Terrurisnl unci Politico/ Violence, 20:415-433, 2008
Copyright t Taylor & Francis Group, LLC
trnal of
ISSN' 0954-6553 print/1556-l836 online
DOl 10.1080/09546550802073367
g:')" (see
mming.
Loring.
Mechanisms of Political Radicalization: Pathways
Toward Terrorism
CLARK McCAULEY AND SOPHIA MOSKALENKO
.:qucncy
58 8~
Psychology Department, Bryn Mawr College, Bryn Mawr,
Pennsylvania, USA
TrOnSt11
8. :20()6.
)tlicl' \)j'
r~:IJlI/J
J\.-'LL':-;~l'\..:
This article conceptualizes political radicalization as a dimension of increasing
extrernit y of beliefs, feelings, and behaviors in support 0.( intergroup conflict and vio­
lence A eros's individuals, groups, and J11aSS publics, twelve mechanisms of radicali­
zation are distinguished. For ten of these mechanisms, radiculization occurs in a
context of group identification and reaction to perceived threat to the ingroup.
The variety and strength of reactive mechanisms point to the need to understand
rudicalization-s-including the extremes 0.1 terrOriSJ11--0S emerging more [rom the
dvnamics of intergroup conflict than from the vicissitudes n.1' individual psychology
Keywords pyramid model, radicalization, terrorism
ln this article we describe mechanisms of radicalization relevant to understanding the
origins or terrorism. We must immediately acknowledge that the idea of mechunistn
IS somewhat different in different domains of social science, sometimes with conflict­
ing definitions used within a single discipline. 1,2 Here we use mechanism in the gen­
eral sense traditionally employed in Psychology: "the means or mariner 111 \\ hieh
something is accomplished. Thus, the mechanism of vision includes the physica.
stimulus and the physiological and neural processes involved ,,~
Clark iv1cCaule) IS Professor of Psychology and a director of the Solo-non Asch Center
for Study of Eihnopolitical Conflict at Bryn Mawr College. and a co-director or the ~Jtl\..)n(l]
Consortium for Stud) of Terrorism and Responses to Tcr rorism (NC-STi\RT). Sophia
ivloskaknku received her Ph.D. In SOCIal Psychology: from the Umvcrsu . uf Pl'IHb) h .uu.:
111 200-l She I" a pos.doctoral fellow at the Department of Homeland Sccurit . au.l ;t IC''lI..'dfL'h
fcllov at the National Consortium for Study of T er rorrsm and Resp(ln~,-'~ t\..) 'I crr oru.m
(NC-STA.RT)
I'll b research was supported by the Unncd States Department \..11' IL)lnelltnd \l'CUfl t~
through the National Consortium for the Study of Tcrr orrsm a nd Re~l'nn~c\ to l'l'IT\)rI:-.Ill
(STA RT ). g. ran t n u III be r N 00 1405 10629 H 0 VV cve r. any 0 pin Ion ~. ri nd in g:-.. d n<..1 (() 111...:1 Ll:-; I (I n"
or rccornmcudauorr- In this document are those of the author., and do nut l1ecc:-:<\nl,\ rc't'kl't
\1('\\:"> ott h« U,S l)eJ!tlrtlneIll of Homeland Securu ,
Ihe authors arc grutci ul
r.u , 1,!f'll'l-­
and Tony Marsella fur comments and suggestions in response to draft \·C'r:-']\..)!1", llf thi-. p~lpl'r
and thank three a nonyITIOUS reviewers for n umcrous helpful suggestion- .
.t\ddrc"s correspondence to Clark Mct..aule y SOl0I110n Asch Center fur St ud , l)!' Ft hno­
political Confl.ct. Bryn Mawr College, ]01 N Merro n Avc. Bryn \LHy r P;\ 19()]() l -rnail
CInCC~l uley((~: brynmawr.ed u
1\) (
4/5
-! Jt)
Funcuonally. political rudicahza tion is Increased preparation for and commit­
men t to Intergroup conflict Descnpn vely. r.idicnhzation means change in beliefs,
feelings, and bchuvrors In directions that increasingly justify intergroup violence
and demand sacnficc 111 defense of the ingroup. I-lo\v does this happen? How do
indrviduals. groups. and mass publics move toward contlict and violence'?
This question a pphes as well to sta te prepara tion for conflict as to non-state pre­
p.uauon for confhct A state and Its citizens are radicalized in the run-up to interstate
conflicts and war. and. as evident III the lJ .S. after the attacks of September 11, 200 1.
111 state response to tcrronsm as well. But con11110n drscourse about radicalizauon
focuses on non-state groups that represent a challenge or threat to the state. Similarly,
tcr rorism as a means of poliucal control is predorrunantly government work." but
CCHnIllC)Jl dISCO urse aSSOCICi tcs tcrr orrsm with the acnons of non-sta te groups. In this
pa per. we focus on the conl111011 usage in which radicahzation refers to increasing
l'\trcllllt) of non-state challenges to state authority. We aim to show. however.
j-h)\\ state action can contribute to radicahzation of non-state groups.
Becausc t
fee II n g ~. t hC t
the pyramid
the) are figh1
support 1'0 I' t 1
of the pyrun
Amencans \V
powerful that
citizens.·· 1 ..;
From ba­
n urn bel'S but
way of think
terror.st , fron
to the ex t rem
R aliicali;a tiou
Radicalization in the Pvramid Model
Individual and Mass Radicalit.ation
Ihere are many possible meanings of rudrcalizntion. but 1l10St of the relevant distmc­
t Il)llS can be represcn ted wi th the usual social psychological distinctions among
bchcl. feelIng. and behavior. Of course it is rndicalization of behavior that is of
greatest practical concern In a poliucai context this means increasing ume. money,
r isk-tak mg. and violence In support of a political cause. As every political cause is
assocratec wuh a particular group that cares about this cause. \ve may equally
,')(l~, that behavioral radicahza tion means increasmg time. money. risk-taking. and
violence III support of Ll political group.
If at a given point III time VVt' compare those who are more and less behaviorally
cornnuued. \VC are likely to find differences in both behefs and feelings. Social move­
men t activists arc Ii kely to share more than non-activists the beliefs or "frames " that
the movement uses to summarize and convey its mission. 5 Anti-poverty activists. for
Instance, tend to set' eli ffcrer: t ca uses of poverty than non-activists. 6 Radicalization of
many kinds may be associated WIth a syndrome of behefs about the current situauon
d nd its history: i We arc a special or chosen grou p (su pcriorit y) who hu ve been
unf.urly treated and betrayed (mjusticc). no one else cares about us or will help us
(dr-.trust ). and the SItuation IS dire our group and our cause are in danger of cxtinc­
non (vulnerability).
Similarly those who do I110rC arc 1ike ly to have different and stronger feellngs
l)
a ho u t the confhct tha n those who do less. (, Acuvixts are likely to feel 1110re sadness
and humihation with group Iuilurc-. 1110re joy and pride with group success. more
a nger and fear at the perfidy or VIolence of the enemies of their ca use
'These tcehngs are the cxprcs-ion of gruup rdcn t ification: caring about wha t ha p­
pens to the group. cspcciallv in rclut ions with other groups.]O Group Identification
can even lead to feelings of guilt about \vrongdoing perpetrated by others. if the
lHhcrs are mcn1bcrs of the group Identified \vith. 1 1 l~he hlunan capacity to care about
large and Inlpcrsonal collectI\'itlcs as If they \vere an extended fanlily is the founda­
tIon of mass politics. and the pre-requIsite for national. ethnic~ and relIgIOUS group
conflict l~, l.~
~
Lcononu-n., d
poinung out 1
money, (\ nd )
advancing th.
borne by the
Ll group CdU\\:
an) a(1\ancc
T'hc CL1SS
sornc kind or
ILl \\
0
r g (1 \
\.'
ITl
vid u;l! n« .r..l:
gro Llp no 1'111:-' )
can 1Jt' difTil'\.!
kind of gruup
cacl:
111l'1l1bcr
SOCIa! puru..ln
all When grn
a larger 11111111
-rhll~ r~tdl
xma 11 g.n)Ll ps.
al« ay-. l'hc ~I
glll~l! /\] ()~le(
TIl 0 S t I ~
\. )J)
1hl
born hers \\ l'rL
III .1\1
()acd~\
Radicaliz::ltiu
/\s I nJ ll'(l ll'd 1
Indl\'iduab ~lt
~b cOll\eYLd r
Mcrhanistn,
C0111 muin beheis,
p VIolence
? How do
j
)
-state pre­
) interstate
r 11,200 L
icalization
Similarlv,
vork," b:lt
ps. In Lhb
increasmg
however.
(JI
R(I(li('LII!~(I[il)11
4ji
Bcca Ll set err U rl;) 1s arc t'e\\ 111 rc 1a tIC) n t o all tho Sl' \\.11 u ~ h d ret h ell' bel iefs and
feel.nus. the terronsts ma , be thought of as the apc.. of ~I pvr.unid )..+ 'The base of
the pvrunud is COll1 posed of ttl I \\'110 sym pa t l:IlC \\'it It the gu~l!~ the terrorist; say
t h C) ( Ire fi g h tin g Io 1'. InN 0 r the r n Ire 1and. 1'0 r Ins tan cc. the h l.I~ cur t h c IIYra n11 d 0 f
~ upport for the IRA was all those who agreed .. Br:ts OLI t '. In the U.S .. t he base
of the pvr.umd of support Cor anu-govcrnrnent acuon I~ the flIrt) percent or
Amcrtca ns who agree that "The federal governmen t has become xo large and
powerful that It poses an immediate threat to the rluhts and lrccdomx of ord.narv
.,j"
~
.;
cu izcns.
~. 1'0111 base to apex. lughcr levels or the pyr.nrud arc assocIated wu h decreased
number- but increased radicahzauon or beliefs, feellng:~. and beha , II.' r"-l Thus one
\\'a) o f t hInk 1ngab0 U t rad 1c a 117 a tI U 11 I~ t 11 a tit b t hl' g l' ad 1C n t t h ~ ltd 1St 111 g u1she s
1crrun~t~ from then base or sympatlu/crx. How do mdrv idua ls mov c Irorn the base
to the extremes of tcrronst violence at the apex'?
