international system and technologies of rebellion

7/28/2019 International System and Technologies of Rebellion

1/15

American Political Science Review Vol. 104, No. 3 August 2010

doi:10.1017/S0003055410000286

International System and Technologies of Rebellion:How the End of the Cold War Shaped Internal ConflictSTATHIS N. KALYVAS Yale UniversityLAIA BALCELLS Institute for Economic Analysis, CSIC

Because they are chiefly domestic conflicts, civil wars have been studied primarily from a perspective

stressing domestic factors. We ask, instead, whether (and how) the international system shapescivil wars; we find that it does shape the way in which they are foughttheir technology of

rebellion. After disaggregating civil wars into irregular wars (or insurgencies), conventional wars, andsymmetric nonconventional wars, we report a striking decline of irregular wars following the end of theCold War, a remarkable transformation of internal conflict. Our analysis brings the international systemback into the study of internal conflict. It specifies the connection between system polarity and the ColdWar on the one hand and domestic warfare on the other hand. It also demonstrates that irregular waris not the paradigmatic mode of civil war as widely believed, but rather is closely associated with thestructural characteristics of the Cold War.

In 1975, most civil wars were located in Asia, andall but two were guerrilla warscontests entailingan asymmetric rebel challenge launched from the

countrys rural periphery. In 1993, in contrast, mostinternal conflicts were located in sub-Saharan Africa,and guerrilla wars comprised less than half of them.Much more common were conventional wars usingheavy armor and artillery in a landscape dominatedby siege warfare and trenches, or primitive wars be-tween poorly armed and trained militias. We argue thatthis dual geographic and military shift is symptomaticof a broader transformation of internal conflicttheresult of a major structural change in the internationalsystem: the end of the Cold War.

This transformation has been overlooked for threereasons: (1) the literature has stressed the determinantsof civil war onset over the way in which civil wars arefought, (2) it has treated civil wars as a homogeneousphenomenon over time and space, and (3) it has em-phasized the impact of domestic factors compared tointernational ones. We show instead that incorporatingthe international system into the analysis of civil wars

Stathis N. Kalyvas is Arnold Wolfers Professor of Political Science,Yale University, 201 Rosenkranz Hall, 115 Prospect Street, NewHaven, CT 06520-8301 ([email protected]).

Laia Balcells is Researcher, Institute for Economic Analysis,Spanish Higher Council for Scientific Research, Campus UAB, 08193Bellaterra, Barcelona, Spain ([email protected]).

For their helpful comments, we thank Ana Arjona, ChrisBlattman, Robert Brenner, Mario Chacon, Jesse Driscoll, ThadDunning, Abel Escriba, Tanisha Fazal, James Fearon, JoseFern andez-Albertos, Chris Haid, Paul Kenny, Matthew Kocher,Adam Lockyer, Jason Lyall, Shivaji Mukherjee, John Roemer,Dominic Rohner, Ian Shapiro, and Elisabeth Wood, as well as theOrder, Conflict, and Violence Group at Yale University, along withaudiences at several lectures, workshops, and conferences. We alsothank the 23 country experts who assisted us in our coding, and BenPasquale and Malika Rakhmankulova for their research assistance.We areparticularly grateful toArthur Steinand thethreeanonymous

APSR reviewers for their expertsuggestions and guidance. The usualcaveat applies.

is critical for understanding the evolution and transfor-mation of internal conflict.We identify the technology of rebellion as the

dimension best capturing the ways in which the in-ternational system has affected civil wars. A centralassumption in the literature is that civil war onset is afunction of structural factors that facilitate insurgency,a technology that can be deployed to serve all kinds ofpolitical ends (Fearon and Laitin 2003, 75). We showthat insurgency (guerrilla or irregular war) is nei-ther the only technology available to rebels nor is itas time invariant as assumed. In addition to irregularwarfare, we identify two overlooked technologies ofrebellion: conventional warfare and symmetric non-conventional (SNC) warfare. Although insurgency isan instance of asymmetric warfare, conventional andSNC wars are both forms of symmetric warfaretheformer militarily moresophisticated than the latter. Wefind that although irregular warfare is the dominanttechnology of rebellion between 1944 and 2004, it isjust barely so: it was used only in 54% of all civil wars.1

Furthermore, we find a major, but hitherto concealed,transformationof civil wars: 66%of all civil wars foughtduring the Cold War were irregular wars compared toonly 26% of those fought after 1991. Why?

