Why Defining Terrorism Matters

This is a guest post by Karolina Lula, a PhD student at Rutgers-Newark.

The terrorism industry has grown exponentially since 9/11.  Whenever a terrorist attack occurs, a plethora of terrorism scholars eagerly spoon out their collective wisdom.  The chance to be included in the over-caffeinated media spotlight justifies decades cooped up in small offices pouring over data. In a certain respect, terrorism scholars mirror their subject.  They both love an audience.

Despite their growing presence in the media, academics fail to persuade others about what terrorism is in the first place.  Language evolves and academia is only one source of influence.  The media would do well to adopt their stricter definition.

Academics have their own set of rules for defining terrorism.  Despite intra-field debate, most North American scholars adopt the three-prong definition of terrorism:  it is politically motivated, perpetrated by non-state actors like lone wolves or organizations, and targets civilians rather than the military.  This means that when a government attacks civilians like in Assad’s Syria, when the perpetrators are motivated by pecuniary gain like on the streets of Detroit, or when they target military assets like the USS Cole, academic purists would distinguish such acts of violence from terrorism.

When it comes to defining terrorism, motives therefore matter.  Mass shootings—like the one in Tucson by Jared Loughner, the one in the Aurora movie theater by James Holmes, the Sandy Hook elementary school shooting by Adam Lanza, or the New Orleans Mother’s Day shooting—would be treated as something else.  Some scholars provide no distinction between rampage violence and terrorist acts.  But in reality, there is an important difference—rampage shooters are not politically motivated.

Another important criterion is target selection.  Guerilla attacks on military targets are often distinguished from terrorist attacks, which are directed against civilian targets.  Critics of the Obama administration have hammered him for his hesitancy to label Benghazi as a terrorist attack.  In fact, Benghazi was not a terrorist attack.  It was a guerilla attack against high-level U.S. diplomats, hardly a case of indiscriminate violence.  When most academics think about a terrorist attacks, we recall 9/11 and the Boston marathon because ordinary citizens were targeted, rather than agents of the state.

If Benghazi was not terrorism, was the Woolwich murder?  The video footage of the perpetrator reveals a clear political motive.  Bloody cleaver in hand, he speaks to the camera. “The only reason we have killed this man today is because Muslims are dying daily by British soldiers,” he says.  “Remove your government, they don’t care about you.”    But did he target civilians?  The soldier was struck en route to work wearing a military shirt.  And yet the man in the video shows little interest in harming civilians.  He even expresses a weird concern for the witnesses: “I apologize that women had to witness this today but in our lands, our women have to see the same.”  No, the perpetrator didn’t target aimlessly.

This stringent definition may seem silly to non-academics, but its value lies in predictive power. Those who lump other forms of violence with terrorism are clouding their ability to make accurate predictions.  Consider the divergent political effects of terrorist campaigns versus guerrilla campaigns.  Whereas terrorists have an abysmal track record of getting what they want, guerillas sometimes win against capable opponents—like when the U.S. withdrew from Somalia in 1994 after a black hawk was shot down.

Academic research at its worst is obscure and elitist.  At its best, it impacts policy.  Research that yields accurate predictions creates policy solutions that could prevent violent attacks.  Researchers who lump all violence with terrorism, on the other hand, are like doctors who can’t cure a misdiagnosed affliction.

Politicians are notorious for using language to disguise policy choices. Words, the DNA of language, can be exercised to change the way we feel.  Steven Pinker says a “taboo word” may be used instrumentally to trigger an emotional response.  In The Stuff of Thought: Language as a Window into Human Nature he notes there is “something about the pairing of certain meanings and sounds [that] has a potent effect on people’s emotions” (326).   Although Pinker explains that the “pairing between a sound and meaning is arbitrary,” humans “can use a taboo word to evoke an emotional response in the audience quite against their wishes” (333).

