Studies of the Youth Gangs, Drugs, and Violence Connection
The relationship between drugs, drug trafficking, and violent crime is the subject of much debate and research (see De La Rosa, Lambert, and Gropper, 1990, for an exhaustive review). Goldstein (1985) suggested three possible relationships: (1) the “pharmacological” effects of the drug on the user can induce violent behavior; (2) the high cost of drug use often impels users to commit “economic compulsive” violent crime to support continued drug use (e.g., robbery for the purpose of securing money to buy drugs); and (3) “systemic” violence is a common feature of the drug-distribution system, including protection or expansion of the drug distribution market share, retaliation against market participants who violate the rules that govern transactions, or maintenance of the drug-trafficking organization.
Collins (1990) summarized the research evidence supporting each of the three types of drug violence Goldstein suggested. First, there is virtually no evidence of the pharmacological effects of drugs (excluding, perhaps, alcohol) on violence. Second, there is considerable evidence of a relationship between drug use and economic compulsive violence. Third, although research is scarce on “systemic” (drug distribution) violence, this form appears to be the most predominant. “Drug distribution system violence tends to occur (at least most visibly) in areas that: are socially disorganized, that is, in which formal and informal social control is absent or ineffective; have traditionally high rates of interpersonal violence; and are economically disadvantaged” (Collins, 1990:266). Collins noted that the Goldstein typology has its limitations, mainly because there are other important sources of violence. This is an especially important point with respect to the gang context. A review of these other sources is divided into two parts: gang homicide and the causes and correlates of youth gang violence.
Youth Gang Homicide and Drug Trafficking
Although youth gang homicides are characterized by periodic spurts and declines, they have been increasing nationwide and evidence an overall growth trend in certain cities (Maxson, 1998a). These spurts are explained largely by “turf” disputes between warring gangs (Block and Block, 1993; Block and Christakos, 1995; Block et al., 1996). The spurts are not citywide—they occur in specific neighborhoods and involve particular youth gangs in escalating incidents of provocation, retaliation, and revenge. The annual number of homicides involving Chicago street gangs increased almost fivefold between 1987 and 1994 (Block et al., 1996). Youth and adult gang-related homicides in Los Angeles County more than doubled from 1987 to 1992, then dropped in 1993 and 1994 (Maxson, 1998a).
To what extent is the large volume of and increase in gang homicides caused by drug trafficking? This popular assumption is tied to the image of youth gangs as entrepreneurial drug-trafficking operations that began to spread across the country during the crack cocaine epidemic.
Klein and his colleagues were the first researchers to test the popular assumption of a strong relationship between youth and adult gang drug trafficking and homicide. In a series of Los Angeles studies, they found that the connection between gang-related homicides and drug trafficking is not strong.13 This relationship has also been found to be weak in several other studies in Boston, Chicago, Los Angeles, Miami, and St. Louis (see Howell, in press[a] for a review of this research).
There are exceptions to this general conclusion. Some ongoing drug market wars account for a significant number of homicides (Block et al., 1996). Block and her colleagues also noted an indirect relationship among homicides, drug trafficking, and street gang activity. Many of the street gang-related homicides might not occur without the existence of drug markets, which routinely bring members of opposing gangs into contact with one another. How can the increase in gang related homicides over the past decade be explained?
The role of firearms in gang violence. The presence of firearms significantly increases the likelihood of murder. The routine use of guns in gang conflict is a fairly recent development, having occurred in the past decade (Miller, 1992). Recent studies show that firearms are now prevalent in youth gangs (Bjerregard and Lizotte, 1995; Howell, 1998; Lizotte et al., 1994). There also is evidence that the impact of drug selling on illegal gun carrying is greater than the impact of gang membership and that drug selling increases with age. Thus, “unlike the diminished role of gangs, drug selling grows as the subjects get older and this enhances hidden gun carrying” (Lizotte et al., 1997:388). A strong association is found between illegal gun use and gang membership and between illicit drug sales and illegal gun use (Decker, 1996; Decker, Pennell, and Caldwell, 1997; Sanders, 1994; Sheley and Wright, 1993, 1995).