During the Cold War, the two superpowers raisedthe military capacity of both states and rebels world-

wide. This mutual improvement in military capacitynevertheless favored the rebels, a result of the rise anddiffusion of a particularly robust version of the technol-ogy of insurgency. The end of the Cold War spelled thedecline of this technology. Of the states that had beenvulnerable to insurgency during the Cold War, manyexperienced civilpeace, whereas others became vulner-able to a different form of internal conflict, SNC war.

1 Like most cross-national research on civil wars, we focus on thepost-1944 period. The pre-1944 period includes a significant numberof irregular wars, many of which were wars of colonial conquestpitting modern against primitive armies, rather than civil wars.

415


2/15


3/15

American Political Science Review Vol. 104, No. 3

FIGURE 1. Civil Wars Starting, Ending, and Ongoing (19442004)

on the interaction of factors such as the role of su-perpowers in stoking or reducing conflict during theCold War; the depth of the Cold War competition; andthe regional penetration of superpowers. The complexand simultaneous occurrence of multiple, even contra-dictory, processes led seasoned observers to suspectthat the end of the Cold War could have indeterminateeffects and recommend caution in formulating predic-tions (Huntington 1993, 187; Jervis 1994, 76970).

What happened? Figure 1 tracks the total number ofcivil war onsets, terminations, and ongoing civil warsper year from 1944 to 2004.5 Ongoing civil wars in-

5 To build our data set, we relied on Sambanis (2001) data set ofcivil wars, extended to 2004 and with some modifications. This dataset is based on the standard definition of civil war, with the fol-lowing criteria: (1) there were more than 1,000 war-related deathsduring the entire war and in at least one single year of the war,(2) the war challenged the sovereignty of an internationally recog-nized state, (3) it occurred within the territory of that state, (4) thestate was one of the principal combatants, and (5) the rebels wereable to mount an organized military opposition to the state. Ourargument applies to conflicts that have already reached a certainlevel of intensity; at very low levels of intensity, conflicts fail toreach the level of a military contest. As a result, we use a dataset with a 1,000 battle deaths threshold rather than one with a muchlower threshold. For a full description, see the Appendix availableat http://www.journals.cambridge.org/psr2010002.

creased steadily after the late 1950s and peaked inthe early 1990s; civil war onsets also peaked in 1991.Immediately after 1991, the number of civil war on-sets declined, whereas terminations wentup. These twotrends converged to produce a decline in the numberof ongoing civil wars in the postCold War period.6

Two interpretations emerged. On the one hand, thespike of civil wars following the end of the Cold Warended the euphoria of the early 1990s and gave wayto frustration and disillusionment in the mid-1990s(Brown 1996, 11); it also popularized the view thatthe new era spelled a coming anarchy through theeruption of new wars (Kaldor 1999; Kaplan 1994).On the other hand, the subsequent emergence of adownward trend in both civil war onsets and ongoingcivil wars led scholars to speak of an extraordinaryand counterintuitive improvement in global security,arguing that the end of the Cold War was the singlemost critical factor in this decline (Human Security

6 Although our data extend only until 2004, the decline of civil waronsets has not been reversed in the 200510 period. For instance,only five civil wars were active worldwide in 2008: Sri Lanka (whichended in 2009), Afghanistan, Iraq, Pakistan, and Somalia. A sim-ilar downward trend can be discerned for lower-intensity conflicts(Harbom and Wallensteen 2009).

417


4/15

International System and Technologies of Rebellion August 2010

Centre 2005). The divergence between these two inter-pretations suggests the danger of extrapolating fromshort-term trends, an atheoretical exercise that hingeson the timing of the observations: the postCold Warera appeared to be a disaster in 1992, but struck ob-servers as a clear improvement by 2005.

