaramond-Italic /AGaramond-Regular /AGaramond-RomanScaps /AGaramond-Semibold /AGaramond-SemiboldItalic /AGar-Special /AkzidenzGroteskBE-Bold /AkzidenzGroteskBE-BoldEx /AkzidenzGroteskBE-BoldExIt /AkzidenzGroteskBE-BoldIt /AkzidenzGroteskBE-Ex /AkzidenzGroteskBE-It /AkzidenzGroteskBE-Light /AkzidenzGroteskBE-LightEx /AkzidenzGroteskBE-LightOsF /AkzidenzGroteskBE-Md /AkzidenzGroteskBE-MdEx /AkzidenzGroteskBE-MdIt /AkzidenzGroteskBE-Regular /AkzidenzGroteskBE-Super /AlbertusMT /AlbertusMT-Italic /AlbertusMT-Light /Aldine401BT-BoldA /Aldine401BT-BoldItalicA /Aldine401BT-ItalicA /Aldine401BT-RomanA /Aldine401BTSPL-RomanA /Aldine721BT-Bold /Aldine721BT-BoldItalic /Aldine721BT-Italic /Aldine721BT-Light /Aldine721BT-LightItalic /Aldine721BT-Roman /Aldus-Italic /Aldus-ItalicOsF /Aldus-Roman /Aldus-RomanSC /AlternateGothicNo2BT-Regular /AmazoneBT-Regular /AmericanTypewriter-Bold /AmericanTypewriter-BoldA /AmericanTypewriter-BoldCond /AmericanTypewriter-BoldCondA /AmericanTypewriter-Cond /AmericanTypewriter-CondA /AmericanTypewriter-Light /AmericanTypewriter-LightA /AmericanTypewriter-LightCond /AmericanTypewriter-LightCondA /AmericanTypewriter-Medium /AmericanTypewriter-MediumA /Anna /AntiqueOlive-Bold /AntiqueOlive-Compact /AntiqueOlive-Italic /AntiqueOlive-Roman /Arcadia /Arcadia-A /Arkona-Medium /Arkona-Regular /ArrusBT-Black /ArrusBT-BlackItalic /ArrusBT-Bold /ArrusBT-BoldItalic /ArrusBT-Italic /ArrusBT-Roman /AssemblyLightSSK /AuroraBT-BoldCondensed /AuroraBT-RomanCondensed /AuroraOpti-Condensed /AvantGarde-Book /AvantGarde-BookOblique /AvantGarde-Demi /AvantGarde-DemiOblique /Avenir-Black /Avenir-BlackOblique /Avenir-Book /Avenir-BookOblique /Avenir-Heavy /Avenir-HeavyOblique /Avenir-Light /Avenir-LightOblique /Avenir-Medium /Avenir-MediumOblique /Avenir-Oblique /Avenir-Roman /BaileySansITC-Bold /BaileySansITC-BoldItalic /BaileySansITC-Book /BaileySansITC-BookItalic /BakerSignetBT-Roman /BaskervilleBE-Italic /BaskervilleBE-Medium /BaskervilleBE-MediumItalic /BaskervilleBE-Regular /Baskerville-Bold /BaskervilleBook-Italic /BaskervilleBook-MedItalic /BaskervilleBook-Medium /BaskervilleBook-Regular /BaskervilleBT-Bold /BaskervilleBT-BoldItalic /BaskervilleBT-Italic /BaskervilleBT-Roman /BaskervilleMT /BaskervilleMT-Bold /BaskervilleMT-BoldItalic /BaskervilleMT-Italic /BaskervilleMT-SemiBold /BaskervilleMT-SemiBoldItalic /BaskervilleNo2BT-Bold /BaskervilleNo2BT-BoldItalic /BaskervilleNo2BT-Italic /BaskervilleNo2BT-Roman /Baskerville-Normal-Italic /BauerBodoni-Black /BauerBodoni-BlackCond /BauerBodoni-BlackItalic /BauerBodoni-Bold /BauerBodoni-BoldCond /BauerBodoni-BoldItalic /BauerBodoni-BoldItalicOsF /BauerBodoni-BoldOsF /BauerBodoni-Italic /BauerBodoni-ItalicOsF /BauerBodoni-Roman /BauerBodoni-RomanSC /Bauhaus-Bold /Bauhaus-Demi /Bauhaus-Heavy /BauhausITCbyBT-Bold /BauhausITCbyBT-Heavy /BauhausITCbyBT-Light /BauhausITCbyBT-Medium /Bauhaus-Light /Bauhaus-Medium /BellCentennial-Address /BellGothic-Black /BellGothic-Bold /Bell-GothicBoldItalicBT /BellGothicBT-Bold /BellGothicBT-Roman /BellGothic-Light /Bembo /Bembo-Bold /Bembo-BoldExpert /Bembo-BoldItalic /Bembo-BoldItalicExpert /Bembo-Expert /Bembo-ExtraBoldItalic /Bembo-Italic /Bembo-ItalicExpert /Bembo-Semibold /Bembo-SemiboldItalic /Benguiat-Bold /Benguiat-BoldItalic /Benguiat-Book /Benguiat-BookItalic /BenguiatGothicITCbyBT-Bold /BenguiatGothicITCbyBT-BoldItal /BenguiatGothicITCbyBT-Book /BenguiatGothicITCbyBT-BookItal /BenguiatITCbyBT-Bold /BenguiatITCbyBT-BoldItalic /BenguiatITCbyBT-Book /BenguiatITCbyBT-BookItalic /Benguiat-Medium /Benguiat-MediumItalic /Berkeley-Black /Berkeley-BlackItalic /Berkeley-Bold /Berkeley-BoldItalic /Berkeley-Book /Berkeley-BookItalic /Berkeley-Italic /Berkeley-Medium /Berling-Bold /Berling-BoldItalic /Berling-Italic /Berling-Roman /BernhardBoldCondensedBT-Regular /BernhardFashionBT-Regular /BernhardModernBT-Bold /BernhardModernBT-BoldItalic /BernhardModernBT-Italic /BernhardModernBT-Roman /BernhardTangoBT-Regular /BlockBE-Condensed /BlockBE-ExtraCn /BlockBE-ExtraCnIt /BlockBE-Heavy /BlockBE-Italic /BlockBE-Regular /Bodoni /Bodoni-Bold /Bodoni-BoldItalic /Bodoni-Italic /Bodoni-Poster /Bodoni-PosterCompressed /Bookman-Demi /Bookman-DemiItalic /Bookman-Light /Bookman-LightItalic /Boton-Italic /Boton-Medium /Boton-MediumItalic /Boton-Regular /Boulevard /BremenBT-Black /BremenBT-Bold /BroadwayBT-Regular /CaflischScript-Bold /CaflischScript-Regular /Caliban /CarminaBT-Bold /CarminaBT-BoldItalic /CarminaBT-Light /CarminaBT-LightItalic /CarminaBT-Medium /CarminaBT-MediumItalic /Carta /Caslon224ITCbyBT-Bold /Caslon224ITCbyBT-BoldItalic /Caslon224ITCbyBT-Book /Caslon224ITCbyBT-BookItalic /Caslon540BT-Italic /Caslon540BT-Roman /CaslonBT-Bold /CaslonBT-BoldItalic /CaslonOpenFace /CaslonTwoTwentyFour-Black /CaslonTwoTwentyFour-BlackIt /CaslonTwoTwentyFour-Bold /CaslonTwoTwentyFour-BoldIt /CaslonTwoTwentyFour-Book /CaslonTwoTwentyFour-BookIt /CaslonTwoTwentyFour-Medium /CaslonTwoTwentyFour-MediumIt /CastleT-Bold /CastleT-Book /Caxton-Bold /Caxton-BoldItalic /Caxton-Book /Caxton-BookItalic /CaxtonBT-Bold /CaxtonBT-BoldItalic /CaxtonBT-Book /CaxtonBT-BookItalic /Caxton-Light /Caxton-LightItalic /CelestiaAntiqua-Ornaments /Centennial-BlackItalicOsF /Centennial-BlackOsF /Centennial-BoldItalicOsF /Centennial-BoldOsF /Centennial-ItalicOsF /Centennial-LightItalicOsF /Centennial-LightSC /Centennial-RomanSC /Century-Bold /Century-BoldItalic /Century-Book /Century-BookItalic /CenturyExpandedBT-Bold /CenturyExpandedBT-BoldItalic /CenturyExpandedBT-Italic /CenturyExpandedBT-Roman /Century-HandtooledBold /Century-HandtooledBoldItalic /Century-Light /Century-LightItalic /CenturyOldStyle-Bold /CenturyOldStyle-Italic /CenturyOldStyle-Regular /CenturySchoolbookBT-Bold /CenturySchoolbookBT-BoldCond /CenturySchoolbookBT-BoldItalic /CenturySchoolbookBT-Italic /CenturySchoolbookBT-Roman /Century-Ultra /Century-UltraItalic /CharterBT-Black /CharterBT-BlackItalic /CharterBT-Bold /CharterBT-BoldItalic /CharterBT-Italic /CharterBT-Roman /CheltenhamBT-Bold /CheltenhamBT-BoldCondItalic /CheltenhamBT-BoldExtraCondensed /CheltenhamBT-BoldHeadline /CheltenhamBT-BoldItalic /CheltenhamBT-BoldItalicHeadline /CheltenhamBT-Italic /CheltenhamBT-Roman /Cheltenham-HandtooledBdIt /Cheltenham-HandtooledBold /CheltenhamITCbyBT-Bold /CheltenhamITCbyBT-BoldItalic /CheltenhamITCbyBT-Book /CheltenhamITCbyBT-BookItalic /Christiana-Bold /Christiana-BoldItalic /Christiana-Italic /Christiana-Medium /Christiana-MediumItalic /Christiana-Regular /Christiana-RegularExpert /Christiana-RegularSC /Clarendon /Clarendon-Bold /Clarendon-Light /ClassicalGaramondBT-Bold /ClassicalGaramondBT-BoldItalic /ClassicalGaramondBT-Italic /ClassicalGaramondBT-Roman /CMR10 /CMR8 /CMSY10 /CMSY8 /CMTI10 /CommonBullets /ConduitITC-Bold /ConduitITC-BoldItalic /ConduitITC-Light /ConduitITC-LightItalic /ConduitITC-Medium /ConduitITC-MediumItalic /CooperBlack /CooperBlack-Italic /CooperBT-Bold /CooperBT-BoldItalic /CooperBT-Light /CooperBT-LightItalic /CopperplateGothicBT-Bold /CopperplateGothicBT-BoldCond /CopperplateGothicBT-Heavy /CopperplateGothicBT-Roman /CopperplateGothicBT-RomanCond /Copperplate-ThirtyThreeBC /Copperplate-ThirtyTwoBC /Coronet-Regular /Courier /Courier-Bold /Courier-BoldOblique /Courier-Oblique /Critter /CS-Special-font /DellaRobbiaBT-Bold /DellaRobbiaBT-Roman /Della-RobbiaItalicBT /Della-RobbiaSCaps /Del-NormalSmallCaps /Delphin-IA /Delphin-IIA /Delta-Bold /Delta-BoldItalic /Delta-Book /Delta-BookItalic /Delta-Light /Delta-LightItalic /Delta-Medium /Delta-MediumItalic /Delta-Outline /DextorD /DextorOutD /DidotLH-OrnamentsOne /DidotLH-OrnamentsTwo /DINEngschrift /DINEngschrift-Alternate /DINMittelschrift /DINMittelschrift-Alternate /DINNeuzeitGrotesk-BoldCond /DINNeuzeitGrotesk-Light /Dom-CasItalic /DomCasual /DomCasual-Bold /Dom-CasualBT /Ehrhard-Italic /Ehrhard-Regular /EhrhardSemi-Italic /EhrhardtMT /EhrhardtMT-Italic /EhrhardtMT-SemiBold /EhrhardtMT-SemiBoldItalic /EhrharSemi /ELANGO-IB-A03 /ELANGO-IB-A75 /ELANGO-IB-A99 /ElectraLH-Bold /ElectraLH-BoldCursive /ElectraLH-Cursive /ElectraLH-Regular /ElGreco /EnglischeSchT-Bold /EnglischeSchT-Regu /ErasContour /ErasITCbyBT-Bold /ErasITCbyBT-Book /ErasITCbyBT-Demi /ErasITCbyBT-Light /ErasITCbyBT-Medium /ErasITCbyBT-Ultra /Euclid /Euclid-Bold /Euclid-BoldItalic /EuclidExtra /EuclidExtra-Bold /EuclidFraktur /EuclidFraktur-Bold /Euclid-Italic /EuclidMathOne /EuclidMathOne-Bold /EuclidMathTwo /EuclidMathTwo-Bold /EuclidSymbol /EuclidSymbol-Bold /EuclidSymbol-BoldItalic /EuclidSymbol-Italic /EUEX10 /EUFB10 /EUFB5 /EUFB7 /EUFM10 /EUFM5 /EUFM7 /EURB10 /EURB5 /EURB7 /EURM10 /EURM5 /EURM7 /EuropeanPi-Four /EuropeanPi-One /EuropeanPi-Three /EuropeanPi-Two /EuroSans-Bold /EuroSans-BoldItalic /EuroSans-Italic /EuroSans-Regular /EuroSerif-Bold /EuroSerif-BoldItalic /EuroSerif-Italic /EuroSerif-Regular /Eurostile /Eurostile-Bold /Eurostile-BoldCondensed /Eurostile-BoldExtendedTwo /Eurostile-BoldOblique /Eurostile-Condensed /Eurostile-Demi /Eurostile-DemiOblique /Eurostile-ExtendedTwo /EurostileLTStd-Demi /EurostileLTStd-DemiOblique /Eurostile-Oblique /EUSB10 /EUSB5 /EUSB7 /EUSM10 /EUSM5 /EUSM7 /ExPonto-Regular /FairfieldLH-Bold /FairfieldLH-BoldItalic /FairfieldLH-BoldSC /FairfieldLH-CaptionBold /FairfieldLH-CaptionHeavy /FairfieldLH-CaptionLight /FairfieldLH-CaptionMedium /FairfieldLH-Heavy /FairfieldLH-HeavyItalic /FairfieldLH-HeavySC /FairfieldLH-Light /FairfieldLH-LightItalic /FairfieldLH-LightSC /FairfieldLH-Medium /FairfieldLH-MediumItalic /FairfieldLH-MediumSC /FairfieldLH-SwBoldItalicOsF /FairfieldLH-SwHeavyItalicOsF /FairfieldLH-SwLightItalicOsF /FairfieldLH-SwMediumItalicOsF /Fences /Fenice-Bold /Fenice-BoldOblique /FeniceITCbyBT-Bold /FeniceITCbyBT-BoldItalic /FeniceITCbyBT-Regular /FeniceITCbyBT-RegularItalic /Fenice-Light /Fenice-LightOblique /Fenice-Regular /Fenice-RegularOblique /Fenice-Ultra /Fenice-UltraOblique /FlashD-Ligh /Flood /Folio-Bold /Folio-BoldCondensed /Folio-ExtraBold /Folio-Light /Folio-Medium /FontanaNDAaOsF /FontanaNDAaOsF-Italic /FontanaNDCcOsF-Semibold /FontanaNDCcOsF-SemiboldIta /FontanaNDEeOsF /FontanaNDEeOsF-Bold /FontanaNDEeOsF-BoldItalic /FontanaNDEeOsF-Light /FontanaNDEeOsF-Semibold /FormalScript421BT-Regular /Formata-Bold /Formata-MediumCondensed /ForteMT /FournierMT-Ornaments /FrakturBT-Regular /FrankfurterHigD /FranklinGothic-Book /FranklinGothic-BookItal /FranklinGothic-BookOblique /FranklinGothic-Condensed /FranklinGothic-Demi /FranklinGothic-DemiItal /FranklinGothic-DemiOblique /FranklinGothic-Heavy /FranklinGothic-HeavyItal /FranklinGothic-HeavyOblique /FranklinGothicITCbyBT-BookItal /FranklinGothicITCbyBT-Demi /FranklinGothicITCbyBT-DemiItal /FranklinGothicITCbyBT-Heavy /FranklinGothicITCbyBT-HeavyItal /FranklinGothic-Medium /FranklinGothic-MediumItal /FranklinGothic-Roman /Freeform721BT-Bold /Freeform721BT-BoldItalic /Freeform721BT-Italic /Freeform721BT-Roman /FreestyleScrD /FreestyleScript /Freestylescript /FrizQuadrataITCbyBT-Bold /FrizQuadrataITCbyBT-Roman /Frutiger-Black /Frutiger-BlackCn /Frutiger-BlackItalic /Frutiger-Bold /Frutiger-BoldCn /Frutiger-BoldItalic /Frutiger-Cn /Frutiger-ExtraBlackCn /Frutiger-Italic /Frutiger-Light /Frutiger-LightCn /Frutiger-LightItalic /Frutiger-Roman /Frutiger-UltraBlack /Futura /FuturaBlackBT-Regular /Futura-Bold /Futura-BoldOblique /Futura-Book /Futura-BookOblique /FuturaBT-Bold /FuturaBT-BoldCondensed /FuturaBT-BoldCondensedItalic /FuturaBT-BoldItalic /FuturaBT-Book /FuturaBT-BookItalic /FuturaBT-ExtraBlack /FuturaBT-ExtraBlackCondensed /FuturaBT-ExtraBlackCondItalic /FuturaBT-ExtraBlackItalic /FuturaBT-Heavy /FuturaBT-HeavyItalic /FuturaBT-Light /FuturaBT-LightCondensed /FuturaBT-LightItalic /FuturaBT-Medium /FuturaBT-MediumCondensed /FuturaBT-MediumItalic /Futura-CondensedLight /Futura-CondensedLightOblique /Futura-ExtraBold /Futura-ExtraBoldOblique /Futura-Heavy /Futura-HeavyOblique /Futura-Light /Futura-LightOblique /Futura-Oblique /Futura-Thin /Galliard-Black /Galliard-BlackItalic /Galliard-Bold /Galliard-BoldItalic /Galliard-Italic /GalliardITCbyBT-Bold /GalliardITCbyBT-BoldItalic /GalliardITCbyBT-Italic /GalliardITCbyBT-Roman /Galliard-Roman /Galliard-Ultra /Galliard-UltraItalic /Garamond-Antiqua /GaramondBE-Bold /GaramondBE-BoldExpert /GaramondBE-BoldOsF /GaramondBE-CnExpert /GaramondBE-Condensed /GaramondBE-CondensedSC /GaramondBE-Italic /GaramondBE-ItalicExpert /GaramondBE-ItalicOsF /GaramondBE-Medium /GaramondBE-MediumCn /GaramondBE-MediumCnExpert /GaramondBE-MediumCnOsF /GaramondBE-MediumExpert /GaramondBE-MediumItalic /GaramondBE-MediumItalicExpert /GaramondBE-MediumItalicOsF /GaramondBE-MediumSC /GaramondBE-Regular /GaramondBE-RegularExpert /GaramondBE-RegularSC /GaramondBE-SwashItalic /Garamond-Bold /Garamond-BoldCondensed /Garamond-BoldCondensedItalic /Garamond-BoldItalic /Garamond-Book /Garamond-BookCondensed /Garamond-BookCondensedItalic /Garamond-BookItalic /Garamond-Halbfett /Garamond-HandtooledBold /Garamond-HandtooledBoldItalic /GaramondITCbyBT-Bold /GaramondITCbyBT-BoldCondensed /GaramondITCbyBT-BoldCondItalic /GaramondITCbyBT-BoldItalic /GaramondITCbyBT-BoldNarrow /GaramondITCbyBT-BoldNarrowItal /GaramondITCbyBT-Book /GaramondITCbyBT-BookCondensed /GaramondITCbyBT-BookCondItalic /GaramondITCbyBT-BookItalic /GaramondITCbyBT-BookNarrow /GaramondITCbyBT-BookNarrowItal /GaramondITCbyBT-Light /GaramondITCbyBT-LightCondensed /GaramondITCbyBT-LightCondItalic /GaramondITCbyBT-LightItalic /GaramondITCbyBT-LightNarrow /GaramondITCbyBT-LightNarrowItal /GaramondITCbyBT-Ultra /GaramondITCbyBT-UltraCondensed /GaramondITCbyBT-UltraCondItalic /GaramondITCbyBT-UltraItalic /Garamond-Kursiv /Garamond-KursivHalbfett /Garamond-Light /Garamond-LightCondensed /Garamond-LightCondensedItalic /Garamond-LightItalic /GaramondNo4CyrTCY-Ligh /GaramondNo4CyrTCY-LighItal /GaramondThree /GaramondThree-Bold /GaramondThree-BoldItalic /GaramondThree-BoldItalicOsF /GaramondThree-BoldSC /GaramondThree-Italic /GaramondThree-ItalicOsF /GaramondThree-SC /GaramondThreeSMSIISpl-Italic /GaramondThreeSMSitalicSpl-Italic /GaramondThreeSMSspl /GaramondThreespl /GaramondThreeSpl-Bold /GaramondThreeSpl-Italic /Garamond-Ultra /Garamond-UltraCondensed /Garamond-UltraCondensedItalic /Garamond-UltraItalic /GarthGraphic /GarthGraphic-Black /GarthGraphic-Bold /GarthGraphic-BoldCondensed /GarthGraphic-BoldItalic /GarthGraphic-Condensed /GarthGraphic-ExtraBold /GarthGraphic-Italic /Geometric231BT-HeavyC /GeometricSlab712BT-BoldA /GeometricSlab712BT-ExtraBoldA /GeometricSlab712BT-LightA /GeometricSlab712BT-LightItalicA /GeometricSlab712BT-MediumA /GeometricSlab712BT-MediumItalA /Giddyup /Giddyup-Thangs /GillSans /GillSans-Bold /GillSans-BoldCondensed /GillSans-BoldExtraCondensed /GillSans-BoldItalic /GillSans-Condensed /GillSans-ExtraBold /GillSans-ExtraBoldDisplay /GillSans-Italic /GillSans-Light /GillSans-LightItalic /GillSans-LightShadowed /GillSans-Shadowed /GillSans-UltraBold /GillSans-UltraBoldCondensed /Gill-Special /Giovanni-Bold /Giovanni-BoldItalic /Giovanni-Book /Giovanni-BookItalic /Glypha /Glypha-Bold /Glypha-BoldOblique /Glypha-Oblique /Gothic-Thirteen /Goudy /Goudy-Bold /Goudy-BoldItalic /GoudyCatalogueBT-Regular /Goudy-ExtraBold /GoudyHandtooledBT-Regular /GoudyHeavyfaceBT-Regular /GoudyHeavyfaceBT-RegularCond /Goudy-Italic /GoudyOldStyleBT-Bold /GoudyOldStyleBT-BoldItalic /GoudyOldStyleBT-ExtraBold /GoudyOldStyleBT-Italic /GoudyOldStyleBT-Roman /GoudySans-Black /GoudySans-BlackItalic /GoudySans-Bold /GoudySans-BoldItalic /GoudySans-Book /GoudySans-BookItalic /GoudySansITCbyBT-Black /GoudySansITCbyBT-BlackItalic /GoudySansITCbyBT-Bold /GoudySansITCbyBT-BoldItalic /GoudySansITCbyBT-Light /GoudySansITCbyBT-LightItalic /GoudySansITCbyBT-Medium /GoudySansITCbyBT-MediumItalic /GoudySans-Medium /GoudySans-MediumItalic /Granjon /Granjon-Bold /Granjon-BoldOsF /Granjon-Italic /Granjon-ItalicOsF /Granjon-SC /GreymantleMVB-Ornaments /Helvetica /Helvetica-Black /Helvetica-BlackOblique /Helvetica-Black-SemiBold /Helvetica-Bold /Helvetica-BoldOblique /Helvetica-Compressed /Helvetica-Condensed /Helvetica-Condensed-Black /Helvetica-Condensed-BlackObl /Helvetica-Condensed-Bold /Helvetica-Condensed-BoldObl /Helvetica-Condensed-Light /Helvetica-Condensed-Light-Light /Helvetica-Condensed-LightObl /Helvetica-Condensed-Oblique /Helvetica-Condensed-Thin /Helvetica-ExtraCompressed /Helvetica-Fraction /Helvetica-FractionBold /HelveticaInserat-Roman /HelveticaInserat-Roman-SemiBold /Helvetica-Light /Helvetica-LightOblique /Helvetica-Narrow /Helvetica-Narrow-Bold /Helvetica-Narrow-BoldOblique /Helvetica-Narrow-Oblique /HelveticaNeue-Black /HelveticaNeue-BlackCond /HelveticaNeue-BlackCondObl /HelveticaNeue-BlackExt /HelveticaNeue-BlackExtObl /HelveticaNeue-BlackItalic /HelveticaNeue-Bold /HelveticaNeue-BoldCond /HelveticaNeue-BoldCondObl /HelveticaNeue-BoldExt /HelveticaNeue-BoldExtObl /HelveticaNeue-BoldItalic /HelveticaNeue-Condensed /HelveticaNeue-CondensedObl /HelveticaNeue-ExtBlackCond /HelveticaNeue-ExtBlackCondObl /HelveticaNeue-Extended /HelveticaNeue-ExtendedObl /HelveticaNeue-Heavy /HelveticaNeue-HeavyCond /HelveticaNeue-HeavyCondObl /HelveticaNeue-HeavyExt /HelveticaNeue-HeavyExtObl /HelveticaNeue-HeavyItalic /HelveticaNeue-Italic /HelveticaNeue-Light /HelveticaNeue-LightCond /HelveticaNeue-LightCondObl /HelveticaNeue-LightExt /HelveticaNeue-LightExtObl /HelveticaNeue-LightItalic /HelveticaNeueLTStd-Md /HelveticaNeueLTStd-MdIt /HelveticaNeue-Medium /HelveticaNeue-MediumCond /HelveticaNeue-MediumCondObl /HelveticaNeue-MediumExt /HelveticaNeue-MediumExtObl /HelveticaNeue-MediumItalic /HelveticaNeue-Roman /HelveticaNeue-Thin /HelveticaNeue-ThinCond /HelveticaNeue-ThinCondObl /HelveticaNeue-ThinItalic /HelveticaNeue-UltraLigCond /HelveticaNeue-UltraLigCondObl /HelveticaNeue-UltraLigExt /HelveticaNeue-UltraLigExtObl /HelveticaNeue-UltraLight /HelveticaNeue-UltraLightItal /Helvetica-Oblique /Helvetica-UltraCompressed /HelvExtCompressed /HelvLight /HelvUltCompressed /Humanist521BT-Bold /Humanist521BT-BoldCondensed /Humanist521BT-BoldItalic /Humanist521BT-ExtraBold /Humanist521BT-Italic /Humanist521BT-Light /Humanist521BT-LightItalic /Humanist521BT-Roman /Humanist521BT-RomanCondensed /Humanist521BT-UltraBold /Humanist521BT-XtraBoldCondensed /Humanist531BT-BlackA /Humanist531BT-BoldA /Humanist531BT-RomanA /Humanist531BT-UltraBlackA /Humanist777BT-BlackB /Humanist777BT-BlackCondensedB /Humanist777BT-BlackItalicB /Humanist777BT-BoldB /Humanist777BT-BoldCondensedB /Humanist777BT-BoldItalicB /Humanist777BT-ExtraBlackB /Humanist777BT-ExtraBlackCondB /Humanist777BT-ItalicB /Humanist777BT-LightB /Humanist777BT-LightCondensedB /Humanist777BT-LightItalicB /Humanist777BT-RomanB /Humanist777BT-RomanCondensedB /Humanist970BT-BoldC /Humanist970BT-RomanC /HumanistSlabserif712BT-Black /HumanistSlabserif712BT-Bold /HumanistSlabserif712BT-Italic /HumanistSlabserif712BT-Roman /ICMEX10 /ICMMI8 /ICMSY8 /ICMTT8 /Iglesia-Light /ILASY8 /ILCMSS8 /ILCMSSB8 /ILCMSSI8 /Imago-Book /Imago-BookItalic /Imago-ExtraBold /Imago-ExtraBoldItalic /Imago-Light /Imago-LightItalic /Imago-Medium /Imago-MediumItalic /Industria-Inline /Industria-InlineA /Industria-Solid /Industria-SolidA /Insignia /Insignia-A /IPAExtras /IPAHighLow /IPAKiel /IPAKielSeven /IPAsans /ITCGaramondMM /ITCGaramondMM-It /JAKEOpti-Regular /JansonText-Bold /JansonText-BoldItalic /JansonText-Italic /JansonText-Roman /JansonText-RomanSC /JoannaMT /JoannaMT-Bold /JoannaMT-BoldItalic /JoannaMT-Italic /Juniper /KabelITCbyBT-Book /KabelITCbyBT-Demi /KabelITCbyBT-Medium /KabelITCbyBT-Ultra /Kaufmann /Kaufmann-Bold /KeplMM-Or2 /KisBT-Italic /KisBT-Roman /KlangMT /Kuenstler480BT-Black /Kuenstler480BT-Bold /Kuenstler480BT-BoldItalic /Kuenstler480BT-Italic /Kuenstler480BT-Roman /KunstlerschreibschD-Bold /KunstlerschreibschD-Medi /Lapidary333BT-Black /Lapidary333BT-Bold /Lapidary333BT-BoldItalic /Lapidary333BT-Italic /Lapidary333BT-Roman /LASY10 /LASY5 /LASY6 /LASY7 /LASY8 /LASY9 /LASYB10 /LatinMT-Condensed /LCIRCLE10 /LCIRCLEW10 /LCMSS8 /LCMSSB8 /LCMSSI8 /LDecorationPi-One /LDecorationPi-Two /Leawood-Black /Leawood-BlackItalic /Leawood-Bold /Leawood-BoldItalic /Leawood-Book /Leawood-BookItalic /Leawood-Medium /Leawood-MediumItalic /LegacySans-Bold /LegacySans-BoldItalic /LegacySans-Book /LegacySans-BookItalic /LegacySans-Medium /LegacySans-MediumItalic /LegacySans-Ultra /LegacySerif-Bold /LegacySerif-BoldItalic /LegacySerif-Book /LegacySerif-BookItalic /LegacySerif-Medium /LegacySerif-MediumItalic /LegacySerif-Ultra /LetterGothic /LetterGothic-Bold /LetterGothic-BoldSlanted /LetterGothic-Slanted /Life-Bold /Life-Italic /Life-Roman /LINE10 /LINEW10 /Linotext /Lithos-Black /LithosBold /Lithos-Bold /Lithos-Regular /LOGO10 /LOGO8 /LOGO9 /LOGOBF10 /LOGOSL10 /LOMD-Normal /LubalinGraph-Book /LubalinGraph-BookOblique /LubalinGraph-Demi /LubalinGraph-DemiOblique /LucidaHandwritingItalic /LucidaMath-Symbol /LucidaSansTypewriter /LucidaSansTypewriter-Bd /LucidaSansTypewriter-BdObl /LucidaSansTypewriter-Obl /LucidaTypewriter /LucidaTypewriter-Bold /LucidaTypewriter-BoldObl /LucidaTypewriter-Obl /LydianBT-Bold /LydianBT-BoldItalic /LydianBT-Italic /LydianBT-Roman /LydianCursiveBT-Regular /Machine /Machine-Bold /Marigold /MathematicalPi-Five /MathematicalPi-Four /MathematicalPi-One /MathematicalPi-Six /MathematicalPi-Three /MathematicalPi-Two /MatrixScriptBold /MatrixScriptBoldLin /MatrixScriptBook /MatrixScriptBookLin /MatrixScriptRegular /MatrixScriptRegularLin /Melior /Melior-Bold /Melior-BoldItalic /Melior-Italic /MercuriusCT-Black /MercuriusCT-BlackItalic /MercuriusCT-Light /MercuriusCT-LightItalic /MercuriusCT-Medium /MercuriusCT-MediumItalic /MercuriusMT-BoldScript /Meridien-Bold /Meridien-BoldItalic /Meridien-Italic /Meridien-Medium /Meridien-MediumItalic /Meridien-Roman /Minion-Black /Minion-Bold /Minion-BoldCondensed /Minion-BoldCondensedItalic /Minion-BoldItalic /Minion-Condensed /Minion-CondensedItalic /Minion-DisplayItalic /Minion-DisplayRegular /MinionExp-Italic /MinionExp-Semibold /MinionExp-SemiboldItalic /Minion-Italic /Minion-Ornaments /Minion-Regular /Minion-Semibold /Minion-SemiboldItalic /MonaLisa-Recut /MrsEavesAllPetiteCaps /MrsEavesAllSmallCaps /MrsEavesBold /MrsEavesFractions /MrsEavesItalic /MrsEavesPetiteCaps /MrsEavesRoman /MrsEavesRomanLining /MrsEavesSmallCaps /MSAM10 /MSAM10A /MSAM5 /MSAM6 /MSAM7 /MSAM8 /MSAM9 /MSBM10 /MSBM10A /MSBM5 /MSBM6 /MSBM7 /MSBM8 /MSBM9 /MTEX /MTEXB /MTEXH /MTGU /MTGUB /MTMI /MTMIB /MTMIH /MTMS /MTMSB /MTMUB /MTMUH /MTSY /MTSYB /MTSYH /MTSYN /MusicalSymbols-Normal /Myriad-Bold /Myriad-BoldItalic /Myriad-CnBold /Myriad-CnBoldItalic /Myriad-CnItalic /Myriad-CnSemibold /Myriad-CnSemiboldItalic /Myriad-Condensed /Myriad-Italic /MyriadMM /MyriadMM-It /Myriad-Roman /Myriad-Sketch /Myriad-Tilt /NeuzeitS-Book /NeuzeitS-BookHeavy /NewBaskerville-Bold /NewBaskerville-BoldItalic /NewBaskerville-Italic /NewBaskervilleITCbyBT-Bold /NewBaskervilleITCbyBT-BoldItal /NewBaskervilleITCbyBT-Italic /NewBaskervilleITCbyBT-Roman /NewBaskerville-Roman /NewCaledonia /NewCaledonia-Black /NewCaledonia-BlackItalic /NewCaledonia-Bold /NewCaledonia-BoldItalic /NewCaledonia-BoldItalicOsF /NewCaledonia-BoldSC /NewCaledonia-Italic /NewCaledonia-ItalicOsF /NewCaledonia-SC /NewCaledonia-SemiBold /NewCaledonia-SemiBoldItalic /NewCenturySchlbk-Bold /NewCenturySchlbk-BoldItalic /NewCenturySchlbk-Italic /NewCenturySchlbk-Roman /NewsGothic /NewsGothic-Bold /NewsGothic-BoldOblique /NewsGothicBT-Bold /NewsGothicBT-BoldCondensed /NewsGothicBT-BoldCondItalic /NewsGothicBT-BoldExtraCondensed /NewsGothicBT-BoldItalic /NewsGothicBT-Demi /NewsGothicBT-DemiItalic /NewsGothicBT-ExtraCondensed /NewsGothicBT-Italic /NewsGothicBT-ItalicCondensed /NewsGothicBT-Light /NewsGothicBT-LightItalic /NewsGothicBT-Roman /NewsGothicBT-RomanCondensed /NewsGothic-Oblique /New-Symbol /NovareseITCbyBT-Bold /NovareseITCbyBT-BoldItalic /NovareseITCbyBT-Book /NovareseITCbyBT-BookItalic /Nueva-BoldExtended /Nueva-Roman /NuptialScript /OceanSansMM /OceanSansMM-It /OfficinaSans-Bold /OfficinaSans-BoldItalic /OfficinaSans-Book /OfficinaSans-BookItalic /OfficinaSerif-Bold /OfficinaSerif-BoldItalic /OfficinaSerif-Book /OfficinaSerif-BookItalic /OnyxMT /Optima /Optima-Bold /Optima-BoldItalic /Optima-BoldOblique /Optima-ExtraBlack /Optima-ExtraBlackItalic /Optima-Italic /Optima-Oblique /OSPIRE-Plain /OttaIA /Otta-wa /Ottawa-BoldA /OttawaPSMT /Oxford /Palatino-Bold /Palatino-BoldItalic /Palatino-Italic /Palatino-Roman /Parisian /Perpetua /Perpetua-Bold /Perpetua-BoldItalic /Perpetua-Italic /PhotinaMT /PhotinaMT-Bold /PhotinaMT-BoldItalic /PhotinaMT-Italic /PhotinaMT-SemiBold /PhotinaMT-SemiBoldItalic /PhotinaMT-UltraBold /PhotinaMT-UltraBoldItalic /Plantin /Plantin-Bold /Plantin-BoldItalic /Plantin-Italic /Plantin-Light /Plantin-LightItalic /Plantin-Semibold /Plantin-SemiboldItalic /Poetica-ChanceryI /Poetica-SuppLowercaseEndI /PopplLaudatio-Italic /PopplLaudatio-Medium /PopplLaudatio-MediumItalic /PopplLaudatio-Regular /ProseAntique-Bold /ProseAntique-Normal /QuaySansEF-Black /QuaySansEF-BlackItalic /QuaySansEF-Book /QuaySansEF-BookItalic /QuaySansEF-Medium /QuaySansEF-MediumItalic /Quorum-Black /Quorum-Bold /Quorum-Book /Quorum-Light /Quorum-Medium /Raleigh /Raleigh-Bold /Raleigh-DemiBold /Raleigh-Medium /Revival565BT-Bold /Revival565BT-BoldItalic /Revival565BT-Italic /Revival565BT-Roman /Ribbon131BT-Bold /Ribbon131BT-Regular /RMTMI /Rockwell /Rockwell-Bold /Rockwell-BoldItalic /Rockwell-Italic /Rockwell-Light /Rockwell-LightItalic /RotisSansSerif /RotisSansSerif-Bold /RotisSansSerif-ExtraBold /RotisSansSerif-Italic /RotisSansSerif-Light /RotisSansSerif-LightItalic /RotisSemiSans /RotisSemiSans-Bold /RotisSemiSans-ExtraBold /RotisSemiSans-Italic /RotisSemiSans-Light /RotisSemiSans-LightItalic /RotisSemiSerif /RotisSemiSerif-Bold /RotisSerif /RotisSerif-Bold /RotisSerif-Italic /RunicMT-Condensed /Sabon-Bold /Sabon-BoldItalic /Sabon-Italic /Sabon-Roman /SackersGothicLight /SackersGothicLightAlt /SackersItalianScript /SackersItalianScriptAlt /Sam /Sanvito-Light /SanvitoMM /Sanvito-Roman /Semitica /Semitica-Italic /SIVAMATH /Siva-Special /SMS-SPELA /Souvenir-Demi /Souvenir-DemiItalic /SouvenirITCbyBT-Demi /SouvenirITCbyBT-DemiItalic /SouvenirITCbyBT-Light /SouvenirITCbyBT-LightItalic /Souvenir-Light /Souvenir-LightItalic /SpecialAA /Special-Gali /Sp-Sym /StempelGaramond-Bold /StempelGaramond-BoldItalic /StempelGaramond-Italic /StempelGaramond-Roman /StoneSans /StoneSans-Bold /StoneSans-BoldItalic /StoneSans-Italic /StoneSans-PhoneticAlternate /StoneSans-PhoneticIPA /StoneSans-Semibold /StoneSans-SemiboldItalic /StoneSerif /StoneSerif-Italic /StoneSerif-PhoneticAlternate /StoneSerif-PhoneticIPA /StoneSerif-Semibold /StoneSerif-SemiboldItalic /Swiss721BT-Black /Swiss721BT-BlackCondensed /Swiss721BT-BlackCondensedItalic /Swiss721BT-BlackExtended /Swiss721BT-BlackItalic /Swiss721BT-BlackOutline /Swiss721BT-BlackRounded /Swiss721BT-Bold /Swiss721BT-BoldCondensed /Swiss721BT-BoldCondensedItalic /Swiss721BT-BoldCondensedOutline /Swiss721BT-BoldExtended /Swiss721BT-BoldItalic /Swiss721BT-BoldOutline /Swiss721BT-BoldRounded /Swiss721BT-Heavy /Swiss721BT-HeavyItalic /Swiss721BT-Italic /Swiss721BT-ItalicCondensed /Swiss721BT-Light /Swiss721BT-LightCondensed /Swiss721BT-LightCondensedItalic /Swiss721BT-LightExtended /Swiss721BT-LightItalic /Swiss721BT-Medium /Swiss721BT-MediumItalic /Swiss721BT-Roman /Swiss721BT-RomanCondensed /Swiss721BT-RomanExtended /Swiss721BT-Thin /Swiss721BT-ThinItalic /Swiss921BT-RegularA /Symbol /Syntax-Black /Syntax-Bold /Syntax-Italic /Syntax-Roman /Syntax-UltraBlack /Tekton /Times-Bold /Times-BoldA /Times-BoldItalic /Times-BoldOblique /Times-Italic /Times-NewRoman /Times-NewRomanBold /TimesNewRomanPS-BoldItalicMT /TimesNewRomanPS-ItalicMT /TimesNewRomanPSMT /Times-Oblique /Times-PhoneticAlternate /Times-PhoneticIPA /Times-Roman /Times-RomanSmallCaps /Times-Sc /Times-SCB /Times-special /TimesTenGreekP-Upright /TradeGothic /TradeGothic-Bold /TradeGothic-BoldCondTwenty /TradeGothic-BoldCondTwentyObl /TradeGothic-BoldOblique /TradeGothic-BoldTwo /TradeGothic-BoldTwoOblique /TradeGothic-CondEighteen /TradeGothic-CondEighteenObl /TradeGothicLH-BoldExtended /TradeGothicLH-Extended /TradeGothic-Light /TradeGothic-LightOblique /TradeGothic-Oblique /Trajan-Bold /TrajanPro-Bold /TrajanPro-Regular /Trajan-Regular /Transitional521BT-BoldA /Transitional521BT-CursiveA /Transitional521BT-RomanA /Transitional551BT-MediumB /Transitional551BT-MediumItalicB /Univers /Universal-GreekwithMathPi /Universal-NewswithCommPi /Univers-BlackExt /Univers-BlackExtObl /Univers-Bold /Univers-BoldExt /Univers-BoldExtObl /Univers-BoldOblique /Univers-Condensed /Univers-CondensedBold /Univers-CondensedBoldOblique /Univers-CondensedOblique /Univers-Extended /Univers-ExtendedObl /Univers-ExtraBlackExt /Univers-ExtraBlackExtObl /Univers-Light /Univers-LightOblique /UniversLTStd-Black /UniversLTStd-BlackObl /Univers-Oblique /Utopia-Black /Utopia-BlackOsF /Utopia-Bold /Utopia-BoldItalic /Utopia-Italic /Utopia-Ornaments /Utopia-Regular /Utopia-Semibold /Utopia-SemiboldItalic /VAGRounded-Black /VAGRounded-Bold /VAGRounded-Light /VAGRounded-Thin /Viva-BoldExtraExtended /Viva-Regular /Weidemann-Black /Weidemann-BlackItalic /Weidemann-Bold /Weidemann-BoldItalic /Weidemann-Book /Weidemann-BookItalic /Weidemann-Medium /Weidemann-MediumItalic /WindsorBT-Elongated /WindsorBT-Light /WindsorBT-LightCondensed /WindsorBT-Roman /Wingdings-Regular /WNCYB10 /WNCYI10 /WNCYR10 /WNCYSC10 /WNCYSS10 /WoodtypeOrnaments-One /WoodtypeOrnaments-Two /ZapfCalligraphic801BT-Bold /ZapfCalligraphic801BT-BoldItal /ZapfCalligraphic801BT-Italic /ZapfCalligraphic801BT-Roman /ZapfChanceryITCbyBT-Bold /ZapfChanceryITCbyBT-Demi /ZapfChanceryITCbyBT-Medium /ZapfChanceryITCbyBT-MediumItal /ZapfChancery-MediumItalic /ZapfDingbats /ZapfDingbatsITCbyBT-Regular /ZapfElliptical711BT-Bold /ZapfElliptical711BT-BoldItalic /ZapfElliptical711BT-Italic /ZapfElliptical711BT-Roman /ZapfHumanist601BT-Bold /ZapfHumanist601BT-BoldItalic /ZapfHumanist601BT-Demi /ZapfHumanist601BT-DemiItalic /ZapfHumanist601BT-Italic /ZapfHumanist601BT-Roman /ZapfHumanist601BT-Ultra /ZapfHumanist601BT-UltraItalic /ZurichBT-Black /ZurichBT-BlackExtended /ZurichBT-BlackItalic /ZurichBT-Bold /ZurichBT-BoldCondensed /ZurichBT-BoldCondensedItalic /ZurichBT-BoldExtended /ZurichBT-BoldExtraCondensed /ZurichBT-BoldItalic /ZurichBT-ExtraBlack /ZurichBT-ExtraCondensed /ZurichBT-Italic /ZurichBT-ItalicCondensed /ZurichBT-Light /ZurichBT-LightCondensed /ZurichBT-LightCondensedItalic /ZurichBT-LightExtraCondensed /ZurichBT-LightItalic /ZurichBT-Roman /ZurichBT-RomanCondensed /ZurichBT-RomanExtended /ZurichBT-UltraBlackExtended ] /NeverEmbed [ true ] /AntiAliasColorImages false /DownsampleColorImages true /ColorImageDownsampleType /Bicubic /ColorImageResolution 300 /ColorImageDepth -1 /ColorImageDownsampleThreshold 1.50000 /EncodeColorImages true /ColorImageFilter /DCTEncode /AutoFilterColorImages true /ColorImageAutoFilterStrategy /JPEG /ColorACSImageDict << /QFactor 0.15 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /ColorImageDict << /QFactor 0.15 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /JPEG2000ColorACSImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /JPEG2000ColorImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /AntiAliasGrayImages false /DownsampleGrayImages true /GrayImageDownsampleType /Bicubic /GrayImageResolution 300 /GrayImageDepth -1 /GrayImageDownsampleThreshold 1.50000 /EncodeGrayImages true /GrayImageFilter /DCTEncode /AutoFilterGrayImages true /GrayImageAutoFilterStrategy /JPEG /GrayACSImageDict << /QFactor 0.15 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /GrayImageDict << /QFactor 0.15 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /JPEG2000GrayACSImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /JPEG2000GrayImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /AntiAliasMonoImages false /DownsampleMonoImages true /MonoImageDownsampleType /Bicubic /MonoImageResolution 1200 /MonoImageDepth -1 /MonoImageDownsampleThreshold 1.50000 /EncodeMonoImages true /MonoImageFilter /CCITTFaxEncode /MonoImageDict << /K -1 >> /AllowPSXObjects false /PDFX1aCheck false /PDFX3Check false /PDFXCompliantPDFOnly false /PDFXNoTrimBoxError true /PDFXTrimBoxToMediaBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXSetBleedBoxToMediaBox false /PDFXBleedBoxToTrimBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXOutputIntentProfile (U.S. Web Coated \050SWOP\051 v2) /PDFXOutputCondition () /PDFXRegistryName (http://www.color.org) /PDFXTrapped /Unknown /DetectCurves 0.100000 /EmbedOpenType false /ParseICCProfilesInComments true /PreserveDICMYKValues true /PreserveFlatness true /CropColorImages true /ColorImageMinResolution 150 /ColorImageMinResolutionPolicy /OK /ColorImageMinDownsampleDepth 1 /CropGrayImages true /GrayImageMinResolution 150 /GrayImageMinResolutionPolicy /OK /GrayImageMinDownsampleDepth 2 /CropMonoImages true /MonoImageMinResolution 1200 /MonoImageMinResolutionPolicy /OK /CheckCompliance [ /None ] /PDFXOutputConditionIdentifier () /SyntheticBoldness 1.000000 /Description << /FRA <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> /JPN <FEFF3053306e8a2d5b9a306f30019ad889e350cf5ea6753b50cf3092542b308000200050004400460020658766f830924f5c62103059308b3068304d306b4f7f75283057307e30593002537052376642306e753b8cea3092670059279650306b4fdd306430533068304c3067304d307e305930023053306e8a2d5b9a30674f5c62103057305f00200050004400460020658766f8306f0020004100630072006f0062006100740020304a30883073002000520065006100640065007200200035002e003000204ee5964d30678868793a3067304d307e30593002> /DEU <FEFF00560065007200770065006e00640065006e0020005300690065002000640069006500730065002000450069006e007300740065006c006c0075006e00670065006e0020007a0075006d002000450072007300740065006c006c0065006e00200076006f006e0020005000440046002d0044006f006b0075006d0065006e00740065006e0020006d00690074002000650069006e006500720020006800f60068006500720065006e002000420069006c0064006100750066006c00f600730075006e0067002c00200075006d002000650069006e0065002000760065007200620065007300730065007200740065002000420069006c0064007100750061006c0069007400e400740020007a0075002000650072007a00690065006c0065006e002e00200044006900650020005000440046002d0044006f006b0075006d0065006e007400650020006b00f6006e006e0065006e0020006d006900740020004100630072006f0062006100740020006f0064006500720020006d00690074002000640065006d002000520065006100640065007200200035002e003000200075006e00640020006800f600680065007200200067006500f600660066006e00650074002000770065007200640065006e002e> /PTB <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> /DAN <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> /NLD <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> /ESP <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> /SUO <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> /ITA <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> /NOR <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> /SVE <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> /ENU (Use these settings for creating PDF files for submission to The Sheridan Press. These settings configured for Acrobat v6.0 08/06/03.) >> >> setdistillerparams << /HWResolution [2400 2400] /PageSize [612.000 792.000] >> setpagedevice
12 sources/Kawachi et al.pdf
Crime: social disorganization and relative deprivation
Ichiro Kawachi a, *, Bruce P. Kennedy
b, 1 , Richard G. Wilkinson
c, 2
a Department of Health and Social Behavior, Harvard School of Public Health, 677 Huntington Avenue, Boston, MA 02115, USA
b Department of Health Policy and Management, Harvard School of Public Health, 677 Huntington Avenue, Boston, MA 02115, USA
c Tra�ord Centre for Medical Research, University of Sussex, Brighton BN1 9RY, UK
Abstract
Crime is seldom considered as an outcome in public health research. Yet major theoretical and empirical developments in the ®eld of criminology during the past 50 years suggest that the same social environmental factors
which predict geographic variation in crime rates may also be relevant for explaining community variations in health and wellbeing. Understanding the causes of variability in crime across countries and across regions within a country will help us to solve one of the enduring puzzles in public health, viz. why some communities are healthier than others. The purpose of this paper is to present a conceptual framework for investigating the in¯uence of the
social context on community health, using crime as the indicator of collective wellbeing. We argue that two sets of societal characteristics in¯uence the level of crime: the degree of relative deprivation in society (for instance, measured by the extent of income inequality), and the degree of cohesiveness in social relations among citizens
(measured, for instance, by indicators of `social capital' and `collective e�cacy'). We provided a test of our conceptual framework using state-level ecologic data on violent crimes and property crimes within the USA. Violent crimes (homicide, assault, robbery) were consistently associated with relative deprivation (income inequality) and
indicators of low social capital. Among property crimes, burglary was also associated with deprivation and low social capital. Areas with high crime rates tend also to exhibit higher mortality rates from all causes, suggesting that crime and population health share the same social origins. Crime is thus a mirror of the quality of the social environment. # 1999 Elsevier Science Ltd. All rights reserved.
Keywords: Crime; Income inequality; Social capital; Social disorganization; Relative deprivation; Collective wellbeing; Mortality
1. Introduction
Crime is a social mirror. A long tradition in crimi-
nological research suggests that crime is most prevalent
in societies that permit large disparities in the material
standards of living of its citizens (Hsieh and Pugh,
1993; Kawachi et al., 1994). If the level of crime is an
indicator of the health of society, then the US provides
an illustrative case study as one of the most unhealthy
of modern industrialized nations. The purpose of this
paper is to sketch out a framework for understanding
the role of the social environment in producing healthy
communities, using crime (or the lack of it) as an indi-
cator of collective wellbeing. Crime is seldom con-
sidered as a an outcome in public health research. Yet
as we hope to demonstrate, crime Ð especially violent
crime Ð is a sensitive indicator of social relations in
society. It is an integral aspect of what Sol Levine
termed `the quality of life' of a community. Where
social relations are strained, we also tend to observe
increased rates of crime, as well as unhealthier and
unhappier citizens. Public health researchers can learn
much from the theories developed by criminologists to
explain variations in crime across nations, and across
geographic regions within countries. Every student of
Social Science & Medicine 48 (1999) 719±731
0277-9536/99/$ - see front matter # 1999 Elsevier Science Ltd. All rights reserved. PII: S0277-9536(98)00400-6
PERGAMON
* Corresponding author. Tel.: +1-617-432-0235; fax: +1-
617-432-3755; e-mail: [email protected]. 1 E-mail: [email protected]
2 E-mail: [email protected]
Sol Levine remembers how he urged us to engage in what he fondly called `creative integration', i.e. the ap-
plication of ideas and concepts across disciplines in ways that shed new light on established problems in one's own discipline. In our case, we want to argue
that crime is a window through which we can begin to address the unsolved problem in public health of explaining why some communities are healthier than
others. Finding the keys to explain community vari- ations in crime is part of the same endeavor as unlock- ing the social and ecological antecedents of ill health.
In Fig. 1, we have mapped out the conceptual framework of this paper, indicating the major variables that we will refer to throughout this piece. In the spirit of Sol Levine's lifelong dedication to the practice of
`creative integration', many of the concepts we will use in the following analysis are imported from disciplines traditionally considered remote from public health,
including political science, criminology, and economics. Key terms will be explained as they come up in the text, but we have also provided an Appendix of de®-
nitions.
1.1. The conceptual framework: crime, relative deprivation and social disorganization
1.1.1. Relative deprivation and anomie Beginning with the work of R. Merton (1968), one
strand of sociological theory has attributed the high crime rate in this country to the sense of anomie engen- dered by the cultural high value placed upon competi-
tive achievement, while at the same time there are wide disparities in the actual standard of living within the population:
What makes American culture relatively distinc- tive... is that it is a society which places a high pre-
mium on economic a�uence and social ascent for
all its members... This patterned expectation is regarded as appropriate for everyone, irrespective
of his initial lot or station in life... This leads to the subsidiary theme that success or failure are results wholly of personal qualities, that he who fails has
only himself to blame, for the corollary to the con- cept of the self-made man is the self-unmade man. To the extent that this cultural de®nition is assimi-
lated by those who have not made their mark, fail- ure represents a double defeat: the manifest defeat of remaining far behind in the race for success and
the implicit defeat of not having the capacities and moral stamina needed for success... It is in this cul- tural setting that, in a signi®cant portion of cases, the threat of defeat motivates men to the use of
those tactics, beyond the law or the mores, which promise `success'... The moral mandate to achieve success thus exerts pressure to succeed, by fair
means and by foul means if necessary.
If Merton's theory is correct, one would expect to observe higher crime rates in societies that exhibit greater degrees of inequality, for instance where the
size of the gap between the material assets of the rich and poor was greater. In other words, the greater the extent of status inconsistency, the more intense the pressure to close the gap, and correspondingly, the
more strenuous the e�orts expended to succeed by means `fair or foul'. As a test of the status inconsis- tency thesis, Kennedy et al. (1998) recently examined
the relationship between state-level homicide rates and the extent of income inequality. They found that the greater the disparity among household incomes, the
higher was the homicide rate at the state level. Relative deprivation was an even stronger predictor of homicide rates than household poverty rates. The correlations
between income inequality and household poverty rates with homicide were, respectively, 0.74 and 0.53 (P < 0.001 for both). Even after adjusting for poverty, income inequality accounted for 52% of the between-
state variance in homicide rates. Homicide rates may not provide the best test of the status inconsistency thesis; nonetheless, these ®ndings are consistent with
previous reports in the criminological literature of an association between income inequality and violent crime (Baily, 1984; Messner, 1989).
1.1.2. Social disorganization and social capital There is some support, then, for the theory linking
relative deprivation (and by implication, status incon- sistency) to crime. An alternative, though not necess- arily mutually exclusive, theoretical tradition in
criminological research focuses on the importance of social cohesion as a preventive force. Scholars working in the Chicago-school tradition of social disorganiz-Fig. 1. Conceptual framework of paper.
I. Kawachi et al. / Social Science & Medicine 48 (1999) 719±731720
ation theory have written at length on the link between
breakdown in social cohesion and high crime rates. Social disorganization theory was developed by the
Chicago School researchers Shaw and McKay (1942)
in their work, Juvenile Delinquency and Urban Areas. According to Shaw and McKay's classic work, the
same socio-economically disadvantaged areas in 21 US cities continued to experience high delinquency rates over a span of several decades despite changes in their
racial and ethnic composition. This demonstrated the persistent contextual e�ects of these disadvantaged communities on crime rates, regardless of what popu-
lations experienced them. These ®ndings forced the researchers to reject individualistic explanations of de-
linquency and focused them instead on community processes leading to the apparent transgenerational transmission of criminal behavior. The social organiz-
ational approach arising out of Shaw and McKay's work views local communities and neighborhoods in terms of systems of friendship, kinship and acquaint-
anceship networks, as well as formal and informal associational ties rooted in family life and ongoing
socialization processes (Sampson, 1995). Social disor- ganization has been de®ned as the ``inability of a com- munity structure to realize the common values of its
residents and maintain e�ective social controls'' (Sampson and Groves, 1989). From the perspective of crime control, a major dimension of social disorganiz-
ation is the ability (or lack of it) of a community to supervise and control teenage peer groups, especially
gangs. Shaw and McKay (1942) argued that residents of cohesive communities were better able to control the youth behaviors that set the context for gang vio-
lence. Examples of such controls include the supervi- sion of leisure-time youth activities, intervention in street-corner congregation and challenging youth `who
seem to be up to no good'. Social disorganization theory has recently been
linked to the emerging concept of social capital (Sampson, 1995). Social capital is de®ned by its princi- pal theorists (Coleman, 1990; Putnam, 1993) as ``those
features of social organization, such as networks, norms of reciprocity and trust in others, that facilitate cooperation between citizens for mutual bene®t''. It
follows that depletion in the stock of social capital is one of the distinguishing characteristics of socially dis-
organized communities (Sampson, 1995). Although the conceptualization and measurement of social capital are still evolving, two critical features of the concept
appear to be the level of trust among citizens and the density and rate of participation in voluntary associ- ations and local organizations (Putnam, 1995).
A growing number of studies in the criminological ®eld, reviewed by Sampson (1995), appear to support
the link between low stocks of social capital and high crime rates. Taylor et al. (1984) examined violent
crimes (such as mugging, assault, homicide, rape) across 63 street blocks in Baltimore. Based on inter-
views with 687 household respondents, they con- structed block-level measures of the proportion of respondents who belonged to an organization to which
coresidents also belonged, and the proportion of respondents who felt responsible for what happened in the area surrounding their home. Both variables were
signi®cantly and negatively associated with rates of violence. Simcha-Fagan and Schwartz (1986) collected survey information on 553 residents of 12 neighbor-
hoods in New York City, and found a signi®cant nega- tive relationship between the rate of self-reported delinquency and rates of organizational participation by local residents. A third set of studies conducted by
Sampson and Groves (1989) in the UK reported that density of local friendship networks had a signi®cant negative e�ect on robbery rates, while level of organiz-
ational participation by residents had signi®cant inverse e�ects on both robbery and stranger violence. Not only does participation in local associations
increase the level of community control, they may fa- cilitate the capacity of communities to obtain extralo- cal resources Ð such as police and ®re services, as well
as block grants Ð that have indirect consequences for crime control (Bursik and Grasmick, 1993). Taken together, these studies corroborate the social
disorganization thesis that the level of social cohesion,
or social capital, is critical to our understanding of rates of crime in neighborhoods, communities and even larger societies. We have thus two strands of evi-
dence linking societal characteristics to crime rates: the relationship between relative deprivation (as measured by income inequality) and crime, and the relationship
between social cohesion (as assessed by indicators of social capital) and crime. What has been lacking to this point is evidence linking income inequality to social cohesion.
1.1.3. Income inequality and social cohesion
One reason why greater income equality is related to lower crime rates (and better health outcomes in gen- eral) seems to be that it tends to reduce social div-
isions, thereby improving social cohesion (Wilkinson, 1997; Kawachi et al., 1997b). In other words, the notion has existed for some time that visibly high inequalities in material assets tend to produce resent-
ment that, in turn, disrupts the social fabric. In an essay on the dysfunctions of social strati®cation, soci- ologist M. Tumin (1953) speculated that ``to the extent
that inequalities in social rewards cannot be made fully acceptable to the less privileged in society, social strati- ®cation systems function to encourage hostility, suspi-
cion and distrust among the various segments of society and thus too limit the possibilities of extensive social integration''. Similarly, Blau and Blau (1982)
I. Kawachi et al. / Social Science & Medicine 48 (1999) 719±731 721
noted that pronounced and highly visible inequalities
lead to a sense that there are great riches within view but not within reach of many people destined to live in poverty. As a consequence, there is much `resentment,
frustration, hopelessness and alienation', producing a `sense of injustice, discontent and distrust' (Blau and Blau, 1982, p. 119). The greater the level of distrust
among individual members, the less cohesive is that so- ciety. Wilkinson (1996) has adduced a variety of quali-
tative and quantitative evidence which suggests that more egalitarian societies tend to be more cohesive. There are several case studies of societies at certain
points in history Ð for example, UK during the two world wars and postwar Japan Ð that underwent a
rapid compression of the income distribution at the same time as experiencing a rise in the sense of solidar- ity and improving life expectancy. In a 20-year study
of administrative regions within Italy, Putnam (1993) reported a strong correlation (r = 0.81) between income inequality and an index of social capital Ð as
measured by the density of citizens' participation in community organizations (choral societies, soccer lea-
gues, Rotary clubs and the like). Citizens living in regions characterized by high levels of social capital were more likely to trust their fellow citizens and to
value solidarity, equality and mutual tolerance. In the US, Brehm and Rahn (1997) analyzed pooled
data from the nationally representative General Social Surveys from 1972 to 1994 in an attempt to uncover the determinants of civic participation (membership in
voluntary associations) and interpersonal trust. Among other variables such as educational attainment and the unemployment rate, the distribution of income (as
measured by the Gini Index) was found to be a statisti- cally signi®cant predictor of levels of interpersonal
trust reported by respondents to the survey. Using the same General Social Survey data, Kawachi et al. (1997a) examined the cross-sectional relationships
between regional income inequality and levels of social capital. The General Social Survey asked respondents in 39 states whether they agreed that `Most people can
be trusted Ð or that you can't be too careful in deal- ing with people?'` The percentage of citizens who
believed that most people could not be trusted was highly correlated with the degree of income inequality in each state (r = 0.7; P < 0.0001). Similarly, density
of civic participation, as gauged by the per capita membership of groups (church groups, fraternal organ-
izations, labor unions and so on) was correlated with income inequality (r =ÿ0.4, P < 0.01) (Kawachi et al., 1997a).
It would thus appear that the two theoretical tra- ditions in criminology Ð one linking crime to relative deprivation, and the other linking crime to social cohe-
sion Ð can be reconciled. In the analyses to follow, we will examine the associations between relative depri-
vation, social cohesion and the full range of crimes, spanning from violent crimes to property crimes. In
doing so, our objectives are two-fold: to present a social ecological approach towards conceptualizing the impact of socioeconomic inequalities on collective well-
being (where crime is taken as an indicator of commu- nity wellbeing) and secondly, to provide an ecologic test of the theory linking social cohesion to lower
crime rates. Our use of the term `social cohesion' will be admittedly somewhat loose in this paper. In the lit- erature, the term has been used to describe commu-
nities that are high in stocks of social capital (Wilkinson, 1996; Kawachi and Kennedy, 1997), as well as low in the extent of social disorganization, and strong in levels of collective e�cacy (Sampson et al.,
1997) (see Section 4). A full discussion of the concepts of `social cohesion' is beyond the scope of the present paper. However, it is important to note that social
capital and social cohesion are not the same thing. In a well-known example, criminal gangs may provide social capital to its members at the same time as dis-
rupting the extent of social cohesion within a commu- nity. Whilst acknowledging the existence of such special cases, in the rest of the paper we propose to
use markers of social capital to indicate the extent of social cohesion within communities.
2. Data and methods
2.1. Measurement of social capital
A core feature of social capital, as presented by its principal theorists (Coleman, 1990; Putnam, 1993) con- sist of levels of interpersonal trust among community members. Following Putnam (1993, 1995), we used
data from the General Social Surveys (GSS), con- ducted by the National Opinion Research Center to estimate state variations in levels of interpersonal trust.
The GSS is a nationally representative survey of non- institutionalized adults over 18 years living in the US. The surveys have been repeated 14 times over the last
two decades, and have included a set of questions on interpersonal trust. In the present study, we averaged 5 years of cumulated data (1986±1990), representing 7679 individual observations. Of the 50 states, only 39
are represented due to the small population of some states (e.g. Delaware, Rhode Island). Because the sampling design of the GSS was intended to be repre-
sentative of regions rather than states, we adjusted in- dividual responses using poststrati®cation weights to re¯ect the age, race/ethnic, and educational compo-
sition of each state. Detailed procedures for the post- strati®cation weighing are described elsewhere (Kawachi et al., 1997a).
I. Kawachi et al. / Social Science & Medicine 48 (1999) 719±731722
Interpersonal trust was assessed from responses to the GSS item that asked: ``Generally speaking, would
you say that most people can be trusted, or that you can't be too careful in dealing with people?'' For each state, we calculated the percentage of residents who
agreed that `most people cannot be trusted'. Belief in the good will and benign intent of others facilitates collective action and mutual cooperation, and therefore
adds to the stock of a community's social capital. In turn, collective action further reinforces community norms of reciprocity.
In addition to the GSS-derived trust variable, we also obtained from the 1990 US Census a measure of family social capital, i.e. the percentage of households in each state headed by a single mother. Sampson
(1987), Wacquant and Wilson (1989), Coleman (1990), and others have argued that marital and family disrup- tion may decrease informal social controls both within
the family as well as at the community level. In con- trast to two-parent households, single parent-headed households might be expected to provide less supervi-
sion and guardianship for their children and property. Moreover, a community consisting of many single parent-headed households might tend to lack collective
family control and supervision of peer-group and gang activity.
2.2. Measurement of poverty and relative deprivation
Poverty and household income data for each state
were obtained from the 1990 US Census Summary Tape File STF 3A. The poverty variable represents the percentage of households in a state that were living
below the Federal poverty threshold. The Federal pov- erty index is a wage-income based measure that does not include income from other sources such as public assistance programs. The index is updated annually
using the Consumer Price Index, to re¯ect changes in the cost of living. In 1990, this represented a household income of less than US$13,359 for a household of
four. We also obtained median household income for each state. Data from the Census provides annual household
income for 25 income intervals (US$0±5,000 at the bottom and US$150,000 or more at the top). To calcu- late income inequality, counts of the number of house- holds falling into each of the 25 income intervals were
obtained for each state. These interval data were con- verted into income deciles using a program developed by Welniak of the US Census Bureau. Our measure of
income inequality, the Robin Hood Index (RHI) was estimated for each state from the income decile distri- bution, which represents the share of total household
income in each decile. The Robin Hood Index is then calculated by summing the excess shares of income for those deciles whose shares of the aggregate income
exceed 10% (Atkinson and Micklewright, 1992). For example in Massachusetts, the RHI is 30.26%. This
represents the share of total income that would have to be transferred from households above the mean to those below the mean in order to achieve a perfectly
equal income distribution. The higher the value of RHI, the greater is the degree of income inequality. For a fuller description of the calculation of the Index,
see Kennedy et al. (1996).
2.3. Measures of crime
Age-adjusted homicide rates were obtained for each state from the Compressed Mortality Files (ICD 9th
revision codes E965) compiled by the National Center for Health Statistics, Centers for Disease Control and Prevention (CDC). Data for years 1987±1991 were
combined to provide more stable estimates of homicide rates. The ®rearm homicide rates were directly age- standardized to the US population, and are expressed as the number of deaths per 100,000 population.
Incidence rates of other crimes Ð rape, robbery, aggravated assault, burglary, larceny and motor vehicle theft Ð were obtained from the Federal Bureau of
Investigation's Uniform Crime Reports (UCR) for the years 1991±1994. These rates (expressed per 100,000 population) are based on the incidents of crime
reported to the police, and subsequently to the FBI through the crime reporting program. Incidence data, while subject to various biases related to variations in reporting, are nonetheless considered less susceptible to
bias compared to arrest data (Reiss and Roth, 1993). The de®nitions of each type of crime are provided in Appendix B. The category of violent crime includes
homicide, aggravated assault, robbery and rape; while property crime includes burglary, larceny and motor vehicle theft.
