In 1898, H. G. Wells wrote The War of the Worlds, a novel that imagined the destruction of a great city and the extermination of its inhabitants by ruthless invaders. The invaders in Wells' story were, of course, Martians. But no aliens were needed to make such devastation a reality. In the decades that followed the book's publication, human beings repeatedly played the part of the inhuman marauders, devastating city after city in what may justly be regarded as a single hundred-year "war of the world."

The twentieth century was the bloodiest era in history. World War I killed between 9 million and 10 million people, more if the influenza pandemic of 1918­19 is seen as a consequence of the war. Another 59 million died in World War II. And those conflicts were only two of the more deadly ones in the last hundred years. By one estimate, there were 16 conflicts throughout the last century that cost more than a million lives, a further six that claimed between 500,000 and a million, and 14 that killed between 250,000 and 500,000. In all, between 167 million and 188 million people died because of organized violence in the twentieth century -- as many as one in every 22 deaths in that period.

Other periods matched the twentieth century's rate of killing, if not its magnitude: consider the reigns of tyrants such as Genghis Khan and Tamerlane; some crises in imperial China, such as the An Lushan Rebellion in the eighth century and the Taiping Rebellion in the mid-nineteenth century; and some cases of Western imperial conquest, such as Belgian rule in the Congo and the German war against the Herero in German Southwest Africa. Yet the twentieth century differs from those earlier ages in one key way: it was supposed to be -- and in a great many ways was -- a time of unparalleled material progress.

In real terms, average per capita GDP roughly quadrupled between 1913 and 1998. By the end of the twentieth century, human beings in many parts of the world enjoyed longer and better lives than had been possible at any time before, thanks mainly to improved nutrition and health care. Rising wealth meant that more and more people were able to flee what Karl Marx and Friedrich Engels had called "the idiocy of rural life": between 1900 and 1980, the proportion of the world's population that lived in large cities more than doubled. And by working more productively, people had more time available for leisure. Some spent their free time successfully campaigning for political representation and the redistribution of income. As a result, governments ceased to confine themselves to providing only basic public goods, such as national defense and a fair judicial system, but instead became welfare states that sought nothing less than the elimination of poverty.

It might have been expected that such prosperity would eliminate the causes of war. But much of the worst violence of the twentieth century involved the relatively wealthy countries at the opposite ends of Eurasia. The chief lesson of the twentieth century is that countries can provide their citizens with wealth, longevity, literacy, and even democracy but still descend into lethal conflict. Leon Trotsky nicely summed up the paradox when reflecting on the First Balkan War of 1912­13, which he covered as a reporter. The conflict, Trotsky wrote, "shows that we still haven't crawled out on all fours from the barbaric stage of our history. We have learned to wear suspenders, to write clever editorials, and to make chocolate milk, but when we have to decide seriously a question of the coexistence of a few tribes on a rich peninsula of Europe, we are helpless to find a way other than mutual mass slaughter." Trotsky later made his own contribution to the history of mass slaughter as the people's commissar for war and as the commander of the Red Army during the Russian Civil War.

Will the twenty-first century be as bloody as the twentieth? The answer depends partly on whether or not we can understand the causes of the last century's violence. Only if we can will we have a chance of avoiding a repetition of its horrors. If we cannot, there is a real possibility that we will relive the nightmare.


There are many unsatisfactory explanations for why the twentieth century was so destructive. One is the assertion that the availability of more powerful weapons caused bloodier conflicts. But there is no correlation between the sophistication of military technology and the lethality of conflict. Some of the worst violence of the century -- the genocides in Cambodia in the 1970s and central Africa in the 1990s, for instance -- was perpetrated with the crudest of weapons: rifles, axes, machetes, and knives.

Nor can economic crises explain the bloodshed. What may be the most familiar causal chain in modern historiography links the Great Depression to the rise of fascism and the outbreak of World War II. But that simple story leaves too much out. Nazi Germany started the war in Europe only after its economy had recovered. Not all the countries affected by the Great Depression were taken over by fascist regimes, nor did all such regimes start wars of aggression. In fact, no general relationship between economics and conflict is discernible for the century as a whole. Some wars came after periods of growth, others were the causes rather than the consequences of economic catastrophe, and some severe economic crises were not followed by wars.

