The causes of violence.

I have been asked to summarize everything that we really know about the biological, sociological, and psychological causes of violence--in 20 minutes or less. Unfortunately, I think I can do it.

But, I warn you in advance what I cannot do-what no one can honestly do--and that is to offer a neat, simple story that explains why so many Americans are afraid to walk home alone at night. Only people on the extremes of the political spectrum have that luxury and that conceit.

The political right believes that the root cause of violent crime is bad genes or bad morals. Not so, says the left. The root cause of violent crime is bad housing or dead-end jobs. And, I tell you that while doing something about the causes of violence surely requires a political ideology, the only way we can determine what those causes are in the first place is to check our ideologies at the door and to try to keep our minds open as wide, and for as long, as we can bear.

I realize that this is not easily done. But, if you give it a try, which I urge you to do, I think that you will find that violence does not have one root cause. Rather, violence has many tangled roots. Some grow toward the left and some grow toward the right. We have to find the largest ones, whichever way they grow, and only then can we debate how to cut them off.

Biological Causes

First, the biological causes. These are the easiest to talk about, because there is not much to say.

Many biological factors have been nominated as candidates for causes of violence. Hormones like testosterone, transmitters in the brain like serotonin, and blood abnormalities like hypoglycemia are only a few that have been mentioned.

Biological factors do not have to be hereditary. They could be caused by a head injury, poor nutrition, or environmental events, such as exposure to lead paint. Fortunately, the National Academy of Sciences just reviewed hundreds of studies on the relationship between biology and violence, and it came to one clear bottom-line conclusion: "No patterns precise enough to be considered reliable biological markers for violent behavior have yet been identified."(1) The National Academy of Sciences found many promising leads that should be vigorously pursued by researchers, but so far, it could point to nothing as a proven, or even close to proven, biological risk factor for future violence. Sociological Causes

Next come the sociological causes. We know the most about social factors and violence, because social factor, such as demography, are relatively easy to measure and because people have been measuring them for a long time. What do we know? We know a great deal about a relatively small number of things.

We know that to live in America is to live in the land of the brave, as well as in the home of the free. We are all familiar with depressing statistics about the U.S. trade deficit with Japan. But more depressing is this Nation's crime surplus. Compared with Japan, a nation of roughly comparable industrialization, with cities much more crowded than ours, the U.S. homicide rate is over 5 times higher, the rape rate is 22 times higher, and the armed robbery rate is an astounding 114 times higher.(2)

We also know that within America, violence is subject to great regional variation. The murder rate, for example, is almost twice as high in the South as it is in the Northeast, but the robbery rate is almost twice as high in the Northeast as it is in the South.(3)

We know that communities within all regions of America differ drastically among themselves in how violent they are. In general, the smaller the community, the lower the rate of violence. Within the same city, some neighborhoods have rates of violent crime 300 times higher than other neighborhoods.(4) We know that people who commit violence on the street are disproportionately poor and unemployed. Prior to their arrest, jail inmates had, on the average, an annual income at the Federal Government's official "poverty level," and about one-half were unemployed at the time they committed a violent crime.(5)

We know that the overwhelming majority---close to 90 percent--of the people arrested for crimes of violence are men and that despite enormous changes in gender roles in recent decades, this figure has not budged for as long as criminal records have been kept.(6) Indeed, there is no place in the world where men make up less than 80 percent of the people arrested for violence, now or at any time in history.(7)

We know that violence is primarily the work of the young. People in their late teens and twenties are much more likely to be arrested for violence than younger or older people.(8)

We know that the arrest rate--and the victimization rate--for violent crime for African-Americans is now about six times higher than for whites.(9)

Finally, we know that official violent crime rates, as high as they are, drastically underestimate the actual rate of violence in America, particularly violence within the family.(10)

After this, what we know about the sociological correlates of violence falls off rapidly. Note that I said "correlates," not "causes."

Two problems keep us from knowing which factor really matters as a cause of violence and which is irrelevant. One problem is that each factor relates not only to violence but to other sociological factors as well. Call this the "ball of wax" problem. Poverty and race, for example, are related not just to violence but also to each other. If poverty is taken into account, the effect of race on violence decreases drastically, and in some studies, disappears entirely. The second problem is that it is sometimes hard to tell which came first, the sociological factor or the violence. Call this the "cause and effect" problem.

It is true, of course, that violence does not cause people to be male or to be young. But it is not clear whether unemployment leads people to commit violent acts or whether, for at least some people, their violent acts lead employers to not want to hire them. It is also possible that, at least for some people, a third factor--like an "impulsive" temperament---causes them both to be violent and to be unlikely to keep a steady job.(11)

Psychological Causes

Finally, the psychological causes. If research on violence were like stock on Wall Street, then I would put my money right now on psychology. By this, I most emphatically do not mean mental disorder. The best epidemiological evidence indicates that major mental disorder accounts for, at most, 3 percent of the violence in American society.(12)

What I mean, instead, are the developmental processes that we all go through, most of us more or less successfully, but some of us with great difficulty. I mean particularly the family(13)--the filter through which most of the sociological factors, such as a parent's being unemployed, and many of the biological factors, like poor nutrition, seem to have their effect on a child growing up.

