THE problem of school violence exploded into public consciousness in the United States in the late 1990s with a rash of highly publicized school shootings like the one at Columbine High School in Littleton, Colorado. Because these shooting incidents took place in suburban or semirural settings, they touched off tea national mood of self-searching about the roots of youth violence that a decade of inner-city carnage had not" (Currie 1999, 234).

Despite the horrific nature of these specific acts, there is no evidence to suggest that the overall level of school violence in America has increased dramatically in recent years (Chandler et al. 1998). In fact, serious youth violence, particularly homicide, has actually declined in the United States during the past few years (Blumstein and Rosenfeld 1998). Lethal violence in America, however, is still exceedingly common and one of our most serious social problems. As Zimring and Hawkins (1997) demonstrate, "Lethal violence rather than high rates of crime is the disabling problem that sets the United States apart from other developed countries in the 1990s" (1).

Given this "American exceptionalism" concerning violence (Messner and Rosenfeld 1997) and the fact that violent acts disproportionately involve young people aged 15 to 24, we should indeed engage in some self-searching about the roots of youth violence. This article will examine some of the more general social, economic, and cultural conditions that give rise to serious crime and violence in the United States. Specifically, the article will explore the role of poverty, economic inequality, and social exclusion in shaping the problem of youth violence by summarizing and integrating the recent theory and research of a number of sociological criminologists, such as Elliott Currie, John Hagan, and Francis Cullen.

POVERTY, INEQUALITY, AND SOCIAL EXCLUSION

Why does the United States have exceptionally high rates of violent crime, particularly youth homicide, compared to other industrial nations? Conservative commentators frequently assert that it is a lenient criminal justice and juvenile justice system that causes high crime rates or that crime and violence are the result of cultural decline and something called moral poverty. But the American justice system is one of the harshest in the world, and, although the cultural and moral condition of American families and communities is important to take into account in understanding crime, these conditions are strongly affected by larger social and economic forces. These larger social structural conditions are the factors that sociological criminologists point to as the roots of violence. As Currie (1998) observes, "For there is now overwhelming evidence that inequality, extreme poverty, and social exclusion matter profoundly in shaping a society's experience of violent crime. And they matter, in good part, precisely because of their impact on the close-in institutions of family and community" (114).

When we look at the research on poverty and economic inequality, we find that the United States has by far the highest poverty rate and the biggest gap between the rich and the poor of any of the developed nations (Kerbo 1996). Currie (1998) notes the findings of the Luxembourg Income Study (LIS), an international survey of poverty, inequality, and government spending in industrial countries (Rainwater and Smeeding 1995). The LIS shows that the United States, while a very wealthy society, has far more inequality and is far less committed to providing a decent life for the poor than are other developed nations. The LIS also demonstrates that, in particular, children and families in the United States are far more likely to be poor than those in other industrial democracies. Furthermore, poor American children are more likely to be extremely poor compared to children in other advanced countries.

According to the LIS and other studies, there are several reasons why poor children and families in the United States find themselves in such a plight. First, many Americans in the so-called urban underclass are trapped in a system of concentrated unemployment that results in an increasingly isolated poverty (Wilson 1996). Second, those who do work, primarily in the secondary labor market, earn very low wages compared to their counterparts in other countries. This creates the problem of the working poor. Finally, the United States provides fewer government benefits to either the underclass or the working poor to offset the problems of concentrated unemployment and poor wages. Recent changes in the welfare system are likely to aggravate the situation.

This deprivation and social exclusion are related to the high rates of violence found in the United States. Currie reviews both studies of international differences in violent crime and studies of violence within the United States and other countries to demonstrate the connection. Crossnational studies show that countries with a high degree of economic inequality have higher levels of violence (Gartner 1990). Other studies show that, even within a generally deprived population, it is the most deprived children who face the greatest risks of engaging in crime and violence (Werner and Smith 1992). Finally, Currie notes the research of Krivo and Peterson (1996), who suggest that it is the link between extreme disadvantage and violence that underlies much of the association between race and violent crime in the United States. After reviewing these and other studies, Currie (1998) concludes,

