Murder in America has fallen into a historic slump, and that’s a fact more people can live with. Murder rates have dropped to levels not seen since the mid-1960s, punctuating the end of a bloody 20th century where more than twice as many died in American homicides as U.S. troops did in wars.
“No one has a good theory that explains the drop,” says David Baldus, a University of Iowa law school professor and recognized expert on murder prosecutions.
“Police take credit for it, but we don’t know the answer,” says Mr. Baldus, who is among several analysts citing a decline in drug-related shootings.
The picture of who is dying and who is killing, assembled by the federal Bureau of Justice Statistics (BJS), shows that in the past quarter century about three of every four victims were male and 10 percent of killers are females.
The 25-year total includes 1,912 police officers killed on the job and 94,142 slain young men ages 18 to 24 — most killed in the kind of “gang-related” incidents that perplex big-city police chiefs.
Among findings in BJS’ comprehensive examination of 507,681 murders from 1976 through 2000 (the last year in which the data have been analyzed completely):
2000 was considered a good year because criminals took “only” 15,317 lives, down by 9,000 from annual tolls just a decade before.
Blacks make up 12.1 percent of the nation’s population but commit most of the murders and are over-represented among homicide victims. They are six times more likely to be murdered, and seven times more likely to kill.
White criminals dominate among those executed and those waiting on death row. They most often commit serial or mass murders.
The best way to prevent murder is to avoid arguments. They are the single leading factor that accounts for four of every 10 homicides.
As for the demographics of death, the federal official who analyzes the data finds them far too disparate to pick out a “typical” victim, or those who spark the most emotions.
“They all bother me. I work with it all the time, but you never forget the victims,” says Marianne W. Zawitz, the BJS statistician who lectures on techniques of researching homicide trends. “It’s troubling. I use examples from my work when I lecture on data presentation and these professionals are horrified by what I’m presenting to them.”
The homicide toll of 15,317 for 2000 was a dramatic decline from 1991’s all-time high of 24,495. Number-crunchers point out that 3,400 more homicides occurred in the study’s baseline year of 1976 than during 2000, when the nation had 80 million more residents.
Put another way, the nation statistically is safer now than a quarter-century ago, and for many gruesome years in between. With a population increase of 39 percent, the murder rate is down 41 percent.
“I’m not going to speculate on why. We don’t offer projections and we don’t give any causality,” says Ms. Zawitz, who tracks homicide trends with criminologist James Alan Fox, a Northeastern University professor and visiting fellow at BJS.
But the good news may not last.
Low totals are holding steady since 2000, but a few criminologists who track the grim numbers in much the same way Wall Street follows the stock market predict the most encouraging homicide figures in two generations are only a prelude to renewed gang warfare.
“I see lots of reasons why the murder rate could go up and few reasons why it would go down,” says Alfred Blumstein, director of the National Consortium on Violence Research and author of “The Crime Drop in America.” “We have solved the easy stuff. It is going to be a lot harder to go beyond that.”
Because murder often is an offshoot of other crime, Mr. Blumstein anticipates a deadly impact from diversion of police resources to terrorism duty, tighter government budgets and new anxieties about the economy.
Who the victims are
BJS’ torrent of detailed numbers on victims and killers reflects race, age, sex and circumstances of the crimes. The data also include such unpublished minutiae as the number of victims pushed out windows (193) or strangled (7,670); the most murderous month (August); and the number of killings by snipers (480).
The BJS figures also include the most helpless victims.
The latest year had 601 murder victims who hadn’t reached their fifth birthday. Typically parents or others close to them do the killing, including baby sitters, who have killed 677 children during the 25-year period.
At the other end of the scale is “eldercide” — 368 men and 326 women older than 64 killed in 2000. Since 1976, BJS has counted 120 murder victims who were 90 years old and 2,755 in their 80s. Eldercides declined sharply from 5 percent of all deaths to about half that in recent years and now occur primarily during another felony, such as armed robbery.
The Americans statistically safest from being murdered are those ages 5 through 11, less-vulnerable children old enough to have outside contacts. About one in 100 victims is in this age group, compared with three in 100 who are younger than 5. (About 40 million children are under 12.)
Numbers cluster most notably in one age bracket, 14-24. That group includes 45 percent of all murderers and 29 percent of victims, but is only 17 percent of the population.
