Review of the Roots of Youth Violence: Literature Reviews

Volume 5, Chapter 9:

The Subculture of Violence8

The search for the social causes of violence has been an ongoing preoccupation of criminologists since the early 20th century, when researchers attempted to look beyond biological and psychological explanations to understand crime. The shift in understanding violence as a social phenomenon, rather than an individual one, emerged from observations that incidents of violence tend to not be evenly distributed within society. Rather, rates of crime and violence vary spatially and demographically. The endeavour to understand these patterns has generated a range of theories that highlight various social processes, including how crime is learned and taught and how it emerges from social inequalities. Amongst these various explanations, few have been as durable as the explanation of culture.

Cultural explanations for violence first emerged in the works of American delinquency theorists in the 1930s who were attempting to account for the concentrations of crime and violence in poor, urban African-American neighbourhoods in the 1930s. Sellin (1938), Miller (1958) and Cohen (1951) were amongst the first of many cultural deviance theorists to develop the concept of a criminal subculture and to link it to social problems such as poverty and inequality. However, it was not until the late 1960s that Wolfgang and Ferracuti (1967) launched a distinct and comprehensive subculture of violence thesis. Possibly the most well-known theory in this genre of work, Wolfgang and Ferracuti’s Subculture of Violence attempted to outline a methodological framework for the empirical examination of violent subcultures. Central to their discussion was the idea that higher rates of violence amongst lower-class and racialized populations could be explained by the fact that these groups have embraced values and norms that are more permissive of violence. This theorization assumes the existence of distinct subcultural, pro-violent values that develop in opposition to dominant or middle-class norms and values.

This literature review will provide a detailed discussion of the subculture of violence thesis and trace its development from the work of Wolfgang and Ferracuti (1967) to its more current applications. Though the subculture of violence thesis was originally devised to explain and examine high rates of violence amongst structurally marginalized populations and neighbourhoods, since then, this framework has been applied and evaluated in relation to a variety of other demographics and locales, such as the American South (Nisbett and Cohen, 1996; Hayes, 2005), athletes (Smith, 1979), middle schools and high schools within the United States and Iceland (Felson et al., 1994; Bernburg and Thorlindsson, 2005; Ousey and Wilcox, 2005). In addition, many have also continued with Wolfgang and Ferracuti’s (1967) original focus and have attempted to determine the existence of a Black subculture of violence (Anderson, 1999; Stewart and Simons, 2006; Brezina et al., 2004; Cao et al., 1997). This discussion will explore the implications of these findings and whether there is adequate support to suggest the existence of a subculture of violence.

Values, Norms and Violence: The Subculture of Violence Thesis

In The Subculture of Violence, Wolfgang and Ferracuti draw upon Wolfgang’s earlier work on inner-city African-American neighbourhoods in Philadelphia in order to formulate an operational definition of the concept of a subculture. The key objective of their work is to develop a way to identify and measure subcultures of violence in order to scientifically prove their existence. In order to do so, the authors propose an integrated methodological and theoretical approach, which involves drawing from a variety of existing criminological theories as well as from insights from sociology and psychology.

With regard to explaining how subcultures cause violence, Wolfgang and Ferracuti argue that violence is a product of conformity to a pro-violent subculture that is in direct conflict with the dominant culture. While they do not suggest that subcultures are in total conflict with the societies of which they are a part, the authors note that the “overt use of force or violence, either in interpersonal relationships or in group interaction, is generally viewed as a reflection of basic values that stand apart from the dominant, the central or parent culture” (Wolfgang and Ferracuti, 1967: 158). Wolfgang and Ferracuti offer no explanation as to how subcultures of violence evolve in the first place, despite their assertions that they tend to be a lower-class, racialized and masculine phenomenon.

The proliferation of violence within this context is believed to result from a tendency amongst subcultural offenders to embrace values and norms that are more permissive of the use of violence under certain conditions. Implicit in this proposition is the concept of disputatiousness, which suggests that violence is a central means for subcultural affiliates to maintain and protect their status. According to Wolfgang and Ferracuti, violent reactions to perceived threats to reputation or honour are culturally prescribed, given that a failure to react defensively may result in life-threatening consequences. In this sense, violent values act as a mechanism of social control, given that they require members of a subculture to engage in violence for their own protection and survival. As a result, equipped with the values to justify their violent actions, subcultural offenders engage in violence frequently and guiltlessly, with little provocation (Wolfgang and Ferracuti, 1967: 314).

Though they draw from the insights of cultural deviance theorists before them, empirically Wolfgang and Ferracuti base their theorizations entirely upon inferences from available statistics on homicide at that time, which illustrated a concentration of violent crime in primarily poor, racialized neighbourhoods in Philadelphia. Through these statistics, the authors deduce that subcultural violence is a uniquely marginalized male phenomenon. As opposed to the random and anomalous violence of the “mainstream” world, this particular strain of violence, they argue, is a collective phenomenon, a normal experience for poor, non-white men.

The proposed methodology in The Subculture of Violence focuses on identifying “proviolent” values. It is believed that the observation of values will, in turn, provide insight into the group norms, since the latter sustains the former through rewarding conformity and penalizing non-conformity. Thus, individual action, attitude and perception are considered to be the key to understanding the collective phenomena of culture. With this in mind, Wolfgang and Ferracuti propose the following:

We suggest that, by identifying the groups with the highest rates of homicide, we should find in the most intense degree subcultures of violence; and, having focused on these groups we should subsequently examine the value system of their subculture, the importance of human life in the scale of values, the kinds of expected reaction to certain types of stimulus, perceptual differenced in the evaluation of stimuli, and the general personality structure of the subcultural actors (Wolfgang and Ferracuti 1967: 153).

Thus, following the identification of a potential subculture of violence based on statistics indicating concentrations of violence amongst marginalized men, whether “pro-violent” values are in existence should be assessed through determinations of whether there is collective approval, encouragement and reward for engaging in violence. Finally, in order to examine whether a variance exists between suspected subcultural offenders and middle-class individuals, Wolfgang and Ferracuti propose measuring “social values using a ratio scale (as in psychophysics) focused on items concerned with behavioural displays of violence” (Wolfgang and Ferracuti, 1967: 315). Although they do not actually apply their methods or test their thesis in The Subculture of Violence, the authors express certainty regarding the potential of their method to identify subcultural offenders, predict who might become a subcultural offender, and ascertain the location of such offenders (Wolfgang and Ferracuti, 1967: 315).

Empirical Examinations of the Subculture of Violence Thesis

Over the past three decades, the subculture of violence thesis has undergone considerable scrutiny, given that Wolfgang and Ferracuti’s claims at the time they were writing were, for the most part, unsubstantiated. Those who have undertaken evaluations of this framework have pointed out the tautological or circular reasoning that underlies the assumed link between violent values and violent actions and the importance of analyzing individual data on values in order to asses the validity of this thesis (Erlanger, 1974; Kornhauser, 1978). In addition, Wolfgang and Ferracuti’s assertions have been widely criticized for perpetuating stereotypes of young African-American males, for failing to consider the emergence of subcultures of violence, and for their theoretical negligence of social structural factors in their discussions of root causes of violence (Convington, 2003; Surratt et al., 2004).

