CHAPTER 13

TERRORISM

When I wrote in the preceding chapter that we are living in the safest time in history, I was aware of the incredulity those words would evoke. In recent years, highly publicized terrorist attacks and rampage killings have set the world on edge and fostered an illusion that we live in newly dangerous times. In 2016, a majority of Americans named terrorism as the most important issue facing the country, said they were worried that they or a family member would be a victim, and identified ISIS as a threat to the existence or survival of the United States.¹ The fear has addled not just ordinary citizens trying to get a pollster off the phone but public intellectuals, especially cultural pessimists perennially hungry for signs that Western civilization is (as always) on the verge of collapse. The political philosopher John Gray, an avowed progressophobe, has described the contemporary societies of Western Europe as “terrains of violent conflict” in which “peace and war [are] fatally blurred.”²

But yes, all this is an illusion. Terrorism is a unique hazard because it combines major dread with minor harm. I will not count trends in terrorism as an example of progress, since they don’t show the long-term decline we’ve seen for disease, hunger, poverty, war, violent crime, and accidents. But I will show that terrorism is a distraction in our assessment of progress, and, in a way, a backhanded tribute to that progress.

Gray dismissed actual data on violence as “amulets” and “sorcery.” The following table shows why he needed this ideological innumeracy to prosecute his jeremiad. It shows the number of victims of four categories of killing—terrorism, war, homicide, and accidents—together with the total of all deaths, in the most recent year for which data are available (2015 or earlier). A graph is impossible, because swatches for the terrorism numbers would be smaller than a pixel.

Table 13-1: Deaths from Terrorism, War, Homicide, and Accidents

“Western Europe” is defined as in the Global Terrorism Database, comprising 24 countries and a 2014 population of 418,245,997 (Statistics Times 2015). I omit Andorra, Corsica, Gibraltar, Luxembourg, and the Isle of Man.

Sources: Terrorism (2015): National Consortium for the Study of Terrorism and Responses to Terrorism 2016. War, US and Western Europe (UK + NATO) (2015): icasualties.org, http://icasualties.org. War, World (2015): UCDP Battle-Related Deaths Dataset, Uppsala Conflict Data Program 2017. Homicide, US (2015): Federal Bureau of Investigation 2016a. Homicide, Western Europe and World (2012 or most recent): United Nations Office on Drugs and Crime 2013. Data for Norway exclude the Utøya terrorist attack. Motor vehicle accidents, All accidents, and All deaths, US (2014): Kochanek et al. 2016, table 10. Motor vehicle accidents, Western Europe (2013): World Health Organization 2016c. All accidents, Western Europe (2014 or most recent): World Health Organization 2015a. Motor vehicle accidents and All accidents, World (2012): World Health Organization 2014. All deaths, Western Europe (2012 or most recent): World Health Organization 2017a. All deaths, World (2015): World Health Organization 2017c.

Start with the United States. What jumps out of the table is the tiny number of deaths in 2015 caused by terrorism compared with those from hazards that inspire far less anguish or none at all. (In 2014 the terrorist death toll was even lower, at 19.) Even the estimate of 44 is generous: it comes from the Global Terrorism Database, which counts hate crimes and most rampage shootings as examples of “terrorism.” The toll is comparable to the number of military fatalities in Afghanistan and Iraq (28 in 2015, 58 in 2014), which, consistent with the age-old devaluing of the lives of soldiers, received a fraction of the news coverage. The next rows down reveal that in 2015 an American was more than 350 times as likely to be killed in a police-blotter homicide as in a terrorist attack, 800 times as likely to be killed in a car crash, and 3,000 times as likely to die in an accident of any kind. (Among the categories of accident that typically kill more than 44 people in a given year are “Lightning,” “Contact with hot tap water,” “Contact with hornets, wasps, and bees,” “Bitten or struck by mammals other than dogs,” “Drowning and submersion while in or falling into bathtub,” and “Ignition or melting of clothing and apparel other than nightwear.”)³

