Keywords

The issue of exactly what constitutes a gang is fraught with debate and lack of consensus across researchers and policy makers (see Wood and Alleyne 2010 for a fuller discussion). However, the Eurogang network has captured the elements broadly agreed by many researchers and policy makers by stating that, “a street gang (or troublesome youth group corresponding to a street gang elsewhere) is any durable, street-oriented youth group whose identity includes involvement in illegal activity” (Weerman et al. 2009, p. 20). The important elements of this definition are that gangs generally consist of young people who are collectively involved in illegal activities on the streets which create problems for communities and authorities. To address the problems that gangs create, a whole host of strategies have been developed to focus on, for example, prevention, intervention and suppression of gangs and their activities. The problem remains, however, that despite the concerted efforts of justice systems, gangs appear to be flourishing. This leaves criminal justice systems struggling to devise effective programs that will contain the disturbing menace that gangs pose in communities, schools, and prisons and the threat that they pose to individual gang members—on both a physical and psychological level. The current discussion considers programs that aim to deter existing gang activity and the political policies that underpin them: Gang prevention programs, although perhaps including elements of deterrence, will not be included since deterring current gang membership is not their key feature. The aim of this chapter is to provide an overall evaluation of deterrence strategies and to examine more closely the psychological effects that group processes may have in insulating gang members against the intended effects of legal deterrence efforts. Although gang membership is a worldwide phenomenon, the focus in this chapter will be on programs and policies that have been devised and implemented in the U.S. and the UK.

Deterrence

Deterrence has two forms: general and specific. General deterrence refers to the idea that nonoffending populations will be deterred from committing offenses because of the punishments that may follow committing crime. Specific deterrence refers to punishments designed to dissuade individual offenders from committing further offenses (Stafford and Warr 1993). However, for deterrence methods to be effective, they rely on the offenders being rational actors who make rational choices about their involvement in crime (i.e., rational choice theory; see Clarke and Cornish 1985). They make these choices by weighing up the pros [which may be material (e.g., financial), mental (e.g., thrill) and/or reputational (e.g., peer approval, status enhancement; see Matsueda 1992)] and the cons [which may be legal (e.g., imprisonment), financial (loss of earnings—fines), emotional (e.g., feeling shame) and/or reputational (e.g., loss of respect from significant others)]. If more benefits than costs from crime are anticipated then offenses will be repeated. For instance, research shows that if young people anticipate that they will experience thrills from committing acts of theft they are likely to commit more thefts, and if they perceive that a valued status such as being seen as ‘cool’ can be gained via theft and/or violence then they maintain an intention to commit further acts of theft and/or violence (Matsueda et al. 2006). On the other hand, if the costs of crime are perceived as outweighing any anticipated benefits then both youth and adults are likely to make a rational choice not to commit further crimes (Zimring and Hawkins 1973). In short, it is the anticipated costs of crime that should act as a deterrent to future offending.

In order to quantify these rational choices, the Expected Utility Model of crime explains that a calculation of expected costs of crime may be defined as the probability of getting caught multiplied by the anticipated punishment, whilst the expected benefits are defined as the probability of getting away with the offense multiplied by its anticipated gains (Becker 1968). Injected into this equation is the probability parameter, which is a subjective evaluation of an individual’s perceived chances of being caught and punished for the offense. In other words, an essential element of an expected utility calculation is what the individual perceives to be the potential costs and rewards of criminal activity. For this calculation to effectively reduce crime by deterring offending it is necessary that people’s rational choice calculi favor the perceived costs rather than the perceived benefits side of the equation, and that they anticipate a real likelihood of being caught and punished for an offense. In short, the expected risks associated with crime must be perceived to outweigh the expected returns (Becker 1968).

The rational choice calculus used for decision making may include several estimations of offense outcomes. Research findings support a rational choice perspective of offending by showing that, for youth, if they believe that they will be arrested for committing acts of violence or theft then they will commit fewer such offenses—provided that they view arrest as a bad outcome (Matsueda et al. 2006). However, this same piece of research also showed how it is the perceived benefits of crime that are the ultimate drivers of rational choices. As such, potential offenders may pay little attention to the anticipated costs of offending. Nonetheless, individuals’ risk perceptions are not static; they are fluid and malleable and updated when new information comes to light. For example, research shows that juvenile offenders judge their risks of arrest according to their previous experiences. That is, if they have previously been arrested they are more likely to increase their risk perceptions to levels higher than they had before arrest (Anwar and Loughran 2011). However, this updating process may have limitations. As Anwar and Loughran note, updating risk perceptions seems to relate primarily to the crimes for which they had been arrested (e.g., acts of violence) rather than to criminal activity generally or crimes for which they had not been arrested (e.g., income-generating offenses).

Exactly why this is so, is not clear. It could be that offenders believe that because they are already known to the police for committing a specific type of offense, they are likely to be obvious suspects following similar offenses and as a result consider it wise to avoid committing such offenses. Equally, it could be that offenders know the penalties for committing offenses for which they have already been convicted but have little knowledge of the punishments associated with other offenses. What does seem clear is that policies that aim to deter offenders from any form of offending seem far less likely to hit the mark than policies that target specific offense types (Anwar and Loughran 2011). Also, as the number of crimes committed by an offender increases, a corresponding drop occurs in the effect of arrest on risk perceptions even though it remains positive. As a result, compared to less experienced offenders, more experienced offenders are likely to have weaker perceptions of risk across time and experience.

The idea that risk perceptions emerge from a systematic cognitive model bounded by rationality is intuitively appealing but it is not without its problems. For instance, it is well documented in psychology that people are cognitive misers (Fiske and Taylor 1991) who frequently resort to heuristics (Tversky and Kahneman 1974) when making decisions. People are also prey to the gambler’s fallacy (Matsueda et al. 2006) in which they assume that if something has happened many times recently then it is less likely to occur as frequently in the future. So, offenders using a gambler’s fallacy as a decision base for committing crime may erroneously judge that if, for instance, they have escaped arrest then they are due to be caught, or if they have been arrested several times then they are due to get away with an offense (Pogarsky and Piquero 2003).

The source of information on which risk perceptions are founded is also an important feature in decisions to offend. Some may gain an idea of arrest certainty from popular media which depicts the restoration of justice via the consistent arrest, prosecution, and punishment of offenders (Parker and Grasmick 1979). Such deterrent effects stemming from a diet of media consumption are, however, more likely to be effective with naïve individuals with no familiarity of criminal justice systems than it is with more seasoned offenders who factor in personal experience (Matsueda et al. 2006).

Deterrence effects may also be influenced by a trade-off between immediate rewards and delayed punishments (Apel 2013). For example, the rewards of crime are relatively immediate whereas criminal sanctions are likely to follow lengthy criminal proceedings and fall far in the future. Deterrence may also be negatively influenced by ‘dimensional preferences’ which strongly push individuals towards the potential benefits of crime than towards its potential costs (Carroll 1978). Even without the above potential influences, a rational choice perspective assumes that people base their decisions on an accurate knowledge of facts such as knowing which punishments follow which crimes. In reality people are not good at judging the likelihood of being caught and equally poor at knowing the penalties associated with offending (Kleck et al. 2005).

Anti-gang Programs

Findings overall indicate that the strongest deterrent effects develop more from the certainty of apprehension rather than from the severity of punishment (e.g., Klepper and Nagin 1989), even when people are aware of which penalties follow which crimes. Indeed, research shows that police presence can be effective in preventing crime (Braga 2005). Some claim that high police presence and visibility increases offenders’ perceptions of risk of apprehension and has such a substantial deterrent effect that it justifies shifting resources from imprisonment to policing (Durlauf and Nagin 2011). Consequently, it could/should be expected that high police presence together with an increased likelihood of punishment would have particularly high deterrent effects on those involved in high levels of offending, such as gang members. The remainder of this section first outlines and then evaluates two of the main approaches used to deter gang membership and activity and continues by considering the development of anti-gang policies that enhance the criminality of gang membership.