Using data gathered from interviews in 1995 with arrested juveniles in the Drug Use Forecasting (DUF) study, Decker and colleagues (1997) found that gang members are much more likely than other juveniles to carry guns most or all of the time (31 percent versus 20 percent). Percentages of arrestees who reported using a gun to commit a crime were higher among adolescents who sold drugs (42 percent) or belonged to a gang (50 percent) than among other juveniles (33 percent). One-third of gang members said it was okay to shoot someone who disrespected them. These findings confirm the importance of gun ownership and use among gang members.
In a 3-year field study of active youth gang members in St. Louis, Decker and Van Winkle (1996) reported that 81 percent owned guns. The mean number of guns owned was more than four. Two-thirds of gang members had used their guns at least once. The most common use was in gang fights; infrequent use was reported in drive-bys, defense against attacks by strangers, and other incidents. Only four members mentioned a drug-related motive. In each of these incidents, the gang members used their guns to prevent a drug customer from robbing them.
Decker (1996) contended that gang interactions, mainly the threat a rival gang presents, help to explain the increasing sophistication of weapons used by gang members. The Blocks showed that most of the increase in Chicago street gang homicides is attributable to an increase in more lethal weapons, not an increase in assaults (Block and Block, 1993; Howell, in press [a]; Hutson et al., 1995; Zimring, 1996). Rosenfeld and Decker (1996:200) found that the St. Louis youth (under age 24) homicide problem “is largely a gun homicide problem.”
Causes and Correlates of Youth Gang Violence
Some studies support the notion that youth and adult gang involvement in drug trafficking has led to more violent crime.14 Other studies suggest that the connection between youth and adult gang drug sales and violence is indirect or weak.15 Some of these studies that shed light on the gangs, drugs, and violence connection are reviewed below.
Huff (1996) studied two samples of Cleveland adolescents: currently or formerly active youth gang members and a second group of youth who had not joined gangs but were deemed similarly at risk of delinquency. Major Cleveland gangs were well represented in the sample. Gang youth were significantly more involved in marijuana and cocaine drug sales and in more serious and violent crimes than nongang adolescents. Gang members were far more likely to sell high-profit drugs and to sell drugs more frequently than nongang adolescents. Huff asked both groups about the source of the drugs they sold. Gangs were not the primary source for either group. A majority of both gang and nongang youth said “others” controlled drug supplies. Gang sellers were far more likely than nongang sellers to go out of State for their supply.
In a unique aspect of this study, police gang experts identified 83 gang members who were leaders in 1986. Huff (1996) compiled their arrest histories from 1980 to 1994. The overwhelming majority of arrests (which averaged 10 per leader) began at or near the time of their initial gang involvement. Most of the arrests (37 percent) were for violent crimes, 29 percent for property crimes, 18 percent for drug offenses, and 6 percent for weapons offenses.
In his investigation of possible crime progression, Huff (1996) determined the year in which gang leaders’ arrests for property, drug, and violent offenses peaked. Peaks for all three offenses clustered within less than 2 years. His discovery that violent crime arrests peaked about 3 months before drug offenses led Huff (1996:99) to suggest that this might be evidence of “a close connection between drug trafficking and violence that is often associated with conflict over ‘turf.’”
Venkatesh’s (1996) ethnography of gangs in Chicago’s Robert Taylor Homes described one of the worst cases of gang drug trafficking and violence. His study documented the transformation of gangs in this low-income public housing development from turf gangs to drug gangs and the escalation of gang violence with the advent of crack cocaine. In the 1960’s and 1970’s, these gangs fought over pride or turf in hand-to-hand conflicts, sometimes using zip guns (homemade, single-shot pistols). Their violence was controlled largely by tenant networks. When crack cocaine was introduced in the 1980’s, a notable escalation in gang violence occurred. Several gangs controlled drug-trafficking turfs in one or more buildings in the housing development. Previously contained fights then burst into the open, endangering residents in gang-related crossfire. In 1992, several children, all innocent bystanders, were shot and killed. Neither police nor tenant organizations were able to contain the gang violence. Rival gangs continued fighting. Eventually, community leaders, youth workers, and tenants were able to effect a truce that Venkatesh predicted would not last.