In Western Europe, the relative danger of terrorism was higher than in the United States. In part this is because 2015 was an annus horribilis for terrorism in that region, with attacks in the Brussels Airport, several Paris nightclubs, and a public celebration in Nice. (In 2014, just 5 people were killed.) But the relatively higher terrorism risk is also a sign of how much safer Europe is in every other way. Western Europeans are less murderous than Americans (with about a quarter their homicide rate) and also less car-crazy, so fewer die on the road.⁴ Even with these factors tipping the scale toward terrorism, a Western European in 2015 was more than 20 times as likely to die in one of their (relatively rare) homicides as in a terrorist attack, more than 100 times as likely to die in a car crash, and more than 700 times as likely to be crushed, poisoned, burned, asphyxiated, or otherwise killed in an accident.

The third column shows that for all the recent anguish about terrorism in the West, we have it easy compared with other parts of the world. Though the United States and Western Europe contain about a tenth of the world’s population, in 2015 they suffered one-half of one percent of the terrorist deaths. That’s not because terrorism is a major cause of death elsewhere. It’s because terrorism, as it is now defined, is largely a phenomenon of war, and wars no longer take place in the United States or Western Europe. In the years since the attacks of September 11, 2001, violence that used to be called “insurgency” or “guerrilla warfare” is now often classified as “terrorism.”⁵ (The Global Terrorism Database, incredibly, does not classify any deaths in Vietnam in the last five years of the war there as “terrorism.”)⁶ A majority of the world’s terrorist deaths take place in zones of civil war (including 8,831 in Iraq, 6,208 in Afghanistan, 5,288 in Nigeria, 3,916 in Syria, 1,606 in Pakistan, and 689 in Libya), and many of these are double-counted as war deaths, because “terrorism” during a civil war is simply a war crime—a deliberate attack on civilians—committed by a group other than the government. (Excluding these six civil war zones, the terrorism death count for 2015 was 11,884.) Yet even with the double counting of terrorism and war during the 21st century’s worst year for war deaths, a global citizen was 11 times as likely to have died in a homicide as in a terrorist attack, more than 30 times as likely to have died in a car crash, and more than 125 times as likely to have died in an accident of any kind.

Has terrorism, whatever its toll, increased over time? The historical trends are elusive. Because “terrorism” is an elastic category, the trend lines look different depending on whether a dataset includes civil war crimes, multiple murders (which include robberies or mafia hits in which several victims are shot), or suicidal rampages in which the killer ranted about some political grievance beforehand. (The Global Terrorism Database, for example, includes the 1999 Columbine school massacre but not the 2012 Sandy Hook school massacre.) Also, mass killings are media-driven spectacles, in which coverage inspires copycats, so they can yo-yo up and down as one event inspires another until the novelty wears off for a while.⁷ In the United States, the number of “active shooter incidents” (public rampage killings with guns) has wobbled with an upward trend since 2000, though the number of “mass murders” (four or more deaths in an incident) shows no systematic change (if anything, it shows a slight decline) from 1976 to 2011.⁸ The per capita death rate from “terrorism incidents” is shown in figure 13-1, together with the messy trends for Western Europe and the world.

Figure 13-1: Terrorism deaths, 1970–2015

Sources: “Global Terrorism Database,” National Consortium for the Study of Terrorism and Responses to Terrorism 2016, https://www.start.umd.edu/gtd/. The rate for the world excludes deaths in Afghanistan after 2001, Iraq after 2003, Pakistan after 2004, Nigeria after 2009, Syria after 2011, and Libya after 2014. Population estimates for the world and Western Europe are from the European Union’s 2015 Revision of World Population Prospects (https://esa.un.org/unpd/wpp/); estimates for the United States are from US Census Bureau 2017. The vertical arrow points to 2007, the last year plotted in figs. 6–9, 6–10, and 6–11 in Pinker 2011.

