Terrorism Is Not About Terror

Terrorists act irrationally from a rational activism perspective, and groups act in ways most consistent with terrorism being about social status and belonging
sociology, politics
2009-04-092017-04-21 finished certainty: likely importance: 8


Sta­tis­ti­cal analy­sis of ter­ror­ist groups’ longevi­ty, aims, meth­ods and suc­cesses reveal that groups are self­-con­tra­dic­tory and self­-s­ab­o­tag­ing, gen­er­ally ineffec­tive; com­mon stereo­types like ter­ror­ists being poor or ultra­-skilled are false. Super­fi­cially appeal­ing coun­ter-ex­am­ples are dis­cussed and reject­ed. Data on moti­va­tions and the dis­so­lu­tion of ter­ror­ist groups are brought into play and the sur­pris­ing con­clu­sion reached: ter­ror­ism is a form of social­iza­tion or sta­tus-seek­ing.

There is a com­mon­ly-be­lieved “strate­gic model” of ter­ror­ism which we could describe as fol­lows: ‘ter­ror­ists are peo­ple who are ide­o­log­i­cally moti­vated to pur­sue spe­cific unvary­ing polit­i­cal goals; to do so, they join together in long-last­ing orga­ni­za­tions and after the fail­ure of ordi­nary polit­i­cal tac­tics, ratio­nally decide to effi­ciently & com­pe­tently engage in vio­lent attacks on (usu­al­ly) civil­ian tar­gets to get as much atten­tion as pos­si­ble and pub­lic­ity for their move­ment, and inspire fear & ter­ror in the civil­ian pop­u­la­tion, which will pres­sure its lead­ers to solve the prob­lem one way or anoth­er, pro­vid­ing sup­port for the ter­ror­ists’ favored laws and/or their nego­ti­a­tions with involved gov­ern­ments, which then often suc­ceed in gain­ing many of the orig­i­nal goals, and the orga­ni­za­tion dis­solves.’

Unfor­tu­nate­ly, this mod­el, is in almost every respect, empir­i­cally false. Let’s look in some more detail at find­ings which cast doubt on the strate­gic mod­el.

The problem

From “What Ter­ror­ists Really Want: Ter­ror­ist Motives and Coun­tert­er­ror­ism Strat­egy”, Max Abrahms 2008:

Does the ter­ror­ist’s deci­sion-mak­ing process con­form to the strate­gic mod­el? The answer appears to be no. The record of ter­ror­ist behav­ior does not adhere to the mod­el’s three core assump­tions. Seven com­mon ten­den­cies of ter­ror­ist orga­ni­za­tions flatly con­tra­dict them. Togeth­er, these seven ter­ror­ist ten­den­cies rep­re­sent impor­tant empir­i­cal puz­zles for the strate­gic mod­el, pos­ing a for­mi­da­ble chal­lenge to the con­ven­tional wis­dom that ter­ror­ists are ratio­nal actors moti­vated fore­most by polit­i­cal end­s…The seven puz­zles…are:

  1. ter­ror­ist orga­ni­za­tions do not achieve their stated polit­i­cal goals by attack­ing civil­ians;
  2. ter­ror­ist orga­ni­za­tions never use ter­ror­ism as a last resort and sel­dom seize oppor­tu­ni­ties to become pro­duc­tive non­vi­o­lent polit­i­cal par­ties;
  3. ter­ror­ist orga­ni­za­tions reflex­ively reject com­pro­mise pro­pos­als offer­ing sig­nifi­cant pol­icy con­ces­sions by the tar­get gov­ern­ment1;
  4. ter­ror­ist orga­ni­za­tions have pro­tean polit­i­cal plat­forms;
  5. ter­ror­ist orga­ni­za­tions gen­er­ally carry out anony­mous attacks, pre­clud­ing tar­get coun­tries from mak­ing pol­icy con­ces­sions;
  6. ter­ror­ist orga­ni­za­tions with iden­ti­cal polit­i­cal plat­forms rou­tinely attack each other more than their mutu­ally pro­fessed ene­my; and
  7. ter­ror­ist orga­ni­za­tions resist dis­band­ing when they con­sis­tently fail to achieve their polit­i­cal plat­forms or when their stated polit­i­cal griev­ances have been resolved.

Ter­ror­ism has­n’t impressed many observers both on case-s­tud­ies & in gen­eral2. On sta­tis­ti­cal grounds, it’s incon­tro­vert­ible that ter­ror­ism is a shock­ingly ineffec­tive strat­e­gy; from Abrahms 2012:

Jones and Libicki (2008) then exam­ined a larger sam­ple, the uni­verse of known ter­ror­ist groups between 1968 and 2006. Of the 648 groups iden­ti­fied in the RAND-MIPT Ter­ror­ism Inci­dent data­base, only 4% obtained their strate­gic demands. More recent­ly, Cronin (2009) has reex­am­ined the suc­cess rate of these groups, con­firm­ing that less than 5% pre­vailed…Chenoweth and Stephan (2008, 2011) pro­vide addi­tional empir­i­cal evi­dence that met­ing out pain hurts non-s­tate actors at the bar­gain­ing table. Their stud­ies com­pare the coer­cive effec­tive­ness of 323 vio­lent and non­vi­o­lent resis­tance cam­paigns from 1900 to 2006. Like Gaibul­loev and San­dler (2009), the authors find that refrain­ing from blood­shed sig­nifi­cantly raises the odds of gov­ern­ment com­pli­ance even after tac­ti­cal con­founds are held fixed. These sta­tis­ti­cal find­ings are rein­forced with struc­tured in-case com­par­isons high­light­ing that esca­lat­ing from non­vi­o­lent meth­ods of protest such as peti­tions, sit-ins, and strikes to deadly attacks tends to dis­suade gov­ern­ment com­pro­mise. Chenoweth and Stephan employ an aggre­gate mea­sure of vio­lence that incor­po­rates both indis­crim­i­nate attacks on civil­ians and dis­crim­i­nate attacks on mil­i­tary per­son­nel or other gov­ern­ment offi­cials, which are often differ­en­ti­ated from ter­ror­ism as guer­rilla attacks (Abrahms 2006; Cronin 2009; and Moghadam 2006). Other sta­tis­ti­cal research (Abrahms, 2012, Fort­na, 2011) demon­strates that when ter­ror­ist attacks are com­bined with such dis­crim­i­nate vio­lence, the bar­gain­ing out­come is not addi­tive; on the con­trary, the pain to the pop­u­la­tion sig­nifi­cantly decreases the odds of gov­ern­ment con­ces­sions.3

effec­tive­ness is its own top­ic; we can note that many of the same cog­ni­tive biases like the that skew our beliefs on ter­ror­ism also apply to guer­rilla war­fare as well—ev­ery­one remem­bers the suc­cess­ful Amer­i­can Rev­o­lu­tion, but who ever invokes the scores or hun­dreds of other revolts & failed rev­o­lu­tions in the British Empire which involved guer­rilla tac­tics? (Or , for that mat­ter—eg. , the , or ? How well did the Amer­i­can South suc­ceed in seced­ing, in a con­flict with quite as many irreg­u­lar forces as the Amer­i­can Rev­o­lu­tion?) Does a close exam­i­na­tion of the , where the much-her­alded were destroyed after the and before the crushed the ARVN and con­quered South Viet­nam, reveal it to have been more effec­tive than con­ven­tional war­fare? A cur­sory look through any list of guer­rilla move­ments does not reveal it to be a list of lumi­nar­ies. “Nobody likes a loser”, least of all in war. But to return to ter­ror­ism.

Worse, ter­ror­is­m—of any kind like hostage-tak­ing4, and includ­ing con­ven­tional war­fare tac­tics like civil­ian atroc­i­ties or strate­gic bomb­ing—re­li­ably pro­duces a polit­i­cal back­lash towards con­ser­vatism and bol­sters hard­lin­ers’ approaches to ter­ror­ism56, pos­si­bly due to a / where the usage of vio­lence is inferred to indi­cate a group is intrin­si­cally vicious/intransigent/hateful7, so there’s a dou­ble-wham­my—the ter­ror­ism makes any kind of com­pro­mise harder to reach, and if there is dan­ger of an agree­ment, the extrem­ists will try to sab­o­tage it, which intran­si­gence nat­u­rally makes any future agree­ments less like­ly.

To this we could add that there are many fewer ter­ror­ists than one might expect, even for the most appar­ently suc­cess­ful and glob­ally pop­u­lar groups like Al Qaeda8.

Terrorist ineffectiveness

In a [pre­vi­ous study of mine]9 assess­ing ter­ror­is­m’s coer­cive effec­tive­ness, I found that in a sam­ple of 28 well-known ter­ror­ist cam­paigns, the ter­ror­ist orga­ni­za­tions accom­plished their stated pol­icy goals 0% of the time by attack­ing civil­ians.

The al-Qaida mil­i­tary strate­gist, , com­plained that with its “hasty chang­ing of strate­gic tar­gets”, al-Qaida was engaged in noth­ing more than “ran­dom chaos”. Other dis­grun­tled al-Qaida mem­bers have reproached the orga­ni­za­tion for espous­ing polit­i­cal objec­tives that “shift with the wind”.

Who is effec­tive? How could ter­ror­ists be more effec­tive? Eas­ily. (See my essay.) The strange thing is that we know, and they know, per­fectly well that there are attacks which do the US tremen­dous dam­age, yet they hardly ever use them. Why are there so few , so few 9/11s, so few s? Their eco­nomic mul­ti­plier is tremen­dous:

In his Octo­ber 2004 address to the Amer­i­can peo­ple, bin Laden noted that the 9/11 attacks cost al Qaeda only a frac­tion of the dam­age inflicted upon the United States. “Al Qaeda spent $795,258 on the event,” he said, “while Amer­ica in the inci­dent and its after­math lost—ac­cord­ing to the low­est esti­mates—­more than $795 bil­lion, mean­ing that every dol­lar of al Qaeda defeated a mil­lion dol­lars.”10

The ?

“Two Nokia mobiles, $197 each, two HP print­ers, $394 each, plus ship­ping, trans­porta­tion and other mis­cel­la­neous expenses add up to a total bill of $4,200. That is all what Oper­a­tion Hem­or­rhage cost us,” the [AQ] mag­a­zine [Inspire] said.11

Iron­i­cal­ly, it was cheaper for Pales­tini­ans to launch sui­cide attacks:

Has­san cites one Pales­tin­ian offi­cials pre­scrip­tion for a suc­cess­ful mis­sion: “a will­ing young man. . . nails, gun­pow­der, a light switch and a short cable, mer­cury (read­ily obtain­able from ther­mome­ter­s), ace­tone. . . . The most expen­sive item is trans­porta­tion to an Israeli town” (30). The total cost is about $239.12

Other air­line plots?

It is rec­og­nized that the cost of the actual equip­ment used in an attack can be quite low. For exam­ple, the ingre­di­ents used to build each bomb intended to blow up air­lin­ers bound for the United States from the United King­dom in 2006 are esti­mated to have cost only $20.13 The cost of an has been esti­mated to be $34 to $41.14

Sim­i­lar­ly, the mate­r­ial cost for con­duct­ing a sui­cide bomb has been esti­mated at only $194.15…the FATF esti­mated that the bomb­ings of two U.S. embassies in East Africa had direct costs of $92,676.16 Other esti­mates, even for car-bomb sui­cide ter­ror­ists, are in sim­i­lar ranges, although prices vary greatly over time and so all of the above is out of date.17

Con­sid­er­ing Euro­pean ter­ror­ism inci­dents as a whole, they are all uni­formly cheap, with the cheap­est being knife/axe attacks (~$0); 3⁄4s cost <$10,000, with only 3 exceed­ing $20,000.

Fund­ing seems to be a con­stant issue for spree killers or ter­ror­ists, even when objec­tively there is no rea­son to think about it:

So ter­ror­ists want to hurt the US, they know many effec­tive ways to do so, and… hardly any­thing hap­pens. The work of ratio­nal actors?

The solution

“When peo­ple see a strong horse and a weak horse, by nature, they will like the strong horse.”

Osama bin Laden, 2001

So, then, what is the expla­na­tion for such self­-de­feat­ing, irra­tional actions? Can we explain the self­-de­feat­ing as delib­er­ate, due per­haps to ? No; even if false flag attacks were more com­mon than every­one believes and made up—a uni­ver­sal cen­tu­ry-long in every coun­try (de­spite the absence of evi­dence)—say 20% of the scores of thou­sands of ter­ror­ist attacks in the 20th & 21st cen­turies, that still leaves count­less orga­ni­za­tions & ter­ror­ists inex­plic­a­bly incom­pe­tent19 & igno­rant20. In the spirit of X Is Not About X posts (see “Pol­i­tics isn’t about Pol­icy”), I’d like to offer one of my own: is not about ter­ror; it’s not even about pol­i­tics. It’s about social­iz­ing.

