AS A PULITZER PRIZE-WINNING AUTHOR and evolutionary psychologist at Harvard, Steven Pinker has already flexed his intellect in bringing us The Blank Slate: a very large treatment of human nature and the folly of intellectuals who try to deny its existence. In that book Pinker won round one: human nature is here to stay — even if it happens to be much less “determinative” of our behavior than many fear, it is far too important to ignore. The special targets for such enlightenment were a certain breed of intellectual, fairly rife in academia, who loves the idea of human possibilities being totally open and naively denies any major hint that genes might be constraining our behavior.

Pinker continues that same theme in his remarkable book Better Angels, an exceptionally well written evaluation of our political world that is driven by a controversial premise: that human beings are becoming less violent in their actual behavior. What drives us, as a species, to defy our better angels? Why do we have the capacity from one generation to the next to be cruel to animals, to beat and abuse spouses and children, to place costly acts of vengeance above acts of profitable cooperation, to engage in public acts of torture with or without guillotines, or to commit murder or mayhem or engage in full-scale war? This formidable list makes it seem quite logical to write off humanity as a species hopelessly given to violence, and as a species that, with two enormously destructive world wars under its belt just in the past century, surely must be getting worse instead of better.

Not so, Pinker tells us. He argues that we are in fact becoming less prone to these same practices, be it in the practice of warfare, or in our treatment of women or minorities, or in the vehemence with which we oppose torture in a political world in which many governments remain uninhibited. In a variety of ways, violence is on the decline in this violent-seeming world of ours — in spite of an astoundingly ugly first half of the twentieth century.

To back up this claim, Pinker uses a unique metric for how to evaluate rates of violence. In spite of the difficulty of measuring past violence quantitatively, an enormous amount of research effort has gone into the task, and Pinker’s mission is to add up all the facts. For his analysis, Pinker has decided to “normalize” our violent track record, in the sense that the quantitative measure is not simply how many people are killed in a given year in warfare. Rather such figures are corrected to reflect how many people were available to be killed, so that a ratio can be identified. This yields a rate, in terms of the number of people killed per thousand per population, and that proportion is what Pinker uses as he assesses patterns of violence over time. His analysis holds that rates of violence per capita are declining significantly, and the figures certainly support his thesis. Surprisingly, he says, they have been declining ever since Paleolithic hunter-gatherers developed even higher rates than modern nations.

This might be quite surprising to anyone save for an evolutionary anthropologist like myself. I already knew that hunter-gatherers, even those lacking any semblance of intensive warfare, tend to have internal homicide rates as high as Los Angeles or New York on a Saturday night. What I didn’t know was that the Mongol Conquest killed the same number of people as died in World War Two, but if you turn the killing into rates p...