Search the Site

Author Steven Pinker Answers Your Questions

Photo credit: Rebecca Goldstein

Last week we solicited your questions for author and Harvard psychology professor Steven Pinker on his new book, The Better Angels of Our Nature: Why Violence Has Declined. You responded quickly with more than 50 questions. Now, Pinker is back with his answers to 10 of them. The result is a fascinating discussion (exactly the kind we like to have around here) on the roots of violence, the rationale for wars of the past and what a decrease in violence says about modern society. As always, thanks to everyone for participating.


Q Any thoughts on the negative side effects of decreased violence? Overpopulation? More sedentary populations? Decreased role for survival of the fittest? Not to say that violence is preferable, just wondering about the downsides of peace. – BL1Y

A A hundred years ago most people would have thought that the answer to your question was obvious. Peace would lead to decadence, effeminacy, materialism, selfishness, and intellectual and artistic stagnation. War was necessary for men to develop courage, manliness, self-sacrifice, loyalty, solidarity, and obedience to authority. Many in the nineteenth-century positively gushed about how holy and thrilling war was, including Tocqueville, Zola, Ruskin, Hegel, Mann, Stravinsky, and Nietzsche.

World War I pretty much put an end to all that.

As far as I can see, there are no downsides to peace. Unless one thinks that it’s a good idea to control population size by machine-gunning and blowing up a random sample of the population, the moral harm of war will always outweigh any good it does in limiting population size. Also, except for truly horrendous wars, population size depends far more on birth rates and on deaths from disease and famine than it does on deaths in war. The “survival of the fittest” is also a dubious benefit of war. Not only does war kill pretty indiscriminately, but it makes no sense to think of Darwinian natural selection as a human good, unless we believe that pumping out the maximum number of babies, killing people who get in our way, and letting sick and weak people die are good things.

Q Is it possible this “Long Peace” is merely the calm before the storm? World War II happened less than a century ago and was by far the highest casualty war in recorded history (if Wikipedia is to be trusted, of course). I’ve been reading G. K. Chesterton lately, and he had some choice words about the false hope of progressive morality… and he was writing before both WWI and WWII. –Don

A It’s certainly possible that another horrendous war could break out; the observation that war has decreased a lot is not the same as the prediction that it will never happen again. But if you’re asking whether war is cyclical, or whether during periods of peace the urge for war inexorably builds up until it bursts, the answer is no. Statistical studies of war show no reliable cycles, and countries can remain at peace indefinitely. It’s not as if Canada, the Netherlands, New Zealand, Sweden, and other peaceable countries are getting itchier and itchier for war with every passing year.

Q I loved your TED lecture. You do an amazing job explaining how modernity has led to a precipitous decline in violence. What explains the relatively high level of violence in the United States compared to other developed nations? The U.S.’s intentional homicide rate is 3-4x most other developed nations in Western Europe, Australia, Japan. Guns? Frontier history? Income inequality? –vimspot

A Thanks! See here for a more recent lecture on the topic. You ask a good question about violence in the United States, though it’s in large part a question about the American south and west, and about African Americans—the homicide rates of northern states are not much greater than those of Europe. It isn’t just guns, because even if you subtract all the killings with firearms and count only the ones with rope, knives, lead pipes, wrenches, candlesticks, and so on, Americans still kill at a higher rate than Europeans.

Measures of inequality certainly correlate with homicide rates across nations, provinces, and states, though the reason is not clear. One possibility is that unequal societies beget more status competition, with marginalized men vesting all their self-worth in their reputation with their peers, leading them to retaliate violently against any insult or perceived maltreatment. But unequal societies also provide less schooling, medical care, and perhaps most important, policing to their lower classes.

My own guess is that Americans (particularly in the south and west) never really signed on to a social contract that gave government a monopoly on the legitimate use of violence, as Europe did. Americans not only retain the right to bear arms but believe it is their responsibility, not the government’s, to deter harm-doers. With private citizens, flush with self-serving biases, acting as judge, jury, and executioner, body counts can pile up as trigger-happy vigilantes mete out rough justice. This may be a legacy of the long periods of anarchy in the mountainous south and frontier west, and of the historical failure of the police and courts to serve African American communities.

Q Other than writing best-selling books what can people do to help society at large resist the urge to think things are worse and worse and the world is less and less safe when this is manifestly not the case? –Joshua Northey

A A small portion of the population is willing to be reasoned with, but when I tell my reasonably intelligent sister that “children are probably safer today than at any time in human history” she scoffs at me as if I am telling her that cigarettes have nothing to do with lung cancer. She is so dismissive she won’t even read the few things I have given her about it, and her attitude is not uncommon.

One necessity is greater statistical literacy among the population and especially among journalists. People need to think in terms of proportions rather than salient examples, to appreciate orders of magnitudes (ten thousand deaths versus ten million deaths), to distinguish random blips from systematic trends, and to be aware of—and thereby discount—their own cognitive biases. When Harvard revamped its undergraduate curriculum a few years ago, I lobbied (unsuccessfully) for a statistical and analytic thinking requirement.

