Teddy Roosevelt didn’t fear assassins. Though his predecessor in the Presidency, William McKinley, was killed by anarchist Leon Czolgosz’s bullet, Teddy shook off security. If an assassin was cowardly enough to attack from behind, he would simply “go down into the darkness;” if the attack came from the front, President Neo planned to simply dodge the bullet. After all, as Roosevelt biographer Edmund Morris recounts:
He had confidence in the abnormal speed of his reflexes and the power of his 185 pound body. Last winter in Colorado, he had leapt off his horse into a pack of hounds, kicked them aside, and knifed a cougar to death. What a fight that had been!
Roosevelt’s plan worked, after a fashion. When John Schrank shot him during a speech in 1912, Teddy Roosevelt simply stood back up and finished the 90 minute speech. “I don’t know whether you fully understand that I have just been shot,” Teddy said, “but it takes more than that to kill a bull moose.”
Five decades later, JFK’s murder laid Roosevelt’s boasts to rest. Modern weapons technology is not exactly amenable to dodging. Yet a surprising thing happened since Kennedy: not one would-be Presidential assassin has found his mark.
Reagan, the only President to be injured in an assassination attempt since JFK, was lucid enough to narrate his own surgery. “You’re ruining my suit,” the Gipper snapped at the doctors slicing through his formalwear to begin operating.
The assassins’ rut is not unique to the United States. Around the world, the assassination of heads of state is on the wane. In advanced democracies, assassination, common as recently as the 70s, is now unheard-of — though those democracies may be targeting more people for death than ever.
The decline of assassination tells us a lot about our age, standing above all else a testament to democracy’s power to legitimize leaders and the government that anoints them. But while it may seem like assassination has gone out of style, don’t count it out just yet. There’s good reason, courtesy of the same awesome technology that powers FourSquare, to think assassination is merely in remission rather than cured. It may yet metastasize into something new, a style of assassination focused on prominent private citizens rahter than public officials.
The Assassins Have Forgotten Their Creed
You’ve heard of Cain and Abel, but have you met Ötzi? He’s a roughly 5,300 year old caveman, frozen and hence remarkably well-preserved for his age. As a consequence of his relative intactness, scientists have managed to identify Ötzi as one of history’s first recorded murder victims.
Naturally, he’s a media darling. “The First Assassinated Man In History Was Killed By an Arrow 5300 Years Ago,” blared a headline about poor old Ötzi on the tech blog Gizmodo. The phrasing is revealing, in that it points to two important truths about assassination: it’s been around forever, but it’s damn near impossible to define.
The English word itself comes from a breakaway 11th century Muslim cult notorious for their killing prowess, but The Assassins couldn’t exactly patent the idea of killing your political opposition’s leaders. The Romans did it (“et tu, Brute?”), and the Indian strategist Kautilya used it to great effect against two of Alexander the Great’s governors. “Great effect” is, on some accounts, an understatement: the Greek killings allowed Kautilya and his boss Chandragupta’s quest to unify the Indian Subcontinent under one banner for the first time in history.
“There was a high point with the anarchist campaigns in the 1890s,” Byman says, “and it’s all been downhill from there.”
But was Kautilya’s plot actually “assassination” or some other kind of killing? You tell me. Much like its cousin “terrorism,” “assassination” can and has been defined in lots of different ways. Do assassins need some kind of political motive — we have no proof Ötzi’s killers, for instance, did, but Gizmodo was comfortable calling that an assassination. Can governments assassinate people, or is it the province of non-state actors? Is targeting an enemy leader during wartime, a la the killing of Admiral Yamamoto in 1943, assassination or merely solid tactics?
Depending on how you answer these questions, you might end up with a different answer about the state of assassination today. “The question of what you count gets so messy,” Daniel Byman, a Georgetown professor whose work focuses on assassination and terrorism, told me. “Do you count the conflict in the Congo?” What about “attacks on diplomats, if they’re high profile enough, [or] attacks on religious figures?” Or, for that matter, the U.S. targeted killing campaign against suspected terrorists?
These are all important, fraught questions, and certainly worth thinking about. But, for the moment, let’s simplify things, and talk only about the targeted killing of heads of state. Are there any general trends we can point out there?