Radicalization in G'roups
nt distinc­
ns among
that is of
e, money.
d cause is
ly equally
king, and
haviorally
:ial 111ove­
nles" that
tivists. for
lization of
, situation
1ave been
III help us
of extinc­
~r
feelIngs
-e sadness
:ess. 1110rc
vvhat hap­
lconomr-t-, and political scicntist« uSing a ra tiona l-chouc lr.unc« or k arc fond 01
p U I11 tI n gUll t t11 at in d IVId ua b s h o uld be re lu ctel nttl) L' U n1n11 t re ~ 11 r l':--'0 II rce S 0 f t rmc ,
n10nc y. a 11 d ris k - ta kin g to a d van c C t h C c (I usc 0 r ~ I LIr gl' g r 0 Ll P T h l' b c 11 C fi1S 0 f
"lhancing the group are available 10 all group member». \\ hcrcu : the costs arc
borne h) the acti \ IstS. Th us the ru tiona 1 choice for (111 uid I\ lei uul \\ ho cares abo U 1
a group cause 1~ tl) do nothing. let other individuals P~l~ thl' l'llsl"-l. (lIH1 ncuefit froll1
till) advance for the group as (I Ircc-ridcr.!"
'The' cL.iSSIC answer to the pro blern of n10bIlI71 ng ind I\'Id u.: b 1'1.) r ~()CI~d action I"
some k ind of coercion. that is. punishment for frcc-ridlng. ('~k't-l'](111 nl~l) C0I11e fn)111
la« or gu\crn111enl regulation (if Iree-rrdcr-, l'(111 be dl"l'L1r(ltl~l) Ilk'[l1iIICdL from l11dI­
\ Id ua 1 Ino r ali 1) (111 tern a I nor 111S). 0 r Ir U1u i n Co r 111 (tl r~ IL'l' - t u - LIl'i..' "-I (Inc'tio n ~ (sn 1~ III
gruup norms). Particularly 111 a sma ll grlHIl'. pt'r~Ji.)}l(tl I1lura]il) dIHl group l1orn1~
can be d i1'fi cult to Scpa rat e ~ bec a II ~ c in d i\ id II (I] Jn n r:tI i t ~ i ~ us Ll (Ill) a n l~ h u rc d 111 son 1C
kind of group COllsensus. And in ~lnall LIce-to-Ltc\..' ~l'l)UP \\ Iil"Il' l~dl'h 111c111ber and
each n1cn1ber·s behavior i~ knO\\1l to 01hl'r~. SOCI~tl rl'\\ard~ for partll'lpatIun and
SOCIal puni:--,hn1cnts for fret'-riding Ct111 n1ake hch:t\ ior~tl CUJnnlltnJl'nt ra110nal aner
all. \Vhcn g.roups can be linked through conln10n JllC1l1blT"-I or l\Jlll111l1n lcaJer~ Intl)
1 II ,,~ I bk' U 11 a 1a r g c l' SC (11 c .
(-I Ia r g t' r n 1 lilt 1- g r 0 up 0 r g a 111 z a t i 0 11. SOC i(tI , l C t I 0 11 bee 0 1'nC s 1
-r h Ll ~ r a die (l 11 L a 11 0 11 and t e IT l)r is111 arc 111 d d e pl)~ s ih L: h ~ b ri 11 g 111 ~ III d 1\ 1d II a bin t0
sIn a 11 g r I.HI P~, Son 1et i III est 11 esc g rOllp ~ ;II\' 1I n k l'd 1n 1() a LIr g L' r u r ~ dIll/ d 11011. b II t 11 () 1
dh\'dYS. -rhL' sl11all gn)llp is ncceSS:ll") rlH' dl'tlun. hut 1hl' l)rt2:1111/<llil)/1 IS J1l)t fhe 01'1­
gin a 1 l\ I Qa cda \\ a san 0 rga 11 1Z(I tion 0 f g r n Ll Jl~' 0 I" l'l'11:--, . b 1I ttl) cLI~ th\_' f) r 0 up,",> a r l'
n1()~ 11 Y 011 thel r ()\\ nand disco 11nect cd fro In (l n) L~ rgl'r ()rgal11/~ltll)11 l'he 1\/1 ad nd
h01l1 bel'S \\'l're ~I pparell tly 1110 rc a scIr-·o rgd 111/1 ng ~ll1d II f2ru uI' 1 ha 11 ([ l,:e 11 c 111 hed ded
111 Al ()aclLI
,I
1t i fica t ion
:[s, if the
are about
e founda­
)US group
l~adicalization
of Individuals., (;roups., and
l\;las~es
i\S ll1(jlcated 111 the rrecelhllg diSCUSSIon. r(ldlC(llIz~111l)11 Cdl1 UCl'ur at different k'veb.
I n d IVi d ua ls are r (I die ali ze d by pel's 0 n (11 g rI l'V (1 nee S (I 11 ci b~, ]lll' 11 tit Y- p- n Hlp gn ev(I ncc S
as conveyed by nlass nlcdia, runlOL or the testI1l1()n~ l)r utheJ':1, 11ld1\ Idllal~ (lrl~ abl)
'Table 1. Pathways to violence: Mechanism, of political radicalization at individual.
group, and mass-public levelLevel of radicalization
that pers
is frame;
Mechanism
1. Personal victimiza tion
Indivrdua
Group
2. Political grievance
3. Joining a radical gro up-ithc slippery slope
4. Joining a radical groupv the power of love
5. Extremity shift in like-minded groups
6. Extreme cohesion under isolation and threat
7. C 0 In petit ion for the same bas e 0 f sup port
8. Competition with state power--condensation
l\!Iass
9. Within-group cornpctitionvIissioning
10. Jujitsu politics
11. Hate
12. Martyrdom
rudicahzed as mern bel'S uf small face-to-face grou ps Poliucal groups and ITIaSS
publics are rudicahzed In confhct with states and with other political groups. Each
or these levels requires Separate attention Table 1 identifies the twelve mechanisms
Lit three levels that \VC wil] now describe.
I. Individual Radicalization by Personal Victimization
Tlux IS a path much CIted In explanations of suicide terrorists. Chechen BLJck
Widows are described as seeking revenge against Russians for their own experience
of rape or for the deaths of their menfolk. 17 Tamil Tigers of the suicide brigades
called "Black Tigers" are often described as survivors of Sinhalese atrocities.
Accounts of Palestinian suicide terrorists often CIte revenge for IDF attacks on
neighbors or loved ones as a motive for self-sacrifice.
The importance of personal grievance as a motive for terrorism goes back a t least
as fa r as R uss1a 11 le 1'1'0 rists 0 f the 1ate 18OOs. Th us And rc 1 Zh elya bo v, a lead er 0 f ter­
rorist organization People's Will and a mastermind of a n urnber of political assassina­
nons, including the coordinated bombs that killed Czar Alexander II, sought out
terrorist activity in a pledge to revenge the many wrong» by the monarchist regime
he experienced first hand. 10 The rape of his fa vori tc a un t by their landrnastcr. Ignored
by local pohce: his dismissal from universi ty wi thout righ t to reapply for participa ting
In an innocent protest ag.unst arbitrary grading practices: and finally, a four-month
j.ul sentence for sending a friendly note to an imprisoned friendv-thcsc grievances
shaped and hardened Zhclyabovs resolve to usc violence against the ruling elite.
Data are hard to come hy on how many terrorists, or how many suicide terror­
ist», have a personal hist ory of victimization that miuh t explain their sacrifice or
course there may be individuals with such a history who nevertheless would not have
moved to violence without seeing their victimization joined to the victimization of
their ethnic or national group. That is, the percentage with a history of personal
victimization is an upper bound of the power of a personal-revenge explanation.
rather than a reliable cst imate of this power. J\ social psychological view would be
Sornenrn
response
example.
cabin to
feared a:
Ano
1999 aft:
postman
alone in
SiI1l1
people 11
2002. Ivl
extort te
Muham
titled \VI
Cast
individu
the inch
Ill e 11 t---- <:1
with wh
of Islam
Mo
degree c
suffered
al ty fur
disorde:
to recru
radicals
An
gncvan(
Hale gr
but lost
In 2005
Lcfko»
impo-s:
3. huliv
J\S
Just
dcnly u
Into a
trust~d
asked t.
20. r =
Mechanisrns oj' Rudicalizatio.:
.lividual,
41c)
that personal grievance IS unlikely to account for group sacrifice unless the personal
framed and interpreted as representative of group grievance.
b
2. Individual Radicalization by Political Grievance
slope
if love
:i threa t
rport
ensation
id mass
)s. Each
hanisms
n Black
penence
»rigades
roci ties.
ack s on
at least
or of ter­
sassma­
.ght out
regime
ignored
cipating
<month
evanccs
elite.
~ terror­
fice. or
lot have
rtion of
iersonal
mation,
ould be
S0111etlITIeS an individ ual IS moved to individ ual radical acuon and VIolence in
response to political trends or events. Ted Kaczynski, the Unabomber. is one
example Over eighteen years, Kaczynski emerged occasionally frOITI his wilderness
cabin to send letter bombs to people representing the technological progress he
feared and detested.
Another example IS Buford Furrow, who turned himself In to police in August
1999 after wounding five at a Jewish Community Center and later killing a Filipino
postman He seems to have been a devotee of white supremacist groups but acted
alone In planning and carrying out these attacks.
Similarly. John Allen Muhammad, with his protege Lee Boyd rv1alvo, killed ten
people in the Washmgton area in 47 days of sniper attacks in September and October
2002. M uh cUD 111 ad, a con ver t to] sIaman d blac k sepa rat iSIll. wa sat t em p tm g to
extort ten million dollars with \VhICh to found a pure black comrnunuy in Canada.:"
Muhurr mad has not been forthcoming about his motivation. but It appears he Iden­
tified WIth what he perceived to be the victimization of black people in the U.S.
