Why are adolescents violent?


Por que os adolescentes são violentos?



James Garbarino

Psychology Department, Loyola University. 6525 N. Sheridan Road. Chicago, IL 60626. jgarbar@luc.edu




This article discusses how adolescents become violent from the perspective of human development, in which the process of formation of the child and the youth depends on diverse biological, psychological e social variables that constitute the context of life of these individuals. The ecological perspective of human development opposes simple cause-effect relations between antisocial adversities and behaviors and believes that factors such as gender, temperament, cognitive ability, age, family, social environment and culture combine in a complex way influencing the behavior of the child and the adolescent. Some conclusions point to the fact that violence in adolescence usually starts from a combination of early difficulties in relationships associated with a combination of temperamental difficulties. It is concluded that the young seem to be as bad as the social environment surrounding them.

Key words: Adolescent, Violent behavior, Ecological perspective


O artigo discute como adolescentes tornam-se violentos a partir de uma perspectiva do desenvolvimento humano na qual o processo de formação da criança e do jovem depende de diversas variáveis biológicas, psicológicas e sociais que constituem o contexto de vida desses indivíduos. A perspectiva ecológica do desenvolvimento humano contraria simples relações de causa-efeito entre adversidades e comportamentos anti-sociais e acredita que fatores como gênero, temperamento, competência cognitiva, idade, família, entorno social e cultura combinam-se de forma complexa, influenciando o comportamento da criança e do adolescente. Algumas conclusões apontam para o fato de que a violência na adolescência usualmente começa a partir de uma combinação de dificuldades precoces nos relacionamentos, associadas a uma combinação de dificuldades temperamentais. Além disso, coloca-se que os jovens parecem ser tão maus quanto é o seu entorno social.

Palavras-chave: Adolescente, Comportamento violento, Perspectiva ecológica



How can we best approach the question, "why are adolescents violent?" To do so effectively, we need a perspective on human development that begins with the realization that there are few hard and fast simple rules about how human beings develop; complexity is the rule rather than the exception. Rarely, if ever, is there a simple cause-effect relationship that works the same way with all people in every situation. Rather, we find that the process of cause and effect depends upon the child as a set of biological and psychological systems set within the various social, cultural, political, and economic systems that constitute the context in which developmental phenomena are occurring.

This insight is the essence of an "ecological perspective" on human development as articulated by scholars such as Urie Bronfenbrenner1. It is captured in these words: If we ask, does X cause Y?" the best scientific answer is almost always "it depends." It depends upon all the constituent elements of child and context2:

  • gender (for example, the amount of infant babbling predicts childhood IQ in girls but not in boys);
  • temperament (for example, about 10% of children are born with a temperamental proneness to becoming "shy," but this predisposition can be overcome in most children with strong, supportive and long term intervention) ;
  • cognitive competence (for example, abused children who exhibit a pattern of negative social cognition were found to be eight times more likely to develop problems with anti-social aggressive behavior than were abused children who exhibited positive social cognition);
  • age (for example, on average, "unconditional maternal responsiveness" at three months of age predicts "obedience" at 12 months of age, but such unconditional maternal responsiveness at nine months of age does not);
  • family (for example, while nurse home visiting that began prenatally was effective in reducing child abuse in first time births to single mothers from 19% to 4% among a high risk sample, this effect did not occur when there was an abusive man in the mother's household);
  • neighborhood (for example, the correlation between poverty and infant mortality is conditioned by neighborhood factors, being "higher that would be predicted" in some areas and "lower than would be predicted" in others, as a function of the degree to which prenatal services are accessible);
  • society (for example, a study comparing the United States and Canada found that the correlation between low income and child maltreatment was higher in the United States than in Canada);
  • culture (for example, native Hawaiians see the goal of child rearing as producing an interdependent person while most Americas are seeking to create rugged individualists, and as a result, Hawaiian mothers place a high value on infants sleeping with parents and Americans discourage the practice).