2.4. Other variables
We obtained from the 1990 Census data two other socioeconomic indicators that have been previously linked to crime rates: unemployment rate (Freeman,
1995) and lower educational attainment, as gauged by the percent of state residents who are high school graduates. Based on the well-established association between male youth and violent crime rates, we also
obtained Census-derived estimates of the proportion of the population in each state who were males aged 15± 24 years. Finally, we obtained state-speci®c estimates
of average alcohol consumption (mean self-reported drinks per day) from the Behavioral Risk Factor Surveillance System (BRFSS) conducted by the
National Center for Chronic Disease Prevention and Health Promotion (CDC). The BRFSS is a state- based, random telephone survey conducted from 1993±
I. Kawachi et al. / Social Science & Medicine 48 (1999) 719±731 723
1996, of over 350,000 community-dwelling US adults aged 18 years and over.
2.5. Data analysis
We constructed a correlation matrix to examine the relationships between measures of relative deprivation,
indicators of social capital, and various categories of crime. A principal components analysis with varimax rotation was also carried out in order to examine
whether particular clusters of ecologic variables tended to be associated with particular types of crime.
3. Results
3.1. Relative deprivation, social capital and crime (Table 1)
Our measure of relative deprivation (Robin Hood Index) was strikingly associated with rates of violent crime. The greater the degree of income disparity in a
given state, the higher were the rates of homicide (r = 0.74), aggravated assault (0.50) and robbery (0.36) (Table 1). The only category of violent crime with which income inequality was not associated was
rape (0.13). Income inequality was also correlated with higher rates of burglary (0.44), but not other types of property crime.
Both indicators of social capital Ð proportion of households headed by single females, and level of interpersonal mistrust Ð were strongly correlated with
violent crime as well as property crime. Greater inter- personal mistrust was linked with higher homicide (r = 0.82), assault (0.61) and robbery (0.45), as well as
burglary (0.54) (all correlation coe�cients, P < 0.05). The proportion of single female-headed households predicted virtually the same range of crimes associated
with interpersonal mistrust, but in addition, rape (r = 0.43) and motor vehicle theft (r = 0.41) (Table 1).
3.2. Determinants of social capital (Table 2)
Consistent with theory, indicators of deprivation, both absolute and relative, were correlated with reduced levels of social capital. Higher unemployment
rates and higher poverty rates were correlated with higher levels of mistrust as well as single female-headed households. Higher educational attainment (as
measured by the proportion of a state's residents who were high school graduates) was inversely correlated with both indicators of social capital. Income inequal-
ity was also strikingly correlated with lower levels of
Table 1
Correlations between state-level indicators of relative deprivation, social capital and crime
Crime indicators Relative deprivation indicator: Social capital indicators:
Robin Hood Index % single mothers mistrust
Violent crimes
Homicide 0.74 *
0.80 *
0.82 *
Rape 0.13 0.43 *
0.17
Assault 0.50 *
0.65 *
0.61 *
Robbery 0.36 *
0.58 *
0.45 *
Property Crimes
Burglary 0.44 *
0.53 *
0.54 *
Larceny ÿ0.04 0.12 0.01 MV-theft 0.24 0.41
* 0.31
* P < 0.05. Abbreviations % Single mothers means the percentage households in each state headed by single mothers, Mistrust
is the % residents in each state responding on General Social Surveys that `people cannot be trusted', MV-theft means motor ve-
hicle theft.
Table 2
Relationships between indicators of deprivation and social
capital
Indicators of deprivation Social capital indicators
% single mothers mistrust
% Unemployment 0.38 *
0.36 *
% High School Graduates ÿ0.53* ÿ0.70* % Poverty 0.36
* 0.49
*
Median income ÿ0.08 ÿ0.24 Robin Hood Index 0.58
* 0.73
*
* P < 0.05. Abbreviations: % single mothers means the
percentage households in each state headed by single mothers,
mistrust is the % residents in each state responding on
General Social Surveys that `people cannot be trusted'.
I. Kawachi et al. / Social Science & Medicine 48 (1999) 719±731724
trust (r = 0.73) and more female-headed households (r = 0.58). The correlation between the two indicators of social capital, female-headed households and dis-
trust, was 0.64 (not shown in Table 2).
3.3. Relationships of other socioeconomic indicators to crime (Table 3)
Interestingly, indicators of deprivation Ð % unem- ployment, % poverty and % high school graduates in a state Ð were only weakly or inconsistently related to
either violent or property crimes (Table 3). The sole exception was the correlation of these variables with homicide rates, though even these were not as strong
as those found with either the Robin Hood Index or the social capital indicators. Higher socioeconomic sta- tus of a population (as gauged by the proportion of
high school graduates in a state) was associated with lower homicide rates (r =ÿ0.67), whereas higher pov- erty and unemployment rates were associated with increased rates (r = 0.53 and 0.35, respectively).
Higher population educational attainment was also re- lated to lower rates of assault (r =ÿ0.34), but also higher rates of larceny (r = 0.34) (all correlations,
P < 0.05). Median income was positively associated with robbery rates and motor vehicle theft. No other signi®cant correlations were noted.
3.4. Alcohol consumption and crime (Table 3 continued)
Contrary to the known association between alcohol and aggression (Boyum and Kleiman, 1995), mean
alcohol consumption was not correlated with any form of violent crime or property crime (with the sole excep- tion of motor vehicle theft). This may re¯ect the posi-
tive correlation between mean alcohol intake and higher socioeconomic status. Mean intake of alcohol was higher in states with a higher proportion of high
school graduates (r = 0.36), lower rates of poverty (r =ÿ0.38) and higher median income (r = 0.54). This pattern of correlation is most likely explained by
the failure of mean alcohol intake to capture the U- shaped pattern of drinking among lower SES groups, i.e. lower SES groups are overrepresented among both
abstainers and alcohol abusers. Higher SES groups in the US are more likely to drink regularly and in mod- eration, and thus to report higher average alcohol con- sumption. The positive correlation between mean
alcohol intake and motor vehicle theft (r = 0.32, P < 0.05) should not be interpreted to mean that hea- vier drinkers commit more auto theft (the ecologic fal-
lacy). Rather, a more likely explanation is that communities with higher mean alcohol intake (a proxy for higher SES) are more likely to present opportu-
nities for auto theft. As a proxy for problem drinking, we examined the
state-speci®c rates of drunk driving arrests. As
expected, drunk driving correlated only modestly with mean alcohol consumption (r = 0.18). Perhaps more surprisingly, no statistically signi®cant correlation emerged between drunk driving and any of the cat-
egories of crime, either violent or property-related (data not shown).
3.5. Principal components analysis (Table 4)
Three factors emerged in the principal components
analysis with varimax rotation, together explaining 73.7% of the total variance (Table 4). Variables load- ing on the ®rst factor included median income, robbery
Table 3
Correlations between other socioeconomic variables and crime
% 15±24 M % HS-grad % Unemploid % Poverty Med-income ALC
Violent crimes
Homicide 0.32 * ÿ0.67* 0.35* 0.53* ÿ0.24 ÿ0.12
Rape 0.14 ÿ0.02 0.10 0.18 ÿ0.07 ÿ0.06 Assault 0.19 ÿ0.34* 0.18 0.18 ÿ0.01 0.01 Robbery ÿ0.04 ÿ0.14 0.27 ÿ0.08 0.38* 0.17
Property crimes
Burglary 0.13 ÿ0.25 0.04 0.25 ÿ0.04 0.08 Larceny ÿ0.07 0.34* ÿ0.15 ÿ0.07 0.12 0.14 MV-theft ÿ0.05 0.00 0.25 ÿ0.17 0.54* 0.32*
* P < 0.05. Abbreviations: % 15±24 M is the percentage of state population who are males 15±24 years old, % HS-grad the per-
centage of state population who are high school graduates, % unemploid the percentage unemployment in state, % poverty the
percent of households in state living below Federal poverty threshold, Med-inc is the median household income in state and ALC
is the mean alcohol consumption among adult residents in state (drinks per day); MV-theft is motor vehicle theft per 100,000 in-
habitants in state.
I. Kawachi et al. / Social Science & Medicine 48 (1999) 719±731 725
and motor vehicle theft. The direction of the coe�-
cients suggested that this factor captured the e�ect of
higher opportunities for property crime in more a�u-
ent states. Variables loading on the second factor
included income inequality (Robin Hood Index), level
of mistrust, percent high school graduates, poverty,
homicide rate and, to a lesser extent, percent female-
headed households and median income (negative). This
factor captured the e�ects of deprivation, both absol-
ute and relative, on disinvestment in social capital and
the associated high homicide rates. Finally, the vari-
ables loading on factor 3 were larceny, rape and bur-
glary. Of these, the ®rst two categories of crime did
not appear to be consistently correlated with any of
the indicators of deprivation or social capital. For
instance, in the zero-order correlations, larceny was
not associated with any of the ecologic variables exam-
ined, except for % high school graduates (Table 3),
where there was a modest positive correlation
(r = 0.34), the meaning of which was unclear. Rape
was positively associated with % single mothers
(r = 0.43; Table 1), but with little else. It thus appears
that this third factor is comprised of types of crime for
which we can so far ®nd little evidence of ecologic cor-
relates.
4. Discussion
4.1. Social disorganization theory and crime
Consistent with social disorganization theory, the strongest correlates of violent crime (homicide, aggra- vated assault and robbery Ð in descending order of
the strengths of correlations) turned out to be our two indicators of social capital: the level of interpersonal trust and the proportion of female-headed households
in a state. Our ®nding ®ts with a variety of quantitat- ive and qualitative evidence linking social cohesion at the community level to rates of crime. According to
W. J. Wilson (1987, 1991), E. Anderson (1990) and others, a critical factor explaining the high incidence of delinquency and crime in urban settings has been the loss of social bu�ers that normally exist in middle class
neighborhoods. Such bu�ers consist of formal and informal networks of organizations (church groups, business groups, neighborhood associations), as well as
the presence of social norms concerning labor force participation and educational attainment. Most recently, social disorganization theory has
been explicitly tested in the context of a community- based study of juvenile delinquency and crime in Chicago neighborhoods. Sampson et al. (1997) sur-
Table 4
Principal components analysis with varimax rotation of crime and state-level indicators of absolute and relative deprivation, and
social capital (Marked loadings are >0.70)
Variable Factor 1 Factor 2 Factor 3
Mistrust 0.338937 0.805899 *
0.107933
% Sngl_moth 0.464193 0.650934 0.262222
Robin Hood 0.123513 0.881895 *
0.083478
%HS-grad 0.022885 ÿ0.908551* 0.154662 % unemploid 0.415339 0.374696 ÿ0.253769 % poverty ÿ0.365372 0.830094* 0.088424 Med-inc 0.701641
* ÿ0.618578 ÿ0.067691 % 15±24 M ÿ0.112597 0.467352 0.031557 Homicide 0.388988 0.797839
* 0.349554
Rape 0.027644 0.154281 0.773805 *
Robbery 0.854248 *
0.221680 0.227146
Assault 0.505076 0.461596 0.553043
Burglary 0.336995 0.371190 0.750570 *
Larceny 0.124399 ÿ0.173388 0.902461* MV-theft 0.855976
* 0.046282 0.258674
Explained variance 3.155768 5.199452 2.712212
Proportion total 0.210385 0.346630 0.180814
Abbreviations: mistrust is the % residents in each state responding on General Social Surveys that `people cannot be trusted';
%Sngl-moth is % households in each state headed by single mothers; Robin Hood is Robin Hood Index of income inequality in
each state; % HS-grad is % of state population who are high school graduates; % unemploid is % unemployment in state; % pov-
erty is % of households in state living below Federal poverty threshold; Med-inc is the median household income in state; % 15±
24 M is the % of state population who are males 15±24 years old; MV-theft is motor vehicle theft per 100,000 inhabitants in state.
I. Kawachi et al. / Social Science & Medicine 48 (1999) 719±731726
veyed 8782 residents of 343 Chicago neighborhoods in 1995 to ask about their perceptions of social cohesion
and trust in the neighborhood. Residents were asked how strongly they agreed (on a ®ve point scale) that ``people around here are willing to help their neigh-
bors'', ``this is a close-knit neighborhood'', ``people in this neighborhood can be trusted'', ``people in this neighborhood generally don't get along with each
other'' and ``people in this neighborhood do not share the same values'' (the last two items were reverse- coded). The resulting scale was then combined with re-
sponses to questions about the level of informal social control (whether neighbors would intervene in situ- ations where children were engaging in delinquent behavior) to produce a summary index of `collective
e�cacy'. In turn, collective e�cacy was signi®cantly (P < 0.01) related to organizational participation (r = 0.45) and neighborhood services (r = 0.21). In
hierarchical statistical models adjusting for individual characteristics (age, SES, gender, ethnicity, marital sta- tus, home ownership and years in neighborhood), the
index of collective e�cacy was signi®cantly inversely associated with reports of neighborhood violence, vio- lent victimization, as well as homicide rates. For
example, a 2 standard deviation elevation in neighbor- hood collective e�cacy was associated with a 39.7% lower rate of homicide.
4.2. The reciprocal e�ects of crime on disinvestment in social capital
One dynamic aspect of social capital that cross-sec- tional analyses fail to address is the reciprocal relation-
ship between crime and social capital. In other words, high crime rates may themselves produce disinvestment in community social capital. Skogan (1991) identi®ed several feedback processes that contribute to declining
social capital, including: fear of crime leading to physi- cal and psychological withdrawal from community life; deteriorating conditions leading to the exit of
businesses, with accompanying loss of jobs (and norms of labor market attachment); and further change in the composition of localities (e.g. middle class ¯ight). If
people shun their neighbors due to fear of crime, fewer opportunities exist for local networks and associations to take hold. The resulting disorganization of commu- nity structure in turn fuels further crime, producing a
vicious cycle of declining social capital, followed by rising crime, followed by further disinvestment in social capital.
The trend toward increasing residential segregation during recent decades Ð whether by race or class Ð has been further hypothesized as a mechanism by
which traditional social bu�ers have become depleted (and social capital eroded) in many inner-city areas of the US (Kawachi and Kennedy, 1997). Demographic
evidence suggests that, accompanying the surge in income inequality in the US since the mid 1970's, there
has been sharp increase in the spatial concentration of poverty. Between 1970 and 1990, the percentage of urban poor Americans living in nonpoor neighbor-
hoods (de®ned as having poverty rates below 20%) declined from 45 to 31%, while the percentage living in poor neighborhoods (poverty rates between 20 and
40%) increased from 38 to 41%. Meanwhile, the share of the urban poor living in very poor neighborhoods (over 40% poverty) grew from 17 to 28% (Massey,
1996). Such patterns of residential concentration impose a double burden on the poor Ð not only do they have to grapple with the multiple problems arising from their own lack of income; they also have to deal
with the social e�ects of living in a neighborhood where most of their neighbors are also poor (Wilson, 1987; Kawachi and Kennedy, 1997). Examples of such
`concentration e�ects' include lack of role models of labor force attachment (caused by persistently high unemployment), and exposure to unsupervised peer
group activity as well as high rates of delinquency.
4.3. Race and crime
Studies of crime in the US have consistently found that the percentage of blacks in a neighborhood is
positively and strongly associated with rates of violent crime. The dispute arises over the meaning and in- terpretation of the variable `percent black'. Several stu- dies have reported a sharply attenuated e�ect of
neighborhood racial composition on rates of violence once family structure and socioeconomic characteristics have been taken into account (Sampson, 1995). It is an
established fact that African-Americans are dispropor- tionately represented among the poor, as well as among single female-headed households. As Wilson
(1987) and others have argued, race (and racial dis- crimination) is causally prior to socioeconomic status and family structure. Hence studies of socio-structural
characteristics and crime ought not to control for `per- cent black'; instead, an observed correlation between the variable `percent black' and violent crime rates is itself likely to be the product of confounding by socio-
economic status, family structure and residential segre- gation.
4.4. Crime over the life course
At each point in life transitions Ð schooling, entry
into labor market, job security Ð relative deprivation produces adverse risks in the most vulnerable segments of population (Bartley et al., 1997). Thus, societies
characterized by high income inequality tend to disin- vest in schooling and education (human capital), re¯ecting the diverging political interests of the a�uent
I. Kawachi et al. / Social Science & Medicine 48 (1999) 719±731 727
(whose interests are to keep the tax burden low) and the poor. Thus, according to Kaplan et al. (1996),
states with large income disparities show evidence of reduced investment in human capital, including more high school dropouts, lower 4th grade reading and
math pro®ciency, lower state spending on public edu- cation, and fewer library books per capita. In turn, truncated educational attainment is itself a powerful
predictor of diminished levels of both interpersonal trust and civic participation. The resulting social disor- ganization produces greater risk of delinquency and
unemployment, and weaker labor force attachment, and hence disadvantages tend to cumulate throughout the life course.
4.5. The relationship of poverty to crime
The present analysis failed to disclose a strong re- lationship between household poverty rates and prop- erty crime. We observed a moderately strong
correlation between poverty and homicide (r = 0.53), though the relationship was weaker than that between income inequality and homicide (r = 0.74). Hsieh and Pugh (1993) carried out a meta-analysis of the 34
aggregate data studies on the relationships between poverty, income inequality and violent crime. Despite di�erences in methodology, the majority of studies
agreed that violent crime is related to both poverty (pooled r = 0.44) as well as to income inequality (pooled r = 0.44). Of note, the e�ects of poverty were
more consistent across the studies using lower levels of aggregation (e.g. cities). The opposite was true for income inequality, which yielded more consistent as-
sociations with violent crime at state and national levels of aggregation. Our failure to observe associ- ations between poverty and property crime may re¯ect problems of data aggregation; we could not exclude
the presence of such an association at lower levels of aggregation, for example, cities or counties. For some categories of property crime, such as
motor vehicle theft, poverty was modestly inversely correlated (r =ÿ0.17). Furthermore, higher median income was signi®cantly positively correlated with
motor vehicle theft (r = 0.54) as well as robbery (r = 0.38). Previous studies, such as the 1984 national survey of 300 localities in the UK (Sampson and Groves, 1989), have similarly reported that higher
community socioeconomic status is associated with statistically signi®cant higher rates of property crime (burglary, auto theft and vandalism). An erroneous in-
terpretation of such ®ndings is that better o� individ- uals commit motor vehicle theft (an ecologic fallacy). A more likely interpretation of these ®ndings at this
level of aggregation is that states with higher poverty rates (and lower median household incomes) present fewer opportunities for motor vehicle theft due to lower
ownership of cars and other material possessions. If in- dividual data were available, we would conjecture that
motor vehicle theft is more likely to be committed by individuals with less access to material means. The in- terpretation of indicators of individual access to
resources, such as poverty rates or median income, are therefore more prone to the ecological fallacy.
4.6. Limitations
Several important limitations should be noted in our analyses. Foremost among these is the level of data aggregation. As Sampson (1995) noted, most macro-
level research on crime has relied on census data that rarely provide measures for the social variables hy- pothesized to mediate the relationship between com-
munity structure and crime. Few data-sets exist which permit a re®ned test of the social disorganization the- ory. Ideally, data on social capital and relative depri-
vation ought to be collected at theoretically relevant levels of aggregation, such as neighborhood blocks. The Chicago-based study of Earls and colleagues men- tioned earlier (Sampson et al., 1997) represents perhaps
the most sophisticated example of this approach. In the present study, we took advantage of social
trust data that happened to be available at the state
level through the General Social Surveys. However, it is unclear whether data aggregated at this level truly characterize the neighborhoods and communities
within a state. A second limitation of ecological analyses of our
type is the lack of individual information about socioe-
conomic characteristics as well as incidence of criminal o�enses or victimization. Again, as Sampson (1995, p. 205) noted, ``an apparent ecological or structural e�ect may in fact arise from individual-level causal processes.
For example, an observed result such as the correlation of median income or percentage black with violence may simply represent the aggregation of relationships
occurring at lower levels of social structure and not a manifestation of processes taking place at the level of the community as a whole''. To overcome this limi-
tation requires carrying out multi-level studies that gather information from individuals as well as the communities in which they reside (again, the Chicago study is an example of this type of design).
Obviously, the cross-sectional nature of the evidence we have presented does not permit causal inferences to be drawn. Thus, it is not possible to distinguish
whether the correlations we observed are due to genu- ine ecological-level e�ects on crime (as posited by social disorganization theory), or whether there is
di�erential selection of individuals into communities based on prior antisocial behavior. Undoubtedly, there are reciprocal feedback e�ects of crime on social cohe-
I. Kawachi et al. / Social Science & Medicine 48 (1999) 719±731728
sion that fail to be captured by cross-sectional ana- lyses.
Ecological studies, perhaps more so than individual- level studies, are susceptible to omitted variable biases. A more rigorous test of the hypothesized ecological re-
lationships would have included other potentially rel- evant variables including access to ®rearms, the role of police and punitive laws, as well as drug abuse. Some
of these data (such as illegal substance abuse) are di�- cult to obtain on a state basis. Finally, data on incidence rates of crime are suscep-
tible to various errors. Both available police statistics and victim surveys focus on a fairly narrow range of common law crimes such as homicide, sexual assault, robbery and theft. Most white collar and victimless
crimes are excluded from routine data collection (Lynch, 1995). Participation in the so-called `second economy' Ð illegal drugs, gambling, prostitution Ð is
di�cult to ascertain, yet disadvantaged communities rely disproportionately on this sector for their econ- omic survival. Police statistics obviously exclude crimes
that do not come to the attention of the police. In some instances, such as homicide, this undercount is trivial, while in others Ð e.g. larceny Ð a substantial
proportion of crimes may be omitted. For example, the National Crime Victimization Survey indicates that only 14.6% of personal larcenies involving household property worth under US$50 are reported to the
police, compared to 58.4% of aggravated assaults (Bureau of Justice Statistics, 1992). The generally weak correlations we observed between larceny and other
socio-structural variables (Table 1) may be attributable to poor reporting of this category of crime. Unfortunately, the National Crime Victimization
Survey is not disaggregated by state, so that an inde- pendent check of our ®ndings using victim reports was not possible.
5. Summary and conclusions
The present paper has outlined an ecological theory of crime. Using what are inherently ecological vari-
ables Ð income inequality and social cohesion Ð we have attempted to merge and synthesize our emerging epidemiological understanding of the role of these fac- tors for societal wellbeing, with an established tradition
in criminological literature that dates back to at least 1942, when Shaw and McKay ®rst developed social disorganization theory.
Although much of policy discussions surrounding crime prevention in the US tends to focus on gun con- trol, more policing and more incarceration, our socio-
structural analysis suggests an alternative avenue of approach. Building social capital is obviously a more challenging task than putting more police on the
streets. However, an ecological approach to thinking about crime suggests several innovative policies that,
on the face of it, have little to do with crime. Housing- based neighborhood stabilization (through tenant buy- outs, rehabilitation of existing low income housing,
low income housing tax credits) and dispersing the concentration of new public housing (avoiding residen- tial segregation) are just two examples of a ground-up
approach to promoting social stability and safer neigh- borhoods (Sampson, 1995). At the same time, policy- makers should not neglect top-down approaches, such
as policies to reduce income inequality, as a potential means to promote social cohesion. Bartley et al. (1997) have cogently argued that ``keeping public health on the political agenda is made more di�cult by the fact
that chronic illness is not epidemic and does not spread out from poor people to rich people''. However, crime is one of the social ills that has the
tendency to spill over into the general walks of life. Increasing evidence points to social cohesion as a vital ingredient for the maintenance of collective wellbeing,
and crime is the mirror of the quality of social re- lationships among citizens.
Acknowledgements
IK and BPK are recipients of the Robert Wood Johnson Foundation Investigator Awards in Health
Policy Research. RGW is supported by the Paul Hamlyn Foundation. The authors would like to thank Alvin R. Tarlov, MD, for his careful reading of, and
insightful suggestions on an earlier draft of this paper.
Appendix A. De®nitions of Key Social Ecologic
Variables
. Social disorganization: de®ned as the ``inability of a community structure to realize the common
values of its residents and maintain e�ective social controls'' (Sampson and Groves, 1989).
. Social capital: de®ned by its principal theorists (Coleman, 1990; Putnam, 1993) as those features of social organization, such as networks, norms of reciprocity, and trust in others, that facilitate co-
operation between citizens for mutual bene®t. Lack of social capital is thus thought to be one of the primary features of socially disorganized com-
munities (Sampson, 1995).
. Collective e�cacy: a community-level concept developed by Sampson et al. (1997) that combines a neighborhood's level of social cohesion with its extent of informal social control. Residents were
I. Kawachi et al. / Social Science & Medicine 48 (1999) 719±731 729
asked how strongly they agreed (on a ®ve point scale) that ``people around here are willing to help
their neighbors'', ``this is a close-knit neighbor- hood'', ``people in this neighborhood can be trusted'', ``people in this neighborhood generally
don't get along with each other'' and ``people in this neighborhood do not share the same values'' (the last two items were reverse-coded). The
resulting scale was then combined with responses to questions about the level of informal social control (whether neighbors would intervene in
situations where children were engaging in delin- quent behavior) to produce a summary index of `collective e�cacy'.
. Social cohesion: in the literature, `socially cohe- sive' has been used to describe communities that are high in stocks of social capital (Wilkinson,
1996; Kawachi and Kennedy, 1997), as well as low in the extent of social disorganization, and strong in levels of collective e�cacy (Sampson et
al., 1997). Although no single de®nition exists, it is important to note that social capital and social cohesion are not the same thing. In a well known
example, gangs may provide social capital to its members without contributing to the level of social cohesion in a community.
Appendix B. De®nitions of crime
. Homicide: age-standardized homicide rate per 100,000 inhabitants in each state.
. Aggravated assault: reported aggravated assault incidence in state (per 100,000 population); de®ned as an unlawful attack by one person upon
another for the purpose of in¯icting severe or aggravated bodily injury.
. Rape: reported rape incidence in state (per 100,000 population).
. Robbery: reported robbery incidence in state (per 100,000 population); de®ned as the taking or attempting to take anything of value from the care, custody or control of a person or persons by
force or threat of force or violence and/or putting the victim in fear.
. Burglary: reported burglary rate in state (per 100,000 population); de®ned as the unlawful entry of a structure to commit a felony or theft.
. Larceny: reported larceny±theft rate in state (per 100,000 population); de®ned as the unlawful tak-
ing, carrying, leading or riding away of property from the possession or constructive possession of
another. It includes crimes such as shoplifting, pocket-picking, purse-snatching, thefts from motor vehicles, thefts of motor vehicle parts,
bicycle thefts, etc., in which no use of force, vio- lence or fraud occurs. This crime category does not include embezzlement, `con' games, forgery
and worthless checks.
. Motor vehicle theft: reported rate of motor vehicle theft in state (per 100,000 population).
References
Anderson, E., 1990. Street Wise. University of Chicago Press,
Chicago.
Atkinson, A.B., Micklewright, J., 1992. Economic
Transformation in Eastern Europe and the Distribution of
Income. Cambridge University Press., Cambridge.
Baily, W.C., 1984. Poverty, inequality and city homicide
rates: some not so unexpected ®ndings. Criminology 22,
531±550.
Bartley, M., Blane, D., Montgomery, S., 1997. Health and the
life course: why safety nets matter. British Medical Journal
314, 1194±1196.
Blau, J.R., Blau, P.M., 1982. The cost of inequality: metropo-
litan structure and violent crime. American Sociological
Review 47, 114±129.
Boyum, D., Kleiman, M.A.R., 1995. Alcohol and other
drugs. In: J.Q. Wilson, Petersilia J. (Eds.), Crime. Institute
for Contemporary Studies, San Francisco, pp. 295±326.
Brehm, J., Rahn, W., 1997. Individual-level evidence for the
causes and consequences of social capital. American
Journal of Political Science 41, 999±1023.
Bureau of Justice Statistics, 1992. Drugs, Crime and the
Justice System: a National Report. Bureau of National
Statistics, Washington DC.
Bursik Jr., R.J., Grasmick, H., 1993. Neighborhoods and
Crime: the Dimensions of E�ective Community Control.
Lexington, New York.
Coleman, J.S., 1990. The Foundations of Social Theory.
Harvard University Press, Cambridge, MA.
Freeman, R.B., 1995. The labor market. In: J.Q. Wilson,
Petersilia, J. (Eds.), Crime. Institute for Contemporary
Studies, San Francisco, pp. 171±191.
Hsieh, C.C., Pugh, M.D., 1993. Poverty, income inequality,
and violent crime: a meta-analysis of recent aggregate data
studies. Criminal Justice Review 18, 182±202.
Kaplan, G., Pamuk, E., Lynch, J.W., Cohen, R.D., Balfour,
J.L., 1996. Inequality in income and mortality in the US:
analysis of mortality and potential pathways. British
Medical Journal 312, 999±1003.
Kawachi, I., Levine, S., Miller, S.M., Lasch, K., Amick III,
B., 1994. Income inequality and life expectancy: theory,
research and policy. Joint Program on Society and Health
Working Paper series. Working Paper No. 94±2. New
England Medical Center, Boston, MA.
I. Kawachi et al. / Social Science & Medicine 48 (1999) 719±731730
Kawachi, I., Kennedy, B.P., 1997. Health and social cohesion:
why care about income inequality?. British Medical Journal
314, 1037±1040.
Kawachi, I., Kennedy, B.P., Lochner, K., Prothrow-Stith, D.,
1997a. Social capital, income inequality and mortality.
American Journal of Public Health 87, 1491±1498.
Kawachi, I., Kennedy, B.P., Lochner, K., 1997b. Long live
community: social capital as public health. The American
Prospect November/December, 56±59.
Kennedy, B.P., Kawachi, I., Prothrow-Stith, D., 1996. Income
distribution and mortality: cross sectional ecological study
of the Robin Hood Index in the US. British Medical
Journal 312, 1004±1007.
Kennedy, B.P., Kawachi, I., Prothrow-Stith, D., Gibbs, B.,
Lochner, K., 1998. Social capital, income inequality and
®rearm violent crime. Soc. Sci. Med. 47, 7±17.
Lynch, J., 1995. Crime in international perspective. In: J.Q.
Wilson, Petersilia, J. (Eds.), Crime. Institute for
Contemporary Studies, San Francisco, pp. 11±38.
Massey, D.S., 1996. The age of extremes: concentrated a�u-
ence and poverty in the 21st century. Demography 33, 395±
412.
Merton, R.K., 1968. Social Theory and Social Structure. Free
Press, New York.
Messner, S.F., 1989. Economic discrimination and societal
homicide rates: further evidence on the cost of inequality.
American Sociological Review 54, 597±611.
Putnam, R.D., 1993. Making Democracy Work: Civic
Traditions in Modern Italy. Princeton University Press,
Princeton, NJ.
Putnam, R.D., 1995. Bowling alone: America's declining
social capital. J. of Democracy 6, 65±78.
Reiss, A.J., Roth, J.A. (Eds.), 1993. Understanding and
Preventing Violence. National Academy Press, Washington
DC.
Sampson, R.J., 1987. Urban black violence: the e�ect of male
joblessness and family disruption. American Journal of
Sociology 93, 348±382.
Sampson, R.J., 1995. The community. In: J.Q. Wilson,
Petersilia, J. (Eds.), Crime. Institute for Contemporary
Studies, San Francisco, pp. 193±216.
Sampson, R.J., Groves, W.B., 1989. Community structure
and crime: testing social-disorganization theory. American
Journal of Sociology 94, 774±802.
Sampson, R.J., Raudenbush, S.W., Earls, F., 1997.
Neighborhoods and violent crime: a multilevel study of col-
lective e�cacy. Science 277, 918±924.
Shaw, C., McKay, H., 1942. Juvenile Delinquency and Urban
Areas. University of Chicago Press, Chicago.
Simcha-Fagan, O., Schwartz, J., 1986. Neighborhood and de-
linquency: an assessment of contextual e�ects. Criminology
24, 667±704.
Skogan, W., 1991. Disorder and Decline. Free Press, New
York.
Taylor, R., Gottfredson, S., Brower, S., 1984. Block crime
and fear: defensible space, local social ties and territorial
functioning. Journal of Research in Crime and Delinquency
21, 303±331.
Tumin, M.M., 1953. Some principles of strati®cation: a criti-
cal analysis. American Sociological Review 18, 387±394.
Wacquant, L.J.D., Wilson, W.J., 1989. The cost of racial and
class exclusion in the inner city. Annals of the American
Academy of Political and Social Science 501, 8±25.
Wilkinson, R.G., 1996. Unhealthy Societies: the A�ictions of
Inequality. Routledge, London.
Wilkinson, R.G., 1997. Health inequalities: relative or absol-
ute material standards?. British Medical Journal 31, 591±
595.
Wilson, W.J., 1987. The Truly Disadvantaged: the Inner City,
the Underclass and Public Policy. University of Chicago
Press, Chicago.
Wilson, W.J., 1991. Studying inner-city social dislocations: the
challenge of public agenda research. American Sociological
Review 56, 1±14.
I. Kawachi et al. / Social Science & Medicine 48 (1999) 719±731 731
12 sources/Kennedy et al.pdf
SOCIAL CAPITAL, INCOME INEQUALITY, AND FIREARM
VIOLENT CRIME
BRUCE P. KENNEDY,1* ICHIRO KAWACHI,2 DEBORAH PROTHROW-STITH,1
KIMBERLY LOCHNER2 and VANITA GUPTA1
1Division of Public Health Practice, Harvard School of Public Health, 718 Huntington Avenue, Boston, MA 02115, U.S.A. and 2Department of Health and Social Behavior, Harvard School of Public Health,
718 Huntington Avenue, Boston, MA 02115, U.S.A.
AbstractÐStudies have shown that poverty and income are powerful predictors of homicide and violent crime. We hypothesized that the e�ect of the growing gap between the rich and poor is mediated through an undermining of social cohesion, or social capital, and that decreased social capital is in turn associated with increased ®rearm homicide and violent crime. Social capital was measured by the weighted responses to two items from the U.S. General Social Survey: the per capita density of mem- bership in voluntary groups in each state; and the level of social trust, as gauged by the proportion of residents in each state who believed that ``most people would take advantage of you if they got the chance''. Age-standardized ®rearm homicide rates for the years 1987±1991 and ®rearm robbery and assault incidence rates for years 1991±1994 were obtained for each of the 50 U.S. states. Income inequality was strongly correlated with ®rearm violent crime (®rearm homicide, r = 0.76) as well as the measures of social capital: per capita group membership (r =ÿ0.40) and lack of social trust (r = 0.73). In turn, both social trust (®rearm homicide, r = 0.83) and group membership (®rearm homi- cide, r =ÿ0.49) were associated with ®rearm violent crime. These relationships held when controlling for poverty and a proxy variable for access to ®rearms. The profound e�ects of income inequality and social capital, when controlling for other factors such as poverty and ®rearm availability, on ®rearm violent crime indicate that policies that address these broader, macro-social forces warrant serious con- sideration. # 1998 Elsevier Science Ltd. All rights reserved
Key wordsÐsocial strati®cation, poverty, social capital, inequality, homicide, crime, ®rearms
INTRODUCTION
``Take the people of Brianc° on. They allow the needy, the widows and orphans, to cut their hay three days earlier than the rest. When their homes are in ruins they repair them for nothing... In the past hundred years they have not had a single murder''.