There is com­par­a­tively strong the­o­ret­i­cal and empir­i­cal evi­dence that peo­ple become ter­ror­ists not to achieve their orga­ni­za­tion’s declared polit­i­cal agen­da, but to develop strong affec­tive ties with other ter­ror­ist mem­bers. In other words, the pre­pon­der­ance of evi­dence is that peo­ple par­tic­i­pate in ter­ror­ist orga­ni­za­tions for the social sol­i­dar­i­ty, not for their polit­i­cal return.

In Under­stand­ing ter­ror net­works (sum­mary), he writes:

Ibrahim com­mented on the supe­rior attrac­tive­ness of a reli­gious revival­ist orga­ni­za­tion over a sec­u­lar polit­i­cal one, namely the strong sense of com­mu­nion that Mus­lim groups pro­vided for their mem­bers…‘The mil­i­tant Islamic groups with their empha­sis on broth­er­hood, mutual shar­ing, and spir­i­tual sup­port become the func­tional equiv­a­lent of the extended fam­ily to the young­ster who has left his behind. In other words, the Islamic group ful­fills a de-alien­at­ing func­tion for its mem­bers in ways that are not matched by other rival polit­i­cal move­ments’ (Ibrahim, 198: 448)." "The Saidi branch was com­posed of sev­eral groups, based in provin­cial uni­ver­sity towns. They recruited heav­ily accord­ing to kin­ship and tribal bonds.

…Friend­ships cul­ti­vated in the jihad, just as those forged in com­bat in gen­er­al, seem more intense and are endowed with spe­cial sig­nifi­cance. Their actions taken on behalf of God and the umma are expe­ri­enced as sacred. This added ele­ment increases the value of friend­ships within the clique and the jihad in gen­eral and dimin­ishes the value of out­side friend­ships. To friends hov­er­ing on the brink of join­ing an increas­ingly activist clique, this promised shift in value may be diffi­cult to resist, espe­cially if one is tem­porar­ily alien­ated from soci­ety…once they become mem­bers, strong bonds of loy­alty and emo­tional inti­macy dis­cour­age their depar­ture.

From Scott Atran’s 2003 review (ibid):

Stud­ies by psy­chol­o­gist Ariel Mer­ari point to the impor­tance of insti­tu­tions in sui­cide ter­ror­ism (28). His team inter­viewed 32 of 34 bomber fam­i­lies in Palestine/Israel (be­fore 1998), sur­viv­ing attack­ers, and cap­tured recruiters. Sui­cide ter­ror­ists appar­ently span their pop­u­la­tion’s nor­mal dis­tri­b­u­tion in terms of edu­ca­tion, socioe­co­nomic sta­tus, and per­son­al­ity type (in­tro­vert vs. extro­vert). Mean age for bombers was early twen­ties. Almost all were unmar­ried and expressed reli­gious belief before recruit­ment (but no more than did the gen­eral pop­u­la­tion). Except for being young, unat­tached males, sui­cide bombers differ from mem­bers of vio­lent racist orga­ni­za­tions with whom they are often com­pared (29: R. Ezekiel, The Racist Mind). Over­all, sui­cide ter­ror­ists exhibit no socially dys­func­tional attrib­utes (fa­ther­less, friend­less, or job­less) or sui­ci­dal symp­toms. They do not vent fear of ene­mies or express “hope­less­ness” or a sense of “noth­ing to lose” for lack of life alter­na­tives that would be con­sis­tent with eco­nomic ratio­nal­i­ty. Mer­ari attrib­utes pri­mary respon­si­bil­ity for attacks to recruit­ing orga­ni­za­tions, which enlist prospec­tive can­di­dates from this youth­ful and rel­a­tively unat­tached pop­u­la­tion. Charis­matic train­ers then intensely cul­ti­vate mutual com­mit­ment to die within small cells of three to six mem­bers. The final step before a mar­tyr­dom oper­a­tion is a for­mal social con­tract, usu­ally in the form of a video tes­ta­ment.

Psy­chol­o­gist Brian Bar­ber sur­veyed 900 Moslem ado­les­cents dur­ing Gaza’s first Intifada (1987–1993) (31: B. Bar­ber, Heart and Stones). Results show high lev­els of par­tic­i­pa­tion in and vic­tim­iza­tion from vio­lence. For males, 81% reported throw­ing stones, 66% suffered phys­i­cal assault, and 63% were shot at (ver­sus 51, 38, and 20% for females). Involve­ment in vio­lence was not strongly cor­re­lated with depres­sion or anti­so­cial behav­ior. Ado­les­cents most involved dis­played strong indi­vid­ual pride and social cohe­sion. This was reflected in activ­i­ties: for males, 87% deliv­ered sup­plies to activists, 83% vis­ited mar­tyred fam­i­lies, and 71% tended the wounded (57, 46, and 37% for females). A fol­low-up dur­ing the sec­ond Intifada (2000–2002) indi­cates that those still unmar­ried act in ways con­sid­ered per­son­ally more dan­ger­ous but socially more mean­ing­ful. Increas­ing­ly, many view mar­tyr acts as most mean­ing­ful. By sum­mer 2002, 70 to 80% of Pales­tini­ans endorsed mar­tyr oper­a­tions (32)…In con­trast to Pales­tini­ans, sur­veys with a con­trol group of Bosn­ian Moslem ado­les­cents from the same time period reveal markedly weaker expres­sions of self­-es­teem, hope for the future, and proso­cial behav­ior (30). A key differ­ence is that Pales­tini­ans rou­tinely invoke reli­gion to invest per­sonal trauma with proac­tive social mean­ing that takes injury as a badge of hon­or. Bosn­ian Moslems typ­i­cally report not con­sid­er­ing reli­gious affil­i­a­tion a sig­nifi­cant part of per­sonal or col­lec­tive iden­tity until seem­ingly arbi­trary vio­lence forced aware­ness upon them.

Con­sider data on 39 recruits to Harkat al-Ansar, a Pak­istani-based ally of Al-Qai­da. All were unmar­ried males, most had stud­ied the Quran. All believed that by sac­ri­fic­ing them­selves they would help secure the future of their “fam­ily” of fic­tive kin: “Each [mar­tyr] has a spe­cial place-a­mong them are broth­ers, just as there are sons and those even more dear” (34: D. Rhode, A. Chivers, New York Times, 2002-03-17, p. A1).

From the RAND study “Derad­i­cal­iz­ing Islamic Extrem­ists”, Rabas et al 2010 (em­pha­sis added):

In a study of Colom­bian insur­gent move­ments, Flo­rez-Mor­ris found that mem­bers who remained in the group until it col­lec­tively demo­bi­lized did so as a result of social and prac­ti­cal needs, shared beliefs, and the group’s role in boost­ing their self­-i­den­tity by mak­ing them feel impor­tant. In addi­tion to these ben­e­fits, insur­gents were also deterred from leav­ing by the lack of other options, a result of the clan­des­tine nature of the orga­ni­za­tion (Mauri­cio Flo­rez-Mor­ris, “Why Some Colom­bian Guer­rilla Mem­bers Stayed in the Move­ment Until Demo­bi­liza­tion: A Micro-So­ci­o­log­i­cal Case Study of Fac­tors That Influ­enced Mem­bers’ Com­mit­ment to Three For­mer Rebel Orga­ni­za­tions: M-19, EPL, and CRS”, Ter­ror­ism and Polit­i­cal Vio­lence, Vol. 22, No. 2010-03-02, p. 218.)21

That study men­tions some inter­est­ing dat­a­points from the Saudi reha­bil­i­ta­tion pro­grams:

The sec­ond study—which focused on indi­vid­u­als who had allegedly par­tic­i­pated in vio­lence in Saudi Ara­bi­a—re­vealed an equally inter­est­ing set of fac­tors. Most sig­nifi­cant­ly, the data show greater domes­tic prob­lems and trou­bled home­lives for this group. Approx­i­mately half came from homes with a father over the age of 50, and one-quar­ter (26%) came from polyg­a­mous house­holds. Saudi author­i­ties stress that they believe there is a cor­re­la­tion between less atten­tion received at home and trou­ble later in life. Sim­i­lar­ly, over a third (35%) of the sec­ond study’s sub­jects came from homes with “fam­ily prob­lems”, and one-fifth were iden­ti­fied as orphans with no tra­di­tional parental over­sight.

Another RAND study () exam­ines detailed finan­cial records of , find­ing that per­son­nel rep­re­sent a major cost of branch­es, which were highly profitable as they engaged in theft & extor­tion, but not enough to com­pen­sate for the risk—even tak­ing into account AQI’s pol­icy of pay­ing salaries to the fam­i­lies of dead or impris­oned mem­bers, mem­bers were for­feit­ing at least half their life­time income. But the RAND researchers also dis­cuss how US Army enlisted per­son­nel—pre­sum­ably bet­ter edu­cated and trained than AQI mem­ber­s—have dis­count rates as flab­ber­gast­ingly high as 22, and that their data did not allow them to esti­mate the edu­ca­tion or skills of the AQI mem­bers or how much the mem­bers might be skim­ming off the mul­ti­far­i­ous crim­i­nal activ­i­ties. Given that the cen­tral Anbar AQI group had to trans­fer $3,545 on aver­age for one of the local groups to launch one attack and the raw mate­ri­als, as quoted pre­vi­ous­ly, are so cheap, one won­ders at the effi­ciency of AQI in turn­ing dol­lars into attacks; how much of the over­head is truly nec­es­sary with mem­bers ded­i­cated to the cause?

Increased spend­ing from the AQI Anbar admin­is­tra­tion to its sec­tors increases the num­ber of attacks in those sec­tors, with one addi­tional attack occur­ring for every addi­tional $3,545 trans­ferred…Putting together an IED or buy­ing a mor­tar for an attack is cheap. How­ev­er, our find­ings add to the mount­ing evi­dence that mil­i­tant group oper­a­tions involve far more than just one-time costs. Main­tain­ing a mil­i­tant orga­ni­za­tion can be quite expen­sive. For AQI, per­son­nel costs for mem­bers con­sti­tuted the bulk of these expens­es. With­out such recur­ring pay­ments, it is unlikely that AQI could main­tain its effec­tive­ness in com­mit­ting vio­lence. The group incurred large costs keep­ing impris­oned mem­bers on the pay­roll as an oblig­a­tion to their fam­i­lies and pay­ing the fam­i­lies of dead mem­bers. Although such pay­ments likely increased the loy­alty of mem­bers, they also diverted large amounts of money that could have oth­er­wise been used to attack Coali­tion and Iraqi forces.

“Psy­chol­ogy of Ter­ror­ism”, Borum 2004:

A sim­i­lar mech­a­nism is one in which a des­per­ate quest for per­sonal mean­ing pushes an indi­vid­ual to adopt a role to advance a cause, with lit­tle or no thought­ful analy­sis or con­sid­er­a­tion of its mer­it. In essence, the indi­vid­ual resolves the diffi­cult ques­tion “Who am I?” by sim­ply defin­ing him or her­self as a “ter­ror­ist,” a “free­dom fight­er,” “shahid” or sim­i­lar role (Della Por­ta, 1992 ; Knut­son, 1981). Tay­lor and Louis (200453) describe a clas­sic set of cir­cum­stances for recruit­ment into a ter­ror­ist orga­ni­za­tion: “These young peo­ple find them­selves at a time in their life when they are look­ing to the future with the hope of engag­ing in mean­ing­ful behav­ior that will be sat­is­fy­ing and get them ahead. Their objec­tive cir­cum­stances includ­ing oppor­tu­ni­ties for advance­ment are vir­tu­ally nonex­is­tent; they find some direc­tion for their reli­gious col­lec­tive iden­tity but the des­per­ately dis­ad­van­taged state of their com­mu­nity leaves them feel­ing mar­gin­al­ized and lost with­out a clearly defined col­lec­tive iden­tity” (p. 178).

