• Zero is the number of times that nuclear weapons have been used in conflict. Five great powers possess them, and all of them have waged wars. Yet no nuclear device has been set off in anger. It's not just that the great powers avoided the mutual suicide of an all-out nuclear war. They also avoided using the smaller, "tactical" nuclear weapons, many of them comparable to conventional explosives, on the battlefield or in the bombing of enemy facilities. And the United States refrained from using its nuclear arsenal in the late 1940s when it held a nuclear monopoly and did not have to worry about mutually assured destruction. I've been quantifying violence throughout this book using proportions. If one were to calculate the amount of destruction that nations have actually perpetrated as a proportion of how much they could perpetrate, given the destructive capacity available to them, the postwar decades would be many orders of magnitudes more peaceable than any time in history.

I have spent a lot of this chapter on the statistics of war. But now we are ready for the most interesting statistic since 1945: zero. Zero is the number that applies to an astonishing collection of categories of war during the two-thirds of a century that has elapsed since the end of the deadliest war of all time. I'll begin with the most momentous.

None of this was a foregone conclusion. Until the sudden end of the Cold War, many experts (including Albert Einstein, C. P. Snow, Herman Kahn, Carl Sagan, and Jonathan Schell) wrote that thermonuclear doomsday was likely, if not inevitable. The eminent international studies scholar Hans Morgenthau, for example, wrote in 1979, "The world is moving ineluctably towards a third world war -- a strategic nuclear war. I do not believe that anything can be done to prevent it." The Bulletin of the Atomic Scientists, according to its Web site, aims to "inform the public and influence policy through in-depth analyses, op-eds, and reports on nuclear weapons." Since 1947 it has published the famous Doomsday Clock, a measure of "how close humanity is to catastrophic destruction -- the figurative midnight." The clock was unveiled with its minute hand pointing at 7 minutes to midnight, and over the next sixty years it was moved back and forth a number of times between 2 minutes to midnight (in 1953) and 17 minutes to midnight (in 1991). In 2007 the Bulletin apparently decided that a clock with a minute hand that moved two minutes in sixty years was due for an adjustment. But rather than tuning the mechanism, they redefined midnight. Doomsday now consists of "damage to ecosystems, flooding, destructive storms, increased drought, and polar ice melt." This is a kind of progress.

• Zero is the number of times that the two Cold War superpowers fought each other on the battlefield. To be sure, they occasionally fought each other's smaller allies and stoked proxy wars among their client states. But when either the United States or the Soviet Union sent troops to a contested region (Berlin, Hungary, Vietnam, Czechoslovakia, Afghanistan), the other stayed out of its way. The distinction matters a great deal because as we have seen, one big war can kill vastly more people than many small wars. In the past, when an enemy of a great power invaded a neutral country, the great power would express its displeasure on the battlefield. In 1979, when the Soviet Union invaded Afghanistan, the United States expressed its displeasure by withdrawing its team from the Moscow Summer Olympics. The Cold War, to everyone's surprise, ended without a shot in the late 1980s shortly after Mikhail Gorbachev ascended to power. It was followed by the peaceful tear-down of the Berlin Wall and then by the mostly peaceful collapse of the Soviet Union.

• Zero is the number of interstate wars that have been fought between countries in Western Europe since the end of World War II. It is also the number of interstate wars that have been fought in Europe as a whole since 1956, when the Soviet Union briefly invaded Hungary. Keep in mind that up until that point European states had started around two new armed conflicts a year since 1400.

• Zero is the number of times that any of the great powers have fought each other since 1953 (or perhaps even 1945, since many political scientists don't admit China to the club of great powers until after the Korean War). The war-free interval since 1953 handily breaks the previous two records from the 19th century of 38 and 44 years. In fact, as of May 15, 1984, the major powers of the world had remained at peace with one another for the longest stretch of time since the Roman Empire. Not since the 2nd century BCE, when Teutonic tribes challenged the Romans, has a comparable interval passed without an army crossing the Rhine.