Victor Hugo, Les MiseÂrables (1862)
Intentional injuries resulting from violence make
a signi®cant contribution to the mortality and mor- bidity of the U.S. population. Injuries caused by violent behavior are estimated to cost American so-
ciety approximately $26 billion dollars a year (Rice et al., 1989). Currently, homicide is the leading cause of death for young African±American males
and females (15±34) and the second leading cause of death for all 10±19 year-olds with an increasing number attributable to ®rearms (Hammett et al., 1992; Fingerhut et al., 1992).
Much of the recent policy debate concerning ways to reduce intentional injuries due to violence has focused on restricting access to lethal means
such as ®rearms. Recent studies suggest a strong as- sociation between gun availability and homicide
rates (Cook, 1991; McDowell et al., 1992;
Kellermann et al., 1993). Furthermore, there has
been an increase in adolescent self-reports of carry-
ing ®rearms that may have contributed to the homi-
cide problem, although whether the increase in
weapon-carrying is due to increased access to weap-
ons or some other factor remains unclear (Reiss
and Roth, 1993).
Unfortunately, the debate about restricting access
to ®rearms has focused attention on individual
behaviors often to the exclusion of other important
determinants of violent crime. The role that broader
social factors, such as income inequality and pov-
erty, play in determining the incidence of violent
crime have been increasingly neglected in the cur-
rent policy debate. Studies have shown that poverty
and income inequality, whether at the city, state, or
national level, are powerful predictors of homicide
and violent crime (Blau and Blau, 1986; Krahn et
al., 1986; Land et al., 1990; Hsieh and Pugh, 1993).
Income inequality, or other indices of relative depri-
vation, are considered to be stronger predictors of
homicide and violent crime than indices of absolute
deprivation, such as poverty (Baily, 1984; Messner,
1989). Recently, Kennedy et al. (1996) found that
the Robin Hood Index, a measure of income
inequality, predicted state-level variations in homi-
Soc. Sci. Med. Vol. 47, No. 1, pp. 7±17, 1998 # 1998 Elsevier Science Ltd. All rights reserved
Printed in Great Britain 0277-9536/98 $19.00 + 0.00
PII: S0277-9536(98)00097-5
*Author for correspondence.
7
cide rates. Even after adjusting for poverty, income inequality accounted for 52% of the between-state
variance in homicide rates. A number of theories have attempted to explain
the observed relationship between income inequality
and violent crime (Shaw and McKay, 1942; Blau and Blau, 1986; Wilson, 1987). Much of this work is built on an initial hypothesis by Shaw and
McKay (1942) that inequality, and the concen- tration of poor economic conditions, lead to social disorganization through a breakdown of social
cohesion and normlessness. It is hypothesized that communities lacking in social cohesion (social capi- tal) are less e�ective in exerting informal means of social control through establishing and maintaining
norms to reduce violence compared to communities with higher levels of social capital (Sampson and Wilson, 1995).
The present study was undertaken to examine two related hypotheses: (1) state-level variations in income inequality predict ®rearm homicide, assault,
and robbery rates independent of poverty and ®re- arm availability; (2) state-level variations in social capital predict ®rearm homicide, assault, and rob-
bery rates independent of poverty and ®rearm avail- ability; and (3) the e�ect of income inequality on violent crime is mediated by its e�ect on social capi- tal.
DATA AND METHODS
Measurement of poverty and income inequality
Poverty and household income data for each state were obtained from the 1990 U.S. Census Summary Tape File STF 3A. The poverty variable represents the percentage of households in a state
that were considered to be below the federal pov- erty index. The federal poverty index is a wage- income based measure that does not include income
from other sources, such as public assistance pro- grams. The index is updated annually to re¯ect cost of living changes in the Consumer Price Index. In
1990, this represented a household income of less than $13,359 for a household of four (U.S. Census, 1993). We also obtained median household income, per capita income, and the percentage of the popu-
lation living in urban areas data for each state. Data from the Census provides annual household
income for 25 income intervals (0±$5,000 at the
bottom and $150,000 or more at the top). To calcu- late income inequality, counts of the number of households falling into each of the 25 income inter-
vals along with the total aggregate income were obtained for the state. The interval data was con- verted into income deciles using a program devel-
oped by Welniak (1988) at the U.S. Census Bureau for this purpose. Our measure of income inequality, the Robin
Hood Index (RHI), was estimated for each state
from the income decile distribution (Atkinson and Micklewright, 1992), which represents the share of
total household income in each decile (see Table 6 for an example derivation). The RHI is calculated by summing the excess shares of income for those
deciles with shares that exceed 10%. In the case of Massachusetts, the RHI is 30.26% (Table 6). This represents the share of income that would have to
be transferred from those above the mean to those below the mean to achieve an income distribution of perfect equality (Atkinson and Micklewright,
1992). Hence, the higher the value of the Index, the greater the degree of inequality in the distribution of incomes.
Measurement of social capital
Two core constructs of social capital, as pre-
sented by its principal theorists (Coleman, 1990; Putnam, 1993a,b, 1995), consist of levels of mutual trust among community members, and civic engage-
ment. Civic engagement refers to the level of com- mitment of citizens to their communities and is re¯ected by their involvement in community a�airs. Typically, this is measured by membership in civic-
related and other associations and groups that bring members of a community together around shared interests. Following Putnam (1993a,b, 1995),
we used weighted data from the general social sur- vey (GSS), conducted by the National Opinion Research Center (Davis and Smith), to estimate
state variations in group membership and levels of social trust. The GSS is a national survey that samples noninstitutionalized English-speaking per-
sons 18 years or older living in the United States. The survey has been repeated 14 times over the last two decades, and has included a set of questions on social trust and organizational membership. In the
present study, we averaged 5 years of cumulated data (1986±1990) from the GSS, representing 7,679 individual observations from 39 states.
Level of civic engagement was measured by the per capita number of groups and associations (e.g., church groups, labor unions, sport groups, pro-
fessional or academic societies, school groups, pol- itical groups, and fraternal organizations) that residents belonged to in each state. The other com- ponent of social capital, trust in others, was
assessed from responses to two GSS items that asked: ``Do you think most people would try to take advantage of you if they got a chance, or
would they try to be fair?'' and ``Generally speak- ing, would you say that most people can be trusted or that you can't be too careful in dealing with
people?'' For each state, we calculated the percen- tage of respondents who agreed with the ®rst part of the above statements. Belief in the goodwill and
benign intent of others facilitates collective action and mutual cooperation, and therefore adds to the stock of a community's social capital. Collective action, in turn, further reinforces community norms
B. P. Kennedy et al.8
of reciprocity. In addition to the social trust items, we evaluated the response to another item on the
GSS as a marker of social capital: ``Would you say that most of the time people try to be helpful, or are they mostly looking out for themselves?''
The GSS was designed to provide a national and census region representative population sample, and as such, responses to the GSS are not necessarily
representative of a state's population. To correct for this potential bias when disaggregating to the state level, we used post-strati®cation weights (as
per Dr. Smith: personal communication) to adjust for the extent to which GSS respondents in a given state were over/under represented. To accomplish this, we developed post-strati®cation weights based
on the distribution of age, race, and educational attainment of GSS respondents. The stratum- speci®c weights were calculated as follows:
wi,j,k,l � Pi,j,k,l=pi,j,k,l where wi,j,k,l is the post-strati®cation weight for the GSS respondent residing in the i-th state, and being
of j-th age-group, k-th race, and l-th level of edu- cational attainment; Pi,j,k,l is the proportion of indi- viduals with these characteristics residing in the i-th
state, obtained from the 1990 U.S. Census; and pi,j,k,l is the corresponding proportion of such respondents in the GSS.
These weights were then used to adjust the indi- vidual responses to the social capital items in the GSS, using the weight procedure in SAS. For
example, in states where the GSS over-sampled younger, black, and less educated respondents, the levels of social trust were adjusted upwards. In all of the subsequent analyses, we used the weighted
responses.
Firearm homicide and violent crime
Age-adjusted overall and race-speci®c homicide
rates attributable to ®rearms (ICD 9th revision codes E965.0±E965.4) were obtained for each state from the Compressed Mortality Files compiled by
the National Center for Health Statistics, Centers for Disease Control and Prevention (CDC). Data for years 1987±1991 were combined to provide more stable estimates of the ®rearm homicide rates.
The ®rearm homicide rates were directly age-stan- dardized to the U.S. population, and are expressed as the number of deaths per 100 000 population.
In addition to age-adjusted ®rearm homicide rates, ®rearm assault and ®rearm robbery incidence rates were calculated by combining data from the
Federal Bureau of Investigation's Uniform Crime Reports (UCR) for the years 1991±1994 (these years were used instead of 1987±91 as the FBI only
began collecting these data in 1991) (Federal Bureau of Investigation, 1991±1994). These rates are based on the number of incidents involving assaults or robberies with a ®rearm reported to
police, and subsequently to the FBI through the crime reporting program. Incidence data, while sub-
ject to various biases due to factors that in¯uence reporting, is considered less susceptible to bias com- pared to arrest data (Reiss and Roth, 1993). The
rates were calculated by summing the number of incidents for each state across the four years and dividing by the sum of the estimated total popu-
lation for each state across the same four years. We used the state population data that is provided along with the incidence data in the UCR. Unlike
the ®rearm homicide rates, these incidence rates could not be adjusted for age as this information was not available in the UCR. Nor could the data be disaggregated by race, so that rates represent
crude overall ®rearm assault and robbery incidence rates.
Firearm availability
As state-speci®c measures of gun ownership are not available, we used the fraction of successful
suicides completed with a ®rearm as a surrogate for gun availability. This measure is the best currently available and has been used in numerous studies as
a proxy for ®rearm availability (Cook, 1978; Lester, 1989, 1991). Suicide data were also obtained from the Compressed Mortality Files by combining years 1987±1991 (ICD-9 codes E955.0±E955.4; E950±
E959).
Data analysis
All analyses were conducted using Pearson's cor- relation and ordinary least squares (OLS) regression with variables in their nontransformed state. Due to
problems of collinearity, the Robin Hood Index and social capital measures were not included in the same model. Instead, separate models were used to
test for their e�ects on ®rearm violent crime. For each of these models we ran two separate re- gressions. In the ®rst, we simply regressed each uni- variate predictor (group membership, social trust,
RHI) on each of the measures of ®rearm violent crime (age-adjusted ®rearm homicide, ®rearm assault, ®rearm robbery). In the second set of re-
gressions we examined the e�ects of each of the pre- dictors adjusting for the e�ects of poverty and ®rearm availability (percentage of suicides com-
pleted with a ®rearm). To model the joint e�ects of income inequality (RHI) and social capital (as measured by social trust), we conducted a path
analysis to decompose their relationship to the age- adjusted ®rearm homicide rate into direct and indir- ect e�ects (Pedhazur, 1973; Alwin and Hauser, 1975).
Social capital and ®rearm violent crime 9
RESULTS
Relationship between income inequality and ®rearm violent crime
There was substantial variation in the degree of income inequality among the states. The overall RHI for the U.S. was 30.22%. New Hampshire
(RHI = 27.13%) had the least income inequality and Louisiana (RHI = 34.05%) the greatest (Fig. 1). In the univariate regression analyses, RHI
was signi®cantly related to both the age-adjusted overall homicide (adjusted R
2 =0.54) and age-
adjusted ®rearm homicide (adjusted R 2 =0.56) rates
(Table 1). RHI was also signi®cantly related to ®re-
arm assault (adjusted R 2 =0.36) and robbery
(adjusted R 2 =0.23) rates, although these relation-
ships were not as strong (Table 1). The association of RHI to all of the ®rearm violent crime variables
remained highly statistically signi®cant after adjust- ing for poverty and ®rearm availability in the multi-
variate regression analyses: a one unit change in the RHI, which is the equivalent to transferring a one
percent share of total income from the wealthy to the less wealthy, was associated with a change in
the age-adjusted ®rearm homicide rate of 1.55 per 100 000 (95% con®dence interval [CI]: 1.12 to 1.99)
Fig. 1(a).
B. P. Kennedy et al.10
(Table 2). This association was stronger for whites
(adjusted R 2 =0.55, p < 0.0001) than for blacks
(adjusted R 2 =0.20, p < 0.0006) in the univariate
regression. RHI continued to be a statistically sig-
ni®cant predictor of age-adjusted ®rearm homicide
rates after adjusting for poverty and ®rearm avail-
ability among both whites (B = 0.69; 95% CI: 0.45
to 0.92; p < 0.0001) and blacks (B = 4.82; 95% CI:
2.57 to 7.07; p < 0.0001).
To further determine the robustness of this re-
lationship, we also examined the e�ects of RHI
after adjusting for state variations in median house-
hold income, per capita income, and percentage of
the population living in urban areas. None of these variables changed the association of the Robin
Hood Index with ®rearm violent crime (data not shown).
Relationships among social capital variables and ®re-
arm violent crime
All of the social capital measures were highly cor-
related with each other and with the violent crime measures (Table 3). Higher levels of social mistrust were associated with higher levels of ®rearm violent
crime, while higher per capita group membership was associated with lower levels of ®rearm violent
Fig. 1(b).
Social capital and ®rearm violent crime 11
crime. Social trust (percentage of people who agreed
that ``most people would try to take advantage of
you if they got a chance'') was more strongly as-
sociated with ®rearm homicide rates (adjusted
R 2 =0.68) than per capita group membership
(adjusted R 2 =0.23) (Table 4). These relationships
remained statistically signi®cant after adjusting for
poverty and ®rearm availability (Table 5). A one
unit change in social trust (percentage of people
who agreed that ``most people would try to take ad-
vantage of you if they got a chance'') was associ-
ated with a change in the age-adjusted ®rearm
Fig. 1(c). State rankings on income inequality, social capital, and ®rearm homicide
Table 1. E�ects of income inequality on overall homicide and ®rearm homicide, assault, and robbery rates (50 states)
Violent crime rates Years B S.e. Adjusted R 2
F1,48 p<
Overall homicide 1987±91 1.77 0.23 0.54 57.94 0.0001 Age-adjusted ®rearm homicide 1987±91 1.31 0.16 0.56 64.63 0.0001 Firearm assault 1991±94 28.56 5.30 0.36 28.08 0.0001 Firearm robbery 1991±94 23.75 6.04 0.23 15.47 0.0002
B. P. Kennedy et al.12
homicide rate of 0.27 (95% CI: 0.24 to 0.36) per
100 000 which is equivalent to about a 5% change
in the age-adjusted ®rearm homicide rate. For per
capita group membership, a one unit change in
group membership was associated with a change in
age-adjusted ®rearm homicide rates of 3.00 (95%
CI: 1.22 to 4.78) per 100 000. As with the RHI, the
social trust variable explained more of the between-
state variance in age-adjusted ®rearm homicide
rates among whites (adjusted R 2 =0.53, p < 0.0001)
than among blacks (adjusted R 2 =0.21, p < 0.002).
These e�ects persisted after adjusting for poverty
and ®rearm availability: B = 0.10 (95% CI: 0.05 to
0.15; p < 0.0001) and B = 0.63 (95% CI: 0.23 to
1.04; p < 0.003) respectively. The results were simi-
lar for per capita group membership (data not
shown).
We also examined the relationships of the other
social trust item (percentage of respondents who
agreed that ``most people can be trusted'') and the
Table 2. E�ects of income inequality on overall homicide and ®rearm homicide, assault, and robbery rates, adjusted for poverty and ®re- arm availability (50 states)
Violent crime rates B S.e. t, p< Adjusted R 2
F3,46 p<
Age-adjusted homicide 2.04 0.33 6.11, 0.0001 0.56 21.80 0.0001 Age-adjusted ®rearm homicide 1.55 0.22 6.92, 0.0001 0.62 27.67 0.0001 Firearm assault 30.82 7.95 3.88, 0.0001 0.35 9.94 0.0001 Firearm robbery 40.22 8.45 4.76, 0.0001 0.30 8.08 0.0002
Table 3. Correlations among indicators of social capital, income inequality, and ®rearm violent crime (39 states)
1 2 3 4 5 6 7 8 9
1 age-adjusted homicide
2 age-adjusted ®rearm
homicide 0.99*
3 ®rearm assault 0.61* 4 ®rearm robbery 0.57* 0.56* 0.53* 5 ®rearm suicide 0.41* 0.48* 0.50* 0.45* 6 income inequality 0.73* 0.76* 0.48* 0.31* 0.43* 7 group membership ÿ0.51* ÿ0.49* ÿ0.34* ÿ0.33* ÿ0.05 ÿ0.40* 8 trust 1
a 0.82* 0.83* 0.55* 0.52* 0.49* 0.73* ÿ0.54*
9 trust 2b 0.72* 0.73* 0.34* 0.46* 0.44* 0.71* ÿ0.65* 0.79* 10 helpfulness
c 0.72* 0.75* 0.60* 0.54* 0.51* 0.71* ÿ0.54* 0.81* 0.78*
*p < 0.05. aPercent responding: ``most people would try to take advantage of you if they got the chance''. b Percent responding: ``you can't be too careful in dealing with people''.
Table 4. E�ects of social capital on overall homicide and ®rearm homicide, assault, and robbery rates (39 states)
Violent crime rates Years B S.e. Adjusted R 2
F1,37 p<
Social trust: Percentage of respondents who agreed that ``most people would try to take advantage of you if they got a chance''. Overall homicide 1987±91 0.41 0.05 0.66 75.42 0.0001 Age-adjusted ®rearm homicide 1987±91
0.30 0.03 0.68 80.39 0.0001
Firearm assault 1991±94 6.17 1.28 0.37 23.27 0.0001 Firearm robbery 1991±94 5.72 1.35 0.31 17.99 0.0001
Per capita group membership Overall homicide 1987±91 ÿ5.06 1.37 0.25 13.67 0.0007 Age-adjusted ®rearm homicide 1987±91
ÿ3.49 1.03 0.23 11.59 0.002
Firearm assault 1991±94 ÿ65.39 29.89 0.09 4.79 0.035 Firearm robbery 1991±94 ÿ72.10 29.75 0.11 5.87 0.02
Table 5. E�ects of social capital on overall homicide and ®rearm homicide, assault, and robbery rates, adjusted for poverty and ®rearm availability (39 states)
Violent crime rates B S.e. t, p< Adjusted R2 F3,35 p<
Social trust: Percentage of respondents who agreed that ``most people would try to take advantage of you if they got a chance''. Age-adjusted homicide 0.38 0.05 6.79, 0.0001 0.67 26.65 0.0001 Age-adjusted ®rearm homicide 0.27 0.04 6.85, 0.0001 0.69 28.69 0.0001 Firearm assault 5.52 1.53 3.59, 0.0009 0.35 7.83 0.0004 Firearm robbery 7.37 1.51 4.85, 0.0001 0.37 8.59 0.0002
Per capita group membership Age-adjusted homicide ÿ4.39 1.21 ÿ3.63, 0.0009 0.44 11.09 0.0001 Age-adjusted ®rearm homicide ÿ3.00 0.88 ÿ3.39, 0.002 0.45 11.22 0.0001 Firearm assault ÿ54.68 28.87 ÿ1.89, 0.066 0.19 4.03 0.014 Firearm robbery ÿ67.34 31.14 ÿ2.16, 0.037 0.08 2.05 0.124
Social capital and ®rearm violent crime 13
perceived helpfulness item (percentage of respon- dents who agreed that ``most of the time people try
to be helpful'') to the ®rearm violent crime vari- ables. The e�ects of both these variables on age- adjusted ®rearm homicide rates were essentially
identical to the social trust variable discussed above: adjusted R
2 =0.54 (F1,37=42.95, p < 0.0001)
and adjusted R 2 =0.56 (F1,37=46.45, p < 0.0001)
respectively.
Relationships among inequality, social capital, and age-adjusted ®rearm homicide: Path analysis
State-level variations in income inequality (RHI) were strongly associated with lack of social trust: states with high inequality also had more respon-
dents who agreed that ``most people would try to take advantage of you if they got a chance'' (r = 0.73, p < 0.0001). The other social capital
measure, per capita group membership, was inver- sely related to the Robin Hood Index: states with low inequality had high per capita group member-
ship (r =ÿ0.40, p < 0.003). The path analysis indicated that the e�ect of
income inequality (as measured by the Robin Hood
Index) on age-adjusted ®rearm homicide is mediated in part by social capital (as measured by level of social trust). According to our model, income inequality exerts a large indirect e�ect on
age-adjusted ®rearm homicide through the social capital variable (Fig. 2). In Fig. 2, as income inequality increases so does the level of social mis-
trust which is in turn associated with increased age- adjusted ®rearm homicide rates.
DISCUSSION
The dominant current in the violence literature has sought to identify the individual factors that
distinguish violent o�enders from nono�enders,
whereas the purpose of the present study was to ask what societal characteristics Ð including income inequality, poverty, and social capital Ð predict
di�erential rates of homicide and violent crime. More than half a century ago, Shaw and McKay (1942) argued in their classic work, Juvenile
Delinquency and Urban Areas, that crime could be linked to broad social forces such as socioeconomic
deprivation. Shaw and McKay demonstrated that in socioeconomically depressed areas in 21 U.S. cities, high rates of crime persisted over several dec-
ades despite changes in the racial and ethnic com- position of the communities. The authors thereby rejected individualistic explanations of crime and
focused instead on the processes by which criminal patterns of behavior were apparently transmitted across generations in areas of social disorganization
and weak social controls. Following the path breaking work of Shaw and
McKay (1942), a number of studies throughout the 1970s and 1980s have investigated the relationships of poverty and income inequality to violent crime,
using cross-sectional, aggregate data at the national or subnational level (e.g., states, census tracts, cities,
and standard metropolitan statistical areas [SMSAs]). Hsieh and Pugh (1993) have provided a meta-analysis of the 34 aggregate data studies that
had been published on poverty, income inequality, and violent crime. Despite di�erences in method- ology, the vast majority of studies agree that violent
crime is related to poverty (pooled r = 0.44), as well as to income inequality (pooled r = 0.44). Interestingly, the e�ects of poverty were more
homogeneous across the studies using lower levels of aggregation (e.g., cities). The converse was true
for income inequality, which yielded more homo- geneous e�ect sizes across studies using state and national levels of aggregation. This may be
accounted for by the di�erences in how the two variables e�ect violent crime rates: income inequal- ity, as a measure of relative deprivation, captures
the e�ect of the individual's relationship to the lar- ger society, whereas poverty, as a measure of absol-
ute deprivation, captures the e�ect of resource deprivation on individuals. This is consistent with Richard Wilkinson's view that at lower levels of
aggregation individual, or absolute income will mat- ter more to health than income inequality Ð it is only within larger geographic areas that the social
heterogeneity which is necessary for the e�ect of income inequality to occur that one ®nds a relation- ship between income inequality and health. As
Wilkinson (1996) points out, it is not the inequality within Harlem that matters to its residents health,
but rather the fact that so many are poor relative to the rest of the U.S.. Thus, there is growing consensus that societal-
level variables such as material deprivation may be a cause of crime. However, much less empirical
Fig. 2. Path coe�cients for the e�ects of income inequality and social capital on age-adjusted ®rearm homicide rates (39 states). Note: zero-order correlations are in parenth- eses, path coe�cients are bold. Inequality is measured by the Robin Hood Index and social capital is measured by the percentage of respondents who agree that ``most people would try to take advantage of you if they got a
chance''
B. P. Kennedy et al.14
work has been carried out to elucidate the pathways
and mechanisms involved. According to one view, homicides and other violent crimes are explained as
an individual's reaction to resource deprivation, subsequent personal frustration, and di�use hosti-
lity directed against targets of opportunity
(Messner, 1983; Crutch®eld, 1989; Arthur, 1991). Another view is that residential segregation and the
concentration of deprived groups in urban slums have given rise to subcultures that value toughness,
excitement, and fatalism, and these subcultural
values supposedly bring young people in con¯ict with the law (Blau and Blau, 1982). Our aim in the
present study was to test yet a third hypothesis: that a pronounced and highly visible gap in the dis-
tribution of income (as distinct from the absolute standard of living) may give rise to social disorgan-
ization and low social cohesion, as indexed by the
level of mutual distrust among members of society, as well as their propensity to associate with each
other. In turn, we hypothesized that lack of social cohesion (or ``social capital'') would predict aggre-
gate rates of homicide and violent crime.
Our ®ndings suggest that income inequality is
powerfully related to the incidence of homicide and violent crimes via the depletion of social capital.
These ®ndings hold even when controlling for pov- erty and access to ®rearms. A variety of sociological
and criminological evidence supports these claims.
In a pooled analysis of 20 years (1975±1994) of data from the General Social Surveys, involving
over 29 000 respondents, Brehm and Rahn (1997) found that rising income inequality was a signi®cant
predictor of declining trust in others. In turn, a decline in social trust was predictive of diminished
levels of group membership. According to the work
of Wilson (1987, 1991), Anderson (1990) and others, a critical factor responsible for the high inci-
dence of delinquency and crime in urban settings has been the loss of social bu�ers that normally
exist in middle class neighborhoods. Such bu�ers
consist of formal and informal networks of organiz- ations (church groups, business groups, neighbor-
hood associations), as well as the presence of social norms concerning work and education. These buf-
fers have become depleted in inner-city areas as a result of the increasing residential segregation of the
poor. Conversely, a youth living in a neighborhood that includes a mixture of working and professional families ``may observe increasing joblessness and
idleness but he will also witness many individuals going to and from work... he may be cognizant of an increase in crime, but he can recognize that
many residents in his neighborhood are not involved in criminal activity'', (Wilson, 1987, p. 56). While the work of Wilson (1987, 1991) and
others refers to the situation of inner-city African± Americans, our data suggest that the relationships of income inequality and social capital to ®rearm violent crime holds equally for whites as well.
Furthermore, these e�ects remain stable when con- trolling for poverty and the percentage of the popu- lation living in urban areas suggesting that income
inequality, through its erosion of social capital, may have a broader social impact that extends beyond speci®c groups in high-risk urban settings. While
our data are based upon analysis of aggregate data, the ``ecological fallacy'' Ð inferring individual re- lations based on grouped data Ð is not an issue
here, since we have not made any cross-level infer- ences (Susser, 1994a,b). Our analyses have used purely ecologic variables (social capital, income inequality, prevalence of poverty) to predict purely
ecological outcomes (population rates of ®rearm homicide and violent crime) and as such, do not provide the means for making predictions about in-
dividual behavior, yet can provide powerful indi- cators of macrosocial determinants of violent behavior (Schwartz, 1994; Susser, 1994a,b).
CONCLUSION
In his review of poverty and inequality and their relationship to crime, Braithwaite (1979) concluded that programs that simply targeted groups living in
poverty would not have a signi®cant impact on the overall crime rates in society. In contrast, he argued ``that gross economic measures to reduce the gap
between the rich and the poor and the rest of the population'' (pp. 231) are necessary if a signi®cant reduction in crime is to be expected. This view runs against the conventional violence
prevention wisdom which usually only targets high risk individuals and groups rather than attempting to shift the underlying societal forces that give rise
to a high incidence of violence in the population. This is not to say that the e�ects of poverty on vio- lent crime are negligible (®rearm homicide,
r = 0.49), and we would certainly not argue against policies to reduce the burden on families living in impoverished settings. Nor would we argue that
policies that restrict access to ®rearms be neglected as one of the means to reduce violent deaths. The proxy for access to ®rearms was highly correlated with ®rearm homicide (r = 0.44) indicating that it
Table 6. Data on derivation of Robin Hood Index (example: Massachusetts)
Decile of households Percent of total income
1 1.08 2 2.48 3 4.13 4 5.74 5 7.33 6 8.97 7 10.83 8 13.09 9 16.41 10 29.93
Social capital and ®rearm violent crime 15
is also a powerful determinant of ®rearm homicide as has been shown in other studies rates (Cook,
1991; McDowell et al., 1992; Kellermann et al., 1993). However, the profound e�ects of income inequality and social capital on ®rearm violent
crime when controlling for both of these factors, in- dicate that policies and interventions that address these broader, macro-social forces warrant serious
consideration.
AcknowledgementsÐKawachi and Kennedy are recipients of the Robert Wood Johnson Foundation Investigator Awards in Health Policy Research.
REFERENCES
Alwin, D. F. and Hauser, R. M. (1975) The decompo- sition of e�ects in path analysis. American Sociological Review 40, 37±47.
Anderson, E. (1990) Street Wise. University of Chicago Press, Chicago, IL.
Arthur, J. A. (1991) Socioeconomic predictors of crime in rural Georgia. Criminal Justice Review 16, 29±41.
Atkinson, A. B., Micklewright, J. (1992) Economic trans- formation in Eastern Europe and the distribution of income. Cambridge University Press.
Baily, W. C. (1984) Poverty, inequality, and city homicide rates: Some not so unexpected ®ndings. Criminology 22, 531±550.
Blau, J. R. and Blau, P. M. (1982) The cost of inequality. Metropolitan structure and violent crime. American Sociological Review 47, 114±129.
Blau, J. R. and Blau, P. M. (1986) The cost of inequality: Metropolitan structure and criminal violence. Sociological Quarterly 27, 15±26.
Brehm, J. and Rahn, W. (1997) Individual-level evidence for the causes and consequences of social capital. American Journal of Political Science 41, 999±1023.
Braithwaite, J. (1979) Inequality, Crime and Public Policy. Routledge and K. Paul, Boston.
Coleman, J. S. (1990) The Foundations of Social Theory. Harvard University Press, Cambridge, MA, pp. 300± 321.
Cook, P. J. (1978) The e�ect of gun availability on rob- bery and robbery murder: A cross-section study of ®fty cities. Policy Studies Review Annual 2, 743±781.
Cook, P. J. (1991) The technology of personal violence: A review of the evidence concerning the importance of gun availability and use in violent crime, self defense, and suicide. Crime and Justice: A Review of Research, ed. M. Tonry, 14, pp. 1±71. University of Chicago Press, Chicago, IL.
Crutch®eld, R. D. (1989) Labor strati®cation and violent crime. Social Forces 68, 489±512.
Davis, J. A., Smith, T. W. General Social Survey Cumulative File. University of Michigan, Interuniversity Consortium and Social Research, Ann Arbor, MI.
Federal Bureau of Investigation (1991±1994) Uniform Crime Reports.
Fingerhut, L. A., Ingram, D. D. and Feldman, J. J. (1992) Firearm and non®rearm homicide among persons 15 through 19 years of age. JAMA 267, 3048±3053.
Hammett, M., Powell, K. E., O'Carroll, P. W. and Clanton, S. T. (1992) Homicide surveillance ± United States, 1979±1988. Morbidity and Mortality Weekly Report 41, 1±33.
Hsieh, C. C. and Pugh, M. D. (1993) Poverty, income inequality, and violent crime: A meta-analysis of recent aggregate data studies. Criminal Justice Review 18, 182± 202.
Kellermann, A. L., Rivera, F. P. and Rushforth, N. B. et al. (1993) Gun ownership as a risk factor for homicide in the home. New England Journal of Medicine 329, 1084±1091.
Kennedy, B. P., Kawachi, I. and Prothrow-Stith, D. (1996) Income distribution and mortality: Test of the Robin Hood Index in the United States. British Medical Journal 312, 1004±1007.
Krahn, H., Hartnagel, T. F. and Gartrell, J. W. (1986) Income inequality and homicide rates: Cross-national data and criminological theories. Criminology 24, 269± 295.
Land, K., McCall, P. and Cohen, L. (1990) Structural cov- ariates of homicide rates: Are there any invariances across time and space? American Journal of Sociology 95, 922±963.
Lester, D. (1989) Gun ownership and suicide in the United States. Psychological Medicine 19, 519±521.
Lester, D. (1991) Crime as opportunity: A test of the hy- pothesis with European homicide rates. British Journal of Criminology 31(2), 186±188.
McDowell, D., Loftin, C. and Wiersema, B. (1992) A comparative study of the preventive e�ects of manda- tory sentencing laws for handgun crimes. Journal of Criminal Law and Criminology 83, 378±394.
Messner, S. F. (1983) Regional di�erences in the economic correlates of the urban homicide rate. Some evidence on the importance of cultural context. Criminology 21, 477±488.
Messner, S. F. (1989) Economic discrimination and so- cietal homicide rates: Further evidence on the cost of inequality. American Sociological Review 54, 597±611.
Pedhazur, E. J. (1973) Multiple Regression in Behavioral Research. Holt, Rinehart and Winston, New York.
Putnam, R. D. (1993a) The prosperous community. Social capital and economic growth. The American Prospect Spring, 35±42.
Putnam, R. D. (1993b) Making Democracy Work. Princeton University Press, Princeton, NJ.
Putnam, R. D. (1995) Bowling alone. America's declining social capital. Journal of Democracy 6, 65±78.
Reiss, A. J., Roth, J. A., eds. (1993) National Research Council. Understanding and Preventing Violence. National Academy Press, Washington, DC.
Rice, D. P., MacKenzie, E. J. and Associates (1989) Cost of Injury in the United States: A Report to Congress. Intitute for Health and Aging, USC and Injury Prevention Center, The Johns Hopkins University, San Francisco, CA.
Sampson, R. J., Wilson, W. J. (1995) Toward a theory of race, crime, and urban inequality. In Crime and Inequality, eds. J. Hagan and R. D. Peterson. Stanford University Press, Stanford, CA.
Schwartz, S. (1994) The fallacy of the ecologic fallacy: The potential misuse of a concept and the consequences. Am. J. Public Health 84, 819±824.
Shaw, C., McKay, H. (1942) Juvenile Delinquency and Urban Areas. University of Chicago Press, Chicago.
Susser, M. (1994a) The logic in ecological: II. The logic of design. Am. J. Public Health 84, 830±835.
Susser, M. (1994b) The logic in ecological: I. The logic of analysis. Am. J. Public Health 84, 825±829.
U.S. Bureau of Census (1993) Income and Poverty (CD- ROM).
Welniak, E. (1988) Unpublished software. U.S. Census Bureau.
Wilkinson, R. G. (1996) Income inequality and social cohesion. American Journal of Public Health 87, 104± 106.
Wilson, W. J. (1987) The Truly Disadvantaged: The Inner City, the Underclass, and Public Policy. University of Chicago Press, Chicago, IL.