Many trace responsibility for the butchery to extreme ideologies. The Marxist historian Eric Hobsbawm calls the years between 1914 and 1991 "an era of religious wars" but argues that "the most militant and bloodthirsty religions were secular ideologies." At the other end of the political spectrum, the conservative historian Paul Johnson blames the violence on "the rise of moral relativism, the decline of personal responsibility [and] the repudiation of Judeo-Christian values." But the rise of new ideologies or the decline of old values cannot be regarded as causes of violence in their own right. Extreme belief systems, such as anti-Semitism, have existed for most of modern history, but only at certain times and in certain places have they been widely embraced and translated into violence.

And as tempting as it is to blame tyrants such as Hitler, Stalin, and Mao for the century's bloodletting, to do so is to repeat the error on which Leo Tolstoy heaped so much scorn in War and Peace. Megalomaniacs may order men to invade Russia, but why do the men obey? Some historians have attempted to answer the novelist's question by indicting the modern nation-state. The nation-state does indeed possess unprecedented capabilities for mobilizing masses of people, but those means could just as easily be harnessed, and have been, to peaceful ends.

Others seek the cause of conflict in the internal political arrangements of states. It has become fashionable among political scientists to posit a causal link between democracy and peace, extrapolating from the observation that democracies tend not to go to war with one another. The corollary, of course, is that dictatorships generally are more bellicose. By that logic, the rise of democracy during the twentieth century should have made the world more peaceful. Democratization may well have reduced the incidence of war between states. But waves of democratization in the 1920s, 1960s, and 1980s seem to have multiplied the number of civil wars. Some of those (such as the conflicts in Afghanistan, Burundi, China, Korea, Mexico, Mozambique, Nigeria, Russia, Rwanda, and Vietnam) were among the deadliest conflicts of the century. Horrendous numbers of fatalities were also caused by genocidal or "politicidal" campaigns waged against civilian populations, such as those carried out by the Young Turks against the Armenians and the Greeks during World War I, the Soviet government from the 1920s until the 1950s, and the Nazis between 1933 and 1945 -- to say nothing of those perpetrated by the communist tyrannies of Mao in China and Pol Pot in Cambodia. Indeed, such civil strife has been the most common form of conflict during the past 50 years. Of the 24 armed conflicts recorded as "ongoing" by the University of Maryland's Ted Robert Gurr and George Mason University's Monty Marshall in early 2005, nearly all were civil wars.

Conventional explanations for the violence of the twentieth century are inadequate for another important reason. None is able to explain convincingly why lethal conflict happened when and where it did. Ultimately, the interesting question is not, Why was the twentieth century more violent than the eighteenth or the nineteenth? but, Why did extreme violence happen in Poland and Serbia more than in Portugal and Sweden, and why was it more likely to happen between 1939 and 1945 than between 1959 and 1965?

In relative terms, Poland and Serbia were exceptionally prone to the ravages of warfare in the first half of the twentieth century. Polish Galicia was one of the killing fields of the eastern front between 1914 and 1917. Serbia had the highest mortality rate of any country during World War I; the war killed just under 6 percent of the country's prewar population. Far worse, the Polish mortality rate in World War II amounted to just under 19 percent; Polish Jews killed in the Holocaust accounted for a large share of that number. The Soviet rate was 11 percent, although the figure for Russians and Ukrainians, who dominated the ranks of the Red Army, was higher. Among other combatant countries, only Germany (including Austria) and Yugoslavia suffered war-induced mortality rates close to 10 percent. The next highest rates were for Hungary (8 percent) and Romania (6 percent). In no other country for which figures have been published did war mortality rise above 3 percent. Indeed, for four of the principal combatants -- France, Italy, the United Kingdom, and the United States -- total wartime mortality was less than 1 percent of the prewar population.

These figures give a good indication of the location and the timing of the worst twentieth-century violence. From around 1904 to 1953, the most dangerous place in the world was the triangle that lies between the Balkans, the Baltic Sea, and the Black Sea. Only slightly less dangerous over the same period was the region at the other end of Eurasia comprising Manchuria and the neighboring Korean Peninsula. Indeed, it was there that the era of large-scale modern warfare began, when Japan attacked Russia in 1904. It was also there that the era came to a close 50 years later, with the end of the Korean War. Thereafter, the location and the type of violence changed. Despite its reputation as having been a "long peace," the Cold War sparked a series of bitter proxy wars around the world, particularly in Central America, sub-Saharan Africa, and Indochina. Those conflicts in effect constituted a Third World War -- or, more aptly, the Third World's War.


Three factors explain the timing and the location of the extreme violence of the twentieth century: ethnic disintegration, economic volatility, and empires in decline.