The death rate for American terrorism for the year 2001, which includes the 3,000 deaths from the 9/11 attacks, dominates the graph. Elsewhere we see a bump for the Oklahoma City bombing in 1995 (165 deaths) and barely perceptible wrinkles in other years.⁹ Excluding 9/11 and Oklahoma, about twice as many Americans have been killed since 1990 by right-wing extremists as by Islamist terror groups.¹⁰ The line for Western Europe shows that the rise in 2015 came after a decade of relative quiescence, and is not even the worst that Western Europe has seen: the rate of killing was higher in the 1970s and 1980s, when Marxist and secessionist groups (including the Irish Republican Army and the Basque ETA movement) carried out regular bombings and shootings. The line for the world as a whole (excluding recent deaths in major war zones, which we examined in the chapter on war) contains a spiky plateau for the 1980s and 1990s, a fall after the end of the Cold War, and a recent rise to a level that still falls below that of the earlier decades. So the historical trends, like the current numbers, belie the fear that we are living in newly dangerous times, particularly in the West.

Though terrorism poses a minuscule danger compared with other risks, it creates outsize panic and hysteria because that is what it is designed to do. Modern terrorism is a by-product of the vast reach of the media.¹¹ A group or an individual seeks a slice of the world’s attention by the one guaranteed means of attracting it: killing innocent people, especially in circumstances in which readers of the news can imagine themselves. News media gobble the bait and give the atrocities saturation coverage. The Availability heuristic kicks in and people become stricken with a fear that is unrelated to the level of danger.

It’s not just the salience of a horrific event that stokes the terror. Our emotions are far more engaged when the cause of a tragedy is malevolent intent rather than accidental misfortune.¹² (I confess that as a frequent visitor to London, I was far more upset when I read the headline RUSSELL SQUARE “TERROR” KNIFE ATTACK LEAVES WOMAN DEAD than when I read RENOWNED ART COLLECTOR DIES AFTER BEING HIT BY BUS IN OXFORD STREET TRAGEDY.) Something is uniquely unsettling about the thought of a human being who wants to kill you, and for a good evolutionary reason. Accidental causes of death don’t try to do you in, and they don’t care how you react, whereas human malefactors deploy their intelligence to outsmart you, and vice versa.¹³

Given that terrorists are not mindless hazards but human agents with goals, could it be rational to worry about them despite the small amount of damage they do? After all, we are justly outraged by despots who execute dissidents, even though the number of their victims may be as small as those of terrorism. The difference is that despotic violence has strategic effects that are disproportionate to the body count: it eliminates the most potent threats to the regime, and it deters the rest of the population from replacing them. Terrorist violence, almost by definition, strikes victims at random. The objective significance of the threat, then, beyond the immediate damage, depends on what the scattershot killing is designed to accomplish.

With many terrorists, the goal is little more than publicity itself. The legal scholar Adam Lankford has analyzed the motives of the overlapping categories of suicide terrorists, rampage shooters, and hate crime killers, including both the self-radicalized lone wolves and the bomb fodder recruited by terrorist masterminds.¹⁴ The killers tend to be loners and losers, many with untreated mental illness, who are consumed with resentment and fantasize about revenge and recognition. Some fused their bitterness with Islamist ideology, others with a nebulous cause such as “starting a race war” or “a revolution against the federal government, taxes, and anti-gun laws.” Killing a lot of people offered them the chance to be a somebody, even if only in the anticipation, and going out in a blaze of glory meant that they didn’t have to deal with the irksome aftermath of being a mass murderer. The promise of paradise, and an ideology that rationalizes how the massacre serves a greater good, makes the posthumous fame all the more inviting.