The links between extreme deprivation, delinquency, and violence, then, are strong, consistent, and compelling. There is little question that growing up in extreme poverty exerts powerful pressures toward crime. The fact that those pressures are overcome by some individuals is testimony to human strength and resiliency, but does not diminish the importance of the link between social exclusion and violence. The effects are compounded by the absence of public supports to buffer economic insecurity and deprivation, and they are even more potent when racial subordination is added to the mix. And this ... helps us begin to understand why the United States suffers more serious violent crime than other industrial democracies, and why violence has remained stubbornly high in the face of our unprecedented efforts at repressive control. (131)

But how do these social and economic forces cause violence? In what specific ways do poverty, inequality, and social exclusion act to produce violent crime by young people? To help answer these questions, I suggest that we utilize the general organizing concepts of social support and informal social control. It is the absence of these two important social processes, in urban, suburban, or rural settings, that allows for the infliction of social and psychic pain on young people and the development of negative attitudes and emotions that can easily lead to violence.

THE ABSENCE OF SOCIAL SUPPORT AND INFORMAL SOCIAL CONTROL

In his presidential address to the Academy of Criminal Justice Sciences in 1994, Francis T. Cullen suggested that a lack of social support is implicated in crime. Cullen argued that social support, if approached systematically, can be an important organizing concept for criminology. He defined social support as "the perceived or actual instrumental and/or expressive provisions supplied by the community, social networks, and confiding partners" (Cullen 1994, 530). Cullen went on to develop a series of propositions, supported by criminological research, about the relationship between the lack of social support and the presence of crime at societal, community, family, and relational levels of analysis.

According to Cullen, a distinction should be made between the concepts of social support and informal social control. Informal social control involves all the sanctions and constraints used in an effort to control another individual's behavior (to make him or her conform to social norms) that fall outside of formal, legal, and bureaucratic systems. Informal social control is generally exercised by significant others, families, friends, neighbors, and community networks. The breakdown or absence of informal social control has also long been cited by criminologists as a factor in the involvement of persons in criminal behavior.

In the following sections, I will review theory and research that examine the relationship between broad structural conditions like poverty, inequality, and social exclusion; institutional-level social support and informal social control; and the problem of youth violence. First I will consider social support as an organizing concept and then informal social control.

Social support

One of the most significant ways in which economic deprivation and social exclusion can lead to youth violence is by inhibiting or breaking down the social supports that affect young people. Cullen (1994) reviews research that supports his proposition that "America has higher rates of serious crime than other industrial nations because it is a less supportive society" (531). He notes studies that have demonstrated the corrosive effect of America's culture of excessive individualism and pursuit of material gain without regard to means (Messner and Rosenfeld 1997). This competitive pursuit of the American Dream not only encourages individuals to obtain material goods "by any means necessary"; it also inhibits the development of a "good society" in which concern for community and mutuality of support dominate. Cullen (1994) also points out that "economic inequality can generate crime not only by exposing people to relative deprivation but also by eviscerating and inhibiting the development of social support networks" (534).

Moving down from the national level, Cullen (1994) argues that "the less social support there is in a community, the higher the crime rate will be" (534). He reviews evidence that "governmental assistance to the poor tends to lessen violent crime across ecological units," and research that reveals "that crime rates are higher in communities characterized by family disruption, weak friendship networks, and low participation in local voluntary organizations" (534-35). Finally, Cullen notes quantitative and ethnographic research on the "underclass" that documents that powerful social and economic forces have created isolated innercity enclaves. These enclaves fray the supportive relations that once existed between adults and youths, supportive relations that previously offered protection to those youths from involvement in crime. We will return to this body of research later in a review ofthe work ofJohn Hagan (1994).

Next, Cullen (1994) addresses the issue of the role of the family in offering social support. He asserts that "the more support a family provides, the less likely it is that a person will engage in crime" (538). This is the critical link between poverty, inequality, exclusion, and violence. Recall Currie's argument that these social forces matter precisely because of their impact on the close-in institutions like the family. As Cullen (1994, 538) notes, there is a considerable amount of evidence that parental expressive support diminishes children's risk of criminal involvement. He cites Loeber and Stouthamer-Loeber's (1986) comprehensive meta-analysis of family correlates of delinquency that clearly shows that indicators of a lack of parental support increase delinquent behavior. This study concludes that youth crime is related inversely to "child-parent involvement, such as the amount of intimate communication, confiding, sharing of activities, and seeking help" (Loeber and Stouthamer-Loeber 1986, 42).

