Why Is the United States the Most Homicidal Nation in the Affluent World?

Greg Ridgeway: Okay, welcome, everyone, to our Research for the Real World. I hope you all had a good Thanksgiving. Welcome back. I’m really excited about this presentation. I went to a National Academy of Sciences meeting about six months ago. Both Professor Roth and I are on the crime trends panel there. NIJ has been really interested in the issue of crime trends. It’s interesting that it’s one of the top-tier, most important questions in criminal justice that we just don’t have an answer to: What makes crime go up and down?

I talk a lot about other fields that have already sort of addressed their top-tier issues, like in dentistry we sort of know what causes cavities and that we need to brush to prevent them. That top-tier question is resolved. In agriculture they know how to grow drought-resistant wheat and drought-resistant soy beans and now they’re moving on to Roundup-resistant soy beans and Roundup-resistant corn. So a lot of those top-tier questions are answered in other fields, but here we are in criminal justice not knowing what makes crime go up and down.

So NIJ, about a year or two ago, started this idea of “let’s really figure out and try to answer this question.” So we started a roundtable at the National Academy of Sciences to address this question. So both Professor Roth and I attended the first meeting, and it was the first time I had heard Randy talk. And I was just fascinated by the discussion. So much of my thinking has been thinking, “Well, what’s made crime go up and down right here in the United States over the last, say, 30, 40 years?”

And he just brought a much more expansive view, from around the world, through different time periods, and in the abstract to today’s talk you saw that he would be talking about the last 450 years of homicide data. And I hope for you, as it was for me, this will be sort of mind-opening and fascinating.

So, a little bit about Professor Roth: He is a professor of history and sociology at Ohio State University. He is a fellow of AAAS, the American Association for the Advancement of Sciences.

His best known book, American Homicide, came out just a couple years ago, maybe 2009? And that won the Michael J. Hindelang Award, which is given out by the American Society of Criminology for most outstanding contribution in the field. So I’m really looking forward to his talk today, and I hope you get as much out of it as I do. Without further ado, Professor Roth.

Professor Randy Roth: Thank you very much. I’ve got to say that since my book came out in 2009, a lot of very nice things have happened. But the coolest thing is I actually made it onto the History Channel for an entire minute. I can’t tell you how excited I was. We don’t have cable, so I’ve never seen the show, but I know that you’ll want to run out and see the show because my mom said I did a really good job. So, you know, as long as you don’t embarrass Mom, you know you did what you needed to do.

I’m real excited to be here, and thanks so much. I mean, I follow your work and I’m just excited. I appreciate the dedication. All of us have been working on these projects on crime. And I think the one thing when we talk about the cost of crime, we’re quite aware of it in a way that most Americans aren’t actually, because many of us count. You know, we count things. So when we talk about the burden of murder or the cost of homicide, when we look at a rate, say, that for two-thirds of the 20th century you’re running a rate of 9 per 100,000 per year.

And people think, “Well! Nine out of 100,000 get murdered  —  that’s kind of small.” But what we know is that that’s a homicide rate; that’s not a homicide risk. You’ve got to multiply that figure by your life expectancy to really find out how likely it is you’re going to be murdered. And so when you look at a rate of 9 per 100,000, what we’re really saying if that’s sustained over the course of your life, right now with life expectancy is up around 76 or 77 years, that’s a lot of death. That means that if we sustain that rate, it means 1 out of every 140 children born in this country will be murdered.

And the way those rates work out if you project them for various groups  —  and, again, I’m using the categories that say the National Center for Health Statistics which is white, non-white, male, female. We know with those rates, 1 of every 460 white females would be murdered, 1 out of every 160 white males, 1 out of every 110 non-white females and 1 out of every 27 non-white males. Those rates are tremendously high, and even if we look at the rate where we’ve been — 5 to 6 per 100,000 — we’re talking about nearly 1 out of every 200 children born in America is going to be murdered. And that’s a tremendously high burden.

And we are aware of that, and I hope that someday we’ll start to talk about those risks. We were talking about mass murder, and that that tends to be a crime that white males commit. It was a teachable moment back in late last year for my criminal justice history class. So I got an entire week behind; we spent an entire week talking about mass murder. I said, “What is it” — we were talking about depressed males — I said, “What is the suicide rate for white males?” Well, it’s been very stable at 20 per 100,000 per year. And what that means is, I told my students, that means 1 out of every 65 white males in our society kills himself.

How hard is it to recruit out of that group some mass murderers? That’s a huge group, a huge reservoir of people that you could recruit somebody depressed enough to do something terrible. So I think when we talk about the cost, it’s tremendous. And what’s important to know about the United States is it hasn’t always been that way. If you take a look at the period after the Revolution, say between the founding of our country — the 1790s, when we achieved political stability after the Revolution — and look down to the Mexican War, the homicide rates in the North and the mountain South were probably the lowest in the Western world.

It’s a stunning thing to think that a society that has a reputation as being the most violent was actually at one point, we think — now that we’ve really done the numbers and done the comparative work — we used to be the least homicidal. How could that be and when did we change is the real question. And even when we look at the slave South in that period — outside of Texas, Florida, the contested frontiers — the homicide rate is moderate by European standards. That doesn’t mean that’s not a brutal, vicious, violent society, but being brutal and vicious doesn’t mean homicidal. We shouldn’t confuse the two.

In part, we see tremendous solidarity within the African-American community in that period right through Reconstruction, where they’re less likely to kill each other than their oppressors are to kill each other. And you’ll see that the economics of it is that masters were far less likely to murder a slave than employers of indentured servants in the 17th century were willing to kill their indentured servants. The most deadly labor system we’ve had in terms of murder was indentured servitude. It was not slavery.

When you think of the economics and that starts to make a little bit of sense — you following me here? — that that could make a little bit of sense. So, in other words, this is not an apology for slavery in any way, shape, or form. It’s a brutal system psychologically and physically, but do you kill the people who work for you and how do African Americans respond to that brutal repression? You see that the homicide rates shifted in the 1890s through the 1930s, when you develop the modern pattern where African Americans are more likely to kill each other than European Americans. But that is, historically in time, a recent phenomenon.

So these patterns have changed dramatically over the course of our history. What we see today has not always been. And have we been a time when a diverse, robust, dynamic, contentious society has had a low homicide rate? Absolutely, it’s possible. So I don’t want to — I see a kind of optimism in this. It doesn’t mean because we’re diverse, because we’re different, because we have a lot of immigrants, because we’re racially diverse, that we need to have a high homicide rate. That is not inevitable.

But we have to think about the complexity of these changes and why these homicide rates change so dramatically over time. That’s what we do. And so I look not just back over the last 40, 50 years — if you want to go back to the Middle Ages, I’ve been working on the Middle Ages, we can go back 1,000 years, that’s okay. Just ask anything you want; I’ve been working on that. Now, one of the first things — and I changed what I wanted to do in my talk a little bit.

So I’m going to talk about child murders by parents or caregivers to begin with, because one of the first things we do as historians is we disaggregate. We say, “Are the causes of intimate partner violence different from the causes of murders of children by parents or caregivers? Is it different from homicides among unrelated adults?” And they are, historically. If you look at broad periods of time, you can start to really disaggregate these patterns. So I want to talk about child murders up, say, from the 16th century through the 19th century. That’s a little odd way to start, but please bear with me, okay?

What we see when we look at those is a very straightforward pattern. And everywhere I’ve looked, these patterns are the same. I could talk about the methods that we use, but we’re not only using our historical epidemiology — looking at coroners’ records, newspapers, jail records, diaries — we look at everything we can. And we use our mathematics because we have independent sources, somewhat independent sources. We can actually estimate the number of these crimes that came to the attention of the public that didn’t end up in the surviving record. So we’re not just going with counts of many times; we’re going with estimates of those that came to the attention of the public.

And we also now have forensic archaeology. We can’t study soft tissue injuries. And many murders of young children, particularly infants, are soft tissue injuries. In other words, this is strangulation or suffocation. It’s not a bone thing. But what we can look for in the historical record is signs of child abuse. One of the tells for that that a forensic archaeologist will look at is repeated twisting spiral fractures of the arms or legs. One of those can happen to anybody — skiing accident, falling down, it could happen. But when it’s repeated twisting spiral fractures, it means this child was being thrown, being tossed. And what we find is, where we have the bone studies, as we do in the 19th century from graveyards in, say, England and the United States, it maps out perfectly with the historical epidemiology that I’ve been able to work on. So in other words, we’ve got bones working together with our historical epidemiology. So we’re starting the field with these multiple sources that we’re finding out how child abuse has changed over time, and it has changed dramatically over time. What is it? Everywhere I’ve looked it’s the inverse of the birth rate.

