Shocker: Is Our World Becoming Less Violent?
Stay up to date with the latest headlines via email.
Humanity's lust for violence has undergone a long, precipitous decline at every level of social interaction, from domestic abuse to violent crime to interstate wars. That's the sweeping and somewhat counterintuitive thesis of psychologist Steven Pinker's new book, The Better Angels of Our Nature: Why Violence Has Declined. The pacification of humanity, says Pinker, is “a fractal phenomenon, visible at the scale of millennia, centuries, decades, and years.”
Pinker writes that the “very idea invites skepticism, incredulity and sometimes anger.” He sets out to overcome that barrier by surveying a broad swath of data, from examinations of ancient bones unearthed in peat bogs and on long-forgotten battlefields, to homicide statistics based on European coroners' inquests and local records dating back 800 years, to databases of modern interstate conflicts and civil wars.
Does Pinker's research validate his thesis? And if so, what forces might explain such a profound shift in human society?
Are We Really Less Violent Today?
Pinker makes his case by combing dozens of disparate datasets to pull out what he proposes is a standard measure of our tendency toward violence: the likelihood of dying at the hand of another human being in a given year. In the long sweep of human history, he makes a compelling case, noting that almost 20 percent of bones uncovered at archaeological excavations of prehistoric societies show evidence of violent trauma – a death rate unparalleled in even the bloodiest episodes in recent history.
With the emergence of the city-state – first in pre-Columbian Mexico in the 15th century -- the rate of violent death declines precipitously, to 5 percent. Wandering closer to modernity, Pinker cites estimates of war deaths (excluding violent crime) in the two most violent regions and centuries of the era of the nation-state: 17th century Europe, “with its bloody wars of religion,” and the 20th century, with its two world wars. Historian Quincy Wright estimated the 17th century death-rate by war at 2 percent, and estimates of that measure in the 20th century run as “low” as 1 percent.
He cites criminologist Manuel Eisner's study of homicides in Europe dating back to 1200 CE, which illsutrated an equally dramatic decline in one-on-one violence, at least on the continent. Eisner estimates that during the Middle Ages, about 100 in 100,000 people were murdered, a figure that has fallen to around 1 in 100,000 today.
That's but a small a fraction of the research Pinker cites, devoting six chapters brimming with graphics and charts to make his case. Yet, as with any thesis spanning the entire history of human existence planet-wide, Pinker ultimately runs into an empirical wall; there isn't sufficient data to justify the claim that violence has fallen precipitously in every culture and at every level of human interaction throughout history. A comprehensive database of violent deaths worldwide for the 20th century, much less for the 11th, simply does not exist.
Elizabeth Kolbert, reviewing the book for the New Yorker ( $$), unfairly charges that “Pinker’s attention is almost entirely confined to Western Europe.” He examines what research is available from a host of societies at different points in history, but he does devote an inordinate amount of space to the decline in violence in the “West,” and in the era of the nation-state, and one might imagine that to be a result of where the best data are to be found.
One also has to question the rigor behind Pinker's contention on the interpersonal level. He can demonstrate that corporal punishment has declined in schools around the world, and show that violent crime has seen a dramatic, decades-long fall in the West, but what about the kind of violence that isn't tracked by governments with any consistency? Has there really been a consistent and global decline in drunken men punching one another in the face? Have we really become more tame at every level of intercourse, and is it really a “fractal” phenomenon that holds true “at the scale of millennia, centuries, decades, and years”?