Both Cullen and Currie warn that any discussion of families and crime must avoid the "fallacy of autonomythe belief that what goes on inside the family can usefully be separated from the forces that affect it from the outside: the larger social context in which families are embedded for better or for worse" (Currie 1985, 185). While any family, regardless of its socioeconomic status, can be affected, both Cullen and Currie stress the social and economic forces like poverty and inequality that have transformed and, in many cases, ripped apart families, particularly families of the underclass, in ways that have reduced their capacity to support children. The Panel on High-Risk Youth (cited in Cullen 1994, 539) states:

Perhaps the most serious risk facing adolescents in high-risk settings is isolation from the nurturance, safety, and guidance that comes from sustained relationships with adults. Parents are the best source of support, but for many adolescents, parents are not positively involved in their lives. In some cases, parents are absent or abusive. In many more cases, parents strive to be good parents, but lack the capacity or opportunity to be so.

In his review of the research on the connections between family deprivation and violent crime, Currie (1998, 135-39) highlights four key findings: "1) extreme deprivation inhibits children's intellectual development; 2) extreme deprivation breeds violence by encouraging child abuse and neglect; 3) extreme poverty creates multiple stresses that undermine parents' ability to raise children caringly and effectively; and 4) poverty breeds crime by undermining parents' ability to monitor and supervise their children." Findings 1 through 3 provide more specific articulation about the ways in which poverty and inequality shape youth violence through the lack of social support. Stunted intellectual development that cripples children's ability to be successful in school or at work, violence and abuse that create angry and fearful children, and the lack of parental care and nurturance all contribute to the production of young people who are prone to strike out at the world through violent acts.

Social and cultural capital

Another important perspective on the relationship between social and economic conditions, the lack of social support, and youth crime is contained in the work of John Hagan. In presenting a "new sociology of crime and disrepute," Hagan (1994) develops the concepts of human, social, and cultural capital and capital disinvestment processes to help us understand the connections between inequality, social institutions, and violent crime. According to Hagan, the general concept ofhuman capital refers to the skills, capabilities, and knowledge acquired by individuals through education and training that allow them to act in new ways. To this he adds the concept of social capital, which "involves the creation of capabilities through socially structured relationships between individuals in groups" (67). Social groups such as intact nuclear and extended families, wellintegrated neighborhoods, stable communities, and even nation-states are the sites for the development of social capital in individuals that provides them with the resources and capacities to achieve group and individual goals. These supportive social networks can lead to the formation of cultural capital such as the credentials of higher education and involvement in high culture like the arts and their supporting institutions. As Hagan points out, "In these community and family settings, social capital is used to successfully endow children with forms of cultural capital that significantly enhance their later life changes" (69).

The ability to endow children with social and cultural capital, however, is linked to economic position. As Hagan (1994) notes, in less advantaged community and social settings, which lack abundant forms of social and cultural capital, parents are far less able to provide resources, opportunities, and supports to their children. Thus, "the children of less advantageously positioned and lessdriven and controlling parents may more often drift or be driven into and along less-promising paths of social and cultural adaptation and capital formation" (70). These "lesspromising paths of social and cultural adaptation," of course, include embeddedness in the criminal economy of drugs and other forms of gang activity and delinquent behavior.

Hagan (1994) emphasizes that "disadvantaging social and economic processes" in the community and broader society, what he calls "capital disinvestment processes," are destructive of social and cultural capital and often produce deviant subcultural adaptations (70). The three capital disinvestment processes that "discourage societal and community level formations of conventional social capital" are residential segregation, race-linked inequality, and concentrations of poverty (70-71). Hagan describes these destructive structural conditions and the dislocations that they produce in community settings. He then reviews a considerable body of new ethnographic and quantitative research that documents and articulates the ways in which these community-level processes of capital disinvestment disrupt and destroy the social capital of families, diminishing their capacity to provide the human and cultural capital their children need to improve their life chances and become stable and productive members of the community. Thus these children are at a much greater risk of becoming embedded in the criminal economy of drugs and the violence that it often entails, as well as becoming involved in other forms of conventional criminality. As Hagan observes, "In communities that suffer from capital disinvestment and in families that have little closure of social networks and social capital to facilitate investment in their children, youths are more likely to drift into cultural adaptations that bring short-term status and material benefits, but whose longer-term consequences include diminished lifechances" (93).

