View metadata, citation and similar papers at core.ac.

uk brought to you by CORE

provided by ValpoScholar

Valparaiso University Law Review

Volume 31
Number 2 Symposium on Juvenile Crime: Policy pp.395-418
Proposals on Guns & Violence, Gangs, & Drugs

Symposium on Juvenile Crime: Policy Proposals on Guns & Violence,

Gangs, & Drugs

Youth Homicide: Keeping Perspective on How Many Children Kill

Eric R. Lotke

Follow this and additional works at: https://scholar.valpo.edu/vulr

Part of the Law Commons

Recommended Citation
Eric R. Lotke, Youth Homicide: Keeping Perspective on How Many Children Kill, 31 Val. U. L. Rev. 395
(1997).
Available at: https://scholar.valpo.edu/vulr/vol31/iss2/5

This Symposium is brought to you for free and open

access by the Valparaiso University Law School at
ValpoScholar. It has been accepted for inclusion in
Valparaiso University Law Review by an authorized
administrator of ValpoScholar. For more information,
please contact a ValpoScholar staff member at
scholar@valpo.edu.
 Lotke: Youth Homicide: Keeping Perspective on How Many Children Kill

YOUTH HOMICIDE:
KEEPING PERSPECTIVE ON
HOW MANY CHILDREN KILL
ERIC R. LoTKE

I. INTRODUCTION

The headlines create the impression of a nation in crisis. Juvenile homicide

hits all-time high, they declare. Scourge of youth violence sweeping the nation.
Politicians lament the death of our youth and vow to keep neighborhoods safe.
Teachers warn students to shun attractive clothing, fearing they will be shot by
children who plan to make it their own. Rarely have alarm bells rung so loudly
or so long; even good news like the recent decline in juvenile homicide was
followed by warnings that the worst is yet to come.'

Two problems of juvenile violence face our nation. The first problem is
that certain neighborhoods have suffered from tremendous increases in youth
violence. In these neighborhoods, youth homicide has doubled or even tripled
in the past decade. The increase in homicide is itself distressing, and it suggests
other troubles lurking beneath.

The second problem is our national response to the first problem. This
problem arises from sympathy for the victim and fear of victimization; it ends
with a loss of perspective on the small scale and a limited range of youth
violence. Although American homicide rates are high and youth homicide is
rising, only a tiny fraction of Americans run a real risk of homicide, and only
a tiny fraction of those homicides are committed by children. Most cities that
show rapid increases in youth homicide have changes on the scale of three

Research Associate, National Center on Institutions & Alternatives. NCIA acknowledges the
assistance of many people who made this report possible. The Annie E. Casey Foundation
recognized the need to provide numerical perspective on the problem of youth homicide and
provided funding to make it possible. Barry Krisberg of the National Council on Crime and
Delinquency provided the support of his organization, especially research associate Dom Del
Rosario, to manipulate data bases on admissions to locked facilities into useable form. Melissa
Sickmund of the National Center for Juvenile Justice provided statistics on court processing and
excellent methodological advice. David Altschuler of the Center for the Study of Social Policy at
Johns Hopkins University and Lindsay Hayes of NCIA helped to edit the report. Mary Cate Rush
tirelessly crunched data, solved technical problems and provided statistical insight as this report
moved from inception to publication. Finally, NCIA wishes to thank the employees of many states
who tried, often in vain, to provide direct information regarding their states.
1. This fear is based on a demographic bulge of children under 10 years old who will be
teenagers entering their crime prone years in the next decade.

395

Produced by The Berkeley Electronic Press, 1997

 Valparaiso University Law Review, Vol. 31, No. 2 [1997], Art. 5

396 VALPARAISO UNIVERSITY LAW REVIEW [Vol. 31

homicides increasing to six homicides-a genuine "doubling" but not one that
warrants nationwide fear. In the sarcastic words of L.A. Youth, a newspaper of
inner city teens: Exclusive . . . The Shocking Truth! Did You Know? Many
Young People Have Never Shot Anybody!

This study attempts to determine how many children kill another human
being in the course of a year. The study focuses on absolute numbers rather
than percentages, percentage changes, or rates per 100,000-abstractions that
leave many Americans with the impression that there are far more killers than
is actually the case. Informal surveys around dinner tables often reveal a belief
that the juvenile killers are numbered in the hundreds of thousands. 2 Many
people are surprised by our finding that approximately 940 children were
convicted of personally taking the life of another human being in the entire
nation in one full year.

Section II provides background on crime rates in America in order to create

the context for the counting that follows. Section III defines the key terms used
in Section IV, which counts how many children kill according to several
different methodologies. Finally, Section V outlines some suggestions on how
to reduce youth violence and homicide.

II. CRiME RATES

Overall crime rates in America have been stable or slightly declining for
most of the past twenty years. Victimization surveys reveal roughly the same
rate of robbery and aggravated assault in 1992 as they did in 1973, 3 and
burglary rates declined precipitously through the 1980s.' Homicide arrest rates
were the same in 1993 as they were in 1973.1 Overall victimization rates
seldom change more than a few percent each year, and the change is more often
downward than upward.

2. Readers are encouraged to conduct such surveys themselves. The author's experience
revealed responses ranging from 25,000 to 400,000.
3. BUREAU OF JusTIcE STATISTICS, U.S. DEP'T OF JUSTICE, CRIMINAL VICrIMIZATION INTHE
UNITED STATES: 1973-92 TRENDS (1994) [hereinafter CRIMINAL VICTIMIZATION].
4. Id.
5. FEDERAL BUREAU OF INVESTIGATION, U.S. DEP'T OF JUSTICE, CRIME IN THE UNITED
STATES--1993 (1994) [hereinafter CRIME-1993].

https://scholar.valpo.edu/vulr/vol31/iss2/5
 Lotke: Youth Homicide: Keeping Perspective on How Many Children Kill
1997] YOUTH HOMICIDE. KEEPING PERSPECTIVE 397

Crime Trends, 1973-1992

Baud on the Natiow Crun Vitmiaatzn Survey
45,00,00

! 40000000

35,000,000

Household

1.5,000,000 - - - - - --
Personal Theft -----------------------
10,000,000
Z ~Violenit

Source: FEDERAL BUREAU OF INVESTIGATION, U.S. DEP'T OF JUSTICE, CRIME IN THE UNITED
STATES-1995, at 218 tbl.38 (1996).

United States Murder Rate, 1970-1995

- .0 --------- ;-a------------------------------------------------------

.. ----
.. .. .. --- - -- *- .. . .. . . .. . .. . .----- ---....

0.0 ... ...

. a. ....
6.0 ---------------------------------------------------------
es0
,.......................

2.0-----------------------------------------------------------------

~
0.0 ~
", ",a" 0% , , 0, cm 4 0, M M !I'

Source: BUREAU OF JUSTICE STATISTICS, U.S. DEP'T OF JUSTICE, CRIMINAL VICTIMIZATION IN THE
UNITED STATES: 1973-92 TRENDS 9 (1994).