Patterns of migration combined with the persistence of religious and cultural traditions have always made some areas of human settlement more ethnically homogeneous than others. In 1900, in parts of the world such as England or France it was possible to travel hundreds of miles without encountering anyone who looked, spoke, dressed, or worshiped in significantly different ways from everyone else. But an ethnic and linguistic map of central and eastern Europe at that time would have resembled a patchwork quilt, reflecting the region's diversity. There, ethnic groups lived next door to one another -- sometimes literally. The many-named town known variously as Czernowitz, Cernauti, and Chernivtsi is a case in point. Located in what is now Ukraine, Chernivtsi was once a multiethnic city inhabited by German professors, Austrian civil servants, Jewish merchants, and workers from many other ethnicities. To the nineteenth-century poet Karl Emil Franzos, Chernivtsi was Czernowitz, "the courtyard of the German paradise," an island of Kultur within "Half Asia." Yet the multiethnic balance on which such places rested tipped over disastrously in the three decades after 1914.

Conflict in multiethnic societies was not and is not inevitable. In fact, central and eastern Europe in the early 1900s saw remarkable advances in the integration and assimilation of ethnic minorities. People of different ethnicities not only lived near one another but also worked, played, and studied together. They even intermarried. For example, in cities such as Hamburg and Trieste, rates of intermarriage between Jews and non-Jews reached remarkable heights. The same trend was apparent after the 1917 revolution in what soon became the Soviet Union, particularly in big cities.

Assimilation, however, can be violently reversed. It is no accident that ethnic tensions increased in places such as the then Romanian city Cernauti in the interwar period. Before World War I, four dynasties -- the Hapsburg, the Hohenzollern, the Ottoman, and the Romanov -- had governed central and eastern Europe. Such regimes cared less about their subjects' nationality than about their subjects' loyalty. After 1918, the political map of the region was redrawn to create or re-create nation-states. Yet the region's ethnic diversity made it impossible to set up homogeneous polities. As a result, in Czechoslovakia, Poland, Romania, and Yugoslavia, the majority population in each accounted for less than 80 percent of the total population. Ethnic minorities in such countries suddenly found themselves treated as second-class citizens. As part of postwar Romania, for instance, Cernauti became the scene of escalating ethnic conflict because of systematic discrimination against German educational institutions and landowners by the Romanian authorities. This was how ethnic disintegration happened.

Economic volatility exacerbated political frictions. The frequency and the amplitude of changes in output and prices peaked between 1919 and 1939. During those years, measures of standard deviation for growth and inflation were nearly double what they were in the preceding and succeeding periods and roughly seven times what they have been since 1990. From the mid-1920s to the mid-1930s, stock markets also experienced their highest levels of volatility of the century. Although it is obvious that low growth or a recession contributes to social instability, rapid growth can also be destabilizing. This is especially true in multiethnic societies, where booms can appear to benefit market-dominant minorities disproportionately, such as the Armenians in Turkey in the early 1900s or the Jews in central and eastern Europe. When booms turn to busts, the prosperous minority can become the target of reprisals by the impoverished majority. As Gregor von Rezzori recalled of his youth in Cernauti during this period, all of the city's other ethnic groups "despised the Jews, notwithstanding that Jews played an economically decisive role." If they could agree on nothing else, Germans and Romanians could agree that Jews were "the natural target of their aggression," Rezzori wrote in his memoir The Snows of Yesteryear.

Of course, not all multiethnic societies that experienced economic volatility in those years descended into sectarian violence. In the still relatively segregated United States, for example, the number of violent acts of racial hatred, such as lynchings, declined between the world wars. The critical third factor determining both the location and the timing of twentieth-century violence was the decline and fall of empires.

The twentieth century was characterized by a remarkably high rate of imperial dissolution. In 1913, around 65 percent of the world's land and 82 percent of its population were under some kind of imperial rule. Those empires soon disintegrated. The Qing dynasty in China was overthrown before World War I, and in 1917 the Romanov dynasty fell from power. They were quickly followed by the Hapsburgs, the Hohenzollerns, and the Ottomans. Little more than two decades later, the British, Dutch, and French empires in Asia were dealt heavy blows by imperial Japan, leaving damage that not even the Allied victory in World War II could repair. The Portuguese empire limped on, but by the early 1970s it too had collapsed. The new empires that grew among the ruins of the old were shorter lived than their predecessors. Whereas some early modern empires lasted for centuries (Ottoman rule, for instance, endured for 469 years), the Soviet Union fell apart after only 69 years, the Japanese overseas empire crumbled after just 50 years, and Hitler's empire beyond Germany's borders hung on for barely six years.