Suppression Strategies

High police presence and swift criminal justice responses provide the bases for suppression strategies aimed directly at deterring gangs from criminal activity. Suppression projects began in the USA during the 1970s and developed over the next few decades to include a wide range of programs such as: Operation Hardcore, Operation Hammer, and Operation Safe Streets (Los Angeles); the Gang Congregation Ordinance (Chicago); the Anti-gang Initiative (Dallas); and the multi-site Youth Firearms Violence Initiative (Los Angeles, Milwaukee, Seattle, and San Antonio). The goal of suppression tactics is to respond swiftly and forcibly to gang activities in order to deter gang members from continued criminal activity. To achieve prosecutorial success, some programs (e.g., Operation Hardcore) included components such as special gang prosecution units and vertical prosecution. Vertical prosecution consists of the allocation of a special gang prosecutor to a case from its outset to conviction and results in an increase in prosecution and conviction rates (OJJDP 1994). Other tactics included suppressing gang members’ ability to associate with one another by threatening them with arrest if they were seen together in public places. For instance, the Chicago Gang Congregation Ordinance stated that,

Whenever a police officer observes a person whom he reasonably believes to be a criminal street gang member loitering in any public place with one or more other persons, he shall order all such persons to disperse and remove themselves from the area. Any person who does not promptly obey such an order is in violation of this section… (Chicago Municipal Code § 8-4-015 June 17, 1992).

Violation of this section was punishable by a fine of up to $500, imprisonment of up to six months and/or up to 120 h of community service. Other programs specifically targeted the most dangerous gangs and most dangerous gang members. For example, the Tri-Agency Resource Gang Enforcement Team (TARGET) in California would identify dangerous gang members, monitor their activities, arrest and prosecute them using vertical prosecution and then closely supervise them whilst on parole to deter them from committing further offenses. Arrest swiftly followed if they did so. In Dallas the Anti-Gang Initiative (1996 and 1997) targeted seven of the city’s most prominent gangs. The program’s tactics included high visibility of police in gang areas, curfews for suspected gang members, and hard-hitting enforcement of any truancy legislation (Lafontaine et al. 2005).

Cooper and Ward (2008) observed that in some areas, suppression tactics resembled full military campaigns. For example, in Los Angeles, police initiated the Community Resources Against Hoodlums (CRASH) during the 1980s which involved ‘gang sweeps’ of gang areas and constant pressure on gangs through intense police patrols. The Order-Maintenance Policing (OMP) project, introduced in New York in 1994, adopted the Broken Window Theory (Wilson and Kelling 1982) approach. The Broken Window Theory proposes that if community standards are not maintained (e.g., broken windows are not mended quickly, litter is allowed to accumulate) then a community becomes vulnerable to criminal activity. The solution lies in community members and police cooperating by attending to minor community problems which should then avert an escalation of anti-social behavior. Using these methods, crime should be prevented, or at least confined whilst at a low level. The OMP project initially aimed to operate via collaborations between justice officials and community members and to favor alternatives to arrest such as education and counselling—all of which is in line with Broken Window theory tenets. In reality, OMP employed aggressive tactics of stop-and-frisk actions to detect less serious crime and gun possession (Fagan et al. 1998) and rejected alternatives to arrest and prosecution (Waldeck 2000).

Multifaceted Strategies

Also in the USA, some programs with a focused deterrence perspective adopted a problem-oriented policing approach which was thought to be particularly effective in reducing gang activity (Braga et al. 1999b). Although facets of this approach included suppression strategies, problem-oriented policing has a more multifaceted approach to tackling gangs by identifying a problem, analyzing it, devising an appropriate response and then evaluating that response to see if adjustments are necessary. For example, during the 1980s and 1990s projects such as The Boston Gun Project and Operation Ceasefire took a broad range approach (e.g., also targeting firearm traffickers who supplied gangs with weapons) to increase gang members’ perceived risk of apprehension and prosecution. The main targets of these interventions were gang youth well known to the authorities and who generally had a history of convictions (Kennedy et al. 1996) and whose violence was driven by inter-gang disputes (Braga et al. 1999a). The program’s aims were outlined to gangs via fliers, street work projects, and interactions with probation department staff. The suppression element of the message was that violence (particularly gun violence) would not be tolerated and a ‘pulling all levers’ approach (i.e., every criminal justice resource available would be used against those involved in violence) was adopted to suppress and deter gang activity. It was hoped that this level of suppression would encourage gang members to ‘police’ each other due to the threat that acts of violence committed by one individual would result in a police crackdown on all members of the gang.

However, unlike suppression-only programs, multifaceted programs included support services such as: probation and parole services, police services, prosecution branches (e.g., the Office of the U.S. Attorney), youth services (e.g., Juvenile Corrections), federal agencies (e.g., the Bureau of Alcohol, Tobacco, and Firearms), and community outreach programs/services. These multi-agency approaches encouraged gang members to take advantage of social service support to access: employment opportunities, mentoring schemes, housing support, substance abuse treatment programs, and vocational training opportunities (McGarrell et al. 2006)—all aimed at helping them to relinquish their gang lifestyles.

This ‘carrot and stick’ (i.e., using persuasion and force simultaneously) approach of multifaceted programs in tackling gangs appeared to be successful, and so other, similar problem-oriented policing programs, functioning on comparable deterrent tenets as Operation Ceasefire, were adopted in Richmond, Chicago, Los Angeles and Indianapolis. For example, the Chicago Alternative Policing Strategy (CAPS 1992) program, just as Operation Ceasefire had done, adopted a democratic police/citizen interactive approach to crime reduction. Police concentrated on their beat areas and got to know residents and the areas’ problems. Meetings were held between residents and police to consider urgent crime-related needs and an advisory body of community leaders presented larger concerns to the area commander. In short, a democratic approach, involving all areas of the community, was adopted to tackle gang activity.

Program Efficacy

Zealous suppression approaches within criminal justice responses do appear, on the face of it, to have achieved at least some of their desired deterrent effects. Operation Hardcore increased charge and conviction rates with fewer dismissals of cases involving gang members. It was later judged as having achieved clear improvements in justice responses to gang defendants and their cases (Dahman 1983). However, its effects on gang-related crime was not evaluated specifically so we cannot know if gang-related crime reduced as a result of the program (Cooper and Ward 2008). In California, TARGET identified 570 gang members and 77 gang leaders during its first two years. Of these, two-thirds were held in custody, 99 % were convicted and there was an associated 62 % reduction in gang-related crime and a 47 % decrease in gang crime over seven years (Cooper and Ward 2008). In Dallas, an analysis of its anti-gang initiative showed that gang-related violence decreased significantly during 1996 and 1997 in the targeted areas (Howell 2000). Similarly, OMP in New York was associated with a dramatic reduction in crime and was met with a euphoria of support (Harcourt 1998), heralded as the Holy Grail of the 1990s (Jones 1997) and “…the new path of deterrence” (Kahan 1997, p. 2479).

On the other hand, some programs did not fare so well. Operation Hammer deployed 1000 police officers to conduct nightly gang sweeps and carry out mass arrests of gang members, but of the 1435 people arrested, nearly half were not gang members and 1350 were released without charge (Klein 1995). The operation, which was never formally evaluated, was subsequently abandoned in the face of public criticism (Klein and Maxson 2006). In Chicago, a lack of clarity regarding what constituted loitering resulted in the Gang Congregation Ordinance being accused of enabling arbitrary and discriminatory targeting by the police. It was eventually ruled by the U.S. Supreme Court (City of Chicago v. Morales 1999) as unconstitutional because its lack of clarity regarding what citizens may or may not do could curtail harmless behavior and thus violate people’s constitutional rights. More recent research also shows that even if laws successfully target gang members they do not necessarily have the desired gang reduction effects, because arresting gang leaders can increase violence by triggering competition among adjacent gangs who aim to take over the territory of leaderless gangs (Vargas 2014).