Belong­ing: In rad­i­cal extrem­ist groups, many prospec­tive ter­ror­ists find not only a sense of mean­ing, but also a sense of belong­ing, con­nect­ed­ness and affil­i­a­tion. Luck­abaugh and col­leagues (1997) argue that among poten­tial ter­ror­ists “the real cause or psy­cho­log­i­cal moti­va­tion for join­ing is the great need for belong­ing.” For these alien­ated indi­vid­u­als from the mar­gins of soci­ety, “join­ing a ter­ror­ist group rep­re­sented the first real sense of belong­ing after a life­time of rejec­tion, and the ter­ror­ist group was to become the fam­ily they never had” (Post, 1984). This strong sense of belong­ing has crit­i­cal impor­tance as a moti­vat­ing fac­tor for join­ing, a com­pelling rea­son for stay­ing, and a force­ful influ­ence for act­ing. “Volkan (1997) .. argued that ter­ror­ist groups may pro­vide a secu­rity of fam­ily by sub­ju­gat­ing indi­vid­u­al­ity to the group iden­ti­ty. A pro­tec­tive cocoon is cre­ated that offers shel­ter from a hos­tile world” (Marsel­la, 2003). Obser­va­tions on ter­ror­ist recruit­ment show that many peo­ple are influ­enced to join by seek­ing sol­i­dar­ity with fam­i­ly, friends or acquain­tances (Della Por­ta, 1995), and that “for the indi­vid­u­als who become active ter­ror­ists, the ini­tial attrac­tion is often to the group, or com­mu­nity of believ­ers, rather than to an abstract ide­ol­ogy or to vio­lence” (Cren­shaw, 1988). Indeed, it is the image of such strong cohe­sive­ness and sol­i­dar­ity among extrem­ist groups that makes them more attrac­tive than some proso­cial col­lec­tives as a way to find belong­ing (John­son & Feld­man, 1982).

Con­clu­sion: These three fac­tors—in­jus­tice, iden­ti­ty, and belong­ing—have been found often to co-oc­cur in ter­ror­ists and to strongly influ­ence deci­sions to enter ter­ror­ist orga­ni­za­tions and to engage in ter­ror­ist activ­i­ty. Some ana­lysts even have sug­gested that the syn­er­gis­tic effect of these dynam­ics forms the real “root cause” of ter­ror­ism, regard­less of ide­ol­o­gy. Luck­abaugh and col­leagues (1997), for exam­ple, con­cluded “the real cause or psy­cho­log­i­cal moti­va­tion for join­ing is the great need for belong­ing, a need to con­sol­i­date one’s iden­ti­ty. A need to belong, along with an incom­plete per­sonal iden­ti­ty, is a com­mon fac­tor that cuts across the groups.” Jer­rold Post (1984) has sim­i­larly the­o­rized that “the need to belong, the need to have a sta­ble iden­ti­ty, to resolve a split and be at one with one­self and with soci­ety—… is an impor­tant bridg­ing con­cept which helps explain the sim­i­lar­ity in behav­ior of ter­ror­ists in groups of widely differ­ent espoused moti­va­tions and com­po­si­tion.”

…Della Porta (1992), for exam­ple, notes that among Ital­ian extrem­ists, “the deci­sion to join an under­ground orga­ni­za­tion was very rarely an indi­vid­ual one. In most cases it involved cliques of friends. In some cases recruit­ment was deter­mined by the indi­vid­u­al’s sol­i­dar­ity with an”impor­tant" friend who was arrested or had to go under­ground." More recent­ly, using open source mate­ri­al, Marc Sage­man (2004) ana­lyzed the cases of approx­i­mately 172 global Salafi muja­hedin and found that nearly two thirds “joined” the jihad col­lec­tively as part of a small group (“bunch of guys”) or had a long­time friend who already had joined.

One last quote (from Abrahms again):

Sec­ond, mem­bers from a wide vari­ety of ter­ror­ist group­s…say that they joined these armed strug­gles…to main­tain or develop social rela­tions with other ter­ror­ist mem­bers. These are not the state­ments of a small num­ber of ter­ror­ists; in the Turk­ish sam­ple, for instance, the 1,100 ter­ror­ists inter­viewed were 10 times more likely to say that they joined the ter­ror­ist orga­ni­za­tion ‘because their friends were mem­bers’ than because of the ‘ide­ol­ogy’ of the group.

There are other inter­est­ing points; both of Abrahm­s’s papers are well worth read­ing, as is Abrahms 2012:

A final expla­na­tion is that ter­ror­ists derive util­ity from their actions regard­less of whether gov­ern­ments com­ply polit­i­cal­ly. This inter­pre­ta­tion is con­sis­tent with the emerg­ing body of evi­dence that although ter­ror­ism is ineffec­tive for achiev­ing out­come goals, ter­ror­ism is indeed effec­tive for achiev­ing process goals (e.g., Abrahms 2008; Arce and San­dler, 2007, 2010; Bloom, 2005; Kydd and Wal­ter 2002). Whereas ter­ror­ist acts gen­er­ally fail to pro­mote gov­ern­ment con­ces­sions, the vio­lence against civil­ians can per­pet­u­ate the ter­ror­ist group by attract­ing media atten­tion, spoil­ing peace process­es, and boost­ing mem­ber­ship, morale, cohe­sion, and exter­nal sup­port­…In­deed, ter­ror­ists tend to ramp up their attacks dur­ing peace process­es, pre­clud­ing con­ces­sions (see Kydd and Wal­ter, 2002).

  • Arce, Daniel. and San­dler, Todd. (2007) “Ter­ror­ist Sig­nal­ing and the Value of Intel­li­gence”. British Jour­nal of Polit­i­cal Sci­ence 37 576–586
  • Arce, Daniel. and San­dler, Todd. (2010) “Ter­ror­ist Spec­tac­u­lars: Back­lash Attacks and the Focus of Intel­li­gence”. Jour­nal of Con­flict Res­o­lu­tion 54 354–373.
  • Bloom, Mia M. (2004) “Pales­tin­ian Sui­cide Bomb­ing: Pub­lic Sup­port, Mar­ket Share, and Out­bid­ding”. Polit­i­cal Sci­ence Quar­terly 119 61–88
  • Kydd, Andrew H. and Wal­ter, Bar­bara F. (2002) “Sab­o­tag­ing the peace: The pol­i­tics of extrem­ist vio­lence”. Inter­na­tional Orga­ni­za­tion 56 263–296

With this per­spec­tive, many things fall into place. For exam­ple, in the In Their Own Words: Voices of Jihad, the authors remark:

The Inter­net offers another exam­ple. It is awash in jihadi web sites, and there is lit­tle ques­tion that it is being exploited for train­ing, fundrais­ing, recruit­ment, and coor­di­na­tion. Yet again, when brows­ing the blogs and chat rooms, one gets the impres­sion that what is being wit­nessed is largely a form of “fan­tasy jihad.” It is not com­fort­ing to see so many obvi­ously edu­cated23 young Mus­lims play­ing the game, but their par­tic­i­pa­tion does not mean that each log-on rep­re­sents a sleeper cell.

Cer­tainly not; indeed, one could well pre­dict that ‘e-ji­had’ users will tend to be rather harm­less. It’s rather harder for online peers (com­pared to meat­space friends) to guilt one into action, after all. And one could well pre­dict that more mate­r­ial fac­tors would, say, influ­ence which cler­ics tend to become rad­i­cal­ized and jihadist, like career suc­cess in work­ing for a gov­ern­ment (Nielsen 2012).

O RLY?

The fore­go­ing was orig­i­nally posted to Less­Wrong.­com, where it was ener­get­i­cally cri­tiqued.

Terrorism does too work!

There are mul­ti­ple mem­o­rable instances where ter­ror­ism seems to work. This should be no sur­prise; after all, if ter­ror­ism never worked, would we ever be con­cerned about it? Of course not. Ter­ror­ism works, darn it!24

Cited exam­ples include the , the , , and . As one com­men­ta­tor wrote: “Let’s stop pre­tend­ing that ter­ror­ism does­n’t work. Do you think Eng­land would ever have talked with the IRA, or that Israel would have given ter­ri­tory to the Pales­tini­ans, if not for ter­ror­ism?”

NO WAI

There are sev­eral pos­si­ble replies. For exam­ple, Pape’s work is focused pretty much only on ; his find­ings on effec­tive­ness, even if cor­rect, may not gen­er­al­ize to the many non-sui­cide attacks. Fur­ther, Abrahms con­sider it unclear how rea­son­able his spe­cific analy­sis is:

Not only is his sam­ple of ter­ror­ist cam­paigns mod­est, but they tar­geted only a hand­ful of coun­tries: ten of the eleven cam­paigns ana­lyzed were directed against the same three coun­tries (Is­rael, Sri Lanka, and Turkey), with six of the cam­paigns directed against the same coun­try (Is­rael). More impor­tant, Pape does not exam­ine whether the ter­ror­ist cam­paigns achieved their core pol­icy objec­tives. In his assess­ment of Pales­tin­ian ter­ror­ist cam­paigns, for exam­ple, he counts the lim­ited with­drawals of the Israel Defense Forces from parts of the Gaza Strip and the West Bank in 1994 as two sep­a­rate ter­ror­ist vic­to­ries, ignor­ing the 167% increase in the num­ber of Israeli set­tlers dur­ing this peri­od-the most vis­i­ble sign of Israeli occu­pa­tion. Sim­i­lar­ly, he counts as a vic­tory the Israeli deci­sion to release Hamas leader Sheik from prison in Octo­ber 1997, ignor­ing the hun­dreds of impris­on­ments and tar­geted assas­si­na­tions of Pales­tin­ian ter­ror­ists through­out the Pape’s data there­fore reveal only that select ter­ror­ist cam­paigns have occa­sion­ally scored tac­ti­cal vic­to­ries, not that ter­ror­ism is an effec­tive strat­egy for groups to achieve their pol­icy objec­tives.

Another is that this is an essen­tially sta­tis­ti­cal argu­ment, over dozens or hun­dreds of ter­ror­ist groups. Adum­brat­ing 4 some­what suc­cess­ful groups would inval­i­date an asser­tion along the lines of “All ter­ror­ist groups are unsuc­cess­ful”, but of course no one is mak­ing that claim. (Just that most are.)

The pre­vi­ously quoted 0% suc­cess rate fig­ure is a bit low. “Why Ter­ror­ism Does­n’t Work” back­tracks a lit­tle, and con­sid­ers a larger group (42, not 20). This larger group has a 7% suc­cess rate.

As fre­quently not­ed, Hezbol­lah suc­cess­fully coerced the multi­na­tional peace­keep­ers and Israelis from south­ern Lebanon in 1984 and 2000, and the [1976–2009] won con­trol over the north­ern and east­ern coastal areas of Sri Lanka from 1990 on. In the aggre­gate, how­ev­er, the ter­ror­ist groups achieved their main pol­icy objec­tives only 3 out of 42 times–a 7% suc­cess rate. Within the coer­cion lit­er­a­ture, this rate of suc­cess is con­sid­ered extremely low. It is sub­stan­tially low­er, for exam­ple, than even the suc­cess rate of , which are widely regarded as only min­i­mally effec­tive.

…This study ana­lyzes the polit­i­cal plights of 28 ter­ror­ist group­s–the com­plete list of (FTOs) as des­ig­nated by the U.S. Depart­ment of State since 2001. The data yield two unex­pected find­ings. First, the groups accom­plished their 42 pol­icy objec­tives only 7% of the time.

(For one par­al­lel in reg­u­lar pol­i­tics, con­sider Gilens & Page 2014.)

Per­haps these stud­ies are sim­ply too harsh and demand­ing?

Using this list pro­vides a check against select­ing cases on the depen­dent vari­able, which would arti­fi­cially inflate the suc­cess rate because the most well known pol­icy out­comes involve ter­ror­ist vic­to­ries (e.g., the U.S. with­drawal from south­ern Lebanon in 198425). Fur­ther­more, because all of the ter­ror­ist groups have remained active since 2001, ample time has been allowed for each group to make progress on achiev­ing its pol­icy goals, thereby reduc­ing the pos­si­bil­ity of arti­fi­cially deflat­ing the suc­cess rate through too small a time frame. In fact, the ter­ror­ist groups have had sig­nifi­cantly more time than five years to accom­plish their pol­icy objec­tives: the groups, on aver­age, have been active since 1978; the major­ity has prac­ticed ter­ror­ism since the 1960s and 1970s; and only four were estab­lished after 1990.

A third coun­ters the appeal with Pape’s author­ity with the obser­va­tion that “ter­ror­ism does­n’t work” is an old vein of thought; in 1976, argued in “The Futil­ity of Ter­ror­ism” that ter­ror­ism is an ineffec­tive strat­e­gy, and said it “almost never appears to accom­plish any­thing polit­i­cally sig­nifi­cant.”26, and con­curs in this pes­simistic take27; Lomasky goes so far as to argue that ter­ror­ism is out­right coun­ter­pro­duc­tive in strength­en­ing the tar­geted gov­ern­ment.

One last con­sid­er­a­tion is that the listed groups may not’ve been very suc­cess­ful at all.

Hitting the broad side

The fol­low quotes are from Wikipedia, about the pre­vi­ously cited groups. Where pos­si­ble, I quote the sum­mary of that group’s aims.

  1. The IRA’s stated objec­tive is to end “British rule in Ire­land,” and accord­ing to its con­sti­tu­tion, it wants “to estab­lish an Irish Social­ist Repub­lic, based on the Procla­ma­tion of 1916.” Until the 1998 Belfast Agree­ment, it sought to end North­ern Ire­land’s sta­tus within the United King­dom and bring about a united Ire­land by force of arms and polit­i­cal per­sua­sion.

  2. In 1988, the PLO offi­cially endorsed a two-s­tate solu­tion, with Israel and Pales­tine liv­ing side by side con­tin­gent on spe­cific terms such as mak­ing East Jerusalem cap­i­tal of the Pales­tin­ian state and giv­ing Pales­tini­ans the right of return to land occu­pied by Pales­tini­ans prior to the 1948 and 1967 wars with Israel.

  3. Hamas wants to cre­ate an Islamic state in the West Bank and the Gaza strip, a goal which com­bines Pales­tin­ian nation­al­ism with Islamist objec­tives. Hamas’s 1988 char­ter calls for the replace­ment of Israel and the Pales­tin­ian Ter­ri­to­ries with an Islamic Pales­tin­ian state.

  4. Hezbol­lah’s 1985 man­i­festo listed its three main goals as “putting an end to any colo­nial­ist entity” in Lebanon, bring­ing the Pha­langists to jus­tice for “the crimes they [had] per­pe­trat­ed,” and the estab­lish­ment of an Islamic regime in Lebanon. Recent­ly, how­ev­er, Hezbol­lah has made lit­tle men­tion of estab­lish­ing an Islamic state, and forged alliances across reli­gious lines. Hezbol­lah lead­ers have also made numer­ous state­ments call­ing for the destruc­tion of Israel, which they refer to as a “Zion­ist enti­ty… built on lands wrested from their own­ers.”

One strik­ing thing about the goals of these groups is how few of them have been accom­plished, and how often they seem to have sab­o­taged and undone real progress towards res­o­lu­tion of their griev­ances. Any­one even slightly famil­iar with Pales­tine and Israel in par­tic­u­lar must won­der whether Hamas or the PLO have helped the Pales­tin­ian cause more than they’ve hurt (an obser­va­tion equally applic­a­ble to Ire­land).

Biases

These ideas and analy­ses can make peo­ple quite angry. They view the pre­vi­ously men­tioned orga­ni­za­tions, as well as al-Qaeda, as being such obvi­ous exam­ples that any­one sug­gest­ing that ter­ror­ism may be use­less is seen as being a naive idiot or per­haps being dis­hon­est. The level of emo­tion seems quite unwar­rant­ed, and makes me think that there may be at play.

The (when “peo­ple base their pre­dic­tion of the fre­quency of an event or the pro­por­tion within a pop­u­la­tion based on how eas­ily an exam­ple can be brought to mind”) seems to apply here. It is much eas­ier to think of claimed attacks than anony­mous ones, even though it was hinted at the begin­ning that ter­ror­ist attacks for which the group claims respon­si­bil­ity are actu­ally in the minor­i­ty! This very coun­ter-in­tu­itive claim seems to be borne out:

Since the emer­gence of mod­ern ter­ror­ism in 1968, 64% of world­wide ter­ror­ist attacks have been car­ried out by unknown per­pe­tra­tors. Anony­mous ter­ror­ism has been ris­ing, with 3 out of 4 attacks going unclaimed since Sep­tem­ber 11, 2001. Anony­mous ter­ror­ism is par­tic­u­larly preva­lent in Iraq, where the US mil­i­tary has strug­gled to deter­mine whether the vio­lence was per­pe­trated by Shi­ite or Sunni groups with vastly differ­ent polit­i­cal plat­forms.28

(Inas­much as peo­ple read about iden­ti­fied attacks and ignore anony­mous attacks, there may also be some at work as well.)

It’s about feeling better

Isn’t it pos­si­ble that many ter­ror­ist acts are really for the pur­pose of mak­ing the ter­ror­ists feel bet­ter about them­selves and their in-groups? Like teenagers play­ing pranks, only with often-lethal con­se­quences.

This is some­what differ­ent from the sug­ges­tion that ter­ror­ists join for a group to spend time with; this hypoth­e­sis is about social net­works, self­-es­teem, and repair­ing injuries to it. Ter­ror­ists are not mad293031 (de­spite an occu­pa­tion con­ducive to it32), nor are they demonic agents of destruc­tion.

That said, the data on ter­ror­ist recruit­ment sug­gests that pres­tige & power of the group or promi­nent mem­bers has more to do with the attrac­tive­ness of being a ter­ror­ist than whether a recruit’s has recently been humil­i­ated by an . Con­sider the . Were Mus­lims deeply offended by the directed against Sad­dam Hus­sein (and the con­se­quent Iraqi suffer­ing and death­s), or by the Pales­tin­ian sit­u­a­tion, then log­i­cally they would join before 9/11 so as to aid al-Qaeda in strik­ing back against the USA. Of course, recruit­ment picked up after 9/11, in the face of enor­mous inter­na­tional pres­sure on any­thing that even was rumored to have Al Qaeda links33. Promis­ing young stu­dents drop their stud­ies to go fight in Soma­li­a—as a group, not one by one.34 This is per­fectly log­i­cal and even pre­dicted by both the pres­tige and social-ties the­o­ries, but it is harder to make it con­sis­tent with the self­-es­teem the­o­ry.

Social net­works can also be . It can be easy to miss this even when the evi­dence is star­ing one in the face. From , “The World of Holy War­craft: How al Qaeda is using online game the­ory to recruit the masses”:

The coun­tert­er­ror­ism com­mu­nity has spent years try­ing to deter­mine why so many peo­ple are engaged in online jihadi com­mu­ni­ties in such a mean­ing­ful way. After all, the life of an online admin­is­tra­tor for a hard-line Islamist forum is not as excit­ing as one might expect. You don’t get paid, and you spend most of your time post­ing links and videos, com­ment­ing on other peo­ple’s links and videos, and then com­ment­ing on other peo­ple’s com­ments. So why do peo­ple like Abu­mubarak spend weeks and months and years of their time doing it? Expla­na­tions from schol­ars have ranged from the inher­ently com­pul­sive and vio­lent qual­ity of Islam to the psy­chol­ogy of ter­ror­ists.

But no one seems to have noticed that the fer­vor of online jihadists is actu­ally quite sim­i­lar to the fer­vor of any other online group. The online world of Islamic extrem­ists, like all the other worlds of the Inter­net, oper­ates on a sub­tly psy­cho­log­i­cal level that does a bril­liant job at keep­ing peo­ple like Abu­mubarak click­ing and post­ing away—and amass­ing all the rank­ings, scores, badges, and lev­els to prove it…It turns out that what dri­ves online jihadists is pretty much exactly what dri­ves Inter­net trolls, air­line ticket con­sumers, and World of War­craft play­ers: com­pe­ti­tion…­Points can result in an array of seem­ingly triv­ial rewards, includ­ing a change in the color of a mem­ber’s user­name, the abil­ity to dis­play an avatar, access to pri­vate groups, and even a change in sta­tus level from, say, “peas­ant” to “VIP.” In the con­text of the gam­i­fied sys­tem, how­ev­er, these pal­try incen­tives really mat­ter.

But for a select few, the addic­tion to win­ning bleeds over into phys­i­cal space to the point where those same incen­tives begin to shape the way they act in the real world. These indi­vid­u­als strive to live up to their vir­tual iden­ti­ties, in the way that teens have re-cre­ated the video game Grand Theft Auto in real life, car­ry­ing out rob­beries and mur­ders.

One man in par­tic­u­lar has been able to take advan­tage of the incen­tives of online gam­i­fi­ca­tion to pur­sue real-life ter­ror­ist recruits: , the Amer­i­can-born al Qaeda cleric hid­ing in Yemen, famous for hav­ing helped encour­age a num­ber of West­ern-based would-be jihadists into action. , the alleged Fort Hood shooter, for exam­ple, mas­sa­cred a dozen sol­diers after exchang­ing a num­ber of emails with Awla­ki. , the Times Square bomber, admit­ted Awlaki influ­enced him, and was one of Awlak­i’s stu­dents prior to attempt­ing to blow up an air­plane on Christ­mas Day 2009…His sup­port­ers vie for the right to con­nect with Awlaki, whether vir­tu­ally or actu­al­ly—a pow­er­ful incen­tive that, from our obser­va­tion, dri­ves many of them into, at the very least, more active lan­guage about jihad.

A user who called him­self “Belaid” on Awlak­i’s now-de­funct blog boasted to oth­ers about what he per­ceived to be a response to his email in Awlak­i’s lat­est blog post, say­ing: “S. Anwar Al-Awlaki i sin­cerely love u for the Sake of Allah for what you are doing, I think you answered my e-mail by giv­ing us this doc­u­ment.” He then fol­lowed up by express­ing his desire to tran­si­tion from vir­tual com­mu­ni­ca­tion to real com­mu­ni­ca­tion. “I ask Allah to make me go visit you so I can see you in real and we in sha Allah go together do jihad insha Allah in our life time!!!” he wrote in Jan­u­ary 2009.

The right inter­pre­ta­tion is almost too obvi­ous to give. World of War­craft is not about com­pe­ti­tion any more than those forums are; the lit­er­a­ture on MMORPGs and since the 1980s (and even video games35) in gen­eral have con­cen­trated on the social aspects36 of online inter­ac­tions. It’s a com­mon­place that long-term play­ers of Ultima Online—no, Everquest, no, World of War­craft—do so not in order to com­pete for the high­est player level37 but in order to con­tinue play­ing with their . (In line with the fol­low­ing sec­tion, mar­riages between play­ers who met in guilds are far from unheard of; they are no longer even news.)

If we see ter­ror­ism as more of a tribal or gang activ­ity than polit­i­cal activism or war­fare, then online con­nec­tions become espe­cially impor­tant to our analy­sis, oth­er­wise we will be fooled by so-called . Ear­lier ‘lone wolves’ like bombers or turn out on closer inspec­tion to have ties, social & oth­er­wise, to like-minded peo­ple; McVeigh lived with sev­eral other extrem­ists and was taught his bom­b-mak­ing skills by the Nichols, who also built the final bomb with him, while Rudolph remained on the run for sev­eral years in a com­mu­nity that wrote songs and sold t-shirts to praise him and was ulti­mately caught clean-shaven & wear­ing new sneak­ers. Lone wolves who gen­uinely had no con­tact with their con­fr­eres, such as , are van­ish­ingly rare excep­tions among the dozens of thou­sands of ter­ror­ist attacks in the 20th cen­tu­ry, and as rare excep­tions, oth­er­wise implau­si­ble expla­na­tions like men­tal dis­ease account for them with­out trou­ble.

About the chicks, man

One com­menter sug­gested that Abrahms almost has it right. Ter­ror­ists are seek­ing social ties, but only as a sub­sti­tute for female com­pan­ion­ship. The spe­cific exam­ple was the Amer­i­can novel/movie ; cer­tain­ly, when one thinks about it, it’s hard to not notice that the nar­ra­tor goes–thanks to lead­ing a ter­ror­ist orga­ni­za­tion–from being a sin­gle loser who has to pre­tend to be ill (men­tally and phys­i­cal­ly) to get any atten­tion or social inter­ac­tion, to being an incred­i­bly pop­u­lar guy with dozens of sub­or­di­nates to hang out with day and night and a girl­friend.

But an even bet­ter exam­ple might be Fatah’s cell.

In ’s “All You Need Is Love”, Bruce Hoff­man writes that a senior Fatah gen­eral told him of how they decided that Black Sep­tem­ber had out­lived its use­ful­ness, and needed to be dis­solved. But that was prob­lem­at­ic. Black Sep­tem­ber likely would not take dis­so­lu­tion lying down:

‘It was the most elite unit we had. The mem­bers were sui­ci­dal—not in the sense of reli­gious ter­ror­ists who sur­ren­der their lives to ascend to heaven but in the sense that we could send them any­where to do any­thing and they were pre­pared to lay down their lives to do it. No ques­tion. No hes­i­ta­tion. They were absolutely ded­i­cated and absolutely ruth­less.’

What, then, did Fatah do in 1973? They must’ve suc­ceed­ed; we all know Black Sep­tem­ber is ancient his­to­ry.

My host, who was one of most trusted deputies, was charged with devis­ing a solu­tion. For months both men thought of var­i­ous ways to solve the Black Sep­tem­ber prob­lem, dis­cussing and debat­ing what they could pos­si­bly do, short of killing all these young men, to stop them from com­mit­ting fur­ther acts of ter­ror.

Finally they hit upon an idea. Why not sim­ply marry them off? In other words, why not find a way to give these men—the most ded­i­cat­ed, com­pe­tent, and implaca­ble fight­ers in the entire PLO—a rea­son to live rather than to die? Hav­ing failed to come up with any viable alter­na­tives, the two men put their plan in motion.

And it worked!

So approx­i­mately a hun­dred of these beau­ti­ful young women were brought to Beirut. There, in a sort of PLO ver­sion of a col­lege mix­er, boy met girl, boy fell in love with girl, boy would, it was hoped, marry girl. There was an addi­tional incen­tive, designed to facil­i­tate not just amorous con­nec­tions but long-last­ing rela­tion­ships. The hun­dred or so Black Sep­tem­berists were told that if they mar­ried these wom­en, they would be paid $12,178; given an apart­ment in Beirut with a gas stove, a refrig­er­a­tor, and a tele­vi­sion; and employed by the PLO in some non­vi­o­lent capac­i­ty. Any of these cou­ples that had a baby within a year would be rewarded with an addi­tional $20,296.

Both Abu Iyad and the future gen­eral wor­ried that their scheme would never work. But, as the gen­eral recount­ed, with­out excep­tion the Black Sep­tem­berists fell in love, got mar­ried, set­tled down, and in most cases started a fam­i­ly…the gen­eral explained, not one of them would agree to travel abroad, for fear of being arrested and los­ing all that they had—that is, being deprived of their wives and chil­dren. ‘And so’, my host told me, ‘that is how we shut down Black Sep­tem­ber and elim­i­nated ter­ror­ism. It is the only suc­cess­ful case that I know of.’

Of course, the base rate for becom­ing a dis­pos­sessed young man becom­ing a ter­ror­ist is so low that it would­n’t be a good use of young women to try to pre­vent ter­ror­ism by mar­ry­ing them off, while if you can tar­get the mar­riages to known ter­ror­ists, you have enough infor­ma­tion that you would be bet­ter off just impris­on­ing or exe­cut­ing them. Sim­i­lar­ly, cases of women falling in love with jihadis online or through Twit­ter and trav­el­ing in groups to the Mid­dle East are not impor­tant in an absolute num­bers sense but for the impli­ca­tions about their psy­chol­o­gy.

Black Sep­tem­ber is inter­est­ing for what the effect of mar­riage says about the moti­va­tions of their mem­bers, not as a pro­to­type of a use­ful sup­pres­sion strat­e­gy—­most coun­tries do not have the same rela­tion to ter­ror­ist groups that Fatah had to Black Sep­tem­ber, and can adopt more effec­tive strate­gies.


  1. As well as delib­er­ate sab­o­tage of pro­duc­tive peace pro­pos­als (to which deci­sion the­o­rists might react in hor­ror, after all, one can always break a peace if it no longer seems like the course of action with the high­est mar­ginal return): Kydd, Andrew H. and Wal­ter, Bar­bara F. (2002) “Sab­o­tag­ing the peace: The pol­i­tics of extrem­ist vio­lence”. Inter­na­tional Orga­ni­za­tion 56 263–296.↩︎

  2. “Does Ter­ror­ism Really Work? Evo­lu­tion in the Con­ven­tional Wis­dom since 9/11”, Max Abrahms 2012:

    In the 1980s, Cren­shaw (Cren­shaw, 1988, 15) like­wise observed that ter­ror­ists do not obtain their given polit­i­cal ends, and “There­fore one must con­clude that ter­ror­ism is objec­tively a fail­ure.” Sim­i­lar­ly, the RAND Cor­po­ra­tion (Cordes et al., 1984, 49) remarked at the time that “Ter­ror­ists have been unable to trans­late the con­se­quences of ter­ror­ism into con­crete polit­i­cal gains. . .[I]n that sense ter­ror­ism has failed. It is a fun­da­men­tal fail­ure.” In the 1990s, Held (1991, 70) asserted that the “net effect” of ter­ror­ism is polit­i­cally coun­ter­pro­duc­tive. Chai (1993, 99) declared that ter­ror­ism “has rarely pro­vided polit­i­cal ben­e­fits” at the bar­gain­ing table. Schelling (1991, 20) agreed, pro­claim­ing that “Ter­ror­ism almost never appears to accom­plish any­thing polit­i­cally sig­nifi­cant.” Since the Sep­tem­ber 11 attacks, a series of large-n obser­va­tional stud­ies has offered a firmer empir­i­cal basis. These indi­cate that although ter­ror­ism is chill­ingly suc­cess­ful in count­less ways, coerc­ing gov­ern­ment com­pli­ance is not one of them.6…Hard case stud­ies (Abrahms, 2010; Cron­in, 2009; Dan­nen­baum, 2011; Moghadam, 2006; Neu­mann and Smith, 2007) have inspected the lim­ited his­tor­i­cal exam­ples of clear-cut ter­ror­ist vic­to­ries, deter­min­ing that these salient events were idio­syn­crat­ic, unre­lated to the harm­ing of civil­ians, or both.

    • Cren­shaw, M. (1988). “The sub­jec­tive real­ity of the ter­ror­ist: Ide­o­log­i­cal and psy­cho­log­i­cal fac­tors in ter­ror­ism”. In R. Slater and M. Stohl (Ed­s.), Cur­rent per­spec­tives on inter­na­tional ter­ror­ism (pp. 12–46)
    • Held, Vir­ginia. (1991) “Ter­ror­ism, Rights, and Polit­i­cal Goals”. In Vio­lence, Ter­ror­ism, and Jus­tice, edited by R.G. Frey and Christo­pher W. Mor­ris
    • Chai, Sun-Ki. 1993. “An Orga­ni­za­tional Eco­nom­ics The­ory of Anti-Gov­ern­ment Vio­lence”, Com­par­a­tive Pol­i­tics 26(1): 99–110.
    • Schelling, Thomas C. (1991) “What Pur­poses Can Inter­na­tional Ter­ror­ism Serve?” In Vio­lence, Ter­ror­ism, and Jus­tice, edited by Ray­mond Gille­spie Frey and Christo­pher W. Mor­ris
    • Moghadam, Assaf. (2006) “Sui­cide ter­ror­ism, occu­pa­tion, and the glob­al­iza­tion of mar­tyr­dom: A cri­tique of dying to win”. Stud­ies in Con­flict and Ter­ror­ism 29 707–729
    ↩︎
  3. Ref­er­ences:

    ↩︎
  4. Abrahms 2012:

    Gaibul­loev and San­dler (2009) ana­lyze a dataset of inter­na­tional hostage crises from 1978 to 2005. They exploit vari­a­tion in whether the hostage-tak­ers esca­late by killing the hostages instead of releas­ing them unscathed. The study finds that hostage-tak­ers sig­nifi­cantly lower the odds of achiev­ing their demands by inflict­ing phys­i­cal harm in the course of the stand­off. The authors con­clude that ter­ror­ists gain bar­gain­ing lever­age from restraint, as esca­lat­ing to “blood­shed does not bol­ster a nego­ti­ated out­come” (19).

    • Gaibul­lo­ev, Khus­rav. and San­dler, Todd. (2009) “Hostage Tak­ing: Deter­mi­nants of Ter­ror­ist Logis­ti­cal and Nego­ti­a­tion Suc­cess”. Jour­nal of Peace Research 46 739–756
    ↩︎
  5. Abrahms 2012:

    In a cou­ple of sta­tis­ti­cal papers, Berrebi and Klor (2006, 2008) demon­strate that ter­ror­ist fatal­i­ties within Israel sig­nifi­cantly boost local sup­port for right-bloc par­ties opposed to accom­mo­da­tion, such as the Likud. Other quan­ti­ta­tive work goes even fur­ther, reveal­ing that the most lethal ter­ror­ist inci­dents in Israel are the most likely to induce this right­ward elec­toral shift. The authors (Gould and Klor, 2010, 1507) con­clude that height­en­ing the pain to civil­ians tends to “back­fire on the goals of ter­ror­ist fac­tions by hard­en­ing the stance of the tar­geted pop­u­la­tion.” These trends do not appear to be Israel-spe­cific. [Gasseb­n­er, Jong-A-Pin, & Mireau 2008 find that esca­lat­ing to ter­ror­ism or with ter­ror­ism helps non-s­tate actors to remove incum­bent lead­ers of tar­get coun­tries from polit­i­cal office. Unfor­tu­nately for the ter­ror­ists, how­ev­er, tar­get coun­tries tend to become even less likely to grant con­ces­sion­s.] Chowani­etz (2010) ana­lyzes vari­a­tion in pub­lic opin­ion within France, Ger­many, Spain, the United King­dom, and the United States from 1990 to 2006. For each tar­get coun­try, ter­ror­ist attacks have shifted the elec­torate to the polit­i­cal right in pro­por­tion to their lethal­i­ty. More anec­do­tal­ly, sim­i­lar obser­va­tions (Mueller, 2006, 184; Neu­mann and Smith, 2005, 587; Wilkin­son, 1986, 52) have been reg­is­tered after mass casu­alty ter­ror­ist attacks in Egypt, Indone­sia, Jor­dan, the Philip­pines, Rus­sia, and Turkey. Hewitt (1993, 80) offers this syl­lo­gism of how tar­get coun­tries typ­i­cally respond: “The pub­lic favors hard-line poli­cies against ter­ror­ism. Con­ser­v­a­tive par­ties are more likely to advo­cate hard-line poli­cies. There­fore, the pub­lic will view con­ser­v­a­tive par­ties as the best.” In a more recent sum­mary of the lit­er­a­ture, RAND (Ber­re­bi, 2009, 189–190) also deter­mi­nes: “Ter­ror­ist fatal­i­ties, with few excep­tions, increase sup­port for the bloc of par­ties asso­ci­ated with a more-in­tran­si­gent posi­tion. Schol­ars may inter­pret this as fur­ther evi­dence that ter­ror­ist attacks against civil­ians do not help ter­ror­ist orga­ni­za­tions achieve their stated goals (e.g., Abrahms, 2006).” Psy­chol­o­gists (e.g., Jost 2006, Jost 2008) have repli­cated these results in lab­o­ra­tory exper­i­ments, fur­ther rul­ing out the pos­si­bil­ity of a selec­tion effect dri­ving the results. Con­sis­tent with these quan­ti­ta­tive stud­ies, his­tor­i­cal research (e.g., Cron­in, 2009; Jones and Libicki, 2008) on ter­ror­ism is also find­ing that the stan­dard gov­ern­men­tal response is not accom­mo­da­tion, but provo­ca­tion par­tic­u­larly after the blood­i­est attacks.

    Per­haps unsur­pris­ing­ly, the most noto­ri­ous rebel lead­ers in mod­ern his­tory from Abdul­lah Yusuf Azzam to Regis Debray, Vo Nguyen Giap, Che Gue­vara, and Car­los Marighela admon­ished their foot-sol­diers against tar­get­ing the pop­u­la­tion since the indis­crim­i­nate vio­lence was prov­ing coun­ter­pro­duc­tive (Rapoport, 2004, 54–55; Wein­stein, 2007, 30–31; and Wilkin­son, 1986, 53, 59, 100, 112). In the months lead­ing up to his death, even Osama bin Laden com­manded his lieu­tenants to refrain from tar­get­ing West­ern civil­ians because in his view the indis­crim­i­nate vio­lence was not hav­ing the desired effect on their gov­ern­ments (“Bin Laden against Attacks on Civil­ians, Deputy Says,” Reuters, 2011-02-25). Accord­ing to con­tem­po­rary news accounts (“For Arab Awak­en­ing, Bin Laden Was Already Dead,” Radio Free Europe, 2011-05-04), this grow­ing con­sen­sus is behind the pri­macy of non­vi­o­lence over ter­ror­ism in the Arab Awak­en­ing engulfing the Mid­dle East and North Africa…­More sys­tem­at­i­cal­ly, Pape (1996) sur­veys the uni­verse of strate­gic bomb­ing cam­paigns from the First World War to the 1990 Per­sian Gulf War. His analy­sis reveals that gov­ern­ments reach an infe­rior bar­gain when their cam­paigns tar­get the pop­u­la­tion, an assess­ment reaffirmed in inde­pen­dent sta­tis­ti­cal analy­sis. In the most com­pre­hen­sive and recent study, Cochran and Downes (2011) exploit vari­a­tion in the use of civil­ian vic­tim­iza­tion cam­paigns on inter­state war out­comes from 1816 to 2007. Their research shows that mil­i­tary lead­ers and politi­cians err in think­ing that civil­ian vic­tim­iza­tion pays. Though obvi­ously suc­cess­ful in stamp­ing out count­less civil­ians, indis­crim­i­nate bomb­ings, sieges, mis­sile strikes, and other painful meth­ods against the pop­u­la­tion do not yield a supe­rior set­tle­ment regard­less of the costs.

    ↩︎
  6. This back­lash effect seems to’ve been delib­er­ately exploited on occa­sions; “When It Pays to Talk to Ter­ror­ists”, NYT:

    Most schol­ars of the now agree that attacks like were designed by rivals to shift power away from mod­er­ates and into the hands of more rad­i­cal fac­tions. The string of attacks attrib­uted to the between Novem­ber 1971 and March 1973, of which Munich was the most dra­mat­ic, were actu­ally an indi­ca­tion of the rifts within the P.L.O. While events like Munich seized head­li­nes, a grow­ing num­ber of mod­er­ates within the P.L.O.—­most notably Arafat—were putting out feel­ers about the prospect of a two-s­tate solu­tion in the Israeli-Pales­tin­ian dis­pute.

    Although their rhetoric con­tin­ued to call for Israel’s destruc­tion, mod­er­ate lead­ers sent pri­vate sig­nals indi­cat­ing a will­ing­ness to com­pro­mise. “We need a change of tac­tics,” Arafat told Soviet offi­cials in 1971. “We can­not affect the out­come of the polit­i­cal set­tle­ment unless we par­tic­i­pate in it.” He then drew a map out­lin­ing a two-s­tate solu­tion for Israel and Pales­tine. As State Depart­ment offi­cials rec­og­nized in June 1972, the “young wolves” in the move­ment had forced Arafat to “back off” from seri­ous peace over­tures in order to remain in pow­er.

    Munich was also engi­neered to elicit vio­lent reprisals from the Israeli gov­ern­men­t—which it did in the form of airstrikes against Pales­tin­ian refugee camps in Lebanon and Syria that killed hun­dreds, mostly civil­ians. Per­suaded of the fun­da­men­tal evil that Pales­tin­ian mil­i­tants rep­re­sent­ed, Amer­i­can lead­ers remained stead­fast in their refusal to con­demn Israel for its attacks on Syria and Lebanon, choos­ing instead to cast Amer­i­ca’s first lone veto of a Secu­rity Coun­cil Res­o­lu­tion on Sept. 10, 1972. The veto affirmed Wash­ing­ton’s posi­tion on the P.L.O.: no recog­ni­tion, no nego­ti­a­tion and no legit­i­macy for ter­ror­ists.

    “Insti­tu­tions will try to pre­serve the prob­lem to which they are the solu­tion”; one sus­pects that the rar­ity of ter­ror­ism plus this back­fire effect is the rea­son so many have been con­ducted or by gov­ern­ments, as epit­o­mized by the “”. This obser­va­tion would also explain other odd­i­ties where we notice that secret police seem to often hold off on death blows, engage in ter­ror­ism them­selves, or sup­port it over­seas despite the pre­dictable risk of cat­a­strophic blow­back: many well-known instances can be listed such as FBI infil­tra­tion of the Ku Klux Klan (peak­ing at infor­mants com­pris­ing up to 20% of its mem­bers, lead­er­ship posi­tions in 7 of 14 groups, head of one state’s Klans­men, run­ning its own splin­ter group, and pos­si­bly shield­ing infor­mants who mur­dered); the FBI’s reg­u­lar cre­ation of ‘ter­ror­ist plots’ dur­ing the War on Ter­ror; Adolf Hitler join­ing the Ger­man Work­ers’ Party at the behest of the Reich­swehr as an infor­mant; the old rumors that Stalin was an Okhrana mole in addi­tion to the Okhrana forg­ing , Ger­man fund­ing & logis­tics sup­port for Lenin & the Bol­she­viks; CIA sup­port for jihadists (part of a net­work of events lead­ing to 9/11, Afghanistan, and Iraq); the (spon­sored by the Ser­bian gov­ern­ment, which then expe­ri­enced the blow­back of WWI); Ger­many con­tin­ues to inves­ti­gate how the was funded by intel­li­gence agen­cies & per­mit­ted to keep killing despite being infil­trated (Neo-Nazi groups in Ger­many are par­tic­u­larly noto­ri­ous for being rid­dled with infor­mants and gov­ern­ment agents); Pak­istan’s has long funded or con­trolled Islamist groups intended for use against India (par­tic­u­larly in the Kash­mir) despite the exis­ten­tial threat those groups pose to the Pak­istani state etc. The case of the Okhrana is suffi­ciently strik­ing as to be worth quot­ing: “From the Okhrana to the KGB: Con­ti­nu­ities in Russ­ian for­eign intel­li­gence oper­a­tions since the 1880s”, Andrew 1989?:

    After its foun­da­tion in 1881, the Okhrana rapidly devel­oped a net­work of agents and agents provo­ca­teurs, ini­tially to pen­e­trate the rev­o­lu­tion­ary Dias­pora abroad. In 1886, Rachkovsky’s agents blew up the Peo­ple’s Will print­works in Geneva, suc­cess­fully giv­ing the impres­sion that the explo­sion was the work of dis­affected rev­o­lu­tion­ar­ies. In 1890, Rachkovsky unmasked a sen­sa­tional bom­b-mak­ing con­spir­acy by Russ­ian emi­gres in Paris; the lead­ing plot­ter was, in real­i­ty, one of his own agents provo­ca­teurs. The most suc­cess­ful intel­li­gence pen­e­tra­tion any­where in Europe before World War I was the Russ­ian recruit­ment of the senior Aus­trian mil­i­tary intel­li­gence offi­cer, Colonel Alfred Redl. The Redl sto­ry, like those of the Cam­bridge moles, has been embroi­dered with a good many fan­tasies. But even the unem­broi­dered story is remark­able. In the win­ter of 1901–1902, Colonel Batyush­in, head of Russ­ian mil­i­tary intel­li­gence in War­saw, dis­cov­ered that, unknown either to his supe­ri­ors or to his friends, Redl was a promis­cu­ous homo­sex­u­al. By a mix­ture of black­mail and bribery of the kind some­times employed later by the KGB; he recruited Redl as a pen­e­tra­tion agent. With the money given him by the Rus­sians, Redl was able to pur­chase cars for him­self and for one of his favorite lovers, a young Uhlan offi­cer to whom he paid 600 crowns a month. Redl pro­vided volu­mi­nous intel­li­gence dur­ing the decade before his sui­cide in 1913, includ­ing Aus­tri­a’s mobi­liza­tion plans against both Rus­sia and Ser­bia.4

    The Bol­she­viks learned from Okhrana files after the Feb­ru­ary Rev­o­lu­tion that almost from the moment the Russ­ian Social Demo­c­ra­tic Labour Party split into Bol­she­viks and Men­she­viks in 1903, they had been more suc­cess­fully pen­e­trated than per­haps any other rev­o­lu­tion­ary group. Okhrana knowl­edge of Bol­she­vik organ­i­sa­tion and activ­i­ties was so detailed and thor­ough that, though some of its records were scat­tered when its offices were sacked in the after­math of the Feb­ru­ary Rev­o­lu­tion, what sur­vived has become one of the major doc­u­men­tary sources for early Bol­she­vik his­to­ry. Of the five mem­bers of the Bol­she­vik Par­ty’s St. Peters­burg Com­mit­tee in 1908 and 1909, no less than four were Okhrana agents.5 The most remark­able mole, recruited by the Okhrana in 1910, was a Moscow worker named Roman Mali­novskv, who in 1912 was elected as one of the six Bol­she­vik deputies in the Duma, the tsarist par­lia­ment. “For the first time,” wrote Lenin enthu­si­as­ti­cal­ly, “we have an out­stand­ing leader (Ma­li­novsky) from among the work­ers rep­re­sent­ing us in the Duma.” In a party ded­i­cated to pro­le­tar­ian rev­o­lu­tion but as yet with­out pro­le­tar­ian lead­ers, Lenin saw Mali­novsky, whom he brought on to the Bol­she­vik Cen­tral Com­mit­tee, as a por­tent of great impor­tance: “It is really pos­si­ble to build a work­ers’ party with such peo­ple, though the diffi­cul­ties will be incred­i­bly great.”…By 1912, Lenin was so con­cerned by the prob­lem of Okhrana pen­e­tra­tion that, on his ini­tia­tive, the Bol­she­vik Cen­tral Com­mit­tee set up a three­-man “provo­ca­tion com­mis­sion” that included Mali­novsky… S. P. Belet­sky, the direc­tor of the Police Depart­ment, described Mali­novsky as “the pride of the Okhrana.” But the strain of his dou­ble life even­tu­ally proved too much. Even Lenin, his strongest sup­port­er, became con­cerned about his heavy drink­ing. In May 1914, the new Deputy Min­is­ter of the Inte­ri­or, V. F. Dzhunkovsky, pos­si­bly fear­ing the scan­dal that would result if Mali­novsky’s increas­ingly erratic behav­ior led to the rev­e­la­tion that the Okhrana employed him as an agent in the Duma, decided to get rid of him. Mali­novsky resigned from the Duma, and he fled from St. Peters­burg with a 6,000-ru­ble pay­off that the Okhrana urged him to use to start a new life abroad. But Lenin had been so thor­oughly deceived that, when proof of Mali­novsky’s guilt emerged from Okhrana files opened after the Feb­ru­ary Rev­o­lu­tion in 1917, he at first refused to believe it.7

    ↩︎
  7. “The Cred­i­bil­ity Para­dox: Vio­lence as a Dou­ble-Edged Sword in Inter­na­tional Pol­i­tics”, Abrahms 2013↩︎

  8. “Why Is It So Hard to Find a Sui­cide Bomber These Days?”, Charles Kurz­man, Sep­tem­ber 2011:

    Recruit­ment diffi­cul­ties have cre­ated a bot­tle­neck for Islamist ter­ror­ists’ sig­na­ture tac­tic, sui­cide bomb­ing. These orga­ni­za­tions often claim to have wait­ing lists of vol­un­teers eager to serve as mar­tyrs, but if so they’re not very long. Al Qaeda orga­nizer Khalid Sheikh Mohammed made this point unin­ten­tion­ally dur­ing a 2002 inter­view, sev­eral months before his cap­ture. Mohammed bragged about al Qaeda’s abil­ity to recruit vol­un­teers for “mar­tyr­dom mis­sions”, as Islamist ter­ror­ists call sui­cide attacks. “We were never short of poten­tial mar­tyrs. Indeed, we have a depart­ment called the Depart­ment of Mar­tyrs.” “Is it still active?” asked Yosri Fouda, an Al Jazeera reporter who had been led, blind­fold­ed, to Mohammed’s apart­ment in Karachi, Pak­istan. “Yes, it is, and it always will be as long as we are in jihad against the infi­dels and the Zion­ists. We have scores of vol­un­teers. Our prob­lem at the time was to select suit­able peo­ple who were famil­iar with the West.” Notice the scale here: “scores”, not hun­dred­s—and most deemed not suit­able for ter­ror­ist mis­sions in the West. After Mohammed’s cap­ture and “enhanced inter­ro­ga­tion” by the CIA, using meth­ods that the U.S. gov­ern­ment had denounced for decades as tor­ture, fed­eral offi­cials tes­ti­fied that Mohammed had trained as many as 39 oper­a­tives for sui­cide mis­sions and that the 9/11 attacks involved 19 hijack­ers “because that was the max­i­mum num­ber of oper­a­tives that Sheikh Mohammed was able to find and send to the U.S. before 9/11.” Accord­ing to a top White House coun­tert­er­ror­ism offi­cial, the ini­tial plans for 9/11 called for a simul­ta­ne­ous attack on the U.S. West Coast, but al Qaeda could not find enough qual­i­fied peo­ple to carry it out. Mohammed’s claim that al Qaeda was “never short of poten­tial mar­tyrs” seems to have been false brava­do.

    …How­ev­er, all these esti­mates must be regarded as exag­ger­a­tions. By the U.S. Jus­tice Depart­men­t’s count, approx­i­mately a dozen peo­ple in the coun­try were con­victed in the five years after 9/11 for hav­ing links with al Qae­da. Dur­ing this peri­od, fewer than 40 Mus­lim Amer­i­cans planned or car­ried out acts of domes­tic ter­ror­ism, accord­ing to an exten­sive search of news reports and legal pro­ceed­ings that I con­ducted with David Schanzer and Ebrahim Moosa of Duke Uni­ver­si­ty. None of these attacks was found to be asso­ci­ated with al Qae­da. A month after Taher­i-Azar’s attack in Chapel Hill, Mueller vis­ited North Car­olina and warned of Islamist vio­lence “all over the coun­try.” For­tu­nate­ly, that pre­dic­tion was also wrong. To put this in con­text: Out of more than 150,000 mur­ders in the United States since 9/11—currently more than 14,000 each year—Is­lamist ter­ror­ists accounted for fewer than three dozen deaths by the end of 2010.

    ↩︎
  9. That pre­vi­ous study is Max Abrahm­s’s “Why Ter­ror­ism Does Not Work”; Inter­na­tional Secu­rity 31.2 (2006) 42–78.↩︎

  10. “Death by a Thou­sand Cuts”, For­eign Pol­icy↩︎

  11. “Qaeda Branch Aimed for Broad Dam­age at Low Cost”, New York Times↩︎

  12. Scott Atran 2003, “Gen­e­sis of Sui­cide Ter­ror­ism” review↩︎

  13. “Craig Whit­lock, "Al-Qaeda Mas­ters Ter­ror­ism on the Cheap," The Wash­ing­ton Post, August 24, 2008.”↩︎

  14. “David Axe, "Sol­diers, Marines Team Up in ‘Trail­blazer’ Patrols", National Defense: NDIA’s Busi­ness and Tech­nol­ogy Mag­a­zine, April 2006.”↩︎

  15. Bjorn Lom­borg, “Is Coun­tert­er­ror­ism Good Value for Mon­ey?” The Mechan­ics of Ter­ror­ism, NATO Review, April 2008.↩︎

  16. pg 91, , RAND 2010.↩︎

  17. Wired, “$265 Bomb, $300 Bil­lion War: The Eco­nom­ics of the 9/11 Era’s Sig­na­ture Weapon”, sourc­ing 2011 esti­mates from :

    …ac­cord­ing to the Pen­tagon’s bomb squad, the aver­age cost of an IED is just a few hun­dred bucks, pocket change to a well-funded insur­gency. Worse, over time, the aver­age cost of the cheapo IEDs have dropped from $1,533 in 2006 to $343 in 2009. A killing machine, in other words, costs less than a 32-gig iPhone…On aver­age, a “vic­tim-op­er­ated” bom­b—one set to explode when its tar­get or a civil­ian inad­ver­tently sets it off—­cost a mere $343…The next most plen­ti­ful cat­e­gory of bomb, those set off with com­mand wires lead­ing from the device, also cost $343 on aver­age in 2009, account­ing for another 23.8% of attack­s…Bombs acti­vated with a remote det­o­na­tor like a cell­phone cost a mere $445 and accounted for a sur­pris­ingly smal­l­—12.6%—of attacks, per­haps owing to the U.S.’ hard-won abil­ity to jam the det­o­na­tor sig­nal…­For insur­gents to turn a car into a bomb or con­vince some­one to kill him­self dur­ing a det­o­na­tion—or both­—the cost shoots up into the thou­sands: $12,952 for a sui­cide bomber; $19,779 for a car bomb; nearly 19 grand to drive a car bom­b…­Most of those bombs have got­ten cheaper to pro­duce. In 2006, vic­tim-op­er­ated IEDs cost an aver­age of $1,533. Com­mand-wire bombs were $1,725. Remote det­o­na­tion bombs? The same…­Car bombs cost $2,283 on aver­age in 2006—which seems absurdly low, given the cost of one involves acquir­ing and then trick­ing out a car. And the going rate on sui­cide bombers appears to have risen, from $8,130 in 2006 to nearly dou­ble that in 2009. Accord­ing­ly, both accounted for over 16% of IED attacks in ’06. And JIEDDO says it has pre­lim­i­nary report­ing indi­cat­ing that sui­cide bombers cost $38,732 as of Jan­u­ary.

    ↩︎
  18. “AP sources: Christ­mas bomber con­sid­ered Hous­ton, Chicago, then picked cheap­est option: Detroit”, Wash­ing­ton Post↩︎

  19. From the ; “12 peo­ple killed, 11 injured, in attack on Paris offices of satir­i­cal news­pa­per”:

    It is under­stood the gun­men ini­tially burst into num­ber 6 rue Nico­las-Ap­pert in a Paris neigh­bour­hood, where the archives of the Char­lie Hebdo are based, shout­ing “is this Char­lie Hebdo?” before real­is­ing they had got the wrong address.

    ↩︎
  20. “Two British men admit to link­ing up with extrem­ist group in Syr­ia: Yusuf Sar­war and Mohammed Ahmed, who were reported to police by fam­ily mem­ber, plead guilty to ter­ror­ism offences”:

    Two Birm­ing­ham men admit­ted on Tues­day to link­ing up with an extrem­ist group fight­ing jihad in Syr­ia, after a fam­ily mem­ber reported them to the police…Po­lice did not know the men had trav­elled to Syr­ia, where they spent eight months, until one of their moth­ers con­tacted detec­tives in May last year, shortly after the pair had left. She had found a note writ­ten by her son say­ing he had gone to fight and wished to “die as a mar­tyr”. The men were arrested on their return to Britain in Jan­u­ary and detec­tives found pic­tures of them pos­ing with weapons and believe the pair had been in or around Alep­po, the scene of heavy fight­ing between forces loyal to Syr­ian pres­i­dent Bashar al-As­sad and rebel groups, among whom are organ­i­sa­tions the west now con­sid­ers as extrem­ist…Their path to rad­i­cal­i­sa­tion involved inspi­ra­tion from mate­r­ial from Osama bin Laden’s men­tor, Abdul­lah Azzam, online mate­ri­al, and using the inter­net to chat with extrem­ists over­seas. As part of their prepa­ra­tions they ordered books online from Ama­zon, includ­ing titles such as Islam For Dum­mies, the Koran For Dum­mies and Ara­bic For Dum­mies.

    ↩︎
  21. Quoted from foot­note 81, page 23↩︎

  22. A rate that may reflect var­i­ous assump­tions the researchers made or cul­tural differ­ences, as a 2011 Dan­ish paper found dis­count rates closer to 5%.↩︎

  23. Atran 2003:

    Research by Krueger and Maleck­ova sug­gests that edu­ca­tion may be uncor­re­lat­ed, or even pos­i­tively cor­re­lat­ed, with sup­port­ing ter­ror­ism (26). In a Decem­ber 2001 poll of 1357 West Bank and Gaza Pales­tini­ans 18 years of age or old­er, those hav­ing 12 or more years of school­ing sup­ported armed attacks by 68 points, those with up to 11 years of school­ing by 63 points, and illit­er­ates by 46 points. Only 40% of per­sons with advanced degrees sup­ported dia­logue with Israel ver­sus 53% with col­lege degrees and 60% with 9 years or less of school­ing. In a com­par­i­son of Hezbol­lah mil­i­tants who died in action with a ran­dom sam­ple of Lebanese from the same age group and region, mil­i­tants were less likely to come from poor homes and more likely to have had sec­ondary-school edu­ca­tion…A Sin­ga­pore Par­lia­men­tary report on 31 cap­tured oper­a­tives from and other Al-Qaida allies in South­east Asia under­scores the pat­tern: “These men were not igno­rant, des­ti­tute or dis­en­fran­chised. All 31 had received sec­u­lar edu­ca­tion. . . . Like many of their coun­ter­parts in mil­i­tant Islamic orga­ni­za­tions in the region, they held nor­mal, respectable jobs. . . . As a group, most of the detainees regarded reli­gion as their most impor­tant per­sonal val­ue. . . secrecy over the true knowl­edge of jihad, helped cre­ate a sense of shar­ing and empow­er­ment vis-a-vis oth­ers.” (35; “White Paper—The Jemaah Islamiyah Arrests”, Sin­ga­pore Min­istry of Home Affairs, 2003-01-09)…in Pak­istan, lit­er­acy and dis­like for the United States increased as the num­ber of reli­gious madrasa schools increased from 3000 to 39,000 since 1978 (27, 38).

    ↩︎
  24. See, for exam­ple, Dying to Win: the Strate­gic Logic of Ter­ror­ism or “The Strate­gic Logic of Sui­cide Ter­ror­ism”, Amer­i­can Polit­i­cal Sci­ence Review, 97(3); August 2003, pg 13.↩︎

  25. See ↩︎

  26. Schelling. “What Pur­poses Can ‘Inter­na­tional Ter­ror­ism’ Serve?”, Vio­lence, Ter­ror­ism, and Jus­tice 1991↩︎

  27. “In almost none of the instances of ter­ror­ist activ­ity is there any gen­uine like­li­hood that the assault on per­son or prop­erty will serve to advance the claimed polit­i­cal ends.” from “The Polit­i­cal Sig­nifi­cance of Ter­ror­ism”, Vio­lence, Ter­ror­ism, and Jus­tice↩︎

  28. Abrahms ref­er­ences his analy­sis of a com­pre­hen­sive RAND dataset of global ter­ror­ism inci­dents, and also “Why Ter­ror­ists Don’t Claim Credit” (in Ter­ror­ism and Polit­i­cal Vio­lence, Vol 9 #1 1997).↩︎

  29. “Does Ter­ror­ism Really Work? Evo­lu­tion in the Con­ven­tional Wis­dom since 9/11”, Max Abrahms 2012:

    A the­o­ret­i­cal pos­si­bil­ity is that ter­ror­ists are sim­ply irra­tional or insane. Yet psy­cho­log­i­cal assess­ments (see Atran 2004; Berrebi 2009; Euben 2007; Hor­gan 2005; Mer­ari 2006; and Vic­to­roff 2005) of ter­ror­ists indi­cate that they are cog­ni­tively nor­mal. An alter­na­tive expla­na­tion with supe­rior empir­i­cal sup­port is that ter­ror­ists sim­ply over­es­ti­mate the coer­cive effec­tive­ness of their actions. By most defi­n­i­tions, ter­ror­ism is directed against civil­ian tar­gets, not mil­i­tary ones (Abrahms 2006; Ganor 2002; Good­win 2006; Hoff­man 2006; Schmid and Jong­man 2005).12 When bar­gain­ing the­o­rists point to cases of suc­cess­ful ter­ror­ist cam­paigns, how­ev­er, their exam­ples are usu­ally of guer­rilla cam­paigns, such as the U.S. and French with­drawals from Lebanon after the 1983 Hezbol­lah attacks on their mil­i­tary instal­la­tions. Inter­est­ing­ly, Osama bin Laden also ref­er­enced his­tor­i­cally suc­cess­ful guer­rilla cam­paigns as proof that ter­ror­ist cam­paigns would pre­vail. Con­tent analy­sis of bin Laden’s state­ments reveals that the 9/11 attacks were intended to emu­late three salient guer­rilla vic­to­ries in par­tic­u­lar: the afore­men­tioned U.S. and French with­drawals from Lebanon in the early 1980s, the Soviet with­drawal from Afghanistan in the late 1980s, and the U.S. with­drawal from Soma­lia in 1994, despite the fact that these cam­paigns were directed against mil­i­tary per­son­nel, not civil­ians. Hamas lead­ers make the same mis­take; they often cite the U.S. and French with­drawals from Lebanon as evi­dence that blow­ing up Egged buses in Jerusalem will like­wise force the Israelis to cave. Accord­ing to Wilkin­son (1986, X, 53, 85), inter­na­tional ter­ror­ism began in the late 1960s because emu­la­tors tried to repli­cate the polit­i­cal suc­cesses of the anti-colo­nial strug­gles.