British Prime Minister David Cameron declared in the immediate aftermath of the Woolwich attack, “We will never give in to terror or terrorism in any of its forms.”  President Obama, by contrast, studiously avoided using the t-word on the heels of Benghazi.  Whereas Cameron was lauded, Obama was vilified.  Both men chose their words carefully.  In the face of terrorism, electorates tend to reward right-wing candidates opposed to government concessions.  It’s no wonder that Obama, a Democrat, wanted to eschew the word “terrorism,” while Cameron, the leader of the Conservative party, readily adopted it.

The media should take a higher moral ground than politicians and avoid politicizing the t-word. The age of Twitter dysentery calls for greater conceptual clarity.  The media are the wellsprings of information, but no longer serve as its gatekeepers.  While academics are still trying to figure out what terrorists really want, they at least agree on the meaning of the word.

20 Responses to Why Defining Terrorism Matters

  1. T Kelkar May 28, 2013 at 3:33 pm #

    The soldier in the Woolwich attack was not wearing a military shirt, he was wearing a Help For Heroes shirt, which is a charity. Anyone can wear such a shirt. We are yet to find out if the soldier was targeted because he was a soldier, or some other reason. Even so, he was not partaking in any armed combat at the time, so does that mean he was a soldier?

  2. RobC May 28, 2013 at 4:01 pm #

    Of some historical interest is Nelson Mandela’s taxonomy of violence, given in his statement from the dock at the opening of the defense case in the Rivonia Trial:

    Four forms of violence were possible. There is sabotage, there is guerrilla warfare, there is terrorism, and there is open revolution. We chose to adopt the first method and to exhaust it before taking any other decision.

    Note, however, that although Mandela said that Umkhonto (the violent arm of the African National Congress, co-founded by Mandela) had engaged only in sabotage, he included in that term not only violence against government targets but also “attacks on the economic life lines of the country,” including “destruction of power plants and interference with rail and telephone communications.”

    • John Collignon May 30, 2013 at 4:40 pm #

      A short Mandela result for libtards, The man finally got his day in court and though it may have been the “right thing to do” , I want to ask liberals who are white,if they wouldn’t mind going with me to say, Capetown and checking out Mandela’s legacy?

  3. W.M. May 28, 2013 at 4:03 pm #

    Jumping off of T Kelkar, I’d like to point out something that I see as a flaw in our definition of terrorism: the definition of “target”. In some instances, the target is not necessarily the victims themselves, but those around the victims/the witnesses.

    Whether that applies to this particular case is up for debate, but the narrow confines we’ve set up for ourselves are perhaps slightly too narrow for our own good. A little broadening without sacrificing the usefulness of the conceptual category of terrorism is probably warranted.

  4. Scott Monje May 28, 2013 at 5:54 pm #

    Much of the political hullabaloo over the Benghazi talking points turns on the definition of terrorism, yet no one on either side of the argument has bothered to offer one. The implied distinction focuses on spontaneity versus planning, with Republicans in Congress seemingly arguing that any suggestion that it wasn’t planned long in advance is an attempt to defraud the electorate. Aside from the fact that there is no evidence that it was planned any earlier than that afternoon or that it was particularly well prepared (they were a hodgepodge of armed and unarmed people, they brought no explosives to get through the gate or break into the safe room, they didn’t attempt to take any hostages, they burned the place down using fuel they found lying around the compound), is there any definition of terrorism that focuses on the degree of planning?

    Shameless self-promotion: http://foreignpolicyblogs.com/2013/05/28/surprises-in-the-benghazi-e-mails/

  5. Andrew Gelman May 28, 2013 at 8:20 pm #

    I wonder about the statement, “The media should take a higher moral ground than politicians.” The media serve an important political role, but “taking a higher moral ground” isn’t always what we’d expect from them. One might as well say that politicians should take a higher moral ground than the media.

  6. Jay Livingston May 28, 2013 at 10:35 pm #

    Would “hate crimes” fit the terrorism definition? The intent — sometimes explicitly acknowledged, sometimes not — is to put some group in its place or to enforce that group’s inferior status. Is that political?

  7. Fabius Maximus May 29, 2013 at 12:57 am #

    Great essay! Three questions…

    (1) “Benghazi was not a terrorist attack. It was a guerilla attack against high-level U.S. diplomats”

    Diplomats are civilians. So why was Benghazi not terrorism?

    (2) The post-Westphalia political norm that only States can legitimately use violence assumed that States would not deliberately target civilians. That was discarded in WWII.

    Given the emotionally and politically loaded nature of terrorism as a label, why should terrorism not include government violence against civilians to accomplish political goals?

    (3) Any thoughts about the observation by Glenn Greenwald and others that “terrorism” has become devoid of objective meaning in US political discourse? It’s violence that we don’t like.

    “The real definition of Terrorism. The word that simultaneously means nothing and justifies everything”

    “Was the London killing of a British soldier ‘terrorism’? What definition of the term includes this horrific act of violence but excludes the acts of the US, the UK and its allies?

  8. W.M. May 29, 2013 at 4:26 am #

    Jay Livingston:

    I’d argue not, because the motive is almost always not political. Hate crimes are usually considered crimes of passion that occur in the moment with no forethought, whereas politically motivates crimes require forethought.

    I would concede that hate crimes oftentimes are precipitated in that moment at least in part by the political atmospherics surrounding the perpetrator, but that is distinguishable from premeditated violence for political concerns.

    This is an interesting idea, though, and I would enjoy scholarship further exploring the possibility of a link.

    • W.M. May 29, 2013 at 4:28 am #

      In layman’s terms, nobody walks around thinking “today I’m going to attack a gay person because dong so would scare a bunch of people into submission”.

  9. lmao tse tung May 29, 2013 at 6:26 am #

    What a load of bs. Diplomats ARE civilians, that civilians were discriminately targeted does not make it less of an attack against civilians and less of a terrorist attack, it makes it more of one.

    The off-duty Woolwich soldier was wearing a Help for Heroes fund-raising t-shirt, NOT military uniform (note the evasive description of what he wore at a useless level of generality in order to make her point more convincing, when it is meritless).

    The whole point of wearing uniforms is to facilitate identification and therefore the principle of distinction (of combatants from civilians) under IHL and the law of war. A soldier out of uniform may not take part in hostilities, and as a corollary, may not be attacked. See Pfanner, Military Uniforms and the Law of War, 86 Int’l Rev. Red Cross 94 (2004) (“In international armed conflicts, members of the armed forces can lawfully take part in combat on the battlefield. Inversely, they can be lawfully attacked. The absence of a military uniform usually indicates that a person is a civilian, is therefore not allowed to perform military functions and must not be attacked.”)

    This is in contrast to the combatant who is never in uniform, who is unusual in that he fights unlawfully absent all markers and indicators of combatancy. This is perfidy, and combatants acting perfidiously may legitimately be targeted. Fund-raising while out of uniform is not an instance of perfidy.

    For a column purporting to be precise about language it has only contributed to the murk.

  10. Simon May 29, 2013 at 12:40 pm #

    As a N. American academic (in-training, as is the author of this post) who researches/publishes on terrorism and adopts only the first of these ‘prongs’ in my preferred definition, let me explain why it is better not to require that the ‘targets’ be civilians and the perpetrators be ‘non-state’.

    On civilians:
    First, ‘civilian’ is a contestable category. It is quite common for ideologues and activists involved in justifying and carrying out what we’d probably want to call terrorism to frame the people they attack as combatants. Consider the line that ‘all Israelis are combatants because they are reservists’. While this doesn’t fly under current IHL, and may appear to most of us to be tortured logic, it illustrates how terrorists themselves understand their actions. If we want to look at terrorism as a political action, we need to understand its strategic character, and that requires understanding what terrorists themselves think they are doing.

    Second, if we look at the history of terrorism, we see that the sort of outcomes which terrorists want to bring about can be produced through similar actions regardless of whether the people being attacked are combatants. From vanguardist ‘propaganda of the deed’ attempts to inspire an uprising to assassinations designed to support a Maoist insurgency to urban guerrilla warfare designed to provoke a disproportionate response rather than intimidate a community into capitulation, terrorism requires that a political community react in a certain way. It may be that the desired reaction can only be produced by targeting patent non-combatants, but this is a contextual, rather than essential, property.

    On non-state groups:
    First, quite a lot of terrorism is committed by insurgent groups whose goal, often partially actualised, is to establish a viable ‘counter-state’. Particularly in ‘weak states’, insurgent groups often take on state-like function. These groups have used terrorism as part of their political and military strategy, but the difference between them and the state which they are fighting is often more or less a formality, and they deliver state-like goods to populations under their control with more efficiency than many actual states.

    Second, there are a number of terrorist groups which are closely directed, equipped, and trained by states. It would be hard to organisationaly distinguish Hizbullah from the Iranian security apparatus, as their current involvement in Syria so aptly illustrates. At the same time, states such as Argentina during the ‘Dirty War’ have established informal or covert paramilitary forces to do what terrorists do as a means of counter-insurgency – basically doing the same thing as insurgents do to the state, in many cases.

    An alternative:

    Since I think a survey of the history of terrorism and its strategic role shows rather unequivocally that restricting the definition to ‘targets = civilians’ and ‘terrorists = non-state actors’ leads to analytical incoherence in many cases, I propose something like this:

    ‘Dramatic acts of violence designed to intimidate, provoke, or inspire a political community.’

    I’m aware that this covers much of what criminal justice systems do or have done, but I actually think this gets at something important about the relationship between terrorism and authority. My definition is based largely on Peter Neumann and M.L.R. Smith’s article and book. I think that starting from here, the best thing to do would then be to taxonomise with ideal-types (eg: state terrorism, insurgent terrorism, vanguardist terrorism, etc).

    • Fabius Maximus May 29, 2013 at 1:35 pm #

      “Dramatic acts of violence designed to intimidate, provoke, or inspire a political community.”

      I applaud your attempt to liberate this potentially useful term from the political and analytical quagmire it’s become.

      Your definition would label as terrorism much military practice from early 20th century colonial warfare (take and hold colonies from the natives), WWII, and especially post-WW2. My guess is that opposition to this from military, governments, and their academic supporters will doom such a re-definition.

    • Scott Monje May 29, 2013 at 3:10 pm #

      “It would be hard to organisationaly distinguish Hizbullah from the Iranian security apparatus, as their current involvement in Syria so aptly illustrates.”

      I’m not so sure about that. Syria’s long and deep involvement in Lebanese politics has given just about every faction there a stake in the outcome of the Syrian civil war. Quite apart from instructions from Iran, Hizballah would be left highly exposed to domestic enemies and to the Israelis if Assad lost the war.

  11. Simon May 29, 2013 at 1:42 pm #

    I like to think that scholars will select definitions for their analytical traction undaunted by the political implications of those definitions, for the most part. This may be because I’ve yet to have my naivity dashed out of me.

  12. Lou Klarevas May 29, 2013 at 2:26 pm #

    Perhaps this piece I wrote right after the Benghazi attack will interest some of you. When determining whether an act of violence is an act of terrorism, the National Counter-Terrorism Center (NCTC) employs a five-element definition that was developed in partnership with terrorism scholars and codified into US statutory law. It offers specific guidance on some of the issues raised in the comments above (e.g., government targets and pre-attack planning).

    In the article, I suggested that perhaps the White House was reluctant to label Benghazi as an act of terrorism because there was no concrete evidence of premeditation — which is a statutory prerequisite — available in the first few days following the incident.

    On a personal level, I want to state I that find the NCTC definition valuable. I would suggest this definition over one that excludes attacks on the military in noncombatant posture. The above post states that most scholars in North America embrace a definition of terrorism that excludes attacks on the military (e.g., the USS Cole attack). I suspect though that most of us do consider the Cole attack an act of terrorism. Moreover, if we were to exclude all attacks on the military, then the 9/11 attack which included a strike against the Pentagon would have to be excluded as well (at least in part). And I am pretty confident that most scholars believe the 9/11 attack was an act of terrorism.

    Anyway, here is my piece. I hope it is of value to those interested in this debate.


    • Scott Monje May 29, 2013 at 3:10 pm #


  13. JPL May 31, 2013 at 6:54 am #

    The intentional targeting of civilians for killing or wounding in order to advance an agenda, of a political or cultural nature, through the action’s presumed effect on public opinion.

    This statement of criteria seems to gather together the phenomena that concern us as problematic, and that perhaps loosely we have been calling “terrorism”. Any category will have borderline cases; language evolves through speakers in a language community making decisions on these cases for reasons that make sense to them, although the reasons may not always be made explicit. E.g., for cases such as the bombing of the marine barracks in Lebanon, the USS Cole bombing, the 9-11 attack on the Pentagon, Benghazi attack, etc., careful attention should be given to the reasons for including or excluding the phenomenon in/from the category; this process will result in refinement of the criteria. If the category becomes heterogeneous over time, a new category will be differentiated with a new term. The above def. for ex. leaves open the possibility for what is called “state terrorism”. It would also be necessary to have an agreed upon definition of the term ‘civilian’, one that would be relatively narrow and not include presumptions of guilt. The question of degree of guilt or innocence of ordinary people for crimes committed by others (e.g., their governments) should be handled separately, as an ethical question.

  14. H bylake June 5, 2013 at 10:59 pm #

    I appreciate the perspectives offered here; useful reflections on a complex issue. it is a term I feel very uncomfortable with, because definitions are unclear and contested, as Greenwald suggests. But it is also a term I also feel the need to use when I teach history courses. I thought I would toss in a few thoughts.

    I think one of the issues we need to keep in mind is: for what purpose is a definition intended? From the predictive power perspective Lula proposes, we want a definition that helps us understand which actions are likely to lead to future terror, as well as perhaps other changes in the overall situation in a country affected by such actions.

    Simon’s point sounds useful for predicting what he wants to study. But it seems at odds with another possible purpose for a definition; the role in popular discourse, and politics. I think that in most places, attacks on civilians are perceived as significantly different than attacks on the military. Thus, scholars might end up using the term differently than others do. (This might lead to a further question about whether that popular definition muddles our ability to predict further actions. But I suspect that when an act is classified as ‘terrorist’ or ‘not-terrorist’ by a society, it leads to a different government response, and thus we need to include that in our analysis?) As JPL notes, we also would want to look at how definitions of terrorism have changed over time; as definitions change, does that lead radical groups to change their behavior?

    On the discussion of hate crimes: I think that is a good question. I see the logic in not treating unpremeditated hate crimes as terrorist acts. But Livingston’s point that acts intend “to put some group in its place or to enforce that group’s inferior status. Is that political” connects us to actions that I think should qualify as terrorism – for instance, the KKK’s actions during Reconstruction, or lynchings. Clearly this included some acts planned by organizations; but also included more spur-of-the-moment violence that sent the same message. At what point does an individual act become connected to a larger context that qualifies it as terrorism? For a more recent example, the attack of Wade Michael Page, who had ties to white supremacist groups, on a Sikh temple in Wisconsin seems to qualify as an example of an attack planned to terrorize.

    Rob C’s point reminds me of another grey area in current definitions. Mandela’s group, like some radical environmentalist groups, considers property damage to be in an entirely different category than violence against humans. Governments, however, often disagree, and classify them as ‘ecoterrorists.’

  15. Aneeq July 17, 2013 at 2:34 am #

    Glad to read this post. Generally Terrorism has been applied profusely to THREAT of or USE of violence by a non-state actor (in the post 9/11 scenario, otherwise originally it included the States also which undertook terrorism in pre-westphalian age). Nevertheless May i ask, why INCITEMENT to VIOLENCE is not added as an attribute of terrorism definition?