Why Groups Use Terrorism: A Reassessment of the Conventional Wisdom
Guest post by Max Abrahms
Over the past decade, political scientists have learned a great deal about terrorism. For a while, the conventional wisdom held that groups commit terrorism because it’s strategically effective. For this reason, the dominant paradigm is sometimes referred to as the Strategic Model of Terrorism. Its logic seemed self-evident: To avert additional pain to their civilians, governments were presumed to adopt a more dovish stance by granting the perpetrators their political demands. Prominent scholars from Robert Pape to David Lake to Andrew Kydd and Barbara Walter promoted this viewpoint until it became the conventional wisdom.
There was only one problem with this emerging scholarly orthodoxy. It wasn’t supported by the evidence. Increasingly, empirical evidence has revealed that terrorism is a remarkably ineffective tactic for groups to induce government concessions. In 2006, I published the first study to examine a sample of terrorist groups in terms of their political effectiveness. What I found is that groups are far more likely to attain their demands when their violence is directed not against civilian targets, but military ones. Since then, other researchers with different samples have confirmed that hardly any of the thousands of terrorist groups since the dawn of modern terrorism around 1970 have achieved their political demands by attacking civilians. The historical record is not entirely barren of such cases, but they are the exception that proves the rule.
Subsequent statistical studies have found that terrorism is not simply correlated with political failure; the attacks on civilians actually lower the odds of government concessions. This is because terrorism tends to shift electorates to the political right, strengthening hardliners most opposed to appeasement. But don’t take my word for it; just look at how target countries have responded to Islamic State and associated Islamist attacks.
- Last year, Islamic State said the purpose of beheading the American journalist James Foley was to persuade the United States into calling off military operations in Iraq. But the terrorist act had the opposite political effect. In the immediate aftermath of the beheading, President Obama declared that the U.S. would consequently ramp up its air campaign in Iraq and extend it into Syria for the first time.
- The Paris attacks had a similar effect on France. The French were the opposite of intimidated. Instead, they were defiant. Attendance at the post-attack Paris march was essentially unprecedented. Crowds that size hadn’t been seen since the end of World War Two. Simultaneously, sales of the Charlie Hebdo magazine soared from about 60,000 to millions worldwide. The Islamophobic far-right Front National picked up numerous supporters. Of course, France also dramatically increased its participation in the anti-ISIS military coalition, reflected best in its deployment of the Charles de Gaulle aircraft carrier to the Gulf. And while Islamic State detests the Assad regime, the French public suddenly warmed to him.
- Canada, too, did the political opposite of what the Strategic Model would predict. After a couple terrorist attacks on Canadian soil, the public gave its spy agency unprecedented powers to disrupt terrorism at home, while suddenly favoring an expanded role in the coalition against Islamic State. Indeed, Canada is now arguably even more hawkishly anti-terrorism than its southern neighbor.
- Jordan was a real question mark. The Jordanian public had been highly ambivalent about fighting Islamic State before its citizen was torched to death in a cage. Would Jordan withdraw from the counterterrorism coalition like the anomalous case of Spain after the 2004 Madrid attacks? Just the opposite — in response to the torching, Jordan began bombing the lights out of Islamic State, even ordering additional fighter planes to help get the job done.
- The beheading of 21 Coptics in Libya had the same counterproductive effect on Egypt. Although not formally a member of the anti-ISIS coalition, Cairo quickly volunteered to lead a pan-Arab military force against Islamic State.
- Even Japan became more bellicose after its citizens were slaughtered. Since 1947, Article 9 of the Constitution has banned Japan from possessing war-making capabilities. But thanks to the terrorist attacks, the Japanese rallied around the flag, pushing for the repeal of Article 9 to better respond to threats like Islamic State.
All of this raises what I’ve coined as The Puzzle of Terrorism: If attacking civilians only encourages governments to dig in their political heels, why do groups do it? In a new study in International Organization, Phil Potter and I propose an original theory that can accurately account for variation in militant group violence against civilians. It turns out that certain kinds of groups are significantly more likely to attack civilians than others – those suffering from leadership deficits in which lower level members are calling the shots. Leadership deficits promote terrorism by empowering lower level members of the organization, who have stronger incentives to harm civilians.
For many reasons, there’s an inverse relationship between the position of members within the organizational hierarchy and their incentives for harming civilians. For starters, lower level members may try to rise up within the group by committing atrocities against civilians. Such organizational ladder-climbing is well documented in gangs, but is also quite common in militant groups – just ask Jihadi John. Furthermore, lower level members have less access to organizational resources than the leadership, incentivizing them to strike softer targets. And leaders tend to have more experience in asymmetric conflict, so they are more apt than their subordinates to understand the political risks of indiscriminate violence in the first place.
In accordance with this new theory for terrorism, our study reveals that decapitation strikes with drones make militant groups more likely to attack civilians by weakening the leadership. Decentralized groups are also prone to civilian targeting because the leadership must delegate tactical decision-making to lower level members. Similarly, we demonstrate that as operatives travel further away from the leadership, they gain a measure of autonomy and are thus more inclined to attack the population. Unlike the Strategic Model, our organizational theory does not rest on the dubious assumption that terrorism helps induce government concessions. But more importantly, it can help to predict which groups will attack civilians, when, and why.
Max Abrahms is an Assistant Professor of Political Science at Northeastern University.