Informal social control

As noted previously, Currie found that the lack of effective parental supervision has a strong relationship to delinquency. This raises the important issue of informal social control. The ability of adults to monitor and supervise, impose sanctions, shame, and otherwise keep young people in line through face-to-face interaction within important social institutions is an important variable in delinquency prevention. There is a considerable amount of criminological evidence that suggests that these informal mechanisms of social control, operating within families, schools, neighborhoods, workplaces, and social networks, play an important role in preventing youth crime and violence. As Minor (1993) points out, "Research has demonstrated that, during the course of childhood and adolescent socialization, the more meaningfully integrated persons become [into] those social institutions which promote informal social control, such as the family, school, and work, the lower the likelihood of delinquency" (59).

As with the lack of social support, social structural forces such as poverty and social exclusion can inhibit or erode the exercise of informal social controls within these intermediate institutions. With the erosion of these controls, the chances for young people to become involved in violent crime increases. As Minor (1993) observes,

Through their impact on social institutions, the macro forces emanating from a society's political economic organization shape the quantity and quality of behavioral choices available to individuals. By diminishing the capacity of institutions, especially the family, to positively influence the choices made by youths and by rendering youths vulnerable to delinquent socialization in peer groups, macro forces can weaken informal mechanisms of social control. (59)

In his excellent review of the theory and research on the political and economic context of delinquency in the United States, Minor (1993) identifies three macro forces that have had important consequences for the problem of youth violence. These forces are the socially defined position of youth, the impact of market relations, and poverty and inequality. First, compulsory education, child labor laws, and the emergence of the juvenile justice system combined to promote youth segregation, impose labor market restrictions on young people, and increase the importance of peer group socialization. Market relations led to the penetration of economic norms into all spheres of life and the fostering of a competitive individualism that undermined interpersonal cooperation and collective social welfare. Poverty and inequality had a disintegrative effect on social institutions through the lack of resources and emotional stress. According to Minor, these three forces evolved together as part of the transformation of the American economy to monopoly capitalism, and they have acted collectively to weaken informal mechanisms of social control and therefore increase youth violence.

Another perspective on the impact of cultural and structural forces on the ability of social institutions such as the family to control youth crime comes from Messner and Rosenfeld (1997). Building on Robert Merton's concept of anomie, Messner and Rosenfeld assert that the core features of the social organization of the United States-culture and institutional structure-shape the high levels ofAmerican crime. At the cultural level, they argue that the core values of the American Dream (achievement, individualism, universalism monetary success) stimulate criminal motivations while promoting weak norms to guide the choices of means to achieve cultural goals (anomie). As Messner and Rosenfeld point out, "The American Dream does not contain within it strong injunctions against substituting more effective, illegitimate means for less effective, legitimate means in the pursuit of monetary success" (76).

At the institutional level, Messner and Rosenfeld (1997) observe that the economy tends to dominate all other social institutions and that this imbalance of institutional power fosters weak social control. There are two ways that this imbalance of power weakens social control. First social institutions such as the family and the schools are supposed to socialize children into values, beliefs, and commitments other than those of the economic system. However, as Messner and Rosenfeld note, "as these noneconomic institutions are relatively devalued and forced to accommodate to economic considerations, as they are penetrated by economic standards, they are less able to fulfill their distinctive socialization functions successfully" (77). Thus, economic domination weakens the normative control associated with culture.

The imbalance of power also weakens the external type of social control associated with social structure. As Messner and Rosenfeld (1997, 78) point out, "External control is achieved through the active involvement of individuals in institutional roles and through the dispensation of rewards and punishments by institutions." When these noneconomic institutions are devalued and rendered impotent, then the attractiveness of the roles they offer to young people is diminished, and the incentives and penalties they can offer for prosocial behavior are limited. Messner and Rosenfeld conclude by noting that the problem of external control by major social institutions is inseparable from the problem of the internal regulation of social norms (anomie):

Anomic societies will inevitably find it difficult and costly to exert social control over the behavior of people who feel free to use whatever means prove most effective in reaching personal goals. Hence, the very sociocultural dynamics that make American institutions weak also enable and entitle Americans to defy institutional controls. If Americans are exceptionally resistant to social controland therefore exceptionally vulnerable to criminal temptations-it is because they live in a society that enshrines the unfettered pursuit of individual material success above all other values. In the United States, anomie is considered a virtue. (79)