Multi-faceted programs also appear to have had some success. For instance, the Chicago Alternative Policing Strategy (CAPS 1992) programme appeared to be promising, as there was a drop in crime rates. However, crime rates that showed a downward trend up until 2007 showed an increase in 2008 (Murphy 2008). An evaluation of Operation Ceasefire conducted by Braga et al. (2001) was lauded as demonstrating that the program was associated with a 63 % reduction in youth homicide in Boston (Wellford et al. 2005). However, as Wellford et al. (2005) also note, in the absence of a random controlled trial, it is difficult to attribute the reduction in youth homicide to the program. Also, other authors argue that the downward trend in gun homicides noted during Operation Ceasefire continued only until 1997; in 1998 it began to rise again (Fagan 2002). Fagan argues that since gun homicides had similarly begun to decline during 1995 in other, non-program cities in Massachusetts, the trends noted in Boston cannot be attributed specifically to Operation Ceasefire. Further, attempts to replicate Operation Ceasefire in other jurisdictions (e.g., Los Angeles and Indianapolis) offered no evidence that the deterrence message embedded in programs had any influence on gang members’ behavior (Greene and Pranis 2007).

Researchers have pointed out problems with the facilitation of programs. In many areas, even when a multifaceted approach to tackling gangs was adopted, problem-oriented policing was often not practiced and police rarely sought citizen input and rarely shared with, or valued information provided by, non-criminal organizations (Katz and Webb 2003). Instead, an emphasis was put on the suppression elements of programs as police departments adopted aggressive responses as a result of pressure from citizens to address the problem with gangs. However, this backfired, as hard-liner approaches often resulted in accusations that the police used excessive force, subjected citizens to unnecessary stop and search routines, and subjected citizens’ children to harsh physical treatment.

Other suppression efforts by the police included the dispersal of gangs, and these were a universal failure (Katz and Webb 2003). For instance, in Las Vegas, the demolition of a public housing development resulted in gang members and their activities being scattered across the city. This led to a worsening of the existing gang problem because gang members continued with their previous behavior in their new residential areas. Katz and Webb (2003) further observed that in the jurisdictions they examined there was little police training, little monitoring, and no police accountability. Suppression tactics generally operated by increasing special gang unit patrols of gang areas. These special units were, however, composed of police officers who knew little about gang members, had little direct contact with gang members, and who made so few arrests that there was little impact on crime in the area. It is understandable that the police, when under the extreme pressure of public outrage about gang activity, may prioritize suppression strategies. However, what is concerning is that although gang unit police and key stakeholders believed that suppression activities justified the existence of gang units, it was only the gang unit police who believed that suppression efforts reduced gang problems (Katz and Webb 2003). Findings such as these have led some researchers to claim that an imbalance exists between the carrot and the stick in multifaceted approaches. For instance, Greene and Pranis (2007) noted that two-thirds of resources used for gang reduction had gone into suppression initiatives in Los Angeles and that “‘Balanced’ gang control strategies have been plagued by replication problems and imbalances between law enforcement and community stakeholders” (p. 6).

However, the delivery of programs is likely to differ according to jurisdiction and even program content. Braga and Weisburd’s (2012) systematic review and meta-analysis of focused deterrence considered the findings of 11 evaluations of programs that employed focused deterrence strategies to reduce crime and observed that although ten programs reported statistically significant crime reductions, a lack of randomized experimental evaluations generated a tendency to overstate positive outcomes. Nonetheless, Braga and Weisburd’s review provides overall support for deterrence principles and notes that taking a multifaceted approach to youth violence generally—and gang violence more specifically—can dramatically hinder violence. Braga and Weisburd add that focused deterrence strategies succeed in adhering to procedural justice principles of mutual respect and consideration by creating a liaison between police and communities to increase social controls, to deflect offenders away from crime and consequently succeed in accomplishing a reduction in crime in the target community. This approach also helps to develop police legitimacy amongst the local community by increasing public awareness that citizens accept the rule of authority (Skogan and Frydl 2004) which, in turn, decreases the likelihood that people will break the law (Paternoster et al. 1997).

However, Braga and Weisburd (2012) observe that focused deterrence programs, which function both on enhancing individuals’ risk perceptions through increased police presence and on encouraging pro-social behavior, make accurate program evaluation problematic. For instance, when suppression strategies are coupled with support (e.g., from youth workers, probation and parole officers, churches and community groups), evaluations of program efficacy are rather muddy and it is difficult to untangle the effects of complex program elements to identify exactly what effect each component has had. As multifaceted programs encourage gang members to take advantage of social service provision, employment opportunities, mentoring, housing, substance abuse treatment programs, and vocational training opportunities (McGarrell et al. 2006), we cannot draw firm conclusions about which elements of the program had deterrence effects. In their meta-analysis, Pratt et al. (2006) note how the effect sizes of deterrence are substantially reduced—sometimes to zero—when other factors such as peer influence, antisocial attitudes and self-control are controlled for in analyses. These authors go so far as to say that:

…the clear drop in predictive power of the deterrence variables from bivariate to multivariate models suggests that empirical support for the effect of formal sanctions on individuals’ criminal behavior is most likely an artefact of the failure to control for other “known” predictors of crime/deviance (p. 384).

Therefore, we cannot say with any certainty that the positive effects observed in some multifaceted programs occurred because they heightened individuals’ risk perceptions or because they provided opportunities that enticed gang members into more prosocial activities, such as jobs. Indeed, the provision of increased support for youth in terms of social service provision, mentoring, housing, substance abuse treatments, and employment and vocational training may indicate that some of the core risk factors that have historically been linked to gang membership (i.e., social deprivation and exclusion—see Klein 1995) are being addressed in focused deterrence programs, or at least they are given more attention. However, it is assumed that it is deterrence, rather than increased opportunities that lead to program success. Either way, it is difficult to effectively gauge exactly what effects lead to reduced gang activity. This is because evaluations of the effects of support and increased opportunities are too often dogged by substantial methodological flaws (e.g., narrative reviews or descriptions of programs). This then prevents even the most rudimentary conclusions being drawn about program effectiveness (see Fisher et al. 2008, for a systematic review).

An evaluation that seems to offer more support for the effectiveness of deterrence principles, albeit with nongang populations, was conducted with a program that included strategies aimed at low-risk offenders on probation who were not substantial drug or alcohol abusers (Project MUSTER; Weisburd et al. 2008). To conduct the evaluation, Weisburd et al. (2008) used a randomized experimental design to compare the effects of the program on three groups of offenders. The first group underwent intensive probation, threats of violation to court, and incarceration and community service (Project MUSTER intervention group), the second group underwent normal probation supervision, and those in the third group were threatened with only one part of the MUSTER treatment, violation of probation (VOP—for non-payment of fines). Findings showed that probationers in the MUSTER cohort and those in the VOP group were more likely to pay their fines than the probationers in the normal supervision group. MUSTER and VOP participants did not differ from each other. The authors justifiably claim that their findings show that if threats of sanctions are associated with non-payment of financial obligations then offenders are more likely to be compliant with legal sanctions.