Yes, as it turns out. “There was a high point with the anarchist campaigns in the 1890s,” Byman says, “and it’s all been downhill from there.” From 1881–1908, anarchists managed to kill a Russian Tsar, both a French and American President, a Spanish Prime Minister, an Austrian Empress, and a Portuguese King and Crown Prince. “The propaganda of the deed,” as it was often called, took Europe and North America by storm. It was assassination’s high water mark in modern history.
Today, the picture is rather different. Benjamins Jones and Olken, economists at Northwestern and MIT respectively, put together a database of 298 assassination attempts directed at national leaders from 1874 till today. They found a higher number of attempts over time, but that’s misleading: there are more countries in the modern world than there were when the anarchists walked the earth, so there are more leaders around to piss off their citizens. Jones and Olken find that, when you control for number of leaders, Byman was right: there’s been a decline in both overall attempts and successful assassinations that goes back “really till the end of World War I,” Jones told me.
Jones and Olken’s work results in something like actuarial assessments for political leaders (if you find that prospect exciting, then congrats — like me, you are a freaky nerd). “At the peak in the 1910s, a given leader had a nearly 1 percent chance of being assassinated in a given year,” Jones and Olken write. “Today, the probability is below 0.3 percent.”
So Europe and North America, once playgrounds for anarchist killers, now almost never see their leaders killed. Assassination, it seems, really has gone out of fashion.
So what happened?
You Come At The King, You Best Not Miss
When Ugandan dictator Idi Amin seized power in a coup d’etat, he developed a nasty way of disposing of soldiers who weren’t quite yet loyal to the new regime. Amin’s forces would round up the prisoners, stick them in a room, and chuck a bushel of live hand grenades in with the ill-fated grunts.
So when some would-be assassins tried to blow him up with one during a parade outside the Nsambya Police Barracks on June 10th, 1976, it seemed like poetic justice. Opinions differ on what happened next: depending on who you believe, the grenade(s) bounced off of Amin’s chest or his “fat stomach;” it either exploded harmlessly, killed bystanders in the crowd, or offed Amin’s driver, prompting the self-styled Last King of Scotland to drive himself to safety.
Amin killed roughly two thousand people in the ensuing crackdown. Some whispered he staged the attempt on his life as an excuse to consolidate power in the face of a secret coup plot.
Whatever the truth, the Nsambya attack itself tells us a lot about assassinating political leaders — how it works, when it doesn’t, and why, outside of war zones, people have stopped trying it.
Start with the simplest point: assassinating a head of state has gotten harder. One does not simply walk up to a presidential parade and chuck grenades into the First Towncar; political security precautions have gotten much tougher than they were in Amin’s Uganda.
In the United States, the Kennedy assassination was the catalyst. Since 1963, the Secret Service’s staff grew by a factor of 10, and its budget by a whopping 273 times. Obama, according to Yahoo’s Chris Moody, rides in a hunkered-down fortress with a mobile blood transfusion lab. It’s a far cry from JFK’s open-top aparade.
Extremist groups, for their part, appear to have recognized this. “There are certain targets that are logical, but hard to hit,” Byman says. “The U.S. Embassy in Saudi Arabia is a prime target, but if you’ve ever been there, it’s a friggin fortress.”
Will McCants, a scholar at the Brookings Institute who spends his days reading al-Qaeda web forums and other jihadi communications, agrees. “When they have an opportunity, they go for it,” but “in many of the countries where they’re operating, the police state is so strong that they have limited ability to get access to those officials.”
But, McCants says, it’s not that they discuss the idea and dismiss it: it’s that al-Qaeda and likeminded groups don’t appear to consider trying to assassinate Western leaders at all. “It is really interesting that a lot of terrorist organizations, at least the jihadi kind that I watch, don’t really make it a high priority in the West.” He pauses. “You don’t see them talk about, you know, ‘let’s rub out President Obama’ or Vice President Biden. I’ve never seen it.”
It’s not just al-Qaeda. While anarchists continue to fixate on the leader-killing glory days, it doesn’t seem like any major extremist faction of any kind nowadays focuses on assassinating democratically-elected leaders. Why?
Well, it might be that asssassination simply isn’t worth the risks — which go well beyond the mere risk of plots being thwarted by the Secret Service.