Cases of individual radicalization to political violence, that is, cases in which the
Individual acts alone rather than as part of a group, are relatively rare. I n such cases.
the Individual is likely to have some association with a larger intellectual InOY'e­
mcnt ---as Kaczynski related to a larger movement of survivalists. as FUITO\V associated
with white supremacists, and as Muhammad participated for a penod in the Nat ion
of Islam.
1\;1 0 re than inany 0 the r ca tego r y 0 f ra d icali zat ion. the re i saprob a bility 0 f so me
degree of psychopathology. Psychiatric testimony at hIS trial indicted that Kaczynsk:
suffered from paranoid schizophrenia. The prosecution did not seek the death pen­
alty for Furrow because he had a history of inpatient hospital treatrnent for mental
disorder. Groups of radicals. especially those who get as far as ter rorrsm. are unlikcl ,
to recruit or tolerate the unreliability that goes with psychopathology. Individualist
radicals can be responding. at least in part, to their private demons.
An interesting example of the difficulty of separatIng personal and group
gnevance is Matt Hale, who in 1998 was the leader of a white supremacist group.
I-fa1e graduated from law school, passed the bar exam. was hired by a lav firm.
but lost his job when the Illinois Bar dented his law license on the basis or racism
In 2005, Hale was sentenced to prison for soliciting the murder of federal Judge Joan
Lcfk ow 'The personal and political are so closely intertw.ncd In this case tha t It h
impossible to say what Hale would have done had he been granted hIS la\\ license
3. Individual Radicalization ill Joining a Radical Group- The
L~"'lippery
Slope
As Just noted, it IS rare that an individual moves from syrnpat luzcr to activist by sud­
denly undertaking some major risk or sacrifice. Typically an mdividuals progress
Into a terrorist group is slow and gradual, with many smaller tests before being
trusted in more important missions. and with many non-violent tasks before being
asked to use gun or bomb (for Red Army Faction and Basque El-A recruus, see note
20. p. 237: for IRA recruits see note 21),
c
j\·/c( 'uu/c'~ and S. Mosk alenl:o
Of course there arc occasior:a. examples of an individual moving from sympathy
to extreme violence In a single grant step. Wafa Idriss, the first female Palestinian
suictdc bOI11beL seems to ha ve carried out her mission within two weeks of deciding
to becon 1cas uiC1deb 0 n1 be r. A1tho ugh 1tis bey 0 n d 1h e sc 0 pe 0 f the prese nt pape r to
exanune tlus question systernaucally. we believe that example, of giant-step transi­
tion to VIolence are notable precisely because they are relatively uncommon.
A VIVId example of gradual radicalization comes from Della Porta." who quotes an
l tahan militant as follows: ..A choice [made] in cold blood, such as "now 1 will become a
terrorist: [did] not exist It \\ias a step- by-step evoluuon, which passed through a kind of
hUI11an relation that I had wit h Guido, and with the people I worked with.'
The power of step-by-step self-persuasion through one's own behavior is well
studrcd In SOCIal psychology. Hundreds of experiments have shown a strong ten­
dency for self-: usti fica uon after an individ ual does something stupid or sleazy.
/\n iudividua. who IS sucked Into saymg a dull experiment is fun, or into writing
.i n essay In favor of' a cause the mdivrdual disagrees with, is likely to find reasons
t tJ ] ustif, the beha VIOl" the experiment wasn't half bad, keeping Communists 1'1'0111
spcakmg on campus IS a good idea. Dissonance theory:" understands this tendency
JS an effort to reduct' the mconsrstency between positive self-Image and bad beha­
\ lUI' In other words, It IS easter to find reasons for what we do than to do what we
rind reason for.
Perhaps the rnust stnking example of the power of self-radicalizauon is found In
one of the exper imen tal varia uons in trod uced by Milgram" in his famous studies of
obedience. In the baSIC paradigm. normal individuals who draw the role of teacher in
~l psychology experiment \\'111 give high levels of shock to a protesting "victim (actu­
ally an accomplice of the experuncnter ) who drew the role of learner. Complete obe­
drencc req uircs the teacher to r.usc the shock administered for rrustak es from 15 to
.+50 volts in 15-volt incrcmcn ts. A bout 601~/1) of teachers are completely obedient.
Less well known is the varration in which it is not the experimenter who comes
up with the idea of raising the shuck level with each mistake. In this va rratron. a "co­
teacher" (another accomplice of the experimenter) asks and grades the questions,
while the naive teacher gi\'t:s the shocks. The experimenter, summoned away for a
"phone call.' IS no longer in the room when the "co-teacher" C0111es up wuh the idea
of raising the shock level wn h each mistake. Despite the absence of the experimenter
and hIS authorny. 20 1; '1) or teachers progress to admimstering 450 volts.
T h e dis son an ce ex p Ia nat ion 0 f the 2O' I~) \vhog 0 all the \N a y is t hat e a ch shoc k
becomes a reason to give the next shock. The closely graded shock levels represent
iJ kind of slippery slope In which refusing to give the next shock requires recognizuig
that there was sometlung wrong wn h giving the last shock. If 30n volts was o k. hov,
can .~ 15 volts be wrong? But If ] 15 volts is wrong, how can 300 volts be right'?
In Milgrams studies, the dependent variable is radicalization In behavior, not In
thoughts or feelings. The latter were not measured. and there is no way of kno\vJng
whether increasing shock levels were assoc.aicd with changes in perception of and
(1t1it ude to \V a r d the victi111
25
In another famous study, LInlbardo was able to demonstrate radicahzauon In
behaVIor of one group of pclrticipant::; (playing roles of prison guards) to\vard
another group (playIng the role of prisoners). PsychologIcally stable nlale student
volunteers \verc ranclon1ly assigned to act as either a guard or a prisoner in a simu­
lated prison enVIronlncnt. Left to their O\VI1 deVIces, over the course of Just a fev\
days. the guards gradudlly escalated their abuse (111 the form of hUIlliliation and
arbitrary pL
to tcrmin.u.
As 111 f\
PrIson Exp\
ing trea trnc
to mak mg
suggestiv« r
his increasi­
to how 11'1 u
fact that hi
for thernscl
1
1~here 1
harms othc
In drssona n
power of
SI
sh ppery sin
and JUS t i fic
'ThIS is the I
about terro
uons wu h l
betr.iy the
net \\ ork 01
'Trust I
but love ~)I
c.m he (IS ~i
about hh n
(BR) mad.:
neal siiu.u i
for the pel
Gcrrn.:
undcrgroui
vr:.. thu: "1l1{
1Il)n~ throu
polillcal
11'
11 urnbel' 01
l)l'\ ~)l
togcl her
BR and ll~
the \'otL' Ll
l\ ftcr
j
grlHlp h 11
group (01
]}zrnlf) 11
lhell ~ruur
Cel U~l.'. \\'e r
I~ t' CIi, IJ1i.\1)1)
[
sympathy
Palestinian
of deciding
~nt paper to
-step tra nsi­
01
lTIOn.
quotes an
fill become a
19h a kind of
10
1. "
rvior is well
strong teo­
1 or sleazy.
into writing
'ind reasons
urns ts from
lis tendency
i bad bcha­ do what \VC
is found In
IS studies of
if teacher in
.tim (actu­
mplete obefrom 15 to
.ibedicm
who comes
tion, a "co­
~ questions.
away for a
lith the idea
xperimenter
t
each shock
Is represen t
recognIzIng
~,7as ole ho\v
~ fIght?
Lvior. not in
of knov\'Ing
tlon of and
:alization in
'ds) toward
tale student
r In a Slmu­
C just a fe\v
,liation and
o] RLI.l i('{/ I i~ ([ t i() n
4:.!1
arbu ra ry punishment) toward the pnsoners to the: degree that Zrmburdo was forced
to terminate t hc experiment
.As in Milgra ms experiment. tcchng« and bchcf-, were not measured durmg the
Prison Lxpcrirncnt. But there IS an obvious progressron toward I110rc' dchumaruz­
Ing tre.u mcnt of the prisoners. starung nom muk iug them do push-ups. moving on
ton 1a kin g the n1 eat fiIth y fo 0 d. an cl end i ng \\' It h fu rc 1ng the m to act 0 ut sex u a 11 J
suggestive plays In post-experiment interviews. one of the "guards" suggested that
hIS increasingly cruel treatment of the "prisoners" \\'ClS the result or hIS curtosity as
to how much the "prisoners" would let him get (1\\ ~l) \\ .th I-ur thl~ guard. the
fact that he \\'cn t too fa r 1Sthe fa uIto f the ' .pn son er ~ ,. \vhod Id not stan d up
for themse Ivcs.
There IS then a pattern of slowly mcreasmg radical behavior behavior that
harms others-- In both the Milgranl experiments and Zrmbardos prison experiment.
In dissonance experiments and in Zimbardos poxt-cx pcrrrncnta. inquirv. \Ve ~ee the
power 0 l self-persuasion In j ustilying one's own be ha \- ior Sell-rad ica li7a tion IS a
slippery slope or increasmgly extreme behaviors. wuh mcrcasmul , extreme reasons
and J list ifica nons icing the slope.
4. Individual Radicalization in Joining a Radical
(;r(}lIp-~ 1'11(:
Power of'Love
ThIS b the path to radicahzation that has received most ,lttCIlthJIl In rcccn t .hcorizing
about tcrrorism.r" Individuals are recruited to a tc r rorrst group \ lei personal connec­
nons with existint; terrorists. No terrorist wants to try to recruit someone who might
betray the terrorists to the authorities. In pracucc. tln- means rcciuiuns; lr om the
network of friends. lovers, and family.