Violence is most likely to occur as empires decline. The late Romans understood well the unpleasantness associated with imperial dissolution. Modern commentators, on the other hand, have generally been too eager to see empires end and too credulous about the benefits of "self-determination" to realize the potentially high costs of any transition from a multiethnic polity to a homogeneous one. As imperial authority crumbles, local elites compete for the perquisites of power. The stakes are particularly high in ethnically heterogeneous provinces. From the point of view of minorities, the prospect of a new political order can be deeply alarming; members of a minority group that has collaborated with the imperial power frequently find that the empire's disappearance leaves them vulnerable to reprisals.

Anyone who doubts whether imperial decline can foment conflict need only reflect on how rarely empires broke up peacefully in the twentieth century. One of the last acts of the Ottoman Empire was to attempt genocide against the Armenians. The Austrian, German, and Russian empires all met bloody ends as World War I drew to a close; indeed, the civil war and ethnic strife that followed the Russian Revolution in 1917 was as costly to the tsar's former subjects as the preceding war against Austria and Germany was. The most violent year in the entire history of British India was 1947, when partition led to the deaths of more than a million people in communal clashes between Hindus and Muslims. Even the relatively calm dissolution of the Soviet Union bred bitter conflict between Armenia and Azerbaijan over the mainly Armenian enclave of Nagorno-Karabakh, as well as the war between Chechen separatists and the Russian government.


If the combination of ethnic disintegration, economic volatility, and empires in decline is the basic formula for twentieth-century conflict, then what are the implications for the twenty-first century?

The good news is that global economic volatility has been significantly lower in recent years than at almost any time in the last century. By widening and deepening international markets for goods, labor, and capital, globalization appears to have made the world economy less prone to crisis. At the same time, financial innovations have improved the pricing and the distribution of risk, and policy innovations such as inflation targeting have helped governments to limit rises in consumer price (if not asset price) inflation. International organizations such as the World Trade Organization and the International Monetary Fund have helped to avert trade disputes and other sources of economic instability.

A second obvious point is that the old zones of conflict are unlikely to be the new ones. Those who engaged in ethnic cleansing in the twentieth century, whether by forced migration or genocide, did their work too well, so that today central and eastern Europe (and Manchuria and Korea) are no longer ethnically heterogeneous. Chernivtsi today is overwhelmingly Ukrainian, for instance, with only a tiny remnant of its once-thriving Jewish community still living there. And third, the world of 2006 is supposed to be a world without empires; the danger of imperial decline should accordingly be much less than in the past century.

Unfortunately, these appearances are deceptive. Today, one region displays in abundance all of the characteristics of the worst conflict zones of the twentieth century. Economic volatility has remained pronounced there even as it has diminished in the rest of the world. An empire (albeit one that dares not speak its name) is losing its grip over the region. Worst of all, ethnic disintegration is already well under way, even though many commentators still conceive of what is currently the main conflict there as an insurgency against foreign invaders or a "clash of civilizations" between Islam and the West.

That place is the Middle East.

Iraq has experienced severe economic volatility in recent years, first as a result of UN sanctions and then because of the U.S. invasion and the subsequent insecurity. Having declined by just under 8 percent in the last year of Saddam Hussein's reign, real GDP plummeted by more than 40 percent in 2003, the year the United States invaded. In 2004, output bounced back by an estimated 46 percent -- but growth slowed last year to under 4 percent. That is about as extreme as volatility can get. Since oil production accounts for about two-thirds of Iraq's GDP, changes in the production and the price of oil have been the primary drivers of these economic swings. Oil prices are now up, meaning Iraq's exports are worth more, but inflation is stuck between 20 and 30 percent, and oil production is running at just 16 percent of its prewar level because of sabotage and other problems.

Other indicators are only slightly less dire. Power generation remains 28 percent below the target that was supposed to have been achieved in 2004. And the unemployment rate is estimated to be between 25 and 40 percent. It is true that there has been some economic good news since the war. More Iraqis now have cars, and many more Iraqis now have telephones thanks to the creation of a cell-phone network. Iraqis today also have a free press. Yet these improvements in communications have also helped insurgents and militias to mobilize.