In contrast, the cross-national literature on civil waronset incorporated the Cold War into the mix of vari-

ables examined and reached the conclusion that its endhad no impact on the likelihood of civil war onset (Col-lier et al. 2003; Fearon and Laitin 2003; Jung, Schlichte,and Seigelberg 2000; Sambanis 2004). In Fearon andLaitins (2003, 7778) words, the prevalence of civilwar in the 1990s was notdue tothe end of the ColdWarand associated changes in the international system.Unlike other findings that generated considerable con-troversy, such as the effects of natural resources orethnic fragmentation, the Cold War noneffect wasnot discussed, let alone questioned.

Yet, this finding raises more questions than it an-swers: we know that the end of the Cold War had someclear effects: (1) it fundamentally changed the role ofexternal actors in civil wars (Byman et al. 2001, xix); (2)it coincided with a regional outbreak of civil wars, espe-cially in Eurasia (Evangelista 1996; Zurcher 2007) andsub-Saharan Africa (Stedman 1996; Young 2006); (3) italso coincided with a surprising reduction of civil warsin Latin America (Castaneda 1993; Chernick 1996) andSoutheast Asia (Findlay 1996); and (4) many seeminglyintractable civil wars terminated with the end of theCold War (Hironaka 2005; Kanet 2006). We argue thatan exclusive and highly aggregated focus on civil waronset has prevented us from realizing how the end ofthe Cold War affected civil wars, and we suggest thatthe best way to grasp this effect is by examining how

civil wars are actually fought.

TECHNOLOGIES OF REBELLION

The Cold War raised the capacity of states worldwide,but it had a similar impact on rebel capacity. The twosuperpowers infused enormous military and economicassistance into allied states (Westad 2005); at the sametime, they also supported a wide range of rebel move-ments throughout the developing world. Although theUnited States supported rebels, such as the UNITAin Angola or the Contras in Nicaragua, it was muchmore common for the USSR to enter into alliances

with Third World rebels who followed some version ofa Marxist political agenda (Westad 1992, 461).

On average, rebels benefited more than states fromthis parallel rise of state and rebel capacity. This imbal-ance was reflected in the proliferation of civil wars thatrelied on the technology of irregular war or insurgency.More specifically, radical entrepreneurs who enjoyedthe support of the USSR and its allies transformedthe time-honored practice of guerrilla warfare into amuch more powerful form of warfare we call robustinsurgency.

Contrary to widespread belief, not all civil warsare insurgencies. When most people in the United

TABLE 1. Technologies of Rebellion inCivil War

Military Technologies of the State

High Low

Military

Technologiesof the Rebels

High Conventional [Successful

military coup]

Low Irregular Symmetricnonconventional

States speak of civil war, they automatically thinkof the American Civil War. This brings up images ofwell-organized, uniformed armies marching in closeformation in the midst of exploding shells. This imageobviously is very different from depictions of conflictsin Vietnam, Iraq, or Afghanistan. Veterans of theseconflicts are likely to evoke an invisible foe avoidingdirect contact, while constantly ambushing them orpoker-faced civilians with inscrutable loyalties. At thesametime, manyjournalists, development workers, andhumanitarian volunteers working in Africa today arelikely to think of an altogether different experienceof civil conflict, one populated by predatory militiaspreying on a defenseless civilian population in a con-text characterized by state collapse. That these threevignettes are not mutually exclusive suggests that thereis considerable heterogeneity in civil wars.

Following this intuition, we disaggregate civil warsbased on their technology of rebellion, which we con-ceptualize as the joint military technologies of statesand rebels engaged in armed conflict. Drawing on a

typology by Kalyvas (2005), we distinguish betweenthree technologies of rebellion emerging at the out-set of a civil war (Table 1).7 Conventional civil wartakes place when the military technologies of statesand rebels are matched at a high level; irregular civilwar emerges when the military technologies of therebels lag vis-a-vis those of the state; and SNC waris observed when the military technologies of statesand rebels are matched at a low level. The fourth cell,where the military technologies of the rebels outstripthe states, describes successful military coups ratherthan civil wars.

Irregular or guerrilla warfare is a technology ofrebellion whereby the rebels privilege small, lightlyarmed bands operating in rural areas (Fearon andLaitin 2003, 75)8; it is an expression of relative asymme-try between states and rebels.9 Rebels have the militarycapacity to challenge and harass the state, but lack thecapacity to confront it in a direct and frontal way. Putotherwise, states can mount a devastating response toa direct armed challenge such that the rebels only

7 By outset, we refer to the first year of the conflict.8 Irregular warfare is seldom relevant in interstate wars (Harkavyand Neuman 2001, 1819).9 Total asymmetry is reflected in the absence (orimmediatesuppres-sion) of an armed challenge.

