Documenting the Decline of Violence

Steven Pinker spends the first few chapters of The Better Angels of Our Nature proving that violence has declined, knowing that just about no one will believe him.

To accomplish this, he first forces the reader to realize just how violent our past was. It's hard to read parts of the book that describe torture or executions in detail, and to be reminded of just how common they were. Because Pinker is a linguist, he often gives examples of idioms that came from our violent past. Phrases don't get into common usage unless they were common activities, right? Think about the meaning of these phrases:

• Burn at the stake
• Hold his feet to the fire
• Racked with pain
• Drawn and quartered
• Cut off your nose to spite your face

How often do we use those words without thinking about where they came from?

Even our hallowed belief in the violence of modern media is overshadowed: In a recent study of genres in children's entertainment, "Television programs have 4.8 violent scenes per hour; the nursery rhymes had 52.2" (page 21).

Government: Part of the Solution

Pre-state people were the most violent, with territorial defense and raids on neighboring people as a way of life. Pre-agricultural, pre-state people may have had healthier diets than their farming descendants, as Jared Diamond argues, but they were much more likely to die from violence instead.

Pinker emphasizes the importance of looking at the relative numbers of deaths from violence. In an absolute sense, the most deaths have occurred in the 20th century because there were so many more people then. But it's most important to look at a person's likelihood of dying a violent death at any point in time, so numbers such as the percentage or rate per 100,000 are a lot more meaningful. "If I were one of the people who were alive in a particular era," he asks, "what would be the chances that I would be a victim of violence?" (page 47).

Using data from ethnographers studying more recent stateless people plus the prehistoric skeleton record, Pinker comes up with stunning graphs showing the decline of violence after the creation of governments. One graph on page 49 gives a visual overview of the percentage of deaths in warfare. Here are some highlights:

• The archaeological record of war deaths at two dozen prehistoric sites ranges from 60 percent to about 4 percent, with an average of about 15 percent
• In eight recent hunter-gatherer societies, the percentage of war deaths ranges from 32 down to about 4, with an average of about 14
• In 10 contemporary hunter-horticulturalist and other tribal groups, the range is 60 to 12 percent; the average is about 25 percent
• In pre-Cortés Mexico (before 1500 C.E.), under the Aztec and other governments, the rate was about 5 percent. Even comparing pre-Colombian remains of hunter-gatherers with city dwellers like the Inca, where both the time period and part of the world are held constant, the percentages are hunter-gatherers 13.4 percent vs city-dwellers 2.7 percent
• The entire world in the 20th century (including the world wars, genocides, and everything else we believe makes our recent past the most violent of all time): the average is about 3 percent
• The 21st century so far has worldwide percentages more like 3 one-hundredths of a percent (that's 0.0003)

The war deaths graph on page 53 shifts the rates from percentages of the dead to focus on the living, listing the rates per 100,000 people. Expressed that way, wars account for an average of 524 per 100,000 of nonstate peoples. It's much lower for states, even in the 20th century:

• Germany: 144
• Japan: 27
• Russia: 135
• United States: 3.7

Okay, so fewer people, relatively, are killed in wars and war-related violence. But what about homicide? There must be more homicide, right?

No.

Homicide rates per 100,000

• Western Europe in 2000: 1 per 100,000
• U.S. at its highest rates in the 1970s and 80s: 10 per 100,000
• Detroit at its worst: 45 per 100,000
• The Semai, called the "nonviolent people of Malaysia": 30 per 100,000
• The !Kung, called the "harmless people": 41 per 100,000
• The Inuit, who were said to act "never in anger": almost 100 per 100,000

At the Detroit rate, Pinker writes, "you would notice danger in everyday life, and as the rate climbed to 100 per 100,000, the violence would start to affect you personally: assuming you have a hundred relatives, friends, and close acquaintances, then over the course of a decade one of them would probably be killed" (page 52). At a rate of 1,000 per 100,000, you'd lose one person per year, and would have a better than even chance of being killed yourself in your lifetime.

Even the least violent nonstate societies studied by anthropologists in the 20th century -- the Semai and !Kung -- have rates per 100,000 much higher than Europe and as high as the most dangerous cities in the U.S. at its worst point in recent history. (The graph on page 55 compares the average of the 10 U.S. cities that had the highest homicide rates in 1990 with the two pre-state peoples with the lowest homicide rates. In other words, the absolute worst that recent civilization has to offer is compared with the best from precivilization. Their rates are about the same, between 30 and 40 deaths per 100,000.)

How did Europe get to that 1 per 100,000 death rate from homicide? Were Europeans always less violent?

For this, Pinker relies on Ted Gurr, whose ground-breaking work plotted homicides in England from 1200–2000 C.E. , based on records kept in various cities. He found an incredible drop from 110 per 100,000 in 1300s Oxford to less than 1 per 100,000 in London in the 1950s. Pinker writes, "This discovery confounds every stereotype about the idyllic past and the degenerate present. When I surveyed perceptions of violence in an Internet questionnaire, people guessed that 20th-century England was about 14 percent more violent than 14th-century England. In fact it was 95 percent less violent" (page 61). Another historian compiled data from a town in Kent and found a similar trend. Rates in other western European countries also dove from around 100 per 100,000 in the 13th century to 1 per 100,000 in 2000.

Why Is the U.S. More Violent than Europe?

It's hard not to ask why U.S. rates are so much higher than European rates.

First Pinker points out that some U.S. states are indeed more like Europe in their homicide rates (New England, Minnesota, Iowa, the Northern Plains, and the Pacific Northwest). The highest rates in the U.S. are in the South and Southwest.

Pinker argues that government never penetrated in these parts of the country as deeply as it did in New England, for instance. They "never fully signed on to a social contract that would vest the government with a monopoly on the legitimate use of force. In much of American history, legitimate force was also wielded by posses, vigilantes, lynch mobs, company police, detective agencies, and Pinkertons…" (page 99).

He has more to say about geography, ethnic roots, economic foundations, and gender and religious differences that fed the varying outcomes in what have become our red and blue states. But the upshot is that "The North is the extension of Europe and continued the court- and commerce-driven Civilizing Process that had been gathering momentum since the Middle Ages. The South and West preserved the culture of honor that sprang up in the anarchic parts of the growing country, balanced by their own civilizing forces of churches, families, and temperance" (p. 106).

Most notably, to me, though, is this: "Southerners do not outkill northerners in homicides carried out during robberies…only those sparked by quarrels" (page 99). It's that culture of honor (think dueling, feuds, revenge killings, and women killed to maintain family honor) that is different in the South, as well as other parts of the world that have higher rates of homicide.

There'll be more to come on the culture of honor later this week, when I summarize Pinker's explanation of the reasons for the decline of violence.

_______

Part 2 of Steven Pinker Week at Daughter Number Three.