Produced by The Berkeley Electronic Press, 1997

 Valparaiso University Law Review, Vol. 31, No. 2 [1997], Art. 5

398 VALPARAISO UNIVERSITY LAW REVIEW [Vol. 31

The declining victimization rates may come as a surprise to Americans
accustomed to hearing that crime is on the rise. Part of the misconception stems
from relying on police records of arrests to show crime trends. 6 Police record
keeping has improved over the years, as staffing has increased and file keeping
has been computerized. Much of the supposed increase in crime is explained
by changes in methodology rather than actual changes in victimization rates.
For instance, between 1973 and 1992, police statistics showed a 120% increase
in the rate of aggravated assault.7 Direct surveys of the American population,
however, indicate that rates of aggravated assault declined 11% during that
period!

When it comes to juvenile crime, arrest trends have been relatively similar
to adult arrest trends in recent years. In 1982, juveniles comprised 18% of all
arrests; in 1995 they comprised 18.3% of all arrests. 9 From 1972 to 1995, the
percentage of overall index crimes-serious crimes such as murder, robbery and
rape-cleared by the arrest of a juvenile decreased from 27.3% to 22.1%.
In the area of property crime, juvenile clearances decreased significantly from
33.8% to 25.0%. For violent index crimes only there has been a slight increase
from 13.2% to 14.1%. Thus, trends in juvenile crime mirror the overall trend
of general stability and marginal declines.

Furthermore, the vast majority of juvenile crime involves non-violent

offenses, primarily relating to property or drugs. Only 6 out of 100 juvenile
arrests are for violent crimes (the same as adults)." Among the small number
of violent offenses, the majority are assaults-a very flexible crime category that
often involves mere threats or fights. Arrests for murder and rape constitute
less than one half of one percent of juvenile arrests."2

6. THE REAL WAR ON CRIME 4 (Steven Donziger et al. eds., 1996).

7. FEDERAL BUREAU OF INVESTIGATION, U.S. DEP'T OF JUSTICE, CRIME IN THE UNITED
STATES-1973 (1974); FEDERAL BUREAU OF INVESTIGATION, U.S. DEP'T OF JUSTICE, CRIME IN
THE UNITED STATES-1992 (1993) [hereinafter CRIME-1992].
8. CRIMINAL VICTIMIZATION, supra note 3.
9. CRIME-1992, supra note 7, at 429 tbl.4.5; FEDERAL BUREAU OF INVESTIGATION, U.S.
DEP'T OF JUSTICE, CRIME IN THE UNITED STATEs--1995, at 218 tbl.38 (1996) [hereinafter
CRIME- 1995].
10. All the clearance data in this section comes from the FBI. CRIME-1995, supra note 9, at
205 tbl.28; FEDERAL BUREAU OF INVESTIGATION, U.S. DEP'T OF JUSTICE, CRIME IN THE UNITED
STATES-1972, at 110 tbl.16 (1973).
11. CRIME-1995, supra note 9, at 218 tbl.38.
12. Id.

https://scholar.valpo.edu/vulr/vol31/iss2/5
 Lotke: Youth Homicide: Keeping Perspective on How Many Children Kill

1997] YOUTH HOMICIDE: KEEPING PERSPECTIVE 399

Arrests for Selected Offenses, 1995

12,000,000 11,416.346

10,000,000
S

8,=A00,0

S6A00,000

Z 40,000

2.8"n2
2,000,oo
1 ,3 64,334 "ISM 4,190 2,56
0. 4-
All Atnests ToWa Pmperty Azprvated Robbery Forcible Murder
Adult & Juvenile Crimes Aftault Juvenile Rapt Juvenile
Juvenle Anst Juvenile Juvenile Juvenile

Source:FEDERAL BUREAU OF INVESTIGATION, U.S. DEP'T OF JUSTICE, CRIME IN THE UNITED

STATES-1995, at 218 tbl.38 (1996).

Yet the overall crime trends mask specific trends within particular
demographic groups. When the focus is narrowed to juvenile homicide, the
picture shifts to genuine and shocking increases. Youth homicide arrest rates
have doubled just since the late 1980s, with the increases sweeping across racial
and ethnic lines. In the four short years between 1987 and 1991, the arrest rate
for homicide among white youth increased by 79%, and the rate among African
American youth increased by 121 %. These increases are most troubling in
the communities that already suffer from high rates of homicide and other
violent crime. In 1992, the victimization rate for homicide among African
American teenagers was nearly eight times the victimization rate among white
teenagers, and five times the victimization rate for the general population.'

13. ALFRED BLUMSTEIN, U.S. DEP'T OF JUSTICE, VIOLENCE BY YOUNG PEOPLE: WHY THE
DEADLY NEXUS? 4 fig.3 (1995).
14. NATIONAL CENTER FOR HEALTH STATISTICS, DEATH FOR SELECTED CAUSES, BY FIVE-
YEAR AGE GROUPS, COLOR AND SEX: UNITED STATES, 1979-1992 (1994); U.S. BUREAU OF THE
CENSUS, U.S. POPULATION ESTIMATES BY AGE, SEX, RACE AND HISPANIC ORIGIN: 1990 to 1993
(1994).

Produced by The Berkeley Electronic Press, 1997

 Valparaiso University Law Review, Vol. 31, No. 2 [1997], Art. 5

400 VALPARAISO UNIVERSITY LAW REVIEW [Vol. 31

Homicide Arrest Rate of 14-17 Year-Old Males

----------------------- ------------

%0 -- -----
---
so -A--------------------------------hite.----.

60 ------- ---
---------
2-...
-- --
-----

0
I

@R CF. V ~ ' O
31 ~ a, ~
a, 0
a, a,0'

Source: ALFRED BLUMSTEIN, U.S. DEP'T OF JUSTICE, VIOLENCE BY YOUNG PEOPLE: WHY THE
DEADLY NEXUs? 4 fig.3 (1995).

Most of the increase in juvenile homicide involved firearms. 5 Between

the 1970s and the mid 1980s, the rate of youth homicide was essentially stable,
and the weapons used in the offense were closely split between guns and other
weapons. In 1987 that started to change. The number of juvenile homicides
involving a firearm started to spiral upwards while the number of non-firearm
homicides held steady. Virtually all of the additional youth homicides since
1987 involved guns, so that in 1994 nearly 80% of the youth homicides were
committed with a firearm. Four times as many children were killed with a gun
in 1994 as in 1984.