There is a risk, of course, that whenever someone talks about families and children, that person invokes images that may never have existed, except perhaps on 1950's television. And, even if these images did once exist, they surely no longer reflect the great variety of relationships in contemporary America. But, whether we prefer Ozzie and Harriet Nelson or Murphy Brown, there is one important thing we should not forget. That is, all types of families share something in common. Whether they are married or cohabitating, biological or adoptive or foster, single or dual, gay or straight, and whatever their ethnicity, virtually all parents try to raise their children to be neither the victims nor the perpetrators of violence.

Fortunately, most families, whatever their type, succeed. Unfortunately, some fail.

Family, Children, and Violence

What do we know about families and children and violence?

We know that while many aggressive children go on to be law-abiding adults, aggression at age 8 significantly predicts violent convictions well into the thirties, in every culture in which it has been studied.(14)

We know that most children who have been physically abused by their parents go on to be perfectly normal adults. Yet, physical abuse doubles the risk that a boy will have convictions for violent crime as an adult.(15)

We know that failure of a child in school is one of the most enduring correlates of later violence. Four out of five violent offenders in prison never finished high school.(16)

We know that stability matters. The more changes of placement a foster child experiences while growing up, the more likely that child will later be arrested for a violent crime.(17)

We know that lack of parental supervision has been consistently related to delinquency, including violent delinquency. One study, for example, found that 10 percent of nondelinquents were poorly supervised by their parents, one-third of one- and two-time delinquents were poorly supervised, and over three-quarters of repeat offenders were poorly supervised.(18) Another study found that for children growing up in very disadvantaged and violent neighborhoods, who look like they have everything going against them, the one factor that seems to protect that child from growing up to be violent is having a parent--overwhelmingly, a mother--who supervises her child very strictly and who nips misbehavior in the bud, rather than waiting for the principal to call or the police officer to knock on the door.(19)

Finally, we know much about the relationship between illegal drugs and violence. But it is important to remember that the connection between one legal drug--alcohol--and violence is beyond dispute. About one-third of all violent offenders are alcoholic, and the earlier an adolescent starts to drink, the more likely that teen will be violent as an adult.(20)

These findings are not immune from either "ball of wax" or "cause and effect" problems. Failure in school, for example, is associated not only with violence but also with poor parental supervision. And, it is not obvious whether frequent changes of placement for a foster child leads to violence, or whether a child's violence at home leads foster parents to give the child back to the agency. But surely, the accumulated findings provide reason to believe that families have an enormous influence, for better or worse, on how children develop. None of these findings in any way negates the influence of social conditions in giving rise to violence. Poor people, for example, without adequate child care, may have a much more difficult time monitoring their children's behavior than affluent people with live-in help.

Nor do the findings necessarily negate the possible influence of biological factors. Nutrition, to give another example, is something that parents literally put on the table for the child to eat. But it is through the family that these things have their effects and through the family that those effects might best be redirected.

We know some important things about violence, particularly about the home environment and violence. But, we do not know nearly enough about how to prevent violence in the first place or how to stop it from happening again once it begins. How can we learn more, so that 10 years from now, it will take a bit longer to summarize the field?

Learning About Violence

We can learn more if we do four things. We need to 1) make a long-term national investment in research and development, 2) have a coherent and coordinated Federal strategy for studying violence, 3) implement a comprehensive and inclusive violence research agenda, and 4) institute a program of rigorously evaluated interventions to reduce violence.

Long-term national investment in research and development for a safer America--It takes resources to isolate the biological, sociological, and psychological factors that are associated with violence, to untangle the bail of wax in which they are found, and to determine which are the causes of violence and which are its effects. The National Academy of Sciences just did an audit and concluded that the Federal Government spends a total of $20 million a year on violence research, which works out to about $3 per violent victimization.(21) Researchers always say that more money is needed for research. But let me point out that the Nation's budget for research on violence is considerably less than one-half what the Federal Government will spend this year on mohair price subsidies.(22) Nothing against goats, but a shortage of fuzzy sweaters is not what is keeping people behind locked doors at night.

A coherent and coordinated Federal strategy for studying violence--Organizational responsibility for research on violence is spread across a number of Federal agencies--the National Institute of Justice, the National Institute of Mental Health, the National Science Foundation, the Centers for Disease Control, and several smaller programs.(23) Surely, we do not need a "violence czar" to provide central management of the Nation's research on violence. But we do need to be sure that all bases are covered and that there is a forum where innovative ideas can be shared and followed up quickly. Partnerships with private foundations may be particularly cost-effective. The collaborations between the MacArthur Foundation and the National Institute of Justice in funding the Program on Human Development and Criminal Behavior and between the MacArthur Foundation and the National Institute of Mental Health in funding the MacArthur Risk Assessment Study are exciting examples of strategic leveraging of public and private resources.(24)

A comprehensive and inclusive violence research agenda--The agenda needs to promote the three kinds of research I mentioned-biology, sociology, and psychology. It has to study them not in isolation from one another but together as different pieces of the same puzzle.