B. P. Kennedy et al.16
Wilson, W. J. (1991) Studying inner-city social dislo- cations: The challenge of public agenda research. American Sociological Review 56, 1±14.
APPENDIX
Table 6 shows the shares of income earned by each decile of household in Massachusetts in 1990. For example, the bottom 10% of households accounted for 1.08% of the total income in that state. If income were distributed per- fectly equally, then each decile of household would account for exactly 10% of the share of total income. In the example of Massachusetts, households at or above the 70th percentile earned more than their ``fair share'' of total income.
The Robin Hood Index (RHI) is calculated by summing
the excesses above the fair share of total income (i.e.,
>10%) earned by each decile of households. In the case
of Massachusetts,
RHI ��10:83ÿ10:0�� �13:09ÿ10:0� ��16:41ÿ10:0���29:93ÿ10:0�
� 0:83�3:09�6:41�19:93 � 30:26%:
The Index is equivalent to the approximate proportion of
aggregate income that must be redistributed from house-
holds above the mean, and transferred to those below the
mean in order to achieve perfect equality in the distri-
bution of household incomes.
Social capital and ®rearm violent crime 17
12 sources/Massey.pdf
STATE OF THE DISCIPLINE
SEGREGATION AND STRATIFICATION A Biosocial Perspective
Douglas S. Massey Office of Population Research, Princeton University
Abstract
Thirty years after the civil rights era, the United States remains a residentially segregated society in which Blacks and Whites inhabit different neighborhoods of vastly different quality. Given high levels of racial segregation and elevated rates of Black poverty, it is axiomatically true that African Americans will experience more neighborhood poverty than other groups. Moreover, because poverty is associated with crime and delinquency, they will also be exposed to far higher rates of social disorder and violence. In this article I argue that long-term exposure to social disorder and violence because of segregation produces a high allostatic load among African Americans, which leads, in turn, to a variety of deleterious health and cognitive outcomes. After summarizing recent research on stress and allostatic load, I specify a biosocial model of racial stratification and draw upon it to explicate well-documented racial differentials with respect to health and cognition.
Keywords: Segregation, Stress, Stratification, Allostatic Load, Health
W+ E+ B+ DuBois himself first noted the close connection between a group’s ecolog- ical and social circumstances ~see Anderson and Massey, 2001!+ This fundamental insight was subsequently elaborated and extended by theorists and researchers asso- ciated with the Chicago School of Urban Sociology, beginning with the seminal work of Park ~1926! and extending through studies by Drake and Cayton ~1945!, Duncan and Duncan ~1957!, and Taeuber and Taeuber ~1965!+ All of these research- ers recognized that within an urban residential landscape governed by market trans- actions, social mobility was to a great extent built on an underlying foundation of spatial mobility+
This perspective has come to be known as the spatial assimilation model ~Massey 1985; Massey and Denton, 1985; Massey and Mullan, 1984!+ In order to gain access to better schools, safer streets, beneficial peer influences, lower insurance rates, and greater housing wealth, individuals and households move residentially+ As they move
Du Bois Review, 1:1 (2004) 7–25. © 2004 W. E. B. Du Bois Institute for African and African American Research 1742-058X004 $9.50 DOI: 10.10170S1742058X04040032
7
up the economic ladder, they seek to translate their socioeconomic gains into improved neighborhood circumstances, which puts them and their children into a better posi- tion to progress further up the ladder of social mobility+ Most new arrivals in American cities started out in central city neighborhoods of modest circumstances and then relied on this interplay between socioeconomic and residential mobility to ratchet themselves up the class hierarchy over time+
Because of pervasive racial discrimination, strong anti-Black prejudice, and con- tinuing high levels of residential segregation, however, this path of upward mobility has been largely inaccessible to Blacks+ Despite this fact, the connection between Black segregation and racial stratification remained largely unexamined from the mid-1960s through the mid-1980s, until the 1987 publication of William Julius Wilson’s book, The Truly Disadvantaged+ Wilson triggered renewed interest in the ecological bases of stratification by specifying the neighborhood as a critical factor mediating access to social, economic, and human capital ~Massey 2001a!+
Wilson noted that by the mid 1980s a remarkable transformation had taken place in urban America—poor Black neighborhoods were themselves getting steadily poorer, yielding a new geographic concentration of poverty that undermined the life chances of ghetto residents+ Subsequent work showed a powerful interaction between high segregation and high poverty rates which caused poor African Americans to experience much higher concentrations of poverty than other groups ~Massey et al+, 1991; Massey and Eggers, 1990; Massey and Fischer, 2000!+ Subsequent research has confirmed the importance of neighborhoods in the process of stratification+ In gen- eral, people who grow up and live in areas of concentrated poverty display lower levels of school completion, college attendance, and employment, and higher rates of incarceration, single parenthood, and welfare dependency ~Brooks-Gunn et al+, 1997; Sampson et al+, 2002!+
To date, theoretical speculation on how these deleterious outcomes are produced has focused on social mechanisms such as peer influences, cultural diffusion, the imitation of role models, access to networks, and collective efficacy ~ Jencks and Mayer, 1990; Sampson et al+, 1997, 1999!+ Much less attention has focused on potential biosocial pathways+ Recent research, however, suggests that biosocial mech- anisms may be quite important in stratifying individuals across a variety of dimen- sions, not simply in the dimensions of health and mortality, but also in cognition and social status ~Bremner 2002; McEwen and Lasley, 2002; Sampson 2003!+
In this article, I review recent evidence to establish the continued salience of racial segregation in American society and link its perpetuation to ongoing prejudice and discrimination+ Having done so, I outline a biosocial model that connects resi- dential segregation to a variety of social, psychological, and health outcomes through its intervening effects on neighborhood poverty and allostatic load+ My review of research on the causal linkages that comprise this model reveals only one link that remains to be established empirically+ I conclude by outlining a research agenda to corroborate this link and suggest the potential importance of biosocial research to understanding the process of racial stratification in the United States+
THE PERSISTENCE OF RACIAL SEGREGATION
Some observers have considered trends in average Black–White segregation across all metropolitan areas+ After noting the downward drift in the mean segregation values since 1970, they have concluded that segregation is declining in importance and that there is little cause for action or concern ~Thernstrom and Thernstrom,
Douglas S. Massey
8 DU BOIS REVIEW: SOCIAL SCIENCE RESEARCH ON RACE 1:1, 2004
1997!+ This reading of the data, however, confounds two distinct trends+ In the subset of U+S+ metropolitan areas that contain a relatively small percentage of Black residents, levels of Black–White segregation have indeed been declining, often quite precipitously ~Krivo and Kaufman, 1999; Massey and Gross, 1991!+ The downward trend has been especially pronounced in newer and smaller metropolitan areas of the south and west with relatively small Black populations, and in areas that contain colleges, military bases, and large stocks of post-1970 housing ~Farley and Frey, 1994!+ In contrast, among older metropolitan areas that contain a disproportionate share of the nation’s Black population, segregation levels remain stuck at very high levels+ Considering the average across all metropolitan areas misrepresents the situ- ation of African Americans because the declines are concentrated in places where few African Americans live+
Iceland et al+ ~2002! provide the latest data on levels and patterns of racial segregation+ I draw upon their figures to consider the degree of residential segrega- tion experienced by African Americans in the United States, dividing the population into three groups: those living in metropolitan areas characterized by low to moder- ate levels of segregation; those living in areas of high segregation; and those living in areas characterized by an intense form of racial isolation known hypersegregation ~Massey and Denton, 1989, 1993!+ Iceland and colleagues present Black–White dissimilarity indices for all U+S+ metropolitan areas in the year 2000+ The dissimilar- ity index measures the unevenness of Black and White settlement across neighbor- hoods, and is the most widely used measure of residential segregation ~see Massey and Denton, 1988!+
Following convention, I define dissimilarity values of sixty or more as “high” and those below this mark as either low or moderate+ I then divide highly segregated metropolitan areas into two categories: those that are hypersegregated and those that are not+ Hypersegregation exists whenever Blacks display an index value of sixty or greater on four of the five dimensions of segregation defined by Massey and Denton ~1988, 1989!: evenness, isolation, clustering, centralization, and concentration+ Some thirty metropolitan areas satisfy the criteria for hypersegregation+
Rather than considering the share of cities experiencing each level of segregation, I focus on the percentage of African Americans living under each segregation regime+ At the dawn of the twenty-first century, a clear majority of African Americans ~almost 60%! lived in a metropolitan area that was highly segregated, and a substantial minority ~some 41%! lived under conditions of hypersegregation+ Not all Blacks live in metropolitan areas, of course, so in Figure 1, I present a pie chart that includes only metropolitan residents+
As can be seen, nearly half of all metropolitan Blacks ~48%! live under conditions of hypersegregation, and another fifth ~21%! live under a regime of racial segrega- tion that is “merely” high+ In other words, among African Americans who reside in U+S+ metropolitan areas, a distinct minority—less than a third—currently enjoy conditions of low or moderate residential segregation+
Recent trends in Black–White segregation within the nation’s hypersegregated metropolitan areas are not very encouraging+ Figure 2 presents Black–White dis- similarity indices for 1980, 1990, and 2000 for the five most segregated metropol- itan areas and for the aggregate of all thirty areas that satisfy the criteria for hypersegregation+ Across all hypersegregated areas, the average level of Black- White segregation went from seventy-seven in 1980 to seventy-one in 2000, a drop of just 8% in twenty years+ In some metropolitan areas, change was barely detect- able+ For example, Chicago, Detroit, Newark, and Milwaukee display indices above eighty throughout the period+ No other group in the history of the United States
Segregation and Stratification
DU BOIS REVIEW: SOCIAL SCIENCE RESEARCH ON RACE 1:1, 2004 9
Fig. 1. Segregation experienced by URBAN African Americans in 2000+ Source: Iceland et al+ ~2002+!
Fig. 2. Trends in Black-White residential segregation for hypersegregated metropolitan areas 1980–2000+ Source: Iceland et al+ ~2002!+
Douglas S. Massey
10 DU BOIS REVIEW: SOCIAL SCIENCE RESEARCH ON RACE 1:1, 2004
has ever experienced such high levels of segregation, even for a brief historical moment ~Lieberson 1980!+
In their 1993 book, Massey and Denton referred to the regime of Black-White segregation in the United States as “American Apartheid+” Figure 3 brings this metaphor to life by comparing levels of Black–White dissimilarity in hypersegre- gated U+S+ metropolitan areas with the degree of segregation experienced by Afri- cans in the Union of South Africa under apartheid ~taken from Christopher 1993!+ Whereas the de jure apartheid of South Africa produced an average dissimilarity index of ninety in South African urban areas as of 1991, the de facto apartheid in the United States yielded values that were not much lower: Eighty-six for Detroit in the year 2000, eighty-three in Milwaukee, and eighty-one in both Chicago and Newark+ The average across all hypersegregated areas was seventy-two+
EXPLAINING RACIAL SEGREGATION
Data thus reveal that a majority of all African Americans, and the large majority of urban African Americans, continue to experience high levels of residential segrega- tion in U+S+ cities, and that about half of all urban Blacks and more than 40% of all African Americans experience hypersegregation, a degree of racial separation that is little different from that achieved in South Africa under apartheid+ A variety of hypotheses have been offered to explain persistent Black segregation+ The easiest hypothesis to dismiss is the hypothesis that racial segregation reflects socioeconomic differences between African Americans and Whites+ This explanation proposes that because the former generally have lower incomes than the latter, more African Americans are channeled into lower-class neighborhoods, on average, than Whites+
To test this hypothesis, Figure 4 presents Black–White dissimilarity indices ~taken from Massey and Fischer, 1999! that were computed within income categories
Fig. 3. Degree of Black-White segregation in hypersegregated metro areas of the U+S+ compared with metro areas in South Africa under apartheid+ Sources: Christopher ~1993!; Iceland et al+ ~2002!+
Segregation and Stratification
DU BOIS REVIEW: SOCIAL SCIENCE RESEARCH ON RACE 1:1, 2004 11
of the fifty largest metropolitan areas+ A line corresponding to a high level of segregation is positioned just above the Segregation Index of 60 to facilitate inter- pretation+ As can be seen, at all income levels the degree of Black–White segregation remains “high+” Although we observe a slight decline from the poorest to the lower- middle income category, thereafter the trend is flat+ At all levels of income, Blacks are highly segregated+ In contrast, among Latinos and Asians, the level of segregation is moderate among the poorest families and falls even lower as income rises+ Indeed, the poorest Latinos and Asians ~those earning under $15,000! are more segregated than the most affluent African Americans ~those earning at least $50,000!+
Other social scientists have argued that persistent racial segregation reflects the preference of African Americans for living in segregated Black neighborhoods ~Clark 1992; Patterson 1998!+ When Charles ~2003! tabulated nationally representative survey data on housing preferences, however, she found that Blacks expressed weaker preferences for co-residence with members of their own group than did Whites, Asians, or Latinos+ The bar chart in Figure 5 shows the ideal neighborhood racial composition expressed by White and Black respondents to the General Social Sur- vey+ The data come from a “show card experiment” where respondents were shown a picture of a neighborhood containing blank houses and were asked to color them in to indicate their preferred distribution of Black, White, Asian, and Latino neighbors+
This figures makes it clear that Whites very strongly prefer same-race neighbors, but that Blacks do not+ Whereas the ideal neighborhood for the typical White person is 57% White ~containing just a smattering of other groups!, the ideal neighborhood for Blacks is only 30% Black and would, in fact, contain a larger share of Whites ~42% on average!+ As of the year 2000, therefore, the degree of in-group preference ex- pressed by Whites was about twice that of Blacks whereas the willingness of African Americans to tolerate out-group neighbors was 2+6 times that of Whites+
Fig. 4. Segregation of Blacks, Latinos, and Asians from Whites by income in 1990+ Source: Massey and Fischer ~1999!+
Douglas S. Massey
12 DU BOIS REVIEW: SOCIAL SCIENCE RESEARCH ON RACE 1:1, 2004
The exceptional nature of White racial intolerance is indicated forcefully by Figure 6, which shows the percentage of Whites reporting an ideal neighborhood that is all White ~with no other groups present!, and the percentage reporting an ideal neighborhood with no Blacks present ~although allowing in other minorities!+ Some
Fig. 5. Ideal neighborhood desired by Whites and Blacks in 2000+ Source: Charles ~2003!+
Fig. 6. Preference for all in-group and no-outgroup neighborhoods in 2000+ Source: Charles ~2003!+
Segregation and Stratification
DU BOIS REVIEW: SOCIAL SCIENCE RESEARCH ON RACE 1:1, 2004 13
thirty years after the civil rights era, about a fifth of White Americans would still prefer to inhabit a neighborhood that was all White, and a quarter would prefer to live in a neighborhood that had no African Americans+ In contrast, only 6+5% of African Americans wished to live in an all-Black neighborhood and just 9% preferred one with no Whites+
These racial preferences appear to be driven more by negative stereotyping toward African Americans than by attachment to other Whites. When Charles ~2003! developed indicators of stereotyping, in-group attachment, and perceptions of class difference and used them to predict White avoidance of Blacks within neighborhoods, she found that avoidance was most powerfully explained by the holding of negative images about Blacks+ As shown in Figure 7, the standardized effect of racial stereotyping on neighborhood preferences ~0+390! was about seven times that of perceived class differences ~0+056!, and about four times that of in-group preferences ~0+091!+
Black residential segregation is not only a function of anti-Black attitudes, of course+ Substantial evidence suggests that discrimination remains a powerful force in American housing ~Galster 1990a, 1990b; Ross and Yinger, 2002; U+S+ Department of Housing and Urban Development 2002; Yinger 1993!+ The limita- tion of housing opportunities for African Americans was clearly demonstrated in a recent analysis done by Massey and Lundy ~2001!, who assigned auditors to call advertised rental units in the Philadelphia metropolitan area and inquire about the availability of apartments+ Male and female speakers of White, middle-class English, Black-accented English, and Black English Vernacular called selected list- ings and read a standardized script inquiring about the unit’s cost and availability+ Results showed that callers who spoke an identifiably “Black” linguistic register achieved far less access to rental housing than callers speaking White, middle-class English+
Fig. 7. Explaining White avoidance of Black neighbors+ Source: Charles ~2003!+
Douglas S. Massey
14 DU BOIS REVIEW: SOCIAL SCIENCE RESEARCH ON RACE 1:1, 2004
Figure 8 shows the percentage of auditors who reached a rental agent and were told that a unit was still available+ Access is always greater for Whites and a signifi- cant interaction between race, class, and gender appears to exist+ Middle-class White males always achieve the greatest access, followed by middle-class White females and middle-class Black males+ Behind them are lower-class Black males, and in last place are lower-class Black females ~assuming that Black English Vernacular indicates lower class origins!+ Whereas White, middle-class males gained access to rental housing on 76% of their attempts, Black lower-class females did so on only 38% of theirs+ Moreover, having gained access, Black females were far more likely to have the issue of credit problems raised and to be assessed application fees+ Whereas rental agents mentioned credit worthiness as a potential problem to 3% of White, middle- class males, they did so to about a quarter of lower-class Black females+
SEGREGATION AND STRATIFICATION
Persistent residential segregation undermines the social and economic well-being of African Americans in a variety of ways+ First, by restricting spatial mobility it neces- sarily limits social mobility because of the close interconnection between the two processes ~Massey et al+, 1987; Massey and Fong, 1990!+ Second, by segmenting Black housing demand and channeling White buyers away from Black neighbor- hoods, it reduces the value of Black housing, making it more difficult for African Americans to accumulate wealth in the form of home equity ~Conley 1999; Oliver and Shapiro, 1995; Yinger 1993!+ As a result, Black wealth remains a small fraction of White wealth despite improvements in employment and earnings ~Keister 2000!+ Third, segregation contributes to the spatial mismatch between the geographic placement of jobs and the residential location of the people who need them ~Kain 1968; Preston and McLafferty, 1999!+
Consistent with Wilson’s ~1987! emphasis on concentration effects and their role in perpetuating socioeconomic disadvantage, perhaps the most important mecha-
Fig. 8. Percentage gaining access to rental units and percentage having credit raised as an issue, Philadelphia 2000+ Source: Massey and Lundy ~2001!+
Segregation and Stratification
DU BOIS REVIEW: SOCIAL SCIENCE RESEARCH ON RACE 1:1, 2004 15
nism of racial stratification operates through segregation’s role in promoting the spatial concentration of poverty+ As already noted, high levels of racial segregation interact with shifts in the distribution of income to concentrate poverty geographi- cally ~Massey and Fischer, 2000!+ Under conditions of high or rising Black poverty, segregation necessarily produces neighborhoods of concentrated poverty because the disadvantage created during economic downturns is confined to a small number of racially isolated neighborhoods that are clustered together in space and concen- trated in high densities at the center of the metropolitan area+
The interactive effect of rising segregation and increasing poverty is illustrated in Figure 9 ~drawn from the simulation developed by Massey 1990!+ The bottom ~solid! line shows what happens to the spatial concentration of poverty as the level of Black-White segregation increases from minimum to maximum, assuming a constant Black poverty rate of 20% and a fixed but moderate level of class segregation between poor and non-poor Black households+ Under conditions of racial integra- tion with a 20% poverty rate, the average poor African American lives in a neigh- borhood that is 25% poor ~owing to modest class segregation!+ As racial segregation increases, however, the concentration of poverty nearly doubles+ Under conditions of complete segregation, the average poor African American lives in a neighborhood where 40% of the families are poor+
The top ~dashed! line shows what happens to the concentration of poverty when the rate of Black poverty is increased to 30%+ Under conditions of racial integration, this shift in the distribution of income raises the concentration of poverty somewhat: the share of poor in the neighborhood of the typical poor black person goes from 25% to 30%+ Under conditions of total segregation, in contrast, an already disad- vantaged neighborhood environment becomes markedly worse, with the concentra- tion of poverty rising from 40% to 60%+ The difference between a neighborhood where the poverty rate is 25% and one where it is 60% is a slightly higher rate of poverty, and a much higher level of segregation+
Fig. 9. Effect of racial segregation on concentration of Black poverty: simulation results for city of 128,000 inhabitants that is 25% Black and has class segregation+ Source: Massey ~1990!+
Douglas S. Massey
16 DU BOIS REVIEW: SOCIAL SCIENCE RESEARCH ON RACE 1:1, 2004
SEGREGATION, STRESS, AND STRATIFICATION
As segregation concentrates poverty, it also concentrates anything that is correlated with poverty to create a uniquely disadvantaged social environment characterized by high rates of joblessness, welfare dependency, substance abuse, and single parent- hood+ Because crime is also associated with poverty, segregation likewise ends up concentrating social disorder and violence, yielding an unusually hostile and threat- ening environment to which poor African Americans must adapt ~Anderson 1999; Massey 1995!+
Figure 10 shows how segregation increases exposure to major crimes within neighborhoods because of the observed correlation between the poverty rate and crimes such as murder, rape, assault, robbery, burglary, larceny, and auto theft ~taken from Massey 2001b!+ As can be seen, given complete integration and a Black poverty rate of 20%, the average poor African American is predicted to reside in a neighbor- hood with a crime rate of around 56 per 1000 inhabitants+ In contrast, given total racial segregation and a Black poverty rate of 30%, the typical poor African American is expected to live in a neighborhood where the crime rate is 84 per 1000, and in some cases, 50% or higher+
Because African Americans experience elevated rates of poverty and high levels of segregation, they are fated to live in environments characterized by much higher rates of crime and violence compared with other groups+ This fact is clearly illus- trated by recent data from the National Longitudinal Survey of Freshmen, which interviewed African Americans and Latinos entering twenty-eight selective colleges and universities in the Fall of 1999 ~see Massey et al+, 2003!+ Respondents were asked to estimate the racial composition of the schools and neighborhoods they inhabited at ages 6, 13, and 18, and to report the frequency with which they witnessed various examples of violence ~shootings, stabbings, beatings, etc+! and social disorder ~graf- fiti, prostitution, drunkenness, etc+!+
Fig. 10. How segregation increases exposure to major crime ~murder, rape, assault, robbery, burglary, larceny, auto theft!+ Source: Massey ~2001a, 2001b!+
Segregation and Stratification
DU BOIS REVIEW: SOCIAL SCIENCE RESEARCH ON RACE 1:1, 2004 17
Using these data, Massey and Fischer ~2002! constructed severity-weighted indi- ces of exposure to violence and disorder+ Figure 11 presents the resulting indices for African Americans and Latinos classified by the average level of segregation they ex- perienced while growing up, and then compares these measures to indices computed for Asians and Whites+ The figure clearly reveals that African Americans and Latinos who grew up in segregated schools and neighborhoods ~.70% minority, on average! experienced far greater exposures to dangerous and threatening events than those who grew up in integrated circumstances ~,30% minority!+ Those who came of age in ra- cially mixed schools and neighborhoods ~30%–70% minority! generally fell in-between+
Consider the index of exposure to social disorder+ African Americans and Latinos who grew up in integrated circumstances were exposed to about the same level of social disorder as Whites and Asians+ Whereas the index of exposure to social disorder was 18+5 for Whites and 18+3 for Asians, it was only slightly higher at 19+5 for those African Americans and Latinos who grew up in integrated schools and neighborhoods+ Among those coming of age in racially mixed settings, in contrast, the index was 25+4; and among those growing up under conditions of segregation, the index was 31+7+ Thus, moving from integration to segregation increased a stu- dent’s exposure to social disorder by around 63%+
The effect of segregation on exposure to violence is even more pronounced+ Whereas the severity-weighted index of exposure to violence stood at 11+8 for African Americans and Latinos from integrated backgrounds—only slightly more than the values of 10+9 and 10+3 observed for Asians and Whites—it was 18+1 for minorities from racially mixed backgrounds and 26+0 for those from segregated backgrounds+ In other words, segregation was responsible for increasing a student’s prior exposure to violence by a factor of around 2+5 compared with Whites, Asians, and minorities who grew up within integrated schools and neighborhoods+ Recall that these particular African Americans and Latinos had already been admitted into the most elite segment of American higher education, suggesting that the differential in exposure to disorder and violence by level of segregation would probably be even greater among African Americans and Latinos generally+
Fig. 11. Exposure of freshmen at selective schools to social disorder and violence in schools and neighborhoods while growing up+ Source: Massey and Fischer ~2002!+
Douglas S. Massey
18 DU BOIS REVIEW: SOCIAL SCIENCE RESEARCH ON RACE 1:1, 2004
Large differences in lifetime exposure to disorder and violence carry important implications for the process of stratification because of the well-documented effects of chronic stress on human capacities+ Frequent or prolonged exposure to disorder and violence within schools and neighborhoods because of racial segregation is quite likely to produce a chronic activation of the human stress response+ To understand the manifold effects of stress on human beings, biomedical researchers have devel- oped the concept of allostasis, which refers to the tendency of organisms to perpetuate their survival and maintain stability through bodily change in response to changes in the environment ~McEwen and Lasley, 2002; Sterling and Ayer, 1988!
