Why do we kill each other? Humans are a homicidal species, and Americans are uniquely murderous among people who enjoy similar levels of wealth and freedom. For American Homicide, Randolph Roth, a professor of history and criminology at Ohio State University, crunched a vast amount of data from American counties and cities from colonial times to the present to come to a conclusion even he didn’t expect at first. After 25 years of research, he found that poverty, unemployment, discrimination, race, class, substance abuse, gun laws, and all the other factors we say cause murder actually don’t. Instead, as he explained to Zócalo, it’s about whether we trust the government, how close we feel to each other, and whether we think government and social hierarchies are legitimate.
Q. You note in your book that all the usual factors we blame for violence – gun laws, poverty, discrimination – don’t capture the big picture. Why do we continue to believe these factors impact homicide levels?
A. It comes down to the fact – mea culpa here, this is my generation’s fault more than yours – that we’re politically polarized. We tend to look at the world through liberal and conservative lenses, even when we’re people of goodwill and have friends on both sides of that divide. This is even my first impulse. When I started out, my original theory died a horrible death in the face of the evidence: I thought that if we eliminate poverty, if we eliminate discrimination, if we build equal opportunity, if we end unemployment, we’ll solve the homicide problem. Well, we’d make the world a better place, but we won’t solve violence. Having swift and certain justice, deterring crime, having a strong police force are all necessary things to give us a sense that society works. But law enforcement and policing have not solved the homicide problem. I have posted diagrams online where you can see the size of police forces over time in major cities – there’s no correlation between that and the homicide rate. You can’t find a long-term correlation with incarceration rates, either. But we keep going back to what we believe about the world. We believe that substance abuse is the issue – but alcoholism and drunkenness were worst in the 1790s to 1820s, which was a period of very little homicide. In one book about the time, my friend William Rorabaugh calls it “The Alcoholic Republic” – we drank three times as much as we do today. Delirium tremens was a problem, liver damage was a problem, but the homicide rate was not that bad. We want to blame unemployment, but when FDR became president, the homicide rate dropped like a rock.
When we’re trying to be catholic about our explanations, we tend to do what conservatives and liberals did in the late 1990s-they split the difference, agreeing to build a strong economy, lower unemployment, hire more police, and imprison more people. But at the end of the effort our homicide rate is still six per 100,000 per year, which is way out of line with the rest of the affluent world.
I started out studying northern New England, and I noticed how non-homicidal it was in the early 19th century. I thought, wow, this has to be because they created the most democratic society on the face of the earth. But when I went in and looked at the data, I saw they started killing each other soon after – it was worse than in Manchester or Liverpool. And I looked at other American cities and rural communities – they were all nonviolent in the early 19th century, from the rural Midwest to Philadelphia and Boston to the mountain South. The only exception was the slave South. My theory died.
Q. What was special about that time period that led to low homicide levels, and is that when you began to form your theory about homicide?
A. When it really hit me was when I realized that one of the most horrific events in American history lowered the homicide rates. There was a horrible Indian war in New England called King Philip’s War, from 1675 to 1676. We believe that one out of 10 male colonists died in that war, a third of New England’s towns were destroyed, and between 25 and 50% of the native population of New England perished in massacres, by starvation, or by warfare or terrorism. There had been peace between Native Americans and New England colonists since the end of the Pequot War in 1637.
Puritan New England was far more homicidal than I thought – it was as homicidal as America is today. After that war, New Englanders just about stopped killing each other. The homicide rate dropped again in the 1690s after the Glorious Revolution in England. You go from having a homicide rate of seven to nine per 100,000 adults per year one per 100,000, within 17 years. What in the world did that? I had to start to think about it – it was coming to grips with a sort of terrible truth, that it was because of a sense of kinship, empathy, patriotism, fellow-feeling, along racial or national or religious lines. Puritans in New England previously persecuted colonists who were not one of the group. Then after the war, they stopped whipping young women who committed fornication, they stopped persecuting Baptists. There is a wonderful poem I quote in the book by Peter Folger, who was Ben Franklin’s grandfather. He said God has punished us with this war because we’re fighting over the meaning of Christianity. We have to stop that.
It’s eerie because it is about race – New Englanders felt unified by race and nationality. They were white, patriotic Britons who believed in the government of William III, that it was legitimate. Something I didn’t want to learn was that being on the winning side of a race war can drop the homicide rate among the winners. We’re talking now about the elements of human nature. When you have political stability, when you have a sense of legitimacy of government, when you have a sense of fellow-feeling, the homicide rate drops. I tested the theory again with the rise of racial slavery in the Chesapeake. If my theory was right, the homicide rate would drop, and bang, it did, right on cue. The white elites start to treat poor Virginians and Marylanders better. It’s in their diaries, it’s in their language – they don’t talk about them as rogues and whores, the vocabulary they used to describe working class indentured servants. They increased their political enfranchisement. And they gave them a gun when they attained freedom to use against blacks. They didn’t pull together as successfully as New England did – the class divide was too deep, and the early years of racial slavery did lead to terrible violence against African American people.
That’s when I started to think about the present.
Q. On the face of it, it seems like some of the reasons you say homicide is high – a sense that government and the social hierarchy aren’t legitimate – shouldn’t apply to the U.S. How do you reconcile this assumption with your theory?
A. What’s extraordinary about the U.S. is the level of distrust in government, which has been with us since the Civil War. It’s a weird thing to see in a democracy – it’s an odd form of self-hate. We vote for a government, and we despise it. You hear all the time, “Oh, the damn government,” and that sentiment has been strong since the political crisis of the 19th century. What you see, and this is criminologist Gary LaFree’s work, is that the best correlates for homicide rates from the 1950s to the present is the percentage of Americans who in polls say “I trust the government to do the right thing most of the time,” and, “I believe most public officials are honest.” When people trust the government and believe public officials are honest, the homicide rate has been low. When those numbers are down, the homicide rate peaks. Black mistrust of government peaked under Nixon – and that’s when the black homicide rate peaked in the post-World War II era. The white homicide rate peaked in 1980, when the anger accumulated over affirmative action, busing, Vietnam, welfare, the humiliation of the hostage taking, hit its peak, particularly on the right. That’s when you see the homicide rates for whites triple compared to the 1950s – the black homicide rate only doubled. Even successful politicians of that period ran against Washington, against government and bureaucracy. Instead of re-legitimizing government, they were, in their way, tearing it down. I don’t mean to fault them for it, but when you see so many attacks on the legitimacy of government, that’s a problem, as far as homicide is concerned.
When we look at the social hierarchy problem, it’s not about the justice of it. The U.S. is certainly one of the wealthiest places on Earth, and are poor people better off here? In many instances yes. My grandparents were immigrants, English is not my parents’ first language, and where did I get to go to college? I’m not arguing that this isn’t a wonderful country, but the question is, how do we feel about the social hierarchy? What’s stunning is that in a caste society, I’m thinking of Sri Lanka in the early 20th century, before independence, you’ll most likely end up where you’re born. But if people believe that’s legitimate – if they believe people are willing to play their roles and respect one another’s roles, people are not upset. They’re not anxious. When a system loses legitimacy, and people no longer believe it’s fair, that’s when things fall apart. When we look at deindustrialization and agribusiness and what they’ve done, they have put another huge distance between the poor and the middle class, greater than my parents and grandparents faced. Losing those entry-level industrial jobs and small farms, by which you can build a middle class family, is a problem.
But it’s also about how people feel. How do you measure legitimacy? In the 1960s, as soon as civil rights legislation passed, you would think people’s reaction would be, “Wow, now we have opportunities.” But the reaction, especially among minorities, was that this is intolerable. Nothing has changed. Anger shot up, even though things should have been getting better. When you make a commitment as a society to change, and you don’t, it’s one of the most homicidal situations you can create. When the North committed itself to end slavery and empower common people after the American Revolution, homicide went down. Even though there was still discrimination and poverty, people felt things were going in the right direction. But in the slave South, where nothing changed, homicide went up. Throwing democratic ideas into that society was like throwing a match in a can of gas. The homicide rate went way up in the slave South after the Revolution. Imagine if you’re an African American and you didn’t get freed. You’re a poor white man and you have no power. Or you’re a wealthy white man and you’re being challenged. They started killing. The same thing happened in the southwestern borderlands after the Mexican Revolution. When you throw democratic ideas into a society still bound by class and caste, you delegitimize the social hierarchy, and people are more upset. It’s the feeling that matters – it’s not an objective response to reality.
I’m shocked at how little legitimacy our government has. Friends from Europe say, “You know, it’s always struck me how much Americans hate their government.” It’s crazy.
Q. Why does anger at the government or the hierarchy manifest in murdering other people, rather than – and not that this is a better alternative – attacks on the government, or on the powerful?
A. It’s a diffuse kind of anger. The person who a murderer attacks is very seldom the person that got the murderer angry in the first place. Why can the slightest thing set murderers off? It’s not that the slightest thing set them off. They’ve been set off before then. They are predisposed to respond. And because murderers, like everyone else, spend most of their time with friends, in their neighborhoods, at the neighborhood bar – when they’re angry, who are they likely to kill? No one associated with the government – and we don’t want that, that’s very chaotic. They’re more likely to kill a friend or acquaintance.
This is true of other primates as well. Jane Goodall found, looking at chimps, that if one is disrespected, humiliated, beaten – especially by somebody above them who they can’t get back at-it could take a day, hours, or a week, but suddenly they will whack somebody of the same status or a lower status. We’ve seen this enough times to know that who you’re humiliated by doesn’t indicate who you will end up hurting. And all primates have politics. If you look at the literature on vervet monkeys, when there is a struggle for dominance, when you lose political stability, male and female vervets secrete more hormones that we associate with aggression. It’s not like a neurotransmitter – it’s not as though I think a thought, therefore I commit a violent act. Hormones just increase your predisposition toward violence. They ready you for certain things. The vervet monkeys started hitting each other – not just in ways related to the dominance struggle, but in all sorts of ways. As soon as the new dominance structure sets up, everybody stops secreting and the violence goes down. The return of political stability decreases violence.
What we’re seeing is forensic anthropology, social science history, primatology, and endocrinology starting to speak to each other. We’ve evolved to be intensely cooperative but also intensely competitive. I think Frans de Waal, the great primatologist, is right when he says we’re “bipolar.” Particular situations bring out particular aspects of our being, of our chemistry. When you look at homicide in lots of different places over time, you see deep structures in human behavior. I think people who are frustrated, who feel things are unjust, who feel no connection with the people around them, can lash out at anybody. It’s a fact.
Q. Do you think Barack Obama’s election, given that it evoked both a sense of togetherness and extreme opposition, will lower or raise homicide rates?
A. What I predicted at conferences last year was that if he won the election, the homicide rate would drop dramatically in American cities, particularly among minorities, not just because of him personally but because of what it would say about the political system – that an African American could win the presidency. I think you can see the same thing among the Hispanic community after Sonya Sotomayor was appointed to the Supreme Court. It wasn’t just Obama – it was that all these people who weren’t African American voted for him. It’s an empowering thing.
I did think homicide rates in cities would drop. We’ve tried to get the FBI’s preliminary data but they won’t give it out until next year. I also thought that if Obama won, in areas where a lot of white conservatives live, the homicide rate would be stable or go up.
So what happened? I’ve found data on 30 American cities’ homicide rates in the first six months of 2009 – three were flat. None went up, and the rest were down between 11 and 67%. If we build on this event, if Obama can reach out successfully, if we can get beyond political polarization, this could be a long-term transformative event, the way the Great Depression and World War II and the Cold War were, and the leadership provided by FDR and Eisenhower were critical. It took 25 years of effort to rebuild trust to get homicide rates down to the level they were in the late 1950s and early 1960s. Then it exploded on us again. It’s going to take a similar effort over a long period of time, and it’s going to take good leadership. Already we see vicious, unfair attacks to delegitimize the presidency and the government, even questioning Obama’s place of birth. When you are willing to go that far, it’s a bad sign for us. Where we’re going to end up, where our feelings and beliefs will end up next year, I can’t tell. I could tell in the short run because I had a sense of it and I guessed right. I’m not asking people to give up their political convictions, but if we keep up with this political hatred, and questioning the legitimacy of anyone who doesn’t share our way of looking at the world, we’ll continue to have the homicide problem we’ve had since the Civil War.
*Photo of police tape courtesy kaje_yomama.