Many of the studies that have evaluated the subculture of violence thesis do not test it in its entirety. Rather, most examine elements of this perspective, focusing on specific propositions such as the contention that there is a link between violent behaviour and violent values. So far, empirical evaluations of Wolfgang and Ferracuti’s propositions have not yielded conclusive support for the subculture of violence thesis. This section will review some of the most widely cited tests of this framework, as well as more recent applications of this theory which attempt to go beyond the purviews of Wolfgang and Ferracuti’s original propositions regarding the specificity of subcultural violence to lower-class, racialized men. This discussion will be followed by a review of some of the more in-depth evaluations of whether the subculture of violence thesis is an appropriate explanation for the high rates of violence amongst poverty-stricken black neighbourhoods in the United States. Finally, recent studies evaluating the existence of a subculture of violence in the southern United States will be reviewed in detail.

Ball-Rokeach (1973) conducted one of the earliest examinations of the subculture of violence thesis, focusing specifically on whether individuals who engage in violence hold favourable attitudes towards violence and are committed to subcultural, pro-violent values (Ball-Rokeach, 1973: 737). She notes the distinctions between the concepts of “values” and “attitudes.” While the former is defined as “a belief either about a ‘desired end-state of existence,” “or a belief about a ‘preferred mode of conduct” such as being honest, courageous or loving,” “attitude” is defined as “an organization of beliefs about a specific object, ... or about an ongoing belief about an ongoing activity or situation” (Ball-Rokeach, 1974: 737). Ball-Rokeach tests the hypothesized relationships between violent values, attitudes and violent behaviour through two independent studies. The first involves assessing the values and attitudes underlying interpersonal violence in a probability sample of 1,429 adult Americans, and the second consists of examining disparities in values amongst 363 male offenders incarcerated for various violent and non-violent crimes (Ball-Rokeach, 1973: 737). An additional component of the first study entailed determinations of the degree to which participants in the national sample engaged in violent actions and behaviours. In the latter study, the key objective was to determine whether violent and non-violent offenders do in fact hold differing values towards violence.

The extent to which respondents engaged in violence in the first study was assessed through self-report questions about their participation and experiences in interpersonal violence as both assailants and victims. In the second, engagement in violence was assessed through distinguishing between those offenders who were convicted of violent offences and those who were not. Ball-Rokeach determined whether participants in the national sample held pro-violent attitudes through their responses to three hypothetical scenarios. Respondents were asked whether they approve of the use of violence in the following scenarios: “a teenage boy punching or beating another teenage boy;” “a public school teacher hitting a student;” and “a judge sentencing a person to death” (Ball-Rokeach, 1973: 738). Whether those who engaged in violence in both the national and prison samples were motivated by pro-violent values was assessed through a value survey compiled by Ball-Rokeach in an earlier study. For this portion of the experiment, respondents were asked to rank, in order of priority, a list of concepts that are indicative of various human values, as well as a variety of human characteristics they found important or appealing. The survey did not include values specific to the use of violence for the purposes of protecting or maintaining reputation amongst peers.

With regard to the national sample, the researchers found no correlation between participation in violence and attitudes towards violence, leading them to reject the hypothesis that pro-violent attitudes lead to violent behaviour. Respondents who had engaged in violence at one point in their lives were no more likely to approve of the use of violence within the given hypothetical scenarios. With regard to her assessment of a causal relationship between values and participation in violence, Ball-Rokeach found that amongst participants in the national sample, those who “report moderate or high participation [in violence] place significantly more importance on An Exciting Life, Mature Love and Being Imaginative” and “less importance on Social Recognition, A Comfortable Life and A World at Peace” (Ball-Rokeach, 1973: 739). Ball-Rokeach interprets these findings as lending little support to the proposition that violent values are conducive to violent behaviour. She argues that if individuals who engage in violence do indeed hold the machismo values that Wolfgang and Ferracuti claim generate violence, it is more likely that these respondents would express a commitment to experiences associated with the idealized masculine persona, such as “An Exciting Life, Pleasure, Social Recognition, being Courageous and Independent” (Ball-Rokeach, 1974: 741). A lack of commitment to values characterized as “non-machismo,” such as being forgiving, would lend further support to the subculture of violence thesis. Though individuals who engaged in violence expressed some commitment to “An Exciting Life,” Ball-Rokeach advises that the differences in commitment to this value amongst this group and those that did not engage in violence was minimal (Ball-Rokeach, 1974: 741). Similar findings were apparent in relation to the value comparison experiment amongst violent and nonviolent offenders in the prison sample.

Finally, although she did not intend to test for this, Ball-Rokeach notes her supplementary finding that questions Wolfgang and Ferracuti’s assumption of a correlation between socio-economic status and violent conduct. In contrast to the expected correlation between approval of violence and lower-class status, Ball-Rokeach found a more significant relationship between pro-violent attitudes and education (Ball-Rokeach, 1973: 743). Thus, Ball-Rokeach concludes that “the relatively weak association between attitudes and violent behaviour (hypothesis 1) taken together with the fact that there is little or no evidence of a relationship between values and violent behaviour (hypothesis 2), suggests that the subculture of violence thesis is, at best, incomplete and at worst, invalid as an explanation of violent behaviour” (Ball-Rokeach, 1973: 748).

It is important to note some distinctions between Wolfgang and Ferracuti’s original propositions and Ball-Rokeach’s research design. In so doing, whether Ball-Rokeach’s study is a valid test of the subculture of violence thesis should be considered. First, Wolfgang and Ferracuti do not suggest that all those who participate in violence have violent values. Rather, they argue that only subcultural offenders hold pro-violent values. This serves as the basis of their assertion that empirical assessments of their thesis should only be conducted on those offenders who are believed to be part of this subgroup. As noted earlier, Wolfgang and Ferracuti argue that those enmeshed within a subculture of violence are more likely to be low-income racialized men. In her study, Ball-Rokeach employs a random sample of men in the general population, which does not serve as an accurate sample for the purposes of this analysis.

An additional problem emerges in relation to Ball-Rokeach’s examination and interpretation of subcultural pro-violent values. This study does not take the idea of disputatiousness into consideration, which is central to understanding the use of violence within subcultures. Wolfgang and Ferracuti assert that the key difference between subcultural violent offenders and general violent offenders is that for the former group, violence for the purposes of protecting or maintaining reputation is not considered to be wrong. Rather, it is encouraged. Thus, a more precise evaluation of the subculture of violence proposition regarding the relationship between violent behaviour and pro-violent values should reflect this instrumental understanding of violence. The scenarios used in Ball-Rokeach’s study to assess whether research participants had engaged in violence are too general. Rather than inquiring about how research participants felt about violence in reaction to a “jostle, a slightly derogatory remark, or the appearance of a weapon in the hands of an adversary,” which Wolfgang and Ferracuti suggest are “stimuli differentially perceived” by subcultural offenders, Ball-Rokeach analyzes participant approval for general violence amongst teenage boys, corporal punishment, and capital punishment (Wolfgang and Ferracuti, 1967: 153).

In another early study of the subculture of violence thesis, Erlanger (1974) examines structural differences in attitudes towards violence and attempts to go beyond studying the influence of pro-violent values through considering the impact of norms on violent behaviour. He tests Wolfgang and Ferracuti’s proposition that African-American and lower-class men are more likely to hold subcultural values condoning violence than white and middle class men are. In addition, through a consideration of the influence of norms, Erlanger examines whether those participants who report engaging in violence are motivated by peer pressure, sanctions or a desire to impress their peers. He assesses the role of pro-violence norms in the occurrence of violence through studying how those who have engaged in violence view their conduct and their “perceived esteem accorded by others” (Erlanger, 1974: 284).

