History holds the key to understanding why America is so homicidal

Cover Interview of

In a nutshell

No matter where Americans live, their risk of being murdered is higher than it is in any other first-world democracy.

From 1965 to 1992, the homicide rate in the United States averaged 9 per 100,000 persons per year. It declined in the mid-to-late 1990s to 6 per 100,000 persons per year, but it remains comparatively high. The next most homicidal affluent democracy, Canada, has had only a quarter of the homicides per capita that the United States has had since World War II. The world’s other affluent democracies have had homicides per capita rates ranging from a fifth of that of the U.S. (Australia) to less than a tenth (Ireland).

The United States ranks first even when populations are compared by gender and ethnicity. Both women and men are far more likely to be murdered in the United States, and Americans of European descent, the least likely victims of homicide in recent decades, were murdered at a rate of 5.5 per 100,000 persons per year from 1965 to 1992. By itself that rate is high enough to make America two and a half to eight times more homicidal than any other affluent democracy.

Americans are exposed to that high annual homicide rate for their entire lives, an expected 78 years for children born today. The likelihood that a newborn American will be murdered if the homicide rate of the recent past persists—as it did for most of the twentieth century—is not 9 in 100,000, but 78 times that. In practical terms, that means one of every 142 children born today will be murdered—one of every 460 white girls, one of every 158 white boys, one of every 112 nonwhite girls, and one of every 27 nonwhite boys.

Why do homicide rates vary so drastically from one society to another, from one time to another, if murders are so alike in their motives and circumstances? Why, if humans have roughly the same capacity for violence, does murder claim one in 10,000 adults in some societies, and one in 20 in others?

To find out what circumstances ultimately foster high homicide rates we first have to go back through history, chart their course, and then make the connections between historical circumstances and the human beings who commit murders.

Doing so shows that homicide rates among adults are not determined by proximate causes, like poverty, drugs, unemployment, alcohol, race, or ethnicity. Instead, factors that seem on the face of it to be impossibly remote—like the feelings that people have toward their government, the degree to which they identify with members of their own communities, and the opportunities they have to earn respect without resorting to violence—determine homicide rates.

In other words, history holds the key to understanding why the United States is so homicidal today.

“One of every 142 children born today will be murdered—one of every 460 white girls, one of every 158 white boys, one of every 112 nonwhite girls, and one of every 27 nonwhite boys.”

The wide angle

The theory of homicide I develop in the book first emerged because my initial hypothesis “died a horrible death” in the face of the evidence from Vermont and New Hampshire, where my study began.

I set out to understand why northern New Englanders were virtually non-homicidal. As I gathered data from beyond the time period I first studied, however, I discovered to my dismay that they were not. By the mid-nineteenth century American New Englanders had become more homicidal than their counterparts in England.

The book I had planned to write on northern New England’s “non-violent” culture was in ruins. However, when I separated by type the homicides I had found in New Hampshire and Vermont, I discovered that the patterns of homicide made sense in terms of New England’s history.

Murders of children by adult relatives or caregivers followed a long, smooth curve that was the inverse of the birth-rate: high fertility meant a low child murder rate and low fertility meant a high murder rate.

Marital homicides and romance homicides jumped suddenly in the 1830s and 1840s: decades in which jobs opened to women in education and industry and in which the ideal of companionate marriage took hold.

Homicides among unrelated adults peaked during periods of political turmoil and of loss of faith in government: the Revolution, the Embargo crisis, and the sectional crisis.

It appeared that state breakdowns and political crises of legitimacy produce surges in nondomestic homicides; the restoration of order and legitimacy produces declines in homicides.

The same pattern was evident on the national level in the twentieth century, a period for which comprehensive homicide statistics are available. The establishment of government legitimacy through the New Deal, World War II, and the Cold War appeared to have reduced homicide rates through the 1950s; the crisis of government legitimacy in the 1960s and 1970s (especially in the eyes of African-Americans) may have contributed to soaring homicide rates.

I knew that it would take more to prove my theory than evidence drawn from the history of Vermont and New Hampshire, my area of expertise. So I extended my research to the colonial period, to early modern Europe, and outward to the South, the Midwest, the West, and the urban East.

Everywhere I looked, the domestic murder rate for children followed the inverse of the birth rate up to the end of the nineteenth century. Marital and romance homicides increased suddenly in the 1830s and 1840s across the northern United States—as well as in England and northern France. Everywhere I looked, homicides among unrelated adults correlated with political events.

I conducted “natural experiments” to prove that correlation. I hypothesized, for instance, that the homicide rate would soar during the American Revolution and remain high for decades afterwards in the Georgia-South Carolina backcountry, where the Revolution was a genuine civil war. I also hypothesized that the homicide rate would hold steady or fall in the Shenandoah Valley of Virginia, which enjoyed political stability under patriot control throughout the Revolution, and where support for the war effort and the new federal government was stronger than anywhere else in the South. My research in local archives confirmed these and other hypotheses. And every measure I could find of changes in people’s feelings and beliefs supported the theory.

August 30, 2010

A close-up

Historically, only a small minority of murders of unrelated adults have had their origins in long-term, hostile relationships. People who killed non-relatives who lived with them—boarders, landlords, slaves, servants, masters, mistresses—had deep-seated, personal reasons for doing so. Killer and victim were bound together emotionally, interacted with each other almost daily, and could not easily sever ties. The vast majority of murders committed by women have always been personal.

However, in the vast majority of homicides of non-relatives in both America and Western Europe there was no long-term, hostile relationship between murderer and victim. The violence wasn’t generated by the relationship between the victim and the perpetrator. The killers were already predisposed to violence. They were already prepared to view people as enemies or rivals or prey.

The strength and prevalence of that predisposition determines whether men in a given society are homicidal or non-homicidal, whether they are emotionally prepared to be violent at the slightest provocation or whether they refrain from violence even if they are brutalized or humiliated.

Where does the predisposition to violence come from? What causes men to be so alienated that they can kill passersby for money or sex? What causes men to view every encounter with another man as having the potential to be a life-and-death struggle for supremacy or self-preservation?

The predisposition to violence is not rooted in objective social conditions. Men who are poor, oppressed, or unemployed can be disposed to violence in one historical situation and nonviolence in another. The same is true of men who are well-off. The predisposition to violence is rooted in feelings and beliefs.

“State breakdowns and political crises of legitimacy produce surges in nondomestic homicides; the restoration of order and legitimacy produces declines in homicides.”


Four correlations emerge from an examination of homicide rates in parts of the United States and Western Europe throughout the past four centuries.

One, a correlation with a belief that government is stable and that its legal and judicial institutions are unbiased and will redress wrongs and protect lives and property;

Two, a correlation with a feeling of trust in government and the officials who run it, and a belief in their legitimacy;

Three, a correlation with patriotism, empathy, and fellow feeling arising from racial, religious, or political solidarity;

Four, a correlation with a belief that the social hierarchy is legitimate, that one’s position in society is or can be satisfactory and that one can command the respect of others without resorting to violence.

The feelings and beliefs in these correlations are closely related—especially the first three; the absence of one usually means at least a partial absence of another. They also have synergistic relationships with the homicide rate. When the homicide rate rises, for instance, because of a loss of government legitimacy or a decline in fellow-feeling, the rise in homicide itself can undermine the belief that government can protect lives and that citizens care about each other and thereby bring about a further increase in homicide.

An increase in homicide can also change the character of a society’s social hierarchy and make violence a means of winning respect. Homicide rates can then soar into hundreds per 100,000 adults per year. Alternatively, when nearly all citizens believe their government is stable and legitimate, when they feel a strong bond with their fellow citizens, and when they believe their society’s social hierarchy is just and violence is not necessary for respect, homicide rates can fall below 2 per 100,000 adults per year. In most societies, these beliefs and emotions have been neither entirely absent nor widely shared, which is why most historical homicide rates have fallen between the extremes.

Encouragingly, these conclusions are supported by the findings of scholars in other fields—notably Tom Tyler in psychology, Gary LaFree in criminology, and the later Roger Gould in sociology—who have discovered independently that political legitimacy and political stability can deter homicide (and other crimes) among unrelated adults.

Criminologist Gary LaFree, for example, confirms the fundamental importance of feelings and beliefs when he points out that of all the variables social scientists have collected data on in the past fifty years, homicide rates among unrelated adults in the United States have correlated perfectly with only two: the proportion of adults who say they trust their government to do the right thing and the proportion who believe most public officials are honest. When those proportions fell, as they did in the late 1960s and early 1970s, the homicide rate among unrelated homicide rate rose.

As I say in the foreword to American Homicide, the theory we are developing is after all a “working hypothesis.” I plan to test it further, and I hope other researchers will put it at risk in against new evidence.

© 2010 Randolph Roth
Randolph Roth American Homicide Belknap Press of Harvard University Press 672 pages, 9 1/4 x 6 1/2 ISBN 978 0674035201
Randolph Roth is Professor of History and Sociology at Ohio State University, where he has founded and co-directs the Historical Violence Database at the Criminal Justice Research Center. He is the recipient of numerous awards, including the 2007 Distinguished Teaching Award from the Ohio Academy of History, the 2009 Ohio State University Alumni Distinguished Teaching Award, and the 2011 Michael J. Hindelang Award from the American Society of Criminology. Roth received his Ph.D. in History from Yale University.
Cover Interview of
August 30, 2010