And I was stunned to find this. When the birth rate is high, the rate at which children at all ages, from neonates to infants to older children aged 15, are murdered at a lower rate. When that birth rate’s high, essentially the world is welcoming to young parents and their children, that murder rate is low. And when that birth rate goes down, either because of economic pressures or rising ambitions, that’s one of the things about the dynamics of it. If my ambitions are higher, the cost of that children, that child is more for me, isn’t it?

In other words, so if my ambition is going up, those children can be at risk if they come in the wrong number at the wrong times with the wrong qualities. So what we see is that it’s the inverse of the birth rate, and it measures out, it correlates with every measure that we have of the well-being of children and their parents. In other words, the murder rate of children is correlated inversely with life expectancy. In other words, when children grow up to be tall because they’re well nourished, when they have a longer life expectancy at birth, they’re less likely to be murdered.

What we see is that when there’s a high percentage of women who were able to marry at least once, the murder rate is low because they can form families. One of the things that’s surprising is when the premarital pregnancy rate is high, the murder rate is low, because it means that people aren’t being as cautious. It means there is more hope, if you get pregnant, so what. And we can see in periods when they re-stigmatize births out of wedlock, like in the great spiritual awakening of the 18th century, you see a bump over those 15 years in the number of neonaticides. When you re-stigmatize that sort of thing, young women feel the pressure to kill newborns.

So, in other words, we’re seeing a pattern that’s extremely robust. And it even comes down to the present. If you take a look at the new studies that are being done by my colleagues at Nationwide Children’s Hospital in Columbus, we’re partnering with the Children’s Hospital in Pittsburgh and with law enforcement and with social services to really get a more accurate look at child abuse. You can actually, they said, “We knew when the recessions of the mid-2001 and the recession of December 2007 started before the economists did. At least half a year to a year before the economists knew we were in recession, we knew we were in recession. Why? Because you see it in the battered bodies of children.”

If we really want to use our knowledge to make money in the stock market for most of the public employees on a public pension plan, and we know what’s happening to those, we could actually use this as an investment group, I’d hate to say it. But we could actually look at that and we could actually make money in the stock market by following child abuse, because it’s a leading economic indicator of familial distress. And when we think about this biologically, it starts to make sense, too. We’re trying to link not only the historical epidemiology with the contemporary social science of medical science, but when we think about this biologically, what happens?

When we think about the first two years of life, it’s the mothers who are most likely to commit these murders. And we know what cortisol does, what a high level of stress does to a human body. What’s interesting is when a young woman is feeling stressed and has a high level of cortisol in her body, which can come from chronic stress in utero that she suffered as a child. It could come from a damaged relationship, economic stress; anything can raise your cortisol level. What happens to your body, the science shows, that you are actually less likely to conceive a child. In other words, that high cortisol rate is telling your body, “This is not the time to have another child. Focus on the ones you have or wait.”

If you have a high level of cortisol in your body, you are three times more likely to miscarry in the first three weeks of pregnancy because your body is telling you, “This is not a good time.” And we find that high level of cortisol interferes with your ability to bond with your child. If you have a normal stress response, when your child cries you are suddenly attentive, alert and engaged. But if your stress level is here, that noise comes one level more agitation, and you can’t bond. So, in other words, what we’re seeing when you look at the stress that young parents and families are under at certain periods of time, and think of what that does to their cortisol level.

And this is very common, interestingly enough, among primates, and here I’m getting into evolution. It’s very common among primates like us who are what we call collective childrearers. In other words, there’s a group of primates, one out of five primates, where the mother cannot produce enough calories on her own to raise a child. In other words, it takes 13 million calories or something to raise a human child to maturity. That’s a lot of calories. You need help, you need significant others, you need a community, you need help. And this is the same for tamarins; this is the same for marmosets.

And they’re, interestingly enough, the great colonizers. When there’s space out there to colonize, they take over. We take over. But when there’s not space there, we cut back, we limit our family size. And so what you’ll find is the tamarins and the marmosets respond in the same way that human beings do. When there’s not enough around there, when the litters are too high, they commit maternal infanticide, neonaticide, whereas those other primates where the mother can produce by herself enough calories, there’s very low levels of maternal aggression in the first year or two of life.

Does this make sense? You see what I’m trying to get at. So what we’re trying to do is pull together the biology, the historical epidemiology, the forensic archaeology, the medical science, contemporary social science to try to build an understanding of how these things change over time. And we’re starting to feel now that when it comes to violence against children by parents or caregivers, that we’re getting much closer to actually having the answer. So I think we’re really making progress. But where do we really stand out?

What’s stunning for the United States is, if you look in the 18th century, the 19th century for African Americans and European Americans, our children may have suffered the least violence in the Western world. In part that’s because for African Americans, raising that child in the face of oppression is our way of resistance. You can see that in the literature, you can see that in the testimony, you can see that in behavior. “My goal in life is to keep my child alive so someday they will be free.”

You see that kind of investment, you see that kind of collective investment in childrearing, so even though those children are undernourished, because the slave owners do not allow them a sufficient diet, their growth is stunted. There’s high rates of stillbirth; there’s high rates of early infant death because of that nutritional deprivation. This is a brutal system, but in terms of parental violence, it looks like those levels are very low. And when you look at European Americans, because of the tremendous economic opportunities for self-employment in the 18th century, the early 19th century, we had very low levels of violence against children.

Now what happens to Native Americans who lose their land? You take a look at the Native Americans in New England when they’re tremendously under stress, when they really finally lose all their land in the 1740s, ’50s and ’60s. Their rate of child homicide is 20 times that of African Americans and European Americans in New England. So in other words, that kind of stress does a lot of damage. So they’re on the wrong side of history. But for those other people, they’re not committing those homicides. So we used to be one of the best places for a child, at least African American children, and European American children to be.

But what’s happened to our ranking internationally since World War II? It has really gone downhill, whereas right now I think among affluent nations we have the third highest rate of child abuse and child murder. Things have gone in a different direction, and we have to think about what that is, what exactly the plight is for young parents and children in this society versus others. So we can talk about that somewhat.

And if you have questions, feel free to break in. I’m totally happy to be interrupted. I’m a teacher; I’m used to that.

So, but now what I’m going to do is switch over to really talk about homicides among unrelated adults, because that’s what we want to think about. You look at that chart there. What do you get? Look at that giant hole in the middle of that chart. Why is that? And we’ve got another hole in it. We’re still have, by far — even where we are right now at about 5 per 100,000 — we still have by far the highest homicide rate of affluent nations. And we’re looking at a rate that has been fairly stable, really, since 2000. It’s really not a constant drop. We had a sudden drop in the 1990s and we’ve been pretty flat since then. So we’re trying to figure out what happens to these dramatic things — you notice how rapidly it goes up and down? Like, you can mark this — that first drop, 1933. Then 1934 is actually when it went down. It’s the second year of FDR’s presidency that it drops. And it keeps dropping pretty much, except for the little postwar peak, right on down to 1959, 1960, ’61. And then you see this sudden surge in the mid-1960s boom. And this is typical — that is dominated by murder rates among unrelated adults. That’s what, really — those rates, unlike child murder rates, move jaggedly. When you look at the murder rates of children by parents or caregivers, it follows these long, smooth curves that take hundreds of years to inflect as fertility behavior changes.

But these rates can double or triple almost overnight and then drop suddenly overnight, murder rates among unrelated adults. It’s a very different phenomenon. And if we look at intimate partner violence, what I can see is the major shifts in that happened in the periods, the decades when the shift in the economic balance of power, in the cultural balance of power, shift towards women: the 1830s and ’40s, the 1960s and ’70s. And you find in those periods of tremendous shift towards empowerment of women, towards greater equality of income, towards more opportunities outside the home, you find that even though the typical relationship becomes, I believe, less violent and more satisfying, that minority of men who can’t make that transition, who can’t accept that transition to new way of life, become more violent. And so you’ll see less day-to-day abuse, but on the other hand more lethal violence by the men who can’t change. So those are different patterns. We could talk about that more. But this is homicides among unrelated adults. What seems to be driving this? And that’s the big puzzle. And one thing that is really difficult here, and I talk about this a lot, is that the causes confound our political intuition as conservatives and liberals.

I certainly am one who thinks that both conservativism and liberalism have made valuable contributions to human progress. I deeply believe that; I engage in nonpartisan politics on a local level; I’ve campaigned for Republicans and Democrats; and I believe in this sort of — I wish we could do things more like that that we work on the local level. But can liberal and conservative ideology explain why murder rates go up and down? I would argue the real reason we have trouble understanding this, Greg, is because most of us are liberals or conservatives. And that we are so deeply committed to our ideologies that we can’t see — we can’t accept failure. We just can’t accept failure.

So if I try, if I were a liberal, do I believe that it would be a better world if we had full employment, if we end discrimination, if we have more empowerment of people, would that be a better world? Absolutely, yes. Would it necessarily drop the homicide rate among unrelated adults? No. In other words, full employment — where were we towards full employment? In the 1960s. We were at full employment right to ’73. What happens to the homicide rate? Poverty rate was dropping because of the Great Society programs. There’s lot of misrepresentations about that, but the number of people in poverty declined.

Of course, it’s primarily among senior citizens that it declined, but it still declines. There’s more help for poor families. Full employment and, boom, what happened to the homicide rate? And look what happened in the 1930s. Are things rosy? Did the New Deal solve the unemployment problem? We know it didn’t, despite its efforts. But the homicide rate went down. That makes no sense in terms of liberalism, does it? In other words, you look at that chart and you think about it as a liberal and you have to confront a certain fact. Your ideology can help build a better world — I believe that — but doesn’t solve the homicide rate.

And the same thing goes for conservatives. Do I believe that swift and certain punishment would create a better world? I deeply believe that. Do I believe that law enforcement is essential? When you look at what happens in societies where law enforcement breaks down, the homicide rates reach catastrophic levels. But can a strong policing drive homicide rates down to very low levels? No, it takes more than that. Law enforcement can cap those levels at about 10 per 100,000 a year, but beyond that, law enforcement needs the help of the society. It needs more than it can do, so it doesn’t line up.

And when you look at the 1950s and ’60s, who had the most people in prison? The United States did. Who had the most personnel in law enforcement and criminal justice? We did. What help did that do when the homicide rate went up? Nothing. And you see tremendous — in the ’70s and ’80s — you see that sudden surge in investment in law enforcement, imprisonment. Did the homicide rate go down in the next 20 years? It didn’t. It went down later. So when I take a look at these sorts of arguments, they just don’t fit the data. And so what we have — the toughest thing that I have to sell, because one of the great things about bringing this message is I find I just get attacked by everybody. It’s so wonderful.

And I don’t mean this as an attack, but man, people say nasty things. You’re in Washington, you know how it goes. But the thing is is that our ideologies fail and that’s what I try to tell my students, you know? Our ideologies fail us, and what we try to do in college is not only ask our students to overcome their ideology, we ask ourselves as faculty to overcome our ideology, to try to say look at the evidence, be an empiricist, and it doesn’t mean that your ideology is immoral. It doesn’t mean that your ideology doesn’t have important things to contribute to human progress, but it means you’re wrong about crime, damn it. And you’ve got to get that through your heads.

And so, in other words, you know, when we have these debates about all these laws and things, I never get asked by any politician to speak to it. Why would they? I’m going to tell them, “Well, you know, it just doesn’t work — the world doesn’t work the way you say it works.” No one wants to hear that, and I think that’s the plight you’re in, right, at the NIJ? You do all this work and then it kind of goes into the publications, and then public policy’s a hard thing to affect, isn’t it? That’s why I got into local politics, I’ve managed campaigns. And I find if you manage campaigns that elect your friends, you deal direct. So this is it, so of course as public employees you can’t get that involved, but it’s something to think about.

So when I look at broad patterns over long periods of time, what do we see? What we see is just, you know, we see all these kinds of trends that I talked about. Here’s our whole thing; does policing, incarceration, do abortion rates fit? Well, abortion rates don’t fit. Why? Because all you have to do is go back to the 1850s. Abortion rates soared tremendously in the 1840s, ’50s and ’60s. Beginning in the 1830s it starts to go up and I can do that through abortion-related deaths; I’ve actually got abortion-related death rates, and they’re stunningly high. And the fact is is when did we surge in our murder rate as a nation? The late 1840s and ’50s and ’60s. Exactly when those children who were first being aborted started to come of age. So, in other words, it doesn’t work. It just doesn’t work when you look at it historically. When we look at alcohol consumption, here’s my favorite one, you know, think drunkenness. When were we the most drunk in the United States? When did we — we actually know this. By God, we work hard as historians.

When were we — in fact, one of my colleagues wrote the wonderful book on this, Bill Rorabaugh, and the title of his book is The Alcoholic Republic. The time when it peaked, we drank — right after the American Revolution down to the 1820s, we drank three times as much ethanol per year per capita as we did today. Think of the horse and buggy accidents. Think of if they actually could drive. It would have been horrible. Lots of firearms accidents, you know? They especially drank on patriotic holidays. So they’d get out for the militia training days and these are young, drunk guys. They’d actually — your officers — the goal is, is if you’re elected officer, you have to treat the men beginning at 6 in the morning. You have grog and you have whiskey, and they’re drinking all day with their firearms. Yeah, they shoot each other, they shoot into the crowd. Their girlfriends are there to watch the parade, they shoot their girlfriends. They’re totally drunk. And so, in other words, this is an alcoholic republic. That’s our lowest murder rate in American history. And, you know, of course, as my students say, that’s because they were too drunk to use a muzzle-loading gun. You know, maybe that is.

But the thing is is you look at these substance abuse patterns and they don’t line up the way you would want them to either. So, in other words, all these theories just, you know, they don’t work. So, this is version 12B. I’d have to say that every theory I had died a horrible death in the face of the evidence. My book took me 25 years to write. I’m glad I had tenure, I would have been fired. Everybody assumed that, “Well, what’s Randy doing? He’s not doing much of anything.” Well, I was being terribly confused and I couldn’t quite figure this out, but a number of us independently in the mid-’90s started to come to this conclusion.

Manuel Eisner, who teaches at Cambridge, Manuel started to see this; Gary LaFree at University of Maryland started seeing it; the late Roger Gould, Yale, he started to see these patterns. We each started to realize we were looking at the same elephant. We were touching it from different points of view. But all of us were independently thinking along these lines. And I came to the conclusion that if I looked at these homicide rates, I’d be looking at political stability as the key. The belief that government is stable and that its legal and judicial institutions are unbiased and will redress wrongs to protect lives and property.

In other words, political stability seemed to be critical; feelings and beliefs towards government seemed to drive things more than the economy or policing or imprisonment. Secondly, a legitimate government, a feeling of confidence in government and the officials who run it and a belief in their legitimacy. Fellow feeling, the third thing, empathy, a feeling that patriotism, empathy and kinship with other members of society are rising from racial, religious or political solidarity. And I think a legitimate status hierarchy 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.

What I’m saying is is these, if you think about these, are the attributes of successful nation-building. If you want to have a successful nation — we talk about nation-building in Afghanistan, we talk about nation-building in Iraq. What are we trying to achieve? That’s what we’re trying to achieve, isn’t it? That’s what we’re trying to achieve. And this is not always a benign process. In other words, we can think of ways to build a strong sense of nationality as we did after our Revolution in the northern United States by empowering, including, opening up opportunities, building togetherness, building patriotism in a powerful, strong way.

But you can also create a strong sense of nationality like Napoleon did, like Mussolini did. When they came into power and got things engaged, did the homicide rate go down? Yes, it did. One of the horrible things about nation-building is sometimes the best way to drop a homicide rate is to be on the winning side of a race war. You know, the winning side stops killing each other in those circumstances. I don’t want to know that about human nature. I didn’t want to know that, I still don’t want to know that, but that’s the truth of it. We know that nation-building is not always a moral process, don’t we? Not all one that we would cheer, but in many cases it has been.

And so that’s what we’re trying to think about. And when these components are in place, homicide rates can get down to 1 or 2 per 100,000 per year. Very low levels. But, boy, when these are not in place, when the bottom three are not in place, you get up towards 10 per 100,000. And when that top one is missing, political stability, you can go up to 100 per 100,000 a year, or even thousands per 100,000 per year. Yes?

David Marimon: Do you mind if I ask a question?

Roth: No!

MARIMON: When I look at this, all I can think is that we’re heading for more murder because we have less political stability — the legitimacy of the government, yes, but there are plenty who do not see our President as legitimate. There is no empathy or patriotism, and we are not where we were at 9/11.

Roth: Okay, I will get to that because, you know, we’re getting ahead here, but in a divided society, legitimacy is perceived differently by different people. I went out on a limb on the History News Network, you can look it up, what would happen if candidate Obama won the election to the homicide rates in the United States? It said right after the election the homicide rate in American cities is going to drop like a rock right after the election, because of what this historic moment means to African Americans and other minorities.

And I thought there would a broader, feel-good feeling in the country at large about doing this, but I said I’m deeply worried about what’s going to happen in those places that are most resistant to the idea of an African American President. And I was very worried what would happen to homicide rates, particularly in parts of the South, the former Confederacy. And when the statistics came out — we don’t have the right statistics yet — but the global statistics, that’s exactly what happened. Fifteen percent drop in homicide in American cities right after the election. By February, we were down 15 percent in American cities, but in the upper South, where you see that most resistant, where the birther movement turned out to be, what happened to homicide rates? It was up 10 percent. It went the other way of the nation as a whole. And you take a look at the opinion polls for African Americans in 2009, despite the depression — because all my friends said, “Homicide rate’s going to go up because the economy’s doing badly.” I said, “No, that’s not what drives the homicide rate.”

And I said what this election would mean to African Americans, and you look at those opinion polls in 2009: African Americans were more optimistic about their future and their children’s future than they had been in a generation. Well, duh. What do they see every day on TV? What are young people seeing? They’re seeing a black President and First Family there doing the best they can, a dignified job. That’s very important, and it had a powerful effect, I believe, on young African Americans. And since they commit most of the murders in this society, that had a powerful downward effect. So certainly, but what does it do to those groups that feel disempowered?

They’re more likely to kill each other, and that’s the irony of all of this. I don’t want people to kill people that they don’t like politically, but the fact is you kill your best friend, you kill your acquaintance, you kill the people in your own neighborhood when you’re mad. And so the theory here is, as you know, it’s about emotions. In other words, if I feel included, if I feel empowered, if I feel that I matter, it doesn’t mean that I get my way on everything, but it means that I matter to the society, that society somehow has my back. And I feel a kinship with people I don’t know really well beyond my friends.

If I get challenged, somebody disrespects me, it can roll off my back. But if I feel disempowered, disincluded — disenfranchised, every little slight matters and I can react violently. And so that’s the kind of psychology that we’re looking at here. And when we take a look at this, this goes back to — we’re getting a little bit ahead — but we look at the biology of this. When we think about how chimpanzees behave — I get into a lot of trouble with this, so I’m just going to say it, I’m going to talk about evolution and, by God, that’s it.

Someone spliced this together to try to turn people against me. This is all, I mean, really nice; people spliced me so that I call certain political groups chimpanzees. They did that to me on YouTube; I really appreciate that deeply, that level of dishonesty. But, you know, this is what happens. But if you take a look at what happens when there’s political instability in a chimpanzee troop, in other words their dominant structure is in play. What they found is that the levels of testosterone go up in both males and females, the level of serotonin goes down and they start hitting each other at a rate three times, four times they would otherwise. And they don’t just fight for position. They hit their friends. They’re beating everybody up. And in fact, you can manipulate their diet so their serotonin goes down. If you want to have a high level of serotonin and feel good about life, eat a lot of ground turkey. Tryptophan, they’d eat a high level of tryptophan. And what they did is they created a — they didn’t tell the observers watching the chimpanzees that they had lowered the level of tryptophan in the diet, which you need to synthesize serotonin. And the observers are saying, “Gee, the chimpanzees are hitting each other at a horrible rate, three or four times!” And then they put the diet back: “Gee, they’re not killing — they’re not hitting each other; what’s going on?” Well, it’s about those hormone levels.

And what’s interesting is when you have political instability, the hormones that facilitate, prepare us for aggression, testosterone goes up, the serotonin goes down, which — serotonin, you know, prevents impulsive aggression. And you see it go up. And what happens when the dominant structure comes back? Serotonin goes up, testosterone goes down, and there’s peace. You take a look when that dominant structure’s in play, who goes to the top when it’s most violent? It happens to be the males with the lowest levels of serotonin and the highest levels of testosterone.

But once you have political stability, which chimpanzees come to the top? Those that have moderate levels of both. In other words, they’re not generally aggressive, but they can’t be imposed upon and they happen to be the best coalition builders, the best at befriending people, the best at reaching out to other people. So when I look at chimpanzees and I look at us, I’ll show you how we behave. I go, “Oh, that’s who we are.” There’s a lot of these things that we could start to see in the way that we behave. So let me show you big patterns in the past, and then we’ll talk more about the present, okay?

But — and, again, everybody wants me to say what’s going to happen. You know, what happened just the last two years as the bloom is off the rose? What’s our 2011, 2012 [inaudible]? Starting to go back up. I said — 9/11: People said the homicide risk is going to go down, and I said no. And it didn’t; it was flat. So, in other words — and I said child murders would go up with the recession, but the others would go in a different direction and they did. I mean it’s sort of like what do I have to do? I mean, I just, you know, I keep saying this is what’s going to happen and it happens, for goodness sakes. I’m trying. So anyway, let me go on and do this first and then I’ll get back to you.

Okay, here we have France. So let’s go outside of the United States. Look at political instability. If you know French history in the 19th century, when would you expect the homicide rate to spike? 1830, ’31; 1848, ’50; 1870, ’71, right? What does it do? It does that, and notice what this is. This is not Paris, where the political violence is. This is in Corsica. They can’t even speak French. All of this is honor violence and political violence, and they respond exactly the way people do in the capital. This is the same thing I found in the 19th century: When the homicide rate goes up in New York City, when the political system starts to break down after the Mexican War, it goes up in New York City.

But it doesn’t just go up in New York City. It goes up in Amish and Mennonite country in Ohio! In other words, this is felt so deeply across. Of course the homicide rate’s lower in Amish and Mennonite country in Ohio than it is in New York City, but you watch them going up and down in synchronicity and the question is how do they know? How would Corsica know that this is the time to kill people? This is what happens. Political instability has tremendous political effect on our bodies, on our minds. And so when that state breaks down, this happens.

And when we take a look at lower homicide rates, I’m going to show you some things about — I’m going to try to measure political instability by looking at death rates and riots and rebellions. How many people have died in the United States since 1992 in a riot? 1992 was the last year we’ve seen any significant loss of life in a public demonstration. And when did the homicide rates drop? Right after ’92. And so I look at this sort of thing, and when did we start dying in riots? The mid-’60s, right? Right when this went up was ’64, ’5.

And so I look at this, looking at measures of political instability, and I’d say that despite the fact we’re politically polarized, we’re not killing each other over politics right now, are we? We aren’t. Thank God, you know. And except for a few terrorists like Timmy McVeigh, you know, and Terry Nichols, we’re not doing this. In other words, we’ve made a decision that we don’t want to go back to that. And I think we have to talk about seeing more progress in our society than we’re willing to give ourselves credit because of this political polarization that we have. It’s bad-spirited, but it’s not necessarily a murderous kind of aggression. Does that make sense?

So that is what I tracked out, so I’m looking — I really look for dead bodies.

Here is Great Britain. If you know your British history, this makes tremendous sense. There was a contract really executed during the Napoleonic wars that the British elite made it clear: “We don’t want to go the way of revolution in France. But once this war is over, we’re going to empower and we’re going to include and we’re going to change.” And what happened at Peterloo in 1819 was the democracy demonstrators were killed. Oppression of the democracy demonstrators.

And if you take a look at what happens to the homicide rate in Great Britain from the 1820s through the 1860s, the period of Chartist protest, upset with the lack of inclusion. You see the homicide rate in England and Wales was higher than in the United States right into the 1840s. It’s higher there. When does it drop radically? See the radical drop right here? That’s the Second Reform Act, 1867, which enfranchised propertyless household heads in the cities. The second big drop, suddenly: the Third Reform Act, which enfranchised propertyless household heads in the countryside. In other words, enfranchising, including, legitimizing suddenly dropped the homicide rate by 50 percent and then by 50 percent again.

You’ll see that what many people keep trying to drive what they call a “civilization thesis”; they say, “Oh look, that’s a gradual decline.” It’s not a gradual decline. It’s blam, it’s blam. It’s very jagged, and it revolves around politics. So that’s what we’re taking a look at. And if we take a look at how high homicide rates can be, let me just look at this one here. Let’s take a look at what happened in the former Soviet Union in — with a breakdown of the state there, the failure of the Soviet Union. Look at how those homicide rates spiked. You’re getting up, look up here, you’re getting in the Russian Federation and in Estonia, you’re getting homicide rates 25 to 35 per 100,000 per year. Disastrous and catastrophic. And everywhere you see them spike.

What is interesting about this is that the rates spike most dramatically where? Where there’s a high percentage of ethnic Russians in the population, of people who saw their political system and their place in the world shattered. In other words, Estonia is 50 percent ethnic Russian; Latvia is a third ethnic Russian; but Lithuania is only 9 percent and it’s down with the others at 10 to 15 per 100,000. It’s only those three places with a large number of ethnic Russians who saw their world shattered that went up to 25 to 35. And if you take a look at this, every one of these places that made a swift and smooth transition to democratic capitalism went back to a low rate very quickly: the Czech Republic, Slovakia, Slovenia.

So, in other words, what you’re seeing there is politics being played out in day-to-day murder rates, and it has to do with these questions of solidarity. Okay. Question, go ahead.

DIANA TURONIS: How does this compare to the child-on-child murders?

Roth: Child-on-child murders. That’s a good question. What I find is that, in other words if we’re talking about an unrelated child killing an unrelated child, it follows the patterns of murder among unrelated adults. In other words, they mimic the behavior of elders, and I think I can even see about an eight-year lag. In other words, once the kids start to see the adults doing it to each other, they do it. So, in other words, you see that rise in the late 1840s in the homicide rate among unrelated adults, and you’ll see the child-on-child murder rates going up in the late 1850s and ’60s. They start to move it.

In other words, you’ll see one fellow — one young man was upset that his — one 6-year-old boy was upset that his 8-year-old sister in Vermont was playing with an Irish girl. Irish are dirt. He told the Irish girl, the little 8-year-old girl, to get off his porch. “We don’t want any damned Irish on my porch.” She said, “Well, your sister said” she could stay here. So he went inside, got dad’s gun, blew her away. That’s what happens. In other words, he’s imitating the behavior he’s seeing in adults, and one of the things that I tend to say if I have to put it simply: What youth homicide problem? It’s an adult homicide problem, and the kids imitate our behavior.

Sometimes the trends can go slightly different, as they did in the ’80s. You know, it’s going up faster for the children, but again there’s that lag. So, in other words, they’re seeing what adults are doing to adults and they mimic that behavior. If adults behave themselves, the kids do.

TURONIS: Has it been normalized for the size of the population?

Roth: Oh, I do everything by rates, absolutely. And I do age-specific rates. Every time I can get an age-specific rate, I do. I try to normalize it. I’m a — I love math.

TURONIS: So the most dangerous age bracket becomes — ?

Roth: In the 19th century, the most dangerous age bracket is in their 30s. You know, homicide, and it’s one of the problems with the whole homicide spike in the ’60s. You know, it’s blamed on the baby boomers. The baby boomers were too young to be responsible for the spike between ’65 and ’72. This homicide spike was being done by people well into their 20s and 30s. And they were born before the baby boom. So in other words, there’s a lot of this stuff that’s just mathematically — what can I say. It just frustrates me how badly people will look at that data, because they want to have a demographic understanding. It’s simpler to say it’s hydraulic, the number of young people in the society. That’s an easy way to think about it. It’s compelling, it’s common sense, and it’s wrong. But, you know, arguing against common sense is as hard as arguing against ideology. It’s just, you know, you, just — anyway, I’ll stop.

          I get upset about it because I think we really have knowledge that is very valuable here, and that we just can’t get it across to people, because of the way they want to think in common sense. Does that help? Is there another question? Yeah, go ahead.

NAZGOL GHANDNOOSH: Hi, my name is Nazgol Ghandnoosh. I’m from the Sentencing Project. And I wanted to ask you about when you were first talking about parental homicide rates, you seemed to be really focused on economics and perceptions of economic stability.

Roth: And ambition, yeah.

GHANDNOOSH: And then in focusing on the unrelated adults, you made a shift to politics.

Roth: Yeah, all of that’s irrelevant, yeah. It doesn’t correlate.

GHANDNOOSH: Even perceptions of economic stability?

Roth: It doesn’t. We’re trying now to see if the consumer price index — and this is Rick Rosenfeld’s work — because we’re trying to find a better measure. We used to have a wonderful measure of trust, and I’ll show you that, because people used to say, “Do you trust the government to do the right thing most of the time?” This is in the Michigan surveys. And, “Do you believe most public officials are honest?” People answered that question in a deep sense, not in a partisan sense, right up until this political polarization really hit in 2008, ’9.

And now that question that we used for years is totally fouled, because it’s a way of saying, “I hate Obama.” And what we find is that this does not track with presidential approval ratings. It’s something deeper about how we feel with our country. So in other words, the fact that Obama administration is right now very unpopular, including with Democrats who are so frustrated they didn’t hand this whole thing to Amazon. You know, why when we have the best people in the country — and for goodness sakes, the people, they’re Democrats, what are you thinking, you know? Put it with people who give a darn whether or not you succeed.

So in other words, everybody’s angry right now. But that’s not going to translate into a higher homicide rate, because that’s not how we feel about our country; that’s how we feel at this moment about something. So you’ll watch, I’ve actually done the homicide rates back to 1938 against the approval ratings, the presidential approval ratings, Gallup poll — very low level of correlation. But if you take those trust-in-government issues — I’ll come to that — very high level of correlation. Because it has to be something more deeply about how we feel about the country.

And as I said, since ’92 we were killing each other in political demonstrations from 1964 to 1992, and then we stopped. We weren’t doing it in the ’50s or early ’60s. And think about that. That’s what I look at as a historian, so I look at those deep things and we get caught up in the moment and think, “Oh, that’s how we’re feeling.” Well no, look at our behavior. Our behavior suggests that we believe in the legitimacy of the system. And what you can see across the world, just to get ahead of ourselves, across the world homicide rates have dropped everywhere in the affluent world.

What’s the turning point? 1988 to 1992, the end of the Cold War. What did the collapse of communism say about capitalism and democracy? How many people envisioned a different world than democracy and capitalism? The alternatives have died, and what’s interesting is when you look at how much more popular capitalism is right now in this country than it was in the ’60s and ’70s, it’s amazing to think about these deep changes. And we don’t — when we are in a position of political polarization, we often don’t see the deep story. That’s what I’m getting at, okay?

Let me show you how this works, then, in an evidentiary sense. If I take a look at the homicide rates in the 17th and 18th century of the United States, what you’ll find is that our homicide rates were fairly high right until the mid-1670s, and they drop suddenly between the mid-1670s and the early 1690s. They drop to a third of what they were before in Maryland and Virginia. They drop to a sixth of what they were in New England. So suddenly you see this sudden drop, and so we become less violent suddenly in the late 17th century and then what happens when the Revolution comes? You come to 1765 and the Stamp Act, it starts to go up again. Stamp Act crisis. It peaks during the Revolution and then as we achieve political stability, it goes down again.

I was trying to figure out, how would I measure political stability? How would I measure community solidarity? So what I tried to do is, I said, “Let’s take a look at the number of deadly riots and rebellions.” How many people die in deadly riots or rebellions? And some of these are political; they have to do with contests for power. And those I’m putting in the political category. There are also deadly riots over community issues. You know, they riot to kill people they think are immoral. You know, there’s a house of prostitution, they murder, they riot, and they kill people there. There’s fighting amongst different religious groups. There’s fighting between the Tories, you know, the Whigs; that would be political. There’s fighting about this religious stuff; that’s more communitarian violence. So I map that out. I look at banishments and forced exiles. In other words, “You’re on the wrong side of this.” So when Tories are forced out and they have to move to Canada or the Antinomians are forced out of Massachusetts, forced out. In other words, we’re saying, “This community is deeply divided and they’re willing to write other people out of their community completely, drive them out on threat of death.”

We can also look at executions for sedition or treason, which I see as a measure of political instability. We can look at executions for moral offenses or religious offenses. In other words, you’re a Baptist, so we’ll hang you; you’re a Quaker, we’ll hang you; you’ve committed adultery, we’ll hang you; you are gay, you’ve committed the crime of buggery, we’ll execute you. What you’ll find is they were doing those executions, they were killing each other, doing all of these things right up into the mid-1670s. And then by the 1690s they virtually stop. And then suddenly — so, in other words, this is the period of low murder rates. Look at that. Here’s your period of higher murder rates, and for these political ones, the community regulation ones, you’ll notice it maps out the homicide rate.

If you look at the best correlate of the homicide rate in the United States from the 1640s through end of the 1920s is the percentages of new county in any decade that we named after national heroes. Is that weird? In other words, you know if we name our counties after national heroes, which would be British heroes, British noble citizens in the Colonial period or about American heroes in the American period. When we just unconsciously name people after our country and its great leaders, we don’t kill each other. And when we stop doing that, we do.

So after the Glorious Revolution of 1688, when there’s this surge of patriotism throughout the British Empire and a sense that this government is more constitutional and legitimate and Protestant, what happens to the murder rate? It drops like a rock. But what happens to the number of new counties that are named after national heroes? It goes from 40 percent to 80 percent. When does that drop? During the Revolution. It drops in the 1760s and ’70s when we start to kill each other. The new American patriotism doesn’t come in until the 1810s and ’20s, but what did I tell you about the 1810s and ’20s? It was the least homicidal period in our history, and we think whatever kinds of measures we can come up with, that’s the most patriotic we’ve been in our history. The most believing. And, again, did we have political polarization in the 1830s? Of course we did. Rambunctious democracy. But did we hate democracy? Did we hate the country? No. Despite the fact we were deeply divided politically, we behaved in a way that showed faith in this great experiment. And so we didn’t kill each other at as high a rate.

Hate speech. You take a look at the hate speech in the 17th century. It was on religious lines. Calling someone a heretic or a zealot. Or it’s on class lines. Working-class people were commonly called rogues and whores. If you take a look at this measure, we knew this as historians, as humanists, but now with Google Ngrams we can map out the use of these words. We’re using those words into the 1670s and then that hate speech disappeared.

What replaces it? Racism. In other words, we have the disastrous Indian Wars in the 1670s and ’80s, we have the rise of racial slavery, and who’s the outgroups? So, in other words, what’s going to happen is the rate at which European Americans kill each other drops with the rise of racial slavery and these destructive Indian Wars. They pull together and all of these divisions over class or race start to go away. And people who disagree with you religiously, you don’t call them a heretic or a zealot; you don’t call working-class people rogues or whores, because they’re white. And so that stops. This is not a benign process, this is not happy time. Catholics are not included. This is not a more perfect world, but it is a less homicidal world, because you’re starting to build a stronger sense of nationality throughout the British Empire to those who are included, and it’s based on race.

That make sense? Starting to see how that pattern goes? Okay, so here’s the 19th century. And again, the break point in terms of these deadly riots and rebellions is exactly the period I talked to you about. The late 1840s is when we start to spike, and if you want to look at one word that maps out the homicide rate in the 19th century, it’s the n-word. Now the n-word is used to describe not only hatred of African Americans, but hatred of whites who are allied with African Americans. And when you see the country break down in the late 1840s, the Mexican war, when the homicide rates spike, boom, the n-word. End of Reconstruction, homicide rate goes down, 1890s and 1900 when Jim Crow, disenfranchisement comes back in, all that hatred, homicide rate goes up.

And we can see the same way for the Slave Power, which is the hate word that the anti-slavery people and the pro–civil rights people used to describe those whites who they hated who were either slave owners or were sympathetic to the slave owners. Look at it; it’s the same chart, essentially. So, in other words, it maps out with hate speech, it maps out with patriotism, it maps out with deadly riots and rebellions. So we are really trying in every way — well, you know, Doug and I are trying, my friend Doug Eckberg, we are trying, we’re historical criminologists. And so we’re really trying to come up with measures of this.

And, again, what I would say is humanistic historians have known about these break points for years, but because we’re humanists, people who are scientifically or social scientifically inclined tend not to believe us when we say, “When did white solidarity become important? 1670s or ’80s. When did British patriotism flourish? In the 1690s and early 1700s.” We’ve known that, and if I had to choose between a political opinion poll or the work of a humanistic historian, to understand the feelings of people at a time, I’d pick a humanist historian any day of the week. But what we need to do as historians is try to market to scientists and social scientists and to journalists, so we need to come up with these proxies.

But what I’m trying to tell you is this really is what we’ve known as historians for a long time. I’m the one who’s trying to quantify it because I’m a quant, and I deal in this kind of world. But if we take a look at the trust-in-government polls — this is work that Gary Lafree did, in 1998 he published this. This is the homicide rate and the percentage of people who trust the government to do the right thing the most time. So in other words, the homicide rate, the dashed line goes up when trust in government goes down. Trust in government went up in the 1990s, homicide rate went down.

And, interestingly enough, divided government is preferable; divided government, you know, both conservatives and liberals, oddly enough, can feel empowered at the same time, even with all this complaining. That’s something that when you look at deep history you can see that. You know, as a historian, I look at the United States today and we’re not killing each other over politics, and I’m thinking, “Wow, that’s great.” And, of course, everybody who’s not a historian is saying, “This is horrible!” I’m thinking, “Well, no, no, we’re not killing each other. Thank God.”

 [audience laughter]

That’s — oh, I know what that is. That’s showing you what’s happening with the end of the Cold War. That’s what’s happening with the end of the Cold War across the Western world. This is Manuel Eisner’s work, my friend Manuel Eisner. So in other words, we’re all working on a very similar wavelength. Do you see what my argument is and how this is going? So in other words, we’re trying to figure this out and we’re working on it at a very deep level trying to sort this all out. Okay, well, I’ll stop now. Thank you.

TURONIS: I’m Diana Turonis with Immigration and Customs Enforcement Office of Human Capital Strategy and Technology. And I’m wondering if you’ve looked at the correlation with the psychotropic drugs, the pharmaceuticals, the different changes in the American Psychological Association diagnosis criteria and how that relates to the trends.

Roth: With drugs?

TURONIS: No, the whole psychological profile and the changes in the pharmaceutical industry and the changes in the AMA diagnostic codes.

Roth: I’m not sure that I understand exactly what you’re asking me, because I’m not really familiar with all of that terminology. What exactly do you want — I’m sorry I don’t understand, I’m being very dense.

TURONIS: Okay, you’ve looked at political, economic, I presume education levels and those different drivers, age levels, abortion rates, things like that.

Roth: Have I looked at schizophrenia, for instance, you’re asking me?

TURONIS: Well, those things, other things that may end up correlating seem to be more predictive. I do a lot with predictive analytics.

Roth: Thank you; well, that’s a good question. One of the things that I’ve found and one of the issues that Greg and I talked about before, is we know on an individual level that there are predictors that are very predictive of how the likelihood of violence is in an adult offender. We know that there’s a certain minority, small group, of mental illnesses that predispose one to violence against oneself and others, particularly forms of schizophrenia and delusions. We see that with certain forms of bipolar disorder, deep psychosis, depression. Absolutely those things — so if you look at any period in time and you look at the murderers, a higher percentage than randomly of the murderers would be those people.

          But what’s interesting is how highly suggestible people who are schizophrenic are. In other words, if they grew up in an — if they’re in an adult society where people aren’t killing each other, they don’t kill at as high a rate, even though those who do kill are more likely to have those disorders, the whole homicide rate goes down because they’re highly suggestible. I’ll give you an example from the 1830s and I have a confidentiality agreement, can’t use names, but looking at the hospital records of the state hospital in New Hampshire.

But in the 1830s when the homicide rate is low, what did delusional people do? They fantasized about how to make the world a better place. So, in other words — and what’s interesting is I’m going to talk about a guy who actually everybody thought was crazy but actually had the absolute right idea. Back then everybody had a different kind of bank note, the 1830s, and they’re arguing about the currency. And Andrew Jackson vetoes the bank and how are we going to have a single bank note, and every bank has its own currency, so counterfeiting is running rampant. And the currency, deflation, inflation, it’s a mess!