This ecological perspective is frustrating: we all would prefer a simple "yes or no" to the question "does X cause Y?" But reality is not obliging on this score. One important corollary of our ecological perspective is the fact that generally it is the accumulation of risks and assets in a child's life that tells the story about developmental progress, not the presence of absence of any one negative or positive influence. For example, Arnold Sameroff's3 classic study included eight risk factors (both parental characteristics— educational level, mental health status, absence, substance abuse— and family characteristics— economic status, race, maltreatment, and number of children).

The results indicated that the average IQ scores of children were not jeopardized by the presence of one or two risk factors. Since research indicates that what matters for resilience is that children reach an "average" level of cognitive competence, (about 100), it is highly significant that children with zero, one or two risk factors averaged 119, 116 and 113, respectively. But IQ scores declined significantly into the dangerous range with the presence of four or more (averaging 90 with four risk factors and 85 with five). In Sameroff's3 research each risk factor weighed equally in the effect; it was the accumulation of risk factors that accounted for the differences. The same is true of developmental assets. Standing against the accumulation of risk are the number of developmental assets in a child's life and the components of resilience. Research conducted by the Search Institute4 has identified 40 developmental assets positive characteristics of family, school, neighborhood, peers, culture, and belief systems. As these assets accumulate, the likelihood that a child or adolescent will be engaged in anti-social violence declinesfrom 61% for kids with 0-10 assets to 6% for kids with 31-40, for example. Asset accumulation predicts resilient response to stress and challenge.

For example, if we ask, "does absence of a parent produce long lasting negative effects?" the answer is, as it always is, "it depends." That's always the answer in general, but once we know what else a particular child is facing— poverty? drug abuse in a parent? child abuse? racism? too many siblings? We can move closer to "yes, probably" or "no, probably not." And one important influence on the developmental impact of these contingencies is always the temperament of the child.

Each child offers a distinctive emotional package, a temperament. Each child shows up in the world with a different package of characteristics. Some are more sensitive; others are less so. Some are very active; some are lethargic. Why are these differences important? For one thing, they affect how much and in what direction the world around them will influence how they think and feel about things.

What one child can tolerate, another will experience as highly destructive. What will be overwhelming to one child will be a minor inconvenience to another. Knowing a child's temperament goes a long way towards knowing how vulnerable that child will be in the world, particularly in extreme situations. Thomas and Chess5 classic research on temperament in the United States reported that while about 70% of "difficult" babies evidenced serious adjustment problems by the time they entered elementary school, for "easy" babies the figure was 10%.

Most children can live with one major risk factor; few can handle an accumulation of them. Getting from a generalized "it depends" to a more specific assessment of the likely fate of any child lies in accounting for all the elements of accumulated risk factors, developmental assets, and temperament to determine the odds of success or failure.

Although it is defined in numerous ways, resilience generally refers to an individual's ability to stand up to adverse experiences, to avoid long term negative effects, or otherwise to overcome developmental threats. Many of us know a child whose life is a testament to resilience. The concept of resilience rests on the research finding that while there is a positive correlation between specific negative experiences and specific negative outcomes, in most situations a majority (perhaps 60%-80%) of children will not display that negative outcome. All children have some capacity to deal with adversity, but some have more than others and are thus more "resilient", while others are more "vulnerable" in difficult times. But some children face relatively easy lives while others face mountains of difficulty with few allies and resources.

Resilience is not absolute. Virtually every child has a "breaking point" or an upper limit on "stress absorption capacity". Kids are "malleable" rather than "resilient", in the sense that each threat costs them something–and if the demands are too heavy the child may experience a kind of psychological bankruptcy. What is more, in some environments virtually all children demonstrate negative effects of highly stressful and threatening environments. For example, in his Chicago data psychologist Patrick Tolan6 reports that none of the minority, adolescent males facing the combination of highly dangerous and threatening low income neighborhoods coupled with low resource/high stress families was resilient at age 15 when measured by continuing for a two year period as neither being more than one grade level behind in school nor having sufficient mental health problems so as to warrant professional intervention.

What is more, resilience in gross terms may obscure real costs to the individual. Some children manage to avoid succumbing to the risk of social failure as defined by poverty and criminality but nonetheless experiences real harm in the form of diminished capacity for successful intimate relationships. Even apparent social success - performing well in the job market, avoiding criminal activity, and creating a family- may obscure some of the costs of being resilient in a socially toxic environment such as if faced by millions of children. The inner lives of these children may be fraught with emotional damage - to self-esteem and intimacy, for example. Though resilient in social terms, these kids may be severely wounded souls.

Why are adolescents violent? The simplest answer is this: they are violent because children they did not learn to succeed in non-violent strategies for meeting their needs and responding to emotions like anger, frustration and fear. Of course most adolescents indeed–most human beings of any age–are capable of violent behavior. The success of military training in producing soldiers who are capable of lethal violence "on command" is evidence of that. But the main story when it comes to adolescent violence starts in childhood.

Virtually all children express aggressive behavior in infancy and early childhood, so the real issue is not "how do children become aggressive?" but rather "why do some children continue to be aggressive and thus set the stage for becoming violent teenagers?" Research reveals that the two principal processes that control the developmental pathway for aggression in childhood are the ideas a child learns about aggression ("cognitive structuring") and the experiences a child has in situations where aggressive behavior is modeled and reinforced ("behavioral rehearsal"). Some children receive consistent messages about that reduce the legitimacy of aggression ("don't hit"), while others receive messages that legitimatize aggression ("fight back when attacked" and "aggression is successful"). Similarly, some children observe parents, siblings and peers resolving conflict non-aggressively while others observe abuse and fighting. Just mapping patterns of cognitive structuring and behavioral rehearsal goes a long way towards understanding why some kids arrive at adolescence with a high level of aggressive behavior while others don't. We must always remember that virtually everyone who commits an act of violence believes it is justified—at least at the time of committing the act. All this helps explain why boys have traditionally engaged in more physical aggression than girls. Girls have been taught "girls don't hit" and were generally excluded from situations where they could practice being physically aggressive, while boys were taught "boys do hit—it's just a matter of learning who, when and where to hit" and were welcomed into situations where physical aggression is normal (most notably competitive sports). And these same processes help explain why the gap between boys and girls in the matter of physical aggression is narrowing as girls are told and shown that "girls do hit" and have a chance to participate in settings where they can practice being physically aggressive. The ratio of girls' to boys' participation in high school sports changed in the past 30 years changed from 1: 32 to 1: 1.5. All this may help explain findings like these: The ratio of male to female arrests for assaults changed in the last 20 years from 10:1 to 4:1, and research on the effects of televised violence on aggressive behavior in children shows that in the 1960s girls were immune to the effect and now show the same effects as boys2.

The kids most at risk from bringing a pattern of serious childhood aggression into adolescence are those who have developed a chronic pattern of bad behavior and violating the rights of others, kids who might be described as meeting the criteria for what mental health professionals call "conduct disorder". Research on the links between "risky thinking" reveals that the odds that a child will develop conduct disorder reflect the intersection of being abused and having lower levels of social information processing (being hypersensitive to negative social information and oblivious to positive social information as well as having very limited ideas of alternatives to physical aggression as a social tactic and also carrying around the belief that aggression is a successful social tactic).

Knowing all this, what conclusions can parents draw?2

The first of these conclusions is that no matter how effective, motivated, and attentive any of us is as a parent, our children go to school with kids who are prone to behavior violently–some of whom have access to lethal weapons. There are kids–mostly boys–in every school who have developed a pattern of aggressive behavior, who have established an internal state in which they see themselves as victimized by peers and society, and whose emotions and moral judgments have become harnessed to their aggressive rage. These youth can make the transition to murder readily if weapons are available and they reach a crisis state. Knowing how these kids reach this point and what we can do reclaim them empowers us to reduce the odds that they will commit acts of lethal violence.

My second conclusion is that the problem of lethal youth violence usually starts from a combination of early difficulties in relationships that are linked to a combination of difficult "temperament" and negative experience. Every parent who knows children knows that children come equipped with different temperaments. Some are sunny and easy; others are stormy and difficult. Some children are easy to parent; others are very challenging. Some are so difficult that no "normal, average" parent will be able to succeed without expert professional advice and support. When it comes to developing patterns of aggression, some of the difficulties lie in being impulsive, emotionally insensitive, having a high activity level, being of less than average intelligence, and being relatively fearless.

These temperamental problems do not spell doom, however. What matters is how well the parenting and educational experiences of these children meet the challenges posed by their difficult temperaments. Of special concern are two patterns. The first is a pattern of escalating conflict in the parent-child relationship, in which parent and young child get caught up in mutually coercive and aversive interactions. The second is a gradual process of emotional detachment arising when parents and teachers abandon these children by withdrawing from them in the face of their negative behavior.

These patterns of response increase the odds that these vulnerable children will become increasingly frustrated and out of sync as they meet up with the challenges of paying attention in school. In a culture likes ours, in which there is such intense cultural imagery that legitimizes and models violence, this emotional abandonment is particularly dangerous. Parent education starting before children are born and continuing through until adolescence is crucial for preventing violence.

Once they are "lost" this way they tend to form into aggressive and anti-social peer groups that build negative momentum throughout childhood and into adolescence. This can be avoided. For example, research by Sheppard Kellam7 and his colleagues demonstrates that if the first grade class room is well organized and provides clear messages about behavior, aggressive kids are reclaimed and their aggressive behavior tamed. If the classroom is chaotic these boys form negative peer groups and their problems with aggression intensify. My own effort—"Let's Talk About Living in a World With Violence"8—has demonstrated its ability to reduce aggression among third graders when used by a teacher who is comfortable dealing with issues of aggression and who integrates these concerns (and the program materials) into the general classroom curriculum.

Children whose difficult temperament and experience put them on track for problems with aggressive behavior need help from parents and teachers to learn to manage their behavior. Teachers need special skills and a high level of motivation to create classroom environments that prevent violence.

Research shows that patterns of aggression start to become stable and predictable by the time a child is eight years old: unless we do something to intervene, children identified as aggressive at this age will tend to be aggressive 30 years later (becoming adults who are violent in their families, get involved in fights in the community, and drive their cars aggressively).

My third conclusion is that the most common pathway to this pattern of aggression at age eight is for temperamentally vulnerable children to be the victims of abuse and neglect at home and as a result to develop a negative pattern of relating to the world in general. This maltreatment can be both physical abuse (beatings) and psychological abuse (rejection).

The negative pattern that results has four parts: 1) being hypervigilant to the negatives (such as threatening gestures) in the social environment around them; 2) being oblivious to the positives (such as smiles); 3) developing a tendency to respond aggressively when frustrated; and, 4) drawing the conclusion that aggression is successful in the world. According to research by psychologist Kenneth Dodge9 and his colleagues, this negative pattern is the most potent link between a child being the victim of maltreatment and developing a pattern of chronic bad behavior and aggression (what will be diagnosed by mental health professionals as "conduct disorder"). Being abused produces a seven fold increase in the odds of developing conduct disorder. About a third of these children with conduct disorder will eventually become violent, delinquent youth (and about 90% will go on to demonstrate some serious problem in adulthood). In juvenile prisons, typically about 80% will have shown this negative pattern. Child abuse prevention is the cornerstone of preventing lethal youth violence.

My fourth conclusion is that troubled kids will be as bad as the social environment around them. I have identified this as the issue of "social toxicity", the presence of social and cultural "poisons" in the world of children and youth, to which troubled kids—boys and girls—are especially susceptible. Just as asthmatic children are most affected by air pollution, so "psychologically asthmatic" children are most affected by social toxicity.

The glorification of violence on television, in the movies, and in video games is part of this social toxicity, and it affects aggressive kids more than others. The same is true for the size of high schools. Academically marginal students are particularly affected in a negative way by being in big schools (with more than 500 students grades 9-12). The availability of drugs and guns is another example. Mobilizing community leaders, parents, professionals, and youth themselves can provide a rallying point for improving the social environment. Where gangs operate and vulnerable kids live in a "war zone" the social environment can draw many kids into patterns of severe—even lethal —violence. Detoxifying the social environment of children and youth is essential to protect children and youth from the problem of lethal violence.

My fifth conclusion is that at the core of the youth violence problem is a spiritual crisis. Human beings are not simply animals with complicated brains. Rather, we are spiritual beings having a physical experience. This recognition directs our attention to the multiple spiritual crises in the lives of violent adolescents. They often have a sense of "meaninglessness," in which they are cut off a sense of life having a higher purpose. By the same token, they often have difficulty envisioning themselves in the future. This "terminal thinking" undermines their motivation to contribute to their community and to invest their time and energy in schooling and healthy lifestyles. Finally, they often have lost confidence in the ability and motivation of the adults in their world to protect and care for them. This leads them to adopt the orientation of "juvenile vigilantism". A teenager says, "If I join a gang I am 50% safe; if I don't join a gang I am 0% safe". The point is that adults don't enter into the equation.

Non-punitive, love-oriented religion institutionalizes spirituality and functions as a buffer against social pathology, according to research reviewed by psychologist Laura Andrew Weaver10. On the other hand, the shallow materialist culture in which we live undermines spirituality and exacerbates these problems. One way to deal with these issues is to have schools join with community leaders to embrace the national character education campaign, as developed, for example, by psychologist Thomas Lickona11. Character education offers all positive elements within a community a focal point for their actions. It provides a framework in which to pursue an agenda that nourishes spirituality.

Over the past 25 years there has been a doubling of the percentage of children and youth who have mental health and developmental adjustment problems severe and chronic enough to warrant professional intervention, according to the research of psychologist Tom Achenbach12. The spreading problem of youth violence is related to this larger development. Dealing with it will require both a broadly-based prevention perspective on community life, and a conscious focusing of attention of dealing humanely and effectively with troubled aggression children in childhood lest they fall in line to proceed down the pathway to youth violence.



1. Bronfenbrenner U. The ecology of human development. Cambridge, MA: Harvard University Press; 1979.         

2. Garbarino J. Lost Boys: Why our sons turn violent and how can we save them. New York: The Free Press; 1999.         

3. Sameroff AJ. Environmental Risk Factors. Pediatrics 1998;102(Supl 5):1287-1292.         

4. Search Institute. 40 Developmental Assets® for Adolescents (ages 12-18) [site on the Internet], 1997. [cited 2008 Oct 04].[about 1p.]. Available from: http://www.search-institute.org/system/files/40AssetsList.pdf         

5. Thomas A, Chess S. Temperament and Development. New York: Brunner-Mazel; 1977.         

6. Gorman-Smith D, Tolan P. Positive Adaptation among Youth Exposed to Community Violence. In: Luthar SS. Resilience and vulnerability: Adaptation in the Context of Childhood Adversities. New York: Columbia University; 2003.p. 392-413.         

7. Kellam SG, Rebok GW, Ialongo N, Mayer LS. The course and malleability of aggressive behavior from early First Grade into Middle School: Results of a Developmental Epidemiology-Based Preventive Trial. Journal of Child Psychology and Psychiatry and Allied Disciplines 1994; 35(2):259-281.         

8. Garbarino J, Holden M. Let's Talk About Living in a World With Violence. New York: Cornell University;1995.         

9. Dodge KA, Pettit GS, Bates JE. How the experience of early physical abuse leads children to become chronically aggressive. In: Cicchetti C, Toth SL, editors. Developmental psychopathology: developmental perspectives on trauma. Rochester, NY: University of Rochester Press; 1997. p. 263-288.         

10. Weaver LAJ, Flannelly KT, Flannelly HJ, Koenig DG, Larson B. An analysis of research on religious and spiritual variables in three major mental health nursing journals, 1991-1995. Issues in Mental Health Nursing 1998; 19(3):263-276.         

11. Lickona T. Educating for character: how our schools can teach respect and responsibility. New York: Bantam; 1991.         

12. Achenbach T, Howell C. Are American children's problems getting worse? A thirteen-year comparison. J Am Acad Child Adolesc Psychiatry 1993; 32(6):1145-1154.         



Artigo apresentado em 05/11/2007
Aprovado em 02/06/2008
Versão final apresentada em 30/09/2008

ABRASCO - Associação Brasileira de Saúde Coletiva Rio de Janeiro - RJ - Brazil
E-mail: revscol@fiocruz.br