Hagedorn (1991, in press) found that few (mostly adult) Milwaukee gang members were involved in cocaine sales in 1987. But by 1991, 75 percent of them were reported as having been involved in cocaine trafficking. Adult gang members said that one-half or more of the dope houses in gang neighborhoods were run by gangs (Hagedorn, 1994b). He estimated that about one-quarter of all homicides and from one-third to one-half of all adult gang violence in which gang members were involved or which they witnessed were drug related (Hagedorn, 1996).
In one of the most detailed studies of the gangs, drugs, and violence connections, Decker and Van Winkle (1994, 1996) found that the St. Louis gangs to which youth belonged, mostly local Crips and Bloods, were extensively involved in drug trafficking, especially cocaine. Members of these gangs fought often, generally using guns. Ammunition, drugs, and guns were sometimes obtained from gangs in Los Angeles and Detroit. Rival gangs often and Van Winkle found, however, that gang violence has many other sources related to everyday gang social processes.16 They saw three main sources of violence among St. Louis gang members (Decker and Van Winkle, 1996). First, violence is a part of everyday life in their neighborhoods and families. Second, conflict differentiates gangs from other delinquent groups. Third, violence is an endemic part of their status as individuals and as gang members. In St. Louis gangs, “members are expected to always be ready to commit violence, to participate in violent acts, and to have engaged in some sort of violence in their initiation” into the gang (Decker and Van Winkle, 1996:173).
Decker (1996) offered a more detailed explanation of the origin and spurt pattern of gang violence that Block (1993) discovered. He used Loftin’s (1984) “contagion” concept and the notion that gang cohesion grows in proportion to the perceived threat represented by rival gangs (Klein, 1971). Loftin argued that three conditions must be present if contagion is to occur: a spatial concentration of assaultive violence, a reciprocal nature to assaultive violence (see Miller, 1958), and escalations in assaultive violence. Decker (1996) explained how the threat of attack by another group ignites the gang, increases cohesion, and produces deadly consequences. Most gang violence, he argued, is retaliatory, a response to violence—real or perceived—against the gang. Spurts of gang violence appear to follow predictable patterns, in a sequence that is initially motivated by the perceived threat that another gang poses, then instigated by a precipitating event, followed by escalation of activity, a violent event, rapid deescalation, and finally, retaliation.
Long-Term Studies of Adolescent Samples
Most of the studies reviewed thus far focus on specific gangs or individual gang members, capturing the significance of their experiences. A different view of the connection between gang drug trafficking and violence is obtained by studying large representative samples of adolescents over a long period of time. OJJDP’s Program of Research on the Causes and Correlates of Delinquency, which studied large adolescent samples in the emerging gang cities of Denver, CO, and Rochester, NY, has produced a number of important findings on the gangs, drugs, and violence connection. Although these studies were not designed specifically to examine youth gangs, they permit comparisons between gang and nongang members in larger samples.
Each of these studies addresses the extent to which gang membership facilitates drug trafficking. Similar patterns were observed in both cities. In Rochester, Thornberry and his colleagues (1993) found that gang members were involved in three to five times as many drug sales as nongang youth in sequential time periods. In Denver, gang members reported nearly seven times as many drug sales as nongang youth (Huizinga, 1997). In another study, supported by OJJDP and several other agencies and organizations, Seattle gang members reported involvement in 10 times as many drug sales as nongang youth (Hill, Howell, and Hawkins, 1996). In Seattle (Hill et al., 1996) and in Rochester (Bjerregaard and Lizotte, 1995), drug use and trafficking rates still remained high after individuals left the gang, indicating that gang influence on drug trafficking extends beyond the period of gang membership. Gang members in all three study sites reported from three to seven times as many serious and violent delinquent acts as nongang youth (Howell, 1998).
