Murder By Strangers

Faced with police department's in some of the nations largest cities (New York, Chicago, Houston, and Los Angeles) reporting lower violent crime rates for the first time in decades the FBI reacted quickly to (1) attribute the lower crime rates to increased numbers of police officers and longer prison sentences and (2) counterbalance the good news with data designed to sustain the fear of crime: namely that (1) people were in more danger than ever being victimized by strangers and (2) demographic changes in the most criminogenic population foretold a crime wave in the near future.

The FBI sent a news release to media outlets across the country in 1994 claiming that for the first time murders were more often committed by strangers than acquaintances and that the percentage of murders committed by non-family members had increased:

"Historical statistics on relationships of victims to offenders showed that the majority of murder victims knew their killers. However, in the last few years (1991 through 1994) the relationship percentages have changed. In 1994, less than half of murder victims were related to (12 percent) or acquainted with (35 percent) their assailants. Thirteen percent of the victims were murdered by strangers, while the relationships among victims and offenders were unknown for 40 percent of the murders." (Federal Bureau of Investigation, 1995, p.17)

The Washington Post, along with newspapers across the country, reported in a front page article, _The New Face of Murder in America: Family Slayings Decline; Fewer Cases Solved; Killers Are Younger" that the _number of people unknown persons has grown in the 1990s." (Thomas, 1995, p. A1)

The FBI news release and the medias knee-jerk reproduction of the findings is a classic case of law enforcement propaganda masquerading as fact. The increase in "murders by strangers" is a statistical artifact accounted for in part by an increase in "unsolved murders" Between 1991 and 1994 the number of murders cleared by arrest dropped by more than five percent compared to the preceding ten year period (See Table I). Table 1

Murder and Non-Negligent Manslaughter

Incidents Cleared By Arrest: 1981-1985 Many of the unsolved murders are "drive by shootings." Given that drive by shootings often are brought about by turf battles between competing gangs selling drugs, the chances are very good that the assailant was well known to the victim. Well known enough to be killed for competing or "snitching" on those who shot him.

The FBI news release which was also reported by newspapers including the Washington Post goes on to state that _In 1965, nearly a third of the murders in this country were family related...{but by] 1992, a little more than one out of 10 of the nation_s homicides were family related._ (Thomas, 1995, A4). The Post quotes Gilford S. Gee, a contributor to the UCR, who noted that: _Criminologists and sociologists used to point to the fact that most murders were committed by family members or acquaintances . . .

That was indeed the case, but no longer._ (Thomas, 1995, A4).

The Post accurately reported the data they received from the Justice Department. What makes the conclusion that fewer murders are the result of family violence questionable is that the percentage of people who live together as married couples with and without children has increased substantially in recent years. The Census Bureau estimates that the number of unmarried couples grew from 52,000 in 1970 to over 360,000 in 1994 (Bureau of the Census, 1995).

Furthermore, the proportion of unmarried couples living together is highest in the poorest social classes, precisely where the greatest likelihood of murder among family members takes place.

Thus the FBI claim that family and acquaintance murders no longer constitute the majority of homicides is erroneous. FBI data show that 47 percent of all murders were between relatives (12%) or acquaintances (35 percent). If the number of acquaintances in the unsolved murders were combined with the number of murders in the acquaintance and relative category it is likely that the "criminologists and sociologists" observation that most murders are between "family members or acquaintances" is as true today as it ever was.

Selective Reporting

Selecting years for comparison that show increases in crime is common practice in Department of Justice reports. In the 1994 Uniform Crime Reports, for example, the FBI presented a comparison of homicide rates between 1991 and 1994 which showed an increase in "stranger" homicides for those years. In fact, a comparison of 1991 and 1994 was about the only comparison that would have shown the increase in stranger homicides the FBI sought.

In 1980 the FBI reported that in 36% of homicides the assailant was unidentified. The rate drops to 27% in 1985 and rises again in 1988 to 33%. In fact, the 1994 rate is lower than the 1980 rate (Federal Bureau of Investigation, 1981, 1986, 1989, 1995).

In the FBI news releases there is no mention of the fact that the category of substantiated stranger homicides: that is, the number of homicides where it was determined that the assailant was a stranger, has remained fairly constant. The figure stands at 13% in 1994, the same as 1980. The number rose to 15% in 1985 before falling to 12% in 1988. Since these data will not serve to fan public paranoia about crime, the FBI prefers to draw faulty conclusions about the nature of unknown murder assailants.