The Weisburd et al. (2008) study suggests that threats of legal sanctions do act as a deterrent with offenders. However, the offending population involved in MUSTER was, as noted, low-risk offenders. This then begs the question of how effective such strategies are likely to be with more criminally entrenched offenders. Gangs are widely accepted as being deeply engrained criminal groups, whose members are disproportionately involved in crime, particularly serious and violent crimes (Thornberry and Krohn 1997), and there is something about gang membership that promotes higher rates of offending whilst in a gang than before or after membership (Thornberry et al. 2003). It seems unlikely, then, that the promise that the MUSTER approach appears to offer would apply as well to gang members who are deeply embroiled in criminal activity as it does to low-risk offenders. In addition, participants in the MUSTER program were not substantial substance users whereas research indicates that compared to nongang youth, youth in gangs have far higher levels of substance use (Gatti et al. 2005). Consequently we may expect that the need for substances and the associated criminal acts employed to acquire those substances are likely to negate the influence of threats embedded in programs such as MUSTER. In short, analogies between participants in MUSTER and gang members are difficult to see.

Anti-gang Policies

Criminal Enhancement of Gang Activity

To support and complement the intended deterrent effect of anti-gang strategies and special gang prosecution units, anti-gang legislation has been introduced in a number of countries. In the US, at a federal level in (2010), the statute specific to gangs, “Criminal Street Gangs” (18 U.S.C. § 521.), states that anyone participating in a street gang whose members have engaged in a continuing series of offenses such as violence against others, drug offenses, or conspiracy to commit such offenses, may receive sentences of up to 10 years imprisonment. At a state level, as of 2013, all 50 states and the District of Columbia had enacted some form of legislation regarding gangs and gang activity (National Gang Intelligence Center 2013). Of these, 28 had passed gang prevention laws, 30 had laws that provided enhanced penalties for gang-related crime, 26 states and the District of Columbia had legislation relating to gang activity in school settings, and 12 states had laws to deal with gang-related databases. Since so many anti-gang laws have been developed across the US, it is not possible for this chapter to describe them all, so a few illustrative examples will be used. For example, Nevada introduced anti-gang statutes in 1989 and 1991 which made it illegal to: possess weapons on school property; discharge firearms out of vehicles; aim a firearm at a person; discharge a weapon in a way that could endanger people; coerce minors into illegal activity; commit certain violations at or near schools, bus stops, or recreational facilities; and promote criminal gang activity (Lafontaine et al. 2005). In California, Florida, Georgia, Illinois, and Louisiana, the introduction of the Street Terrorism Enforcement and Prevention Act (STEP) enabled police and prosecutors to identify if a gang fits the Act’s definitions and notify individuals categorized as gang members that the gang has been identified and that any subsequent law violations will be treated more harshly than usual (Cooper and Ward 2008).

Civil Gang Injunctions

In addition to gang-specific laws, Civil Gang Injunctions (CGIs; i.e., court-issued restraining orders) were introduced to prohibit and deter gangs from specific legal activities (e.g., loitering at schools, carrying pagers, and riding bicycles) or from entering certain areas. CGIs have been gaining in popularity since the 1990s in the U.S. and in the UK where they were introduced as part of the Police and Crime Act (2009). In the UK, between 2011 and 2014, 88 gang injunctions were put in place in 25 areas identified by the government as Ending Gang and Youth Violence priority areas (Home Office 2014). UK gang injunctions aim to “prevent a person from engaging in, encouraging or assisting gang-related violence and to protect them from gang-related violence” (Home Office 2014, p. 3).

In the US, CGIs predominantly target gangs’ territoriality (Hennigan and Sloane 2013). They typically prohibit/limit movement within known “gang” areas and the scope of the CGI can cover a single street or an entire city. For instance, one injunction in Los Angeles covered 16 square miles (Los Angeles County District Attorney’s Office 2011). The composition of a CGI involves a good deal of flexibility and variation (Maxson et al. 2005). The behaviors listed may be tailored to address the specific behavior of specific gang members in specific areas and/or they may target entire gangs by name. They may also forbid gang members from associating with each other by standing, sitting, walking, driving, bicycling, gathering, or appearing anywhere in public with any other named individual or any other known gang member (O’Deane 2011). Police have the discretion to name 10–30 specific people in the injunction but hundreds of unidentified ‘John Does’ may also be cited—and identified and named at a later date (Myers 2009). All individuals identified as members of a gang issued with a CGI are threatened with arrest if they transgress any part of the order and they face enhanced sentences of ten years following conviction. Between 1980 and 2000, 37 CGIs were issued in Southern California and 22 in Los Angeles (Maxson et al. 2005), and by 2003, 47 % of African American men in Los Angeles County aged between 21 and 24 had been included in the Los Angeles County gang database (Siegel 2003). By 2013, Southern California had 46 CGIs targeting 72 neighborhoods in Los Angeles (Office of the City Attorney of Los Angeles 2013).

The use of CGIs has been criticized for the harsh effects that they have on youth identified as gang members. First, there is no consistency in how gang members are identified (Caldwell 2010) and others have accused officials of using injunctions to criminalize ethnic minorities so as to control their movements (Muniz 2014). Other accusations include that CGIs are used to dehumanize gang members to the point where members of the public take little note when they are shot by police (Muniz 2014). Injunctions are also cited as causing already marginalized individuals to be even further marginalized by being prevented from contributing to positive community activities such as family activities outside the home and even hindering employment prospects (Caldwell 2010). Once labelled as a gang member and subject to a CGI, individuals also have few options. Since CGIs are civil orders, those subject to them are not entitled to a public defender to help them appeal the order unless they are already on probation or parole (Muniz 2014). Consequently, once named in an injunction, those labelled as gang members have an uphill, if not impossible struggle to remove their names from an injunction (Crawford 2009) and in practice few have managed to do so (Hennigan and Sloane 2013). It is not surprising then that some authors argue that labelling an individual as a gang member encourages them to adopt the identity (Huff and McBride 1993).

However, if CGIs achieve the desired deterrent effect then they are likely to continue to be supported in criminal justice circles—regardless of the criticism that they attract. In deterrence terms, the knowledge that emanates from a gang injunction (i.e., that individuals are being closely monitored) should, according to deterrence principles, prevent gang members from involvement in gang-related activities because they will heighten gang members’ risk perceptions. Evaluations of gang injunctions suggest that they do indeed achieve some of the desired reductions in crime. For instance, a comparison of crime data from 14 areas in Los Angeles covered by injunctions with matched comparison neighborhoods found that violent crime fell by 5–10 % in injunction areas during the first year of injunction use. There was also no evidence that crime had been displaced to adjoining areas (Grogger 2002). A Los Angeles Civil Grand Jury examined the effect of gang injunctions and reached similar conclusions in (2004). An additional explanation as to why gang injunctions may reduce crime is that they promote changes in the community. For instance, they are thought to increase community cohesion, collective efficacy, and community members’ motivation to call the police should they feel threatened by gang activity (Maxson et al. 2005). Maxson et al. (2005) also observed a corresponding reduction in residents’ fear of, and intimidation by, gangs. However, of the two gang injunction neighborhoods examined by Maxson et al. (2005), only one experienced these improvements and this was the one which had had a gang injunction the longest. The second neighborhood experienced more gang visibility, more anxiety amongst residents, more social disorder, more property victimization, and less belief that the neighborhood could resolve its problems. Maxson et al. (2005) note that the disparity between the two neighborhoods could be due to: (1) gangs from the original injunction neighborhood spilling their activities to the second neighborhood; (2) methodological issues; or (3) police suppression backfiring and generating gang oppositional defiance (see Jankowski 1991) and/or strengthening gang members’ cohesion as suggested by Klein (1995) which, in turn, increases violent crime (see Klein 2011; Wilson and Chermak 2011).