And so this guy’s completely schizophrenic. He’s delusional. And what he does is he walks around his home town of Hudson, New Hampshire — and you’ll love this as someone in D.C. He’s marking a six-mile-on-a-side perimeter, which he’s marking out the District of Columbia around Hudson, and he says, “This is where we’re going to put our national bank, because we have to have one currency.” And he said, “Everybody needs to send me their bank notes now and I will send you back a bank note of the National Bank of Hudson, New Hampshire. Then we’ll all have the same currency. And we’ll have one monetary system and everybody will believe in it because I printed these notes.”

Now, of course, everybody said he’s crazy. He was absolutely right; that’s the Federal Reserve System. But the thing is is here’s a guy in a nonviolent society who is totally just trying to fix something. And, again, you know, when we look at it, you know, with the two recent incidents in D.C., all this concern about the NSA. We are a violent society. Both those people thought that President Obama was spying on them. That’s a delusion, of course. But in a violent society, they’re going to take that in a violent direction. Same with “Taxi Driver,” same with the Batman movie.

In other words, but again, so what I see is that mentally ill people are highly suggestible, and so that what we who are normal, what we do gets picked up by them in either a violent or a nonviolent way so that individual predisposition washes — is overwhelmed by this broader effect. And that’s what you see. In other words, when we talk about child abuse, are the generations that are most like — the most heavily abused in history, do they grow up to be the most violent? If you take a look at the homicide rates in the 19th century, children were very unlikely to be abused in the 1820s and the early 1830s. What do they do when the political crisis comes? They become the most violent generation in our nation’s history.

What happens to the kids who are raised in that most violent period against children, the 1860s, and ’70s? Homicide rate drops in the ’80s and ’90s. So in other words, it’s not that abused children are — if you’ve got a group of murderers, there’s always going to be disproportionate representation with children who’ve been abused. But that doesn’t mean that abused generations become violent adults. Does that make sense? So in other words, it’s one of the real challenges is to deal with this disjunct between those individual-level facilitators and these big historical events that just swamp them out.

And how we get those two levels of scholarship to talk to each other, it’s one of the things we talked about is with the Frontiers for Research for Young Scholars is to get those systems to talk to each other, okay? So I did understand your question finally. I am so sorry that I was dense. But then, it took me 25 years to figure this out. I’m not a quick study. So I say, the great thing about doing historical research is you don’t have to be a genius, because once you — it’s going to tell you you’re wrong and you’re wrong and you’re wrong. But finally, the data’s just out there.

Like, you look at Eastern Europe. And you know anything about the history of Eastern Europe and you look at that data and you just go, “Well duh!” But, you know, it takes time to get it through your thick skull. Okay, another question.

CARRIE MULFORD: I’m Carrie Mulford at the National Institute of Justice. I couldn’t help but notice that you didn’t mention gun availability, gun laws, anything like that, so I was curious —

Roth: Because I don’t like being abused and having my life threatened. [audience laughter] Because I’ve had both. I’ve been, you know, my scholarly credentials have been, you know, when I — because guns aren’t the problem. I mean, I say this, and as I say, I went to graduate school, many of us went to graduate school. We had very expensive and rigorous educations to learn that the answer to every important question is yes, no and maybe, right? And the answer to the question, “Are guns responsible for our homicide rate?” Yes, no and maybe. And that’s the truth.

And you try to tell that truth, and you get — I mean from the left and from the right. I mean it’s been more vicious from the right because they’re more vicious, but I’m sure if I said that in the 1960s when the left was more violent than the right, I would have gotten the death threats from them. So, you know, it’s a hard thing to really deal with, but I do talk about it in the book and I think about it extensively. Do you want me to talk about this? Because I know that people will watch this video, and they’ll doctor it. You can look at YouTube at how they’ve chopped things I’ve said to make me look — like one of their favorites was to chop what I said about Democrats and Republicans with what I said about chimpanzees. So they chopped it so I called Republicans chimpanzees, so they started a campaign to get me fired in Ohio. This is the kind of stuff that people will do, and I don’t want to be a coward about it, so I’ll answer your question, but this won’t go well for me if this gets on the Web. They will make sure that I get punished for saying what I’m going to say.

We’ve had, we’ve always had, a high level of gun ownership from, say, the colonial period right down into the 1940s. And at times we’ve been the most violent society in the affluent world, and at times we’ve been the least homicidal, haven’t we? So is it just guns?

When you take a look at the muzzle-loading era, and this is in the book, when it takes you time to load that gun? You know, “I’m mad now, but you’ve got three minutes to run, I suggest you do that.” Did you see, you know, the movie “Lincoln,” did you see that wonderful scene where he shoots and misses, and the guy’s trying to reason with him while he’s trying to reload his derringer, because you only got one shot? And he’s waiting, you know, he gives him about 20 seconds, and then he decides, “Uh oh, it’s time to run.” And he does. That’s how it goes. So when you look at that, what’s interesting is that gun use, the percentage of homicides committed with guns, goes up and down with the murder rate.

When that murder rate is high as it is in the mid- and early 17th century, the majority of homicides among unrelated adults are committed with guns. You look at this low period from the 1690s to the 1760s, only 10 percent of all homicides among unrelated adults are committed with guns. You go back up to the Revolution, it goes up to over half. Up and down in the Embargo crisis, up and down going into the crisis of the 1840s. So what it means is when people are feeling hostile or defensive, they will go to that dispute with their neighbor with their gun loaded. When they’re not feeling hostile or defensive, they go and cuss them out when their cattle come across the line and destroy their crop. They’ll go to the law.

So you’ll see that, you know, they’ll go to property disputes, they’ll go to political disputes with the guns loaded, and they kill each other. And so you have to kind of plan that out. And so what you see is that it goes up and down like that. And one of the experiments I would love to do, I’d love to run American history back to 1857 and dis-invent modern firearms. What do I mean by modern firearms? Those, the great invention of Smith and Wesson when they put everything together in 1857 and the first rimfire handgun that the black powder was totally enclosed within the cartridge, so you could keep your gun loaded all the time. Because if you know, black powder’s hygroscopic, it absorbs water, it corrodes your barrel, you can’t keep it loaded. Why do they always show the gun over the fireplace? Because that’s the warmest, driest place in the house. You’re trying to keep that gun going. So the thing is is when you see that firearm’s gunstock change between 1857 and 1910, it took that long for us to move to the breech-loading guns with self-contained ammunition, reliable manufactured ammunition. You will see the percentage of homicides committed with guns go up and up and up regardless of whether the homicide rate was going up or down.

And what you see, the dramatic thing is all through the colonial period, when you look at intimate partner homicides, family homicides, only 10 percent were committed with guns whether the homicide rate was high or low among unrelated adults. But when you see that modern firearm come in, you’ll see the rate at which intimate partner violence was committed goes up to be the same level as with unrelated violence. And you’ll see who’s most likely to be killed with a handgun in the late 19th century is not an unrelated adult. It’s a woman who rejected her lover. Because now I can take this gun around and I can stalk her. I can go with this gun — concealed if I am suicidal, I can conceal it, I can go talk to her. And very often she wants to be friends, she want to — rejected me, but the family wants to be friends. You go over to her house, he’ll say, “Will you take me back?” She gives the wrong answer, she says, “No, I can’t come back, but I’d like to be friends.” Shoots her, shoots himself, done. Seventy percent of those homicides are being committed with a handgun in the 1850s, ’60s and ’70s because that’s the perfect murder weapon. And because they love you, of course, the guys love you, so they don’t want to shoot you in the face, they don’t want to disfigure you. They love you, they shoot you in the back of the head or through the heart. They want it to be quick, they want it to be relatively painless, and they want to go too.

So in other words, what I would you say is that when you have this gender problem coming up, you can see that throwing guns into that is deadly. And when that homicide rate goes up, having guns there means the completion rate of an assault goes up. So, yeah, I think that when you have that homicide rate go up, having this many guns in the society makes it works than it would otherwise be. And I say one of the reasons why we probably had a tough time during the 1950s getting down to 1 or 2 per 100,000, why were we stuck at 4 to 5, part of that is the fact that we were so heavily armed, and because we’ll engage in this kind of impulsive violence.