Whenever a person perceives an external threat, a brain organ known as the hypothalamus triggers an allostatic response, which is a complex interaction between the brain, the endocrine system, and the immune system+ Upon perceiving the threat, the hypothalamus immediately signals the adrenal glands to release adrenaline ~McEwen and Lasley, 2002!+ The flow of this hormone into the bloodstream accel- erates the heartbeat, constricts blood vessels in the skin, increases blood flow to internal organs, dilates the bronchial tubes, triggers the release of fibrogen into the circulatory system ~to promote clotting!, releases glucose and fatty acids into the bloodstream from stored fats ~to provide a ready source of energy!, and signals the brain to produce endorphins ~to mitigate pain!+
While all this is going on, the hypothalamus simultaneously signals the pituitary gland to release an adrenocorticotropic hormone, which, in turn, causes the adrenal glands to secrete cortisol into the blood ~McEwen and Lasley, 2002!+ Cortisol acts to replace the energy stores depleted by adrenaline, converting energy into glycogen and fat+ Cortisol also promotes the conversion of muscle protein to fat, blocks insulin from taking up glucose, subtracts minerals from bones, and changes the external texture of white blood cells to make them “stickier” and more adhesive+
The allostatic response is nature’s way of maximizing an organism’s resources to meet an immediate, short-term threat+ Long-term functions such as the building of muscle, bone, and brain cells are temporarily sacrificed to put more energy into the bloodstream for evasive or aggressive action ~McEwen and Lasley, 2002!+ The hypothalamic-pituitary-adrenal ~HPA! axis is common to all mammals and is designed for infrequent and sporadic use+ Unlike most mammals, however, humans are capa- ble of keeping the HPA axis turned on indefinitely because humans are capable of experiencing stress from ideas in addition to actual events+ Human beings can antici- pate threatening circumstances mentally—imagining events that might occur or recall- ing past traumas ~Bremner 2002; McEwen and Lasley, 2002!+
Repeated triggering of the allostatic response through chronic exposure to stress- ful events—as when someone is compelled by poverty and discrimination to live in a dangerous and violent neighborhood—yields a condition known as allostatic load+ As allostatic load increases and persists over time, it has powerful negative effects on a variety of bodily systems ~McEwen and Lasley, 2002!+
One important set of effects is cardiovascular+ Chronically elevated levels of adrenaline increase blood pressure and raise the risk of hypertension+ Elevated fibro- gen levels increase the likelihood of blood clots and increase the likelihood of thrombosis+ The build-up of “sticky” white blood cells causes the formation of arterial plaques that contribute to atherosclerosis+ Elevated cortisol levels, meanwhile, cause the production of excess glycogen and fat, raising the risk of obesity, while the suppression of insulin leads to excessive blood sugar and a greater risk of Type II diabetes ~McEwen and Lasley, 2002!+
Chronically elevated levels of adrenaline also disrupt the functioning of the vagal nervous system+ This system is responsible for slowing down the heart rate and
Segregation and Stratification
DU BOIS REVIEW: SOCIAL SCIENCE RESEARCH ON RACE 1:1, 2004 19
reducing bodily tension, acting as a “brake” for the ACH axis+ Disruption of the vagal system contributes to the expression of a Type A personality, which is associated with aggressiveness, impulsiveness, frustration, and a low threshold for anger+ People with Type A personalities often try to reduce tension by self-medicating with drugs, alcohol, and tobacco, and through these poor coping choices end up exacerbating allostatic load and causing secondary damage to vital organs such as the liver, lungs, and heart ~McEwen and Lasley, 2002!+
Allostatic load also compromises the human immune system+ Long term exposure to elevated cortisol usually lowers the immune response to increase susceptibility to illness and infection ~Schulz et al+, 1998!+ In some circumstances, however, cortisol appears to overstimulate the immune system to mistakenly goad it into attacking targets within the body that don’t normally pose a threat, leading to the expression of inflammatory diseases such as asthma and autoimmune diseases such as multiple sclerosis, arthritis, and Type I diabetes ~McEwen and Lasley, 2002!+
Finally, allostatic load has serious consequences for a variety of brain systems, and hence, influences cognitive functioning+ The organ of the brain that is primarily responsible for the consolidation and storage of memory is the hippocampus ~Carter 1999!+ Because stressful events are important to remember, the hippocampus is rich in cortisol receptors and people are indeed more likely to remember things that are associated with strong emotions ~McEwen and Lasley, 2002!+ Our ancestors who recalled where and under what circumstances danger occurred were more likely to survive and pass on their genes+ Chronically elevated cortisol, however, causes the receptors to become permanently saturated, leading to atrophy of the hippocampus and an impairment of memory, both short-term and long-term ~Bremner 2002!+
Excessive cortisol also appears to interfere with the normal operation of neuro- transmitters such as glutamate, which is a critical ingredient in the formation of synaptic connections+ By disrupting the production and operation of glutamate at the synapse, allostatic load inhibits long-term potentiation—the formation of a relatively permanent neural connection—which is the fundamental chemical event in human learning+ In this way, chronic exposure to disorder and violence may compromise the very process of learning itself ~McEwen and Lasley, 2002!+
Finally, the hippocampus plays an important role in shutting down the HPA axis by reducing cortisol production+ As a result, damage to it is doubly detrimental+ Through its effect on the hippocampus, chronic stress creates a viscous cycle whereby excessive cortisol causes shrinkage of the hippocampus, which causes less inhibition of cortisol production, which also causes more hippocampal shrinkage ~McEwen and Lasley, 2002!+ Over the long run, this cycle leads to dendritic remodeling, wherein neurons become shorter and sprout fewer branches, as well as to the suppression of neurogenesis, or the creation of new brain cells ~Gould et al+, 1998!+ Simply put, people who are exposed to high levels of stress over a prolonged period of time are at risk of having their brains re-wired in a way that leaves them with fewer cognitive resources ~Bremner 2002; McEwen and Lasley, 2002!+
A BIOSOCIAL APPROACH TO STRATIFICATION
The foregoing review suggests a biosocial model of stratification that connects elements of social structure ~racial segregation and income inequality interacting to produce concentrated poverty and its correlate, spatially concentrated violence! to distinctively high allostatic loads among African Americans ~through their involun- tary confinement in areas of concentrated poverty and violence! to an elevated risk of
Douglas S. Massey
20 DU BOIS REVIEW: SOCIAL SCIENCE RESEARCH ON RACE 1:1, 2004
coronary heart disease ~hypertension, thrombosis, atherosclerosis, diabetes, and obe- sity!, a greater likelihood of inflammatory disorders ~asthma, multiple sclerosis, arthritis!, and impaired cognition ~atrophy of memory, inhibition of synaptic learn- ing, dendritic remodeling, and suppression of neurogenesis!+
This hypothesized biosocial model is summarized in Figure 12 with the various links in the causal chain labeled A through E+ Pathway A, the interaction of segre- gation and inequality to produce the concentration of poverty and its correlates is very well established in the research ~Massey 1990, 2001b; Massey and Denton, 1993; Massey and Fischer, 2000!+ Likewise, pathways C, D, and E, which connect allostatic load to compromised health and cognitive outcomes, have been confirmed in a growing number of clinical and laboratory studies ~reviewed in Bremner 2002; McEwen and Lasley, 2002!+ At the same time, Black-White differentials in mortality, and morbidity from a variety of causes are well-documented ~Collins and Hawkes, 1997; Hayward and Heron, 1999; Hummer 1996; Manton et al+, 1987; Stockwell and Goza, 1996!, and persistent gaps in measured cognitive skills are similarly well known ~ Jencks and Phillips, 1998!+
To date, these stubborn racial differentials with respect to health and cognition have been resistant to full explanation using the usual array of socioeconomic and demographic control variables ~Geronimus et al+, 1996; Hummer 1993; Navarro 1990; Phillips et al+, 1998!+ Even after exhaustive background controls are added to statistical models, a significant racial gap generally remains, leading some observers to fall back on genetic explanations ~Herrnstein and Murray, 1999; Rushton 2000!+
The only link in the model that has not been established empirically is pathway B, the connection between concentrated poverty0violence and high allostatic loads+ No matter how reasonable or logical this pathway might seem, researchers have not yet documented it empirically, though there is substantial evidence connecting segrega- tion to excess Black mortality ~Collins and Williams, 1999; Fang et al+, 1998; Guest et al+, 1998; Polednak 1997!+ The absence of empirical evidence for pathway B is not because investigators have tried and failed to produce such evidence+ Rather, owing to a lack of appropriate data,no one has yet been in a position to document the connection+
What the field needs at this point is a dataset that contains biosocial markers indicating allostatic load gathered from a large multi-racial sample whose individual, family, and neighborhood characteristics are well-defined and measured at various points in time+ Compiling such a dataset should be a top priority for stratification research+ The leading candidate for such a dataset is the National Longitudinal Survey of Adolescent Health, a nationally representative survey of students enrolled in grades 7 through 12 during September 1994 through April 1995 ~when they were roughly 12–18 years of age!+ Eligible respondents were re-interviewed during April–
Fig. 12. Biosocial model of racial stratification+
Segregation and Stratification
DU BOIS REVIEW: SOCIAL SCIENCE RESEARCH ON RACE 1:1, 2004 21
August of 1996 ~when they were aged roughly 14–20! and again during August 2001 through April of 2002 ~when they were aged 19–25!+ Some 15,197 respondents participated in the last wave of the survey, and each person was asked to contribute 15 cc of urine to test for the presence of sexually transmitted diseases+ However, 2 ml were set aside and frozen with the date and time of the collection recorded+ The existence of these urine samples provides a unique opportunity to generate data on the missing link between neighborhood conditions and health outcomes+
Although measurement of allostatic load from a single urine specimen is likely to be unreliable ~Stewart and Seeman, 2000! and influenced by a variety of sources of unobserved heterogeneity ~individual differences in diet, sleep, diurnal patterns of hormonal secretion, etc+!, these problems are common in social science+ The validity and reliability of measurement will probably be no less than for widely used social science indices of self-esteem, racial prejudice, liberalism-conservatism, and other attitudes+ As with these indices, access to a large sample size compensates for the lack of reliability and provides sufficient statistical power to separate patterns from noise+ To the extent that the measure is unreliable, however, error will mitigate against finding any significant relationship between neighborhood conditions and allostatic load+ Thus, if statistically significant relationships are found between neighborhood conditions, cortisol, and health or cognitive outcomes, they can therefore be regarded as conservative+ A better approach, of course, would be to build measurement of allostatic load into the next round of the Adolescent Health survey ~currently planned for 2004–2005!, using multiple biomarkers and assays to achieve greater validity and reliability+
Compiling a multi-level, longitudinal data file that links individuals with mea- sures of allostatic load is important because the biosocial model just outlined offers plausible, objective accounts of racial differentials with respect to health, cognition, and mortality that do not require one to fall back on essentialist genetic theories, which make little sense when applied to the socially constructed category of race+ In the biosocial model of stratification I have sketched, racial differentials are explained by the unique social structure to which African Americans are subjected in the United States+ Among all U+S+ social groups, only they simultaneously experience high rates of poverty and high levels of segregation+ As a result, they experience far higher rates of neighborhood poverty than members of other groups ~Massey and Eggers, 1990; Massey and Fischer, 2003!, thus exposing them to higher levels of violence and disorder and driving up their allostatic loads to produce a host of negative health and cognitive outcomes that undermine their ability to compete in the socioeconomic order+
In the past, many social scientists have shunned biologically grounded explana- tions of racial gaps for fear of legitimizing racist theories or out of a fear of being labeled a racist; but an appreciation of the biosocial mechanisms by which racial differentials are produced turns these fears on their heads+ Indeed, by understanding and modeling the interaction between social structure and allostasis, social scientists should be able to discredit explanations of racial difference in terms of pure heredity+ In an era when scientific understanding is advancing rapidly through interdisciplin- ary efforts, social scientists in general—and sociologists in particular—must abandon their hostility to biological science and incorporate its knowledge and understand- ings into their work+
Corresponding author : Professor Douglas S. Massey, Woodrow Wilson School of Public and International Affairs, Princeton University, 239 Wallace Hall, Princeton, NJ 08544. E-mail: [email protected]
Douglas S. Massey
22 DU BOIS REVIEW: SOCIAL SCIENCE RESEARCH ON RACE 1:1, 2004
REFERENCES Anderson, Elijah ~1999!+ Code of the Street: Decency, Violence, and the Moral Life of the Inner City+
New York: Norton+ Anderson, Elijah and Douglas S+ Massey ~2001!+ The sociology of race in the United States+ In
Elijah Anderson and Douglas S+ Massey ~Eds+!, Problem of the Century: Racial Stratification in the United States, pp+ 3–13+ New York: Russell Sage Foundation+
Bremner, J+ Douglas ~2002!+ Does Stress Damage the Brain?: Understanding Trauma-Related Disorders from a Neurological Perspective+ New York: Norton+
Brooks-Gunn, Jeanne, Greg J+ Duncan, and J+ Lawrence Aber ~1997!+ Neighborhood Poverty: Context and Consequences for Children+ New York: Russell Sage Foundation+
Carter, Rita ~1999!+ Mapping the Mind+ Berkeley, CA: University of California Press+ Charles, Camille Z+ ~2003!+ The dynamics of racial residential segregation+ Annual Review of
Sociology, 29: 67–207+ Christopher, A+ J+ ~1993!+ Segregation levels in the late-apartheid city 1985–1991+ Tijdschrift
voor Economische en Sociale Geografie, 83: 111–134+ Collins, Chiquita A+ and David R+ Williams ~1999!+ Segregation and mortality: The deadly
effects of racism? Sociological Forum, 14: 495–523+ Collins, James W+ and Elizabeth Hawkes ~1997!+ Racial differences in post-neonatal mortality
in Chicago: What risk factors explain the black infant’s disadvantage? Ethnicity and Health, 2: 117–125+
Clark, William A+ V+ ~1992!+ Residential preferences and residential choices in a multi-ethnic context+ Demography, 30: 451–466+
Conley, Dalton ~1999!+ Being Black, Living in the Red: Race, Wealth, and Social Policy in America+ Berkeley, CA: University of California Press+
Drake, St Clair and Horace R+ Cayton ~1945!+ Black Metropolis: A Study of Life in a Northern City+ New York: Harcourt, Brace, and Co+
Duncan, Otis D+ and Beverly Duncan ~1957!+ The Negro Population of Chicago: A Study of Residential Succession+ Chicago: University of Chicago Press+
Fang, Jing, Shantha Madhavan, William Bosworth, and Michael H+ Alderman ~1998!+ Residential segregation and mortality in New York City+ Social Science and Medicine, 47: 469–476+
Farley, Reynolds and William H+ Frey ~1994!+ Changes in the segregation of whites from blacks during the 1980s: Small steps toward a more integrated society+ American Sociological Review, 59: 23–45+
Galster, George C+ ~1990a!+ Racial discrimination in housing markets during the 1980s: A review of the audit evidence+ Journal of Planning Education and Research, 9: 165–175+
Galster, George C+ ~1990b!+ Racial steering in urban housing markets: A review of the audit evidence+ The Review of Black Political Economy, 18: 105–129+
Geronimus, Arline T+, John Bound, Timothy A+ Waidmann, Marianne Hillemeier, and Patri- cia B+ Burns ~1996!+ Excess mortality among Blacks and Whites in the United States+ New England Journal of Medicine, 335~21!: 1522–1558+
Gould, Elizabeth, P+ Tanapat, Bruce S+ McEwen, G+ Flugge, and E+ Fuchs ~1998!+ Proliferation of granule cell precursors in the dentate gyrus of adult monkeys is diminished by stress+ Proceedings of the National Academy of Sciences USA, 95: 3168–3171+
Guest, Avery M+, Gunnar Almgren, and Jon M+ Hussey ~1998!+ The ecology of race and socio- economic distress: Infant and working-age mortality in Chicago+ Demography, 35: 23–34+
Hayward, Mark D+ and Melonie Heron ~1999!+ Racial inequality in active life among adult Americans+ Demography, 36: 77–91+
Herrnstein, Richard, J+ and Charles Murray ~1999!+ The Bell Curve: Intelligence and Class Structure in American Life+ New York: Free Press+
Hummer, Robert A+ ~1993!+ Racial differentials in infant mortality in the U+S+: An examination of social and health determinants+ Social Forces, 72: 529–554+
Hummer, Robert A+ ~1996!+ Black-White differences in health and mortality: A review and conceptual model+ Sociological Quarterly, 37: 105–125+
Iceland, John, Daniel H+ Weinberg, and Erika Steinmetz ~2002!+ Racial and Ethnic Residen- tial Segregation in the United States 1980–2000+ Washington, DC: U+S+ Bureau of the Census+
Jencks, Christopher and Susan E+ Mayer ~1990!+ The social consequences of growing up in a poor neighborhood+ In Laurence E+ Lynn, Jr+ and Michael G+ H+ McGeary ~Eds+!, Inner City Poverty in the United States, pp+ 111–186+ Washington, DC: National Academy of Sciences+
Segregation and Stratification
DU BOIS REVIEW: SOCIAL SCIENCE RESEARCH ON RACE 1:1, 2004 23
Jencks, Christopher and Meredith Phillips ~1998!+ The Black–White test score gap: An intro- duction+ In Christopher Jencks and Meredith Phillips ~Eds+!, The Black-White Test Score Gap, pp+ 1–54+ Washington, DC: The Brookings Institution Press+
Kain, John F+ ~1968!+ Housing segregation, Negro employment, and metropolitan decentral- ization+ Quarterly Journal of Economics, 82: 175–197+
Keister, Lisa A+ ~2000!+ Wealth in America: Trends in Wealth Inequality+ New York: Cambridge University Press+
Krivo, Lauren J+ and Robert L+ Kaufman ~1999!+ How low can it go? Declining Black–White segregation in a multiethnic context+ Demography, 36: 93–109+
Lieberson, Stanley+ ~1980!+ A Piece of the Pie: Blacks and White Immigrants Since 1880+ Berkeley, CA: University of California Press+
Manton, Kenneth G+, Clifford H+ Patrick, and Katrina W+ Johnson ~1987!+ Health differen- tials between Blacks and Whites: Recent trends in mortality and morbidity+ Milbank Quar- terly, 65: 129–199+
Massey, Douglas S+ ~1985!+ Ethnic residential segregation: A theoretical synthesis and empir- ical review+ Sociology and Social Research, 69: 315–350+
Massey, Douglas S+ ~1990!+ American apartheid: Segregation and the making of the underclass+ American Journal of Sociology, 96: 329–358+
Massey, Douglas S+ ~1995!+ Getting away with murder: Segregation and violent crime in urban America+ University of Pennsylvania Law Review, 143~5!: 1203–1232+
Massey, Douglas S+ ~2001a!+ The prodigal paradigm returns: Ecology comes back to sociology+ In Alan Booth and Ann C+ Crouter ~Eds+!, Does it Take a Village? Community Effects on Children, Adolescents, and Families, pp+ 41–48+ Mahwah, NJ: Lawrence Erlbaum Associates+
Massey, Douglas S+ ~2001b!+ Segregation and violent crime in urban America+ In Elijah Anderson and Douglas S+ Massey ~Eds+!, Problem of the Century: Racial Stratification in the United States, pp+ 317–346+ New York: Russell Sage Foundation+
Massey, Douglas S+, Camille Z+ Charles, Garvey A+ Lundy, and Mary J+ Fischer ~2003!+ Source of the River: The Social Origins of Freshmen at America’s Selective Colleges and Universities+ Princeton, NJ: Princeton University Press
Massey, Douglas S+, Gretchen A+ Condran, and Nancy A+ Denton ~1987!+ The effect of residential segregation on Black social and economic well-being+ Social Forces, 66: 29–57+
Massey, Douglas S+ and Nancy A+ Denton ~1985!+ Spatial assimilation as a socioeconomic process+ American Sociological Review, 50: 94–105+
Massey, Douglas S+ and Nancy A+ Denton ~1988!+ The dimensions of residential segregation+ Social Forces, 67: 281–315+
Massey, Douglas S+ and Nancy A+ Denton ~1989!+ Hypersegregation in U+S+ metropolitan areas: Black and Hispanic segregation along five dimensions+ Demography, 26: 373–393+
Massey, Douglas S+ and Nancy A+ Denton ~1993!+ American Apartheid: Segregation and the Making of the Underclass+ Cambridge, MA: Harvard University Press+
Massey, Douglas S+ and Mitchell L+ Eggers ~1990!+ The ecology of inequality: Minorities and the concentration of poverty, 1970–1980+ American Journal of Sociology, 95: 1153–1189+
Massey, Douglas S+ and Mary J+ Fischer ~1999!+ Does rising income bring integration? New results for Blacks, Hispanics, and Asians in 1990+ Social Science Research, 28: 316–326+
Massey, Douglas S+ and Mary J+ Fischer ~2000!+ How segregation concentrates poverty+ Ethnic and Racial Studies, 23: 670–691+
Massey, Douglas S+ and Mary J+ Fischer ~2002!+ The long-term consequences of segregation+ Paper presented at the Annual Meetings of the Population Association of America, Atlanta, Georgia+
Massey, Douglas S+ and Mary J+ Fischer ~2003!+ The geography of inequality in the United States 1950–2000+ In William G+ Gale and Janet Rothenberg Pack, eds+, Brookings-Wharton Papers on Urban Affairs 2003, pp+ 1–40+ Washington, DC: Brookings Institution+
Massey, Douglas S+ and Eric Fong ~1990!+ Segregation and neighborhood quality: Blacks, Hispanics, and Asians in the San Francisco metropolitan area+ Social Forces, 69: 15–32+
Massey, Douglas S+ and Andrew B+ Gross ~1991!+ Explaining trends in residential segregation 1970–1980+ Urban Affairs Quarterly, 27: 13–35+
Massey, Douglas S+, Andrew B+ Gross, and Mitchell L+ Eggers ~1991!+ Segregation, the concentration of poverty, and the life chances of individuals+ Social Science Research, 20: 397–420+
Massey, Douglas S+ and Garvey Lundy ~2001!+ Use of Black English and racial discrimination in urban housing markets: New methods and findings+ Urban Affairs Review, 36: 470–496+
Douglas S. Massey
24 DU BOIS REVIEW: SOCIAL SCIENCE RESEARCH ON RACE 1:1, 2004
Massey, Douglas S+ and Brendan P+ Mullan ~1984!+ Processes of Hispanic and Black spatial assimilation+ American Journal of Sociology, 89: 836–873+
McEwen, Bruce and Elizabeth N+ Lasley ~2002!+ The End of Stress as We Know It+ Washington, DC: The Joseph Henry Press+
Navarro, Vincente ~1990!+ Race or class versus race and class: Mortality differentials in the United States+ Lancet, 336: 1238–1240
Oliver, Melvin L+ and Thomas M+ Shapiro ~1995!+ Black Wealth/White Wealth: A New Perspec- tive on Racial Inequality+ New York: Routledge+
Park, Robert E+ ~1926!+ The urban community as a spatial pattern and a moral order+ In Ernest W+ Burgess and Robert E+ Park ~Eds+!, The Urban Community, pp+ 3–18+ Chicago: Univer- sity of Chicago Press+
Patterson, Orlando ~1998!+ The Ordeal of Integration: Progress and Resentment in America’s “Racial” Crisis+ New York: Perseus+
Phillips, Meredith, Jeanne Brooks-Gunn, Greg J+ Duncan, Pamela Klebenov, and Jonathan Crane ~1998!+ Family background, parenting practices, and the Black–White test score gap+ In Christopher Jencks and Meredith Phillps ~Eds+!, The Black–White Test Score Gap, pp+ 103– 148+ Washington, D+C+: Brookings Institution Press+
Polednak, Anthony P+ ~1997!+ Segregation, Poverty, and Mortality in Urban African Americans+ New York: Oxford University Press+
Preston, Valerie and Sara McLafferty ~1999!+ Spatial mismatch research in the 1990s: Progress and potential+ Papers in Regional Science, 78: 387–402+
Ross, Stephen L+ and John Yinger ~2002!+ The Color of Credit: Mortgage Discrimination, Research Methodology, and Fair-Lending Enforcement+ Cambridge, MA: MIT Press+
Rushton, J+ Philippe ~2000!+ Race, Evolution, and Behavior: A Life History Perspective ( Third Edition)+ Port Huron, MI: Charles Darwin Research Institute+
Sampson, Robert J+ ~2003!+ The neighborhood context of well-being+ Perspectives in Biology and Medicine, 46: S53–S64+
Sampson, Robert J+, Jeffrey Morenoff, and Felton Earls ~1999!+ Beyond social capital: Spatial dynamics of collective efficacy for children+ American Sociological Review, 64: 633–660+
Sampson, Robert J+, Jeffrey D+ Morenoff, and Thomas Gannon-Rowley ~2002!+ Assessing ‘neighborhood effects’: Social processes and new directions in research+ Annual Review of Sociology, 28: 443–478+
Sampson, Robert J+, Stephen Raudenbush, and Felton Earls ~1997!+ Neighborhoods and violent crime: A multilevel study of collective efficacy+ Science, 277: 918–924+
Schulz, P+, C+ Kirschbaum, J+ Prüssner, and D+ Hellhammer ~1998!+ Increased free cortisol secretion after awakening in chronically stressed individuals due to work overload+ Stress Medicine, 14: 91–97+
Sterling, P+ and J+ Ayer ~1988!+ Allostatis: A new paradigm to explain arousal pathology+ In S+ Fischer and J+ Reason ~Eds+!, Handbook of Life Stress, Cognition, and Health, pp+ 629–649+ New York: John Wiley and Sons+
Stewart, J+ and T+ Seeman ~2000!+ Salivary cortisol measurement+ John D+ and Catherine T+ MacArthur Research Network on Socioeconomic Status and Health+ Webpage located at www+macses+ucsf+edu0research0allostatic0+
Stockwell, Edward G+ and Franklin W+ Goza ~1996!+ Racial differences in the relationship between infant mortality and socioeconomic status+ Journal of Biosocial Science, 28: 73–84+
Taeuber, Karl E+ and Alma F+ Taeuber ~1965!+ Negroes in Cities: Residential Segregation and Neighborhood Change+ Chicago: Aldine+
Thernstrom, Stephen and Abigail Thernstrom ~1997!+ America in Black and White: One Nation, Indivisible+ New York: Simon and Schuster+
U+S+ Department of Housing and Urban Development ~2002!+ Discrimination in metropolitan housing markets: National results from phase I of HDS2000+ Washington, DC: U+S+ Depart- ment of Housing and Urban Development+
Wilson, William J+ ~1987!+ The Truly Disadvantaged: The Inner City, the Underclass, and Public Policy+ Chicago: University of Chicago Press+
Yinger, John ~1993!+ Closed Doors, Opportunities Lost: The Continuing Costs of Housing Discrimi- nation+ New York: Russell Sage+
Segregation and Stratification
DU BOIS REVIEW: SOCIAL SCIENCE RESEARCH ON RACE 1:1, 2004 25
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
12 sources/Mills.pdf
C. Wright Mills, “The Promise [of Sociology]” Excerpt from The Sociological Imagination (originally published in 1959)
This classic statement of the basic ingredients of the "sociological imagination” retains its vitality and relevance today and remains one of the most influential statements of what sociology is all about. In reading,
focus on Mills' distinction between history and biography and between individual troubles and public issues.
Nowadays men often feel that their private lives are a series of traps. They sense that within their everyday
worlds, they cannot overcome their troubles, and in this feeling, they are often quite correct: What ordinary
men are directly aware of and what they try to do are bounded by the private orbits in which they live; their
visions and their powers are limited to the close-up scenes of job, family, neighborhood; in other milieux, they
move vicariously and remain spectators. And the more aware they become, however vaguely, of ambitions and
of threats which transcend their immediate locales, the more trapped they seem to feel.
Underlying this sense of being trapped are seemingly impersonal changes in the very structure of continent-
wide societies. The facts of contemporary history are also facts about the success and the failure of individual
men and women. When a society is industrialized, a peasant becomes a worker; a feudal lord is liquidated or
becomes a businessman. When classes rise or fall, a man is employed or unemployed; when the rate of
investment goes up or down, a man takes new heart or goes broke. When wars happen, an insurance salesman
becomes a rocket launcher; a store clerk, a radar man; a wife lives alone; a child grows up without a father.
Neither the life of an individual nor the history of a society can be understood without understanding both.
Yet men do not usually define the troubles they endure in terms of historical change and institutional
contradiction. The well-being they enjoy, they do not usually impute to the big ups and downs of the societies
in which they live. Seldom aware of the intricate connection between the patterns of their own lives and the
course of world history, ordinary men do not usually know what this connection means for the kinds of men
they are becoming and for the kinds of history-making in which they might take part. They do not possess the
quality of mind essential to grasp the interplay of man and society, of biography and history, of self and world.
They cannot cope with their personal troubles in such ways as to control the structural transformations that
usually lie behind them.
Surely it is no wonder. In what period have so many men been so totally exposed at so fast a pace to such
earthquakes of change? That Americans have not known such catastrophic changes as have the men and
women of other societies is due to historical facts that are now quickly becoming "merely history." The history
that now affects every man is world history.....
The very shaping of history now outpaces the ability of men to orient themselves in accordance with cherished
values....Is it any wonder that ordinary men feel they cannot cope with the larger worlds with which they are so
suddenly confronted? That they cannot understand the meaning of their epoch for their own lives?...Is it any
wonder that they come to be possessed by a sense of the trap?
It is not only information they need--in this Age of Fact, information often dominates their attention and
overwhelms their capacities to assimilate it....What they need, and what they feel they need, is a quality of
mind that will help them to use information and to develop reason in order to achieve lucid summations of
what is going on in the world and of what may be happening within themselves. It is this quality, I am going to
contend, that journalists and scholars, artists and publics, scientists and editors are coming to expect of what
may be called the sociological imagination.
The sociological imagination enables its possessor to understand the larger historical scene in terms of its
meaning for the inner life and the external career of a variety of individuals. It enables him to take into account
how individuals, in the welter of their daily experience, often become falsely conscious of their social
positions. Within that welter, the framework of modern society is sought, and within that framework the
psychologies of a variety of men and women are formulated. By such means the personal uneasiness of individuals is focused upon explicit troubles and the indifference of publics is transformed into involvement
with public issues.
C. Wright Mills, “The Promise [of Sociology]” Excerpt from The Sociological Imagination (originally published in 1959)
The first fruit of this imagination--and the first lesson of the social science that embodies it--is the idea that the
individual can understand his own experience and gauge his own fate only by locating himself within his
period, that he can know his own chances in life only by becoming aware of those of all individuals in his
circumstances. In many ways it is a terrible lesson; in many ways a magnificent one. We do not know the
limits of man's capacities for supreme effort or willing degradation, for agony or glee, for pleasurable brutality
or the sweetness of reason. But in our time we have come to know that the limits of 'human nature' are
frighteningly broad. We have come to know that every individual lives, from one generation to the next, in
some society; that he lives out a biography, and that he lives it out within some historical sequence. By the fact
of his living he contributes, however minutely, to the shaping of this society and to the course of its history,
even as he is made by society and by its historical push and shove.
The sociological imagination enables us to grasp history and biography and the relations between the two
within society. That is its task and its promise. To recognize this task and this promise is the mark of the
classic social analyst. It is characteristic of Herbert Spencer-turgid, polysyllabic, comprehensive; of E. A.
Ross-graceful, muckraking, upright; of Auguste Comte and Emile Durkheim; of the intricate and subtle Karl
Mannheim. It is the quality of all that is intellectually excellent in Karl Marx; it is the clue to Thorstein
Veblen's brilliant and ironic insight, to Joseph Schumpeter's many-sided constructions of reality; it is the basis
of the psychological sweep of W.E.H. Lecky no less than of the profundity and clarity of Max Weber. And it is
the signal of what is best in contemporary studies of man and society.
No social study that does not come back to the problems of biography, of history and of their intersections
within a society has completed its intellectual journey. Whatever the specific problems of the classic social
analysts, however limited or however broad the features of social reality they have examined, those who have
been imaginatively aware of the promise of their work have consistently asked three sorts of questions:
(1) What is the structure of this particular society as a whole? What are its essential components, and how are
they related to one another? How does it differ from other varieties of social order? Within it, what is the
meaning of any particular feature for its continuance and for its change?
(2) Where does this society stand in human history? What are the mechanics by which it is changing? What is
its place within and its meaning for the development of humanity as a whole? How does any particular feature
we are examining affect, and how is it affected by, the historical period in which it moves? And this period-
what are its essential features? How does it differ from other periods? What are its characteristic ways of
history-making?
(3) What varieties of men and women now prevail in this society and in this period? And what varieties are
coming to prevail? In what ways are they selected and formed, liberated and repressed, made sensitive and
blunted? What kinds of 'human nature' are revealed in the conduct and character we observe in this society in
this period? And what is the meaning for 'human nature' of each and every feature of the society we are
examining?
Whether the point of interest is a great power state or a minor literary mood, a family, a prison, a creed-these
are the kinds of questions the best social analysts have asked. They are the intellectual pivots of classic studies
of man in society-and they are the questions inevitably raised by any mind possessing the sociological,
imagination. For that imagination is the capacity to shift from one perspective to another-from the political to
the psychological; from examination of a single family to comparative assessment of the national budgets of
the world; from the theological school to the military establishment; from considerations of an oil industry to
studies of contemporary poetry. It is the capacity to range from the most impersonal and remote
transformations to the most intimate features of the human self and to see the relations between the two. Back
of its use there is always the urge to know the social and historical meaning of the individual in the society and
in the period in which he has his quality and his being.
That, in brief, is why it is by means of the sociological imagination that men now hope to grasp what is going
on in the world, and to understand what is happening in themselves as minute points of the intersections of
C. Wright Mills, “The Promise [of Sociology]” Excerpt from The Sociological Imagination (originally published in 1959)
biography and history within society..... They acquire a new way of thinking, they experience a transvaluation
of values: in a word, by their reflection and by their sensibility, they realize the cultural meaning of the social
sciences.
Perhaps the most fruitful distinction with which the sociological imagination works is between 'the personal
troubles of milieu' and 'the public issues of social structure.' This distinction is an essential tool of the
sociological imagination and a feature of all classic work in social science.
Troubles occur within the character of the individual and within the range of his immediate relations with
others; they have to do with his self and with those limited areas of social life of which he is directly and
personally aware. Accordingly, the statement and the resolution of troubles properly lie within the individual
as a biographical entity and within the scope of his immediate milieu-the social setting that is directly open to
his personal experience and to some extent his willful activity. A trouble is a private matter: values cherished
by an individual are felt by him to be threatened.
Issues have to do with matters that transcend these local environments of the individual and the range of his
inner life. They have to do with the organization of many such milieux into the institutions of an historical
society as a whole, with the ways in which various milieux overlap and interpenetrate to form the larger
structure of social and historical life. An issue is a public matter: some value cherished by publics is felt to be
threatened. Often there is a debate about what that value really is and about what it is that really threatens it.
This debate is often without focus if only because it is the very nature of an issue, unlike even widespread
trouble, that it cannot very well be defined in terms of the immediate and everyday environments of ordinary
men. An issue, in fact, often involves a crisis in institutional arrangements, and often too it involves what
Marxists call 'contradictions' or 'antagonisms.'
In these terms, consider unemployment. When, in a city of 100,000, only one man is unemployed, that is his
personal trouble, and for its relief we properly look to the character of the man, his skills, and his immediate
opportunities. But when in a nation of 50 million employees, 15 million men are unemployed, that is an issue,
and we may not hope to find its solution within the range of opportunities open to any one individual. The very
structure of opportunities has collapsed. Both the correct statement of the problem and the range of possible
solutions require us to consider the economic and political institutions of the society, and not merely the
personal situation and character of a scatter of individuals.
Consider war. The personal problem of war, when it occurs, may be how to survive it or how to die in it with
honor; how to make money out of it; how to climb into the higher safety of the military apparatus; or how to
contribute to the war's termination. In short, according to one's values, to find a set of milieux and within it to
survive the war or make one's death in it meaningful. But the structural issues of war have to do with its
causes; with what types of men it throws up into command; with its effects upon economic and political,
family and religious institutions, with the unorganized irresponsibility of a world of nation-states.
Consider marriage. Inside a marriage a man and a woman may experience personal troubles, but when the
divorce rate during the first four years of marriage is 250 out of every 1,000 attempts, this is an indication of a
structural issue having to do with the institutions of marriage and the family and other institutions that bear
upon them...
What we experience in various and specific milieux, I have noted, is often caused by structural changes.
Accordingly, to understand the changes of many personal milieux we are required to look beyond them. And
the number and variety of such structural changes increase as the institutions within which we live become
more embracing and more intricately connected with one another. To be aware of the idea of social structure
and to use it with sensibility is to be capable of tracing such linkages among a great variety of milieux. To be
able to do that is to possess the sociological imagination.....
12 sources/Peterson and Krivo.pdf
Racial Segregation and Black Urban Homicide Author(s): Ruth D. Peterson and Lauren J. Krivo Source: Social Forces, Vol. 71, No. 4 (Jun., 1993), pp. 1001-1026 Published by: Oxford University Press Stable URL: http://www.jstor.org/stable/2580128 Accessed: 19-06-2016 04:56 UTC
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact [email protected].
Oxford University Press is collaborating with JSTOR to digitize, preserve and extend access to Social Forces
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:56:26 UTC All use subject to http://about.jstor.org/terms
Racial Segregation and Black Urban Homicide*
RUTH D. PETERSON, Ohio State University LAUREN J. KRIVO, Ohio State University
Abstract
Social deprivation and social isolation perspectives provide grounds for expecting residential segregation to increase violent crime among oppressed minorities. Un- fortunately, scholars seldom have included residential segregation in analyses of violent crime. In addition, most research has analyzed the relationship between general rather than race-specific or ethnic-specific crime rates and social and demographic predictors. To address these shortcomings, we examine the impact of racial residential segregation on rates of African-American homicide victimization for large U.S. central cities. The analyses demonstrate that black-white segregation leads to higher rates of black killing although the relationship exists only for stranger and acquaintance homicides. This suggests that social isolation, rather than social deprivation, is the mechanism by which segregation leads to higher levels of homicide among African Americans.
Lethal violence among African Americans is a serious problem in the United States. For example, a 1986 report based on national health statistics revealed that over the 14-year period between 1970 and 1983 the black homicide rate ranged from five to nine times that for whites (Centers for Disease Control Homicide Surveillance 1986). Similarly, O'Carroll and Mercy's (1986) analysis of Federal Bureau of Investigation (FBI) statistics indicates that over the period from 1977 to 1984 the risk of homicide victimization was six times greater for blacks than for whites (see also recent analyses by Hawkins 1986 & Riedel 1984). FBI Supplementary Homicide Report (SHR) data for 1990 indicate that the size of this differential has not diminished in more recent years. For 1990, the ratio of black to white homicide victimization rates was 6.81. Clearly then, in the
* The research reported in this article was partially completed while Ruth Peterson was on research leave at the Department of Sociology, Cleveland State University and Lauren Krivo
was on research leave at the Social WelfareResearch Institute, Boston College. This article was
presented at the annual meetings of the American Sociological Association, Cincinnati, 23-27
August, 1991. We would like to thank William C. Bailey and Robert L. Kaufman for many
extremely helpful comments on this article. Robert J. Wilger generously provided data on segregation indices for central cities. We also gratefully acknowledge valuable comments provided by two anonymous referees. Direct correspondence to Ruth D. Peterson, Department
of Sociology, Ohio State University, Columbus, OH 43210.
i) The University of North Carolina Press Social Forces, June 1993, 71(4):1001-1026
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:56:26 UTC All use subject to http://about.jstor.org/terms
1002 / Social Forces 71:4, June 1993
aggregate, African Americans are victims of lethal violence at a far greater rate
than whites. This substantial racial difference in homicide is a major cause of the lower life expectancy of blacks in the U.S. (Potter 1991).
While the seriousness of the black homicide problem is well recognized, its determinants are not well understood. Among contemporary scholars, a major perspective on black violence focuses upon the criminogenic consequences of the social and economic inequality that African Americans experience in the U.S. (Balkwell 1990; Blau & Blau 1982; Braithwaite 1979). This explanation has its roots in Merton's (1968) social structure and anomie thesis and Coser's (1963)
analysis of deprivation, discrimination, and diffuse aggression. In general, such perspectives link criminal and deviant behavior to the deprivation that certain segments of the population experience when there is a disjuncture between cultural goals (economic success) and structural arrangements (socioeconomic resources). This disjuncture results in feelings of frustration and alienation that are reflected in criminal and deviant patterns.
Blau and Blau (1982) argue that the expression of frustration in the form of violent crime is particularly pronounced when economic inequality is based upon ascriptive characteristics like race. Criminal violence occurs because in democratic societies such as the U.S., access to resources should not be based on ascribed criteria. Ascriptive inequality reinforces ethnic and class differences and engenders pervasive conflict. Due to the weak political power of "have-nots," which accompanies ascriptive inequality, racially and ethnically disadvantaged groups are unable to organize successful collective action (e.g., strikes, boycotts). Instead, the conflict and hostilities engendered by ascriptive inequality find expression in diffuse aggression, including criminal violence.
A relatively large body of recent empirical research has evaluated this argument by examining the relationship between a variety of measures of socioeconomic deprivation and rates of violent crime including homicide. Many analyses demonstrate that one or more measures of general and/or racial socioeconomic inequality are associated with higher homicide rates for the general population (Balkwell 1990; Blau & Blau 1982; Blau & Golden 1986; Blau & Schwartz 1984; Crutchfield 1989; Land, McCall & Cohen 1990; Messner & Golden 1985; Sampson 1986). Studies have also demonstrated that absolute deprivation (ow income or poverty) is linked to higher levels of homicide (Bailey 1984; Huff-Corzine, Corzine & Moore 1986; Loftin & Parker 1985; Messner 1983a, 1983b; Sampson 1985; Williams 1984; Williams & Flewelling 1988).
While these studies are instructive about the determinants of overall homicides, they do not inform us fully about the structural causes of variation in black killings for three primary reasons. First, they do not examine that variation directly. Instead, scholars tend to analyze homicides for the total population rather than for the black population. In so doing, they fail to take into account the contribution of black killings to overall homicide rates, and they disregard the considerable variation in black homicide rates across jurisdic- tions. For example, Supplementary Homicide Reports (SHR) data show that for the period from 1979 to 81 average homicide victimization rates ranged from a low of 6.6 to a high of 62.7 per 100,000 blacks for the sample of 125 large U.S.
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:56:26 UTC All use subject to http://about.jstor.org/terms
Segregation and Black Urban Homicide / 1003
central cities analyzed below. A better understanding of black homicide requires that we examine this variation directly.
Second, many homicide researchers have used general population rather than race-specific measures of social and economic deprivation as explanatory variables. Similar to black homicides, there is substantial variation in the social, economic, and demographic conditions of blacks across communities in the U.S. For example, 1980 census figures for the sample of cities examined in the current study indicate that the percent of black families living in poverty ranged from a low of 8.0% to a high of 36.1% with a mean rate of 26.0%. Also, on average about 15% of employed blacks held professional or managerial positions in 1980, but the range was from a low of 9% to a high of 31%. As a final example, on average for 1980 the median family income of whites exceeded that of blacks by a factor of 1.63. However in some cities, the average income of whites and blacks was almost identical, and in others white income was more than twice that of blacks. (More specifically, the range of the white/black income ratio was 1.1 to 2.18.) By no means then are African Ameri- cans similarly situated in terms of their socioeconomic status in U.S. cities. The question is whether the variation in black homicides is connected systematically to the variation in the social and economic conditions of blacks across com- munities in this society. Few studies have attempted to answer this question (for some exceptions see Harer & Steffensmeier 1992; LaFree, Drass & O'Day 1992; Messner & Golden 1992; Sampson 1985, 1987).
A third reason why past studies have not fully examined structural causes of variation in black killings is that scholars have tended to conceptualize social deprivation in terms of a limited number of economic factors (e.g., poverty, general income inequality & racial income inequality). In so doing, they have ignored potentially important noneconomic aspects of inequality that may influence lethal violence. A critical analysis of the structural determinants of African-American violence requires that such killings be linked directly to measures of the socioeconomic status of African Americans, including important noneconomic aspects of social inequality.