In California the American Civil Liberties Union (ACLU 1997) analyzed the Blythe Street gang injunction and concluded that the injunction did not result in a reduction in violent crime and/or drug trafficking. Also in California, of the 80 % of gang members who committed offenses after being named in an injunction, over half committed crimes in the injunction area (Hale 2006). This suggests that injunctions do not deter since gang members simply continue to offend either in the injunction area or in adjoining areas (Myers 2009). It is also proposed that even when gang injunctions are effective in deterring gang activity, their impact may not be enduring and so research needs to examine reductions in crime for more than a year (Grogger 2005). Other authors point out that the lack of evidence of any long-term improvements (i.e., longer than one year) begs the question of whether gang injunctions are responsible at all for any crime reduction and that in some communities no significant reduction in crime has occurred, and even where reductions do occur other factors could be responsible for observed changes (Caldwell 2010). Maxson et al. (2005) caution that any gains made by gang injunctions will be continually threatened by the persistence of gang activities and that gang injunctions (i.e., the stick) are an opportunity to bring about change, but positive opportunities (i.e., the carrot) may be necessary before lasting changes to gang activity will be effective.

Grogger (2005), on the other hand, seriously doubts that deterrence effects are responsible for any observed reductions in crime. As Grogger notes, gang members are named in injunctions due to police suspicions that they have been involved in serious crimes such as robberies, drug offenses, or rape, and this means that they have already run the risk of attracting serious and lengthy sanctions. Therefore, Grogger doubts that threatening gang members, who seem willing to run the risk of lengthy sanctions, with lesser threats for transgressing gang injunctions (maximum of 6 months), will deter gang members’ activities. More recent findings support Grogger’s contentions by showing that deterrence principles (i.e., the threat of arrest and punishment) have less influence on gang members than on nongang youth living in CGI areas (Hennigan and Sloane 2013). Instead, Hennigan and Sloane found that CGIs seemed to disrupt the strength of members’ gang identity and because identity mediated the relationship between a gang’s cohesion and its criminal activities, gang members living in a CGI area compared to gang members living in a non-CGI area were involved in lower levels of crime.

The findings regarding the success or failure of CGIs are, to say the least, equivocal. Even if gang injunctions succeed in initially denting gang activity as gang members warily adjust to this new initiative in their communities, there is no evidence so far to suggest that such dents have longevity. Equally, residents in communities where gang injunctions are issued may initially perceive the heightened police interest in gang activity in their neighborhood as a motivation to report gangs and crime, but again, there is no evidence so far that this effect is enduring. As Grogger (2005) suggests, more research is needed to examine the perspectives of residents in gang injunction communities before conclusions can be made regarding the effects of this legislation overall. In short, any effects that injunctions might have might simply wear off over time and we need more research before any firm conclusions can be drawn.

The Doctrine of Joint Enterprise or Common Purpose

A further policy that targets gang members’ activities is the doctrine of joint enterprise or common purpose which is employed in several countries. For the purpose of this chapter, discussion will focus particularly on the UK where this 300-year-old law has undergone something of a popular revival in recent years. Originally introduced to combat illegal practices such as dueling, the doctrine holds responsible all individuals present at the time for a crime and has been enthusiastically applied to gang members. For example, in the UK, gang members (secondary parties) who are present at the time of an offense (e.g., homicide) but who do not participate in the actual crime, can be held as responsible as the gang member (principal offender) who performs the conduct element of the offense (murders the victim). By being held as responsible as the gang member who committed the offense, secondary parties will also be convicted of murder, which in the UK carries a mandatory life sentence. These secondary liability principles—that is, the shared responsibility of principal and secondary parties—can be applied to any offense but are used principally for offenses involving violence. In short, it means that all those present at the scene of a crime may be equally sentenced even if they all had different roles in the offense. There is a great deal of unease regarding what is often called the ‘parasitic liability’ component of joint enterprise as this means that if all those involved in a crime (e.g., robbery) are aware that one of their group may, during the original crime, commit another crime (e.g., murder) then all are guilty of that murder; proof of their participation in that murder is not necessary (McClenaghan et al. 2014). In essence, the doctrine functions on the presumption of group members having full and pre-existing knowledge of each other’s intentions and actions.

In the UK, it is estimated that almost 500 people have been convicted of murder as secondary parties between 2005 and 2013 (The Guardian 2014), but no formal records are kept of the actual numbers of prosecutions and convictions brought under the joint enterprise doctrine (House of Commons Justice Committee 2014). Also in the UK, in 2011, a Justice Select Committee set up to examine joint enterprise practice and effects heard from a number of people that joint enterprise was often used to address gang incidents as it enabled the prosecution of several members of one gang at a time for any member’s criminal activity (McClenaghan et al. 2014). Given that research shows how gangs may include tens, and in some instances hundreds of members (see Klein and Maxson 2006 for a fuller discussion of gang sizes in the USA), it defies reason to assume that all gang members can possibly be fully aware of other members’ intentions. Even though it is unlikely that such vast numbers would be present at the same event, it seems highly unlikely that all who are present, especially if they happen to be peripheral or fringe gang members, will be fully aware of others’ intentions. As Jeremy Corbyn, the Member of Parliament appointed as the Justice Select Committee’s special rapporteur (an independent expert) on joint enterprise and a key figure in the governmental inquiry, stated to McClenaghan et al. (2014):

I came to the issue because of dealing with young people in my constituency who are peripheral members of gangs. These are young people who are peripheral to some, often horrific, incidents. That doesn’t make them all guilty. It makes them in the wrong place at the wrong time, often, but it doesn’t make them all guilty. Unless there is incontrovertible proof that they took an active part in an attack then we end up prosecuting and ruining the life chances of quite a lot of young people, who are frankly bored and hanging around the streets (p. 29).

Research conducted by Eady (2013) for the organization Joint Enterprise Not Guilty by Association (JENGbA) adds to the argument that joint enterprise prosecutions often involve inconsistencies that are difficult to comprehend. In this study, of the 101 people convicted under the joint enterprise doctrine, 70–80 % of those aged under 22 maintained that during trial the prosecution mislabeled them as gang members. Eady argues that there are potentially four explanations for why the prosecution would allege a defendant’s gang involvement: (1) The defendant is a gang member and denying membership is less destructive to their case; (2) Prosecutors perceive gangs are dangerous and responsible for many violent crimes; (3) The prosecution gains a presentational advantage in portraying defendants as gang members; (4) The idea of gang responsibility helps to justify shared responsibility and joint enterprise prosecutions. Interestingly, this study also flagged sentence inconsistencies in which defendants who denied being present at the offense or even knowing the perpetrators received the longest prison sentences (average, 24.54 years) whilst those who admitted being present at the offense, but claimed involvement only in legal activity received an average sentence of 17.65 years, and those who admitted being involved in the offense but to a lesser extent or committing a lesser crime received an average sentence of 18.64 years. As Eady notes,

In most cases where the Respondent claimed not to have been present, the basis for conviction seems to lie in the Respondent’s previous history or associations rather than in the prosecution claiming that they were in fact present. In around 30 % of these cases mobile phone records provided evidence of this association but not, it is argued by Respondents, of involvement with the crime. (p. 11)

In its joint enterprise follow-up report, the House of Commons Justice Committee (December, 2014) examined evidence provided from a number of sources to discern the suitability of applying the joint enterprise doctrine. Amongst the evidence examined in the follow up report was the potential of joint enterprise laws to deter youth from becoming involved in group offenses. Providing oral evidence Simon Natas (a solicitor) argued that:

The use of punitive law and order policies to combat youth crime is a blunt instrument which runs the risk of making matters worse, not only because young people are much less likely to cooperate with the authorities if they perceive the legal system to be unjust, but also because custodial sentences have been shown to increase, not reduce, reoffending (p. 13).

Giving evidence to the hearing, researcher Dr. Ben Crewe from the Cambridge Institute of Criminology, added to this point by pointing out that for deterrence to be effective people have to be aware of the sanction and that this awareness must shape their risk perception. In their research, Crewe et al. (2015) note how only few of the youth interviewed had any idea what joint enterprise was, even when they had been prosecuted under the law, and even those who were aware of it had limited understanding of its implications. Crewe pointed out that the main deterrent effect of any law functioned on people’s belief in the certainty of being caught. Consequently, legislation that holds all gang members responsible for an individual member’s actions will have little deterrent impact if the two main principles of deterrence (i.e., awareness of sanctions and a corresponding increase in risk perception) are not understood by the target populations. The Committee concluded that publicly available information about the use of joint enterprise was incomplete and that the use of joint enterprise needs an urgent review since the effects of the doctrine are likely to be negative for the reputation of the justice system, wider society, those convicted and victims of crime.

When reviewing the statistics on gang membership, we can see that stringent police tactics and governmental policies such as those outlined above do not succeed in deterring gang membership. Although the exact numbers of gangs and gang members is largely unknown in the UK, a policy report (Centre for Social Justice 2009) highlighted a host of research findings that suggest that gang culture and its associated violence had increased in the previous decade, with up to 6 % of 10–19-year-olds claiming to be gang members (Sharp et al. 2006) and at least half of the murders of young people in London during 2007 relating to gang activity (Metropolitan Police Authority 2008). A policy paper by the Centre for Social Justice (2012) published after the London riots of 2011 goes on to note how raids leading to the arrest of more senior gang members had left vacuums that younger gang members fought to fill, leading to a corresponding increase in violence as younger gang members “….vie for status and respect using the currency of violence” (p. 7). Echoing the conclusions of their earlier report (Centre of Social Justice 2009), the policy paper maintains that the police cannot arrest their way out of the gang problem. The report also argues that preventative work that could discourage young people from gang involvement has ‘fallen off the radar’ (Centre of Social Justice 2012, p. 3).

Unsurprisingly, in an atmosphere of suppressive and punitive prosecution tactics, relations between the police and youth in the UK have experienced repeated blows as police who work on databases of ‘permanent suspects’ label and target youth as gang accomplices if they are seen to associate with ‘known’ gang members, even if they have committed no offense (Ralphs et al. 2009). In gang areas where it is unlikely to be possible or wise for non-gang youth to avoid any contact with gang members for reasons of personal safety, non-gang youth run the risk of attracting police suspicions that they are gang members. Non-gang youth then view the associated constancy of being stopped and searched as police harassment. In turn, these anti-police attitudes manifest into negative consequences for all because police lose any co-operation they could have had from non-gang youth, and the youth suffer barriers to leading a normal life by being prevented from attending social events and sometimes even school (Ralphs et al. 2009). This may then lead into a further negative effect and encourage non-gang youth to join gangs as noted by Huff and McBride (1993; see above). It is unsurprising then that government policies and the associated police strategies have received scathing criticism that they fail because they do not take an evidence-based approach to gang reduction (Shute and Medina 2014).

In the US, gang figures bode just as badly as they do in the UK for oppressive police and government strategies. The National Gang Intelligence Center (NGIC 2011) notes that gangs are expanding across the U.S. with approximately 1.4 million active gang members belonging to more than 33,000 gangs which pose a threat to communities nationwide. The NGIC goes on to note how gangs are expanding their criminal profiles and that even when incarcerated gang members are not deterred from continuing gang involvement via friends and family; a finding also noted in UK research (see Wood et al. 2010). None of this bodes well for the effects of deterrence on gang membership as it seems that policy, suppression and even threats of punishment fail to deter as intended. As the NGIC noted in 2011:

Local neighborhood, hybrid and female gang membership is on the rise in many communities. Prison gang members, who exert control over many street gang members, often engage in crime and violence upon their return to the community. Gang members returning to the community from prison have an adverse and lasting impact on neighborhoods, which may experience notable increases in crime, violence, and drug trafficking (p. 11).

Consequently, it appears that police, policy, and even punishment deterrence strategies are ineffective, at least in an overall sense because gangs continue to thrive, adapt, and evolve. However, there are many potential explanations as to why deterrence effects are apparently ineffective in addressing what appears to be a global and flourishing phenomenon.

Why Deterrence Strategies May Not Be Effective with Gang Members: Psychological Explanations

Little research has directly examined gang membership and deterrence effects. What research has shown so far is that gang members are unlikely to be deterred from carrying and using guns (Watkins et al. 2008), unlikely to perceive certainty of arrest, are not influenced by the potential severity of punishment and expect little condemnation from significant others for offending (Maxson et al. 2011). Other research shows that deterrence efforts have less impact on gang members than on other youth (e.g., Hennigan and Sloane 2013). However, what we do not yet fully understand are the reasons why gang members hold such perceptions of chance of arrest, feel a need to carry weapons, and are seemingly immune from the effects of deterrence. Gangs are groups that have the potential to exert a powerful and enduring influence over individual members by shaping the way that they think and behave. This makes a closer consideration of the psychological effects of gang membership on individual gang members vital to the analysis of deterring gang membership. However, to date, a comprehensive psychological exploration of gang membership remains remarkably under-researched. The following section considers issues of identity, cohesion, morality, status enhancement, and oppositional culture, which are a few of the many possible psychological explanations that could help us understand why deterrence strategies may not be effective with gang members.

Social Identity

Research indicates that youth who feel alienated from legitimate social controls such as families, education, and prosocial community endeavors are those most likely to join gangs (e.g. Marshall et al. 2005). As they reject or reduce familiar childhood groups and practices in favor of gang activities, they are likely to feel uncertain about their attitudes, their future, and importantly about who they are (i.e., their identity). Uncertainty-identity theory (Hogg 2000) explains that when people feel uncertain about their personal identity they are motivated to affiliate with a group and, in line with social categorization principles (Abrams and Hogg 2010; Tajfel and Turner 1986), they use their group membership to categorize themselves according to the attitudes and behaviors that epitomize that group. Since gangs offer friendship, pride, a sense of identity, heightened self-esteem, excitement, and potentially access to financial benefits (Goldstein 2002) it is not difficult to see why youth struggling with legitimate controls (e.g., school) and personal identity are attracted to gang membership. According to social identity (Tajfel and Turner 1986) and self-categorization theory (Turner et al. 1987) the shaping of group members’ self-views and personal identity continues during membership until harmony between personal and group identities is achieved.

Once the process of intertwining personal identity with the gang’s identity is underway, gang members may begin to put the group’s aims first. For instance, Hennigan and Spanovic’s (2012) examination of social identity and gang membership showed how, compared to nongang youth, gang youth who identify with their group put the group norms of criminal activity ahead of personal fears of punishment for criminal activity. As the authors note, “Since crime and violence are normative among gang-involved youth, personal estimates of getting caught and punished have little or no influence on their criminal and violent behaviors” (Hennigan and Spanovic 2012, p. 143). Hence, police, policy and punishment practices that aim to deter gang involvement are likely to be ineffective with youth who disregard threats of punishment in favor of adhering to group norms.

Pluralistic Ignorance

Of course it is always possible that gang members may not actually agree with many of the group’s norms. However, research suggests that even if group members do not fully accept or agree with group norms, they will still adhere to them. The concept of pluralistic ignorance suggests that even when individual group members do not agree with their group’s activities, they still publicly accept and abide by them because they believe (often wrongly) that the rest of the group does agree with the activities (O’Gorman 1986; Reid et al. 2005b). There is also a great deal of social psychological evidence showing that when people want to be accepted by a group, they will conform to others’ decisions—even if they believe those decisions are wrong (Cooper et al. 2004). Research examining gang members’ reports also suggests that they experience pluralistic ignorance. For instance, even when gang members express privately that they feel extreme discomfort with some of their criminal activities—they still participate in those activities (Matza 1964). However, social psychological findings suggest that as group members’ identification with the group intensifies, they increasingly believe in the group’s norms (Reid et al. 2005a). Also, as violence is an intrinsic part of gang life, gang members may be inclined to adhere to ingroup norms due to fears of being on the receiving end of violent sanctions that their gang imposes on norm-violating members (Rimal and Real 2003). This threat is also more immediate and potentially greater than any that the criminal justice system poses. Plus, since the potential rejection of friends or admired others is an especially threatening aspect of norm violation (Baron and Kerr 2003), youth who are vulnerable, socially disenfranchised and whose self-esteem is embodied in the group are likely to feel any threat of group rejection keenly. In short, a gang is likely to wield its own form of deterrence on members (i.e., deterrence of law-abiding behavior) and this may well trump the deterrence efforts of the criminal justice system.

Cognitive Dissonance and Moral Disengagement

Consequently, given the power of ingroup influences and the eagerness with which aspiring gang members accept them, it is likely that gang youth, unlike nongang youth, feel compelled to abide by group norms—such as crime and violence (Viki and Abrams 2012)—even if individual criminal justice deterrence effects on that member are strong and his/her risk perception is high. Nonetheless, any unease that members may feel regarding their involvement in acts of violence is likely to cause them internal turmoil. Cognitive dissonance is a state of stress or mental discomfort that results when people’s behavior, thoughts and attitudes are in conflict. Because cognitive dissonance is uncomfortable, people are motivated to reduce it (dissonance reduction) by bringing thoughts/attitudes in line with their behavior (Festinger 1962). By not rejecting gang norms and continuing to be involved in gang activity (behavior), gang members are likely to adapt their thoughts and attitudes to bring them in line with their behavior to achieve cognitive consistency.

One strategy that gang members may use to reduce their dissonance is to set aside their existing moral standards (i.e., morally disengage). Moral disengagement is a strategy that involves the use of up to eight strategies (e.g., moral justification, euphemistic language, dehumanizing victims) to justify harmful acts and resolve the cognitive dissonance and self-condemnation associated with violating personal moral standards (Bandura et al. 1996). Moral disengagement involves eight socio-cognitive mechanisms which operate on three levels of social processing. The first enables inhumane acts (e.g., violence) to be reinterpreted as worthy. Socio-cognitive mechanisms at this level include: moral justification (behavior is for a good reason—i.e., furthering gang status), euphemistic language (sanitizing descriptions of harm—e.g., violence is just “gang business”), and advantageous comparisons in which personal behavior is favorably compared with others’ apparently worse behavior (e.g., our group only assaults—others kill). The second level allows the displacement of responsibility for personal actions onto authority figures (behavior is at the direction of authority figures which then negates personal responsibility), diffusion of responsibility (responsibility for harm is shared by several perpetrators and so individuals are absolved from blame), and distorting the consequences of harm (ignoring, minimizing, or disbelieving that harm has been done). The third level distorts the view of any victims via dehumanization processes (the victim is thought of as subhuman and thus devoid of human qualities—see also Alleyne et al. 2014) or via victim blaming (they got what they deserved). Research has shown that in order to be accepted by a chosen group, youth may set aside their moral standards (Emler and Reicher 1995), and gang research shows that moral disengagement is higher in street gang members (Alleyne and Wood 2010; Esbensen and Deschenes 1998; Maxson, et al. 2011) and in prisoners involved in prison gang activity (Wood et al. 2009, 2014) than it is in non-gang youth or non-gang prisoners. Moral disengagement strategies have also been empirically linked with increased violence (Bandura et al. 1996), and have been shown to mediate pathways between impoverished neighborhoods strongly associated with gang membership and anti-social behavior (Hill et al. 2001) and between low levels of empathy and anti-social behavior. As Hyde et al. (Hyde et al. 2010) observe:

In more modern contexts, urban youth living in impoverished homes and neighborhoods that offer them little hope or opportunity for socially acceptable pathways to success may develop a moral code of behavior that is not bound by mainstream prohibitions against committing antisocial actions, particularly when such actions are associated with the means to obtain financial success (e.g., dealing illicit drugs) or ensuring safety (e.g., joining a gang). (p. 198)

Group Commitment and Cohesion

As strategies such as moral disengagement embed and gang members begin to think, as well as act alike, the group will become more cohesive. Klein (1995) considers cohesion to be “…the quintessential group process” (p. 43; see also Wood 2014 for a fuller discussion) which influences members to become loyal, committed and ready to make sacrifices for a group that they view with pride and respect (Crocker et al. 1994). A meta-analysis examining cohesion concludes that highly cohesive groups are more productive than those that are less cohesive (Evans and Dion 1991) and, as Klein (1995) observes, gangs produce crime. Importantly, this effect is enduring—continuing even after members have left a gang (Pyrooz et al. 2014). Group commitment and the ensuing cohesiveness are not only likely to result in efficient criminal activity; they are also likely to have negative connotations for criminal justice efforts to deter gangs and their activities, since highly cohesive gangs are efficient at mobilizing membership and accessing commodities such as drugs and weapons (Hughes 2013). Such is the effect of cohesion that gang experts such as Klein and Maxson (2006) contend that it is their “….very strong opinion that levels of gang cohesiveness correlate directly with levels of gang crime and with gang responses to our efforts at gang control. Greater cohesion leads to greater crime involvement and greater resistance to gang control” (p. 196).

Criminal justice officials not only have to contend with gang cohesion potentially undermining their deterrence efforts, they also have to consider that gangs may well have their own agendas—primarily those involving other gangs—and these agendas may take priority over individual members’ personal risk perceptions. Gang researchers note how a gang’s identity is often defined in reference to other gangs—particularly rival gangs (Papachristos et al. 2013). Outgroups may be used as points of reference as a gang assesses its actions and, importantly, its status (Decker and Van Winkle 1996). Social dominance theory (SDT: Sidanius and Pratto 1999) explains that when group members have high social dominance orientation (SDO) they feel compelled to enhance, or reinforce, their group’s place in a social hierarchy of groups (such as gangs) which have been arbitrarily constructed to respond to situational factors such as competition for valued resources. So, for example, gangs may compete with each other in an effort to enhance or reinforce their group’s status regarding control over valued illegal resources (e.g., narcotics). Although research examining group processes such as SDO in the context of gangs is still in its infancy, findings so far indicate that, compared to nongang individuals, those involved in gangs and gang activity have higher levels of SDO (Densley et al. 2014; Wood et al. 2014).

Reputation and Status Enhancement

As gangs work to assert/enhance their group’s status in this arbitrary-set system they are likely to respond violently to rivals who threaten their aims (Aldridge et al. 2008; Decker and Van Winkle 1996). Densley’s (2013) research shows how gang members see violence as the fundamental response to protecting territory and/or gang business. Accordingly, gangs’ intergroup enmity is funded by violence which stems from competition for power, domination, reputation, respect, and status (Harding 2012). Intergroup transgressions cannot be ignored if a gang is to maintain its status. This makes status, together with previous conflicts, the most common reasons for intergang violence (Hughes and Short 2005; Papachristos 2009). In turn, violence against rivals helps a gang save face, protects its members by discouraging attacks from rivals, and exacts revenge on opposing groups for any transgressions (Papachristos et al. 2013). Criminal justice efforts to deter gang members’ involvement in gang activity are therefore likely to encounter the considerable obstacles of status, reputation, and resource-motivated commitment.

Being a gang member can also enhance individual reputation and status. We know from research that status is highly prized by both street gang members (Alleyne and Wood 2010) and prisoners involved in prison gang activity (South and Wood 2006; Wood et al. 2009, 2014). Reputation enhancement theory posits that group membership facilitates individual behavior as members select an image to display in front of specific others (Emler and Reicher 1995). These others then provide positive feedback that reinforces the individual’s image within the group. For gang members, crime is likely to be key to a positive personal reputation in a gang world where criminal activity is a prized gang product. Violence in particular will also protect individual members from future victimization (Emler and Reicher 1995). In their study of gangs in the UK, Harris et al. (2011) note:

Not reacting with often extreme violence was experienced as tantamount to abject failure. There was a sense of being worse than nothing if a once-held status is lost. This was not only due to loss of respect, but also a sense of inevitable attacks and victimization from others. (p. 20)

In a culture of inter-group rivalry where it is important to protect personal and group status and criminal enterprises it is unlikely that criminal justice efforts to deter gang activity via suppression strategies and their accompanying policy infrastructure will have enduring effects. They may well have temporary influences as gangs, wary of new police tactics, effectively lie low—for a while, but the durability of these effects is likely to be short lived. Gangs have their own agendas, be they social, criminal, or conflict-related, and they are likely to adapt so that they can adhere to these agendas. It also needs to be borne in mind that, even if deterrence strategies are effective in heightening individual gang members’ risk perceptions, gang members are also prospective victims—particularly from rival gangs (Sanders 1994). So, even if gang members feel deterred from gang activity and want to rescind their membership to lead law-abiding lives, the probability of continued victimization by rival gangs may compel them to continue their membership to gain the protection that their gang can offer (Vigil 1988).

Oppositional Culture

By employing deterrence strategies, the authorities also run the risk of feeding an “oppositional culture” that has long been acknowledged as endemic in gang norms. Oppositional culture means that the group is set in opposition to legitimate authorities (Moore and Vigil 1987). As Short and Strodtbeck (1965) observed, “…. gang members are less favorably disposed toward adult incumbents of legitimate roles such as teacher, religious leader, policeman, businessman, and politician than are their non-gang, lower class counterparts” (pp. 275–276; see also Drury 2010). Research examining gang membership shows how street gang members (Alleyne and Wood 2010), and those involved in prison gang activity (Wood et al. 2014), hold strong anti-authority attitudes. Although many adolescents hold anti-authority attitudes (Drury 2010) which are therefore not unique to gang members, the targeting of gangs in gang suppression programs is likely to bring gang members into frequent and negative contact with authority figures which, in turn, will exacerbate their hostile views of authority. The net effect here is that gangs may come to view themselves as unfairly targeted victims of oppression (Lien 2005), which will help to reinforce their gang identities (McAra and McVie 2005; Ralphs et al. 2009), strengthen their oppositional culture (Klein and Maxson 2006), and may even increase the number of gangs and hence, criminal activities (Hagedorn 2008) as gangs adopt even greater criminal responses in an effort to defend their group identity (Ayling 2011). Members may consider themselves as defenders of their group which they see as being victimized. For instance, identifying this effect in Norwegian gang members, Lien (2005) notes:

He (sic) develops ideas of compassion, love, and sacrifice in relation to his friends, and he explains his acts through a construction of himself as a victim of society. The victimization point is necessary in order to justify the criminal act. He cannot be blamed, the act is heroic rather than evil, and the victims get what they deserve. (p. 121)

As Lien (2005) explains, gang members develop ‘heroism’ for their group and for what they perceive as their victimization by society. Victimizing opponents is then all the more justified as each gang member strives to demonstrate his/her loyalty to the group and to fulfil his/her perceived obligation to other members. And, in a climate where abject loyalty, cohesion, identity, self-esteem and personal protection abound it is easy to see why gang members may not be as vulnerable to deterrence strategies as criminal justice agents would wish.

Concluding Remarks

The aim of this chapter was to examine the effectiveness of deterring gang members from their gang lifestyles. To this end, it began by considering some of the myriad of criminal justice initiatives that, at great cost to public purses, have striven to deter gang activity and gang membership. However, so far neither programs nor policies seem to have successfully hit the mark and deterred gang activity in any enduring way. Some programs appear to have some effects on some gangs, leading to euphoric claims that they are effective deterrents (e.g., Kahan 1997). However, this euphoria was premature. A lack of methodologically sound evaluations and accusations that deterrent effects are, at best short-term and at worst counter-productive, has corroded the initial enthusiasm that deterrence programs provide a panacea to the gang problem. Zealous approaches such as suppression tactics and the variety of policy infrastructures supporting them have, all too frequently, backfired amidst accusations of an excessive use of force on citizens—gang and nongang—and unfair/unsound convictions of individuals who are guilty of failing to anticipate others’ actions (e.g., Eady 2013). Unsurprisingly, these programs and policies have undergone several incarnations as justice officials and politicians strive for amendments that will uphold justice principles.

Even programs that include carrots as well as sticks and aim not only to suppress gang activity but also to offer gang members viable nongang alternatives, have struggled to keep afloat (e.g., Greene and Pranis 2007). As appealing as these multi-faceted programs are inasmuch as they also help address key gang-related risk factors such as social disadvantage, they have been dogged by accusations that they lack balance between their carrots and their sticks and these have undermined their program integrity. If we feed into this a lack of consistency between and within programs and the methodological problems which have consistently hindered conclusive program evaluations and definitive meta-analyses, we reach a point where, despite vast resource expenditure, we still know little, if anything, about what is truly effective in reducing/preventing gang membership.

In a climate where gang membership appears to be flourishing despite deterrence strategies, a further aim of this chapter was to outline some of the reasons why gang members may not be deterred from gang membership and activity. There is little research directly examining deterrence effects with gang members and little research that specifically examines the psychological effects that gangs have on their membership. Although gangs may appear to be homogeneous groups with homogeneous members—they are not. Individual differences mean that gangs and gang members are heterogeneous and that strategies that work with one gang member may fail with another. Consequently we must consider gang members as individuals—and work to devise interventions tailored to their specific needs. Despite a host of social psychologists (e.g., Abrams and Hogg 2010; Hogg 2000; Tajfel and Turner 1986; Turner et al. 1987) noting the powerful impact that group processes have on individuals, little work has examined the specific nature of those group processes within the gang context (Wood 2014). Effects such as identity development, pluralistic ignorance, resolution of cognitive dissonance via moral disengagement strategies, and the development of group cohesion are all likely to have powerful roles in inoculating gang members from criminal justice efforts to deter their behavior. As noted, gangs have their own agendas and even if those agendas are as simple as wanting to ‘hang out’ together, gang members will, due to group process influences, stick to their agendas, regardless of the threat of negative consequences (e.g., Hennigan and Spanovic 2012).

Deterrence strategies operating on threat are unlikely to be effective. Gangs are all too familiar with threats, and threats that are more immediate and potentially deadly, such as those radiating from rival gangs, may take precedence over threats that emanate from criminal justice systems. As gangs perceive the multitude of threats leveled at them they are likely, as researchers such as Klein (1995) argue, to become cohesive and the more cohesive a gang is, then the more likely it is to continue, develop and adapt to the cultural climate that undulates around it. We are already seeing gangs adapting to social media methods of functioning (see Patton et al. 2013), and it is reasonable to expect that gangs will continue to adapt and change other fundamental operational methods in response to perceived threats. However, if research can help us to understand more about gang members and identify with robust empirical examinations the group processes that profoundly influence gang membership then we will have the infrastructure to devise effective interventions to tackle gang membership at an individual level. So far, criminal justice systems have attempted to hamstring gangs with suppression strategies, to prosecute gang members collectively via joint enterprise policies, and to offer alternatives to gang life via community support measures; yet, gangs still seem to thrive. Although understanding more about the psychology of gang membership is a research agenda that is in its infancy, more research examining the psychological influences of gangs on gang members is crucial if we are to adequately and consistently deter gang membership and gang activity.