Erlanger relies on data from a 1969 survey involving both African-American and white males between the ages of 21 and 64, in which respondents were asked about their participation in violence and their perceived status within their peer group. Participation in violence was gauged through the following question: “How often do you get in angry fist fights with other men” (Erlanger, 1974: 284)? The perceived status was determined through responses to the following questions: “How do you compare with most men you know on being respected and listened to by other people? How do you compare with most men you know on being well liked by other people and having lots of friends” (Erlanger, 1974: 284)?

Erlanger’s findings suggest some support for the subculture of violence thesis with regard to participation in violence. He found that racially and economically marginalized men were more likely to participate in violence than white and middle class men were (Erlanger, 1974: 285). In addition, with regard to the question of norms, his findings illustrate that “low-income blacks and low-income whites show a positive net effect of fighting on perceived esteem” (Erlanger, 1974: 286). However, Erlanger hesitates to embrace the subculture framework, arguing that his findings are not statistically significant. While the relationship between violence and perceived self-esteem was relatively weak, evidence supporting a counter-norm of non-violence amongst the white middle-class participants was also minimal. Based on the weak support for these predicted patterns, he argues that “none of the predictions of the subcultural theory is supported” (Erlanger, 1974: 286).

Although Erlanger offers a more precise study of Wolfgang and Ferracuti’s framework, his research design is not without its flaws. One problem lies in Erlanger’s reliance on a general question about fist-fighting to evaluate his research participants’ engagement in violence, rather than on one that captures the concept of disputatiousness. In addition, though Erlanger attempts to assess disputatiousness through relating participation in violence to self-esteem, the assumed link between self-esteem and violence in this study is debatable. It can be argued that factors other than participation in violence may impact an individual’s self-esteem, which might influence how questions were answered. Given this possibility, questions about self-esteem may not say much about whether participants in violence are under the influence of pro-violent norms.

Subcultures of Violence, Peer Support and Social Control

Despite Wolfgang and Ferracuti’s propositions that subcultures are most likely to develop amongst marginalized populations, some of the more recent evaluations of the subculture of violence thesis have examined its relevance within a variety of contexts and amongst diverse populations. The following section will review the findings of three studies that were conducted amongst middle and high school students in the United States and Iceland. All explore the impact of group norms on levels of violence and the presumed relationship between violent values and violent behaviour. While Felson et al. (1994) and Berburg and Thorlindsson (2005) found some support for a link between group norms and values permissive of violence and violent behaviour, Ousey and Wilcox (2005) found the opposite to be true. Finally, all three studies raise doubts about the role of socialization processes in the transmission of subcultural violence. While Ousey and Wilcox (2005) point to the importance of considering additional factors such as impulsivity and exposure to violent peers, the other two studies suggest the significance of social control processes in encouraging disputatiousness amongst subcultural affiliates.

The degree to which group norms exert an influence on individuals who commit violence has been the subject of more recent evaluations of the subculture of violence thesis. One of the most widely cited studies in this genre is one conducted by Felson et al. (1994) on high school students in the early 1990s. Felson et al. are interested in the “process by which the subculture of violence operates,” particularly in relation to degree to which subcultural values act as a mechanism of social control on group members (Felson et al., 1994: 157). Rather than focusing on socialization processes, which theorize violence as an occurrence of the internalization of pro-violent values, the authors examine Wolfgang and Ferracuti’s related proposition that subcultural violence may also result from a desire or need to defend or maintain honour. By distinguishing between these two processes, Felson et al. make an analytic distinction between group and individual values and are able to control for the latter when examining the causal relationship between values and violence.

Felson et al. conducted two waves of interviews with male high school students across 87 high schools. In the first wave, 2,213 sophomore males were interviewed. Of the original group, 1,886 students participated in the second round of interviews. The researchers relied on personal interviews as well as questionnaires in order to obtain data on their subjects’ participation in acts of interpersonal violence, theft and vandalism, and general delinquent behaviour. Information on the degree to which research participants engaged in violent or aggressive actions to maintain honour or viewed violence as a legitimate reaction to some form of provocation was obtained through their responses to a variety of hypothetical scenarios. For each, research participants were asked whether they thought it was a “good thing” to: “turn the other cheek and forgive others when they harm you,” “reply to anger with gentleness,” and “be kind to people even if they do things against one’s own belief” (Felson et al., 1994: 159). Felson et al. controlled for race, socio-economic status and family and neighbourhood instability when analyzing the relationships between violent and non-violent delinquent acts and the responses to each scenario.

Felson et al. list three key objectives to their study. First, they sought to gain a sense of the degree to which interpersonal violence amongst males varied across the high schools in their sample. Second, they conducted aggregate analyses in order to explore whether any variations in rates of violence between the schools in the sample could be linked to the presence, absence or influence of subcultural values. Finally, the impact of school context on rates of individual violence was assessed. This analysis was conducted for the purposes of examining the relationship between the research participant’s personal values and his participation in violence, as well as any correlation between violence and his schoolmates’ values. The purpose of this contextual analysis was to assess the impact of immediate peer group values on individual violent behaviour.

Felson et al. found significant support for the contention that group norms that encourage violence exert far more influence via social control processes than through socialization processes. First, Felson et al. observed distinctions amongst high schools with regard to peer approval of violent or aggressive responses to provocations. While some school environments were conducive to this, others were not. When analyzed further, their contextual analysis revealed that “a boy’s violence and delinquency are related to the values prevalent in his school, independent of his own values” (Felson et al., 1994: 168). Thus, in those schools marked by high rates of violence, it appears as though boys engage in violence to maintain their reputation within their school peer groups. Based on this finding, the authors question the idea that boys commit violence as a result of their internalization of school-based, pro-violent values.

Though they deviate from Wolfgang and Ferracuti’s original proposition that subcultures of violence are more likely to exert an influence on lower-class racialized men, Felson et al. reject this contention from the outset, citing inconclusive findings from earlier studies, including Erlanger’s (1974), as evidence. The authors hypothesize that the subculture of violence thesis may be more appropriate as an explanation for concentrations of violence within small groups in which members engage in ongoing social interaction. They advise that “aggregates based on race or class and regions may not provide the necessary conditions for such a subculture to develop” (Felson et al., 1994: 168). Felson et al. interpret their findings as an additional impetus to reject Wolfgang and Ferracuti’s original proposition regarding the likely locales of violent subcultures.

Like Felson et al. (1994), the work of Ousey and Wilcox (2005) also questions the degree to which pro-violent values instigate violent behaviour within subcultures of violence. Ousey and Wilcox examine the influence of individual and aggregate-level pro-violence values on incidents of violent behaviour within a school environment. The authors replicate Felson et al.’s (1994) study with a few alterations. They criticize Felson for overlooking two other possible explanations for violent behaviour that are theoretically notable in the literature on crime and delinquency: exposure to violent peers and low self-control. Incorporating elements of differential association theories of crime and delinquency, the authors hypothesize that individuals may model the violent behaviour of their peers, rather than actually embracing pro-violent values or engaging in violence for the purposes of impression management. In addition, based on the work of Gottfredson and Hirschi (1990), the authors consider the impact of low self-control on the occurrence of subcultural violence. Ousey and Wilcox criticize Felson et al. (1994) for not controlling for impulsivity in their study and state that a failure to do so may exaggerate the indicators of pro-violence values or attitudes.

Ousey and Wilcox’s objective is to address these limitations through examining various relationships between pro-violent values on an individual and aggregate level and aggressive and violent behaviour within schools. In challenging the idea that violent values are primarily the cause of violent behaviour within subcultures of violence, they predict the following: 1) Schools with students who appear to hold values that are more accepting of violence will, on average, have higher rates of violent incidents than will schools with fewer students with pro-violent values. 2) Individuals who hold values or attitudes favourable to violence are more likely to report frequent participation in violence. 3) A distinction between variation in pro-violence values at the individual level and school level can be observed. 4) Once impulsivity and exposure to violent peers are controlled for, it is expected that the effects of school and individual level pro-violence values will be attenuated. 5) The link between pro-violent values and violent behaviour will be stronger in those schools that present greater evidence of violent subcultures. (Ousey and Wilcox, 2005: 9).