• Zero is the number of developed countries that have expanded their territory since the late 1940s by conquering another country. No more Poland getting wiped off the map, or Britain adding India to its empire, or Austria helping itself to the odd Balkan nation. Zero is also the number of times that any country has conquered even parts of some other country since 1975, and it is not far from the number of permanent conquests since 1948 (a development we'll soon examine more closely). In fact the process of great power aggrandizement went into reverse. In what has been called "the greatest transfer of power in world history," European countries surrendered vast swaths of territory as they closed down their empires and granted independence to colonies, sometimes peacefully, sometimes because they had lost the will to prevail in colonial wars. As we will see in the next chapter, two entire categories of war -- the imperial war to acquire colonies, and the colonial war to keep them -- no longer exist.

• Zero is the number of interstate wars that have been fought since 1945 between major developed countries (the forty-four with the highest per capita income) anywhere in the world (again, with the exception of the 1956 Hungarian invasion). Today we take it for granted that war is something that happens in smaller, poorer, and more backward countries. But the two world wars, together with the many hyphenated European wars from centuries past (Franco-Prussian, Austro-Prussian, Russo-Swedish, British-Spanish, AngloDutch) remind us that this was not always the way things worked.

The point of this chapter is that these zeroes -- the Long Peace -- are a result of one of those psychological retunings that take place now and again over the course of history and cause violence to decline. In this case it is a change within the mainstream of the developed world (and increasingly, the rest of the world) in the shared cognitive categorization of war. For most of human history, influential people who craved power, prestige, or vengeance could count on their political network to ratify those cravings and to turn off their sympathies for the victims of an effort to satisfy them. They believed, in other words, in the legitimacy of war. Though the psychological components of war have not gone away -- dominance, vengeance, callousness, tribalism, groupthink, self-deception -- since the late 1940s they have been disaggregated in Europe and other developed countries in a way that has driven down the frequency of war.

• Zero is the number of internationally recognized states since World War II that have gone out of existence through conquest. (South Vietnam may be the exception, depending on whether its unification with North Vietnam in 1975 is counted as a conquest or as the end of an internationalized civil war.) During the first half of the 20th century, by comparison, twenty-two states were occupied or absorbed, at a time when the world had far fewer states to begin with. Though scores of nations have gained independence since 1945, and several have broken apart, most of the lines on a world map of 1950 are still present on a world map in 2010. This too is an extraordinary development in a world in which rulers used to treat imperial expansion as part of their job description.

A long peace, to be sure, is not a perpetual peace. No one with a statistical appreciation of history could possibly say that a war between great powers, developed countries, or European states will never happen again. But probabilities can change over spans of time that matter to us. The house odds on the iron dice can decline; the power-law line can sink or tilt. And in much of the world, that appears to have happened.

Some people downplay these stunning developments by pointing out that wars still take place in the developing world, so perhaps violence has only been displaced, not reduced. In the following chapter we will examine armed conflict in the rest of the world, but for now it's worth noting that the objection makes little sense. There is no Law of Conservation of Violence, no hydraulic system in which a compression of violence in one part of the world forces it to bulge out somewhere else. Tribal, civil, private, slave-raiding, imperial, and colonial wars have inflamed the territories of the developing world for millennia. A world in which war continues in some of the poorer countries is still better than a world in which it takes place in both the rich and the poor countries, especially given the incalculably greater damage that rich, powerful countries can wreak.

The same statistical consciousness, though, alerts us to alternative possibilities. Perhaps the odds haven't changed at all, and we're overinterpreting a random run of peaceful years in the same way that we are liable to overinterpret a random cluster of wars or atrocities. Perhaps the pressure for war has been building and the system will blow at any moment.

But probably not. The statistics of deadly quarrels show that war is not a pendulum, a pressure cooker, or a hurtling mass, but a memoryless game of dice, perhaps one with changing odds. And the history of many nations affirms that a peace among them can last indefinitely. As Mueller puts it, if war fever were cyclical, "one would expect the Swiss, Danes, Swedes, Dutch, and Spaniards to be positively roaring for a fight by now." Nor are Canadians and Americans losing sleep about an overdue invasion across the world's longest undefended border.

What about the possibility of a run of good luck? Also unlikely. The postwar years are by far the longest period of peace among great powers since they came into being five hundred years ago. The stretch of peace among European states is also the longest in its bellicose history. Just about any statistical test can confirm that the zeroes and near zeroes of the Long Peace are extremely improbable, given the rates of war in the preceding centuries. Taking the frequency of wars between great powers from 1495 to 1945 as a baseline, the chance that there would be a sixty-five-year stretch with only a single great power war (the marginal case of the Korean War) is one in a thousand. Even if we take 1815 as our starting point, which biases the test against us by letting the peaceful post-Napoleonic 19th century dominate the base rate, we find that the probability that the postwar era would have at most four wars involving a great power is less than 0.004, and the probability that it would have at most one war between European states (the Soviet invasion of Hungary in 1956) is 0.0008.