The time is ripe to give some priority to studying developmental influences and the effect of the family environment on violence. But this has to include health-related and biological factors that are mediated through the family, as well as social and psychological influences. You cannot paint a full, lifelike picture of the causes of violence if you mark a corner of the canvas ideologically off limits before you start painting.

A program of rigorously evaluated interventions to reduce violence--This goes to the top of the agenda. We will finally understand the causes of violence when we can take a group of children at high risk of becoming violent and ethically offer them opportunities and services to defy our predictions.

The interventions should be intensive and broadly based in practice, but initially, small-scale in scope. We simply do not know enough to mount major national programs to attack the causes of violence, even if we had the money to do so. But we certainly do know enough to start trying many things in a completely voluntary way, without unnecessarily labeling anyone, and see what works.(25)

One Approach

One modest idea is derived from the research on child rearing that finds parental supervision so important in preventing crime and violence. Taking a cue from studies like this, we could offer an intensive, long-term, state-of-the-art education program to a random group of parents whose children are enrolled in Federal child care programs.(26) This program would teach parents how to effectively monitor their children's behavior, how to recognize potentially serious misbehavior when it occurs, and how to consistently, but fairly, discipline their children in response to misbehavior.(27) If this worked, if children whose parents received the program had lower levels of aggression and other social problems when compared to a control group, we could gradually expand the program, rigorously evaluating its effects each step of the way. If it did not work, we would go back to the drawing board, roll up our sleeves, and try something different.

A dozen ideas like this--none of them panaceas--could be derived from research on children and families and tried simultaneously in different parts of the country. If even a few of them worked, we would have taken a giant leap forward in violence prevention.

Conclusion

The short of it is that first, we need to make a national scientific commitment to understand the causes of violence. Once this happens, we need to make a national political commitment to do something about them.

Endnotes

1 A. Reiss and J. Roth, eds., Understanding and Preventing Violence (Washington, DC: National Academy Press, 1993), 116.

2 T. Westermann and J. Burfeind, Crime and Justice in Two Societies: Japan and the United States (Pacific Grove, CA: Brooks/Cole, 1991).

3 R. Nisbett, "Violence and U.S. Regional Culture," American Psychologist, 48, 1993, 441-449.

4 Supra note 1, p. 88.

5 Bureau of Justice Statistics, Report to the Nation on Crime and Justice, 2d ed. (Washington, DC: U.S. Department of Justice. 1988) 149.

6 Supra note 1. p. 72.

7 J. Wilson and R. Herrnstein, Crime and Human Nature (New York: Simon and Schuster, 1985).

8 A. Blumstein, J. Cohen, J. Roth, and C. Visher, Criminal Careers and "Career Criminals" (Washington, DC: National Academy Press, 1986).

9 Supra note 1. p. 71.

10 J. Weis, "Family violence research methodology and design," in Family Violence, L. Ohlin and M. Tonry, eds. (Chicago. IL: University of Chicago Press, 1989). 117-162.

11 See J. Monahan and L. Walker, Social Science in Law: Cases and Materials, 3d ed. (Westbury, NY: Foundation Press, 1994).

12 J. Monahan, "Mental disorder and violent behavior: Perceptions and evidence." American Psychologist, 47, 1992, 511-521.

13 R. Loeber and M. Stouthamer-Loeber. "Family factors as correlates and predictors of juvenile conduct problems and delinquency," in Crime and Justice: An Annual Review of Research, vol. 7, M. Tonry and N. Morris, eds. (Chicago, IL: University of Chicago Press, 1986), 29-149.

14 D. Farrington, "Childhood aggression and adult violence: Early precursors and later-life outcomes," in The Development and Treatment of Childhood Aggression, D. Pepler and K. Rubin, eds. (Hillsdale, NJ: Erlbaum, 1991). 5-29. 15 C. Widom, "The Cycle of Violence," Science, 244, 1989, 160- 166.

16 Supra note 5, p. 48.

17 Supra note 1, p. 243.

18 G. Patterson and M. Stouthamer-Loeber, "The correlation of family management practices and delinquency," Child Development, 55, 1984, 1299-1307. 19 H. Wilson, "Parenting in Poverty," Journal of Social Work, 4, 1974, 241-254. 20 Supra note 1, p. 185.

21 Supra note 1, p. 345.

22 Budget of the United States Government--Fiscal Year 1993 (Washington, DC: Government Printing Office, 1992), Appendix One-349.

23 Supra note 1, p. 349.

24 Ibid.

25 E. Mulvey, M. Arthur, and N. Reppucci, "The prevention and treatment of juvenile delinquency: A review of the research," Clinical Psychology Review, 13, 1993, 133-167.

26 E. Zigler and S. Styfco, eds., Head Start and Beyond: A National Plan for Extended Childhood Intervention (New Haven, CT: Yale University Press, 1993).

27 M. Gottfredson and T. Hirschi, A General Theory of Crime (Stanford, CA: Stanford University Press, 1990).

Dr. Monahan, a psychologist and a professor at the School of Law at the University of Virginia in Charlottesville, delivered this speech at the U.S. Sentencing Commission's Inaugural Symposium on Crime and Punishment in the United States in Washington, DC.