Ousey and Wilcox engaged in survey research with 3,690 seventh-grade students from 65 middle schools in randomly selected counties in the state of Kentucky. The students participating in the project completed the survey between March and May of 2001. The final sample size was 2,904 students following the deletion of surveys with missing data. Measurements of the degree to which the students in the sample engaged in violence were obtained through a series of questions requesting information on their levels of participation in various forms of violence while at school. Students were asked how often they had assaulted, sexually assaulted or stolen from other students over the course of one school year. In order to measure pro-violence values on an individual level, respondents were asked about their level of agreement with the following statements: 1) “In order to gain respect from friends, it is sometimes necessary to beat up on other kids.” 2) “It is alright to beat up on another person if he or she started the fight.” 3) “It is alright to beat up on another person if he or she called you a dirty name.” 4) “Hitting another person is an acceptable way to get him or her to do what you want.” (Ousey & Wilcox, 2005: 10). In addition, indicators of pro-violence values at the school level were obtained through averages of individual responses for students within each school, or the school-level mean. The researchers then employed both individual scores and the school-level means to estimate any distinctions amongst schools with regard to pro-violent values. Finally, the researchers controlled for exposure to violent peers, impulsivity, parental attachment and school attachment.

Ousey and Wilcox found little support for the thesis that a school subculture of violence was responsible for the violent behaviour of the students in their sample. Their findings also question the central proposition of the subculture of violence with regard to the relationships between violent values and violent behaviour. Although they did find that students who were more approving of and justified the use of violence did indeed commit more violence, the researchers assert that this relationship only appears to be significant at the individual level. It does not appear, the author’s conclude, that the violent value context of the school exerts much of an impact on individual violent behaviour. These findings contradict Felson et al.’s (1994) conclusions, which found the opposite to be true. Finally, as hypothesized, Ousey and Wilcox found that controlling for low self-control and impulsivity impacted the strength of the relationship between violent values and violent behaviour. Based on this finding, the authors conclude that “the impact of violent values is somewhat exaggerated when violent peers and impulsivity or low self control are excluded” (Ousey and Wilcox, 2005: 15).

Finally, in contrast to Ousey and Wilcox’s conclusions, a study conducted by Berburg and Thorlindsson (2005) found support for Felson’s et al.’s (1994) findings regarding the significance of social control processes in perpetuating subcultural violence. Berburg and Thorlindsson examine Wolfgang and Ferracuti’s assertion that individuals within a group might be motivated to engage in aggressive behaviour due to the influence of social control stemming from others’ adherence to conduct norms. Like Felson et al. (1994), the authors explore whether those presumed to be enmeshed within a subculture of violence experience social pressures or expectations to engage in physical aggression for the purposes of showing courage or protecting status. In examining this aspect of the subculture of violence thesis, the authors measure violent values, conduct norms, and aggression on both an individual and aggregate level. By distinguishing between individual and group values, the researchers were able to test the influence of subcultural values while controlling for personal values.

Berburg and Thorlindsson distributed surveys to Icelandic adolescents during one school day in March of 1997. Their sample consisted of 1,493 boys and 1,448 girls in 49 public schools, 32 of which were located in a more urban district of Reykjavik. All of the research participants were 15–16 years of age. Berburg and Thorlindsson’s study of the subculture of violence departs from previous research, including Wolfgang and Ferracuti’s original thesis, in two primary ways. First, they break from the assumption that aggression is a masculine characteristic through the inclusion of girls in their sample. Second, the researchers undertake their study amongst a racially and structurally homogenous group of youths.

The researchers tested the impact of neutralization and retribution values as well as group conduct norms. Whether the research participants embraced neutralization values was determined by their agreement or disagreement with the following statements: “Sometimes there are situations that justify people being beaten up or hit,” and “When someone treats me badly, I think it’s okay to beat up him/her or hit him/her” (462). Researchers assessed participants’ retribution values through asking whether it was appropriate or a “good thing” to forgive people or respond kindly when someone either treats them badly or gets angry with them (462). Finally, through determining a group’s conduct norms, Bernburg and Thorlindsson hoped to examine the extent to which high school students believed status and reputation to be contingent upon aggressive behaviour. Perceived conduct norms were assessed through the agreement or disagreement with the following question: “He/she who does not respond to a personal attack by hitting or beating up the person is considered a coward in my group of friends” (462). Through distinguishing between conduct norms and personal values, the researchers hoped to acquire a better sense of the influence of peer pressure on aggressive behaviour. Finally, the authors tested aggressive behaviour by asking research participants how often they engaged in a range of threatening and physically violent acts, such as fighting, kicking, punching and threatening with a weapon.

Bernburg and Thorlindsson found support for the subculture of violence proposition that group acceptance of neutralization values promote aggressive behaviour in part due to the individual internalization of these norms. However, they found no evidence to support a link between retribution values and violent behaviour. Finally, the authors found that the impact of conduct norms on aggressive behaviour was significant, particularly for boys, lending support to the hypothesis that group pressure to respond to personal attacks with aggression or violence acts a form of social control. These findings suggest that adolescent boys often engage in aggressive behaviour for the purposes of protecting and maintaining their reputations amongst their friends and within the wider community. With regard to gender differences, Bernburg and Thorlindsson found that conduct norms and neutralization values were far more influential in relation to male aggression. However, their data suggest that these factors play a role in adolescent female aggression as well.

Occupational Subcultures of Violence

In addition to examining whether or not the subculture violence thesis could be employed as an explanation for violent behaviour within the school environment, researchers have also wondered whether this framework could account for concentrations of violence within occupational subcultures. Smith (1979) considers this possibility in relation to professional hockey players in one of the earliest studies on the subculture of violence thesis. More recently, Surratt et al. (2005) examine the applicability of this thesis in relation to sex work. Unlike the school studies reviewed earlier, it can be argued that the social and cultural settings examined in these evaluations are for more appropriate terrains for assessing the relevance of the subculture of violence thesis, given that both professional hockey and sex work are characterized by high rates of violence when contrasted with dominant culture. However, both deviate from Wolfgang and Ferracuti’s initial prescriptions for establishing the existence of subcultures of violence. Surrat et al. (2005) are primarily concerned with female sex workers’ experiences with violence and the degree to which violence is normalized in their social sphere. They do not consider the concept of disputatiousness or the use of violence to maintain or protect honour, nor do they offer any evidence to support or challenge a direct link between values and violence. A strength of Smith’s (1979) study is his research site: machismo is an organizing principle within “jock” culture. Although violence is not explicitly encouraged amongst professional hockey players, at the same time, “violence works as an occupational tool and expresses moral character” (Smith, 1979: 239). Smith, however, does not analyze the subculture of violence thesis amongst a structurally marginalized population.

Hockey Violence and the Subculture of Violence Thesis

Smith (1979) employs interviews and survey research to examine the applicability of the subculture of violence thesis in relation to professional hockey players in Toronto, Ontario. His sample consisted of three different populations of men between the ages of 12 and 21. While the first two consisted of house-league hockey players and competitive hockey players, the third comprised non-hockey players. A total of 520 hockey players and 180 non-hockey players participated in the study, which was conducted in April of 1976. The interviewees in all groups hailed from various socio-economic backgrounds.