Iraq is not the only Middle Eastern country with a volatile economy. All six countries that border Iraq have seen their own share of ups and downs. In the past five years, per capita GDP growth in Iran has ranged between 0.3 and 7.3 percent; in Jordan, between 0.3 and 5.1 percent; in Kuwait, between -5.5 and 6.9 percent; in Saudi Arabia, between -6.0 and 3.3 percent; in Syria, between -0.4 and 3.3 percent; and in Turkey, between -9.0 and 7.4 percent. Over the past decade, Kuwait's average annual GDP performance was the lowest (-0.9 percent), while Iran's was the highest (3.0 percent). On the other hand, the average Kuwaiti is nearly ten times richer than the average Iranian. Inflation in the region in 2004 ranged from 0.3 percent in Saudi Arabia to nearly 15 percent in Iran. Unemployment rates are in the low double digits everywhere but in Saudi Arabia (where it is around 5 percent), but youth unemployment rates are generally twice as high -- just under 20 percent in Turkey, for example, compared with an overall rate of 10 percent. Youth unemployment matters because these are relatively youthful societies. Roughly 20 percent of the population of Iraq and its neighbors is between 15 and 24 years old; the equivalent figure for Europe and the United States is about 14 percent.

The Middle East also has the misfortune to be a zone of imperial conflict. Most Americans will probably always reject the proposition that the United States is (or operates) a de facto empire. Such squeamishness may be an integral part of the U.S. empire's problem. To be an empire in denial means resenting the costs of intervening in the affairs of foreign peoples and underestimating the benefits of doing so. It is remarkable that in June 2004, just over a year after U.S. troops toppled Saddam, a majority of Americans already said they regarded the invasion of Iraq as a mistake. Although support for the war has vacillated since then, at no time since September 2004 have more than 55 percent of poll respondents said they approved of it. What makes the U.S. public's misgivings especially remarkable is that the number of U.S. military personnel who have died in this war has been very small by historical standards. The total as of mid-July 2006 was 2,544, of whom 525 had died as a result of non-combat-related causes.

The Islamic Republic of Iran, by contrast, is heir to an imperial tradition dating back to the time of the Safavid dynasty, which ruled Persia from 1501, and beyond. Although Iran's leaders prefer the rhetoric of religious revolution and national liberation, the historian cannot fail to detect in their long-held ambition to acquire nuclear weapons -- and thereby dominate the Middle East -- a legacy of Persia's imperial past. The Iranian president, Mahmoud Ahmadinejad, is not only a devotee of the Hidden, or Twelfth, Imam (who devout Shiites believe will return to the world as the Mahdi, or messiah, for a final confrontation with the forces of evil); he is also a war veteran at the head of a youthful nation. It is not wholly fanciful to see in him a potential Caesar or Bonaparte.

The third and most striking feature of the Middle East today is the acceleration of ethnic disintegration, which is most obvious in Iraq. In 1993, Harvard's Samuel Huntington predicted that in the post­Cold War world, "the principal conflicts of global politics [would] occur between nations and groups of different civilizations," particularly between a decadent "Judeo-Christian" West and a demographically ascendant Islamic civilization. For a time, events seemed to be fulfilling his prophecy. Many Americans interpreted the terrorist attacks of September 11, 2001, in Huntington's terms, while Islamists interpreted the U.S. invasions of Afghanistan and Iraq as wars by Christian "crusaders" against Muslims.

Yet a closer inspection of events since 1993 suggests a post­Cold War trend of clashes within, rather than between, civilizations. Of the 30 major armed conflicts that either are going on now or have recently ended, only nine can be regarded as being in any sense between civilizations. But 19 are in some measure ethnic conflicts, the worst being the wars that continue to bedevil central Africa. Moreover, most of those conflicts that have a religious dimension are also ethnic conflicts.

Events in Iraq suggest that there, too, what is unfolding is not a clash between the West and Islam but, increasingly, a clash within Islamic civilization itself. By some accounts, ethnic disintegration there is already well under way. In a June 6, 2006, cable to Secretary of State Condoleezza Rice, which was leaked to the press, Zalmay Khalilzad, the U.S. ambassador to Iraq, cataloged the evidence of mounting sectarian tension in and around the Green Zone in Baghdad. "Personal fears are reinforcing divisive or sectarian channels," Khalilzad wrote. "Ethnic and sectarian faultlines are becoming part of the daily media fare. One Shia employee told us she can no longer watch TV news with her mother, who is a Sunni, because her mother blamed all the government failings on the fact that the Shia are in charge." Even more worrisome, Khalilzad reported that U.S. embassy employees had "become adept in modifying [their] behavior to avoid Alasas, informants who keep an eye out for 'outsiders' in neighborhoods. The Alasa mentality is becoming entrenched as Iraqi security forces fail to gain public confidence."