One of the most central and enduring dimensions of racial inequality in the U.S. is the high level of black-white residential segregation found in most urban areas (Massey & Denton 1987, 1988; Van Valey, Roof & Wilcox 1977; Wilger 1988). Residence in segregated black neighborhoods is a fact of life for a large portion of African Americans. Segregation appears to be associated with a number of negative social conditions for blacks, including greater poverty, more physical deterioration, poorer schools, and higher crime rates (e.g., Massey 1990; Massey, Condran & Denton 1987). Despite these possible consequences, studies have been more concerned with describing levels of residential segregation and analyzing its determinants than with studying its impact on other social conditions (Clark 1991, 1992; Lieberson & Carter 1982; Logan & Schneider 1984; Massey & Denton 1987; Stearns & Logan 1986a; Van Valey, Roof & Wilcox 1977). In the present study, we attempt to extend our understanding of both the consequences of black-white segregation in the U.S., and the structural determinants of African-American killings. We do so by analyzing the impact of racial residential segregation on the incidence of black homicide victimization in large U.S. central cities.
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:56:26 UTC All use subject to http://about.jstor.org/terms
1004 / Social Forces 71:4, June 1993
Conceptual Argument
Residential segregation between blacks and whites is a pervasive aspect of racial inequality in the U.S. Numerous studies report high levels of racial residential segregation in U.S. cities and suburbs that cannot be explained by black-white income differentials (Denton & Massey 1988; Farley 1977; Massey 1979; Simkus 1978). Rather, prejudice and discrimination in the housing market produce residential environments that are separated largely along racial lines (Galster 1987; Galster, Freiberg & Houk 1987; Lake 1981; Pearce 1979; Schuman & Bobo
1988; Turner, Struyk & Yinger 1991; Wienk et al. 1979; Yinger 1986). A growing body of research suggests that these separate black and white communities provide markedly different social environments for their residents. Higher levels
of unemployment, welfare dependency, dilapidated housing, mortality, unwed motherhood, and crime pervade segregated African-American neighborhoods (Hogan & Kitagawa 1985; LaVeist 1989,1992; Massey 1990; Massey, Condran &
Denton 1987). Similarly, poorer schools and higher taxes plague predominantly black areas - the types of areas where a large portion of blacks reside (Massey 1990; Massey, Condran & Denton 1987; Schneider & Logan 1982). Thus, black-white residential segregation means not only that African Americans live in separate communities, but that they live in areas that are inferior to neighbor- hoods more heavily populated by whites.
Despite indirect evidence that residential segregation is linked closely with inferior community outcomes, few studies have examined directly the influence of this dimension of social stratification on social conditions for the African- American population (for exceptions, see Jiobu & Marshall 1971; LaVeist 1989, 1992; Logan & Messner 1987). It is particularly unfortunate that residential segregation has seldom been considered in studying such important social problems as lethal violence among blacks. If, as Blau and Blau (1982) maintain, ascriptive inequality has a particularly strong impact on violent crime, then the results of most previous analyses of homicide may be biased since they do not include one of the most central components of ascriptive inequality - racial residential segregation.
Drawing on Blau and Blau, Logan and Messner (1987) outline some of the mechanisms by which racial residential segregation could have a critical influence on homicide. As they note:
racial segregation imposes a significant barrier to black upward mobility and quality of life. Place of residence locates people not only in geographical space but also in networks of social opportunities - it influences prospects for employment, for public services, for educational advancement, for appreciation in home values, and more. Residential segregation by race accordingly implies that opportunities for achievement are limited for certain groups, and it conflicts with basic American value commitments which encourage members of all groups to strive for socioeconomic success. Such a "disjuncture" between structural arrangements and fundamental cultural values, Merton argued, tends to
undermine the legitimacy of social norms and thereby promotes deviant behavior. (510)
Logan and Messner (1987) argue that high levels of racial segregation should result in feelings of greater resentment, frustration, hopelessness, and alienation among African Americans, which in turn would be reflected in diffuse forms of aggression, including high rates of criminal violence.
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:56:26 UTC All use subject to http://about.jstor.org/terms
Segregation and Black Urban Homicide / 1005
Wilson (1987) suggests another mechanism by which residential segregation may be linked to violent crime. Given high rates of black poverty in most central cities (higher than the highest white poverty rates in most cities), black-white residential segregation is inextricably linked with concentrated black poverty (Massey 1990; Massey & Eggers 1990). Thus, residential segregation implies isolation from mainstream society, an isolation that ties blacks into a local setting of multiple disadvantages (Anderson 1990; Wilson 1987). These disadvantages are reinforced by the impact of racial residential segregation on types and levels of community social control. At a formal level, individuals in more heavily poor and predominantly black neighborhoods may be unable to obtain adequate police protection. At an informal level, segregated black communities may lack the internal resources to organize neighborhood structures to help prevent crime (e.g., crime-tip hot lines and other reporting projects, home security surveys, volunteer patrol organizations, and neighbor- hood crime watches).' These communities simply lack the monetary, social, and institutional resources to combat crime. At the same time, the condition of anomie implies that the legitimacy, and therefore the effectiveness, of social norms are undermined, which in turn increases the likelihood of deviant behavior.
There are very few studies of the effects of racial residential segregation on violent crime (Logan & Messner 1987; Messner & South 1986; Potter 1991; Rosenfeld 1986; Sampson 1985; South & Felson 1990).2 Among these, Logan and Messner (1987), Rosenfeld (1986), and Sampson (1985) have found that higher residential segregation is associated with higher levels of murder. However, these effects are not always statistically significant. Logan and Messner (1987) found a significant relationship between segregation and criminal homicide in the suburban rings of 54 metropolitan areas for 1980 but not 1970. In contrast, Rosenfeld (1986) found that residential segregation had a significant positive effect on murder rates in standard metropolitan statistical areas (SMSAs) for 1970. Using race-specific homicide arrest rates for cities, Sampson (1985) also showed that residential segregation significantly influenced white but not black homicides for 1970. In a related vein, Potter (1991) found that residential isolation had a sizable significant effect on the difference in homicide between blacks and whites in 27 metropolitan areas for 1980.
Messner and South (1986) and South and Felson (1990) considered the effect of residential segregation on violent crimes in evaluating Blau's macrostructural theory of intergroup contact. According to this perspective, residential segrega- tion should be associated negatively with interracial victimization and positively with intraracial victimization. This pattern of association would occur because segregation reduces the opportunity for interracial contacts and increases the opportunity for intraracial contacts. Both studies found the expected pattern for robbery and rape.
Although these investigations indicate the importance of residential segregation for violent crime, they fall short of fully illuminating the influence of segregation on black homicide. In general, the studies are few; their results are inconsistent; and rarely are they concerned with race-specific crime rates. Also, individual studies have some specific limitations. For example, Logan and Messner (1987) analyze suburban crime patterns for the total population
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:56:26 UTC All use subject to http://about.jstor.org/terms
1006 / Social Forces 71:4, June 1993
although the majority of African Americans reside in central cities where most street crime occurs. Potter (1991) examines the black-white homicide gap rather than the levels of race-specific homicide and does so for a small sample of metropolitan areas (27 places). Sampson (1985) provides a direct test of the effect of segregation on black homicide. However, his analysis is based upon arrest data which raises questions about his conclusions. Reported crimes do not always result in arrests, and all arrested persons are not associated unequivocal- ly with crimes.
In brief, there are important theoretical reasons to expect that racial residential segregation should contribute to higher homicide rates among disadvantaged minorities. Nonetheless, the role of this factor has not been given due attention in the empirical literature. As a result, we have little understan- ding of the merits of theoretical arguments regarding segregation and violent crime for African Americans. The analyses that follow address this important question, and thereby contribute to our understanding of the consequences of segregation and the structural determinants of the killings of blacks.
Data and Methods
Our analysis examines the influence of black-white residential segregation and other dimensions of African Americans' socioeconomic status on rates of black homicide victimization in large cities for 1980. The sample includes the central cities of SMSAs where there was a 1980 central city population of at least 100,000 and a black population of at least 5,000 (N=125).3
DEPENDENT MEASURES
Average black homicide (murder and non-negligent manslaughter) victimization rates per 100,000 blacks are computed for the period from 1979 to 1981 for each city. In addition to examining the overall black homicide rate, we disaggregate killings and compute rates for different types of killing based upon the relationship between the victim and the offender: family, acquaintance, or stranger. All homicide rates are based on data from the FBI's Supplementary Homicide Reports (SHR). These data include homicides known to the police in jurisdictions participating in the SHR Program. SHR data are slightly less complete than the FBI's Uniform Crime Reports (UCR) data, the most common source used for this type of analysis. However, general homicide rates based upon the two sources are correlated very highly for the 125 cities considered here (r >.90). SHR data are used because the UCR do not report the race of either the victims or offenders in homicide incidents. We use victimization rates rather than offending rates because reporting of the race of homicide victims is nearly complete in the SHR files, while the race of the offender is missing for a large number of SHR homicide incidents.
In calculating rates, we restrict our analysis to homicide incidents in which there was a single victim and single offender. Williams and Flewelling (1987) note several reasons why such a restriction is necessary. First, they point out that the "situational circumstances of events involving multiple offenders and/or victims tend to be different from one-on-one event[s]" (546). Therefore, the determinants of multiple-offender/multiple-victim homicides may be
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:56:26 UTC All use subject to http://about.jstor.org/terms
Segregation and Black Urban Homicide / 1007
different from those involving a single victim and a single offender. Until we have improved knowledge of the similarities and differences between the two kinds of killing, Williams and Flewelling recommend that we keep analyses of them separate. Second, they note that multiple offender or victim killings are relatively rare (approximately 16% of incidents in which there was a known offender during 1979 to 1981) and that an understanding of homicides (African- American homicides in this context) should begin with the most common type, that is, one-on-one killings. Third, and of particular relevance for our analysis, Williams and Flewelling note that certain practical considerations preclude examining multiple-offender/multiple-victim homicides when researchers are interested in the determinants of different types of killing. Below we will examine rates for family, acquaintance, and stranger homicides. Unfortunately, in multiple victim situations, the SHR provide information only on the relationship between the first victim and each offender. Therefore, the victim- offender relationship cannot be determined for second and subsequent victims. Moreover, considerable ambiguity would result from attempts to classify victim/offender relationships when events involved multiple kinds of relation- ships. Williams and Flewelling (1987) provide the following example.
Imagine a situation in which a husband, along with two of his friends, kills his wife and their three children. The killing of the wife and children would be family victimizations only if linked to the husband as the offender. These killings would be acquaintance or stranger victimizations, however, if linked to the husband's two friends, depending upon whether the wife and children knew them. (546)
Considering single-offender/single-victim killings avoids such ambiguities and permits the calculation of rates for different types of relationships.
INDEPENDENT VARIABLES
The explanatory variables, their operationalizations, and the predicted directions of effect on homicides are presented in Table 1. Two measures of segregation are examined: (1) the index of dissimilarity (D), and (2) the correlation ratio
(92). The index of dissimilarity is a widely used indicator of evenness in the distribution of two groups (in our case blacks and whites) across census tracts within an urban area. Values of D range from a minimum of 0 when blacks and whites are distributed evenly across tracts to a maximum of 100 when blacks and whites are segregated completely. Index values between 0 and 100 indicate the percent of blacks (or whites) that would have to change their tract of residence to achieve perfect integration (i.e., a D-value of 0).
The correlation ratio (92) is a measure of racial concentration (Stearns & Logan 1986b). It is included in this analysis to capture better our theoretical interest in segregation as an indicator of the concentration of blacks in neighbor- hoods that are socially isolated from more heavily white communities. That r2 indicates this exact aspect of segregation is illustrated by Stearns and Logan's
statement that 92 iS "an aggregate measure of the polarization of neighborhoods toward all-black and all-white" (1986b:128). The correlation ratio (g) ranges from 0 to 1. For a given city, it approaches the maximum of 1 when most tracts are nearly all black or nearly all white.
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:56:26 UTC All use subject to http://about.jstor.org/terms
1008 / Social Forces 71:4, June 1993
TABLE 1: Predictor Variables of Black Homicides: Operationalizations and Expected Directions of Effect
Variables Operationalization Expected Direction of
Effect Segregation
Dissimilarity (D) Index of dissimilarity of black-white residential segregation across census tracts +
Racial concentration (q2) The correlation ratio of black-white
residential segregation across census tracts +
Black socioeconomic status
Black income inequality Gini index of family income inequality among blacks +
Black-white income Ratio of white to black median family inequality income +
Poverty Percent of black families below the federal poverty line +
Education Percent of black population older than 24 that graduated high school
Professionals Percent of employed blacks in professional and managerial occupations
Demographic factors
Percent black Percent of the total population that is black +
Percent of black males, aged 15-34 Percent of black males aged 15 to 34 +
Divorce rate Percent of black population older than 15 that is currently divorced or separated +
Population Natural log of the total city population +
Region South - 1, Non-South- 0 +
Measures of the remaining independent variables indicating blacks' socioeconomic status and demographic composition are straightforward. These variables are similar to those used in many studies of crime and inequality except that most are based upon characteristics for the African American population rather than for the general population. Economic inequality is measured by the gini index of income inequality among blacks, and the black-white income ratio. The gini index uses other blacks as the reference in considering the role of income inequality, while the income ratio compares blacks with whites. Although some studies use the black-white income gap as
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:56:26 UTC All use subject to http://about.jstor.org/terms
Segregation and Black Urban Homicide / 1009
a measure of interracial inequality, we use the income ratio to control for differences across cities in average income. The black family poverty rate, percent high school graduates, and percent of employed blacks in professional
and managerial occupations are included as measures of the level of socio- economic status within the black population. Several demographic factors are considered as control variables because they have been found in previous research to influence general homicide rates: percent black population, percent of black males aged 15-34, percent of blacks divorced, log of population size, and region (South/non-South).
Data for all independent variables except segregation were obtained from the U.S. Bureau of the Census reports for 1980. The segregation indices were obtained from Wilger (1988), Massey and Denton (1988), and Hwang and Murdock (1982). Most of the values of D come from Wilger (1988) although a small portion which are not available from Wilger (central cities of metropolitan areas of less than 4% black) are from Massey and Denton (1988) or Hwang and Murdock (1982). Of note, the indices from Wilger include Hispanics in the counts of whites and blacks within tracts while those from the other sources
exclude Hispanics.4 Values of q2 are available for only 111 of the 125 cities in the full sample because the data from which q2 values were calculated are available only in Wilger (1988).
SrATIST?CAL ANALYSIS
The total black homicide victimization rate is regressed on black-white residential segregation and the other independent and control variables using ordinary least squares (OLS) techniques. However, as Williams and Flewelling (1988) note, use of an aggregate rate "can mask or imprecisely reveal empirical relationships indicative of a differential causal process operating in the social production of criminal homicide" (422). Hence, to further understand the causal processes at work in black homicide, we disaggregate killings and compute
rates for three subtypes based upon the relationship between the offender and the victim: family, acquaintance, and stranger.5
Family, acquaintance, and stranger homicides differ from one another in important ways. Williams and Flewelling (1988) note that family and acquain- tance homicides occur between people with existing relationships during their routine everyday interactions. In contrast, homicides involving strangers take place in more impersonal situations among individuals with no prior contact or relationship. They include such felony homicides as robbery killings. Structural conditions might have varying effects by homicide type because of the differing relationships and situations under which these homicides occur. For example, we might anticipate increasingly weaker effects of our structural variables on homicides as we move from stranger to acquaintance to family killings. This would occur because homicides involving individuals in more primary relationships may be more responsive to a variety of interpersonal and situational factors (Luckenbill 1977). This possibility was illustrated in Williams and Flewelling's (1988) analysis of 1980 general homicide rates disaggregated by the victim/offender relationship and the precipitating incident (conflict or other). The researchers found that the strength of the effects of percent poor
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:56:26 UTC All use subject to http://about.jstor.org/terms
1010 / Social Forces 71:4, June 1993
(and divorce rate) are generally weaker for homicides involving family members than for those involving acquaintances or strangers. By contrast, Smith and Parker (1980) found that their resource measure (the structural-poverty index) had a significant influence on primary (e.g., killings of family members and lovers) but not nonprimary (e.g., gangland slayings and felony murders) homicides for 1970. Methodological differences between the two studies may account for the differential patterns of influence of the resource deprivation measures. For our purposes what is important is the differential effect of these measures across different types of homicide. The variable effects of structural conditions on different types of general homicide also have been demonstrated in other empirical work (Parker 1989; Parker & Smith 1979; Riedel 1987; Zahn & Sagi 1987).
Analyses that disaggregate black homicides by the type of relationship between the victim and the offender (i.e., family, acquaintance, or stranger) are estimated using generalized least squares (GLS) techniques. This makes it possible to perform significance tests of differences in coefficients for the explanatory variables. The model used is a Seemingly Unrelated Regression also referred to as a joint GLS model (Theil 1971). This estimation technique is used because the same cases and independent variables are included in each of the relationship-based homicide equations. To test for the differences in the coefficients across these equations, the standard t-test from OLS results assumes that the covariance of the coefficients is 0. However, this assumption is violated when the cases and values of independent variables are all the same. The GLS model used here corrects for this violation and permits us to calculate the appropriate t-values for tests of significance of the difference in the effects of predictor variables on different types of homicide.
Results
BIVARIATE RELATIONSHIPS
Table 2 presents the zero-order correlation matrix of all variables along with their means and standard deviations.6 These correlations show that almost all of the theoretical variables have the expected bivariate relationships with total and different types of black homicide. Higher levels of relative and absolute deprivation are associated with higher levels of homicide with one exception - black income inequality and family killings. Turning to the control variables, the coefficients for percent black population and region are related consistently to homicide in the expected direction. Southern communities and those with a higher percentage of blacks have higher homicide rates. However, for the remaining demographic factors, the associations are not always consistent with expectations. Of particular note is the pattern for the age distribution variable; a greater percentage of black males in the more crime-prone ages is related to lower homicide rates for total and all types of killing. The multivariate analyses will assess whether this pattern holds when other variables are controlled. The coefficients for the remaining two demographic variables are inconsistent across homicide types.
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:56:26 UTC All use subject to http://about.jstor.org/terms
Segregation and Black Urban Homicide / 1011
For the purpose of this article, the most important finding in the correlation matrix is the relatively strong positive association between residential segrega- tion and black homicide. For all homicide types except those involving acquaintances, the index of dissimilarity has the strongest or second strongest
bivariate association with homicide than any other explanatory variable. Additionally, the bivariate correlations differ in strength across the three types
of homicide, ranging from .189 for family killings to .315 for acquaintance killings. This provides some initial empirical support for disaggregating black homicides into these types.
The bivariate relationships between homicide and racial concentration (r2) (not reported here) show a similar pattern to those found for D and black killings. When examining homicides involving strangers, r2 has a stronger bivariate association with black killings than any other variable (.302). For total, family, and acquaintance homicide rates, racial concentration has the second or third strongest relationship with the homicide victimization of African Amer- icans.
Table 2 also shows that among the independent variables there are several sizable correlations, including those between poverty and black income inequality (r=.787) and poverty and educational level (r=-.762). Such strong bivariate correlations suggest the possibility of problems of serious multi- collinearity. However, examination of changes in the standard errors of coefficients in models that remove and then re-enter the highly correlated variables showed no substantial changes.7 As such, the multivariate results presented below do not appear to be influenced by multicollinearity.
Black Homicide and Resideptial Dissimilarity
TOTAL BLACK HOMICIDE RATES
Table 3 presents the results of the OLS model examining the effects of black- white residential segregation, as measured by the index of dissimilarity, and the socioeconomic status measures on total black homicide rates controlling for other structural characteristics of central cities. Beginning with the demographic variables, the results show that only one of these factors is a significant predictor of black homicide. In line with considerable past research on total homicide rates, percent black has a significant effect on black killings. However, in contrast to the results of studies of general homicides, a larger black population is associated with a lower black homicide rate. This finding directly contradicts the common subcultural interpretation that suggests a larger black population strengthens a black violent subculture and thereby increases the rate of killing. In fact, just the opposite occurs for blacks in central cities. Previous studies of race-specific homicide rates also contradict the subcultural argument by demonstrating either a negative or nonsignificant influence of percent black (or nonwhite) on black killings (Allen, McSeveney & Bankston 1981; Sampson 1985, 1987). The other control factors in our model do not play a role in explaining black homicides. Apparently, the social disorganization implied by higher marital dissolution and the cultural effects often imputed to region do not operate among African Americans.
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:56:26 UTC All use subject to http://about.jstor.org/terms
1012 / Social Forces 71:4, June 1993
TABLE 2: Correlations, Means, and Standard Deviations of Dependent and Independent Variables
Total Family Acquaint. Stranger Segrega- Black Black- Homicide Homicide Homicide Homicide tion Income White
(D) Inequal. Inequal.
Family homicide rate .576
Acquaintance homicide rate .818 .308
Stranger homicide rate .479 .393 .237
Segregation (D) .370 .189 .315 .299
Black income
inequal. .180 -.092 .266 .061 .409 Black-white
income inequal. .226 .058 .221 .087 .242 .422 Poverty .209 .005 .289 .148 .435 .787 .455
Education -.249 -.052 -.382 -.195 -.494 -.546 -.453
Professionals -.324 -.181 -.328 -.285 -.446 -.318 -.125
Percent black .050 .105 .064 .106 .423 .357 285
Percent blacks males 15-34 -.145 -.116 -.188 -.013 -.530 -.367 -.136
Divorce rate .178 -.016 .122 .006 .286 .323 -.092
Population (log) .130 .144 -.066 .182 .374 .091 -.070
Region .161 .223 .199 .169 .224 .026 .478
Mean 27.014 5.381 16.529 2.979 67.814 .419 1.630 Std. dev. 10.228 3.270 7.366 2.710 11.313 .024 236
Turning to the variables of theoretical interest, only one indicator of socio- economic status (occupational composition) is an important determinant of African-American killings. Of particular, note when the analysis is performed for black homicides using race-specific explanatory variables, neither black income inequality nor black-white income inequality is a significant predictor of black killings. As noted, occupational composition is related significantly to homicide, but neither education nor poverty is significant. Further, occupational composition is the most important contributor to black killings. This finding suggests that increased access to professional, managerial, and other supervisory positions has a particularly salient effect on reducing homicide rates for the African-American population.
Most central to the argument of this article is the finding that racial residential segregation (D) has a strong significant positive effect on black homicide rates. More specifically, a 10-point increase in the index of dis- similarity (about a one standard deviation increase in this segregation index) is
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:56:26 UTC All use subject to http://about.jstor.org/terms
Segregation and Black Urban Homicide / 1013
TABLE 2: Correlations, Means, and Standard Deviations of Dependent and Independent Variables (Continued)
Poverty Educ. Prof. Percent Blacks Divorce Pop. Region Black Males Rate
Family 15-34 homicide rate
Acquaintance homicide rate
Stranger homicide rate
Segregation (D)
Black income inequal.
Black-white income inequal.
Poverty
Education -.762
Professionals -.515 .660
Percent black .319 -.467 -.273
Percent black males 15-34 -.579 .627 .507 -.328
Divorce rate .189 -.055 -.257 .012 -.386
Population (log) -.084 .163 .113 .123 -.006 .276
Region .184 -.375 -.049 .325 -.026 -.493 -.084
Mean 25.978 55.354 14.595 24.170 18.197 17.067 12.471 .424 Std. dev. 5.235 9.172 3.388 16.140 3.106 1.970 .785 .496
associated with a 2.4-person increase in the number of homicides per 100,000 blacks. Further, black-white residential dissimilarity has a relative effect on black homicides that is nearly identical to that for occupational composition
(P=.27). It is particularly striking that segregation has more influence than measures (the gini index, the white/black income ratio) of some of the most central theoretical constructs (anomie, relative economic deprivation) in discussions of crime. Therefore, omission of this important dimension of racial inequality from past research may have undermined our ability to understand how inequality influences violent crime.
A DISAGGREGATED ANALYSIS
To explore how structural conditions may vary in their effects on different types of homicide, we estimated models distinguishing among killings of family members, acquaintances, and strangers. The GLS results of these analyses are presented in Table 4. Table 5 reports the t-values for tests of the difference in the effects of the predictor variables between types of homicide.
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:56:26 UTC All use subject to http://about.jstor.org/terms
1014 / Social Forces 71:4, June 1993
TABLE 3: OLS Regression of Black Homicides Rates on Segregation (D), Black Socioeconomic Status, and Demographic Factors: Central Cities, 1980
Unstandardized Standard Standardized Coefficient Error Coefficient
Independent variable
Segregation (D) .244** .112 .702 Black income inequalitya -.001 .069 -.002 Black-white
income inequality 6.457 4.654 .149 Poverty -.085 .369 -.044 Education -.023 .201 -.021 Professionals -.843* .370 -.279 Percent black -.123* .064 -.194 Percent of black males, .531 .435 .161
aged 15-34 Divorce rate .973 .618 .187 Population (logged) .780 1.324 .060 Region 3.720 2.537 .180
Constant -18.596 R2 .257
(N=125)
a The coefficients and standard errors of the gini index of black income inequality are multiplied by 1,000 because the gini index varies only from 0 to 1 and hence produces extremely small parameters.
*p<.05 **p<.01
These analyses show that each type of homicide is influenced significantly by one or more of the structural factors (see Table 4). Family homicides are higher in central cities in the South, and in central cities where there are relatively fewer black professionals and more black high school graduates. Acquaintance killings are more likely in communities where segregation (D) is higher, where fewer blacks have a high school education, where the percent black is relatively smaller and in the South. Homicides involving strangers are influenced by segregation (D), percent of black professionals, percent of black males aged 15 to 34, and population size.
Judging by the R2 values, Table 4 also shows that the set of structural variables provides a better fit moving from family homicides to acquaintance and stranger homicides.8 This pattern is consistent with some previous research (Loftin & Parker 1985; Williams & Flewelling 1988) showing that structural factors are less influential in killings involving family members. Beyond this however, there are few clear patterns in the relationships across the three types of killing. None of the individual explanatory factors is significant for all three types. Nor does the strength of the association for significant variables increase
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:56:26 UTC All use subject to http://about.jstor.org/terms
Segregation and Black Urban Homicide / 1015
TABLE 4: GLS Regression of Types of Black Homicides Rates on Segregation (D), Black Socioeconomic Status, and Demographic Factors: Central Cities, 1980
Family Acquaintance Stranger Homicide Homicide Homicide b p b b p (S.E.) (S.E.) (S.E.)
Independent variable
Segregation ()) .006 .022 .154* .236 .057* .238
(.036) (.077) (.029)
Black income inequalitya -.025 -.184 .059 .191 -.028 -.244
(.022) (.047) (.018)
Black-white income .756 .055 .176 .006 -.297 -.026
inequality (1.487) (3.181) (1.195)
Poverty .045 .072 -.274 -.195 .134 .260
(.118) (.252) (.095)
Education .130* .365 -.267* -.332 -.023 -.078
(.064) (.137) (.052)
Professionals -.328** -.340 -.253 -.116 -.222* -.278
(.118) (.253) (.095)
Percent black .006 .029 -.105* -.230 -.005 -.029
(.020) (.044) (.016)
Percent of black males, -.167 -.158 .383 .161 .316** .362
aged 15-34 (.139) (.297) (.112)
Divorce rate .012 .007 .671 .179 .031 .022
(.197) (.422) (.159)
Population (logged) .670 .161 -1.163 -.124 .646* .187
(.423) (.905) (.339)
Region 2.026** .307 2.982* .201 .486 .089
(.810) (1.734) (.651)
Constant 4.138 4.059 -2.203 R2 .180 .260 .228
(N=125)
a The coefficients and standard errors of the gini index of black income inequality are multiplied by 1,000 because the gini index varies only from 0 to 1 and hence produces extremely small parameters.
* p <.05 **p <.01
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:56:26 UTC All use subject to http://about.jstor.org/terms
1016 / Social Forces 71:4, June 1993
TABLE 5: Tests of the Significance of the Difference in Coefficients between Homicide Types
Family vs. Acquaintance Family vs. Acquaintance vs. Stranger Stranger
Independent variable t t t
Segregation (D) -2.026* 1.237 -1.353 Black income inequality -1.890* 1.812 .113 Black-white income inequality .192 .145 .675 Poverty 1.331 -1.586 -.722 Education 3.043** -1.741* 2.273* Professionals -.310 -.120 -.851 Percent black 2.658** -2.236* .489 Percent of black males, -1.945* .219 -3.314**
aged 15-34 Divorce rate -1.642 1.485 -.089 Population (logged) 2.130* -1.957* .054 Region -.580 1.410 1.812*
*p <.05 **p <.01
(or decrease) monotonically when moving from one type of homicide to another. Moreover, different types of homicide cannot be explained in terms of distinct types of structural factors, i.e., segregation, black socioeconomic status, or demographic characteristics. Rather, the three types of killing simply are affected by discrete individual variables. In sum, the findings of Table 4 indicate that the analysis of aggregate homicide rates obscures the varying influence of structural conditions on black homicides; but the results do not reveal a pattern that leads to clear theoretical conclusions.
Yet, we gain insight by examining the significance of the coefficients for the structural variables within types of homicide, and the significance of the difference in effects between homicide types. First, the disaggregated analysis shows that more of the demographic factors influence the specific types of homicide than was apparent in the analysis of the total rate (see Table 3). Location in the South positively influences killings involving family members. Size of the black population and region affect killings involving acquaintances, and percent of young black males and size of the central city population influence killings involving strangers. A larger black population significantly reduces the level of acquaintance homicide; and this effect is significantly different from the negligible influence of this variable on family and stranger killings. Moreover, as is the case for the total homicide rate, the negative effect for percent black directly contradicts the common subcultural interpretation attributed to this variable. This finding is particularly noteworthy since a violent black subculture should have very pronounced effects on killings involving acquaintances. By contrast, the coefficients for region suggest that there may be a southern cultural effect for family and acquaintance killings.
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:56:26 UTC All use subject to http://about.jstor.org/terms
Segregation and Black Urban Homicide / 1017
An additional difference across homicide types is that a greater percent of young black males is related to more stranger and acquaintance homicides but not family homicides. The impact of age reaches statistical significance only for stranger killings although the age effect for both stranger and acquaintance homicides is significantly greater than that for family killings. This finding makes sense in that the circumstances leading to family killings may be less age restricted (e.g., lover's triangles and arguments over money). In contrast, young males likely comprise proportionately more of the people in contexts in which acquaintance and stranger homicides occur (e.g., bars and street corners) (Collins, Cox & Langan 1987).
The only other noteworthy differences shown in Tables 4 and 5 are those for the variable of primary theoretical interest here - segregation (D). First, the disaggregated analysis makes clear that residential dissimilarity has a significant effect only for acquaintance and stranger killings. Family homicides are not affected by this variable. Further, the influence of segregation is relatively strong -the second strongest predictor of acquaintance killings and the fifth most important factor (of 11) in homicides involving strangers. Of note, there is very little differentiation among the standardized coefficients for segregation, black income inequality, black-white inequality, and education in the model for stranger killings. The magnitude of the unstandardized coefficients illustrates the importance of segregation's effect. A 10-unit increase in the index of dissimilarity (approximately a one-standard deviation increase) is related to an increase of 1.5 and .6 homicides per 100,000 blacks for killings involving acquaintances and strangers, respectively. Such increases are relatively large given mean rates of 16.5 for acquaintance homicides and 3.0 for stranger killings.
One possible explanation for a segregation effect on homicides involving acquaintances and strangers, but not family members, is that segregation is more indicative of social isolation (Wilson 1987) than of relative socioeconomic deprivation (Blau & Blau 1982). To the extent that the most segregated black areas are subject to social isolation and weak social control, this would have an effect largely on acquaintance and stranger killings which tend to occur in more public settings (e.g., bars, street corners, and businesses). Slow or inadequate response by police and emergency medical services, absence of neighborhood crime prevention structures, and unwillingness of bystanders to intervene in conflicts (personally or by calling formal authorities) all would increase the likelihood that conflicts in public places would have a lethal outcome.9
However, indicators of social isolation should have much less influence on violence between family members. Such violence more commonly occurs in the home or another private setting where violent encounters are particularly sheltered from, and therefore not constrained or encouraged by, levels of intervention by police, emergency medical authorities, or other third parties. That is, family killings may be determined more by interpersonal and situa- tional precipitants than by external agents of control.
While the influence of residential dissimilarity is greater for killings involving less intimate parties, the effect of this variable is especially pro- nounced for acquaintance homicides. The reason for segregation having a greater influence on acquaintance than on stranger killings is not clear. Perhaps,
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:56:26 UTC All use subject to http://about.jstor.org/terms
1018 / Social Forces 71:4, June 1993
even in highly segregated contexts in which many aspects of social control are weak, formal mechanisms may decrease stranger violence but not limit violent encounters among individuals that know one another. Thus, police may perceive encounters between strangers as more serious and hence respond more quickly, offer greater assistance, and be less likely to discount such violence than when these encounters involve acquaintances. In other words, variation in segregation may be a better indicator of variation in social control for acquain- tance encounters and hence this factor has a stronger effect on this type of killing.
BLACK HOMICIDE AND RACIAL CONCENTRATION
Additional analyses were performed to examine the effects on homicide of the correlation ratio (q2) as an alternative measure of residential segregation. As noted above, we consider 2 because this measure may tap better our theore- tical interest in the social isolation associated with racially polarized black and
white neighborhoods. The findings from these regressions are very similar to those presented in Tables 3 and 4.10 Therefore, the results of these models are not presented in tabular form, and our discussion focuses only on the findings
for r,2. (Tables are available upon request.) The results for the correlation ratio provide further support for the influence
of segregation on black homicide. Like the previous results, the r2 analysis also underscores the importance of disaggregating homicides by type in assessing segregation's influence. For example, as with residential dissimilarity, racial concentration has a significant positive association with both homicide by acquaintances (b=6.841) and by strangers (b=3.537). Also consistent with the
previous analysis, the correlation ratio (q2) has a small and nonsignificant relationship with family killings (b=.720). Further, examination of the magnitude of the unstandardized coefficients for the correlation ratio leads to remarkably similar conclusions about the absolute size of the effect of segregation on black homicides as that found for the index of dissimilarity: an increase of one standard deviation in racial concentration is related to a 1.2-person increase in the number of acquaintance homicides per 100,000 blacks and a .6-person increase per 100,000 blacks in stranger killings.11 Recall from the results presented in Table 4 that an increase of one standard deviation in the index of dissimilarity (10 points) is related to a 1.5-person increase per 100,000 blacks in the number of acquaintance and stranger homicides, respectively. These coefficients are sizable when considered in light of average black homicide rates for the 111 cities analyzed here of 17.1 for killings by acquaintances and 3.1 for killings by strangers.