Sampson and Laub's innovative reassessment (1993) of the longitudinal data gathered by Sheldon and Eleanor Glueck in the 1940s also supports the proposition that poverty and inequality undermine the ability of informal social controls within the family and school to contain delinquent behavior. Sampson and Laub develop an age-graded theory of informal social control. Their basic thesis is that "structural context mediated by informal family and school controls explains delinquency in childhood and adolescence" (7). Their unified model of informal family social control focuses on three dimensions: discipline, supervision, and attachment. They observe that "the key to all three components of informal family social control lies in the extent to which they facilitate linking the child to family and ultimately society through emotional bonds of attachment and direct yet socially integrative forms of control, monitoring, and punishment" (68).

The second part of Sampson and Laub's theory suggests that structural background factors, such as poverty, influence youth crime largely through their effects on family process. The empirical findings support their theory. They find that negative structural forces have little direct effect on delinquency but instead are mediated by intervening sources of informal social controls in the family and the school. They offer the following summary:

We found that the strongest and most consistent effects on both official and unofficial delinquency flow from the social processes of family, school, and peers. Low levels of parental supervision, erratic, threatening, and harsh discipline, and weak parental attachment were strongly and directly related to delinquency.... Negative structural conditions (such as poverty or family disruption) also affect delinquency, but largely through family and school process variables. (Sampson and Laub 1993, 247)

What Sampson and Laub find in their reassessment of the Gluecks' data on white children born in the 1920s and 1930s is supported by a more recent study of urban black children conducted by Shihadeh and Steffensmier (1994). They studied the links between economic inequality, family disruption, and urban black violence in more than 150 cities across the country. They found that as economic inequality increases, so do arrests of black youths for violent crimes. Shihadeh and Steffensmier suggest that the link between inequality and violence, however, is indirect. Greater income inequality increases the number of black single-parent households, and the increase in single-parent households is related to the level of youth violence. Single parents, with more stress and fewer resources, have a more difficult time monitoring and supervising their children and, in general, exercising effective social control. Rutter and Giller (1983) and Larzelere and Patterson (1990) provide additional evidence on the connection between poverty and poor parenting skills.

A final perspective on the relationship between economic conditions, informal social control, and violent delinquent behavior that should be mentioned is the work of Colvin and Pauly (1983). They develop an integrated structural-Marxist theory of delinquency production that focuses on the structures of control in several locations in the economic production and social reproduction processes: workplaces, families, schools, and peer groups. They argue that "the more coercive the control relations encountered in these various socialization contexts tend to be, the more negative or alienated will be the individual's ideological bond and the more likely is the individual to engage in serious, patterned delinquency" (515). Working-class parents who experience coerciveness in workplace control structures develop alienated bonds, which in turn contribute to the development of more coercive family control structures. Children who experience coercive family control structures develop alienated initial bonds that lead them to be placed in more coercive school control structures, which reinforce the juveniles' alienated bonds. This, in turn, leads to greater association with alienated peers, who form peer group control structures that interact with various community opportunity structures to produce delinquency.

FROM INEQUALITIES TO YOUTH VIOLENCE

Even though the rates of violent crime committed by young people have declined in recent years, youth violence remains a serious social problem in the United States. While many factors must be taken into account as we search for ways to deal with youth violence in general, and school violence in particular, it is imperative to understand the broader social and economic forces that play a critical role in shaping America's experience with this problem. The theory and research that have been reviewed in this article make a compelling case for the thesis that poverty, economic inequality, and social exclusion are causal agents in the production of crime and violence by young people in the United States. Although these structural conditions do not often have a direct effect in producing violent crime, they are important because of the impact they have on social institutions like the family, the school, and the community. While families, schools, and neighborhoods in middle-class, suburban areas can also become disrupted, the evidence shows that poverty, inequality, and exclusion decisively undermine the ability of those close-in institutions to provide the social support and informal social control that produce healthy, well-functioning children and prevent serious violent crime. When these institutions, in whatever socioeconomic setting, are unable to socialize children properly, care for them appropriately, and provide them with human and social capital, violence is a possible result. When these institutions, in whatever socioeconomic setting, are unable to effectively monitor, supervise, and sanction juveniles, violent crimes can take place.