“Clearly what the data show is young men are most likely to be victimized, and generally by other young men,” Ms. Zawitz says.
Mr. Blumstein says violence tapered off even among this group as the impact of “the later phases of the crack-cocaine bubble” diminished along with young criminals carrying guns on the streets.
“The rise we saw from 1985 to 1993 was largely the consequence of young people having been recruited into crack markets, being armed, and doing what young people have always done — which is fight, but with much more lethal weaponry,” Mr. Blumstein says.
While declining to offer personal theories on cause and effect, or to profile victims, Ms. Zawitz says the best prevention advice is simply to avoid the arguments that her research pinpoints as by far the key precipitating factor of homicide.
Arguments are known to have triggered 199,478 homicides, 39.3 percent of all cases in the 1976-2000 database. Factors include alcohol, drugs, money disputes and sexual unfaithfulness.
Arguments epitomize killings by “intimates” — the 58,170 women and men killed in 25 years by past or present lovers. This category too often proves the thesis of 19th-century French author Octave Mirbeau that “murder is born of love, and love attains the greatest intensity in murder.”
These dead include 10,405 girlfriends, 5,744 boyfriends, 20,356 wives and 11,891 husbands. Unpublished data files show that during the quarter-century studied, 1,158 victims were slain in homosexual relationships, about 2 percent of the “intimates” category.
Not all the details are in public reports and that includes the role of intimate same-sex partnerships in slayings. Such raw research often is excluded from the encyclopedic 178-page annual report because numbers are too small, a comparable detail is not reliably reported by enough jurisdictions or, as with the homosexual cases, the relevance is not clear.
Overall the number of “intimates” murders is declining, but the ratio of women being murdered has gone from 1.2:1 in the early 1980s to 2:1 by 1994 and almost 3:1 in the past few years.
“Counseling services in big cities reduced one aspect of domestic killing: women killing men,” Mr. Blumstein says, calling that a surprising result of programs intended to protect women.
“The good news is that women killing men is down,” agrees Carolyn Rebecca Block of the Illinois Criminal Justice Information Authority, who is involved in several homicide studies.
She says the most common triggering factor in the “intimates” category is one partner leaving or announcing the intention to leave.
The rate of homicides per 100,000 persons in 1976 was 9.22, compared with 5.44 in 2000. Until then, levels below 5.5 had not been seen since 1965, according to BJS.
BJS includes all degrees of murder, manslaughter and non-negligent homicide, but excludes “justifiable homicide” by police and civilians, suicides, accidental deaths from such causes as drug overdose, and court-ordered executions that some other agencies record as homicides.
Federal and local analyses also exclude the current official count of 3,016 criminal homicides in the September 11 terror attacks. Maryvictoria Pyne, who heads the FBI Uniform Crime Reports unit, says the omission avoids a statistical burp in tracking common crimes.
“That’s going to skew your trend,” Miss Pyne says.
“September 11th is not relevant to any of the statistics of homicide counts. It’s appropriate to have as an asterisked footnote,” Mr. Blumstein agrees.
What happens to killers
Sources other than BJS show what happens to the killers. Historically, police report some 80 percent of murders “cleared” by arrest or identification of the killer.
Recently, that clearance rate is declining to about two-thirds but the number of killers convicted imposes an enormous burden on prisons.
California alone holds 23,144 persons serving terms for killings, including 9,951 for first-degree murder. An additional 1,986 killers are on parole.
From 1976 through April 1, 842 killers were executed. Another 3,525 remain on death rows operated by the federal government and 37 states, including California, where 622 of the condemned wait.
“Who actually gets executed depends on whether you’re in Texas,” says Richard Dieter, executive director of the Death Penalty Information Center, which advocates letting juries sentence killers to life without parole, an option not officially available in Texas.
California has executed 16 prisoners since 1977, while 301 were executed in Texas, where 454 others were awaiting execution as of April 1.
While many victims’ families appear to favor executions, one group of 1,000 such “family survivors” actively seeks to abolish the death penalty.
“We oppose capital punishment not so much because we care what the death penalty does to perpetrators, but what it does to ourselves. It replicates the kind of violence that already has victimized us,” says Robert Renny Cushing, executive director of Murder Victims’ Families for Reconciliation in Cambridge, Mass.
