Eduardo Munoz / Reuters

In the United States, the years since 9/11 have been a period of heightened awareness about the threat of terrorism, but they haven’t been a period of frequent attacks. The U.S. witnessed a much higher number of terrorist attacks in the 1970s than in the 2000s. But the destruction of the World Trade Center and a piece of the Pentagon was something unprecedented; no single terrorist attack in history up to that point had killed so many. The pipe bombings of the 1970s—when, as Marquette University’s Risa Brooks has written, “the country experienced a rash of bombings by Puerto Rican nationalist groups and the militant left, such as the Weather Underground, which combined were responsible for more than 100 bombings”—were low-casualty affairs that, if they killed anyone, tended to do so one or two people at a time.


Frequency of Terrorist Attacks in the United States, 1970-2014


At the time, the subject of “terrorism” didn’t attract much scholarly attention. Martha Crenshaw was one of the pioneers of terrorism studies and, incidentally, my advisor in graduate school. In the decades since she started researching it, terrorism has gone from a largely ignored subfield to an object of serious, and some might say excessive, concern among policymakers and the public. There are hundreds of ways to define “terrorism,” and they tend to involve violence committed by a non-state actor to achieve some kind of political objective. (Though, to complicate matters further, there is also the notion of “state terror” or “terror from above,” involving politically motivated violence by governments to oppress or intimidate domestic opponents.)  

In a paper published in 1972, Crenshaw defined “revolutionary terrorism” by examining Algeria’s National Liberation Front (FLN) as being “part of a revolutionary strategy ... to seize political power from an existing government,” involving “a consistent pattern of symbolic representation of the victims or objects of terrorism,” with a deliberate intent to “create a psychological effect on specific groups and thereby challenge their political behavior and attitudes.”

Given Crenshaw’s decades’ worth of perspective on the issue, I asked her recently what has changed about terrorism from the FLN to ISIS, whether the Islamic State is truly unique in the history of terrorist organizations (if indeed it can be considered a terrorist organization), and why there still isn’t any agreed-upon definition of what terrorism is. A transcript of our conversation follows, condensed and edited for clarity. I started by asking what attracted her to the field before there was widespread interest in it.


Martha Crenshaw: When I started, which was in the very early 1970s, nobody, practically, was working on the issue at all. David Rapoport had published a tiny little volume, and he made a distinction between assassination and terrorism. And then, there was a chapter in an edited book that was published in 1964 that did talk about terrorism.

Kathy Gilsinan: Given that nobody seemed to care about it, why did you even decide to…?

Crenshaw: When I was an undergraduate I took courses in Russian history, and I thought that the Russian revolutionary movement was very interesting. Then I was in graduate school, and we were reading a book called Guerrilla Warfare in the 1960s. And at the end it said, “nobody has really studied terrorism, we should study terrorism more.” I spoke French, which was helpful, and I knew something about the Algerian War, so I thought, well, why don’t I write about the FLN and terrorism and the Algerian War. So I did, and I turned that into my Ph.D. dissertation.

Gilsinan: I guess the way to phrase this question as a trick question is: What is terrorism? One of the more maddening things about the field to me is that nobody’s working with a consistent definition.

Crenshaw:  The article that I published [based on the dissertation] was precisely that. It was called “The Concept of Revolutionary Terrorism,” and it was an attempt to define terrorism. I took what there was—there was what I told you, there was a book that was written in 1939 about the failure of the League of Nations to deal with terrorism, there was a book about reigns of terror in the Zulu kingdoms of South Africa, called Terror and Resistance, and I looked at literature on “terror from above”—Nazi Germany, Stalinist Russia—everything I could just to try to get a grip on it.

And at that time, if terrorism was mentioned, it would be mentioned as the first stage in an insurgency or revolutionary warfare. There was a literature on insurgency, and so terrorism would be that first stage, and then as soon as the rebels got through with that first stage, they’d move on to guerrilla warfare, and then to conventional warfare, and then national liberation. But nobody really delved into, what is [terrorism]? And so, unfortunately, my article didn’t persuade anybody, and they’re still arguing about it. But I’m quite happy with what I said in 1972.

Gilsinan: Why do you think people are still arguing about the definition of terrorism?

Crenshaw: For one thing, it’s not an easy concept to define. It’s controversial, it’s contested, and people use it in a normative sense that, your enemies do terrorism and your friends are freedom fighters, et cetera. All these awful clichés. But there is that kind of bias, people are unwilling to just say, look, this is a form of violence, and lots of different types of people use it for lots of different types of purposes. It’s not a very nice form of violence, because it often targets civilians on purpose, and I think we’re always shocked when violence deliberately targets civilians. And then it’s just kind of hard empirically to get at what we mean by terrorism—violence that’s symbolic, violence that’s meant to communicate, to send a message—how, just objectively, do you distinguish that from any other sort of violence?

Gilsinan: It really does get to motivations.

Crenshaw: It’s about motivations, which you can’t know. All right, so you assassinate a president, and if you just killed him because you hated him, and you had a personal vendetta, how would we know the difference—say you were killed in the attempt—versus a person who assassinated the president in order to send a message to all American political figures? So usually terrorist attacks target somebody because they are symbolic of a larger class of people, there’s a meaning behind the choice of target, but how do you read that meaning? And different audiences will read that meaning in different ways. A policeman was killed—what does that tell you? If a group takes credit, and they say, “We killed that policeman because he is the agent of the oppressive state,” well, then you’ve got a clue. But what if they don’t say anything? Well, they could have killed him because he arrested their brother, we don’t know. So it’s just hard to get at it.

And, of course, academics love to quarrel about definitions.

Gilsinan: It seems like September 11 really changed the popular conception of what terrorism is.

Crenshaw: It struck a lot of people as just totally unusual and out of the blue, and it was, in so many ways. But yes, we’d had, maybe not a long history, but a history of terrorism that began in the late 1960s just in the United States. And it’s not a lot of it. Terrorism across the board is actually very rare [when] you compare it to, say, homicides, or many other leading causes of death. But it wasn’t unheard of, and it had happened here.

There was [terrorism] in Algeria in the 1950s, in Europe, we have, obviously, the IRA, we had the Italians [including left-wing terrorist groups like the Red Brigades], and the Germans [who had their own left-wing groups, including the Baader-Meinhof Gang], and ETA [Basque separatists in Spain]. You know, occasionally in past years, terrorism had made that kind of public impact, and particularly in Lebanon in the early 1980s in the bombing of the U.S. Marine barracks [and] at the U.S. embassy. So in the ’80s terrorism did make an impression on us in the United States. I think by the end of the Cold War, the first Gulf War, [the] Oslo Accords [between Israel and the Palestine Liberation Organization, under which the PLO renounced terrorism], I think maybe there was a general feeling that perhaps terrorism was subsiding.

Israel was experiencing terrorism in the ’90s, but in the U.S.—I’m giving you my general impressions rather than anything based on research—I think we saw that as a very particular Israeli problem having to do with the conflict with the Palestinians. Many people thought that revolutionary terrorism, nationalist terrorism, which was the sort of terrorism we knew, was going to fade away, and it looked like Hezbollah and that kind of Shia-oriented terrorism was fading away because of the Gulf War and the end to the Lebanese civil war. So the rise of Sunni jihadist terrorism, now we can kind of look back and say the roots were there, they began at least with the siege of the Great Mosque in Mecca, in 1979, but that’s hindsight.

Now we look back at [9/11] and we see the bombings of the embassies in East Africa in 1998, the bombing of the U.S.S. Cole in 2000, and those events made an impression, but not an enormous impression, not everywhere, so that 9/11 just, we were stunned. And again, the enormity of it, compared to anything that had gone before, was shocking.

Gilsinan: And since, right?

Crenshaw: Nothing anywhere approaching it [has happened since]. It’s very rare to find [terrorist] incidents that cause a lot of deaths, and particularly rare to find those outside of conflict zones. So, right after 9/11, there was the Bali bombing, and then we had [bombings in] London and Madrid, but even so, across the board, attacks that are that lethal are, luckily for us, still extremely rare.

Gilsinan: Is there any reason to expect that to change?

Crenshaw: It’s unpredictable. Our perception of the threat I think has sort of shifted over time, so that now what everybody’s concerned about is what we call the lone-wolf phenomenon, that is, people just inspired through social media. So there’s been a great disaggregation of what we perceive as a threat.

Gilsinan: The lone-wolf phenomenon is interesting to me—on the one hand the rise of lone wolves is sort of evidence of the success of cracking down on terror cells, right? It’s very hard to conspire to commit a terrorist attack in a group. And it’s also very hard to acquire explosives. At the same time, though, lone wolves are harder to prevent because they act on their own, and because it’s pretty easy to acquire firearms.

Crenshaw: I think that that’s a good point, that because we’ve really been very effective at sort of preemptive and preventive measures, it really is hard to get together a large-scale conspiracy on the territory of a country with an effective government—a country where police forces and FBI-type agencies are all working efficiently. It’s not totally impossible, obviously the London bombings, Madrid bombings, are cases in point that there can be a conspiracy that escapes official notice. But particularly since then, authorities are very much on alert for it. So what it’s meant is our adversaries, particularly ISIS and al-Qaeda in the Arabian Peninsula, see their way around it, which is to inspire people pretty much to do whatever they can on their own. So we’re probably going to see more of that.

And there are a number of implications. One is that the amount of destruction they can wreak is smaller than [what] can be wreaked by a dedicated, organized, expert conspiracy. But we’re still very alarmed about this. We still feel threatened, and so [ISIS and AQAP are] happy that they can make us feel threatened with very little expenditure of effort, simply by inspiring amateurish people who can’t do too much damage except psychologically.

But I think another point that’s interesting here is, most of these individuals, they don’t think of themselves as lone wolves. They think of themselves as vital instruments for this cause. And in some cases, in the United States in particular, they are led by informers to believe that they are part of al-Qaeda or ISIS or whatever conspiracy.

Gilsinan: It sounds like actually the rise of the lone wolf is kind of a good problem to have. It just depends on how you define the success of a terrorist group; if their goal is to cause psychological damage at great scale with little cost, then the lone wolves are succeeding. But if it is to kill a lot of people and bring down the American empire, then it seems like they’re doomed to fail.

Crenshaw: Right, and so a lot of it would depend on how the countries that are the victims of these sorts of attacks frame the threat. If they said, “Well, we don’t think this is a good thing at all, but it shows that these organizations are so weak that all they can do is appeal to unhappy teenagers in the West to sort of do whatever they can,” that might be wiser than saying “Oh my God, they could be anywhere, they’re springing up in our midst,” and exaggerating the amount of damage they can cause.

Gilsinan: So, ISIS. We had this discussion, I think I asked you, is ISIS a terrorist group? And you [said], “I don’t like to use ‘terrorist group’” as a phrase. Why not?

Crenshaw: I don’t. The problem is, you’ve got an actor like ISIS, like Hezbollah, like Hamas, like AQAP, indeed like a lot of other violent non-state actors, and they don’t just do terrorism, they do a lot of different things. So by saying that they are “a terrorist organization,” you’re lumping together, first of all, so many disparate organizations, and second, you’re obscuring the fact that they might hold territory, like ISIS, or Hezbollah for that matter, or, in some instances, AQAP. And you’re obscuring the fact that they do other things, like sometimes they provide social services. So, if it’s a group that does nothing but terrorism, by and large, I think that the name fits. An example of that is the Abu Nidal organization in the ’70s and ’80s. It’s a Palestinian group, small, and that was all they did was organize acts of terrorism, they didn’t do anything else. Or the Black September organization would be a similar one. So I think that name is justified. And I’m not saying that the other groups are morally any better, but the name just doesn’t fit that well, to call them “a terrorist organization.” That’s not to say that ISIS doesn’t use terrorism—yes it does, obviously, to the extreme. So that’s just, in order not to confuse what are actors that have a whole lot more power than the name “terrorist organization” implies.

Gilsinan: What ISIS is doing right now seems to harken back to some of the early [academic] work on terrorism as a tactic in guerrilla warfare, [as] a step toward something else—toward assembling a conventional army, and then overthrowing the state.

Crenshaw: And you look at Hezbollah, same thing. And Hezbollah’s a case of a group that did use terrorism in the past but it really doesn’t use what we classically think of as terrorism anymore. It shoots rockets at Israel. It’s a pretty well-armed organization, it’s better-armed than the Lebanese Army, so they’ve transcended that stage. That’s not to say they don’t use terrorism when they want to—the [bus bombing that killed five] Israeli tourists in Bulgaria for example—so when they want to, they can still use terrorism, they don’t drop it. ISIS clearly is using terrorism against Shia, against Sunnis whom they consider to be apostates. AQAP is involved in really what’s an insurgency in Yemen, but they were still trying to blow up targets in the United States.

I said from the beginning that I didn’t think that terrorism was necessarily a first stage of a violent campaign—it can continue through the life of an organization. But what we’re dealing with now is actors that are just bigger, they control territory. ISIS aspires to be a kind of state, not a normal kind of state, but a state-like entity that controls territory, and whether it’s going to be successful at that or not is another question. These types of actors that we’re dealing with that are threatening to us are changing a lot too.

Gilsinan: But are they unprecedented?

Crenshaw: You know if you look back at the FLN in Algeria, and again the circumstances were very different then, but the FLN controlled territory, and it aspired to be an alternative government within the areas that it controlled, it used [terrorism] in order to control the local population. It’s not unheard of. You look at the Viet Cong during the Vietnam War. Maybe they didn’t control territory, but they certainly aspired to be a government. So the idea that a kind of rebel group might start out, say, with isolated acts of terrorism and then move to actually controlling people and controlling territory doesn’t seem to me to be that striking or unusual a phenomenon.

And look at the circumstances in which it’s happening, where there’s just an enormous power vacuum. There’s been nobody to resist them. So if it weren’t for the collapse of Syria, we wouldn’t see ISIS controlling the territory that it does.

Gilsinan: What do you think of attempts to find the “profile” of a terrorist?

Crenshaw: I think it’s always been hopeless; I think it’s hopeless now. You can look out and say, well, some of the people we know about in the West who have been attracted to ISIS, they had personal problems, they were poorly assimilated, they were lonely. But there are a whole lot of other people with the same characteristics who weren’t drawn to ISIS. And furthermore there are a lot of people—there appear to be, we still don’t have really good data—there appear to be people [attracted to ISIS] who, to all intents and purposes, seem to be perfectly adjusted and content, and [have] jobs, and [are] doing well in school. It’s an enormous puzzle as to who would be attracted to the message and who would not be. Clearly it’s not all religion, it’s a whole lot of other stuff going on: adventurism, excitement, youth rebellion against the parental generation.

Gilsinan: Do [ISIS recruits] tend to be younger? Is there any pattern at all?

Crenshaw: I don’t have data on it, but I think I heard from a former British official that [ISIS recruits in Britain] appear to be getting younger. Which, you think, well, you can’t recruit people that are a little bit older and a little bit smarter, so they’re going for 13- and 14-year-olds, which is a very vulnerable age, as we all know. They might join a gang. They might go to Syria.   

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.