Participants’ attitudes towards violence, specifically teenage fighting, were assessed through agreement with the following questions: “Are there any situations you can imagine, not counting sport, in which you would approve of a teenage boy punching another teenage boy” (241)? Those who agreed with this statement were then asked whether they agreed with a boy fighting to defend himself from ridicule, in response to a challenge to fight, or in reaction to being shoved by another boy. In order to assess attitudes towards hockey fighting specifically, respondents were presented with the same scenarios, but with hockey players as the hypothetical assailants. Finally, the researchers asked the respondents how many street fights they had been in during the three years prior to their study in order to obtain information on participation in violence. The hockey players in the sample were asked to provide additional information on the number of fights they had been in during the 1975–1976 season. Official game reports indicating penalties for fighting were also considered.

In keeping with the subculture of violence thesis, Smith found that the hockey players in his sample exhibited values and attitudes far more permissive of violence than the non-hockey players did. He also noted a significant relationship between the hockey players’ pro-violent attitudes and values and participation in violence. Smith interprets this finding as supportive of Wolfgang and Ferracuti’s proposition suggesting a causal correlation between violence and pro-violent values and attitudes. However, he does not view his findings as supportive of the subculture of violence thesis in its entirety. Smith counters Wolfgang and Ferracuti’s assertions that subcultural affiliates are more likely to hail from marginalized socio-economic backgrounds through his findings, which indicate that, although the hockey players who were from a lower socio-economic background had participated in more hockey fights, they were in fewer street fights than their higher SES counterparts were (Smith, 1979: 244).

Sex Work, Drug Use and Subcultures of Violence

Surratt et al. explore the subculture of violence theses in relation to female street sex workers in Miami, Florida. In so doing, the authors assess whether the disproportionate amount of violence sex workers experience can be attributed to the fact that they live and work within a subculture of violence. The authors interviewed and conducted focus groups with 325 women, between 2000 and 2001, in order to examine their experiences of violent victimization. The target population for the study consisted of active, drug-using female sex workers between the ages of 16 and 49. In order to qualify as “active,” research participants had to have traded sex for money or drugs at least three times a week for a period of 30 days and used heroin or cocaine at least three times a week during this time. Surratt et al. employed a targeted sampling plan, which involved mapping the neighbourhoods with the highest concentration of prostitution, and relied on active sex workers to recruit participants for their study. Interviews were administered using three data collection instruments: the NIDA Risk Behaviour Assessment, the Childhood Trauma Questionnaire and the Georgia State University Prostitution Inventory. Focus groups were conducted following the interviews for the purposes of contextualizing findings.

With regards to demographics, the mean age of the sample was 38.1. Of the 325 women who participated in the study, 60.3 per cent were African-American, 23.4 per cent were Caucasian and 12.9 per cent were Latina. Most interviewees considered themselves homeless, though some were staying in shelters or at the homes of friends and acquaintances temporarily. Less than half of the sample had obtained a high school diploma and the vast majority lived on less than $1,000 per month. On average, the career span for most of the sex workers in the study was 15.8 years. Alcohol and cocaine were reported to be the most widely used substances amongst this group.

Research participants were asked questions about childhood violence, as well as experiences of violent victimization that had taken place in the past year. The rates of both were extremely high. With regard to childhood experiences, 44.9 per cent of the women reported a history of physical abuse, 50.5 per cent advised of sexual abuse and 61.8 per cent reported emotional abuse. In relation to current incidents of violence, 24.9 per cent reported being beaten, 13.8 per cent advised being threatened with a weapon and 12.9 per cent indicated being raped. In all of these incidents, clients or “dates” were the offenders. Surratt et al. view these figures as supportive of the subculture of violence thesis, given that the rates of sexual and physical assaults reported amongst this group are 43 and 13 times the national averages, based on figures reported in the National Violence Against Women Survey. Surrat et al. interpret their modest but consistent correlation between childhood and current victimization as indicative of a sex work subculture of violence. The researchers found that women who had experienced the most traumatic and severe incidents of physical, emotional and sexual abuse as children also reported the highest rates of violent victimization as adults. Focus group data illustrated that the vast majority of women who reported incidents of violence were rather fatalistic about their experiences. Most regarded the violence in their lives as an inevitable aspect of their work. Based on these findings, Surratt et al. conclude that street sex workers are enmeshed within a subculture of violence. However, whether their evidence is sufficient to support this claim is debatable. As they acknowledge, this study does not include a systematic assessment of values or attitudes towards violence. In addition, Surrat et al. are more concerned with their research participants’ victimization experiences, rather than with social or cultural supports for committing violence. Despite these shortcomings, their findings raise important questions about gender in relation to offending and victimization when assessing subcultural supports for violence.

Disputatiousness in the South and the Inner City

This section will address the subculture of violence thesis as an explanatory device in relation to violence in the Southern United States and the inner city as addressed in Nisbett and Cohen’s, The Culture of Honor (1997), and Anderson’s The Code of the Streets (1999) respectively. Both works offer comprehensive examinations of the concept of disputatiousness. In these studies, the high rates of violence in the inner city and the South are theorized as resulting from cultural codes that permit the use of violence in situations where property, honour and family are at stake. In keeping with Wolfgang and Ferracuti’s original prescriptions for analyzing violent subcultures, Nesbitt and Cohen approach the examination of disputatiousness in the South through quantitative methods, particularly survey research and an array of social psychological tests, which they employ to assess attitudes and values towards violence. In The Code of the Streets, Anderson (1999) conducted ethnographic research to explore the lives of residents in an inner-city Philadelphia neighbourhood in order to gain insight into the meaning and function of violence in their lives. Unlike The Subculture of Violence, both The Culture of Honour and The Code of the Streets present arguments regarding the genesis of the violent cultures in the South and the inner city. Nesbitt and Cohen present a historical explanation, tracing Southern proclivities for violence to the herding economy, which was brought to the area by Scotch-Irish settlers in the late 17th century (Nisbett and Cohen, 1997: 8). For Anderson, the code of the streets is a cultural adaptation to structural constraints, a theorization that is also implicit in the culture of honour thesis. Both studies find conclusive proof for subcultures of violence in the South and the inner city.

The Culture of Honour: The Psychology of Violence in the South

According to Nisbett and Cohen (1997), cultures of honour tend to develop in societies where individuals are at severe risk of losing their resources. Due to a combination of limited state protection and an economic system generative of few resources, the authors contend that a form of frontier mentality emerges. As a consequence, citizens, particularly men, feel the need to embrace hyper-masculine characteristics to ensure the protection of their property, families and themselves. Within this context, reputation, an unwillingness to tolerate insults and the ability to impose one’s will on others are critical to socio-economic well-being. Those who are known or appear to be capable of protecting their resources and themselves are less likely to become the victims of theft and violence (Nisbett and Cohen, 1997: xvi). Considering these conditions, the use of violence for the purposes of protection becomes culturally permissible and, to a certain degree, a necessity.

Nisbett and Cohen note that in the 17th and 19th centuries, cultures of honour existed in various parts of the South and had developed in congruence with herding economies imported to the area by Scotch-Irish settlers. Despite the passing of such lawless times, they argue, the cultural conditions associated with this way of life have persevered and continue to guide the actions and behaviour of the descendents of these early settlers today. After disputing some of the most common explanations for the high rates of Southern violence, such as the legacy of slavery and the hot climate, they examine whether the subculture of violence thesis is an appropriate explanation for the high rates of violence in states such as Kentucky, Tennessee, and West Virginia.

Despite the overlaps between the culture of honour and subculture of violence theses, Nisbett and Cohen depart from Wolfgang and Ferracuti’s propositions regarding the locales and populations presumed to be involved in subcultural violence. Rather than focusing on racialized populations in the inner city or disorganized neighbourhoods, Nisbett and Cohen focus on the high rates of violence amongst Southern white males.

To examine the appropriateness of a cultural explanation for violence in this region, the authors perform a multitude of comparative experiments to determine whether white northern and southern men vary in the degree to which they embrace pro-violent values and norms. Attitudes towards disputatiousness were captured through survey data and questions exploring approval for violence in hypothetical situations depicting incidents of violence committed in the name of honour. Noting the potential disparities between conduct and attitudes, they supplement their study with tests to determine whether southern males do in fact act and perceive confrontation differently from their northern counterparts. Their overall questions are the following: “Do young southern white men today really care about insults? Is their masculinity really put on the line when they are affronted, as survey results would imply? Are they truly prepared to take steps to restore the balance after someone offends them” (Nisbett and Cohen, 1997: 39)?

To answer these questions, the authors performed a series of social-psychological experiments, which involved testing variations in perceptions and reactions towards insults and shoves amongst a sample of northern and southern college students at the University of Michigan. The most noteworthy and intricate experiment involved a staged interaction, in which a confederate of the researchers bumped into each research participant while walking down a hallway and swore at him (Nisbett and Cohen, 1997: 42). Evidence of disputatiousness and concerns about damaged reputation were determined by the following: assessments of whether the research participants were aroused or stressed, which were in turn assessed through variations in levels of cortisol and testosterone; examinations of each participant’s behaviour following the interaction; and finally, the participants’ emotional and cognitive responses. The authors found support for the culture of honour hypothesis: insulted southern participants experienced the greatest increase in cortisol and testosterone levels, expressed far more concerns about being perceived as masculine during an interaction with an individual who had observed the staged altercation, had firmer handshakes, and tended to express more hostility than their insulted northern counterparts did. The same finding applied to a control group composed of southern men who were not insulted (Nisbett and Cohen, 1997: 41–50). The researchers interpret such findings as conclusive support for the culture of honour thesis and indicative of values of “self-protection, sensitivity to insult and a willingness to take matters of punishment into one’s own hands” (Nisbett and Cohen, 1997: 57).

Nisbett and Cohen supplement these experiments with macro-level analyses to assess any institutional supports for violence in the South. In so doing, they observe focus on collective expressions of culture as reflected in laws, social policies and institutional behaviour, relying on both field experiments and archival data to do so. The authors point out that the law and social institutions are critical to shaping the behaviour of citizens by defining what is acceptable and unacceptable. Not only do they reflect individual values, they also “feedback” and shape them (Nisbett and Cohen, 1997: 58). Accordingly, they investigate a variety of policies, including: laws that embody norms of self-defence, particularly those in relation to gun control and national defence; policies on corporal punishment and domestic violence, which they consider to be reflective of attitudes relating to violence for the purposes of social control; and corporal punishment, which, they argue, captures collective attitudes towards the use of violence for retributive purposes. Based upon the voting habits of legislators as well as an array of statistics and opinion polls, Nisbett and Cohen conclude that the South is indeed more favourably disposed toward capital and corporal punishment and far more concerned with national defence than the North is, and is generally opposed to gun control. All of this, they argue, is conclusive evidence that a culture of honour exists in the South.

Nisbett and Cohen found additional evidence for their thesis following a series of field experiments conducted to determine the existence of institutional supports for a culture of honour in the South. One notable study examined perceptions and reactions of various northern and southern employers to a fictional job candidate. The hypothetical job candidate is depicted as appealing for consideration despite his conviction of manslaughter for killing a man who was having an affair with his partner. The researchers crafted this experiment by composing a letter confessing details of the crime, which they sent along with a resume to various employers in both parts of the country. As they had anticipated, the southern employers were far more sympathetic to the plight of the job candidate than their northern counterparts were (Nisbett and Cohen, 1997: 75).

Finally, in an attempt to examine media reinforcements of the culture of honour, Nisbett and Cohen examine how various college newspapers in the South, West and North would write news stories about a fictional murder committed in response to provocation and for the purposes of maintaining reputation and honour. After distributing an outline of the bare facts of the case to the universities involved in the study, the researchers reviewed the stories they produced, combing them for evidence of sympathy and blame. Nisbett and Cohen found that “compared to northern papers, southern and western papers...were more likely to emphasize provoking circumstances than aggravating ones” (Nisbett and Cohen, 1997: 77). Over all, they found that newspaper coverage in the West and South portrayed the offender more favourably than the news stories written by the northern students did.

Nisbett and Cohen provide compelling empirical proof that the high rates of southern violence amongst rural white males in the South can be attributed to subcultural supports for violence committed in the name of honour. A significant methodological strength of this study is that it reaches far beyond survey data and considers not only the attitudes and values of research participants, but also whether institutions reflect cultures of honour.

More recently, Hayes and Lee (2005) revisited Nesbitt and Cohen’s culture of honour thesis and found partial support for their study. The authors relied on data from the GSS, which allowed them to examine the relationship between growing up in the South and an increased likelihood to engage in violence amongst a national sample of over 2,500 US households. Hayes and Lees analyzed the responses of research participants to a series of questions assessing their approval of the use of violence for the purposes of self-defence and maintaining honour and respect. The researchers predicted that individuals who reside in the South, as well as those who were socialized in southern states, would be more supportive of violence. The findings of the study indicate that “Southern white rural males, in particular, will be more approving of violence than other demographic groups when used in defence of honour, family or personal property” (Hayes and Lee, 2005: 613).

The Code of the Streets: Disputatiousness in the Inner City

Perhaps the most well-known, controversial and widely examined ethnographic exploration of the African-American subculture of violence thesis is Anderson’s (1999) A Code of the Streets. Anderson’s work is notable for its use of ethnography and narrative to examine the applicability of the subculture of violence framework to an inner-city, low-income African-American neighbourhood in Philadelphia. Unlike Wolfgang and Ferracuti, Anderson does not infer the existence of subcultures of violence based on statistical indicators of high rates of violence in poor racialized neighbourhoods. Rather, he attempts to explore in detail the meanings of and motivations for violence amongst those who are thought to be enmeshed within the subcultures of violence. Anderson’s key question is, “why it is that so many inner city young people are inclined to commit aggression and violence towards one another” (Anderson, 1999: 90)? Hoping to elucidate the cultural and social dynamics that foster internecine violence in the urban core of Philadelphia, Anderson engaged in four years of participant-observation and in-depth interviews with residents in a neighbourhood along Germantown Avenue.

According to Anderson, the high rates of violence amongst inner-city residents can be attributed to a “code of the streets.” This code, he notes, functions as a “set of informal rules governing interpersonal public behaviour” that encourages the use of violence for the purposes of maintaining honour and defending reputation (Anderson, 1999: 33). Anderson notes the importance of adopting an image that they are not to be “messed with” amongst street-oriented youth. In order to account for the fact that only a very small minority of residents in Germantown actually engage in “street” behaviour, Anderson identifies a second code, which he terms the “code of civility.” Despite their shared social and structural location, unlike their “street” counterparts, “civil” residents are not so quick to engage in violence, preferring instead to adopt “middle-class values” of decency, hard work, responsibility, pacifism, and religion. While both contingents experience the hardships of race and class oppression, “rather than dwelling on the hardships and inequities facing them,” Anderson argues, “civil” individuals tend to “accept mainstream values more fully than street families” and make the best of what they have (Anderson, 1999: 38). Thus, in keeping with Wolfgang and Ferracuti’s (1967) framework, violence is understood to be a consequence of a “street code,” which results in the internalization of pro-violent norms and values amongst those who are active participants in a subculture of violence. These individuals are theorized as rejecting dominant or “middle-class” values in favour of a code of behaviour that valorizes subcultural norms of aggression, hyper-masculinity, idleness and debauchery. While the majority of the sub-proletariat African-Americans within the neighbourhood – the civil residents – reject this way of life, they constantly feel the “pull” of the street code, and are forced to defend themselves and their values as they venture into public spaces on a daily basis.

Building upon Nisbett and Cohen’s (1997) theorizations, Anderson recognizes Germantown Avenue’s inner city as a culture of honour. He theorizes the violence occurring in this context as a form of capital that African-American men employ to demonstrate hyper-masculinity and ensure the preservation of their property and family. This masculine performance is crucial to maintaining status, self-preservation and respect. As Anderson notes, the associations of violence, honour, and respect have a practical application: if “others have little regard for a person’s manhood, his very life and the lives of his loved ones could be in jeopardy” (Anderson, 1999: 91). This instrumental use of violence is closely intertwined with the illegal drug market, which forms the basis of the underground economy in the inner city and operates in opposition to legitimate legal structures. For those individuals involved in this economy, violence is a form of street justice, and appearing tough is critical to their economic survival. Thus, disputatiousness is a norm in the inner city. As Anderson notes, “Many of the forms dissing can take may seem petty to middle class people (maintaining eye contact for too long, for example), but to those invested in the street code, these actions, a virtual slap in the face, become serious indications of the other person’s intentions” (Anderson, 1999: 34).

One major contribution of Anderson’s study is that it draws from a variety of traditions in criminological thought to explain the origins of subcultures of violence. Anderson elaborates on the etiology of the code of the streets, borrowing from Sampson and Wilson’s (1995) theorizations of the relationships between race, crime and inequality. Cultural proclivities for violence are conceptualized as adaptations to structural constraints and social disorganization, which in turn result from race and class oppression. Anderson points to the havoc wreaked in the area due to global economic changes, asserting that deindustrialization and the loss of manufacturing jobs have generated an “oppositional culture” and “ideology of alienation” amongst inner-city African-Americans (Anderson, 1999: 111). This is a departure from Wolfgang and Ferracuti’s circular framework, which assumes that violence is a reflection of pro-violent values.

Despite his avoidance of tautological reasoning and his attempt to account for origins of subcultural violence, an important criticism of Anderson’s work is that he fails to clarify the specific processes that had led the residents of Germantown Avenue’s inner city to embrace pro-violent values and attitudes (Wacquant, 2002). By attributing differences in motivation to commit violence to values, he raises several questions about the relationships between social structure, violence and masculinity. For example, how did the polarized “decent” and “street” groups emerge in the first place, and what exactly led to the development of their distinct value orientations? How and why do some inner city residents succumb to the pull of the streets while others somehow manage to resist it? Are there other social structural differences between these factions, besides race and class, that are causing the disparity and fuelling the violence amongst “street” individuals? Despite his rich ethnographic data and narrative approach, none of these questions is addressed in Anderson’s account.

Over the past few years, Anderson’s code of the street’s thesis has also faced some criticism due to the fact that it rests upon assumptions acquired through qualitative research methods. As a consequence, many have questioned the validity and generality of his findings. In an attempt to address this shortcoming, Brezina et al. (2004) conducted an extensive review of quantitative studies on youth violence in order to assess the validity and generality of Anderson’s claims. In their review, the authors tested two of Anderson’s key hypotheses. First, they explored whether adherence to a code that encourages retaliation for the purposes of protecting and maintaining status is associated with racialized, low socio-economic and socially disorganized neighbourhoods. Second, they examined whether adherence to the code is related to neglectful and abusive parenting, experiences with violence and victimization, exposure to violent peers, and a general belief that legitimate means to attain status or respect are unavailable (Brezina et al., 2004: 307). The authors found that previous studies were largely consistent with Anderson’s work. They conclude that “prior studies have also established links between violent definitions and many of the socio-environmental factors highlighted by Anderson, including parenting practices, peer relations, neighbourhood context, and various demographic factors such as SES and urban residence” (Brezina et al., 2004: 319). However, previous research did not support the contention that the code is solely an inner-city black phenomenon.

Brezina et al. also found support for Anderson’s more specific hypotheses in relation to adherence to the code. The authors examined whether those who had experienced high levels of victimization and youth who perceived a lack of future opportunities through legitimate means, such as school and the labour market, were more likely to use violence to protect and maintain their honour. Three waves of data from a National Youth Survey, a self-report study on delinquent behaviour, were analyzed to assess these claims. The survey was based on a probability sample of youths between the ages of 11 and 17 and had been conducted in the late 1970s. The study sampled households within two Philadelphia neighbourhoods and interviewed youths within the selected households. Brezina et al. found that “youths who perceive a lack of future opportunity and those who report relatively high levels of prior victimization are more likely to than other youths to become associated with aggressive peers” (Brezina et al., 2004: 322). These associations, in turn, increased levels of violence. The authors found that exposure to violent peers increased the likelihood of engagement in violence for the purposes of retaliation to real or perceived slights, a finding which is consistent with Ousey and Wilcox’s (2005) study reviewed earlier.

Finally, in another recent examination of Anderson’s code of the streets thesis, Stewart and Simon (2006) employ two waves of survey data from 720 African-American youth from over 259 neighbourhoods in the United States to explore whether neighbourhood context, family of origin and experiences of discrimination influence adoption of the street code. The authors tested three of Andersons’ key hypotheses. First, they examined whether individuals growing up in a “street” family are far more likely to adopt the code of the streets than individuals growing up in a “decent” family are. Next, they examined whether youths who experience racism are more likely to embrace the code. And finally, the authors observed whether there is a relationship between neighbourhood disorganization and violence and adherence to the street code (Stewart and Simons, 2006: 8). The researchers found support for Anderson’s conclusions and a significant relationship between adoption of the street code and family background, experiences of racism, and neighbourhood context (Stewart and Simons, 2006: 24). While being raised in a “street” family contributed to the likelihood of adherence to the code, a “decent” family background appeared to mute its impact. However, in contrast to Anderson’s suggestion that the code is an inner-city African-American phenomenon, Stewart and Simon’s findings illustrate that this is not the case.

Conclusion and Policy Implications

The subculture of violence thesis has undergone considerable evaluation since its emergence in the mid-20th century. Though Wolfgang and Ferracuti (1967) had outlined a methodology for the scientific evaluation of subcultures, their claims were based on inferences and statistics rather than on actual knowledge of the values and attitudes of individuals presumed to be enmeshed within a subculture of violence. Various propositions of the subculture of violence thesis have been reviewed, with mixed results. Both Ball-Rokeach (1973) and Erlanger (1974) insist that there is little empirical support for Wolfgang and Ferracuti’s claims and question their key assertion that violent values lead to violent behaviour. However, it is debatable whether these studies capture and evaluate the central tenets of the subculture of violence thesis appropriately.

More recent evaluations show support for Wolfgang and Ferracuti’s propositions concerning the role of social control processes in perpetuating subcultural violence rather than the notion that the use of violence emerges from violent values. Studies illustrate that far more predictive of violent behaviour are group norms that lead to the instrumental use of violence amongst subcultural affiliates for the purposes of impression management and maintaining reputation. These findings question the assumption that all individuals engaged in subcultures of violence actually approve of violent actions. Both Felson et al. (1994) and Berburg and Thorlindsson (2005) found this to be the case amongst middle school and high school students in the United States and in Iceland. Smith’s (1979) study of professional hockey players lends support to this conclusion as well. However, in a replication of Felson et al.’s study, Ousey and Wilcox (2005) found little support for the contention that group norms play a significant role in the proliferation of violence. While they do find a relationship between violent values and participation in violence, especially at an individual level, they also insist that claims about the impact of violent values have been exaggerated, given that prior research does not control for exposure to violent peers.

Interest in the instrumental use of violence has led to revisions of the subculture of violence thesis. The culture of honour is one variation of this thesis that has been the subject of considerable qualitative and quantitative examinations. This concept elaborates on Wolfgang and Ferracuti’s proposition on disputatiousness, which not only emphasizes the use of violence for maintaining honour and reputation, but also considers the sanctions associated with failing to comply with norms around violence. Nisbett and Cohen (1996) find considerable support for their hypothesis linking the high rates of violence in the South to a southern culture of honour. Similarly, Anderson (1999) concludes that high rates of violence within poor, socially disorganized, predominantly black inner-city neighbourhoods can also be attributed to a culture of honour. Unlike Wolfgang and Ferracuti (1967), both works provide theories regarding the genesis of cultures of honour. In Nisbett and Cohen, the southern culture of honour is believed to be an artefact of the herding economies brought to the area by Irish and Scottish settlers between the 17th and 19th centuries. Despite socio-economic transformations since this time, norms and values more permissive of the use of violence to maintain honour are believed to have survived. According to Anderson, the culture of honour that rules North Philadelphia’s Germantown Avenue can be linked to structural constraints and social disorganization, both of which are a result of race and class oppression. Recently, Brezina et al. (2004) and Stewart and Simon (2006) tested a variety of Anderson’s claims, due to the fact that his findings were accrued through ethnographic observation. Both studies found considerable support for his hypotheses regarding socio-economic marginalization and embracing the code of the streets.

Finally, most of these studies indicate that subcultures of violence are not unique to economically marginalized or racialized communities, as Wolfgang and Ferracuti had suggested. In contrast, Smith (1979) found that the hockey players in his study who hailed from privileged backgrounds were far more likely to engage in fights than the non-privileged ones were. Similarly, in their re-evaluation of Anderson’s work, Stewart and Simon (2006) found no significant relationship between racial group and cultural supports for violence. Erlanger (1974) was led to similar conclusions based on his analysis of white and African-American populations in the United States. Finally, Nisbett and Cohen (1996) argue that the culture of honour in the South can only explain the high rate of violence amongst white southern men.

The policy implications of subcultural theories focus on targeting the consequences of socio-economic marginalization. In order to improve the lives of sex workers, for instance, Surrat et al. recommend programs that aim to promote safer work environments, assistance in finding alternatives to the street, and sensitivity training for the police and other criminal justice officials. The importance of ensuring access to community-based treatment facilities is also noted. With regard to the incidence of subcultural violence in black and African-American neighbourhoods, Oliver (1994) advocates the implementation of community-based interventions to disrupt social and cultural supports for violence (Oliver, 1994, as discussed in Washington, 1997). He argues that coalitions involving community, religious, and business leaders should work together with local politicians and the private sector to fund and design programs for black youth.

Conflict resolution, mentoring, educational and recreational programs are also listed as potentially empowering solutions; however, observers have noted that in order for these programs to be effective, these interventions must consider the fact that young, at-risk black men “need to simulate behaviour that is valued in the ‘streets’” (Washington, 1997: 12). In addition, it is important that those teaching conflict resolution or leading mentoring programs be perceived as role models for at-risk youth. It has been argued that such efforts will most likely be ineffective if those leading the programs are not admired or respected by participants (Washington, 1997: 13). Along with these recommendations, early intervention is seen as critical to preventing violence in socially disorganized and impoverished neighbourhoods. Identifying at-risk youth early on, it is argued, would allow for community leaders to devote additional resources and attention to those who need it the most. Finally, Oliver (1994) also proposes advocacy efforts to ensure that all those living in socially disorganized and violent neighbourhoods are aware of the geographic areas in which violence generally occurs.

In sum, research indicates that high concentrations of violence amongst specific populations and geographic locations can be attributed to a variety of social and cultural forces. While macro-socio-economic processes play a role, collective norms that govern smaller groups, such as schools, also appear to generate violence. Further research on the subculture of violence thesis could benefit from more qualitative and ethnographic analysis of the specific factors that lead individuals down paths of violence and nonviolence. Doing so could provide greater insight into how certain individuals become more or less immune to norms and values of violence, despite a shared structural position or geographic location.


Anderson, Elijah. (1999). The Code of the Streets: Decency, Violence and the Moral Life of the Inner City. New York: WW Norton.

Ball-Rokeach, Sandra J. (1973). Values and violence: A test of the subculture of violence thesis. American Sociological Review, 38(6), 736−749.

Berburg, Jon Gunnar and T. Thorlindsson. (2005). Violent values, conduct norms and youth aggression: A multilevel study in Iceland. The Sociological Quarterly, 46, 457−478.

Brezina, Timothy. (2004). A quantitative assessment of Elijah Anderson’s subculture of violence thesis and its contributions to youth violence research. Youth, Violence and Juvenile Justice, 2(4), 303−328.

Erlanger, Howard S. (1974). The empirical status of the subculture of violence thesis. Social Problems, 22(2), 280−292.

Felson, Richard, A.E. Liska, S.J. South and T.L. McNulty. (1994). The subculture of violence and delinquency: Individual vs. school context effects. Social Forces, 73(1), 155−173.

Hayes, Timothy C. and Mattew R. Lee. (2005). The Southern culture of honor and violent attitudes. Sociological Spectrum, 25, 593−617.

Nisbett, R.E. and D. Cohen. (1996). Culture of Honor: The Psychology of Violence in the South. Boulder, CO: Westview Press.

Ousey, Graham C. and Pamela Wilcox. (2005). Subcultural values and violent delinquency: A multilevel analysis in middle schools. Youth Violence and Juvenile Justice, 3, 3−22.

Sampson, Robert and William Julius Wilson. (1995). Toward a theory of race, Crime and urban inequality. In John Hagan and Ruther Peterson (Eds.), Crime and Inequality (pp. 37−54). Stanford: Stanford University Press.

Smith, Michael. (1979). Hockey violence: A test of the violent subculture hypothesis. Social Problems, 27(2), 235−247.

Stewart, Eric and Ronald Simons. (2006). Structure and culture in African American adolescent violence: A partial test of the “code of the street” Thesis. Justice Quarterly, 23(1), 1−33.

Surratt, Hilary L., James A Inciardi, Steven P. Kurtz and Marion C. Kiley. (2004). Sex work and drug use in a subculture of violence. Crime and Delinquency, 50(1), 43−59.

Washington, Earl M. (1996). A survey of the literature on the theories of violence and its prevention. Journal of Negro Education, 65(4), 403−07.

Wolfgang, Marvin and Franco Ferracuti. (1967). The Subculture of Violence. London: Tavistock.

8 This section was prepared with the assistance of Rashmee Singh, PhD candidate, Centre of Criminology, University of Toronto.


Volume 1. Findings, Analysis and Conclusions

Volume 2. Executive Summary

Volume 3. Community Perspectives Report

Volume 4. Research Papers

Volume 5. Literature Reviews