Enlightenment Now: The Case for Reason, Science, Humanism, and Progress

That was the goal of my 2011 book The Better Angels of Our Nature, which presented a hundred graphs and maps showing how violence and the conditions that foster it have declined over the course of history. To emphasize that the declines took place at different times and had different causes, I gave them names. The Pacification Process was a fivefold reduction in the rate of death from tribal raiding and feuding, the consequence of effective states exerting control over a territory. The Civilizing Process was a fortyfold reduction in homicide and other violent crimes which followed upon the entrenchment of the rule of law and norms of self-control in early modern Europe. The Humanitarian Revolution is another name for the Enlightenment-era abolition of slavery, religious persecution, and cruel punishments. The Long Peace is the historians’ term for the decline of great-power and interstate war after World War II. Following the end of the Cold War, the world has enjoyed a New Peace with fewer civil wars, genocides, and autocracies. And since the 1950s the world has been swept by a cascade of Rights Revolutions: civil rights, women’s rights, gay rights, children’s rights, and animal rights.

Few of these declines are contested among experts who are familiar with the numbers. Historical criminologists, for example, agree that homicide plummeted after the Middle Ages, and it’s a commonplace among international-relations scholars that major wars tapered off after 1945. But they come as a surprise to most people in the wider world.¹⁷

I had thought that a parade of graphs with time on the horizontal axis, body counts or other measures of violence on the vertical, and a line that meandered from the top left to the bottom right would cure audiences of the Availability bias and persuade them that at least in this sphere of well-being the world has made progress. But I learned from their questions and objections that resistance to the idea of progress runs deeper than statistical fallacies. Of course, any dataset is an imperfect reflection of reality, so it is legitimate to question how accurate and representative the numbers truly are. But the objections revealed not just a skepticism about the data but also an unpreparedness for the possibility that the human condition has improved. Many people lack the conceptual tools to ascertain whether progress has taken place or not; the very idea that things can get better just doesn’t compute. Here are stylized versions of dialogues I have often had with questioners.

So violence has declined linearly since the beginning of history! Awesome!

No, not “linearly”—it would be astonishing if any measure of human behavior with all its vicissitudes ticked downward by a constant amount per unit of time, decade after decade and century after century. And not monotonically, either (which is probably what the questioners have in mind)—that would mean that it always decreased or stayed the same, never increased. Real historical curves have wiggles, upticks, spikes, and sometimes sickening lurches. Examples include the two world wars, a boom in crime in Western countries from the mid-1960s to the early 1990s, and a bulge of civil wars in the developing world following decolonization in the 1960s and 1970s. Progress consists of trends in violence on which these fluctuations are superimposed—a downward swoop or drift, a return from a temporary swelling to a low baseline. Progress cannot always be monotonic because solutions to problems create new problems.¹⁸ But progress can resume when the new problems are solved in their turn.

By the way, the nonmonotonicity of social data provides an easy formula for news outlets to accentuate the negative. If you ignore all the years in which an indicator of some problem declines, and report every uptick (since, after all, it’s “news”), readers will come away with the impression that life is getting worse and worse even as it gets better and better. In the first six months of 2016 the New York Times pulled this trick three times, with figures for suicide, longevity, and automobile fatalities.

Well, if levels of violence don’t always go down, that means they’re cyclical, so even if they’re low right now it’s only a matter of time before they go back up.