Other terrorists belong to militant groups that seek to call attention to their cause, to extort a government to change its policies, to provoke it into an extreme response that might recruit new sympathizers or create a zone of chaos for them to exploit, or to undermine the government by spreading the impression that it cannot protect its own citizens. Before we conclude that they “pose a threat to the existence or survival of the United States,” we should bear in mind how weak the tactic actually is.¹⁵ The historian Yuval Harari notes that terrorism is the opposite of military action, which tries to damage the enemy’s ability to retaliate and prevail.¹⁶ When Japan attacked Pearl Harbor in 1941, it left the United States without a fleet to send to Southeast Asia in response. It would have been mad for Japan to have opted for terrorism, say, by torpedoing a passenger ship to provoke the United States into responding with an intact navy. From their position of weakness, Harari notes, what terrorists seek to accomplish is not damage but theater. The image that most people retain from 9/11 is not Al Qaeda’s attack on the Pentagon—which actually destroyed part of the enemy’s military headquarters and killed commanders and analysts—but its attack on the totemic World Trade Center, which killed brokers, accountants, and other civilians.

Though terrorists hope for the best, their small-scale violence almost never gets them what they want. Separate surveys by the political scientists Max Abrahms, Audrey Cronin, and Virginia Page Fortna of hundreds of terrorist movements active since the 1960s show that they all were extinguished or faded away without attaining their strategic goals.¹⁷

Indeed, the rise of terrorism in public awareness is not a sign of how dangerous the world has become but the opposite. The political scientist Robert Jervis observes that the placement of terrorism at the top of the list of threats “in part stems from a security environment that is remarkably benign.”¹⁸ It is not only interstate war that has become rare; so has the use of political violence in the domestic arena. Harari points out that in the Middle Ages, every sector of society retained a private militia—aristocrats, guilds, towns, even churches and monasteries—and they secured their interests by force: “If in 1150 a few Muslim extremists had murdered a handful of civilians in Jerusalem, demanding that the Crusaders leave the Holy Land, the reaction would have been ridicule rather than terror. If you wanted to be taken seriously, you should have at least gained control of a fortified castle or two.” As modern states have successfully claimed a monopoly on force, driving down the rate of killing within their borders, they opened a niche for terrorism:

The state has stressed so many times that it will not tolerate political violence within its borders that it has no alternative but to see any act of terrorism as intolerable. The citizens, for their part, have become used to zero political violence, so the theatre of terror incites in them visceral fears of anarchy, making them feel as if the social order is about to collapse. After centuries of bloody struggles, we have crawled out of the black hole of violence, but we feel that the black hole is still there, patiently waiting to swallow us again. A few gruesome atrocities and we imagine that we are falling back in.¹⁹

As states try to carry out the impossible mandate of protecting their citizens from all political violence everywhere and all the time, they are tempted to respond with theater of their own. The most damaging effect of terrorism is countries’ overreaction to it, the case in point being the American-led invasions of Afghanistan and Iraq following 9/11.

Instead, countries could deal with terrorism by deploying their greatest advantage: knowledge and analysis, not least knowledge of the numbers. The uppermost goal should be to make sure the numbers stay small by securing weapons of mass destruction (chapter 19). Ideologies that justify violence against innocents, such as militant religions, nationalism, and Marxism, can be countered with better systems of value and belief (chapter 23). The media can examine their essential role in the show business of terrorism by calibrating their coverage to the objective dangers and giving more thought to the perverse incentives they have set up. (Lankford, together with the sociologist Erik Madfis, has recommended a policy for rampage shootings of “Don’t Name Them, Don’t Show Them, but Report Everything Else,” based on a policy for juvenile shooters already in effect in Canada and on other strategies of calculated media self-restraint.)²⁰ Governments can step up their intelligence and clandestine actions against networks of terrorism and their financial tributaries. And people could be encouraged to keep calm and carry on, as the British wartime poster famously urged during a time of much greater peril.

Over the long run, terrorist movements sputter out as their small-scale violence fails to achieve their strategic goals, even as it causes local misery and fear.²¹ It happened to the anarchist movements at the turn of the 20th century (after many bombings and assassinations), it happened to the Marxist and secessionist groups in the second half of the 20th century, and it will almost certainly happen to ISIS in the 21st. We may never drive the already low numbers of terrorist casualties to zero, but we can remember that terror about terrorism is a sign not of how dangerous our society has become, but of how safe.