This violence by young people seems to generally take one of three forms: predatory economic crimes, drug industry crimes, or social relationship violence. The first of these forms of violence occurs in the pursuit of monetary or materialistic goals by any means necessary. Given the intense cultural pressures for monetary success in America, economically disadvantaged youths who are blocked from less effective, legitimate means are often inclined to select more effective, illegitimate means to pursue the American Dream. As Messner and Rosenfeld (1997) point out, "This anomic orientation leads not simply to high levels of crime in general but to especially violent forms of economic crime, for which the United States is known throughout the industrial world, such as mugging, car-jacking, and home invasion" (76).

The second form of youth violence, involvement in the illegal drug industry, also stems from the pursuit of monetary success through effective, illegitimate means. Hagan (1994) points out that participation in the illegal drug industry is a subcultural adaptation to processes of capital disinvestment, in effect, a form of recapitalization, an effort to use available (albeit illegal) resources to achieve economic goals. As he notes, "During the period of capital disinvestment when access to legitimate job networks linked to core sector jobs declined in many distressed minority communities, networks of contacts into the world of drugs and drug-related crime proliferated, paving the way for many youths to become embedded in the criminal economy" (96). Hagan also points out that today's illegal drug industry is much more violent and unstable than those of the past. As more young people became embedded in the criminal economy of drugs, the more violent they became as gangs battled for drug markets. Rates of serious violence, including homicide, skyrocketed in the late 1980s and early 1990s, in particular with the rise of the crack cocaine epidemic. In fact, the recent reductions in violent crime rates are often attributed to the stabilization of the crack markets in the mid-1990s (Blumstein and Rosenfeld 1998).

The final form of youth violence is the violence that often flares within the context of frayed and volatile social relationships. Youths who are powerless, angry, frustrated, and alienated often act out in violent ways. For those who are rendered powerless by the social and economic structure, violence within social relationships is one way to reassert power and control in their lives (Pfohl 1994,407). Youths who experience the structural humiliation of poverty and inequality and lack the support and controls of a protective family or community often attempt to transcend their humiliation and shame through violence (Braithwaite 1992; Gilligan 1996). As John Braithwaite (1992) asserts, "When inequality of wealth and power is structurally humiliating, this undermines respect for the dominion of others. And a society where the respect for dominion is lost will be a society riddled with crime" (80). Additionally, Katz (1988) has observed that humiliation can lead to the embrace of righteous violence, which resolves the humiliation "through the overwhelming sensuality of rage" (24). Inequality and social exclusion, among other social and cultural forces, can shred the bonds of community that tie young people to others and can foster the use of violence within their close-in social relationships in the family, at school, or on the street corner.

One final note about these forms of violence. All of them are more likely to become lethal due to the overwhelming presence of guns in American society. The mugging, the bad drug deal, the schoolyard fight-all are more likely to turn deadly due to the easy availability of firearms in the United States. The recent school shooting deaths in Colorado and elsewhere were all made possible by the easy access of these youths to guns. As Zimring and Hawkins (1997) point out, firearms are a contributing cause of violent death and injury from intentional attacks. They note that "current evidence suggests that a combination of the ready availability of guns and the willingness to use maximum force in interpersonal conflict is the most important single contribution to the high U.S. death rate from violence" (122-23).

POLICY IMPLICATIONS

In her influential book, Deadly Consequences, Deborah ProthrowStith (1991) urges that we take a public health approach to the problem of youth violence. A public health approach emphasizes the need for prevention. Rather than waiting for the violence to occur and then intervene, as we often do in the formal criminal justice approach, the public health approach argues that we need to intervene as early in the process as possible, in a variety of ways, to keep the harm from happening in the first place. A prevention strategy toward violence not only saves victims from being victimized; it also saves offenders from the consequences of their involvement in violent crime. It invests money at the front end of the problem in order to keep from paying a lot more at the back end in victimization and criminal justice process costs. In the long run, prevention saves both money and lives.

The problem is that we often do not think about prevention strategies or any type of long-term solutions. We are so caught up in reacting to the high levels of violence occurring all around us that we grab whatever short-term solutions appear to be available at the time. As Currie (1985) notes, we frequently find ourselves in the position of trying to mop up the flood on the bathroom floor while the tub is overflowing and the faucet is still running. We mop as hard as we can, and we buy increasingly expensive mops to help battle the flood, but until we learn how to turn off the faucet and stop the tub from overflowing we had better be prepared to do an awful lot of mopping.