Conclusion
A sizable body of past research on homicide has examined the impact of economic inequality on general homicide rates. The premise of this study is that the tendency to focus on economic deprivation and total homicides limits our understanding of the relationship between inequality and lethal violence in two ways. First, it ignores a central theme regarding how inequality influences
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:56:26 UTC All use subject to http://about.jstor.org/terms
Segregation and Black Urban Homicide / 1019
homicide. Anomie and social isolation perspectives imply that the criminal consequences of inequality are most likely for oppressed minorities and other ascriptively disadvantaged groups. But most prior research has analyzed total rather than group-specific crime rates, and examined general rather than group- specific social and demographic predictors.
Second, prior research has overlooked the empirical consequences of
excluding a major component of inequality in examining crime. It is unlikely that the effect of deprivation on homicide is limited to economic factors. Thus, the exclusion of important noneconomic dimensions of inequality, particularly residential segregation, has limited our understanding of the effect of inequality on crime, and may have resulted in biased conclusions about the criminogenic consequences of economic deprivation. Our analyses sought to address these limitations of past research.
Indeed, our findings underscore that to better understand black lethal violence it is important to: (1) analyze racially disaggregated homicide rates, (2) differentiate types of killing by the victim-offender relationship, and (3) include segregation in models of homicide.
First, our analysis confirms that, like general homicides, African-American killings are associated with a number of structural characteristics. However, black homicides do not have the same structural precipitants as general rates of killing. In models of black homicide that include segregation and measures of blacks' socioeconomic status, many of the structural factors examined in studies of general homicide do not influence black killings, or do so in a manner that is inconsistent with prior research for the total population. Specifically, with the exception of region (South), the demographic factors considered here have weak effects on black homicide. This indicates that the social disorganization, cultural or other crime-producing effects implied by the age composition, divorce rate, size of the city population, and percent black do not operate (or operate differently) within the black population. While in sharp contrast with the consistent effect of demographic factors found in the large body of research on general homicide rates, this negligible role for demographic characteristics is not uncommon in research that considers disaggregated or race-specific homicide rates (Allen, McSeveney & Bankston 1981; Sampson 1985, 1987).
Our analysis also demonstrates that poverty and income inequality
(intraracial and interracial) do not influence black killings. These findings are especially noteworthy in light of prominent arguments concerning the relation- ship between resource deprivation and crime. Perhaps income inequality is not critical for African Americans in assessing their economic well-being given that absolute economic deprivation is generally very high in the black population. The high level of absolute deprivation for African Americans may also explain why the poverty rate has no effect; it may be that poverty is so widespread among blacks that variation across communities is not especially meaningful for predicting homicide rates. Although these interpretations are speculative, our findings make it clear that the underlying structural determinants of black killings are not identical to those for general homicides.
Second, our analyses of homicides disaggregated by type of victim-offender relationship illustrate that examining aggregate patterns of homicide indeed "mask . . . empirical relationships indicative of a differential causal process
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:56:26 UTC All use subject to http://about.jstor.org/terms
1020 / Social Forces 71:4, June 1993
operating in the social production of [black] homicide" (Williams & Flewelling 1988:422). The same structural factors do not underlie all types of black killing. Rather, a different combination of characteristics appears to influence each of the three types of homicide. Further, these characteristics explain less of the variance in family killings than in killings involving either acquaintances or strangers. This suggests that structural factors are more important for explaining acquaintance and stranger killings, and that interpersonal and situational conditions likely play a greater role in family victimizations.
Finally, of central concern here, the analysis confirms our view that racial residential segregation is an important aspect of inequality that should be considered in studies of lethal violence. Consistent with our premise, the analysis of total black homicides demonstrates that racial residential segregation (particularly when measured as residential dissimilarity) has a sizable positive
impact on black homicide rates. This finding is consistent with resource deprivation perspectives derived from Merton's anomie thesis (Blau & Blau 1982; Logan & Messner 1987), and with Wilson's recent arguments regarding the consequences of social isolation for minority communities. However, this result does not permit us to determine whether resource deprivation or social isolation is more salient for explaining the effect of segregation on homicide.
The analysis of different types of homicide contributes to a better under- standing of how segregation influences black lethal violence. Specifically, the disaggregated analyses make clear that the effect of segregation exists only for killings involving strangers and acquaintances, particularly the latter. This is true whether segregation is measured as dissimilarity or racial concentration. This pattern suggests that social isolation and the related lack of social control is the mechanism by which segregation leads to more homicides.
We draw this conclusion for two reasons. First, if the influence of segrega- tion reflects resource deprivation, then we would expect the pattern of influence of this variable to be similar to that of the other inequality measures (the gini index, white/black income ratio). We do not find this to be the case. Second, acquaintance and stranger homicides can be distinguished from family killings in that they often occur in more public settings. Homicides that are more public may be escalated because of inadequate protection i.e., "benign neglect" by law enforcement and other agencies, and the inability of neighborhoods to develop peacekeeping activities (e.g., neighborhood watches). In contrast, family
homicides tend to occur in more private contexts. As such, the presence or absence of more public social control mechanisms should have little effect on these more private killings.
These findings suggest that efforts to reduce black lethal violence should include attempts to alter important structural conditions underlying such violence. In particular, racial residential segregation and the accompanying social isolation are important targets for policymakers seeking to reduce African-American homicide in U.S. cities. Doing so will not be easily or quickly accomplished. Black-white residential segregation is entrenched in the U.S., especially in some of the large metropolitan areas where a very sizable portion of African Americans reside (Massey & Denton 1989). Recent evidence indicates
that various types of housing discrimination, which can reinforce patterns of residential segregation, are still widespread (Munnell et al. 1992; Turner, Struyk
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:56:26 UTC All use subject to http://about.jstor.org/terms
Segregation and Black Urban Homicide / 1021
& Yinger 1991; cf. Clark n.d.). Thus, one step that might be taken to reduce segregation would be to monitor more closely real estate and banking practices. However, segregation is also associated with inequality in economic resources (income and wealth) (Clark 1986; Farley 1977; Krivo 1990) and with differences between blacks and whites in their preferences for neighborhood racial composition (Clark 1991, 1992). These factors are less subject to intervention.
A more direct route to reducing black lethal violence might be achieved through programs aimed at reducing social isolation and increasing formal and informal control in segregated African-American communities. This might include strategies to improve the response time of police and emergency medical services, and to strengthen or re-establish important social institutions (e.g., local churches, banks, schools, stores, recreational facilities) within such neighborhoods (cf. Hagedom 1991). Also, where plausible, efforts might be made to establish a variety of community crime prevention programs within black localities.
In conclusion, the above analysis has helped to refine our understanding of the role of inequality in lethal violence, and particularly, the role of segregation in black homicides. The study makes it clear that consideration of racial residential segregation is important for understanding the alarmingly high rates of black homicide in U.S. cities. More generally, it appears that one important consequence of the "American Apartheid of residence" (Massey 1990) is a higher level of lethal violence for African Americans.
Notes
1. Community crime prevention strategies such as those noted above are predicated on the assumption that citizens can help reduce crime in their own communities through their involvement in neighborhood surveillance, reporting of suspicious activities, social interaction and bystander intervention. While a large number of evaluation projects report success for such
activities, a careful examination of research on citizen-based crime prevention efforts indicates
that there is only weak and inconsistent support for the influence of such programs on crime (see Garofalo & McLeod 1989; Rosenbaum 1987).
Most pertinent to our argument is the fact that even the modest effects of community crime prevention programs are precluded in some neighborhoods. Such strategies are most
likely to be implemented and sustained in neighborhoods characterized by "social homo- geneity, economic well-being, low residential turnover, and homeownership" (Garofalo & McLeod 1989:338). Since many segregated African-American communities exhibit characteristics quite contrary to this description, they are unlikely to have community crime prevention programs, thereby foregoing any benefits that might come from such citizen-based efforts.
2. Messner and Golden (1992) include racial residential segregation in an analysis of homicide offending rates. However, it is not possible to assess the separate effect of segregation in their research because it is included as a variable within an index of racial inequality.
3. We examine central cities rather than SMSAs as units of analysis for three primary reasons. First, according to Blalock (1982) "similarity is basic to the process of aggregation" (238). He notes that in aggregating by geographic proximity, it is always one's objective to select or form units of observation that are as theoretically homogeneous as possible (see also Parker 1989). In considering crime rates, cities are more homogeneous than SMSAs (or states or counties) which may include very diverse areal units, ranging from rural farming communities to very large and densely populated cities.
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:56:26 UTC All use subject to http://about.jstor.org/terms
1022 / Social Forces 71:4, June 1993
Second, within SMSAs, homicides occur more commonly in the central cities. For example, Parker (1989) reported that for 204 SMSAs during 1969 to 1971, 76% of the SMSA homicides occurred within the central cities. For the 125 geographic areas in our study, 84% of the black homicide victimizations within SMSAs occurred in the central cities.
Finally, we consider cities because of our interest in examining the impact of socio- economic deprivation on black homicides. It is unlikely that SMSAs provide the "relevant frames of reference in the assessment of economic [and social] well-being" (Messner 1982:112). As Bailey (1984) argues "It may be that daily associations and experiences are more salient in assessing [one's] well being relative to others than a self assessment based on a 'generalized other' at the county (SMSA) level' (535). Therefore, inequality measures should be based upon less highIy aggregated units such as cities.
4. Among the cases for which the index of dissimilarity was available from more than one source, the values of D differ substantially between the sources in only five cities. In three cases (Birmingham, Greensboro, and Newark), the use of Wilger's indices (which are calculated including Hispanics in the counts of whites and blacks) is preferable because a relatively large number of blacks were apparently identified incorrectly as Mexican American in some cities in the 1980 census (Denton & Massey 1989). If relatively large numbers of blacks who are non- Hispanic are excluded from the total number of blacks because of their incorrect identification as Hispanic, indices calculated for the black non-Hispanic population could be seriously incorrect. Indices based upon data for the entire black population would be preferable in cities for which this is the case (see Table 2 in Denton & Massey 1989:796). Birmingham, Greensboro, and Newark are three such places.
For two cities (Kansas City, Missouri, and Corpus Christi), the inclusion of incorrectly classified Hispanics in the black population is not a problem (Denton & Massey 1989) but inclusion of large numbers of correctly identified Hispanics in the white population could create a problem. To assess whether the use of D calculated from data including Hispanics versus data excluding Hispanics affected our results, we performed the analyses presented below substituting D values from Massey and Denton (which exclude Hispanics) for the D values from Wilger (which include Hispanics) for Kansas City and Corpus Christi. The results of these analyses were virtually identical to those presented below.
5. As operationalized here, family homicides include those that involve a victim who is related by blood or marriage to the offender, including immediate family (parent, child, spouse, sibling) and more distant relatives (e.g., in-law, cousin). The category of acquaintance homicides includes those in which the victim and offender have known each other, but are not related. The range is from close personal friends to neighbors and business associates. Killings by strangers are those where the offender does not know the victim. Homicide incidents where the relationship between the victim and the offender is not known to police are excluded from this part of the analysis.
6. The correlations and other descriptive statistics in Table 2 apply to the full sample of 125 cities and, therefore, include D but not q . Recall that the data needed for calculating q were available only for a subset of 111 cities. Tables providing the bivariate correlations and descriptive statistics for the set of 111 cities are available upon request.
7. More specifically, we estimated models for total black homicides and for each of the three types of homicide that, first, removed poverty but included black income inequality and education, and, second, removed black income inequality and education but included poverty. There were no changes in the significance of poverty, education, or black income inequality across these different models for total homicides, or for any of the three types of homicide.
8. The R2 values reported in Table 4 are Buse's R2 audge et al. 1985). Buse's R2 is used because the models by relationship type were estimated using the joint GLS discussed above.
9. Unfortunately, researchers have not examined the effect of bystander intervention on actual crime levels. However, analyses do indicate that bystanders are less likely to intervene in a crime when the victim is a stranger than they are when the victim is a relative or a very close friend (Bickman & Rosenbaum 1977; Gillis & Hagan 1983; Latane & Darley 1970). A bystander
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:56:26 UTC All use subject to http://about.jstor.org/terms
Segregation and Black Urban Homicide / 1023
is also more likely to intervene when the victimization occurs very close to one's home rather than in more distant places (Gillis & Hagan 1983).
10. There is only one case in which the significance of variables other than segregation differs between the analyses including D and those including 2. In the total homicide equations, region is significant when 2 is included but not when D is in the analysis.
11. These figures are derived by multiplying each of the unstandardized coefficients for 2 for total, family, acquaintance, and stranger homicides (8.386, .720, 6.841, and 3.537, respectively) by the standard deviation for of .178.
References
Allen, H. David, Dennis R. McSeveney, and William Bankston. 1981. "The Influence of Southem Culture on Race-Specific Homicide Rates." Sociological Spectrum 1:361-74.
Anderson, Elijah. 1990. Streetwise: Race, Class, and Change in an Urban Community. University of Chicago Press.
Bailey, William C. 1984. "Poverty, Inequality, and City Homicide Rates: Some Not So Unexpected Findings." Criminology 22:531-50.
Balkwell, James W. 1990. "Ethnic Inequality and the Rate of Homicide." Social Forces 69:53-70.
Bickman, Leonard, and Dennis P. Rosenbaum. 1977. "Crime Reporting as a Function of Bystander Encouragement, Surveillance, and Credibility." Journal of Personality and Social Psychology 35:577-86.
Blalock, Hubert M. 1982. Conceptualization and Measurement in the Social Sciences. Sage.
Blau, Judith R., and Peter M. Blau. 1982. "The Cost of Inequality: Metropolitan Structure and Violent Crime." American Sociological Review 47:114-129.
Blau, Peter M., and Reid M. Golden. 1986. "Metropolitan Structure and Criminal Violence." Sociological Quarterly 27:15-26.
Blau, Peter M., and Joseph E. Schwartz. 1984. Crosscutting Social Circles. Academic Press.
Braithwaite, John. 1979. Inequality, Crime and Public Policy. Routledge & Kegan Paul.
Centers for Disease Control Homicide Surveillance. 1986. High Risk Racial and Ethnic Groups - Blacks and Hispanics, 1970 to 1983. Centers for Disease Control.
Clark, William A.V. 1986. "Residential Segregation in American Cities: A Review and Inter-
pretation." Population Research and Policy Review 5:95-127.
_ . 1991. "Residential Preferences and Neighborhood Racial Segregation: A Test of the Schelling Segregation Model." Demographly 28:1-19.
_ . 1992. "Residential Preferences and Residential Choices in a Multiethnic Context." Demography 29:451-66.
_ . N.d. "Measuring Racial Discrimination in the Housing Market: Direct and Indirect Evidence." Urban Affairs Quarterly. Forthcoming.
Collins, James J., Brenda G. Cox, and Patrick Langan. 1987. "Job Activities and Personal Crime Victimization: Implications for Theory." Social Science Research 16:345-60.
Coser, Lewis. 1963. "Violence and the Social Structure." Pp. 30-42 in Science and Psychoanalysis. Vol. 6, edited by Jules H. Masserman. Grune & Stratton.
Crutchfield, Robert D. 1989. "Labor Stratification and Violent Crime." Social Forces 68:489-512.
Denton, Nancy A., and Douglas S. Massey. 1988. "Residential Segregation of Blacks, Hispanics, and Asians by Socioeconomic Status and Generation." Social Science Quarterly 69:797-817.
_. 1989. "Racial Identity Among Caribbean Hispanics: The Effect of Double Minority Status on Residential Segregation." American Sociological Review 54:790-808.
Farley, Reynolds. 1977. "Residential Segregation in Urbanized Areas of the United States in 1970: An Analysis of Social Class and Racial Differences." Demographly 14:497-518.
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:56:26 UTC All use subject to http://about.jstor.org/terms
1024 / Social Forces 71:4, June 1993
Galster, George C. 1987. "Residential Segregation and Interracial Economic Disparities: A Simultaneous Equations Approach." Journal of Urban Economics 21:2244.
Galster, George C., Fred Freiberg, and Diane L. Houk. 1987. "Racial Differentials in Real Estate Advertising Practices: An Exploratory Case Study." Journal of Urban Affairs 9:199-215.
Garofalo, James, and Maureen McLeod. 1989. "The Structure and Operations of Neighborhood
Watch Programs in the United States." Crime and Delinquency 35:32644.
Gillis, A.R., and John Hagan. 1983. "Bystander Apathy and the Territorial Imperative." Sociological Inquiry 53:449-60.
Hagedorn, John M. 1991. "Gangs, Neighborhoods, and Public Policy." Social Problems 38:529-42.
Hanushek, Eric A., and John E. Jackson. 1977. Statistical Methods for Social Scientists. Academic Press.
Harer, Miles D., and Darrell Steffensmeier. 1992. "The Differing Effects of Economic Inequality on Black and White Rates of Violence." Social Forces 70:1035-54.
Hawkins, Damell F. 1986. "Black and White Homicide Differentials: Alternatives to an Inadequate Theory." Pp. 109-34 in Homicide among Black Americans, edited by Damell F. Hawkins. University Press of America.
Hogan, Dennis P., and Evelyn M. Kitagawa. 1985. "The Impact of Social Status, Family Structure, and Neighborhood on the Fertility of Black Adolescents." American Journal of Sociology 90:825-55.
Huff-Corzine, Lin, Jay Corzine, and David C. Moore. 1986. "Southem Exposure: Deciphering the South's Influence on Homicide Rates." Social Forces 64:906-24.
Hwang, Sean-Shong, and Steve H. Murdock. 1982. "Residential Segregation in Texas in 1980." Social Science Quarterly 63:737-48.
Jiobu, Robert M., and Harvey H. Marshall, Jr. 1971. "Urban Structure and the Differentiation Between Blacks and Whites." American Sociological Review 36:63849.
Judge, George G., W.E. Griffiths, R. Carter Hill, Helmut Lutkepohl, and Tsoung-Chao Lee. 1985. Te Theory and Practice of Econometrics, 2nd edition. Wiley and Sons.
Krivo, Lauren J. 1990. "Family Assistance as a Means to Homeownership: A Comparison of Anglos, Blacks, and Mexican Americans." Paper presented at the annual meetings of the Population Association of America.
LaFree, Gary, and Kriss A. Drass, and Patrick CYDay. 1992. "Race and Crime in Postwar America: Determinants of African-American and White Rates, 1957-1988." Criminology 30:157-88.
Lake, Robert W. 1981. 77Te New Suburbanites: Race and Housing in the Suburbs. Center for Urban Policy Research, Rutgers University.
Land, Kenneth C., Patricia L. McCall, and Lawrence E. Cohen. 1990. "Structural Covariates of Homicide Rates: Are There Any Invariances Across Time and Social Space?" American Journal of Sociology 95:922-63.
Latane, Bibb, and John M. Darley. 1970. ne Unresponsive Bystander: Why Doesn't He Help? Appleton-Century-Crofts.
LaVeist, Thomas A. 1989. "Linking Residential Segregation to the Infant-Mortality Race Disparity in US. Cities." Sociology and Social Researchi 73:904.
. 1992. "The Political Empowerment and Health Status of African-Americans: Mapping a New Territory." American Journal of Sociology 97:1080-95.
Lieberson, Stanley, and Donna K. Carter. 1982. "A Model for Inferring the Voluntary and Involuntary Causes of Residential Segregation." Demography 19:511-26.
Loftin, Colin, and Robert Nash Parker. 1985. "An Errors-in-Variable Model of the Effect of Poverty on Urban Homicide Rates." Criminology 23:269-85.
Logan, John R., and Steven F. Messner. 1987. "Racial Residential Segregation and Suburban Violent Crime." Social Science Quarterly 68:510-527.
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:56:26 UTC All use subject to http://about.jstor.org/terms
Segregation and Black Urban Homicide / 1025
Logan, John R., and Mark Schneider. 1984. "Racial Segregation and Racial Change in American Suburbs: 1970-1980." American Journal of Sociology 89:874-88.
Luckenbill, David F. 1977. "Criminal Homicide as a Situated Transaction." Social Problems 25:176-86.
Massey, Douglas S. 1979. "Effects of Socioeconomic Factors on the Residential Segregation of Blacks and Spanish Americans in US. Urbanized Areas." American Sociological Review 44:1015-22.
. 1990. "American Apartheid: Segregation and the Making of the Underclass." American Journal of Sociology 96:329-57.
Massey, Douglas S., Gretchen A. Condran, and Nancy A. Denton. 1987. "The Effect of Residential Segregation on Black Social and Economic Well-Being." Social Forces 66:29-56.
Massey, Douglas S., and Nancy A. Denton. 1987. "Trends in the Residential Segregation of Blacks, Hispanics, and Asians: 1970-1980." American Sociological Review 52:802-25.
.1988. "Suburbanization and Segregation in U.S. Metropolitan Areas." American Journal of Sociology 94:592-626.
.1989. "Hypersegregation in U.S. Metropolitan Areas: Black and Hispanic Segregation Along Five Dimensions." Demography 26:373-91.
Massey, Douglas S., and Mitchell L. Eggers. 1990. "The Ecology of Inequality: Minorities and the Concentration of Poverty, 1970-1980." American Journal of Sociology 95:1153-88.
Merton, Robert K. 1968. Social Theory and Social Structure. Enlarged ed. Free Press.
Messner, Steven F. 1982. "Poverty, Inequality, and the Urban Homicide Rate: Some Unexpected Findings." Criminology 20:103-14.
. 1983a. "Regional Differences in the Economic Correlates of the Urban Homicide Rate: Some Evidence on the Importance of the Cultural Context." Criminology 21:477-88.
. 1983b. "Regional and Racial Effects on the Urban Homicide Rate: The Subculture of Violence Revisited." American Journal of Sociology 88:997-1007.
Messner, Steven F., and Reid M. Golden. 1985. "Economic Sources of Homicide: Reconsidering the Effects of Poverty and Inequality." Paper presented at the annual meetings of the American Sociological Association.
. 1992. "Racial Inequality and Racially Disaggregated Homicide Rates: An Assessment of Altemative Theoretical Explanations." Criminology 30:421-47.
Messner, Steven, and Scott J. South. 1986. "Economic Deprivation, Opportunity Structure, and Robbery Victimization: A Test of Routine Activities/Lifestyle Theories." American Sociological Review 52:184-94.
Munnell, Alicia H., Lynn E. Browne, James McEneaney, and Geoffrey M.B. Tootell. 1992. Mortgage Lending in Boston: Interpreting HMDA Data. Working Paper no. 92-7. Federal Research Bank of Boston.
O'Carroll, Patrick W., and James A. Mercy. 1986. "Patterns and Recent Trends in Black Homicide." Pp. 29-41 in Homicide among Black Americans, edited by Damell F. Hawkins. University Press of America.
Parker, Robert Nash. 1989. "Poverty, Subculture of Violence, and Type of Homicide." Social Forces 67:983-1007.
Parker, Robert Nash, and M. Dwayne Smith. 1979. "Deterrence, Poverty, and Type of Homicide." American Journal of Sociology 85:614-24.
Pearce, Diana M. 1979. "Gatekeepers and Homeseekers: Institutional Patterns in Racial Steering." Social Problems 26:32542.
Potter, Lloyd B. 1991. "Socioeconomic Determinants of White and Black Males' Life Expectancy Differentials, 1980." Demography 28:303-21.
Riedel, Marc. 1984. "Blacks and Homicide." Pp 51-60 in The Criminal Justice System and Blacks, edited by Daniel Georges-Abeyie. C. Boardman.
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:56:26 UTC All use subject to http://about.jstor.org/terms
1026 / Social Forces 71:4, June 1993
. 1987. "Stranger Violence: Perspectives, Issues, and Problems." Journal of Criminal Law and Criminology 78:223-58.
Rosenbaum, Dennis P. 1987. "The Theory and Research Behind Neighborhood Watch: Is it a Sound Fear and Crime Reduction Strategy?" Crime and Delinquency 33:103-34.
Rosenfeld, Richard. 1986. "Urban Crime Rates: Effects of Inequality, Welfare Dependency, Region, and Race." Pp. 116-30 in The Social Ecology of Crime, edited by James M. Byrne and
Robert J. Sampson. Springer-Verlag.
Sampson, Robert J. 1985. "Race and Criminal Violence: A Demographically Disaggregated Analysis of Urban Homicide." Crime and Delinquency 31:47-82.
. 1986. "Effects of Inequality, Heterogeneity, and Urbanization on Intergroup Victimiza- tion." Social Science Quarterly 67:751-66.
. 1987. "Urban Black Violence: The Effect of Male Joblessness and Family Disruption." American Journal of Sociology 93:348-82.
Schneider, Mark, and John R. Logan. 1982. "Suburban Racial Segregation and Black Access to Local Public Resources." Social Science Quarterly 63:762-70.
Schuman, Howard, and Lawrence Bobo. 1988. "Survey-based Experiments on White Racial Attitudes toward Residential Integration." American Journal of Sociology 94:273-99.
Simkus, Albert A. 1978. "Residential Segregation by Occupation and Race in Ten Urbanized Areas, 1950-1970." American Sociological Review 43:81-93.
Smith, Dwayne M., and Robert Nash Parker. 1980. "Type of Homicide and Variation in Regional Rates." Social Forces 59:136-47.
South, Scott J., and Richard B. Felson. 1990. "The Racial Patterning of Rape." Social Forces 69:71- 93.
Stearns, Linda Brewster, and John R. Logan. 1986a. "The Racial Structuring of the Housing Market and Segregation in Suburban Areas." Social Forces 65:2842.
. 1986b. "Measuring Trends in Segregation: Three Dimensions, Three Measures." Urban Affairs Quarterly 22:124-50.
Theil, Henri. 1971. Principles of Econometrics. Wiley.
Turner, Margery Austin, Raymond J. Struyk, and John Yinger. 1991. Housing Discrimination Study: A Synthesis. Office of Policy Development and Research, U.S. Department of Housing and Urban Development.
Van Valey, Thomas L., Wade Clark Roof, and Jerome E. Wilcox. 1977. "Trends in Residential Segregation: 1960-1970." American Journal of Sociology 82:826-44.
Wienk, Ronald E., Clifford W. Reid, John C. Simonson, and Frederick J. Eggers. 1979. Measuring Racial Discrimination in American Housing Markets: The Housing Market Practices Survey. U.S. Department of Housing and Urban Development.
Wilger, Robert J. 1988. Black-White Residential Segregation in 1980. The University of Michigan.
Williams, Kirk R. 1984. "Economic Sources of Homicide: Reestimating the Effects of Poverty and Inequality."American Sociological Review 49:283-89.
Williams, Kirk R., and Robert L. Flewelling. 1987. "Family, Acquaintance, and Stranger Homicide: Alternative Procedures for Rate Calculations." Criminology 25:543-60.
. 1988. "The Social Production of Criminal Homicide: A Comparative Study of Dis- aggregated Rates in American Cities." American Sociological Review 53:421-31.
Wilson, William Julius. 1987. The Truly Disadvantaged: T7he Inner City, the Underclass, and Public Policy. University of Chicago Press.
Yinger, John. 1986. "Measuring Racial Discrimination with Fair Housing Audits: Caught in the Act." American Economic Review 76:881-93.
Zahn, Margaret A., and Philip C. Sagi. 1987. "Stranger Homicides in Nine American Cities." Journal of Criminal Law and Criminology 78:377-97.
This content downloaded from 155.247.166.234 on Sun, 19 Jun 2016 04:56:26 UTC All use subject to http://about.jstor.org/terms
- Contents
- image 1
- image 2
- image 3
- image 4
- image 5
- image 6
- image 7
- image 8
- image 9
- image 10
- image 11
- image 12
- image 13
- image 14
- image 15
- image 16
- image 17
- image 18
- image 19
- image 20
- image 21
- image 22
- image 23
- image 24
- image 25
- image 26
- Issue Table of Contents
- Social Forces, Vol. 71, No. 4 (Jun., 1993), pp. 851-1120
- Volume Information [pp. 1113-1119]
- Front Matter [pp. 954-1066]
- Comparing World-Systems: Concepts and Working Hypotheses [pp. 851-886]
- The Organization of Survival: Women's and Racial-Ethnic Voluntarist and Activist Organizations, 1955-1985 [pp. 887-908]
- Risk and Recreancy: Weber, the Division of Labor, and the Rationality of Risk Perceptions [pp. 909-932]
- A Sociological Theory of Scientific Change [pp. 933-953]
- The Search for Adolescent Role Exits and the Transition to Adulthood [pp. 955-979]
- Joint Role Investments and Synchronization of Retirement: A Sequential Approach to Couples' Retirement Timing [pp. 981-1000]
- Racial Segregation and Black Urban Homicide [pp. 1001-1026]
- Religious Involvement and Self-Perception among Black Americans [pp. 1027-1055]
- Odds versus Probabilities in Logit Equations: A Reply to Roncek [pp. 1057-1065]
- When Will They Ever Learn That First Derivatives Identify the Effects of Continuous Independent Variables or "Officer, You Can't Give Me a Ticket, I Wasn't Speeding for an Entire Hour" [pp. 1067-1078]
- Book Reviews
- Review: untitled [pp. 1079-1080]
- Review: untitled [pp. 1080-1082]
- Review: untitled [pp. 1082-1083]
- Review: untitled [pp. 1084-1085]
- Review: untitled [pp. 1085-1087]
- Review: untitled [pp. 1087-1088]
- Review: untitled [pp. 1088-1089]
- Review: untitled [pp. 1090-1091]
- Review: untitled [pp. 1091-1092]
- Review: untitled [pp. 1093-1094]
- Review: untitled [pp. 1094-1095]
- Review: untitled [pp. 1095-1097]
- Review: untitled [pp. 1097-1098]
- Review: untitled [pp. 1098-1099]
- Review: untitled [pp. 1099-1100]
- Review: untitled [pp. 1100-1101]
- Review: untitled [pp. 1101-1102]
- Review: untitled [pp. 1102-1103]
- Review: untitled [pp. 1103-1104]
- Review: untitled [pp. 1104-1105]
- Review: untitled [pp. 1106-1107]
- Review: untitled [pp. 1107-1108]
- Review: untitled [pp. 1109-1110]
- Back Matter [pp. 1111-1120]
12 sources/The Racist Origins of Felon Disenfranchisement - The New York Times.pdf
http://nyti.ms/1EXmNy1
The Opinion Pages
The Racist Origins of Felon Disenfranchisement Editorial Observer
By BRENT STAPLES NOV. 18, 2014
The state laws that barred nearly six million people with felony convictions from voting in the midterm elections this month date from the late 19th and early 20th centuries, when Southern lawmakers were working feverishly to neutralize the black electorate.
Poll taxes, literacy tests, grandfather clauses and cross burnings were effective weapons in this campaign. But statutes that allowed correctional systems to arbitrarily and permanently strip large numbers of people of the right to vote were a particularly potent tool in the campaign to undercut African-American political power.
This racially freighted system has normalized disenfranchisement in the United States — at a time when our peers in the democratic world rightly see it as an aberration. It has also stripped one in every 13 black persons of the right to vote — a rate four times that of nonblacks nationally. At the same time, it has allowed disenfranchisement to move beyond that black population — which makes up 38 percent of those denied the vote — into the body politic as a whole. One lesson here is that punishments designed for one pariah group can be easily expanded to include others as well.
The history of disenfranchisement was laid out in a fascinating 2003 study by Angela Behrens, Christopher Uggen and Jeff Manza. They found that state felony bans exploded in number during the late 1860s and 1870s, particularly in the wake of the Fifteenth Amendment, which ostensibly guaranteed black Americans the right to vote.
They also found that the larger the state’s black population, the more likely the state was to pass the most stringent laws that permanently denied people convicted of crimes the right to vote.
These bans were subsequently strengthened as the Jim Crow era began to take hold.
The white supremacists who championed such measures were very clear on their reasons. In 1894, a white South Carolina newspaper argued that voting laws needed to be amended, lest whites be swept away at the polls by the black vote. In 1901 Alabama amended its Constitution to expand disenfranchisement to all crimes involving “moral turpitude” — a vague term that was applied to misdemeanors and even acts not punishable by law. The president of the constitutional convention argued that manipulating the ballot to exclude blacks was warranted, because they were inferior to whites and because the state needed to avert the “menace of Negro domination.”
The official who introduced the new provision at the convention said, “The crime of wife-beating alone would disqualify 60 percent of the Negroes.” This did not mean that only black men committed spousal abuse; it meant that whites were less likely to be prosecuted for this and several other offenses that could lead to disenfranchisement.
Alabama today has one of the highest rates of felony disenfranchisement in the nation: An estimated 7.2 percent of its citizens — and 15 percent of African- Americans — have lost the right to vote.
The disenfranchisement laws flourished in both Northern and Southern states where large black populations were cast in the role of eternal outsiders, and proposals for allowing former felons to vote were often cast as heralding the end of
civilization.
The debate looks a lot different in Maine and Vermont, states where there are no black populations to speak of and racial demonization does not come into the equation. Both states place no restrictions on voting rights for people convicted of even serious crimes and have steadfastly resisted efforts to revoke a system that allows inmates to vote from prison.
Maine residents vigorously debated the issue last year, when the Legislature took up — and declined to pass — a bill that would have stripped the vote from some inmates, whose crimes included murder and other major felonies. Families of murder victims argued that the killers had denied their loved ones the right to vote and therefore should suffer the same fate.
Those who opposed the bill made several arguments: That the franchise is enshrined in the state Constitution and too important to withdraw on a whim; that voting rights keep inmates connected to civic life and make it easier for them to rejoin society; that the notion of restricting rights for people in prison was inconsistent with the values of the state.
A former United States marshal and police chief argued that revoking inmate voting rights would strip imprisoned people of dignity and make rehabilitation that much more difficult. The editorial page of The Bangor Daily News argued against revocation on the grounds that, “Removing the right of some inmates to exercise their legal responsibility as voters in a civilized society would undermine that civilized society.”
The fact that most states view people who have served time in prison as beyond the protection of the bedrock, democratic principle of the right to vote shows how terribly short this country has fallen from achieving its ideals.
Meet The New York Times’s Editorial Board »
A version of this editorial appears in print on November 19, 2014, on page A24 of the New York edition with the headline: The Racist Origins of Felon Disenfranchisement.
© 2016 The New York Times Company