The Better Angels of Our Nature: Why Vio - By Steven Pinker Page 0,121

slave trade, imperial wars throughout Asia, Africa, and the South Pacific, and two major bloodlettings that didn’t even make the list: the American Civil War (650,000 deaths) and the reign of Shaka, a Zulu Hitler who killed between 1 and 2 million people during his conquest of southern Africa between 1816 and 1827. Did I leave any continent out? Oh yes, South America. Among its many wars is the War of the Triple Alliance, which may have killed 400,000 people, including more than 60 percent of the population of Paraguay, making it proportionally the most destructive war in modern times.

A list of extreme cases, of course, cannot establish a trend. There were more major wars and massacres before the 20th century, but then there were more centuries before the 20th. Figure 5–3 extends White’s list from the top twenty-one to the top hundred, scales them by the population of the world in that era, and shows how they were distributed in time between 500 BCE and 2000 CE.

FIGURE 5–3. 100 worst wars and atrocities in human history

Source: Data from White, in press, scaled by world population from McEvedy & Jones, 1978, at the midpoint of the listed range. Note that the estimates are not scaled by the duration of the war or atrocity. Circled dots represent selected events with death rates higher than the 20th-century world wars (from earlier to later): Xin Dynasty, Three Kingdoms, fall of Rome, An Lushan Revolt, Genghis Khan, Mideast slave trade, Timur Lenk, Atlantic slave trade, fall of the Ming Dynasty, and the conquest of the Americas.

Two patterns jump out of the splatter. The first is that the most serious wars and atrocities—those that killed more than a tenth of a percent of the population of the world—are pretty evenly distributed over 2,500 years of history. The other is that the cloud of data tapers rightward and downward into smaller and smaller conflicts for years that are closer to the present. How can we explain this funnel? It’s unlikely that our distant ancestors refrained from small massacres and indulged only in large ones. White offers a more likely explanation:Maybe the only reason it appears that so many were killed in the past 200 years is because we have more records from that period. I’ve been researching this for years, and it’s been a long time since I found a new, previously unpublicized mass killing from the Twentieth Century; however, it seems like every time I open an old book, I will find another hundred thousand forgotten people killed somewhere in the distant past. Perhaps one chronicler made a note long ago of the number killed, but now that event has faded into the forgotten past. Maybe a few modern historians have revisited the event, but they ignore the body count because it doesn’t fit into their perception of the past. They don’t believe it was possible to kill that many people without gas chambers and machine guns so they dismiss contrary evidence as unreliable.19

And of course for every massacre that was recorded by some chronicler and then overlooked or dismissed, there must have been many others that were never chronicled in the first place.

A failure to adjust for this historical myopia can lead even historical scholars to misleading conclusions. William Eckhardt assembled a list of wars going back to 3000 BCE and plotted their death tolls against time.20 His graph showed an acceleration in the rate of death from warfare over five millennia, picking up steam after the 16th century and blasting off in the 20th.21 But this hockey stick is almost certainly an illusion. As James Payne has noted, any study that claims to show an increase in wars over time without correcting for historical myopia only shows that “the Associated Press is a more comprehensive source of information about battles around the world than were sixteenth-century monks.”22 Payne showed that this problem is genuine, not just hypothetical, by looking at one of Eckhardt’s sources, Quincy Wright’s monumental A Study of War, which has a list of wars from 1400 to 1940. Wright had been able to nail down the starting and ending month of 99 percent of the wars between 1875 to 1940, but only 13 percent of the wars between 1480 and 1650, a telltale sign that records of the distant past are far less complete than those of the recent past.23

The historian Rein Taagepera quantified the myopia in a different way. He took a historical almanac and stepped through the pages