Teen-age "Super Predators"

In addition to raising a false alarm about a dramatic increase in "stranger murders," the FBI and local law enforcement agencies periodically point to allegedly dramatic increases in the number of crimes committed by juveniles. Citing FBI sources, U.S. News and World Report published a warning in 1967 that the nation was experiencing an "explosion in Teen-age crimes:_

Deep worry is developing among the nation_s leaders over juvenile delinquency that seems to be getting out of hand across the United States. More youngsters are getting arrested every year - at lower ages and for more serious offenses. Many will be graduating into the ranks of a criminal army that is costing America billions of dollars a year (U.S. News and World Report, 1967, p. 74).

In 1970, U.S. News and World Report published a story claiming that _In Long Beach, Calif., Police Sgt. James D. Reed says that young thugs who _stalk older people, like animals stalking their prey,_ robbing and brutally beating their victims, want _excitement and money in their pockets_ (U.S. News and World Report, p. 19). Look magazine disclosed in 1966 that _More and more youngsters are involved in burglary, auto theft, shoplifting, and a variety of lesser crimes_ (Moskin, p.24).

Panics over youth crime are as persistent in Western society as are panics over the stock market but they, like so many other alarm bells, are based on political-law enforcement propaganda, not facts.

Arrest rates are the best index we have of the extent of juvenile crime and these data show that juvenile crime keeps pace with the number of juveniles in the population.

In 1966, 21% of the arrestees for violent crimes and 23% of the arrestees for all offenses were under eighteen. In 1969 the percentages were 22% and 26%, respectively. Juveniles accounted for 23% of the violent crime arrests and 26% of all arrests in 1971 and 1973. On average, juveniles accounted for around 22% of violent crime arrests and one quarter of the arrests for all offenses from 1966 to 1973 (Federal Bureau of Investigation, 1967, 1970, 1972, 1974).

These data do not support the police and FBI claims reported in the press that there was a dramatic acceleration in juvenile crime in recent years.

Today we are witnessing another spate of law enforcement driven propaganda about the "time bomb" of juvenile crime, a campaign closely linked to the creation of anxiety over the state of the family in the U.S. where children are growing up "fatherless, jobless, and godless," and dependent on "welfare Moms_ (Zoglin, 1996, Butterfield, 1995).

In fact there has been a slight decline in the percentage of arrests accounted for by the arrest of juveniles since the 1960's and 1970's. In 1994, individuals under 18 contributed to 19% of violent crime arrests and 19% of arrests for all offenses compared to 22 and 25 percent respectively in the period 1966 to 1973. Figures from the past few years also show a reduction in the percentage of arrests accounted for by juveniles compared to the 1960s and 1970s. Juvenile arrests accounted for under 20% of total arrests for both violent crimes and all offenses from the mid 1980s to the present (Federal Bureau of Investigation, 1984a, 1986, 1988, 1990, 1992, 1995).

That the percentage of arrests accounted for by juveniles is less today than in the 1960's and 1970's is explained by demographics. In the 1960's and 1970's people under eighteen made up a larger percent of the population than they do in the 1990's. In 1960 35.7% of the population was younger than 18 and this proportion remained relatively stable over the next 10 years, rising to 36.1% in 1966 and falling to 34.2% in 1970 (Bureau of the Census, 1995).

From 1980 to the present, the proportion of the population under eighteen has held steady at about 26% of the population, substantially lower than two or three decades ago (Statistical Abstracts of the U.S., 1995).

Following these demographic changes, the distribution of arrests by age change as well. Between 1971 and 1994 the percentage of adults arrested increased from 74.2% to 81.4% reflecting the increase in the proportion of the population over eighteen.

Current Panics Over Juvenile Crime

While the arrests of juveniles remained relatively stable over the past decade, there has been an unending public diatribe about the increasing danger posed by juvenile crime. This time the panic is directed at the "near future" when demographic changes will supposedly once again create a massive increase in juvenile crime. The panic is fueled not just by law enforcement agencies but publicity seeking criminologists as well.

Newsweek announced in 1995 that _Criminologists are already warning that the United States can expect another wave of violent crime in the coming decade, and some say it will be much worse than the one that is now subsiding_ (Morganthau, 1995, p. 42).). Time warns that individuals between 14 and 17, _the age group that in the early _90s supplanted 18-to-24 year-olds as the most crime prone . . . is precisely the age group that will be booming in the next decade_ (Zoglin, 1996, p. 52). These media cite not only the FBI and local police but John DiIulio of Princeton University who warns that in the near future the nation will face a generation of "superpredator" teenagers. James Alan Fox, head of the Department of Criminology at Northeastern University, joins this chorus:

_So long as we fool ourselves in thinking that we_re winning the war against crime, we may be blindsided by this bloodbath of teenage violence that is lurking in the future_ (Zoglin, 1996, p. 52).

The Bureau of the Census estimates that in 1996 26.2% of the population was under age 18. By the year 2000, it is expected that the percentage will decline to 25.9%, by 2010 people under eighteen are expected to make up 24.6% of the population and 23.6% in 2025. In 1996 14-to-17- year olds represent 5.5% of the population and will grow to 5.7% in 2000 and 5.8% in 2010 before receding to 5.3% in 2025. Over the next 14 years, if population forecasts are correct, there will be a decline in the total population under eighteen and no more than an increase of 0.3% in the number of juveniles between 14 and 17.

So small an increase will contribute negligibly if at all to an increase in crime.

Fox and Pierce maintain that a dramatic increase in youth crime will occur because by the year 2005 _the amount of 15-19 year-olds will rise 28% among blacks and 47% among Hispanics_ (1994, p. 26). These data are highly misleading, implying as they do that any increase in minority populations translates directly into an increase in crime. From 1980 to 1991 the Black population between the ages of 15 and 34 increased while the white population in this age group declined, but the violent crime rate for the age group 15-34 declined during these years.

Fox and Pierce fail, as well, to point out that the increase in the Black and Hispanic population as a percentage of the total population will be only minimal: the percentage of African American youths between 14-24 years old will increase from 2.3% of the population in 1995 to 2.4% in 2005. The percentage of Hispanics will increase slightly from 1.9% to 2.3% (Bureau of the Census, 1995). The Department of Justice also employs percentage increase statements in ways that exaggerate and spread fear among the population. In Juvenile Offenders and Victims:

A National Report, prepared by the Office of Juvenile Justice and Delinquency Prevention (OJJDP), a headline warns that _If trends continue as they have over the past 10 years, juvenile arrests for violent crime will double by the year 2010_ (Snyder and Sickmund, 1995, p. 111). The report estimates 261,000 juvenile arrests for violent crimes in 2010, a 101% increase from the 129,600 arrests for the same offenses in 1992. Once again, such a statistic has little utility since it is not placed in the context of all arrests. Both the number of juvenile and adult arrests will increase in the future because the total population continues to grow, expanding the pool of potential arrestees.

Furthermore, the OJJDP_s dire prediction is based on the premise that arrests of juveniles for violent crimes over the next 15 years will mirror the annual increases in arrests from 1983 to 1992. However, recall that while the juvenile population has remained relatively stable at 26% of the population since 1980, it will decrease to 24.6% by 2010.

Clearly, the Justice Departments assumption that juvenile arrests will keep pace with the past when the number of youths under 18 is declining reveals a desire to fuel the public_s anxiety about a teenage _bloodbath._

Even more misleading are the OJJDP_s statistics about arrest rates for juveniles. The report claims that _The increase in violent crime arrest rates is disproportionate for juveniles and young adults_ and presents six graphs all showing juvenile arrests for violent offenses outdistancing adult arrests for the same categories (Snyder and Sickmund, 1995, p. 112).

These "facts" were then presented by the conservative Council on Crime in America (1996), whose membership includes the right-wing criminologist John DiIulio, and published under the title "The State of Violent Crime in America" (The Council on Crime in America, 1996).

To arrive at the conclusion that the juvenile violent crime rate is accelerating faster than the adult violent crime rate the authors compared juvenile arrests per 100,000 people 10-17, Not, as the title of the figure claims, per 100,000 population. People under 10, the report tells us. were eliminated because they are rarely arrested. The arrest rate for adults, however, is based on a population of everyone over eighteen years of age. By the same logic that led to isolating the 10-17 year age group and comparing it with the youth population most likely to be arrested, it would be necessary to limit the adult arrest rate to adults in the age groups most likely to be arrested:

at the very least, to eliminate persons over 65 because they, like children under ten, are very unlikely to be arrested. Even more interesting would be to compare the 10-17 age group with people aged 18-35 since these are the adult years in which most arrests occur. Once again we see operating a U.S. Department of Justice report being that conducts so-called research in a fashion designed to spread fear and panic. We see as well the media and generously funded right-wing think tanks spreading the news and insisting that:

"Americans must search for better, more cost-effective ways of preventing violent crimes and protecting themselves and their loved ones from violent and repeat criminals, adult and juvenile. But our first order o business must be restraining known convicted, violent and repeat criminals. Restraining violent criminals is a necessary but insufficient condition for meeting America's crime challenges, reforming the justice system, and restoring public trust in the system and in representative democracy itself."

Questionable math also underlies the statements of John J. DiIulio, Jr., of Princeton University. DiIulio has stated on numerous occasions that the number of juvenile male _superpredators_ will increase significantly in coming years. In an article dubiously entitled _Crime in America - It_s Going to Get Worse,_ DiIulio asserts that:

_The current trend in birth rates makes it certain that a new violent crime wave is just around the corner. Today there are some 7.5 million males ages 14 through 17. By the year 2000 we will have an additional 500,000. About six percent of young males are responsible for half the serious crimes committed by their age group, studies reveal. Thus, in a few years we can expect at least 30,000 more murders, rapists, robbers and muggers on the streets than we have today_ (DiIulio, 1995, p. 57).

DiIulio bases his conclusions on studies _that have shown about 6% of all boys are responsible for about half of all the police contacts with minors_ (Zimring, 1996, p. ?). However, studies of the 6% cohort in several cities indicate that _almost no life-threatening violence showed up in the youth sample that were responsible for the majority of all police contacts...[and that] no study of any youth population supports [a] projection of predatory violence_ (Zimring, 1996).

DiIulio also argues that 270,000 _superpredators_ will be added to the U.S. population by the year 2010. However, as Zimring points out, _If 6% of all males under 18 are superpredators, that means we currently have more than 1.9 million juvenile superpredators on our streets. We would hardly notice another 270,000 by 2010_ (Zimring, 1996, p. ?).

Moreover, in estimating the number of "superpredators" DiIulio counts all males under eighteen. But very few violent crimes are committed by youths under the age of thirteen: _Since 93% of all juvenile arrests for violence occur after age 13_ (Zimring, 1996). DiIulio has included toddlers and children in diapers to project an increase of 270,000 potentially violent youths.

Currently there are 450,000 14 to 17 year-olds in the population and there will be 30,000 added to this cohort by the year 2010. This is a substantial increase but nowhere close to DiIulio's estimated. Furthermore, to assume that the proportion of _dangerous_ young males is constant is ludicrous since the social forces that create violence cannot be reduced simply to a person's age. Even the FBI acknowledges that age is only one of the variables associated with an increased likelihood of violence.


The distortion and manipulation of statistics by the Department of Justice is not limited to the FBI and local police departments. Even when data on crime are gathered objectively by the U.S. Census Bureau, the reports based on these data emanating the Department of Justice's Bureau of Justice Statistics are constructed so as to maximize fear and minimize public understanding.

The Bureau of Justice Statistics (BJS) is responsible for constructing the questionnaires and interpreting and reporting the findings. Once the data arrive in the BJS offices they are under the control of a bureaucracy with a vested interest in presenting the data in a particular light.

Characteristics of the NCVS

After pilot studies were conducted between 1967 and 1972, the first official National Criminal Victimization Survey (NCVS) appeared in 1973. Each year the survey asks a random sample of approximately 135,000 U.S. residents in 65,000 households if they have been the victim of a crime during the past year.

Unlike the Uniform Crime Reports, the NCVS can register crimes not reported to or observed by the police. It also tallies all the crimes that occur in a particular incident, not just the most violent or "most serious."

Like any survey instrument, the NCVS findings must be read cautiously. Residents of the highest crime areas may be the least likely to be surveyed and the most reticent to accurately report their experiences.

People may be reluctant to disclose their victimization. On the other side, faced with an interviewer probing to find victims of crime respondents may well conjure responses to fill the interviewer's questionnaire. or may inadvertently recount crimes that transpired more than a year ago.

The main problem with the NCVS reports, however, are not its methodological weaknesses but the use of the reports to buttress the political and bureaucratic interests of the Department of Justice. Unlike the UCR this is not accomplished by creative statistical gathering techniques, it is accomplished by creatively summarizing the results of the annual surveys.

The 1994 BJS Bulletin reporting the latest results of the Criminal Victimization Surveys follows the same format of every report since the surveys were first published in 1973. It begins with the following statement:

"In 1994 residents age 12 or older experienced approximately ...10.9 million crimes of violence...In terms of crime rates, for every 1,000 persons age 12 or older, there were 51 victims of violence..." (Perkins and Klaus, 1994: p.1)

This statement is grossly misleading. Of the 10.9 million "crimes of violence" reported, 7.7 million (71%) were attempts or threats of violence, not completed acts of violence. A less politicized statement of the violent crime rate revealed by the victim survey would state that:

"Overall during 1994 there were 3.2 million crimes of violence and 7.7 million attempts or threats of violence. In terms of crime rates, for every 1,000 persons age 12 or older there were 15 victims of violence and 36 victims of attempts or threats of violence."

Even this modified statement is probably an over-statement of the frequency with which U.S. residents are the victims of serious, violent crimes. Over 58% of the victims of completed, attempted or threatened violent crimes did not report the crime to the police for a variety of reasons. According to the NCVS report "many indicated that they felt the matter was private or personal in nature," others did not report it because either it was "not important enough" or "nothing could be done about it."

In almost every category of crime reported, it is the least serious crime that accounts for the majority of the instances. Rape, robbery and assaults make up the violent crimes reported in the NCVS (murder is not reported since victims are not able to respond).

Assaults are the least serious of the acts categorized as violent crimes and they account for the vast majority (84%) of all violent crimes. Assaults can be divided into aggravated and simple.

Of the 9,128,000 assaults reported in 1994, 6,650,000 (73%) were simple assaults (with and without minor injury), 2,478,000 (27%) were aggravated assaults. Simple assault without injury which is "an attempted assault without a weapon not resulting in injury" accounts for nearly one half (48%) of all violent crimes. Even victims of aggravated assaults rarely experience injuries: among aggravated assaults less than one third resulted in injury.

The systematic attempt to make the problem of crime seem as bad as the data will allow affects the reporting of property crimes as well. The report states:

"In 1994 the NCVS measured 31 million household burglaries, motor vehicle thefts, and thefts of other property...Expressed as rates per 1,000 households, there were 54 burglaries, 18 motor vehicle thefts, and 236 property thefts." (Perkins and Klaus, 1994: p.2-3).

As with violent crimes, these statements are not false but they are clearly designed to give maximum weight to the seriousness of crime and the danger crime poses for individuals. The fact is that the least serious of the property crimes, property theft, accounts for 77% of all property crimes. Property thefts are most likely to be under $50.00 and thefts under $50.00 account for 22% of all victimizations. Only 14% of all property crimes reported by victims in 1994 were of property valued in excess of $250.00. And 40% of property crime victims claimed they did not notify the police because the item was recovered or because they could not prove it was stolen (Perkins, et. al, 1996).

The maxim that the least serious offense makes up the bulk of all offenses holds true for most of the crimes within specific categories. Thus in 1992, petty larceny without contact characterized 62.3% of the crimes of theft and 35% of all crimes. Petty larceny with contact accounted for 2.6% of all thefts and a minuscule 1.4% of the total. Among household crimes, household larceny was responsible for 54.7% of the victimizations within the category and burglary represented 32.1% of household crimes and only 14.1% of all crimes.

Once again, the most serious crime, forcible entry burglary is the least common type, representing only 10.8% of all household crimes and 4.8% of all victimizations. What this analysis of the facts behind the NCVS report reveals is a systematic bias in summarizing the findings to make the frequency and seriousness of crime appear much worse than it is in fact. The most consistent finding of the NCVS is that most crimes are not reported by the victim, that in almost every crime category surveyed the majority of the criminal victimizations are for the least serious offense in the category, and that there is in fact no infliction of actual violence in the majority of so_called _crimes of violence._ Such data should lead the authors of the NCVS report to highly qualified, cautious statements about the extent to which there is a serious crime problem in the United States. But political and bureaucratic interests take priority over accuracy in the hands of the crime control industry. Indeed, the opening statement of the NCVS report could say "Last year 85-90% of all residents in the U.S. were not the victims of any crime. Furthermore, the majority of those who were victimized were the victims of petty theft.

Less than 10% of the population was the victim of any type of violent crime, and the vast majority of these victims were victims of attempted or threatened violence but suffered no actual violence."