Given what we know about the connections between poverty, inequality, and social exclusion and the social problem of youth violence, what can we do to begin to turn off the faucet of this violence? What are the policy implications of the theory and research that have been reviewed in this article for the prevention of violence by our nation's youths? Public health professionals like Prothrow-Stith (1991, 140) stress three levels of intervention: primary, secondary, and tertiary. Primary prevention focuses on the larger social or physical environment that contributes to the problem. Secondary prevention involves interventions with people who are at high risk, identifying practices and situations that put them at risk. Tertiary prevention focuses on those who are already afflicted and seeks to minimize the consequences of the problems they are experiencing. The work reviewed in this article has important implications for strategies of primary and secondary prevention of youth violence.

A primary prevention approach to youth violence would focus on the larger structural conditions that shape the problem. It would target what Hagan (1994) referred to as "disinvestment processes," social and economic forces that often lead to violent crime. Specifically, a primary prevention approach would concentrate on the need to reduce poverty and inequality and develop more inclusionary public policies. Space limitations prevent any extended discussion of how to achieve these ambitious goals, but the most important point is that we need to make a commitment to long-run, permanent intervention in the labor market itself. Currie (1985) has argued that "a commitment to full and decent employment remains the keystone of any successful anticrime policy" (263). He points out that we need direct public job creation, policies to upgrade wages and narrow existing disparities in earnings, an improved national system of job training, greater support for workplace organization through the labor movement, policies to spread the social costs of the transfer of jobs abroad, and legislation to shorten work hours and spread available work time (Currie 1996, 1998).

In addition to these labor market interventions to reduce poverty and inequality, we need to attack the social exclusion that breeds violent crime. Barkan (1997) argues that we need to "provide government economic aid for people who cannot find work or who find work but still cannot lift themselves out of poverty" (538). Currie (1998) suggests "providing more generous, universal social services, particularly in the two areas that most distinguish us from less volatile industrial societieschild care and health care" (157). We need only look to these other industrial democracies to see that more inclusionary government welfare programs can have a major impact on families and communities and prevent serious youth violence (Eitzen and Leedham 1998). Measures to end racial segregation in housing and restore the social integration of urban neighborhoods would also constitute primary prevention strategies (Hagan 1994; Barkan 1997).

While primary prevention focuses on structural conditions like poverty and economic inequality, secondary prevention strategies involve the closein institutions of the family, school, and community and the developmental processes that occur within them. A secondary prevention approach to youth violence would point to the need to establish early-childhood intervention programs for high-risk children and their families (Barkan 1997), invest serious resources in the prevention of child abuse and neglect (Currie 1998), improve urban schools beset by "savage inequalities" (Kozol 1991), and invest in skill-building programs for vulnerable adolescents (Currie 1998).

Three recent federal reports review a wide array of secondary prevention programs that have been attempted, and they note numerous programs that seem to work and many others that show some promise (Howell and Bilchik 1995; Muller and Mihalic 1999; Sherman et al. 1997). However, the Maryland Report (Sherman et al. 1997) points out that many crime prevention programs are most likely to work in communities that need them the least. Secondary prevention programs are much less likely to work in urban centers with a high concentration of poverty that are already swamped by much violent crime. According to Walker (1998), this proves that "the heart of the crime problem lies in the underlying structural conditions in high-crime areas" (277). Thus, primary prevention programs that attack these underlying structural conditions should have the highest priority if we are interested in reducing the amount of youth violence in our society.

But we all know that programs designed to reduce poverty and economic inequality in the United States do not have a high priority these days. As Wilson (1996) observes, there has been a "dramatic retreat from using public policy as a means to fight social inequality" (208). Most legislative proposals to provide greater social inclusion through public programs are met with scorn and are doomed to failure. A conservative moral and political philosophy holds sway, arguing that social inequality is necessary to encourage individual initiative and economic efficiency. This leads to a noninterventionist laissez-faire approach to government that relies heavily on the discipline of largely unregulated market forces backed up by a reliance on severe criminal sanctions (Hagan 1994). As long as this ideology dominates political discourse in the United States, the retreat from public policy as a way to alleviate problems of poverty and inequality will continue, with "profound negative consequences for the future of disadvantaged groups such as the ghetto poor" (Wilson 1996, 209). The violence will continue, and more children will die. The tub is still overflowing, and none of our political leaders appears to have the courage to reach in and turn off the faucet.