He says his membership includes families of executed murderers, whom he also counts as victims. They include Sonia Jacobs, who pleaded guilty in 1992 to second-degree murder in the 1976 slayings of two police officers. Her boyfriend was executed for those crimes in 1990.
“Many of us who reject execution needed to find a voice,” says Mr. Cushing, 51, whose father was killed in 1988. He claims “scores of instances” when MVFR members influenced prosecutors to change a strategy, or sought a governor’s mercy.
The color of murder
The Justice Department says black criminals commit 59.2 percent of felony murders. Although that category accounts for about one of every five homicides, prosecutors seek death sentences for felony murder more often even than for criminals who murder prison guards or kill while resisting arrest.
Those statistics were confirmed in an interview with Mr. Baldus, a social scientist who developed them while studying claims that courts levied more severe punishment for “killing while black.”
When a murder involves rape, the defendant is 12.8 times more likely to face the death penalty, according to research he did for a Supreme Court appeal.
Mr. Baldus found a death-odds multiplier of 4.2 for a murder during an armed robbery, 7.9 for a crime with multiple victims, and even higher for a paid killer, use of torture, insurance motives, or murdering a judge or prosecutor.
Whatever the defendant’s race, when the victim is white it is 4.3 times more likely the defendant will go on trial for his life, Mr. Baldus says.
The latest edition of the National Association for the Advancement of Colored People Legal Defense Fund report “Death Row U.S.A.” says that, since 1977 when executions were resumed, 12 whites have been executed for killing blacks and 182 blacks have been executed for killing whites.
LDF’s records show that overall 477 whites, 290 blacks, 50 Hispanics, 22 Asians and three American Indians were executed from 1977 through April 1. Those criminals’ 1,271 victims — 1,022 of whom are listed as white — were divided almost equally between males and females.
“Studies almost uniformly show there is no statistical difference to the way blacks now are treated vis-a-vis nonblacks. There is no black-defendant effect,” says Mr. Baldus, referring to the upheaval in murder prosecutions since the Supreme Court approved procedures under which states resumed capital trials.
“The things that increase the likelihood of being charged with a death-penalty offense are the number of aggravator circumstances and the level of brutality in the case,” he says.
That is disputed by LDF’s Miriam Gohara.
“No, not at all. I don’t think we agree,” says Miss Gohara, arguing that there is political pressure on prosecutors.
“Race continues to be a factor in a prosecutor’s decision on who to prosecute and, in our view, an impermissible factor.”
Miss Gohara does agree, however, that the victim’s race is “the most statistically quantifiable figure” and that the black minority commits the majority of murders.
But she says the fact that interracial murders are less common makes the imbalance in death sentences more notable.
“The vast majority of crime in general is white on white, and black on black,” the NAACP lawyer says.
Dudley Sharp of the pro-death penalty organization Justice for All says that public acceptance of arguments of racial discrimination involves a misunderstanding of how aggravating factors contribute to which murderers are sentenced to die, or put on trial for their lives in the first place.
“The argument is the system only cares about white murder victims, so they’re only pushing those categories. Guess what? That’s what it should do, because whites are far more often the victims of crimes with aggravating factors, rather than impulse or passion killings,” Mr. Sharp says.
“Everyone’s aware that the overwhelming majority of black murder victims are the result of black murderers. But those cases are rarely combined with secondary aggravating circumstances,” he says.
In taking that dispute to court, the statisticians’ jargon-filled world of numbers, odds, likelihoods and probabilities proved so arcane that practitioners such as Mr. Baldus say it confused Supreme Court justices into misjudging a 1987 case.
“The statistics are so prone to error in interpretation that the United States Supreme Court made the error,” Mr. Baldus says of the court’s rejection of his theory in the failed 1987 appeal of Warren McCleskey, a black man eventually executed for killing a white man during a Georgia armed robbery.
The high court ruled the Baldus study failed to establish unconstitutional discriminatory intent, or significant racial bias by prosecutors.
The Justice Department uses the word “intraracial” to describe murder in America. From 1976-2000, according to BJS, “86 percent of white victims were killed by whites; 94 percent of black victims were killed by blacks.”
And NAACP figures show that of the 175 prisoners put to death since 1977 for multiple slayings that collectively took 618 lives, 131 were whites who had killed 92 percent of all white victims.