Trust may dcterrnmc the network wu hm which radicals and terrOrIsts recrun.
but love often detcrmrncs who will JOIn. The pull or rornauuc and comrudcl , lovl'
can be ass t ron gas pol it ic ~ in m 0 vingin d i vId Ll a Is nit 0 an II ndel' g ro Ll n d grou p .4 sked
a bo ut his III 0 tivat ion s fo r go ing un derg r 0 un d. a m CIII bl' r 0 f the Ituh a n 13 n gat c !Z 0 sse
(BR) made t lus reply: "There are many thIngs I cannot cxpl.un by al1~tlyzlng the polr­
tical situation, .. as far a~ I am concerned it \VC:lS up to cmouon.il feelinE!-~. of p(l~sions
for the people I shared my life wuh. ,-27
German militants of the Red Ann) Fracuon (H./\F) \\LTl' abl) dra\\ 11 111l() the
underground by devotion to friends. "There IS \VldcSI,rc,ld agret-rllcllt among rc-carch­
ers that 'most terrorists . . ulumately became members of [German] tcrror rst org.uuza­
tIons through personal connectlons \vith people or rl'ILltl\'l's LlSS()CI~lt~d \\'lth ctppn,priate
politIcal initIativcs. con1nlunes, self-supportIng org:lnIL;ltioll~,. or CUl11111iUces the
IlllI11ber of cuupk'~ and brothers alld sIsters \V(IS astonlshlngl) hIgh .. _~~ ..~'l
DevotIon to conlrades can lead a clIque 01' rrIl'lld~ to Join ;\ tl~ITl)n~t group
toget her f\ccordlng to della Porta,J() "block rccrLlltln\~I)t" uccurred but 11 ror the
SR. and the RAF. Sonletlnles a sn1all poli tIcal groll p \-\ lHIlLi hold :l lllectlng and If
the v 0 tc fa v 0 re d j 0 i11i 11 g t h(' un de l' g r 0 It nd. a II \\'0 U Id J()111 t l'get h l' r
/\rtcr an Individual joins a radical grOllI'. 10\ e l'or frIeIllb and cornradcs 111 the
group is likely to increase further a~ conlnl0ll ~u;lls and l'0111111011 threat~ Illcrease
group cohesion·~l (sec also Section 6 C;,.OllP R{/di('(/Ii~{[!l(}Jl under !SU!O!lOll ond
]'!?rcof). Intervie\\'s \vilh 30 long-terlll n1clnber~ or SlllIl I'elll led \Vhlte to cOllclLldc~~-""
that group solidarity, along \vith hope of nlaking d difference i"lH' the group and Its
cau se, vve ret he l\v 0 st I' 0 ngest fo rccs hoIdIn g n1II iLlnt" togcthe r III the faceo r a rrest ~
,r
1 .,
/\fl'( (U{,'<l
1 __ ......
anti S, A/oskulcllko
d
nd Loyal istat tuc k ~ , T h u ~ de v0 u 0 n to CO m ra des 1S not 0 n 1y a fo rCe for J 01 nin g a
ru.hcul group, It h equ~tlly or more Ll bar ncr to leaving the group,
vVhilt"-'" quotes one Rcpubhc.u: Lb follows. "Thcrcs times I've said to mysclt. 'Why?
You're mad In the head, like,' But
I Just can't turn my back on it ... there's too many
ol my friends In j.ul. thcrcs too nlany of n1Y mates giver: their lives. and I've walked
behind lvc walkc.i bclund too m.iny funerals to turn my back on it now,"
change the S'
research sho.
ar g umen t S C(1
rather than r,
in a grou p of
5. Group Radicalization ill Like-Minded C;roups
6. Group Rau
I here 1~ a 11 ex perur.cn utl mode 1 of gro II p rad rcaliza non tha t has been referred to
\ dnou~IJ as "risk . ~hlrt." ·'gJ"l.)UP extremity sluft.' or "group polarization," Groups
ot strLlnger~ brought tugether to dISCUSS issues of risk taking or political opinion
~h()\\ cou-rsteru
t\\'()
or change: Increased agreement about the opinion at
l~~lle. dnd (l ~hll'l In the average opinion or group members. The shift is toward
1l1lTelt~Cd ext 1''-'11111) on \VhICh('\ er side of the opinion is favored by I110St individ uals
hc'fore dl~CllS~Iun ~-+ If m.» t md 1\'ld ua ls favor risk before discussion, the sh: It IS
ll)\\ard increased rhk takIng. If 111l)St individuals oppose American foreign .ud before
.hscus-non. the ~hI!'t b to\\ .ud increased opposition to foreign aid,
'The xhilt IS not ju-t Ll m.n ter or go-along-to-get-along compliance: each group
J nc" 11 1he r g 1\ e ~ h l)t 11 pre -ell~ l' U ~ ~ 1l. )nan d poSt-d1
~c ussion 0 pin 1011 0 n a que stiun 11a: r t'
that only the researcher scc-. Thus drscussion among individuals with similar values
produces nucrnuhzco shrft toward more extreme opinions.
There arc current l, t\VO cxpla nauons or group extremity shift.3~ According to
rclivunt orguJllellr\ thcor, , ~l cult ura lly determined pool of arguments favors one SIde
of the Issue 1110rC than t hc other SIde. An individual samples from this pool In asses­
~lng: lu-: or her mdrvidua l opmron. then in discussion hears new arguments 1'1'0111
others, \Vh1c11, COInIng Ir orn the sa nle pool, are mostly in the same di rection ~l s the
indr, idual \\ as lcanlng The result IS that individuals are rationally persuaded by
the imbalance or 11('\\ argun1ents heard In dISCUSSIon.
Accordmg to soc/of comparison theorv, opmion positions have SOCIal vulucs
~ t tt ac h cd tot hen 1 A11 1nd I vi d Ll d 1~ fee I pre ssur c to \\' u rd a g.reC1nen 1. 1hat is, pressurl'
tn nlove their opillil)n~ to\\ard the I11ean opinion of the group. But the pressure is
11l)t un1Corn1 I nellv1d uab Inure cx t rCllle than average in the group-fa \'ored elircc­
tle)n ·the ell rect Iun Ll\ ureJ by nlo~t ind1vld uals before discussion ----are 111Ure
adrn1rcd -,() -rhey Llr~ ~cell a~ 111or~ devoted to the group, lllore able--in sunl, as better
people. 'Thl~ C:\tLt statlls trLlnsLltc~ 11110 11lorc Influence and less change during group
dISCUSSIon, \\'herCds Il1dl\lduLtls less extren1e than average In the gruup-favored dirc'c­
tlon h(l\'l' less Intlul'11cL' ~lIHl chdngl' 1nOiT. No one \\'ants to be belo\\'<tverage III ~up­
1'1l.)rt l)l' the grollp-Ll\'OreL! opInion, and the result I~ that the average opInIon
hl'CC)nll'S Inore e.\tre111l' III the group-favored direction.
f\ \ivHJ d('~l:nptll.lJl ()j the pl)\VCr of SOCIal conlparison In rad}(:alI/ln~ the
\Veather L
d ll.S Llntl-\\,ar group of the 1970s. is provided by ('ollier
(\: Hor~)\vitL." \ViLhll1-group con1pctitlon for the status of being "1110st radIcal"
1l10\'cd the group to tc'rrUrISJl1 ~rhe hLdlnlark of this kind of radicaliL'ation l~ the
cxtent to \VhI(h the pcrsona} beconlcs politiCIzed' every act is judged by political
~tandards. Including \vho ~leeps \vlth 'vvhon1
Both relevant argunll'l1t~ and sOl'lal conlparisol1 explanatIons are necessary to
explaIn the pattern or expcrln1ental re~11lts,3o In support of relevant argunlcnts.
resL'arch sho\\':-\ that 111an1pulatll1g drgurnents \vithout kno\vledge of posltioll~ can
The mode! f(
small cornba t
in the same 1
whose mernb
on one anot
produces ext]
closer than 1
sacri ficed the
nade to save
Very hIg
merit of gr oi:
of attraction
reality create
mern bership.
security. Le~
tions of vali
good and \\ 1
mean that l L
come frorn (\
'Thus hl t
for mter nali.
pull togethcl
ancc - go-all
the social re
value. incl LIe
Groups
group IS \ve,
standards ()!
n1el1l bers art
pen\ erful fell
refonll or hI
\vorld has c(
the socla1 rc:
and nlOra Ii t
reqUIre \-loh
Th IS JOl
tar) traillin~
states. One 1
gro up goes
side threat 1
l.
1ndcrgruund,
k
irids
1
!v!echanisms
yself, '\Vh1")
·'s too many
I've walked
(~(
Radicalr:a t ion
423
chan get hc s 1Z e and d 1rectIon 0 f the gr 0 ups hift. Ins up p 0 rt 0 f soc 1a 1 co 111 parison.
research shows that knowledge of others opinions without knowledge of others'
argurncnts can yet produce group shift. 'The two explanations are complcmentarv
rather than redundant. Both conduce to increased similarity and increased extremity
In (t group of like-minded individ uals.
6. Group Radicalization Linder Isolation and Threat
referred
to
n." Groups
cal opinion
opinion at
is toward
individual­ the shrft b
1 aid before
t
each gruup
restionna: rc
nilar values
2cording to
)rs one side
)01 in asses-­
nents from
.tion as the
rsuaded by
iciai val uc-:
IS, pressure
pressure IS
ored direc­
-are nlore
11, as better
inng group
iored direc­
"age in s.ip­
ge opInIon
alizing the
by Colhcr
1st radical"
1110n is the
oy poliuca 1
ecessary to
arguments.
srtions CCll1
.i hc model for this kind of radicalization is the powerful cohesion that develops in
small combat groups. Soldiers In combat are largely cut off from all but their buddies
In the sa me platoon or squad. This Isolation is characteristic also of terrorist cells.
whose members can trust only one another. As both soldiers and terrorists depend
on one another for their lives In fighting the enemy. extreme Interdependence
prod uces extreme group cohesion. This is a cohesion that can make group members
closer than brothers. Recipients of the U.S. Medal of Honor Include 111(111y who
sacrificed themselves to save others: some literally threw themselves 011 top of a gre­
nade to sa ve t heir buddies. ~C)
Very high levels of cohesion in a group mean very strong pressures for agree­
ment of group members. Group dynamics theory disunguishes between t\VO sources
of attraction to a group: the value of 111 atenal group goals and the value of the social
reality created by the group. Material goals include the obvious rewards of group
membership. such as progress toward common goals, congeniality. status. and
security. Less obvious is the social reality value of the group: there are many ques­
tions of value for which the only source of certainty is group consensus. What is
good and what is evil? What is worth work mg for. worth dying for'? What does It
mean that I am going to die'? Certainty about these crucial h U111an questions can only
come from agreement with others.