The calculation of probabilities, to be sure, critically depends on how one defines the events. Odds are very different when you estimate them in full knowledge of what happened (a post hoc comparison, also known as "data snooping") and when you lay down your prediction beforehand (a planned or a priori comparison). Recall that the chance that two people in a room of fifty-seven will share a birthday is ninety-nine out of a hundred. In that case we are specifying the exact day only after we identify the pair of people. The chance that someone will share my birthday is less than one in seven; in that case we specify the day beforehand. A stock scammer can exploit the distinction by sending out newsletters with every possible prediction about the trajectory of the market. Several months later the fraction of recipients that got the lucky matching run will think he is a genius. A skeptic of the Long Peace could claim that anyone making a big deal of a long run of nonwars at the end of that very run is just as guilty of data snooping.

But in fact there is a paper trail of scholars who, more than two decades ago, noticed that the war-free years were piling up and attributed it to a new mindset that they expected to last. Today we can say that their a priori predictions have been confirmed. The story can be told in titles and dates: Werner Levi's The Coming End of War (1981), John Gaddis's "The Long Peace: Elements of Stability in the Postwar International System" (1986), Kalevi Holsti's "The Horsemen of the Apocalypse: At the Gate, Detoured, or Retreating?" (1986), Evan Luard's The Blunted Sword: The Erosion of Military Power in Modern World Politics (1988), John Mueller's Retreat from Doomsday: The Obsolescence of Major War (1989), Francis Fukuyama's "The End of History?" (1989), James Lee Ray's "The Abolition of Slavery and the End of International War" (1989), and Carl Kaysen's "Is War Obsolete?" (1990). In 1988 the political scientist Robert Jervis captured the phenomenon they were all noticing:

The most striking characteristic of the postwar period is just that -- it can be called "postwar" because the major powers have not fought each other since 1945. Such a lengthy period of peace among the most powerful states is unprecedented.

It is clear that a profound transformation of the international structure in Europe -- and the whole world -- is underway. In the past, such changes have regularly been consummated by war. The argument presented in this essay supports the prediction that this time the changes can take place without war (although not necessarily without domestic violence in the states concerned). So far -- mid-January -- so good. The author and his readers will be eagerly and anxiously testing the prediction each day.

These scholars were confident that they were not being fooled by a lucky run but were putting their finger on an underlying shift that supported predictions about the future. In early 1990, Kaysen added a last-minute postscript to his review of Mueller's 1989 book in which he wrote:

Precocious assessments of the obsolescence of interstate war are especially poignant when they come from military historians. These are the scholars who have spent their lives immersed in the annals of warfare and should be most jaded about the possibility that this time it's different. In his magnum opus A History of Warfare, John Keegan (the military historian who is so habitually called "distinguished" that one could be forgiven for thinking it is part of his name) wrote in 1993:

And the no-less-distinguished Evan Luard, our guide to six centuries of war, had written still earlier, in 1986:

[It has become] quite possible that war in the sense of major, organized armed conflict between highly developed societies may not recur, and that a stable framework for international order will become firmly established.

War, it seems to me, after a lifetime of reading about the subject, mingling with men of war, visiting the sites of war and observing its effects, may well be ceasing to commend itself to human beings as a desirable or productive, let alone rational, means of reconciling their discontents.

The equally distinguished Michael Howard had already written, in 1991:

Most startling of all has been the change that has come about in Europe, where there has been a virtual cessation of international warfare… Given the scale and frequency of war during the preceding centuries in Europe, this is a change of spectacular proportions: perhaps the single most striking discontinuity that the history of warfare has anywhere provided.

Among affluent liberal democracies… a true state of peace appears to have developed, based on genuine mutual confidence that war between them is practically eliminated even as an option. Nothing like this had ever existed in history.

More than two decades later, none of them would have a reason to change his assessment. In his 2006 book War in Human Civilization, a military history that is more sweeping than its predecessors and salted with the Hobbesian realism of evolutionary psychology, Azar Gat wrote: