The Decline of Violence

Neuroscientist Steven Pinker on the triumph of peace and prosperity over death and destruction


You are less likely to die a violent death today than at any other time in human history. In fact, violence has been declining for centuries. That is the arresting claim made by Harvard University cognitive neuroscientist Steven Pinker in his new book, The Better Angels of Our Nature: Why Violence Has Declined (Viking). The title, taken from Abraham Lincoln's First Inaugural Address, refers to the way in which the modern world encourages people to suppress their inner demons and let their better angels fly. 

Just a couple of centuries ago, violence was pervasive. Slavery was widespread, wife and child beating were acceptable practices, heretics and witches were burned at the stake, pogroms and race riots were common, and warfare was nearly constant. Public hangings, bearbaiting, and even cat burning were popular forms of entertainment. By examining collections of ancient skeletons and scrutinizing contemporary tribal societies, anthropologists have found that people were nine times as likely to die violent deaths in the prehistoric period than in modern times, even allowing for the world wars and genocides of the 20th century. Europe's murder rate was 30 times higher in the Middle Ages than it is today. 

What happened? Human nature did not change, but our institutions did, encouraging people to restrain their natural tendencies toward violence. In more than 800 pages of data and analysis, Pinker identifies a series of institutional changes that have led to decreasing levels of life-threatening violence. The rise of states 5,000 years ago dramatically reduced tribal conflict. In recent centuries, the spread of courtly manners, literacy, commerce, and democracy have reduced violence even more. Polite behavior requires self-restraint, literacy encourages empathy, commerce changes zero-sum encounters into mutually beneficial exchanges, and democracy restrains the excesses of government. 

Pinker is the author of seven books, including The Language Instinct, How the Mind Works, and The Blank Slate: The Modern Denial of Human Nature, about which reason interviewed him in 2002 ("Biology vs. the Blank Slate," October 2002). A native of Montreal, Pinker received his bachelor's degree from McGill University in 1976 and his Ph.D. in psychology from Harvard in 1979. After serving on the faculties of Harvard and Stanford, he moved to MIT in the early 1980s. He returned to Harvard in 2003 as the Harvard College Professor and Johnstone Family Professor in the Department of Psychology. 

Pinker dropped by reason's Washington, D.C., office in October to talk with Science Correspondent Ronald Bailey about ideology, empathy, and why you are much less likely to get knifed in the face these days. To see a video of this interview, go to

reason: Why has violence declined? I think most people would be astonished to hear that.

Steven Pinker: First of all, I have to convince people that there's a fact that needs to be explained—namely, that violence has declined. And it has, as I demonstrate with 100 graphs and data sets. The reasons, I think, are multiple. One of them is the spread of government, the outsourcing of revenge to a more or less disinterested third party. That tends to ramp down your rates of vendetta and blood feud for all the reasons that we're familiar with from The Sopranos and The Godfather. If you've got a disinterested third party, they're more likely to nip that cycle in the bud. Not necessarily because they have any benevolent interest in the welfare of their subject peoples, especially in the early governments. Their motive was closer to the motive of a farmer who doesn't want his livestock killing each other. Namely, it's a deadweight loss to him. 

But even without this benevolent interest, you find that with the first states in the transition from hunting and gathering to settled ways of life, violence goes down, and in the consolidation of kingdoms during the transition from medieval times to modernity, rates of homicide go way down.

reason: What else? 

Pinker: A second one is the growth of commerce; opportunities for positive-sum exchange, as opposed to zero-sum plunder. When it's cheaper to buy something than to steal it, that changes the incentives, and you get each side valuing the other more alive than dead—the theory of gentle commerce [that comes] from the Enlightenment.

reason: How much has violence declined? 

Pinker: [During] the transition from tribal societies to settled states, there was a reduction from about a 15 percent chance of dying violently down to about a 3 percent chance in the first states.

reason: One of the claims that I was most struck by is that rates of violence in Europe have dropped 30-fold from the medieval period.

Pinker: That's right. Rates of homicide can be estimated more accurately than crimes with fuzzier definitions, like assault and rape, because a dead body always gets people's attention, or a missing person. The homicide records go back in many parts of Europe to the 1200s. And they all show an astonishing trend. Namely, that the rates of homicide have plummeted, from anywhere from 30 to 100 per 100,000 per year down to the [current] European average, which is between one and two per 100,000 per year.

reason: I was at a conference a while back where I outlined your thesis briefly to a panel of fairly distinguished political scientists and economists. They rejected it out of hand, retorting, "The 20th century was horrible. Millions died." Why is it that most people tend to believe violence is increasing?

Pinker: Well, millions died in centuries before the 20th. People confuse a data point with the trend. They remember the horrific episodes of violence in the 20th century, but one occurrence is not a trend. And despite universal predictions that World War I to World War II was just the beginning of a sequence where World War III would be even worse, World War III didn't happen. And in fact, from the spike of the Second World War, there's been a historically unusual period of peace among developed countries. 

reason: Let's go through some of the reasons and processes by which the world became less violent. It began with what you call the pacification process, which involved the creation of states.

Pinker: The first states seemed to have in their wake a massive reduction of death in tribal raiding and feuding, basically because it's a nuisance to the overlords. So you have things like the Pax Romana, the Pax Islamica, the Pax Sinica, in China, where the emperors would much rather have the peasants alive to stock their tax rolls and armies, and be slaves or serfs. So they had a selfish interest in preventing too much internecine feuding among their subject peoples and basically kept them from each other's throats. Not that it was a life that we would consider particularly pleasant. You're substituting a lot of violence among tribes and villages and clans for a lesser amount—but still a brutal form of violence—from the state against its citizens. 

The next transition, after you have the government preventing people from committing violence against each other, you now have the problem of preventing the government from committing violence against its own peoples. And that was, basically, the advent of democracy and the various reforms of the Enlightenment.

reason: The next reduction in violence occurred as a result of what you call the civilizing process.

Pinker: It's a term that I borrowed from the German sociologist Norbert Elias, in his book by that name, where he figured out—even in the absence of quantitative data —that Europe had become a less violent place in the transition from the Middle Ages to modernity. We now know that he was right, now that historical criminologists have gathered the quantitative data. But he had noticed it just from narrative accounts of what daily life was like. Just people cutting off each other's noses, stabbing each other over the dinner table in response to an insult—there seems to be less now than there was then. He had an immediate explanation and an ultimate explanation. The immediate explanation was a psychological change. Namely that people exercise more self-control and more empathy. They counted to 10 and swallowed their pride rather than lashing out with a dagger when they'd been insulted. They tried to get inside the heads of other people in general, to figure out what they wanted. 

Now that just begs the question of why would there have been this psychological change. He identified two exogenous factors, his first being the consolidation of kingdoms. Instead of a patchwork of little fiefdoms and duchies and principalities and baronies, you had kingdoms where criminal justice was nationalized, and that allows justice to be reckoned by a more disinterested third party, and it keeps warlords from cycles of vendetta and feuding. [This occurred] together with the expansion of the infrastructure of commerce, both the physical infrastructure in the form of roads and wagons and carts and mills, and also a financial infrastructure: currency that could be recognized throughout a kingdom once the king had established control, and enforceable contracts, tilting the incentive structure from conquest to exchange. 

The intermediate link was that in order to get ahead during this transition, you no longer had to be the baddest knight in the land. You had to basically take a trip to the king's court and kiss up to his various minions and bureaucrats. That required inhibiting various impulses—not blowing your nose into your hand and then shaking someone else's hand, or not gnawing on a bone and putting it back into the serving dish—that weren't appropriate to the king's court. So there was a whole set of manners involving self-control that we call courtesy, from the word for court. According to Elias, this habit of self-control—and also empathy, because in an economy based on commerce, you've got to keep the customer satisfied, you've got to anticipate demand of your clients and customers—[meant that] people exercised what psychologists call today "theory of mind," an ability to get into other people's heads. The whole causal chain is government and commerce, [which lead] to self-control and empathy, [which lead] to less impulsive violence.

reason: But with regard to violence, government has been a big problem. It was the solution to tribal warfare, but governments have developed their own more efficient ways of killing people. 