Thus high cohesion bnngs high pressures for both behavioral cornph.mce and
for Internalized value consensus. It is obvious to group members that they have to
pull together in order to reach group goals, and the result In many cases is compli­
ance-> go-along-to-get-along agreement that does not bring in terior certainty. But
the social reality value of the group depends on internalizing group standards of
val Il C, including 1110ral standards.
Groups differ in their power to set moral standards, The sociai reality value of a
group is weak to the extent that members belong to other groups wit h competing
standards of value. Conversely, the social reality val uc of a group is strong when
members are cut off from other groups. ThIS principle IS the foundation of man ,
powerful forms of group-focused persuasion, including cult rccruu mg and thought
reform or brainwashing. When cohesion is very high. as when an mdividuals social
world has contracted to just the few friends In his combat group or his terrorist cell.
the social reality value of the group is maximized. The group's consensus a bout val ue
and morality acquires enormous power, including the power to Justify and even
require violence against those who threaten the group.
This joining of cause and comrades In a high cohesion group IS the goal of nlI1I­
tary training in every state, and is equally the foundauon of terrorist violence against
states. One practical implication IS that something important happens when a radical
group goes underground as a terrorist group, The combination of isolation and out­
side threat makes group dynamics immediately more powerful In the underground
cell than In the radical group that preceded It. 'The power of relevant arguments and
social cornpanson IS rnultipher! In an underground group.
7. Group Radicalirution ill Competition for the I.Sal11e Base 0.( Support
Groups In cornpetiuon for the same base of sympatluzers can, like individuals. gain
status by more radical action 111 support of the cause:+ o Analysts have suggested that
the] 979 assassin.uion of Lord Mountbattcn by the Irish Republican Army in 1979
was an effort to compete wi th escalated attacks by the Irish N ational Liberauon
Arrnv. and that the 1985 hijacking of both a ~rw A plane and the luxury liner Achille
Lauro 'vvere attempts by Palestinian terrorists to gain advantage over rival groups.-+]
'Today It IS common to see more than one group claiming credit for a particular
terrorist attack, even for a particular suicide terrorist attack.":'
Radrcahzation by competition is particularly clear in the case of the Arrncruan
Secret Army for the LIberation of Armenia. ASALl\ first gained diaspora support
by attacking Turks a t a tune when main-line Armenian organizations were only talk­
Ing about rctribuuon for the Turkish genocide of Armenians. One of the older
organizations (Tashriaks) responded to the new competition by establislunp Its
own anu-Turk ish terrorist group. the Justice Commandos of the Armenian
Genocide .c+.~ Sirmlarlv. the Palestinian Front for the Liberation of Palcstme was
forced to take up suicide terrorism despite its materialist Marxist logic when PFLP
began to seem rrrclcvan t in the second intifada.-+-+
It is possible for a group to become too radical and lose its base of support. The
line between higher status from more radicalization and lower status frorr: too much
radicalization is fine and variable over time. That it IS possible to go over the lme is
mdrcatcd by examples when the I R.A expanded us targets beyond what 11s republican
sympat lnzers would accept: on such occasions the IRA would apologize and narrow
ItS target range. a t least for a period of tinle.-+';;
Similarly, Pa lestinian su icidc ter rorisrn attacks against Israel slowed drama ti­
cally in the period after the Oslo Accords. I-lope of a peace agreement was asso­
crated with decreased support for terrorism. as reported III polling of
Palcstinrans. When the promise of the Oslo Accords was lost and the second Inti­
fada began, polls showed support for terrorism rising to new highs even as the
number of terrorrst attacks rose to new highs. It appears that in many cases terror­
ism Increases with popular support for terrorism. but can decline if popular sup­
port for terrorism declines. 1\11 too often, however. more radical action brings
more stat us and more support to a grou p competing wi: h ot her groups to repre­
scnt the sa me ca use.
An often-overlooked aspect of competiuon for a base of support b VIolence
agamst competitors. About one quarter of the k ilung in Northern Ireland \V~lS
Catholics kilhng Catholics and Protestants killtng Protestatlts.-+ cJ Both srdes killed
suspected informers or individuals resisting the discipline militants sought to impose
'The I R. A in pa rtic u 1a I' a tt aeke dan d killed tho sc ignor ing I R A stn c t u res aga i 11St sc11­
ing drugs.
An extreme exam pIe of ingroup VIolence is the Tamil 'Tigers, who. In their rise to
power, killed 1110re Tamils than Sinhalesco-+ 7 The L~rTE early wiped out compet.ng
Tamil militant groups, and continued in 2006 killing individual l-eUl1I1 cntics and
'Tamil political opponents. An example that permeated the Western press was the
Ju1) 29, l0(
the 1-' a 1111 \ l
[Jr. 1-1 ruche
From a
threat [rom
pressures Io
pomt of v n
after some I
our sacrific
'vVh 1ch VI o II
1ngroup enl
8. Group J?
'ThIS form
isis." ,A gr
to make a 1
disobedicn
of police 1
CIvil or hu
rcprcssion
dynaillic I'
uctivists. 1
or el1\
sit-In m.
costs a~ t(
f11ltnlcnt ;
are not \\1
vanc\..' :l rl'
and nOI\­
pu lice'. \\
to face 1
Sf 1t'-Sl' lee I
den~ed ii
rrh\~
her
re~l'{\
in Germ
nl()\'cnlC
111 (1erll1,
the Stud
'Vlctn:lll1
Rae1
bel \\ ccn
uon tn
kIlled In
'rile rcu
\ III \CI1L'l
l)c1
M echanisms
ments and
inals, gam
~ested
that
iy in 1979
jberation
.er Achille
groups,-tl
particular
Armenian
a support
only talk­
the older
ishing its
Armenian
stine was
len PFLP
port. The
too much
the line is
.publican
d narrow
drarnati­
vas asso­
>lling of
ond inti­
.n as the
~s terror­
ular sup­
n brings
to repre­
violence
and was
.es killed
) impose.
iinst sell­
.ir rise to
.mpeting
itics and
was the
of Radicali:a t ion
425
July 29. 1999, suicide-bomb killing of Dr. Neelan Tiruchelvam As a leader of
the Tamil United Liberation Front and a Member of the Sri Lankan Parliament,
Dr. Tiruchclvam was a leading critic of human rights abuses by the LTTE.
From a group dynamics point of view, threat from ingroup competitors is like
threat from an outgroup enemy in producing high cohesion, with resulting high
pressures for conformity and strong sanctions against deviates, From an individual
point of view, when my friends and I are risking all for the cause, and especially
after some of our friends have died for this cause, no one can be allowed to betray
our sacrifice. The competition for in group support is a competition for survival, in
which violence against an outgroup enemy is often joined with violence against
.
.
mgroup enermes.
8. Group Radicalization in Competition with State Power-s-Condensation
This form of radicalization has been a focus of research by social movement theor­
ists.-t8 A group with weak and diffuse popular support attains sufficient organization
to make a public display: a rally ~ a protest march, a sit-in, or some other form of civil
disobedience. The power of the state is exerted to quash the group, often in the form
of police response that may include indiscriminate violence or S0111e abrogation of
civil or human rights. The result is an increase in sympathy for the victims of state
repression and some mobilization of the group's sympathizers toward action. (This
dynamic is considered later in relation to mass radicalization.) For social movement
activists, however, there is another dynamic at work, a dynamic or condensation.
Of all those who take the first radical action--joining an illegal rally or march or
sit-in-s-most are likely to respond to repression by giving up action. They see the
costs as too high to continue. Others will not be deterred and will increase their corn­
mitrnent and escalate their action against the state. The determinants of this choice
are not well studied, but probably those who bring a moral frame and personal grie­
vance are less easily deterred. In any case, the result of the interaction between state
and non-state group is often a mutual escalation of violence between group and
police, with further peeling off of individuals whose radicalization is not sufficient
to face increasing state pressure. The conclusion of this cycle of escalation and
self-selection is likely to be that a tiny fraction of the original protest group has con­
densed into a highly radicalized group that goes underground as a terrorist cell.
This cycle of reaction and counter-reaction has been described by della Porta in
her research on the origins of the Brigate Rossa in Italy and the Red Anny Faction
in Germany." The Red Brigades condensed out of 1960s leftist student protest
movements in Italy; the RAF condensed out of similar leftist student protest groups
in Germany. Sprinzak" has described a similar trajectory by which a tiny fraction of
the Students for a Democratic Society ~ who began with protest against the war in
Vietnam, condensed into the Weather Underground.
Radicalization by condensation depends upon the strength of the affective tICS
between individuals, in particular ties to individuals who suffer from the state reac­
tion to radical challenge. Comrades imprisoned cannot be abandoned: comrades
killed in police shootouts or in prison are martyrs whose deaths demand a response.
The reaction in many eases is increased commitment to violence to pay back state
violence.
Della Porta 5l offers a number of examples of individuals for whom the death or
imprisonment of a comrade was the instigation for joining a terrorist underground.
-126
c.
AJcCou/(!y and S. Moskalcnk o
Auger and revenge are no doubt important in this kind of reaction, but a kind of
"survival guilt" may also contribute. Those alive and free feel guilty that a better
man or woman is dead or in prison. Research toward understanding survival guilt
has only recently begun.i' and may playa part in understanding the political power
of martyrdom.
This power is evident in an example offered by della Portar" "For example,
Volker Speitel, one of the militants who worked in political groups that supported
the RAF militants in prison, described how the death of Meins (by hunger strike)
pushed him to the final step of joining the underground: 'Then the day carne when
Holger Meins died ... For us this death was a key experience ... The death of Holger
Meins and the decision to take arms were one and the same thing. Reflection was not
possible anymore."
9. Group Radicalization ill Within-Group Competition-s-Fissioning
The within-group competition for status represented in social comparison theory can
produce intense conflict. The downside of conflating the personal and the political
is that differences of political opinion can lead to personal animosities-and vice
versa.:" Some observers ha ve suggested that only common action against the state
or another group can save a terrorist group from tearing itself apart.i"
Systematic data are lacking, but examples suggest that intra-group conflict leads
often to splitting or fissioning of a terrorist group into multiple groups. The IRA pro­
vides an obvious example, with many competing factions---Official IRA, Provisional
IRA, Real fRA, Continuity IRA, fNLA--who sometimes targeted one another. Simi­
larlya split within ASA LA was the occasion of killing between former comrades. 56
Intra-group competition can go beyond killing. A threat from members of our
own group is likely to produce a feeling of contamination that requires not just death
but torture and obliteration. Such was the fate, evidently, of 14 members of the Japa­
nese United Red Army who in 1972 were found dead and dismembered in a group
hideout.:"
From a group dynamics perspective, the tendency toward fissioning in radical
groups should not be surprising. As already noted, cohesion leads to pressures for
agreement within the group. When, as in an already radical group, perception of
external threat produces very high cohesion, the pressure for agreement is very
strong. An individual will seldom be able to resist the pressure of a unanimous
majority, but a minority of two or 1110re individuals may be able to resist.i" When
the pressure for agreement is very strong, the minority is likely to be expelled from
the group---or obliterated.
Jo. AJass Radicalization ill Conflict with all Outgroup-r-Jujitsu Politics
This form of radicalization can be understood as a generalization of the group
dynamics theory already described. In small face-to-face groups, outgroup threat
leads reliably to increased group cohesion, increased respect for ingroup leaders,
increased sanctions for ingroup deviates, and idealization of ingroup norms.i" In lar­
ger groups, reference to cohesion is often replaced with reference to ingroup identi­
fication, patriotism, or nationalism, but the pattern in response to outgroup threat is
similar to that seen in sl11a11 groups. Consider the results of the 9/11 attacks on U.S.
politics.i" in
increased su
increased sa
suggesting t
("they hate
Mass 1''­
egy. Some t
beyond the
lized to ac
beyond wh:
using the el
Dr. Ay
Under the 1
opined, An
from behin
countries, 1
Taliban \\:~
had expect
support fo
J1. Mass J
It is often
longed V10
'This tcnde
DehllmanJ
abstract ly.
as follo~'s
are not hi
Dehll
about hal!
the war \'
cohesion (
the Japan
were even
the war y\
Smnl.
the Germ
London ~
Apparent
puni tivc
perhaps l
A lu:
hatred S
anger~ fc
negativ"C
a bad es~
many en
NIechanisms of Radicalization
a kind of
It a better
vival guilt
cal power
example,
supported
~er strike)
une when
of Holger
n was not
leory can
. political
-and vice
the state
4-J 1I
politicsr'" increased patriotism visible in rallies, flags, banners, and bumper stickers;
increased support for the president and for every agent and agency of government;
increased sanctions for Americans challenging the consensus (Bill Maher sacked for
suggesting the 9/11 attackers were not cowards); and reification of American values
C"they hate us for our values").
Mass radicalization by external attack is so reliable that it can be used as a strat­
egy. Some terrorists have explicitly sought to elicit a state response that will carry far
beyond the terrorists to strike terrorist sympathizers who have not yet been mobi­
lized to action.v' The predictable result is to mobilize terrorist sympathizers far
beyond what the terrorists can accomplish alone. We call this strategy jujitsu politics:
using the enemy's strength against him. 62
Dr. Ayman Al Zawahiri enunciated this strategy in his political memoir Knights
Under the Banner of the Prophet. 6 3 If the shrapnel of war reach American homes, he
opined, Americans will either give up their aims in Muslim countries or will come out
from behind their Muslim stooges to seek revenge. If Americans move into Muslim
countries, he predicted, the result will be jihad. Although the U.S. war against the
Taliban was faster and cleaner of collateral damage to civilians than Al Qaeda
had expected, the U.S. move into Iraq has indeed been associated with increasing
support for radical Islam in Muslim countries.
lict leads
[RA pro­ ovisional
ter. Simi­
ades.i"
rs of our
rst death
he Japa­
a group
1
radical
lures for
ption of
IS very
mimous
)8 When
ed from
~ group
) threat
leaders~
) In laridenti­
hreat is
I
JD
U.S.
II. Mass Radicalization in Conflict with all Outgroup-s-Hate
I t is often observed that groups in conflict, especially if the conflict involves pro­
longed violence, become more extreme in their negative perceptions of one another.
This tendency can become so extreme that the enemy is no longer seen as human.I'"
Dehumanization is signaled by referring to targets as "pigs," "dogs," or, more
abstractly, "wheels" in the enemy machine. Della Porta'" quotes an Italian militant
as follows: '" ... enemies are in a category, they are functions, they are symbols. They
are not human beings."
Dehumanization can occur in interstate conflict as well. In WWII, for instance.
about half of American soldiers favored wiping out the whole Japanese nation once
the war was won. This radical opinion did not depend on membership in a high
cohesion combat group, nor did it depend on experience of losses in combat against
the Japanese. Indeed soldiers in training in the U.S., who had never been in combat,
were even more likely than combat soldiers to favor exterminating the Japanese after
the war was won. 66
Similarly it has been observed that residents of English cities never born bed by
the Germans during WWII were more bloody-minded and vengeful than residents of
67
London and other cities of southern England that felt the full fury of the Blitz.
Apparently group identification in the context of group conflict can lead to radically
punitive attitudes even in the absence of personal victimization by the enemy­
perhaps especially in the absence of personal victimization.
A high level of categorical hostility toward another group is often described as
hatred. Some theorists believe that hate is an emotion, perhaps a combination of
anger, fear, and conternpt.i'" A more recent view is that hate is an extreme form of
negative identification that includes the idea that members of the enemy group share
a bad essencc.i" In this view hate is not an emotion but the occasion of experiencing
many emotions, depending on what happens to the hated target. As above, positive
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
428
C. Mcc.auley and S. Moskalenko
emotions are occasioned when bad things happen to the hated group, and negative
emotions are occasioned when good things happen to the hated group.
The idea that the enemy shares a bad essence can make sense of the impulse to
attack all of them, without regard for age, gender, or civilian status. A group's
essence is the hidden something shared by group members that gives them their
tendency toward shared group characteristics.i" A group's essence is understood
to be stable over historical time and immutable for the individual group member.
If the essence is bad, there is nothing to be done--negotiation and education can
no more make a difference than negotiation or education can make a difference in
the essence of a tiger. If tigers threaten us and hurt us, all tigers are targets-old,
young, in uniform or out of uniform.
/2. Mass Radicalization in Conflict with an Outgroup-v-Martyrdom
The root meaning of rnartyr is witness, and there is something particularly powerful
about a form of witnessing that takes the life of the witness. One way to think about
the issue is to consider the psychology of persuasion, in which a credible source com­
bines expertise and trustworthiness." A martyr is trustworthy insofar as it is difficult
to see how an individual giving up life for a cause could be lying for some personal
Interest or advantage. This leaves the question of expertise, and the social construc­
tion of a martyrdom has to rule out the possibility that the martyr is "crazy" or
otherwise unable to choose death freely.72 It follows that higher status martyrs make
better witnesses: better educated, more successful individuals, with more life choices
a vailable, are seen as knowing better what they are doing when they give their lives
for a cause.
Radical groups try to keep salient the memory of their martyrs. The Liberation
Tigers of Tamil Eelam yearly celebrate three days of Martyrs' Day activities, includ­
mg honoring the parents of dead heroes. Palestinians killed by Israel are remembered
with portraits, graffiti, shrines, and rallies such as are often held in Martyr's Square
in Gaza. Palestinian web sites offer videos made by suicide terrorists before their
attacks.
As noted earlier in relation to Holger Meins's death by hunger strike, there is
reason to believe that the political impact of martyrdom can be significant. Mahatma
Gandhi's hunger strike against British rule in India is probably the most famous
example, although this was not a fast to the death. Perhaps the strongest example
is recounted in Ten Men Dead,73 the history of IRA and INLA prisoners who died
on a hunger strike to protest British efforts to treat political prisoners as common
criminals. The men died over a period of 73 days. Several hunger strikers were
elected to the Irish or British parliaments, and many observers believe that the hun­
ger strikes resuscitated a moribund Provisional IRA.
The social construction of martyrdom is under-theorized (but see note 74), and
empirically under-developed, but the impact of martyrdom on mass audiences
deserves close attention.
the value of
we undcrto.
mass levels.
We do r
ones. No do
radicaliza tic
mutual relat
here in rela
understandi
tions for im
Nor do
discussed h.
influences t
framework
anyone of
for a single
aware, mul
neither suff
we suggest
groups to r
This vi
social mobi
ways to e;
S0111e indi.
involvemer
aries) and,
another (1\
past in joir
experience
ment in an
them to je
politics.
Simila
mult i-Iacu
tories." K 1
terrorists
tion and ir
personal 0
blems. 1~h,
and the de
of pathvv'a
This
I
'''Beconlll1
pIex in th.
Radicalization as Opposition Politics
We began with a conceptualization of political radicalization as change in beliefs,
feelings, and action toward support and sacrifice for intergroup conflict. We noted
that these aspects of radicalization are only moderately correlated, and suggested
bear on a
wi th terre
Still,
1
identified
vidual rae;
Mechanisms of Radicalization
rnd negative
e impulse to
. A group's
; them their
understood
up member.
ucation can
jifference in
argets----old,
'Iy powerful
think about
.ource corn­ t is difficult
ne personal
11 construc­
"crazy" or
irtyrs make
life choices
~ their lives
Liberation
ies, includ­
.membercd
vr 's Sq uare
lefore their
ce, there is
Mahatma
ist famous
st example
; who died
s common
ikers were
It the hun­
e 74), and
audiences
the value of differentiating these aspects both in conception and measurement. Then
we undertook a review of mechanisms of radicalization at individual, group, and
mass levels.
We do not suppose that the twelve mechanisms identified are the only important
ones. No doubt more will be uncovered. But we do expect that the more powerful the
radicalization, the more mechanisms will be implicated and the more artful their
mutual relation and reinforcement. Thus we suspect that the mechanisms considered
here in relation to political radicalization and terrorism may also be important in
understanding thought reform, cult recruiting, military training, and state prepara­
tions for interstate war.
Nor do we propose a single underlying theory uniting the twelve mechanisms
discussed here. Indeed it seems unlikely that any single theory can integrate all the
influences that bring individuals to radical political action, although a conceptual
framework in which to view these influences may be possiblc.i:' It is unlikely that
anyone of these mechanisms is sufficient to explain political radicaliza tion, even
for a single individual. In every individual trajectory to terrorism of which we are
aware, multiple mechanisms can be identified. Thus the twelve mechanisms are
neither sufficient causes one by one nor instantiations of some larger theory. Rather
we suggest that there are multiple and diverse pathways leading individuals and
groups to radicalization and terrorism.
This view is consistent with previous research on psychology of terrorism and
social mobilization. For instance, Linden and Klanderrnans" distinguish three path­
ways to extreme-right political activism: continuity, conversion, and compliance.
Some individuals show continuity in a lifetime of consistent political interest and
involvement. Of these, some are consistently involved in the same cause (revolution­
aries) and SODle are consistent only in their involvement in one extreme group after
another iwondcrersv. Other individuals show a trajectory of sudden break with their
past in joining an extreme movement (converts), often following a dramatic personal
experience such as auto accident or rape. Finally there are individuals whose involve­
ment in an extreme movement occurred through friends or relatives who persuaded
them to join icompliamsv, although they had not previously had much interest in
politics.
Similarly, Kimhi and Even 77 argue that " ... not only is suicide terror a complex
multi-factorial phenomenon, but also seems to be a phenomenon of multiple trajec­
tories." Kimhi and Even identify four motive-trajectories among Palestinian suicide
terrorists. Religious motives for jihad and martyrdom, nationalist motives for libera­
tion and independence of the Palestinian people, motives of retribution or revenge for
personal or group victimization by Israelis, and motives of escape from personal pro­
blems. These motives often overlap to considerable extent in particular individuals,
and the degree of overlap gives again an indication of the multiplicity and complexity
of pa thways to terrorism.
This complexity is well summarized in the conclusion of Horgan's chapter,
.
. ,,7R I."TI
-- Becoming
a T errorist.tt'
rnc rea 1"ity IS t1iat t h ere are many f actors ( 0 f ten so COlTI­
plex in their combination that it can be difficult to delineate them) that can come to
bear on an individual's intentional or unintentional socialization into involvement
with terrorism."
Still, it is worth noting that there is a reactive quality to most of the mechanisms
identified. Of the twelve mechanisms, only t\VO are more relatively autonomous. Indi­
vidual radicalization in joining a radical group-s-the slippery slope is a mechanism of
v
in beliefs,
We noted
suggested
429
~r
430
C. NfcCauley and S. Moskalcnko
self-radicalization via self-j ustification, in which new beliefs and values are adopted
in order to make sense of past behaviors. These new reasons then support more
extreme behavior in the same direction. Group radicalization in like-minded groups
is also more an autonomous than a reactive mechanism: the events reacted to occur
within a group as arguments and individuals compete for acceptance.
The other ten mechanisms reviewed are more clearly reactive. They begin from
and depend on a dynamic of opposition in which the significant events are the
actions of others. Individuals react to personal victimization, to group grievance,
and to state action against friends and lovers. Non-state groups react to threat from
the state, threat from other groups competing for the same base of sympathizers, and
threat from internal dissension. Mass publics react to state action that injures indis­
criminately, to martyrs, and, in long conflicts, to a perception of the enemy as less
than human.
The reactive character of these mechanisms is important because, as noted in the
introduction, efforts to understand radicalization usually focus on the non-state
actors who are radicalized. Terrorism research, in particular, tends to focus on
theIn-the terrorists-rather than on the situation they are in--or, more precisely,
the situation they believe they are in. But these mechanisms do not operate only
in non-state groups challenging the state. The same mechanisms moving people
toward radicalization and terrorism will operate as well in those who react to radi­
cals and terrorists.I" Even a cursory look at the experience of the U.S., since the
attacks of September 11, 2001, can suggest that those attacked have not escaped a
radicalization of their own.
The degree to which radicalization of non-state groups occurs in response to the
actions of others must be the starting point for understanding these groups. Political
radicalization of individuals, groups, and mass publics occurs in a trajectory of
action and reaction in which state action often plays a significant role. Radicaliza­
tion emerges in a relationship of intergroup competition and conflict in which both
sides are radicalized. It is this relationship that must be understood if radicalization
is to be kept short of terrorism.
I
Resl
PIJlLt
ExpJ
und
Jv1 ,-.H
Jcwi
tioni
Nat'
Cau
KIrr
Anu
gall
Seal
Poll
goo
Ath
Sen
Th(
Al
(1/
"
l\).­
pOl'
19S
anr
Notes
(St
1. James Mahoney. "Tentative Answers to Questions About Causal Mechanisms.'
Paper presented at the Annual Meeting for Political Science Association, Philadelphia, PA,
2003.
2. Renate Mayntz, "Mechanisms In the Analysis of Social Macro-phenomena," Philoso­
phy ofSocial Science 34 (2004): 237-259.
3. J. P. Chaplin, Dictionary of Psychology, 8th ed. (New York' Dell, 1975),285.
4. R. J. Rummel, Death by Government (New Brunswick, NJ: Transaction, 1994).
5. David A. Snow and Pamela E. Oliver, "Social Movements and Collective Behavior:
Social Psychological Dimensions and Considerations," in Karen S. Cook, Gary Alan Fine and
James S. House, eds., Sociological Perspectives on Social Psychology (Boston: Allyn and
Bacon, 1995), 571--599.
6. Donald W. Hine and Christina Jaymc Montiel. "Poverty in Developing Nations: A
Cross-Cultural Attributional Analysis," European Journal of Social Psychology 29, no. 7
(1999): 943 -959.
7. Roy J. Eidelson and Judith 1. Eidelson, "Dangerous Ideas: Five Beliefs That Propel
Groups Toward Conflict," American Psychologist 58 (2003): 182-192.
8. Jeff Goodwin, James M. Jasper and Francesca Polletta, eds., Passionate Politics:
Emotions and Social Movements (Chicago: University of Chicago Press, 2001).
Pel
H
of
Mechanisms of Radicalization
re adopted
iport more
ded groups
.d to Occur
begin from
Its are the
grievance
hreat from
hizers, and
ures indis­
.my as less
oted in the
non-state
focus on
. precisely.
erate only
ng people
ct to radi­
since the
escaped a
I
inse to the
). Political
jectory of
.adicaliza­
'hich both
calization
chanisms, ~.
.lphia, PA.
. Philoso­
285.
1994).
Behavior'
n Fine and
Allyn and
~ations:
A
29, no. 7
hat Propel
e Politics.
431
9. H. Flum, "Anger in Repressive Regimes: A Footnote to Dornmatior, and the Arts of
Resistance by J ames Scott," European Journal ofSocial Theory, Special Issue on Anger in
Political Life, ed. Mary Holmes, 7, no. 2 (2004): 171-188.
10. Diane M. Mackie, Thierry Devos and Eliot R. Smith, "Intergroup Emotions:
Explaining Offensive Action Tendencies in an Intergroup Context," Journal of Personality
and Social Psychology 79, no. 4 (2000): 602-612.
11. Sonya Roccas, Yechiel KIaI' and Ida Leviatan, "Who Feels Guilt? Collective Guilt,
Moral Outrage, Exonerating Cognitions, Group Identification and Personal Values A1110ng
Jewish-Israelis," in Nyla R. Branscombe and Bertjan Doosje, eds., Collective Guilt: Interna­
tional Perspectives (New York: Cambridge University Press, 2004).
12. Clark Mcf.auley, "The Psychology of Group Identification and the Power of Ethnic
Nationalism," in Daniel Chirot and Martin E. P. Seligman, eds., Ethnopolitical Warfare:
Causes, Consequences, and Possible Solutions (Washington, DC: APA Books, 2001), 343-362.
13. Clark McCauley, "Jujitsu Politics: Terrorism and Response to Terrorism." in Paul R.
Kimmel and Chris E. Stout, eds., Collateral Damage. The Psychological Consequences of
America's rVar on Terrorism (Westport, CT: Praeger, 2006),45--65.
14. McCauley (see note 13 above), 45-65.
15. Gallup (2003) Question qn33_Fonn A. Retrieved March 5, 2006, from http://brain.
gall up.coru/ documents / question.aspx?q uestion==145523&Advanced=0&SearchConType= 1&
Sea rc hTypeA11 = federal (~)2 0go vernment%2 Oso0/0201arge(I'()20andc:!cl2Op0 werfu 1
16. Ashutosh Varshney. "Nationalism, Ethnic Conflict, and Rationality," Perspective on
Politics 1 (2003): 85-99.
17. For this and subsequent references to people and events In the news, readers may try
google or obtain relevant URL from [email protected]
18. K. C. Tessendorf, Kill the Tzarl Youth and Terrorism in Old Russia (New York:
Atheneum, 1986).
19. Carol Morello, "Virginia Court Upholds Muhammad Sentences: Sniper Could be
Sent to Another State," Washington Post, April 23, 2005, 1.
20. Gerald H. Zuk and Carmen Veiga Zuk, "Negation Theory as a Cause of Delusion:
The Case of the Unabomber,' Contemporary Family Therapy 22, no. 3 (2000): 329-336.
21. http://query.nytin1es.com/gst/fullpage.html?sec=health&res=9FOCEED61 E3CF937
A 15752COA9679C8B63&n=Topo/02fReference~~J2ITimes(~)20TopicS(!~l2fSubjectsoA)2fH(/()2fHate
(j';J20Crimes
22. Donatella della Porta, Social Movements, Political Violence, and the State. A C'0111­
parative Analysis of Italy and Germany (Cambridge, UK: Cambridge University Press.
1995)~ 168.
23. John Sabmi, Social Psychology (2nd ed.) (New York: Norton, 1995).
24. Stanley Milgram, Obedience to Authority: An Experimental VieH' (New York: Harper
and Row, 1974).
25. Philip G. Zimbardo, "Quiet Rage: The Stanford Prison Study" Videorecording
(Stanford, CA: Psychology Dept., Stanford University, 1971).
26. Marc Sageman, Understanding Terrorist Networks (Philadelphia: University of
Pennsylvania Press, 2004).
27. Della Porta (see note 22 above), 168.
28. Ibid.
29. Klaus Wasmund, "The Political Socialization of West German Terrorists.' in Peter
H. Merkl, ed., Political Violence and Terror Motifs and Motivations (Berkeley, CA: University
of California Press, 1986), 191-228 (p. 204).
30. Della Porta (see note 22 above), 168.
31. McCauley (see note 12 above).
32. Robert W. White, "Commitment, Efficacy, and Personal Sacrifice Among Irish
Republicans," Journal of Political and Military Sociology 16 (1988): 77-90.
33. White (see note 32 above), 83.
34. Roger Brown, "Group Polarization. ~~ in Social Psychology: The Second Edition (New
York: Free Press, 1986).
35. Ibid.
36. G. Levinger and D. J. Schneider. "Test of the 'risk as a value' hypothesis," Journal of
Personality and Social Psychology 11 (1969): 165--169.
') )
4~-
C. J'vlcCauley and S. Moskalenko
37. Peter Collier and David Horowitz. Destructive Generation (New York: Summit
Books, 1989).
38 Brown (see note 34 above).
39. http://\vw\v.anny.Inil/clnh-pg/moh Lhtrn
40, Della Porta (see note 22 above, Chapter 4).
41. Clark McCauley and Mary Segal, "Social Psychology of Terrorist Groups," in
C. Hendrick, ed.. Review of Personality and Social Psychology 9 (Beverly Hills, CA: Sage,
1987): 231-256 (p. 238).
42. Mia M. Bloom, Dying to Kill: Thf Allure o] Suicide Terror (New York: Columbia
U DIversity Press, 2005).
43. Laura Dugan, Julie Huang, Gary LaFree and Clark McCauley, Sudden Desistence
Front Terrorism. The Armenian Army for the Liberation ofArmenia and the Justice Commandos
o] the Armenian Genocide. Unpublished manuscript (2006).
44. Bloom (see note 42 above).
45. McCauley and Segal (see note 41 above), 237.
46, Marie Therese Fay, Mike Morrissey and Marie Smyth. Mapping Troubles-Related
Death» in Northern Ireland J969---J99~ (Derry Londonderry: INCORE, 1998), Table 12.
47 Rajan Hoo le, Daya Bomasundaram, K. Sritharan and Rajani Thiranagama, The
Broken Palmyra. The Tanzi! Crisis in Sri Lanka, an Inside Account (Claremont, CA: The
Sn Lankan Studies Institute, 1990).
4R Christian Davenport, Hank Johnston and Carol Mueller, eds., Mobilization and
Repression (Vol. 21) (Minneapolis, MN: LJniversity of Minnesota Press, 2005).
49 Della Porta (see note 22 above), 78-82.
50. Ehud Sprinzak, "The Process of Dclegitimation: Toward a LInkage Theory of
Political Terrorism," in Clark McCauley, cd., Terrorism and Public Policy (London: Frank
Cass, 199]). 50-68.
51 Della Porta (see note 22 above), 168-169.
52. Edvvard S. Kubany, Fransis R. Abueg, James M. Brennan, S. N. Haynes, Frederic P.
Manke and C. Stahura, "Development and Validation of the Trauma-Related Guilt Inventory
(TRGI):~ Psychological Assessment 8 (1996): 428--444.
53. Della Porta (see note 22 above), 169.
54. Collier and Horowitz (see note 37 above).
55, Janusz K. Zawodny, "Infrastructures of Terrorist Organizations.' in Lawrence Zelie
Freedman and Yonah Alexander, eds., Perspectives on Terrorism (Wilmington, DE: Scholarly
Resources, 1983), 61- 70.
56. Dugan. et al. (see note 43 above).
57 Walter Laqueur, Terrorism (Boston: Little, Brown, 1977),125.
58. Solomon E. Asch, "Studies of Independence and Conformity: A MInority of One
Against a Unanimous Majority." Psychological Monographs 70, no. 9, 1956 (Whole No. 416).
59. John Duckitt and Kirstin Fisher, "The Impact of Social Threat on Worldview and
Ideological Attitudes," Political Psychology 24~ no. 1 (2003): 199--222.
60. Clark McCauley, "Psychological Issues in Understanding Terrorism and
the Response to Terror ism" in Chris E. Stout, ed., The Psvcliologv of Terrorism Vol III
(Westport, CT: Praeger, 2002), 3 30.
61 Carlos Marighella. Mininianual of the Urban Guerrilla (J. Butt & R Sheed. Trans.)
(Havana: Tricontinental, 1970).
62. McCauley (see note 13 above).
63. Ayman Al Zawahiri, Knights Under the Banner (~l the Prophet (200 1). Retrieved
March 11. 2004, from http) /www.fas.org/irp/world/para/aynlan_bk.htn1.
64. Della Porta (see note 22 above), 173-174.
65. Ibid.
h6. Samuel A. Stouffer, Arthur A. Lumsdaine. Manon H. Lumsdaiue, Robin M.
Williams Jr., M. Brewster Smith, Irving L. Janis, et al., The American Soldier: Combat and
lls Aftermath (Vol. 2) (Princeton, NJ: Princeton University Press, 1949).
67. Robert A. Papc, Bombing to ~VjJl. Air Power and Coercion in VVOf (Ithaca, NY:
Cornell University Press, 1996).
68. Robert J. Sternberg, "A Duplex Theory of I-late: Development and Application to
Terrorism, Massacres, and Genocide." Review ~l General Psychology 7, no. 3 (2003): 299--328.
69. Edward E
Four Ways of 1'1­
(Washington, D.C
70. Nick H~lsl
Categories." Britis
71. Carl I I­
Communication E
72. Lacey Ba
JIVesrern lVorld (E'
73. David Be:
(N ew Yo r k: At 1an
74. Farhad K
(London: Pluto PI
75. Max Tayl
gical Process in th
585--601.
76. Annette 1
Right Activism In
213--228
77 Shaul Kl
Victoroff, ed .. To
(Washington. D.(
78. John I-Il)!
79. Christian
PoE tical Science (\
eds.~ Repression u
vii-xh
Mechanisms ofRudicalization
Groups," 111
ls, CA: Sage,
rk: Columbia
'en Desistcnc»
:/ Commandos
ubles- Related
able 12.
ragarna, The
nt, CA: The
i/i~(lti017 LInd
~
Theory of
idon: Frank
, Fredenc P.
ill Inventory
vrcnce Zelic
E: Scholarly
rity of One
le No. 416).
-ldvievv and
»rism
"Ill!
and
Vul. III
ed, Trans.)
Retrieved
Roblll IV1.
'(JI
71 ba t
and
haca. NY'
lication to
. 299-328
433
69. Edward E. Royzrnan, Clark McCauley and Paul Rozin, "From Plato to Putnam:
Four \Vays of Thinking About Hate" in Robert J. Sternberg, ed., Psychology oj' flate
(Washington, D.C.: APA Books, 2004), J-35.
70. Nick Haslam, Louis Rothschild and Donald Ernst, "Essentialist Beliefs About Social
Categories," British Journal of Social Psychology 39, no. 1 (2000): 113-127.
71. Carl I. Hovland and Walter Weiss, "The Influence of Source Credibility on
Communication Effectiveness," Public Opinion Quarterly i S (1951): 635-650.
72. Lacey Baldwin Smith, Fools, Martyrs, Traitors: The Story of Martyrdom in the
Western World (Evanston, IL: Northwestern University Press, 1997).
73. David Beresford, Ten Men Dead: The Story of the 1981 Irish Hunger Strike
(New York: Atlantic Monthly Press, 1987).
74. Farhad Khosrokhavar, Suicide Bombers: Allah's NCH' Martyrs (D. Macey, Trans.)
(London: Pluto Press, 2005).
75. Max Taylor and John Horgan, "A Conceptual Framework for Addressing Psycholo­
gical Process in the Development of a Terrorist," Terrorism and Political Violence 18 (2006):
585-601.
76. Annette Linden and Bert Klandermans. "Stigmatization and Repression of Extreme­
Right Activism in the Netherlands," Mobilization. An International Journal 11, no. 2 (2006):
213-228.
77. Shaul Kimhi and Shcmuel Even, "The Palestinian Human Bombers.' in Jeff
Vict oroff, cd., Tangled Roots: Social and Psychological Factors in the Genesis of Terrorism
(Washington, D.C.: lOS Press, 2006), 308-323.
78. John Horgan, Psychology ofTerrorism (New York: Routledge, 2005), 105-- 106.
79. Christian Davenport, "Introduction: Repression and Mobilization: Insights from
Political Science and Sociology," in Christian Da venport, Hank Johnston and Carol Mueller,
eds., Repression and Mobilization (Minneapolis, ~1N: University of Minnesota Press, 2005),
vii-xh.