Also, journalists have to rethink their policy of featuring only gory events and terrifying threats. Tensions that fizzle out (e.g, remember how a decade ago India and Pakistan were allegedly on the verge of nuclear war?), wars that sputter to a halt, “war-torn” countries that are no longer torn by war, and other happy events and non-events should be just as newsworthy as things t
hat go bang. A good example is UN peacekeeping, which Joshua Goldstein shows (in his new book Winning the War on War) is demonstrably effective at preventing wars from flaring up—not always, but far more often than not. People read about the failures (e.g., Bosnia), but no one reports the successes, or tabulates what proportion of peacekeeping missions are successful.

Q I am curious to hear his thoughts about the following alternate hypotheses. The decrease in violence could be attributed to a decrease in the number of political actors (or perhaps decrease in the number of viable, competitive actors given massive power imbalances). Or, it could be attributed to a period of temporary economic prosperity.

I’m sure both of these factors have been internalized into many of our political institutions. But I am curious to hear his thoughts about the causal mechanisms for the decrease in violence, and, relatedly, what reasons we have for thinking those institutional changes will persist if material factors start changing. (i.e. what if we start experiencing shortages of vital resources?) –Anthony Kammer

A Yes, the decrease in the number of political actors is one of the historical developments that has reduced the number of wars. Europe went from 5,000 political units in the 15th century to 500 in the 17th and about 30 in the 1950s. And holding territory constant, countries have fewer civil wars within their boundaries than interstate wars which cross them, since governments, armies, and police can keep their own citizens from each others’ throats. So fewer jurisdictions meant fewer wars.

The connection with prosperity is more tenuous. Today civil wars are more certainly more common among countries at the very bottom of the economic scale, but the relationship quickly levels off as GDP per capita increases. In the past lots of rich countries waged war, often with each other, and some periods of economic growth (like the 1960s) had far more violence, both international and domestic, than periods of recession, like the 2000s. Surprisingly few wars are fought over control of resources—they are far more often fought over glory, honor, insult, influence, ideology, emotion, security, and other intangibles.

Q The countryside here in Ireland is dotted with the ruins of castles and Iron Age “forts”. I wondered sometimes if this is evidence that ancient or early medieval Ireland was very insecure and violent, considering that today we live in houses that can be easily invaded through the glass windows.

I’m not sure if I am missing other evidence, however. Are the highly defensive dwellings of the past good indicators of the insecurity and violence of that age? –Shane

A Yes; the castles really do reflect the fact that Iron-Age and medieval Europe was more violent than early modern Europe—as Barbara Tuchman wrote, medieval knights fought their private wars “with furious gusto and a single strategy, which consisted in trying to ruin the enemy by killing and maiming as many of his peasants and destroying as many crops, vineyards, tools, barns, and other possessions as possible.”

Q Does the recent history of violence within a society have any near-term predictive value with respect to future violence? That is to say, would a spike in violence due to a revolution or civil war portend a period of less violence, or of more violence? –David

A The timing of interstate wars is completely random (a Poisson process). That is, the outbreak of war does not make the next one more or less likely. There is some evidence that a country that hosts a genocide has an elevated probability of having a recurrence, but it’s tenuous.

Q What about domestic violence? Is that also less so than in the past? And violence towards women in general. –OK

A Yes, spousal abuse, spousal killings, and rape have all substantially decreased in the past few decades.

Q Hello Steven. Given that violence and the tendency towards violence is partly inherited, do you think that public policy should reflect this more? For instance if two parents have a history of violent offending, should their child – without having committed a crime – be given more attention in terms of public policy, than the child of parents who are peaceful non violent hippies. –Luke Allum

A There are many statistical predictors of violence that we choose not to use in our decision-making for moral and political reasons, because the ideal of fairness trumps the ideal of cost-effectiveness. A rational decision-maker using Bayes’ theorem would say, for example, that one should convict a black defendant with less evidence than one needs with a white defendant, because these days the base rates for violence among blacks is higher. Thankfully, this rational policy would be seen as a moral abomination. I suspect that the same sentiments would prevent any policy from pre-judging a child based on the behavior of his parents, whether one thinks the connection is due to genes or to parenting.

Q In The Blank Slate you seemed to touch on this topic- calling it the “Noble Savage.” I would assume this book extends this work greatly. If violence played a big role in our EEA, what problems do you see arising in the future due to this, now that we can seemingly control our environments so well? –Ryan

A Yes, I present extensive statistics showing the non-state peoples (hunter-gatherers, hunter-horticulturalists, pastoralists, and others living outside the control of states) have far higher rates of violence than modern states, even at their worst. I think this very long prehistory of life under anarchy probably selected for motives that can continue to lead to violence today, particularly dominance and revenge, both of which are adaptive in a state of anarchy but not in societies with well-functioning systems for nonviolent dispute resolution. This does not mean that we harbor a thirst for blood which must periodically be discharged—even the most bellicose societies modulate their violence, and can live for decades in peace. Evolution gave us motives that impel us to violence, such as greed, dominance, revenge, and the urge to mete out moralistic punishment, but it also gave us motives that undermine or control the violent inclinations, such as self-control, empathy, and reason—the better angels of our nature.