Such news should come as no surprise. Baghdad and the provinces around it -- Babil, Diyala, and Salahuddin -- are precisely the regions of Iraq where ethnic conflict could have been predicted to occur given the current conditions of economic volatility and imperial crisis. They are the most ethnically mixed parts of the country, where Sunnis and Shiites or Sunnis and Kurds live cheek by jowl. Under Saddam's secular tyranny, these communities coexisted more or less peacefully. Anecdotal evidence even suggests that there was some intermarriage. Since Saddam's fall, however, integration has been reversed. Around 92 percent of the votes in the December 2005 election were cast for sectarian parties. In such a fractious atmosphere, the odds seem dangerously stacked against the once-dominant Sunni Arab minority. Not only do current constitutional arrangements raise the prospect that Iraq's oil revenue will flow largely to Shiite and Kurdish provinces, but the Sunnis are badly underrepresented in the Iraqi security forces. According to the Brookings Institution, Sunnis make up less than ten percent of the enlisted forces.

Although the atmosphere has cooled somewhat since the Askariya shrine bombing in February, the trend is clearly toward ethnic or sectarian conflict. The number of incidents of sectarian violence recorded in May 2006 was 250, compared with just 20 in the same month the year before and ten in May 2004. There has been a threefold increase in the homicide rate in Baghdad since February 2006, much of it the result of sectarian violence. The assassination of Abu Musab al-Zarqawi, the leader of al Qaeda in Iraq, in June was therefore less of a milestone than President George W. Bush had hoped. As insurgency has given way to civil war, Zarqawi had already ceased to be a key player.


What makes the escalating civil war in Iraq so disturbing is that it has the potential to spill over into neighboring countries. The Iranian government is already taking more than a casual interest in the politics of post-Saddam Iraq. And yet Iran, with its Sunni and Kurdish minorities, is no more homogeneous than Iraq. Jordan, Saudi Arabia, and Syria cannot be expected to look on insouciantly if the Sunni minority in central Iraq begins to lose out to what may seem to be an Iranian-backed tyranny of the majority. The recent history of Lebanon offers a reminder that in the Middle East there is no such thing as a contained civil war. Neighbors are always likely to take an unhealthy interest in any country with fissiparous tendencies.

The obvious conclusion is that a new "war of the world" may already be brewing in a region that, incredible though it may seem, has yet to sate its appetite for violence. And the ramifications of such a Middle Eastern conflagration would be truly global. Economically, the world would have to contend with oil at above $100 a barrel. Politically, those countries in western Europe with substantial Muslim populations might also find themselves affected as sectarian tensions radiated outward. Meanwhile, the ethnic war between Jews and Arabs in Israel, the Gaza Strip, and the West Bank shows no sign of abating. Is it credible that the United States will remain unscathed if the Middle East erupts?

Although such an outcome may seem to be a low-probability, nightmare scenario, it is already more likely than the scenario of enduring peace in the region. If the history of the twentieth century is any guide, only economic stabilization and a credible reassertion of U.S. authority are likely to halt the drift toward chaos. Neither is a likely prospect. On the contrary, the speed with which responsibility for security in Iraq is being handed over to the predominantly Shiite and Kurdish security forces may accelerate the descent into internecine strife. Significantly, the audio statement released by Osama bin Laden in June excoriated not only the American-led "occupiers" of Iraq but also "certain sectors of the Iraqi people -- those who refused [neutrality] and stood to fight on the side of the crusaders." His allusions to "rejectionists," "traitors," and "agents of the Americans" were clearly intended to justify al Qaeda's policy of targeting Iraq's Shiites.

The war of the worlds that H. G. Wells imagined never came to pass. But a war of the world did. The sobering possibility we urgently need to confront is that another global conflict is brewing today -- centered not on Poland or Manchuria, but more likely on Palestine and Mesopotamia.

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now
  • Niall Ferguson is Laurence A. Tisch Professor of History at Harvard University and a Senior Fellow at the Hoover Institution at Stanford University. His latest book is "The War of the World." Copyright 2006 by Niall Ferguson.
  • More By Niall Ferguson