The numbers are in and even experts who study violence are surprised: Cities have not always been meccas for murder, and this is not the first heyday for homicides.
Historians now say that homicide rates were extraordinarily high in Europe during the Middle Ages _ and high in the United States during the early 19th century _ then declined steadily until the 1960s. And for centuries, it was villages that were often the scenes of violence.
These findings, published in recent papers and presented earlier this month at the annual conference of the Social Science History Association here, contradict a basic tenet of criminology, that violence is endemic to densely packed urban and industrial centers where traditional social ties and values necessarily break down.
The new research should "undermine the basic claim that the city, with its anonymity and crowding, is in and of itself the factor in causing violence," said Joan McCord, a professor of criminal justice at Temple University and a past president of the American Society of Criminology.
Eric Monkkonen, a professor of American urban history at UCLA, said: "What we are finding is that violence is not an immutable human problem. There really has been a civilizing process" in which, scholars say, an increase in state power and courtly manners beginning in the 16th and 17th centuries helped curb impulsive, violent behavior.
When the first research on homicide rates was reported more than a decade ago by historians studying England, scholars doubted its accuracy, since it relied on old court records and coroners' data. In recent years, critics have focused on the population base used to calculate homicide rates: Precisely how many people were living in England in the 1400s, for example?
But with researchers reporting similar results from many countries over centuries, including the United States, Australia, England, France, Italy and the Netherlands, there is mounting agreement, said Lawrence Stone, professor emeritus of history at Princeton, that "all these figures just can't be false."
New data presented at the conference by a Dutch scholar, Pieter Spierenburg, showed that the homicide rate in Amsterdam, for example, dropped from 47 per 100,000 people in the mid-15th century to 1 to 1.5 per 100,000 in the early 19th century.
Stone has estimated that the homicide rate in medieval England was on average 10 times that of 20th century England. A study of the university town of Oxford in the 1340s showed an extraordinarily high annual rate of about 110 per 100,000 people. Studies of London in the first half of the 14th century determined a homicide rate of 36 to 52 per 100,000 people per year.
By contrast, the 1993 homicide rate in New York City was 25.9 per 100,000. The 1992 national homicide rate for the United States was 9.3 per 100,000.
After examining coroners' inquests, Barbara Hannawalt, a professor of medieval English history at the University of Minnesota, concluded that most slayings in medieval England started as quarrels among farmers in the field. "They were grubbing for existence," she said. Insults to honor were taken seriously, and violence was the accepted method of settling disputes, since the king's courts were slow, expensive and corrupt.
The knife and the quarterstaff, the heavy wooden stick commonly carried for herding animals and walking on the muddy roads, were the weapons of choice. "Everyone carried a knife, even women," she said, since "if you sat down somewhere to eat, you were expected to bring your own." Given the lack of sanitation at the time, even simple knife wounds could prove deadly.
Why the homicide rate in Europe began to drop in the 16th and 17th centuries is a matter of debate. The most widely accepted explanation stems from the work of Norbert Elias, a sociologist who in the late 1930s introduced the idea of a "civilizing process," in which the nobility was transformed from knights into courtiers, bringing in a new set of manners, and the modern state spread its power over the populace.
Official justice administered by courts replaced private vengeance conducted by feuds, fights and duels. Challenging conventional academic wisdom, Elias suggested, too, that the power of the state extended to cities first, so urban homicide rates would be comparatively low.
Recent research indicates that he was right. In Philadelphia, for example, the annual average rate of indictments for homicide fell from 4 per 100,000 in the 1850s to 2.2 in the early 1890s, according to research by Roger Lane, a history professor at Haverford College.
The drop occurred as the city became industrialized and despite the greater availability of firearms during the period.
As people began to go to work in factories, their behavior was constrained by the foreman and the whistle, Lane has written. Behavior was also improved by the spread of public schools and institutions like the YMCA and Sunday schools, which taught morality.
Similar developments helped lower New York's homicide rate in the late 19th and early 20th centuries, Monkkonen said.
Then came the 1960s. Around the world, homicide rates did an about-face. When President John F. Kennedy was assassinated in 1963, for example, the national homicide rate in the United States was 4.6 per 100,000 people. By 1970, it had doubled, and by 1980 it had reached 10.1 per 100,000.
Researchers point to several possible explanations. Post-World War II baby boomers came of age in the 1960s, increasing the number of young men, the most violence-prone group, in the population.
The '60s also marked a shift among many social, cultural and economic forces that worked against violence in previous eras. America began moving into a post-industrial economy, governmental authority came into question with the Vietnam War, and the traditional family was threatened by things like divorce.
"The good news is violence can go down," said Monkkonen. "The bad news is, we need to learn how to make it happen."