Pinker: Yes. Although probably in fewer numbers, because in no period of history would you have, say, 25 percent of the population killed by their government, which you could easily have in tribal warfare in hunter-horticulturists. 

(Interview continues below video.)

reason: What accounts for what you call the "humanitarian revolution" in the 17th and 18th centuries?

Pinker: My best guess was that it was because of literacy. The first industry to show advances in productivity prior to the Industrial Revolution was bookmaking. Paper got cheaper. Printing is cheaper than handwriting, both because it's faster and because you can squeeze more text on a given amount of paper. Bookbinding, distribution, all of that increased in the 17th century. There were also ships that could move people around as well as ideas. The first post office. 

So you have the republic of letters. You have pamphlets going viral. You have many more books being published. You have higher literacy, so more people can read them. Once you don't live in a pokey little village, where all of your ideas come from the priest or from the elders, you're exposed to a whole world of ideas. And you're allowed to talk about them; you're not burned at the stake for talking about them. You can get together in pubs and coffeehouses and salons, and hash things out. The discussion's going to go in some directions rather than others. It's unlikely that everyone's going to be persuaded, "Hey, the kings rule by divine right. Isn't that obviously true?" It's more likely that a bunch of minds exchanging ideas will maybe see a wee problem in that doctrine.

Likewise [with] slavery. If you don't examine your practices too deeply, if you do things because that's the way my father did it, and that's the way it's always been done, and the Bible says it's OK, and it would be disruptive if we changed, then you might think that slavery is not so bad. But if you start to argue with someone, if someone says, "Hey, slavery really can't be defended," and you try to defend it, you're eventually going to lose the argument, just the way you'd lose a fallacious mathematical argument. 

I argue that both the technologies of exchange of ideas and the political infrastructure, namely the freedom of speech—not getting burned or broken or disemboweled if you come up with a heretical idea—will, just in the nature of social relations, push in certain directions, and they're going to be humanitarian directions. Because humanitarian treatment is just a better way for people to live together than constant war or exploitation.

reason: Moving to this century, you claim we are now in the midst of the Long Peace and the New Peace. What is the Long Peace, and what is the New Peace?

Pinker: Long Peace is a term that I took from the historian John Gaddis, referring narrowly to the absence of war, direct war, between the U.S. and the USSR, confounding all predictions in the late '80s when he coined the phrase. He and a number of military historians, even in the '80s, said, "Hey, something very weird is going on. The U.S. and the USSR aren't going to war. Everyone said they would. How come they haven't?" And more generally, people noticed, "Hey, what about Western Europe, France and Germany? They've gone an awful long time without fighting a war. This is kind of historically unusual." 

There were predictions even then in the late '80s that something had changed historically, that the use of war, as Clausewitz put it, as continuation of policy by other means, had really changed. War had been taken off the table as a live option. The absence of war involving developed countries, say the 40 or 45 richest countries, and between the great powers, was unusual even in the '80s. And here we are 25 years after that, and our luck has held out. 

The New Peace is another phenomenon that very few people are aware of. Namely, there are war nerds who meticulously tabulate the number of battle deaths year by year in each of a number of categories of armed conflict: civil wars, interstate wars, colonial wars, and so on. They have been stunned, in plotting their data, to notice there's been a big decline in wars in the rest of the world, starting around the end of the Cold War. All of those nasty little civil wars in Africa, South Asia, and Central and Latin America kind of fizzled out, and no one's even noticed. More important, the number of people killed has plunged. This past decade, even with all of the wars that we read about, has had the lowest rate of battle deaths of any decade since they started keeping score in 1946. That's the phenomenon I call the New Peace—that the Long Peace is starting to spread to the rest of the world.

What's also [interesting is] a finding that I came across after the book had come out, by Erica Chenoweth and Maria Stephan—nonviolent vs. violent resistance movements. If you want to topple your government, what works better: mass protests in the street or arming guerrillas? The answer is you can't tell if you just think of anecdotes, because there are some on each side that work and don't work. So they tabulated numbers, and they found that about 75 percent of nonviolent resistance movements work, and about 25 percent of violent resistance movements work. Unless you do the counting, you'd never know that there was this massive difference.

reason: Over what period was that?

Pinker: I think it's over the last 40 to 50 years.

reason: With regard to the Long Peace and the New Peace, some people are going to think, "Steven Pinker is over-interpreting this random fluctuation which just happened to be a peaceful period." 

Pinker: There is a big dose of randomness in human history, and some coming fanatic somewhere might be promoting a new movement that will gain adherents and lead to brutal insurgencies and repression. Stuff happens. What doesn't happen, though, is cycles of build-up and release: [the notion that] every year that has passed without a war means we're more and more due for the next one. Tensions have been building up; it's only a matter of time before they'll burst. That doesn't happen. And that's the mental model that a lot of people have. 

reason: Going back to the pacifying effects of gentle commerce that you mentioned, you note that this period could be called the Capitalist Peace. 

Pinker: This is a heretical idea coming out of, of all places, Scandinavia. There are war nerds who run regressions trying to predict what leads to escalation, military tensions, or de-escalation. There was a lot of statistical support for an idea called the Democratic Peace. The extreme form is that no two democracies have ever waged war on one another. There's a new movement to try to argue it's actually the capitalism, more than the democracy, that's doing the work in this correlation. And there seem to be data that capitalist countries are less likely to go to war with each other. They're less likely to go to war, period, including against noncapitalist countries, [and are] less likely to have civil wars, and less likely to have genocides. 

reason: In other words, "Make money, not war." 

Pinker: Yes, exactly.

reason: You cite a nice quotation from Aleksandr Solzhenitsyn. who writes: "Macbeth's self-justifications were feeble—and his conscience devoured him. Yes, even Iago was a little lamb too. The imagination and the spiritual strength of Shakespeare's evildoers stopped short at a dozen corpses. Because they had no ideology." Why is ideology so deadly?

Pinker: There are two features of a lot of ideologies that make them deadly. One of them is demonizing. Namely, if the impediment to a better world is some defined group of people—the Jews, the rich peasants, the capitalists, the left-wingers, the communists—then you're justified in unlimited outlays of violence, because you're still ahead of the game. The costs are outweighed by the benefits. The other is if your ideology promotes some utopia—and often they're linked, because it's usually some demonic group that stands in the way of your utopia—then, at first you might think, well, utopia, who could be against that? Infinite good forever. If you hold out the hope of infinite good forever, then those that oppose it are arbitrarily evil, then you're justified in arbitrary punishment to eliminate them, and you're still doing much more good than harm. The amount of violence that you can perpetrate in pursuit of this utopia can be as high as you like, and you're still ahead of the game, including collateral damage among innocents. "You can't make an omelet without breaking the eggs" was the old slogan. 

reason: When we think about ideologies, the vast ones over the 20th century, fascism and communism come to mind. Do you see others out there now that might pose a problem in the future? 

Pinker: Well, radical religions, millennial religions, radical Islam, radical forms of Christianity, that say that there's some irredeemably evil group standing between the world and perfection: the crusaders, the infidels, the Jews, the polytheists, the nonbelievers, and so on. And in many of these ideologies, both religious and secular, there's the idea that there will be one final spasm of violence before we attain bliss for eternity, the worse the better. And as there's more violent chaos, it just means that utopia's around the corner. Now those are very dangerous ideologies, both religious and secular.

reason: The age of ideology coined the term genocide. You've pointed out that just because we recently invented the term doesn't mean it didn't happen before.

Pinker: What it means is we started caring about it. They used to be called sieges, sackings, conquests, but if you think about what a siege actually meant to the inhabitants of the city, it was a form of genocide. And this is a point on which all historians [who study] genocide are unanimous: Genocide goes way back. 

reason: What is the "rights revolution"?