So in other words, what I told you was that guns aren’t the fundamental problem, we’d be killing each other with rolling pins because we hate each other, we hate our country. Europeans have a tough time understanding why Americans hate their government so much. So I say there’s a very elaborate form of self-hatred in a democratic society, isn’t it? But I think that that’s there. So in other words, you can see why everybody hates me by what I say, okay. It’s not an ideological response, and, you know, it’s based on years of research, it’s based on hard work that they don’t respect. I’ll be blunt.

ARTHUR L. BURNETT, SR.: Good morning, I’m Judge Burnett, Chief Executive Officer of the National African American Drug Policy Coalition. From all this history, what can we do in the future, especially with African Americans, with reference to reducing the homicide rate? And it’s been suggested to me three [inaudible], but I want your comments on them: Number one, full employment with reference to young black males and so forth, to what extent that would deal with that anger and frustration and reduce the homicide rate. Number two, the disintegration of the African American family, and 72 percent of youngsters being born to mothers without fathers, without families at home. And the third one deals with the whole idea of what you talked about, empathy, teaching youngsters from 1 to 3 or 4 the value of life and self-worth and the role of religion. I want to hear what your answers are on that, so what we can do in the future to benefit from your historical analysis?

Roth: Yeah, that’s a great question. I would love for us to have stronger families. You know, having come from one, there’s just something about it that’s just helpful. And at an individual level — some of the things we’re talking about at the National Academy of Sciences — on an individual level, whether or not I become an offender is deeply correlated in the Pittsburgh Youth Study with whether or not I feel connected to my family, whether or not I feel respected by my family, whether or not I’m involved in family activities. At an individual level, that works. But at the global level, does it work?

In other words, the big homicide rate drivers aren’t involved in that. And so I do think that we would certainly have, you know, we could — since that makes someone more likely to be an offender, I would certainly want to see that happen. But the fact is, is that going to eliminate these homicide surges and declines? No, because when that homicide rate goes up, kids from good families, happy families are committing murder, and that’s the puzzle that we have to deal with.

Secondly, when we talk about faith, there’s no question that empathy can come from faith and that that sense of kinship, where there’s a strong sense of commonality as there were with Protestants generally in the mid-18th century, you can see the homicide rate go down. But one of the great puzzles is we are the most churchgoing people in the Western world. We’re the most likely to believe in God, yet we have by far the highest homicide rate. And I have talked to ministers about this, they say, “Well that’s not true religion.” And I think that’s probably true. How many Christians hate their fellow Americans, denounce them, call them traitors? It happens all of the time.

And so — but that’s a politicization of religion. And the question is, is faith powerful enough to overcome and help us pull together politically? So far, you know, I wrote a piece on “can faith change the world?” I used to be a historian of religion. And the answer was yes, no and maybe. [audience laughter] And, you know, I’ve been doing this for years. And it’s true! You take a look at the religions, the religious revival of the early 19th century, it drew people together. But it also divided them in new ways. And so you can see families pulled together by faith, but you can see families torn apart by faith, which leads to violence if one converts and the other doesn’t. If a husband refuses to become Christian, the wife is a Christian, they get in these disputes, it can end even in murder.

So in other words, you know, faith can be destabilizing, as well as integrating. Take a look at what’s happening in the Middle East. There’s no question that Islam is more central to many people’s lives than it was 40 years ago. But is that creating a less violent society? It’s not, because it becomes politicized; it becomes part of this division. And I think you can see it with the polarization in the 1840s. That’s when the Northern and Southern branches of the churches, the Methodist church, the Baptist church, they split apart about slavery. Exactly the time the murder rate went up. They’re all devout Christians. They all believe that they are speaking for God. And they want to kill each other.

Nicholas Zill: Hi, Nick Zill, retired from Westat. I want you to, following up on the previous question, here in the District of Columbia, if you live in upper Northwest on the western side of Rock Creek Park, there are virtually no murders. Your chance of being murdered is zero. If you live on the other side, if you live in Anacostia, your chance is among the highest in the country. So from a practical point of view, no matter what these trends are, no matter what these changes are, I’d do much better for my personal safety, the community I live in, and the stability of it in terms of family stability, educational achievement, than moving across the river. How does that play out in terms of the trends you’ve been talking about?

Roth: Well, I think the thing is I go back to the way I’ve been answering all along. I certainly see that reality. And, you know, my work looking over hundreds and hundreds of years of history, I don’t ignore that. And many of my colleagues who work on neighborhood things say, “Well, I disagree with your theory.” And they don’t see that I agree with theirs. I absolutely agree that there are places that are more dangerous in our society, that there are these neighborhood-level effects, there are these individual-level effects — mental illness, child abuse, deprivation.

We know these things. The problem is how do we integrate that with the macro trends. And what I’m really arguing is that if we build a strong nation, if we build a strong nation, that homicide rate’s going to drop in those dangerous neighborhoods. That’s the most effective way to see that homicide rate drop. And the thing is is how do you build a strong nation? [snaps fingers] You can’t snap your fingers. So, you know, why we want to go after things like unemployment is because we know that’s a risk factor. Why we want to build stronger families is because we know that’s an individual risk factor and we’ll be working around the margins of the problem is what I’m saying.

But very productively on those margins, and I want to do that. But what I’m saying is is that how do you build a strong sense of nationhood? And being a democracy, we have to do it through inclusion and through empowerment. And the problem right now is when some groups are empowered, the others grew less empowered. If you take a look at what’s happening politically on the gun issue, et cetera, you’ll see white men feel that they’re victims. This ideology that white men in America are victims is so powerful today. Speaking as a white male, there’s been no wealthier, more powerful group of people on the planet than white male Americans today.

And we have this giant pity party going on. I don’t understand it, I don’t get it, but it’s there. And we know for many people who’ve lost their place in this world because of deindustrialization, it’s a real sense of disempowerment. If you’re a white supremacist, how do you feel about the way the country’s going? It’s not good. If you believe in traditional gender relations, how’s your life going? It’s not going well. And you see a future where you are losing. And so, yeah, and playing out, I mean, “stand your ground.” That’s a statement about an older notion of masculinity, isn’t it? That’s what it’s about. “I’m a real man because I’m not going to back down in a public place.” It’s a different idea of manhood than one that says, “A real man walks away from a confrontation when it starts to threaten violence. A real man is somebody who walks around with confidence because he believes in his country, courageous, can defend himself without the need for a gun.” That’s a different concept of masculinity than the one that is constantly ready, prepared to use force. And again, you see that playing out. That’s shrinking in our country, but as I said, when that small group feels threatened with change, it can become more violent.

And so even though we see gun use, gun ownership is dropping, we see that older idea of manhood declining, I think, that older idea of gender relationships declining, for those who don’t want to make that change, who can’t make that change, who’ve been disrespected by women, who feel disempowered, their world is falling apart. It’s like for the ethnic Russians who saw their world fall apart. They’re at risk of becoming more violent. And I think that’s what we’re kind of looking at. Does that make sense? I think that’s what we’re talking about.

And so how do you deal with that level of anger? How do you deal with changing that idea of manhood? Because, you know, my dad grew up in one of the most violent, dangerous neighborhoods in the country, Globeville, out in Denver. And they used to have gangs and he has to go out fully armed all the time. I remember when Alex joined the Boy Scouts, Dad sent all his old gang knives, switchblades, everything. “Hey Alex, they’re all yours,” we looked at these like, “Ah! Dad!” But, you know, my father just thought, “My God, I’m so happy to be beyond that.” You know, he wanted nothing to do with that. And he just changed his idea — you know, his idea of manhood was “get me the hell out of here.” You know, “I don’t want to be that kind of man.”

And the thing is, how do you get out of that? It’s very hard. And my dad did not fully escape it. I can’t say that he fully did. He was still conflicted, all through his life about what kind of man he was. You could just see his temper just [snaps fingers] like that, because that’s how he was raised. In a violent, deadly neighborhood, high level of child mortality, half the kids were dead, including in our family, by adulthood. Bad place.

And how do you get out of that? My father got out of it farther than his father did, but it’s a different idea of masculinity. Scary, I’m just thinking about how the only thing I heard about my grandfather — because I didn’t know him, he died before I was born — “There’s nothing your grandfather couldn’t pick up or tear apart with his bare hands.” That’s all I ever knew about my grandfather, I knew he was a very violent guy. So, you know, that kind of manhood, you know, I’ve heard stories. You know, how do you get away from that? It’s very hard.

Ridgeway: Professor Roth, thanks for coming, and thanks to all of you for attending.