15. OFFICE OF JUVENILE JUSTICE AND DELINQUENCY PREVENTION, U.S. DEP'T OF JUSTICE,
JUVENILE OFFENDERS AND VICTIMS: 1996 UPDATE ON VIOLENCE (1996) [hereinafter JUVENILE
OFFENDERS AND VICTIMS: 1996 UPDATE ON VIOLENCE].

https://scholar.valpo.edu/vulr/vol31/iss2/5
 Lotke: Youth Homicide: Keeping Perspective on How Many Children Kill

1997] YOUTH HOMICIDE: KEEPING PERSPECTIVE 401

Source: OFFICE OF JUVENILE JUSTICE AND DELINQUENCY PREVENTION, U.S. DEP'T OF JUSTICE,
JUVENILE OFFENDERS AND VICTIMS: 1996 UPDATE ON VIOLENCE 24 (1996).

Nonetheless, these rapid increases in youth homicide are highly site-specific

and do not present the overall threat to public safety that many people perceive.
Eighty-two percent of the counties in America experienced zero youth homicides
in 1994; 92% experienced zero or one. 6 Just four cities-Chicago, Los
Angeles, New York and Detroit-account for nearly one-third of the juvenile
homicide arrests nationwide, even though they account for only one-twentieth of
the country's juvenile population. 7 Even in these high homicide cities, the
rates of increase are large but the actual numbers are relatively small. Most
states experience just a handful of homicides by juveniles in the course of a
year; many states experience none at all, and large states like New York
experience just over one hundred. Finally, the increases may finally be coming
to an end: data for 1995 suggest that arrests of juveniles for homicide and other
violent crimes have started to decline.

16. HOWARD SNYDER, NATIONAL CENTER FOR JUVENILE JUSTICE, NATIONAL JUVENILE CRIME
TRENDS 1980-1994 (1996) (reporting data from an analysis of FBI Supplemental Homicide Reports).
17. Eric Lotke and Vincent Schiraldi, An Analysis of Juvenile Homicides: Where They Occur
and the Effectiveness of Adult Court Intervention, 12 J. Juv. JUST. & DETENTION SERVICES
(forthcoming 1997).

Produced by The Berkeley Electronic Press, 1997

 Valparaiso University Law Review, Vol. 31, No. 2 [1997], Art. 5

402 VALPARAISO UNIVERSITY LAW REVIEW [Vol. 31

III. DEFINITIONS

Criminal justice in America is not centralized; each state has its own
system, and the federal government has a system of its own. Different states
often define similar issues differently, and the inconsistencies can be confusing.
For example, some states define "juvenile" as a person under age eighteen,
other states define "juvenile" as a person under age sixteen, and still other states
define "juvenile" as a person under eighteen for some purposes and sixteen for
other purposes.

To foster consistency, this study uses the definitions of the Uniform Crime
Reports and the National Corrections Reporting Program. All state data are
conformed to these definitions. The important terms in this study are as
follows:

* Juvenile: a person under eighteen years of age (has not had an eighteenth
birthday). Words such as "children" or "youngsters" are occasionally used
for variety, but they all have the same meaning.

* Homicide: the deliberate killing of another human being, specifically,

murder or non-negligent manslaughter.

IV. How MANY CHILDREN KILL?

Nobody knows exactly how many children kill in the course of a year in
America. Estimates run from as low as 1000 to as high as 3000. To put the
matter in perspective, an average year in America sees a total of between 20,000
and 24,000 deaths by homicide.'" Thus, children appear to commit as little as
5% or as many as 15% of the annual homicides in America.

In this Section we attempt to determine how many children kill. The goal
is as modest as it is fundamental. It is fundamental because the size and location
of the problem shape the response, yet it is modest because with a problem this
important, counting the children is hardly an extravagant goal.

One problem in counting the children who kill is that many records are not
kept in a systematic fashion, and relevant data are often scarce or obsolete. For
this reason, an implicit recommendation underlying this entire report is that
record keeping be improved. It should not be necessary to guess about such a
fundamental matter in such an important area of public policy.

18. BUREAU OF JUSTICE STATISTICS, U.S. DEP'T OF JUSTICE, 1995 SOURCEBOOK OF CRIMINAL
JUSTICE STATISTICS 324 tbl.3.109 (Kathleen Maguire & Ann Pastore eds., 1996).

https://scholar.valpo.edu/vulr/vol31/iss2/5
 Lotke: Youth Homicide: Keeping Perspective on How Many Children Kill

1997] YOUTH HOMICIDE: KEEPING PERSPECTIVE 403

Another problem in counting is the omnipresence of plea bargaining. The

official offense of conviction does not always reflect the actual conduct. As
people move through the system, the charges often shift. A person may be
guilty of one offense, arrested for a related offense, and plead guilty to yet a
different offense. It is difficult to determine how the charges shift over time,
and whether the final charge is a less or more accurate reflection of the actual
conduct. For this reason, statistical records are imperfect instruments with
which to estimate how many children kill.

Fortunately, most plea bargaining occurs within categories: if somebody

is accused of a homicide and is genuinely involved in that homicide, the plea
bargain will likely be for some kind of a homicide offense; it will not shift
categories to, say, a drug sale or a burglary. Anybody who drops out of the
homicide category entirely or drops all the way down to negligent manslaughter
or aggravated assault likely had a peripheral involvement in a homicide, if any.

Another problem is that many offenders are never caught, making it

difficult to count them. Nonetheless, some techniques can estimate the number
of offenders who get away clean. The following Sections use a variety of
methodologies to count how many children are involved in a homicide. Each
Section uses the most recent data available for that methodology, though this
makes it difficult to compare across methodologies. Each methodology leads to
different results, but they cluster around the low thousands.

A. Arrests: 2560 Juveniles Arrestedfor a Homicide Offense in 1995

According to the FBI Uniform Crime Reports (UCR), which tabulate

arrests nationwide, 2560 juveniles were arrested for a homicide offense in
1995.9 Arrests are frequently used as a measure of crime by politicians and
the press. For example, a recent cover story in U.S. News and World Report
warns: "Teenage Time Bombs: Violent Juvenile Crime Is Soaring-and it Is
Going to Get Worse." The statistics behind the warning? "The number of
youths under 18 arrested for murder tripled between 1984 and 1994."20

Unfortunately, such headlines assume too tight an equivalence between

arrests and offenses. The number of arrests and other data provided by police
are so sensitive to police practice that they often measure police conduct better
than the underlying offense.

19. CRIME-1995, supra note 9, at 218. The figure is reported by 9498 agencies, covering an
estimated population of 196,440,000 for 1995.
20. Ted Gest & Victoria Pope, Teenage 7ime Bombs: Wolent Juvenile Crime Is Soaring-and
it Is Going to Get Worse. U.S. NEWS & WORLD REP., Mar. 25, 1996, at 32 (emphasis supplied).

Produced by The Berkeley Electronic Press, 1997

 Valparaiso University Law Review, Vol. 31, No. 2 [1997], Art. 5
404 VALPARAISO UNIVERSITY LAW REVIEW [Vol. 31
One problem is that police often arrest several people en route to
identifying a single perpetrator. Such duplicative arrests can cause a single
homicide to appear as several homicides in statistics based on arrests. Another
problem is that juveniles often act in groups. It only takes one person to pull
a trigger, but more people may be associated with the act: some in the car
alongside the triggerman; others who refuse to cooperate with a police
investigation. The police may arrest all these actors in their effort to identify
the triggerman and determine the various degrees of involvement and culpability.
This practice can make a single incident appear to be several incidents when, at
the end of the year, the police department simply reports a gross number of
homicide arrests. Similarly, an additional arrest is counted each time a person
is taken into custody, notified or cited to appear in court-even if multiple
citations occur for the same underlying incident. 2

Such problems can be aggravated in the context of serious crimes and high
levels of public concern, which often lead the police to intensify their
enforcement practices and increase the frequency of their arrests. If crime is
measured by arrest, the heightened enforcement will appear to be heightened
crime. The apparent increase in crime can lead to increased arrests, which may
lead in turn to an appearance of higher crime in a self-perpetuating upward
spiral.

Arrest rates can, of course, provide a crude measure of crime rates because
they often reflect a response to genuine criminal behavior. Arrest rates cannot,
however, provide too much detail. In the context of extremely fine questions
like the number of a single, exceedingly rare type of crime by people of a single
age group, the error introduced by arrest statistics may outweigh their accuracy.
An increase of a few hundred arrests (in a nation of 270 million people and
15,000 police departments) can create the appearance of a nationwide crime
wave.

A close reading of the UCR provides some insight into the problem. The
UCR reports 2560 arrests of juveniles for homicide in 1995.' It also reports
984 juvenile homicide offenses cleared-that is, referred to the court for
prosecution.' The difference of 1576 arrests could have dropped out for any

21. CRIME-1993, supra note 5, at 216.

22. CRIME-1995, supra note 9, at 218 tbl.38 (9498 agencies reporting; population
196,440,000).
23. Id. Eleven thousand one hundred eighty-one murder offenses were cleared by arrest; 8.8%
were under age 18. Id. at 205 tbl.28 (11,715 agencies reporting, population 210,149,000). cFor
UCR purposes, law enforcement agencies "clear or solve an offense when at least one person is
arrested, charged with the Commission of the Offense, and turned over to the court for prosecution."
Id. at 97.

https://scholar.valpo.edu/vulr/vol31/iss2/5
 Lotke: Youth Homicide: Keeping Perspective on How Many Children Kill

1997] YOUTH HOMICIDE: KEEPING PERSPECTIVE 405

number of reasons ranging from duplicative arrests, to absence of factual basis
to proceed, to failure to apprehend a suspect.24 The important point is that
arrests exceed clearances by the large gap of 160%.'

These problems are especially vexing in the context of children, who tend
to commit crimes in groups. Approximately 16.8% of homicide arrests are of
juveniles, but just 10.5% of homicide clearances are of juveniles. This disparity
suggests that more children than adults are filtered out of the system shortly
after arrest.

In sum, arrest data capture a great number of "false positives"-people who

appear to be offenders but in fact are not. After arrest, more careful factual
determinations are made. Not all arrests merit referral to the courts for
prosecution; of those referred, not all lead to convictions or adjudications. Such
a narrowing indicates that the system is functioning properly, with arrests being
made to protect public safety, and then some arrests being shunted out as the
case develops and more facts come to light. In the end, only a fraction of cases
are left. That is why reporting arrest rates as if they are crime rates can be
deceiving.

B. Convictions: Approximately 1330 Children Convicted of a Homicide

Offense in 1992

Ideally, one could determine how many children were known to kill another
person in the course of a year simply by looking up how many children were
convicted or adjudicated of a homicide offense. In the legal system, the closest
connection between actual conduct and legal status is the conviction: people are
convicted of offenses when they plead guilty to having done it, or when they are
proven to have done it beyond a reasonable doubt.26

Unfortunately, it is not so simple. Statistics on convictions are not kept in

any systematic, nationwide fashion. Many states do not keep such statistics at
all. California and Illinois, for example, are completely unable to provide
information on convictions. The absence of data from these and other large
states makes it impossible to tabulate the number of homicides nationwide by

24. Id. at 205 tbl.28. The 984 cleared offenses include only those offenses where no adults
were involved, which the UCR says will result in a "slight underestimation." Id. at 197. The
Office of Juvenile Justice and Delinquency Prevention (OJJDP), however, contends that 30% of the
juvenile homicides included an adult accomplice; adjusting the UCR clearance figure by OJJDP's
estimate of adult involvement brings it up to 1279, still far fewer than the 2560 arrests. JUVENILE
OFFENDERS AND VICTIMS: 1996 UPDATE ON VIOLENCE, supra note 15, at 25.
25. CRIME-1995, supra note 9, at 205, 218.
26. Distortions relating to plea bargains are particularly relevant in the context of convictions.

Produced by The Berkeley Electronic Press, 1997

 Valparaiso University Law Review, Vol. 31, No. 2 [1997], Art. 5

406 VALPARAISO UNIVERSITY LAW REVIEW [Vol. 31

direct survey of each state. Even in states where conviction statistics are
available, differences in the definition of key terms like "juvenile" and
"homicide" make it impossible to combine them into a single figure.

Nevertheless, enough states are able to provide conviction data to provide

a feel for the issue. Using 1994 as a baseline year, for example, Delaware and
Vermont definitively reported that zero juveniles were convicted of homicide;
New Hampshire reported one juvenile convicted of homicide; and Rhode Island
reported two juveniles convicted of homicide. New York, a state with a
substantial amount of youth crime, reported 134 juvenile homicide convictions
in 1994. Pennsylvania, another sizable state, reported 44 juvenile homicide
convictions in that year.27

A surrogate for convictions data can be found in data for admissions to

locked facilities, which are kept in a systematic fashion nationwide. 8
Admissions are a reasonable surrogate for convictions because 90% of the
children convicted of homicide offenses are admitted to locked facilities; just
10% are released to probation or psychological treatment.29 Admissions data
are available for twenty-four states comprising 66.5% of the U.S. population
under the age of eighteen. It is possible to extrapolate outward. from this data
pool to estimate the number of juveniles convicted of a homicide offense in the
course of a year.' This method leads to the conclusion that 1330 individuals
under age eighteen were convicted of a homicide offense in 1992.

This methodology is novel so additional detail will be provided. The

admissions data reveal that 567 juveniles were admitted to juvenile facilities and
336 juveniles were admitted to adult prisons for homicide offenses in 1992.
That is a total of 903 juvenile admissions for homicide in 1992. Some children

27. These data were provided by the criminal justice data authorities of each state.
28. NATIONAL CORRECTIONS REPORTING PROGRAM, U.S. DEP'T OF JUSTICE, JUVENILES
TAKEN INTO CUSTODY RESEARCH PROGRAM: STATE JUVENILE CORRECTIONS SYSTEM REPORTING
PROGRAM (1992). The states are: California, Georgia, Illinois, Iowa, Kentucky, Louisiana,
Maryland, Massachusetts, Minnesota, Mississippi, Missouri, Montana, Nebraska, New Jersey, New
York, North Dakota, Ohio, Oklahoma, Oregon, South Carolina, Tennessee, Texas, Utah, Virginia,
and Wisconsin. Id. The most recent data is from calendar year 1992. Id.
29. OFFICE OF JUVENILE JUSTICE & DELINQUENCY PREVENTION, U.S. DEP'T OF JUSTICE,
DETAILED SUPPLEMENT TO JUVENILE COURT STATISTICS 1993, at 70 (1995). In 1993, there were
929 homicide cases in juvenile courts. Id. Ninety-three percent of those cases were petitioned, 44%
of the total cases were adjudicated and 33% were transferred. Id. Seventy-five percent of the
adjudicated cases led to placement in a secure facility; presumably nearly all of the transferred cases
led to a secure placement. Id. Applying these percentages, 641 cases out of 716 transferred or
adjudicated led to placement-a placement rate of 90%. Id. (data from 17 states representing 27 %
of the U.S. youth population at risk).
30. The National Council on Crime and Delinquency in San Francisco was particularly valuable
in providing admissions data and helping to analyze it.

https://scholar.valpo.edu/vulr/vol31/iss2/5
 Lotke: Youth Homicide: Keeping Perspective on How Many Children Kill

1997] YOUTH HOMICIDE: KEEPING PERSPECTIVE 407

admitted to prison, however, were admitted from juvenile facilities, so the total
of 903 admissions actually counts a small number of children twice. Data
pertaining to release from locked facilities reveals that 13 % of the 315 children
released from juvenile corrections systems for homicide offenses in 1992 were
certified as adults and transferred to prison. For this reason, the 903 admissions
ake reduced by forty-one people (13% of 315) to yield a revised total of 862
individual juveniles admitted to prisons or juvenile facilities for homicide
offenses in 1992.

That total, however, covers only 66.5% of the U.S. juvenile population.
Extrapolating3 to cover the nation as a whole indicates that 1197 children were
admitted to locked facilities for homicide offenses in 1992.32 The final step is
to adjust for the 10% of the children who were convicted of a homicide but not
committed to confinement. This adjustment leads to the conclusion that 1330
children were convicted of a homicide in 1992.

C. The FBI Supplementary Homicide Reports: 2947 Juvenile Homicide

Offenders in 1994

So far we have only counted offenses that were formally entered into the
system. Neither methods based on arrests nor methods based on convictions can
count offenders who got away clean. Yet, a different means of measuring
juvenile homicide can count those unknown, unarrested perpetrators. This
method, based on the FBI Supplementary Homicide Reports (SHR) is the basis
of the best current research. Supplementary Homicide Reports are filled out
whenever a person is victimized by a homicide. The police at the scene
interview witnesses in an attempt to determine what happened. They then fill
out the SHR form, answering questions like the age and race of both victim and
offender, and the circumstances of the death.

31. The fact that the data set includes less than half the states (24 out of 50) but more than half
the juvenile population (66.5%) indicates that the included states are atypically large. As it turns
out, they also have atypically high homicide rates: the average homicide rate (1993) among the
included states was 10.13 homicides per 100,000 people; the average among the excluded states was
7.82 per 100,000 people. To calculate the total number of juvenile homicide offenders admitted into
institutions for all states, we first multiplied the number of offenders in the 24 included states (862)
by the percentage of excluded states based on population size (33.5%) resulting in 434 additional
offenders. We then multiplied that number by the ratio of the average homicide rate for the included
versus excluded states (.77) which yields an additional 335 offenders. The resulting total of juvenile
homicide offenders admitted into institutions is 1197.
32. There is often a time delay between the commission of an offense and the admission to a
locked facility. Thus, many of the people admitted in 1992 actually committed their offenses in
1991. The delay may be insignificant, however, because the estimate still captures the number of
people who committed the offense in one year, even if that year actually starts sometime in 1991
and ends in the middle of 1992. Twelve months of admissions reflects 12 months of offenses, even
if they are not exactly the same 12 months.

Produced by The Berkeley Electronic Press, 1997

 Valparaiso University Law Review, Vol. 31, No. 2 [1997], Art. 5

408 VALPARAISO UNIVERSITY LAW REVIEW [Vol. 31

Using these forms, the Federal Office of Juvenile Justice and Delinquency
Prevention (OJJDP) counted 2947 juvenile homicide offenders in 1994, nearly
three times the number in the mid-1980s.33 This finding made the headlines
in the Los Angeles Times, the Washington Post, and other newspapers across the
country.34

One advantage of the SHR is its independence of other law enforcement

activity. Even if no arrest is ever made, the SHR will capture the fact that the
police reported to a scene with a person dead in circumstances indicating
homicide. Thus, the SHR counts homicides even if the offender is never
caught. Another advantage is that the SHR provides detail beyond mere
counting. Using these forms, researchers can detect whether the offender and
victim were friends or strangers, near or distant in age, and whether it was a
killing over love or money.

Yet the strengths of the SHR are also its weaknesses. The information that
goes into the forms is largely untested. There is never a hearing to determine
its accuracy; it is never subject to cross-examination; it is never compared to
other facts to find inconsistencies. Just because a witness says the offender was
sixteen years old does not make it so; it may be true, but it also may not. As
the case proceeds, information may come to light that makes the SHR forms
obsolete.

An indication of trouble is the fact that the SHR counted 2947 "known
juvenile homicide offenders" in 1994, but there were only 1283 juvenile
homicide clearances. This gap suggests that less than half of the "known"
offenders counted in the SHR led to a prosecution. No doubt many offenders
got away clean, but probably not that many. It is plausible to assume that many
of the people identified as killers by the SHR were likely filtered out by the
justice process.

33. This figure comes from the statistical software reporting the homicide reports available in
HOWARD N. SNYDER & TERRENCE FINNEGAN, NATIONAL CTR. ON JUVENILE JUSTICE, EASY
ACCESS TO THE FBI's SUPPLEMENTARY HOMICIDE REPORTS: 1980-1994 (1996). The published
figure is 2800. JUVENILE OFFENDERS AND VICTIMS: 1996 UPDATE ON VIOLENCE, supra note 15,
at 22. The difference in the two figures islikely that the statistical software imputes data for
agencies that failed to report and is therefore more complete.
34. The L.A. Times got the zeros wrong; it reported 2600 homicides as 26,000 homicides.
Ronald J.Ostrow, Number of Juvenile MurderersIs Soaring, L.A. TIMES, Mar. 8, 1996, at Al.
This error may signify larger problems of illiteracy around these issues, as people writing and
editing on most matters of public importance generally will not make errors of a factor of ten in the
lead sentence.

https://scholar.valpo.edu/vulr/vol31/iss2/5
 Lotke: Youth Homicide: Keeping Perspective on How Many Children Kill
1997] YOUTH HOMICIDE: KEEPING PERSPECTIVE 409
D. Victimization: Approximately 1942 Children Were Killed by Someone
Their Own Age

The final method of counting is entirely different and, unfortunately, not

strictly comparable because it counts children under nineteen years old instead
of eighteen years old. It will, however, provide useful confirmation of previous
methods because it leads to a result on the same order of magnitude.

According to the National Centers on Disease Control and Prevention, 3532

people under nineteen years of age died of a homicide in 1994. 31 This is a
very different methodology because it focuses on victims rather than offenders.
However, we know from the SHR that in 1994, 55% of the victims under
eighteen years of age were killed by somebody within their age group. If the
SHR ratio holds true, it suggests that 1942 of the young homicide victims were
killed by somebody of their own age.

E. Of Look-Outs and Trigger-Pullers, Disputes and Drive-Bys: How Many

Children Commit What lype of Homicide?

Counting the youth involved in a homicide is just a beginning. Also

relevant is the level of involvement and the type of homicide. Was the youth
the look-out or the trigger-puller? Did the child kill an abusive parent or a rival
drug dealer? This Section investigates such questions.

The number of actual killers is bound to be smaller than the number of

children involved in homicides because several people are often involved in a
single offense. Although there is seldom a legal or statistical difference between
the person who pulled the trigger and the sidekick, many people find an ethical
distinction. The question always remains: would the sidekick have pulled the
trigger? Indeed, might the fact that the sidekick did not pull the trigger, and
may not even have had a gun, indicate that the sidekick acting alone would not
have killed the victim? These questions are unanswerable, but it seems
needlessly clumsy to attribute to the accessory the same ethical qualities as the
principal. The ultimate question is how many children are personally and
individually responsible for taking the life of another human being.

The best source of data for determining individual responsibility is the

SHR. Its detailed breakdown reveals that just half of the juvenile offenders
acted alone; a quarter acted as one of a pair; and the last quarter acted as part

35. Data from the CDC can be accessed through its site on the worldwide web:
< http://www.cdc.gov/ncipc/usp/usmort.hml >.

Produced by The Berkeley Electronic Press, 1997

 Valparaiso University Law Review, Vol. 31, No. 2 [1997], Art. 5
410 VALPARAISO UNIVERSITY LAW REVIEW [Vol. 31

of a group of three or more.36 If those ratios are applied to the number of

children convicted of a homicide offense, it leads to the conclusion that
approximately 940 youngsters personally and individually took the life of another
human being nationwide in 1992.

Furthermore, 10% of all youthful homicide offenders killed their parents,

often in the context of abusive relationships.37 While these are not justifiable
homicides, they are also quite different from the random or gang-related killings
so many Americans fear. Another half of the children killed acquaintances.3"
Sometimes the acquaintances were abusive parent-equivalents, like the mother's
boyfriend; sometimes they were rival drug dealers-the statistics do not say.
But as terrible as it is to kill a family member or acquaintance, these offenses
too differ from those often portrayed in the evening news. Acquaintance killings
are not random, motiveless or unfathomable. All too often they begin as petty
disputes over trivial issues between hot-headed teens; with firearms present, the
dispute sometimes escalates into a murder. Applying the SHR ratios to
convictions suggests that 535 children were known to be principally responsible
for killing somebody they knew nationwide in 1992.

So who are the cold blooded predators of the evening news? How many
children spray gunfire into crowds or lie in wait to ambush unsuspecting
pedestrians? The answer is twofold: more than a country would wish, but not
as many as might appear from the mass media. Subtracting the 12% of the
offenses about which nothing is known, 31% of the juvenile homicides were
committed against a stranger.39 That means in the entire nation in 1992, a total
of 410 children were convicted of such an offense; of them, 290 children
personally committed the crime.

One final point. The media often focus on terrible crimes committed by
very young children. Politicians often use these offenses as arguments for
building new juvenile jails or increasing the punitive nature of the juvenile
justice system. To provide some background for these claims, the number of
very young killers was calculated. The conclusion was that nationwide,
including both family offenses and stranger offenses, approximately eighty
children under age fourteen were convicted of a homicide offense for which they

36. The analysis of the SHR was done by OJJDP. JUVENILE OFFENDERS AND VICTIMS: 1996
UPDATE ON VIOLENCE, supra note 15, at 25.
37. Id. at 23.
38. Id.
39. Id.

https://scholar.valpo.edu/vulr/vol31/iss2/5
 Lotke: Youth Homicide: Keeping Perspective on How Many Children Kill
1997] YOUTH HOMICIDE: KEEPING PERSPECTIVE 411
were principally responsible in 1992.'

The reason to dissect the figures in this fashion is simply to provide

perspective and to demonstrate that in this nation of nearly 270 million people,
the number of juvenile killers is in the hundreds. Youth homicide may not be
of the magnitude many people believe. Implicit in this discussion is one final
thought: it should not be necessary to make such estimates. With all the
attention paid to juvenile homicide, with all the headlines and all the speeches,
it should not be necessary to guess how many children are being discussed.

Number of Convicted Youth Homicide Offenders

Total convicted offenders 1330

(accessories and principals)

Principally responsible for a 940

homicide

Principally responsible for a 290

homicide of a stranger

Principally responsible for a 80

homicide and less than fourteen
years of age

V. How TO REDUCE JUVENILE HOMICIDE

People are easily overwhelmed by the problem of youth homicide.

Hands fly into the air, cries that "nothing works" echo throughout the room.
In the face of this apparently overwhelming problem, one solution emerges:
lock them up, lock them all up, it's the only way to keep us safe.

This Article attempts to measure the problem of juvenile homicide in

hopes of eliciting a more measured response. The focus is on absolute
numbers, rather than percentages, to provide perspective on the scale of the

40. The data for admissions to locked facilities reveal that 74 children under 14 years of age
were admitted for a homicide offense (66.5% of the juvenile population represented) in 1992.
Extrapolating as above, the total number of identified homicide offenders under 14 is estimated to
be 103. Assuming again that roughly half of the juvenile offenders acted alone, a quarter acted as
one of a pair, and the last quarter acted as part of a group of three or more, it follows that
approximately 80 children under age 14 were known to be principally responsible for a homicide.
It is necessary to be careful, however, about the large number of estimates used in this calculation.

Produced by The Berkeley Electronic Press, 1997

 Valparaiso University Law Review, Vol. 31, No. 2 [1997], Art. 5
412 VALPARAISO UNIVERSITYLAW REVIEW [Vol. 31

problem and to make it easier to create policies that will assist the nation's
youth and maximize public safety.

One item that follows directly from the counting is the irrelevance of
large scale transfer of juveniles to adult court. Frustrated with the problem
of youth violence, many legislators have proposed to increase the transfer of
juvenile offenders to the adult criminal justice system. The theory is that an
increasing number of children are simply too vicious to be handled in juvenile
court, with its focus on rehabilitation, and they must be sent to adult court,
with its focus on punishment. In 1995 and early 1996, half the states
considered legislation to increase transfer. Legislation under consideration at
the federal level increased federal transfer and actually required states to
transfer children as young as fourteen years of age in order to qualify for
federal funds.

Such activity is irrelevant for two reasons. First, all fifty states and the
District of Columbia currently have the authority to process violent juvenile
offenders into adult court;4 in fact, between 1989 and 1993, the number of
juvenile offenders transferred increased by 41%, although this activity reduced
neither fear nor juvenile homicide. Indeed, an analysis of juvenile homicide
rates and transfer rates found no apparent correlation between transfer rates
and homicide rates: states with high juvenile homicide rates sometimes have
high transfer rates, sometimes not, and states with low juvenile homicide rates
sometimes have low transfer rates, sometimes not.42

Second, the number of children transferred is already large compared to

the number of juveniles who have been apprehended and convicted of a
homicide offense. More juveniles are transferred for nonviolent offenses than
for violent offenses.43 The majority of juvenile transfers involved property
or drugs; 10% of the transfers involved offenses against public order like
disorderly conduct. These data suggest that focusing transfer more narrowly
would work better than increasing the use of transfer.

This is especially true because widespread transfer of children to the

adult correctional system may make problems worse. A recent study in
Florida found that children transferred to adult court reoffended approximately
30% more frequently than matched children who stayed in the juvenile

41. OFFICE OF JUVENILE JUSTICE AND DELINQUENCY PREVENTION, U.S. DEP'T OF JUSTICE,
JUVENILE OFFENDERS AND VICTIMs: A NATIONAL REPORT (1995).
42. Lotke & Schiraldi, supra note 17.
43. JUVENILE OFFENDERS AND VICTIMS: 1996 UPDATE ON VIOLENCE, supra note 15.

https://scholar.valpo.edu/vulr/vol31/iss2/5
 Lotke: Youth Homicide: Keeping Perspective on How Many Children Kill

1997] YOUTH HOMICIDE: KEEPING PERSPECTIVE 413

system." The study was conducted with youth who were matched for
seriousness of the transfer offense, number of charges, number of prior
offenses, severity of prior offenses, and sociodemographic characteristics.
This finding reflects the harsh and debilitating conditions present in adult
correctional facilities, and the absence of attention paid to teaching children
from their mistakes.

If transfer is not the answer, what is? The rest of this Section is devoted
to exploring the path to make communities safe. The key is to focus on the
scale of the problems. While they are serious, they are not insurmountable.
For example, the city of Miami, often associated with terrible problems of
youth violence, arrested no more than thirty-six children for homicide offenses
in 1994. It is true that some kids got away clean, and it is also true that some
kids shot and wounded without killing, shot and missed, or brandished but did
not shoot. It is also true that this behavior is virtually as problematic as a
completed homicide. Even so, the scale is manageable. If the arrest figure
undercounts by a factor of five then the city is dealing with 180 kids; if it
undercounts by a factor of ten then there are 360 kids. Surely the city of
Miami can find creative ways to manage a few hundred kids. With all that
in mind, this Section highlights some promising approaches and promising
programs.

A. Prevention

The best path to a safe society is to prevent crime before it happens.

Delinquent juveniles do not spring spontaneously, armed and dangerous, onto
the streets. They grow slowly from broken families, disorganized neighbor-
hoods, malfunctioning schools, and unsupervised peer groups. This is not to
say that all children who grow up in such circumstances will become
murderers: most will not. But those who kill typically start moving in the
wrong direction at an early age. It follows that early efforts to point them in
the right direction will reduce violence as the kids grow up.

The landmark High/Scope Perry Preschool study, which tracked high risk
minority youth over a period of twenty-seven years, found that early
interventions more than paid for themselves in reduced crime and social costs.
Those who attended a high quality preschool program performed better in
every category than their non-program counterparts: they had half as many
criminal arrests, higher earnings and property wealth, and greater commitment

44. Donna Bishop et al., The Transfer of Juveniles to Criminal Court: Does It Make a
Difference?, 42 CRIME & DELINQ. 171 (1996).

Produced by The Berkeley Electronic Press, 1997

 Valparaiso University Law Review, Vol. 31, No. 2 [1997], Art. 5

414 VALPARAISO UNIVERSITY LAW REVIEW [Vol. 31

to marriage.45 This study suggests that expanding Head Start or similar
preschool programs will reduce violent crime as the children become
teenagers.

Perry Preschool Study Group Differences in Official Crime

Criminal Involvement Preschool No Preschool

Percentage arrested as 31 51
juveniles or adults

Percentage arrested as 16 25
juveniles

Percentage arrested as 25 40
adults

Percentage convicted 16 21
as adults

Similar findings are available for programs designed for older children.
For example, the federal Children-At-Risk (CAR) program targets high risk
adolescents who live in distressed neighborhoods, delivering services
involving collaboration between police, schools, social workers and other
service providers. Preliminary results show that youths in the CAR program
had fewer contacts with police and courts than a control group, and had
higher rates of high school attendance and promotion to the next grade level.
In three of four cities studied, there was evidence of greater declines in crime
in CAR neighborhoods." These results suggest that skillful intervention by
all those who play a part in the life of a troubled child-from family to
school teachers and police--.can help steer that child away from criminal
involvement.

Promising results are also found in high school conflict resolution or peer
mediation programs. For example, the SCORE program in Boston and the
Resolving Conflicts Creatively program in New York City teach adolescents
how to resolve differences without violence. Kids in the classes play roles
and answer questions designed to lead them to peaceful solutions and better
management of aggression.

45. LAWRENCE J. SCHWEINHART Er AL., SIGNIFICANT BENEFITS: THE HIGHJSCOPE PERRY

PRESCHOOL STUDY THROUGH AGE 222-25 (1993).
46. ADELE HARRELL, NATIONAL INST. OF JUSTICE AND URBAN INST., INTERVENING WITH
HIGH-RSK YOUTH: PRELIMINARY FINDINGS FROM THE CHILDREN-AT-RISK PROGRAM (1996).

https://scholar.valpo.edu/vulr/vol31/iss2/5
 Lotke: Youth Homicide: Keeping Perspective on How Many Children Kill

1997] YOUTH HOMICIDE: KEEPING PERSPECTIVE 415

After-school recreation is also helpful, especially for kids who lack
supportive families and positive entertainment options. Sports and recreation
provide healthy outlets for kids whose time and energy might otherwise go
to hanging out, picking fights, breaking windows, and maybe someday
shooting a gun. When Midnight Basketball first started in 1986 in Glenarden,
Maryland, drug-related crimes were cut in half. No other crime-fighting plan
has achieved such astonishing results-not prison, not police, not chain gangs.
The success indicates that similar, pro-social programs would go far to
reducing violent crime.

Furthermore, the experience of Barrios United, House of Umoja, and

other such groups demonstrate that the best organizations are those that form
locally around the efforts of private citizens. Few children wish to live in
fear, even those who carry weapons and brag about it. Private groups formed
in the neighborhoods have the best chance of youth changing behavior, and
public support can help them achieve their goals.