Jump back to Jones and Olken, the economists who studied assassinations of major leaders. The real goal of their study wasn’t to track the number of assassinations; it was to figure out whether assassination worked.
They came up with a clever way to test this, scientifically speaking. Jones and Olken limited their dataset to cases where the assassin actually managed to fire a gun or detonate a bomb, the theory being that once an assassin gets close enough to use their weapon, the success or failure of any given attempt is more-or-less random. This makes for a natural experiment of sorts, allowing Jones and Olken to compare the effects of “successful” assassinations against failed attempts without worrying about confounding factors like the strength of a nation’s intelligence system.
“You can easily tell a story for decline [in assassinations] largely because of representative government.”
So what did the economists find? First, the vast majority of assassination attempts against major leaders, 75 percent, failed. Second, the effects of both successful and failed attempts can be dramatic — depending on the type of government that’s being targeted.
Aiming at an autocrat is a huge gamble. Jones and Olken find that successfully killing dictators can take down the entire system: killing an autocrat, according to one measure of democracy, tends to make autocracies significantly more likely to translate to democracy. But in the bulk of the cases, the assassin fails, prompting a moderate crackdown on freedoms. Idi Amin’s post-Nsambya killing spree is an extreme example, but it gets the point across.
In democracies, by contrast, assassination attempts, successful or otherwise, simply have no effect on the structure of government. “The successful assassination of democrats produces no change in institutions,” Jones and Olken write. This makes sense, of course: democracies, unlike autocracies, aren’t one-man systems. The entire point of a democracy is that power is concentrated in the people and institutions, not any one person. If you kill a President, there’s always a veep.
In an interview about their research, Jones argued that this democratic resilience tells us something crucially important about the broader decline in assassinations. Democracies don’t just survive because there’s a successor in place; autocracies often have those. Rather, it’s about legitimacy: when people feel like they can change their government through electoral rather than violent means, the death of a leader is less likely to create a vacuum for radical political change. While killing an autocrat may feel like the death of a tyrant to citizens, the murder of a democratically elected leader feels like an attack on the people as much as an attack on the government.
This effect, Jones suggests, also explains why people are less likely to try to kill their leaders nowadays. There are more democracies than there used to be, so there are fewer people dissatisfied with their power to change their governments. “Certainly, the shift towards democracy — while far from complete in the world — has definitely been a main story of the 20th century,” Jones says (he’s right). “You can easily tell a story for decline [in assassinations] largely because of representative government.”
This theory fits with what we know about the psychology of assassins. In 1999, the Secret Service put together a study of everyone who tried to kill an American “prominent public official or figure since 1949.” The 83 subjects, some of whom are referred to only by initials in the study because their identities remain confidential, “rarely had ‘political’ motives.” Only one, in fact — Robert Kennedy’s murderer, Palestinian extremist Sirhan Sirhan — had a classically political motivation, the others being more interested in fame or something more idiosyncratic (a personally developed conspiracy theory, for instance).
So Jones is right, we should expect the pacifying, legitimizing effect of democracy to keep the assassination rate low. But democracy and the 21st century have barely had time to get acquainted. And there may be trouble on the horizon.
“Tiller is the concentration camp ‘Mengele’ of our day and needs to be stopped before he and those who protect him bring judgment upon our nation.”
Scott Roeder wrote those chilling words on an anti-abortion web-forum two years before he assassinated prominent abortion provider George Tiller. Roeder had long been active in right-wing extremist movements, but had grown increasingly extreme over the 2000s.
Roeder is hardly alone in his beliefs. Just this year, another anti-abortion activist, David Leach, interviewed Roeder and posted the video on YouTube. “If someone would shoot the new abortionists, like Scott shot George Tiller,” Leach said, “hardly anyone will appreciate it but the babies.”
If assassination is going to make a comeback in the democratic world, its poster children will look a lot more like Scott Roeder and David Leach than early 20th century anarchists: loosely networked, internet-savvy, and targeted at prominent private citizens rather than heavily guarded political leaders.