Pinker: The rights revolutions are the various social changes [that are] directed largely at violence on smaller scales. Violence against racial minorities, like African Americans, in the form of pogroms and race riots and forms of violent intimidation. The women's rights revolution, which has campaigned against rape and domestic abuse. The children's rights revolution, which has campaigned against both child abuse and milder forms of violence against children—corporal punishment, spanking, paddling in schools. The gay rights revolution, which has tried to do something about gay bashing, both by neighborhood hooligans and by the police, and succeeded in decriminalizing homosexuality in most of the Western world. And the animal rights revolution, extending the circle of concern to all sentient beings, not even only humans.

reason: Why did the rights revolutions happen now?

Pinker: Yeah, why is it the last 50 or 60 years? I suspect that is again the spread of information. It may not be a coincidence [that it started at the same time that Marshall McLuhan came up with] the cliché that we live in a global village thanks to electronic media. We sense other people's ways of life much more immediately. They feel much more like part of the family, even if we are not like them. But also, [there was] the revival of the Enlightenment acceleration of argument, debate, knowledge, rational examination of how we live our lives. Because it's amazing how quickly a lot of these customs crumbled once they were just put under the microscope. When you really come down to it, should homosexuality be illegal? You're going to lose that argument. You can grasp at straws—well, the Bible says so; well our ancestors did it that way—but they aren't very good arguments. 

The other thing that happened in the second half of the 20th century is that more people went to school for longer periods of time. There was a huge increase even in traditional media, like book publishing. More recently, we have the Internet, and social media, and all of those developments. But we just live in a more intellectualized world, and certain customs that were just defended—either by sheer tradition or by appealing to religious sources like the Bible, or they just felt selfishly to be ways of getting what we want, such as getting rid of annoying behavior by children, or rape in the case of sexuality—the more you think about what it's like to be a victim of those forms of exploitation, the more you try to apply some kind of rationally designed rules that apply to everyone, the more you're going to get rid of those forms of traditional oppression and violence.

reason: That reminds me of Tim Ferris' argument in his book The Science of Liberty that both science and liberalism are methods. Liberalism is not an ideology. It is a way of approaching the world and trying to find out what works, what is true. 

Pinker: Absolutely. And that's one of the toxic features of the deadly ideologies of the 20th century. Perhaps the best example is Mao's Great Leap Forward, where it was a screwball idea for all kinds of reasons, but one of them was, as people were dying by the millions, he didn't change his mind. It was stay the course, full speed ahead, I don't care about reality. I've had this brainchild and I'm going to impose it on this country, come what may. It's a sobering lesson for all forms of political belief systems, including libertarianism. That is, you've got to be receptive to signals from the world telling you whether your ideas really are making people better off or not.

reason: How does classical liberalism figure in this decline of violence that you've identified?

Pinker: I go back and forth between the terms classical liberalism, secular humanism, Enlightenment humanism, and Enlightenment liberalism. Basically, the ethical system growing out of the Enlightenment, in which it's the suffering and flourishing of the individual that are the ultimate moral value, that societies are gadgets to enhance the flourishing of individuals, and that other forms of morality, such as loyalty to the tribe, deference to authority, preservation of spiritual purity, are secondary or should be discounted. 

reason: The book is called The Better Angels of Our Nature. What are the better angels of our nature?

Pinker: I identify four of them. One of them is self-control: the ability to anticipate the consequences of your behavior and to inhibit it as a result. That is, someone insults you, you count to 10 and you walk away instead of knifing them. Empathy: the ability to feel others' pain, so that you no longer have fun when you watch someone disemboweled, but you're actually sickened by the thought of it. Morality: admittedly, that's a very multiple-personality angel, because a lot of morality actually leads to violence rather than preventing it. But at least a morality concentrating on fairness and universal rules is one of our better angels. And reason, the ability to deploy our cognitive faculties to figure out the best way of living our lives.