    • Atran, Scott (2004) Trends in Sui­cide Ter­ror­ism: Sense and Non­sense. Paper pre­sented to World Fed­er­a­tion of Sci­en­tists Per­ma­nent Mon­i­tor­ing Panel on Ter­ror­ism, Erice, Sici­ly, August
    • Euben, Rox­anne L. (2007) “Review Sym­po­sium: Under­stand­ing Sui­cide Ter­ror”. Per­spec­tives on Pol­i­tics 5 118–140.
    • Hor­gan, John. (2005) “The Social and Psy­cho­log­i­cal Char­ac­ter­is­tics of Ter­ror­ism and Ter­ror­ists”. In Root Causes of Ter­ror­ism: Myths, Real­i­ties and Ways For­ward, edited by Tore Bjor­go. New York: Tay­lor and Fran­cis
    • Mer­ari, Ariel. (2006) “Psy­cho­log­i­cal Aspects of Sui­cide Ter­ror­ism”. In Psy­chol­ogy of Ter­ror­ism, edited by Bruce Bon­gar et al. New York: Oxford Uni­ver­sity Press
    • Vic­to­roff, Jeff. (2005) “The Mind of the Ter­ror­ist: A Review and Cri­tique of Psy­cho­log­i­cal Approaches”. Jour­nal of Con­flict Res­o­lu­tion 49 3–42
    • Ganor, Boaz. (2002) “Defin­ing ter­ror­ism: Is one man’s ter­ror­ist another man’s free­dom fight­er?” Police Prac­tice and Research: An Inter­na­tional Jour­nal 3 287–304
    • Good­win, Jeff. (2006) “A The­ory of Cat­e­gor­i­cal Ter­ror­ism”. Social Forces 84 2027–2046.
    • Schmid, Alex P. and Jong­man, Albert J. (2005) Polit­i­cal ter­ror­ism: A new guide to actors, authors, con­cepts, data bases, the­o­ries and lit­er­a­ture
    ↩︎
  30. Borum 2004:

    Psy­chol­o­gy, as a dis­ci­pline, has a long his­tory of (per­haps even a bias toward) look­ing first to explain deviant behav­iors as a func­tion of psy­chopathol­ogy (i.e., men­tal dis­ease, dis­or­der, or dys­func­tion) or mal­ad­justed per­son­al­ity syn­dromes. As Schmid and Jong­man (1988) not­ed, “The chief assump­tion under­ly­ing many psy­cho­log­i­cal ‘the­o­ries’…is that the ter­ror­ist in one way or the other not nor­mal and that the insights from psy­chol­ogy and psy­chi­a­try are ade­quate keys to under­stand­ing.” In real­i­ty, psy­chopathol­ogy has proven to be, at best, only a mod­est risk fac­tor for gen­eral vio­lence, and all but irrel­e­vant to under­stand­ing ter­ror­ism. In fact, “the idea of ter­ror­ism as the prod­uct of men­tal dis­or­der or psy­chopa­thy has been dis­cred­ited” (Cren­shaw, 1992).

    …Nev­er­the­less, the research that does exist is fairly con­sis­tent in find­ing that seri­ous psy­chopathol­ogy or men­tal ill­nesses among ter­ror­ists are rel­a­tively rare, and cer­tainly not a major fac­tor in under­stand­ing or pre­dict­ing ter­ror­ist behav­ior (Mc­Cauley, 2002 ; Sage­man, 2004)…In the opin­ion of Fried­land (1992), “as for empir­i­cal sup­port, to date there is no com­pelling evi­dence that ter­ror­ists are abnor­mal, insane, or match a unique per­son­al­ity type. In fact, there are some indi­ca­tions to the con­trary.” The two most sig­nifi­cant schol­arly reviews of the “men­tal dis­or­der” per­spec­tive on ter­ror­ism are that of Ray Cor­rado (198178) and Andrew Silke (1998). Although writ­ten nearly twenty years apart, both reached sim­i­lar con­clu­sions. Acknowl­edg­ing that some stud­ies have found psy­chopatho­log­i­cal dis­or­ders among some ter­ror­ists, Silke (1998), sum­ma­rized his review of the lit­er­a­ture with the fol­low­ing con­clu­sions: “The cri­tique finds that the find­ings sup­port­ing the pathol­ogy model are rare and gen­er­ally of poor qual­i­ty. In con­trast, the evi­dence sug­gest­ing ter­ror­ist nor­mal­ity is both more plen­ti­ful and of bet­ter qual­i­ty.” An even more recent review of the sci­en­tific and pro­fes­sional lit­er­a­ture by Ruby (2002) sim­i­larly “con­cludes that ter­ror­ist are not dys­func­tional or patho­log­i­cal; rather, it sug­gests that ter­ror­ism is basi­cally another form of polit­i­cally moti­vated vio­lence that is per­pe­trated by ratio­nal, lucid peo­ple who have valid motives.”

    …Is­raeli psy­chol­ogy pro­fes­sor Ariel Mer­ari is one of the few peo­ple in the world to have col­lected sys­tem­at­ic, empir­i­cal data on a sig­nifi­cant sam­ple of sui­cide bombers. He exam­ined the back­grounds of every mod­ern era (since 1983) sui­cide bomber in the Mid­dle East. Although he expected to find sui­ci­dal dynam­ics and men­tal pathol­o­gy, instead he found that “In the major­i­ty, you find none of the risk fac­tors nor­mally asso­ci­ated with sui­cide, such as mood dis­or­ders or schiz­o­phre­nia, sub­stance abuse or his­tory of attempted sui­cide .”

    …Nearly a decade lat­er, psy­chol­o­gist John Hor­gan (2003) again exam­ined the cumu­la­tive research evi­dence on the search for a ter­ror­ist per­son­al­i­ty, and con­cluded that “in the con­text of a sci­en­tific study of behav­iour (which implies at least a sense of rigour) such attempts to assert the pres­ence of a ter­ror­ist per­son­al­i­ty, or pro­file, are piti­ful.” This appears to be a con­clu­sion of con­sen­sus among most researchers who study ter­ror­ist behav­ior. “With a num­ber of excep­tions (e.g., Feuer 1969), most observers agree that although latent per­son­al­ity traits can cer­tainly con­tribute to the deci­sion to turn to vio­lence, there is no sin­gle set of psy­chic attrib­utes that explains ter­ror­ist behav­ior” (Mc­Cormick, 2003).

    ↩︎
  31. “MI5 report chal­lenges views on ter­ror­ism in Britain: Sophis­ti­cated analy­sis says there is no sin­gle path­way to vio­lent extrem­ism”:

    It con­cludes that it is not pos­si­ble to draw up a typ­i­cal pro­file of the “British ter­ror­ist” as most are “demo­graph­i­cally unre­mark­able” and sim­ply reflect the com­mu­ni­ties in which they live. The “restricted” MI5 report takes apart many of the com­mon stereo­types about those involved in British ter­ror­ism. They are mostly British nation­als, not ille­gal immi­grants and, far from being Islamist fun­da­men­tal­ists, most are reli­gious novices. Nor, the analy­sis says, are they “mad and bad”. Those over 30 are just as likely to have a wife and chil­dren as to be lon­ers with no ties, the research shows…­Far from being reli­gious zealots, a large num­ber of those involved in ter­ror­ism do not prac­tise their faith reg­u­lar­ly. Many lack reli­gious lit­er­acy and could actu­ally be regarded as reli­gious novices. Very few have been brought up in strongly reli­gious house­holds, and there is a higher than aver­age pro­por­tion of con­verts. Some are involved in drug-tak­ing, drink­ing alco­hol and vis­it­ing pros­ti­tutes. MI5 says there is evi­dence that a well-estab­lished reli­gious iden­tity actu­ally pro­tects against vio­lent rad­i­cal­i­sa­tion. The “mad and bad” the­ory to explain why peo­ple turn to ter­ror­ism does not stand up, with no more evi­dence of men­tal ill­ness or patho­log­i­cal per­son­al­ity traits found among British ter­ror­ists than is found in the gen­eral pop­u­la­tion. Far from being lone indi­vid­u­als with no ties, the major­ity of those over 30 have steady rela­tion­ships, and most have chil­dren. MI5 says this chal­lenges the idea that ter­ror­ists are young men dri­ven by sex­ual frus­tra­tion and lured to “mar­tyr­dom” by the promise of beau­ti­ful vir­gins wait­ing for them in par­adise. It is wrong to assume that some­one with a wife and chil­dren is less likely to com­mit acts of ter­ror­ism. Those involved in British ter­ror­ism are not unin­tel­li­gent or gullible, and nor are they more likely to be well-e­d­u­cat­ed; their edu­ca­tional achieve­ment ranges from total lack of qual­i­fi­ca­tions to degree-level edu­ca­tion. How­ev­er, they are almost all employed in low-grade jobs.

    ↩︎
  32. “Are Ter­ror­ists Men­tally Deranged?”, Ruby 2002:

    Specifi­cal­ly, any psy­chopathol­ogy demon­strated by ter­ror­ists at a higher rate than non­ter­ror­ists may be the effect of ter­ror­ist behav­ior, not its cause. In fact, the unique demands of a ter­ror­ist lifestyle are likely to engen­der the sub­se­quent devel­op­ment of psy­cho­log­i­cal idio­syn­crasies, which could then influ­ence the ter­ror­ist’s behav­ior. These idio­syn­crasies can become patho­log­i­cal, just as any intense and uncon­ven­tional lifestyle can lead to psy­cho­log­i­cal pecu­liar­i­ties. For instance, it is rea­son­able to assume that a ter­ror­ist will want to avoid detec­tion and appre­hen­sion as he/she goes about the plan­ning and exe­cu­tion of ter­ror­ist acts. This surely would lead to an increased level of aware­ness, in order to detect any sur­veil­lance. Such a height­ened level of aware­ness, depend­ing on how intense and chron­ic, could develop into notice­able sus­pi­cious­ness of oth­ers and a cer­tain level of rigid­ity of actions. The accom­pa­ny­ing thought processes and behav­iors could be described as para­noid, obses­sive, and com­pul­sive. More­over, if the ter­ror­ist main­tains a high level of inter­per­sonal cau­tion and sig­nifi­cantly reduces emo­tional and social con­nec­tion with oth­ers, sub­se­quent behav­iors and thought processes could meet the DSM-IV cri­te­ria for para­noid, obses­sive-com­pul­sive, or schizoid per­son­al­ity dis­or­ders.

    ↩︎
  33. This to be a major fac­tor behind US sup­port of the of Soma­lia; the inva­sion crushed the mod­er­ate which had been restor­ing order & jus­tice to the famously anar­chic coun­try. The results can be judged for one­self.↩︎

  34. “A Call to Jihad, Answered in Amer­ica”, The New York Times 2009 ():

    For a group of stu­dents who often met at the school, on the Uni­ver­sity of Min­nesota cam­pus, those words seemed espe­cially fit­ting. They had fled Soma­lia as small boys, escap­ing a cat­a­strophic civil war. They came of age as refugees in Min­neapolis, embrac­ing bas­ket­ball and the prom, hip-hop and the Mall of Amer­i­ca. By the time they reached col­lege, their dreams seemed within grasp: one planned to become a doc­tor; anoth­er, an entre­pre­neur. But last year, in a study room on the first floor of Carl­son, the men turned their ener­gies to a differ­ent enter­prise. “Why are we sit­ting around in Amer­i­ca, doing noth­ing for our peo­ple?” one of the men, Mohamoud Has­san, a skinny 23-year-old engi­neer­ing major, pressed his friends. In Novem­ber, Mr. Has­san and two other stu­dents dropped out of col­lege and left for Soma­lia, the home­land they barely knew. Word soon spread that they had joined the Shabaab, a mil­i­tant Islamist group aligned with Al Qaeda that is fight­ing to over­throw the frag­ile Somali gov­ern­ment.

    …For many of the men, the path to Soma­lia offered some­thing per­sonal as well—a sense of adven­ture, pur­pose and even renew­al. In the first wave of Soma­lis who left were men whose uprooted lives resem­bled those of immi­grants in Europe who have joined the jihad. They faced bar­ri­ers of race and class, reli­gion and lan­guage. Mr. Ahmed, the 26-year-old sui­cide bomber, strug­gled at com­mu­nity col­leges before drop­ping out. His friend Zakaria Maruf, 30, fell in with a vio­lent street gang and later stocked shelves at a Wal-Mart. Mr. Has­san, the engi­neer­ing stu­dent, was a ris­ing star in his col­lege com­mu­ni­ty…“Now they feel impor­tant,” said one friend, who remains in con­tact with the men and, like oth­ers, would only speak anony­mously because of the inves­ti­ga­tion.

    …At the root of the prob­lem was a “cri­sis of belong­ing,” said Mohamud Galony, a sci­ence tutor who was friends with Mr. Ahmed and is the uncle of another boy who left. Young Soma­lis had been raised to honor their fam­i­lies’ tribes, yet felt dis­con­nected from them. “They want to belong, but who do they belong to?” said Mr. Galony, 23. By 2004, Mr. Ahmed had found a new cir­cle of friends. These reli­gious young men, pegged as “born-a­gains” or “fundis,” set them­selves apart by their dress. Their trousers had gone from sag­ging to short, emu­lat­ing the Prophet Muham­mad, who was said to have kept his clothes from touch­ing the ground…The full dimen­sions of the recruit­ment effort also remain unclear. A close friend of sev­eral of the men described the process as “a chain of friend­ship” in which one group encour­aged the next. “They want to bring peo­ple they are close with because they need that famil­iar­i­ty,” the friend said. “They cre­ated their own lit­tle Amer­ica in Soma­lia.”

    ↩︎
  35. One did not play by one­self; one played it with oth­ers. And to a con­sid­er­able degree, one built and hacked on Space­war as much as one com­peted with other play­ers.↩︎

  36. One of the most cited essays in the lit­er­a­ture is “”.↩︎

  37. Com­pet­ing for the high­est level is actu­ally impos­si­ble in those MMORPGs which imple­ment a .↩︎