A key question is, Does gang involvement in drug trafficking cause subsequent violent crime? The Seattle gang studies have examined this issue. Despite a high prevalence of Seattle gang member involvement in drug trafficking, accelerated adolescent involvement in drug trafficking after joining a gang, and strong evidence that gang involvement prolongs drug trafficking (Hill, Howell, and Hawkins, 1996; Hill et al., 1996), an analysis shows that gang member involvement in drug trafficking at age 16 does not predict assaultive violence at age 18 but does predict drug trafficking at age 18 (Howell et al., 1996). Surprisingly (given this finding), the study also showed that drug trafficking at age 16 predicts significantly more assaultive violence and handgun possession at age 18 among nongang youth.17
In Denver, Esbensen and Huizinga (1993:571) reported that drug sales “were not driving” street offending. Both violent (gang fighting, rape, robbery, and aggravated assault) and nonviolent offenses (burglary, theft, fencing stolen goods) composed the “street offending” measure. Although Rochester analyses showed an association between gang drug trafficking and violent offenses (Thornberry et al., 1993), neither the strength of the relationship nor the temporal order of the two behaviors has yet been examined. Several other studies of either gang or nongang samples have shown an association between adolescent drug trafficking and violence.18
These findings make a persuasive case that drug trafficking is strongly associated with other serious and violent crimes but not necessarily that drug trafficking by gang members causes more frequent violent offending. In Pittsburgh—the third site in OJJDP’s Program of Research on the Causes and Correlates of Delinquency—a study of nongang youth suggested that drug use, serious theft, and violence precede drug selling (Van Kammen, Maguin, and Loeber, 1994). Van Kammen and her colleagues also found that sales of illicit drugs started significantly later in adolescence than the other three behaviors. Initiation of drug selling was strongly related to previous involvement in multiple types of delinquency. The authors concluded that “the present study indicated a temporal sequence between the delinquent behaviors and the onset of drug dealing. This does not mean that the relationship is causal. Instead, it is likely that drug dealing and serious forms of delinquency are expressions of similar antisocial tendencies. Whether the same etiological factors apply to each still remains to be demonstrated” (Van Kammen, Maguin, and Loeber, 1994:240).
Although a causal relationship between gang drug trafficking and violence has not yet been demonstrated in the above studies, it is important to remember that, in the main, the findings this Bulletin reviews come from two sources: gang studies in emerging gang cities and nongang samples. A key question is the extent to which gang membership facilitates gun use in drug trafficking—possibly resulting in higher levels of violence—in the same way that the gang facilitates overall violent offending. This may hold true in two cases; gang member drug trafficking may indirectly contribute to more violent encounters with other gangs involving guns when (1) drug trafficking exacerbates the need for guns and (2) the perceived threat of violence from rival groups increases. Resolution of this connection requires further examination.
13 See Klein and Maxson, 1985; Klein, Maxson, and Cunningham, 1988, 1991; Maxson, 1995, 1998a.
14 See Hagedorn, 1996; Padilla, 1992; Sanchez-Jankowski, 1991; Sanders, 1994; Short, 1996; Skolnick 1989, 1990, 1991; Skolnick et al., 1988; Taylor, 1989, 1990; Venkatesh, 1996.
15 See Block and Block, 1993; Chin, 1990, 1995, 1996; Decker, Pennell, and Caldwell, 1997; Decker and Van Winkle, 1996; Esbensen and Huizinga, 1993; Fagan, 1989; Huff, 1989, 1996; Klein, Maxson, and Cunningham, 1991; MacLeod, 1987; Maxson, 1995; Maxson and Klein, 1996; Moore, 1990, 1991; Waldorf and Lauderback, 1993.
16 See also Anderson, 1994; Block and Block, 1993; Chin, 1996; Decker and Van Winkle, 1996; Horowitz, 1983; Kennedy, Piehl, and Braga, 1996; Sanchez-Jankowski, 1991.
17 The researchers selected drug selling at age 16 and violence and other outcomes at age 18 in part because the average ages for joining a gang are 14 to 15 in Seattle. Thus, it was anticipated that gang membership and involvement in gang-related drug trafficking would be very prevalent by age 16. Measuring violence at age 18 would allow time for gang drug trafficking to cause violence—if that were the case.
18 See Altschuler and Brounstein, 1991; Dembo et al., 1993; Padilla, 1992; Van Kammen and Loeber, 1994; Williams, 1989.