Modern anarchists, for their part, get the memo. Andy Greenberg, a Forbes technology reporter, recently profiled Kuwabatake Sanjuro, a “crypto-anarchist” who runs a site called Assassination Market. Assassination Market is basically Kickstarter for murder: People put names on a kill list and others donate money (in Bitcoins, naturally) towards each individual’s demise. Anyone who can successfully prove they were responsible for a hit collects the pot. So far, the most “popular” target — Fed Chair Ben Bernanke — has a $75,000 price on his head.
“At some point, someone is going to be killed based on something like this,” Daveed Gartenstein-Ross told me. “That will absolutely happen.”
Gartenstein-Ross is an expert on violent extremism, fresh off a stint at the International Centre for Counter-Terrorism at The Hague. He’s not particularly worried about Bernanke, given that the full weight of the United States intelligence services has got the Chairman’s back. Rather, Gartenstein-Ross fears for the George Tillers of the world: prominent, intensely controversial private citizens who don’t have the resources to fund Secret Service-like security. “Normal people who have unpopular opinions,” in Gartenstein-Ross’ phrase.
The most obvious way new technology enables this sort of assassination is by making it easier to identify where people are. Suppose, Gartenstein-Ross offers, you want to kill someone famous, and you learn they’re about to give a speech in New York City.
“You take some of the data mining technology that’s already been developed, and you follow Twitter. You can probably follow [her] location throughout Manhattan,” as people will tweet about the person going by. Even something as vague as “’I just saw a convoy of cars coming by,’ it must be important” could be used to pinpoint their location, Gartenstein-Ross suggests.
The target doesn’t have to be famous for Gartenstein-Ross’ point to hold. Apps that locate the user — most obviously something like Foursquare, but some social apps add a location tag to the user’s post automatically — can help a would-be assassin identify a target’s location, either by following the target or her friends and family.
If someone would shoot the new abortionists, like Scott shot George Tiller,” Leach said, “hardly anyone will appreciate it but the babies.”
More simply, someone could find someone’s address and simply post it on a site that murderers might read. “Tiller’s address had been broadcast multiple times before he had been killed online,” Gartenstein-Ross notes.
McCants thinks this scary picture is relatively plausible. “Look at [jihadis] continually trying to kill the Muhammed cartoon guys,” he told me. “That’s high on the wish list.” But, he suggests, anyone who a group would want to target like this would almost certainly know it, and would be able to secure enough security in advance to protect themselves.
Byman, for his part, sees both potential and pitfalls in information technology for terrorist groups. “It enables communication, it enables coordination, it enables learning” between groups of assassins, but “these things are very vulnerable to, as we know from [the] NSA, tremendous surveillance.”
On balance, the proliferation of Web technologies is “probably good for individuals, but bad for groups.” Groups must communicate online to coordinate when using new technology, and online communication is incredibly easy for law enforcement to track. Individuals, by contrast, can make use of data mining and similar technologies with relative ease without any need to send emails the NSA might be reading.
This means, in Byman’s estimation, Gartenstein-Ross’ concerns about the targeting of private citizens are quite well-founded. He recalls a site that put together a list of abortion doctors, together with their home addresses. If anything happened to one of the doctors, the site — designed to appear as if drenched in blood — would cross their name off the list. The information on that site, Byman says, “would not necessarily be easy to access” absent a home on the web.
“There are different skill sets associated with terrorism,” Byman explains. “One is being able to fire a gun, one is being able to live below ground, [and] another is surveillance.” The internet allows people with different skill sets to compliment each other, and not necessarily intentionally: someone may provide surveillance information on an abortion doctor online for the purposes of organizing a protest, but a trigger-puller might well take that information and put it to deadlier use.
So it could be that both Jones’ optimism and Gartenstein-Ross’ pessimism are right. While democracy may be slowly eroding the power of traditional assassinations of political leaders, the internet technologies that flourish alongside democratic freedoms might well enable a new wave of assassinations targeted at citizens rather than states. Though a federal appeals court ultimately ruled that the blood website wasn’t protected speech, the grounds for the court’s ruling would protect other, similar sites.
So fifty years after the JFK assassination, we may be entering a new era of assassination, one that poses unprecedented challenges for law enforcement around the globe just as the old ones are starting to wither away.
Neither the Secret Service nor the FBI responded to requests for comment on this story.