Emory Report

April 13, 1998

 Volume 50, No. 28

First Person

Society needs to get tough
on causes of crime, says Agnew

More than 40 states-including Georgia-have recently passed legislation making it easier for juveniles to be tried in adult court. This legislation is part of a larger trend to get tough with serious juvenile offenders. Most states have also increased the penalties given out to certain offenders in juvenile court. As a consequence, there has been a large increase in the juvenile incarceration rate. Most confined juveniles are now in overcrowded facilities that suffer from a range of problems, as illustrated by the recent federal investigation of Georgia's juvenile correctional system.

This "get tough" legislation was passed in response to a real problem. Juveniles ages 10-17 make up 13 percent of the population but account for 31 percent of all arrests for Index, or serious, crimes and 19 percent of all arrests for violent crimes. The rate of juvenile violence increased sharply from the mid-1980s to 1994, with the juvenile homicide rate more than doubling (juvenile violence, including homicide, is now declining).

On the face of it, it seems this get tough approach should reduce juvenile crime: More severe penalties should deter juveniles from committing crime, and those chronic offenders who are not deterred can be locked away for long periods (most serious crime is committed by a small group of chronic offenders, about 5-10 percent of the adolescent population). Several studies, however, suggest the impact of these get tough strategies has been overstated.

These strategies often fail to deter the juveniles who are punished. Certain data suggest that juveniles tried in adult court are more likely to re-offend than similar youth tried in juvenile court (e.g., "The Transfer of Juveniles to Adult Court," Donna M. Bishop et al., Crime and Delinquency 42 (1996): 171-191). Likewise, studies indicate that many other get tough strategies-like the use of military-style boot camps-often fail to deter offenders (e.g., Boot Camps for Juvenile Offenders, U.S. Department of Justice, 1997).

Further, these strategies are unlikely to deter youth in the general population. The probability of punishment is more important than the severity of punishment in deterring crime. In particular, threatening juveniles with severe punishments has little impact if few juveniles are caught and punished. At present, the large majority of juvenile crimes never become known to the police, and only about 20 percent of those crimes known to the police result in arrest. Threats of severe punishment, then, have little impact. And it is unlikely that we can increase the probability of punishment to the point that it would have a substantial deterrent effect (although this might be done for select types of crime in certain locations-see Juvenile Gun Violence and Gun Markets in Boston, David Kennedy, U.S. Department of Justice, 1997).

Finally, the strategy of confining chronic offenders for long periods has only a modest effect on crime. It is difficult to identify such offenders early in their careers, when confinement would have the largest impact. And the early confinement of chronic offenders does not prevent as many crimes as was first thought (among other things, juveniles usually commit their crimes in groups and the other group members may continue to offend at a high rate). (See Chapters 15 and 22 in Criminology, Joseph Sheley, Wadsworth, 1995).

So what should we do? I suggest that we place less emphasis on getting tough with juvenile offenders and more on attacking the root causes of delinquency (prevention) and providing treatment to juvenile offenders (rehabilitation). This is not a popular strategy at present-many people believe that rehabilitation and prevention do not work and that serious offenders deserve severe punishments.

However, we now have a reasonably good idea of the causes of delinquency. Such causes include traits like impulsivity and low verbal IQ, child abuse, family conflict, poor parenting practices, poor school performance, low commitment to school, association with delinquent peers and gang members, and growing up in poor, disorganized communities. A number of prevention and treatment programs have been developed to address these causes, like pre- and postnatal health care programs, preschool enrichment programs, parent-training programs, and a variety of school-based interventions.

These programs are successful in reducing delinquency or show strong promise of success. One recent review concluded that well-designed treatment programs reduce re-offending rates by about 50 percent on average (see The Psychology of Criminal Conduct, D.A. Andrews and James Bonta, Anderson, 1994). These programs are also cost effective-more than paying for themselves in terms of the crime (and other problems) they prevent. And these programs represent a more just approach to the control of delinquency since they recognize that crime is at least partly caused by factors beyond the juvenile's control-like poverty and poor parenting.

At the same time, I should note that it is true many rehabilitation/prevention programs do not work. Some may even increase re-offending-like Scared Straight-type programs that involve encounters with hard-core prisoners. The prevention and rehabilitation programs we employ need to be carefully selected and implemented, and we should insist that all programs be properly evaluated-something that rarely happens today.

Also I should note that the use of such programs does not mean we need to stop holding juveniles (partly) accountable for their behavior. It simply means we place more emphasis on identifying and helping "at risk" juveniles and providing treatment for those who have already offended. For a strategy that combines accountability and with an emphasis on prevention/rehabilitation, see Combating Violence and Delinquency: The National Juvenile Justice Action Plan, U.S. Department of Justice, 1996.

It is time that we address the underlying problems that cause delinquency rather than simply responding to the symptoms of these problems.

Robert Agnew is a professor in the Department of Sociology.

Return to April 13, 1998 Contents Page