A further dimension involves the children's parents. Many of the

risk-factors that describe children who are likely to kill involve their parents:
parental drug use, parental unemployment, parental neglect and abuse.
Interventions such as drug treatment and job training, as well as a genuine
effort to create jobs in every sector of the economy, will help to reduce youth
crime and violence.

B. Law Enforcement

Optimism about prevention programs need not displace reliance on

traditional law enforcement. Skillful, vigorous enforcement will always be
necessary, and the pattern of youthful offending reveals why.

David Kennedy, studying firearms violence in Boston, Massachusetts,

found that most children who kill have a history of disruptive or dangerous
behavior long before they pull a trigger.47 Before the 125 known murderers
in his study committed their homicide, 77% (ninety-six) had been arraigned
in state courts; 26% (thirty-three) had been locked up; and 54% (sixty-eight)
had been sentenced to probation. Police and probation officers said they
knew all the kids who had killed or were killed by gun violence; the dark
joke among the officers was whether it would be ethical to take out life
insurance on certain individuals. The point, of course, is not to buy life
insurance, but to use the information to design enforcement strategies that

47. David Kennedy et al., Youth Violence in Boston: Gun Markets, Serious Youth Offenders,
and a Use-Reduction Strategy, 59 LAW & CONTEMP. PROBs. 147 (1996).

Produced by The Berkeley Electronic Press, 1997

 Valparaiso University Law Review, Vol. 31, No. 2 [1997], Art. 5

416 VALPARAISO UNIVERSITY LAW REVIEW [Vol. 31

will make insurance unnecessary. It is possible to intervene earlier in the
lives of the kids deemed most likely to commit serious violence.

Working in a coalition including members of the Boston Police

Department, the United States Attorney for the Commonwealth of
Massachusetts, the Federal Bureau of Alcohol, Tobacco and Firearms, the
Suffolk County District Attorney, the Massachusetts Departments of Probation
and Parole, and more, Kennedy helped to develop a plan to reduce firearm
violence. The goal was to keep guns out of the hands of kids, disrupt illegal
gun markets, and reduce fighting among street gangs. Children identified by
officers as problematic were subject to heightened scrutiny and control, and
illicit firearms were tracked to the source to determine how to interrupt
supplies. The work was so successful that it is being replicated in cities
nationwide.

In other cities, community policing has helped to bring crime down and
make communities feel more secure. The crucial idea is to gather quality
information and selectively invoke the apparatus of justice where it will do
the most good. Teenagers must be swiftly and certainly punished for petty
crimes, and probation officers must ensure that no further crimes are
committed while steering kids towards lawful alternatives. The quality of
enforcement is generally more important than the severity of sentence.

C. Punishment

Most people who kill do it only once. From the perspective of public
safety, the question is not how to punish people who commit homicide, but
how to punish lesser offenses so the perpetrators will not go on to commit a
homicide. The answer appears to lie in the creation of a wide variety of
structured programs that punish people for transgressions while pointing them
towards a law-abiding future.

A successful system of graduated sanctions consists of: (1) immediate

sanctions within the community, such as community service or diversion, for
beginning or non-violent offenses; (2) intermediate sanctions within the
community, such as intensive supervision or wilderness camps, for more
serious offenders; (3) secure corrections for the small number of truly serious
or violent offenders; and (4) structured after-care programs for all youth who
have been in contact with the justice system and received any kind of
sanction.

Experience with such structured punishments has been generally positive.

For example, Mark Lipsey's comprehensive meta-analysis of 443 separate
delinquency studies found that behavior-oriented, skills-oriented, and

https://scholar.valpo.edu/vulr/vol31/iss2/5
 Lotke: Youth Homicide: Keeping Perspective on How Many Children Kill

1997] YOUTH HOMICIDE: KEEPING PERSPECTIVE 417

multi-modal treatment methods produced reliably positive results. The

states of Massachusetts and Utah, which have experimented with community-
based corrections rather than large secure facilities, have found declines in the
frequency and severity of offending after correctional intervention. 4' Public
safety is best served by steering adolescents who have drifted from the lawful
path back towards the law with a structured series of graduated sanctions.

D. The Media

The press has no formal role in the formation of justice policy, yet it
influences crime policy as powerfully as the official policymakers. People
bombarded with a steady stream of horrible crime news are more apt to feel
afraid, choose punitive justice policies and think that crime is on the rise
when it is not. Different reporting practices may make it easier to craft sound
justice policy."0

Government censorship is not, of course, the answer. The press may,

however, wish to follow routes like KVUE news television in Austin, Texas.
KVUE pioneered guidelines to help it choose what crime news is truly
newsworthy. KVUE still reports crime, but it has raised its standards on why
to report a crime and how to present it. Is there an immediate threat to public
safety? Should citizens take action? Does it have a significant community
impact? These criteria have the effect of screening out many crimes and
preserving limited news space for important crimes or other civic events. The
public has responded enthusiastically to this development; it seems to have
had enough of "police blotter" news."' If more press outlets became more
selective in their reporting of crime news-holding crime to the same
standards as other topics-it might diminish the sense of panic that leads to
irrational and sometimes harmful policies.

VI. CONCLUSION

There is every reason to believe that a combination of graduated

sanctions, skillful policing, and prevention programs can reduce the incidence

48. OFFICE OF JUVENILE JUSTICE AND DELINQUENCY PREVENTION, U.S. DEP'T OF JUSTICE,
GUIDE FOR IMPLEMENTING THE COMPREHENSIVE STRATEGY FOR SERIOUS, VIOLENT AND CHRONIC
JUVENILE OFFENDERS 138 (1995).
49. Id. at 136. See also JEROME MILLER, LAST ONE OVER THE WALL (1991).
50. THE REAL WAR ON CRIME 69-98 (Steven Donziger et al. eds., 1996).
51. Joe Holley, Should the Coverage Fit the Crime? A Texas TV Station Tries to Resist the
Allure ofMayem, COLUM. JOURNALISM REV., May-June 1996.

Produced by The Berkeley Electronic Press, 1997

 Valparaiso University Law Review, Vol. 31, No. 2 [1997], Art. 5

418 VALPARAISO UNIVERSITY LAW REVIEW [Vol. 31

of youth homicide in America.52 The problems seen on the evening news
or referenced by politicians on tour may appear unmanageable-but analysis
reveals that the number of children involved is not so large and the solutions
are not so far off that hope must be abandoned. With a little creative energy,
this nation can help its children to navigate the difficult path through
adolescence in this turbulent and troubled time.

52. The federal government has provided an excellent guide to youth crime control: OJJDP's
Guide for Implementing the Comprehensive Strategy for Serious, Violent and Chronic Juvenile
Offenders. See supra note 48. Those who are serious about youth crime control may benefit from
studying this plan.

https://scholar.valpo.edu/vulr/vol31/iss2/5