418


5/15


option is to fight asymmetrically. Think of civil warsin El Salvador (197992), Peru (198096), and Nepal(19962006). In those wars, rebels tend to hover justbelow the military horizon, hiding and relying onharassment and surprise, stealth, and raiding (Simons1999); yet, they are frequently able to establish territo-rial control in peripheral areas. Ideally, and followingMaos recommendation, the rebels aim at launching a

major conventional attack after they have managed togrind down the states strength. In practice, irregularwars frequently turn into wars of attrition.

Conventional warfare emerges when rebels are ableto militarily confront states using heavy weaponrysuch as field artillery and armor. In conventional wars,military confrontation is direct, either across well-defined front lines or between armed columns; clashesoften take the form of set battles, trench warfare, andtown sieges (Balcells 2010). There is a clear distinctionbetween offensive and defensive actions and theemphasis onterritoryis alluded to in the use of the termpositional warfare (Lockyer2008,62).10 Classic casesinclude the American Civil War (186165) and the

Spanish Civil War (193639). More recent examplesinclude the Biafra conflict in Nigeria (196770), theAbkhazia conflict in Georgia (199294), the NagornoKarabach conflict in Azerbaijan (199194), and theCroatian and Bosnian wars in the former Yugoslavia(199295). These conflicts saw the deployment ofartillery and tanks in a landscape often dominated bytrenches. The battle of Cuito Cuanavale, which tookplace in Angola in September 1987 between the pro-Soviet MPLA government and South AfricanbackedUNITA rebels, entailed clashes between heavilyarmored columns and has been called the largestconventional land battle in Africa since World War II

(Chester 1992). Likewise, in Bosnia, virtually all ofthe fighting was done by professionally led, relativelywell-organized citizen armies, and the contrary viewis largely the product of mirror-imaging by Westernofficers who regularly disparaged the appearance,discipline, and professionalism of the armies involved.The myth of the so-called paramilitaries has persisted,although few, if any, major paramilitary units operatedafter 1992 (U.S. Central Intelligence Agency 2002,xiiv).

Last, some conflicts do not fit well into the irreg-ular war/conventional war dichotomy. They divergefrom irregular wars because there is no asymmetrybetween state and rebels; they also diverge from con-

ventional wars because the two sides are militarily low

10 A U.S. captain describes his experience in Iraq: The differencebetween the two deployments involved primarily the positioning ofthe enemy relative to ourselves. In the ground war, we had definitivelines of battle. Saddam Fedayeen elements did make things tricky,as they were running around in pick-up trucks and taxis wearingcivilian clothes, but we still knew generally the enemys territoryversus our own. Returning in 2004 with the insurgency in full swing,while driving around Mosul, we never were sure when and wherewe might be attacked. Some neighborhoods were definitively saferthan others, but there was no enemy zone versus friendly zone asmentioned earlier. Except for the U.S. bases, where mortar attackswere frequent but largely harmless, there was no place where onemight feel completely safe (Berschinski et al. 2007, 136).

tech.11 This is the case when states are unable (or, ina few cases, unwilling) to deploy an organized militaryagainst poorly equipped insurgents. This mutual weak-ness producesa type of warfare often describedas pre-modern (Earle 1997, 108) or primitive (Mueller2004), lending itself to comparisons with such conflictsas the Thirty Years War (Munkler 2005, 2; Prunier2009, 336). Often mistakenly described as guerrilla

wars, SNC wars tend to arise in contexts characterizedby extremely weak or collapsed states. Consider thecivil war in Congo-Brazzaville (199397): the electionsthat followed the end of the single-party, Soviet-typeregime produced inconclusive and contested results.The military effectively collapsed in 1992 and partymilitias (with names such as Ninjas, Cobras, Zoulous,etc.) assumed control of distinct areas of the capitalcity that were clearly delineated by checkpoints (calledbouchons). Even the president of the country relied ontwo militias alongside the scattered remnants of theold national army. By 1997, the armed actors involvedin this conflict included the leftovers of the old mili-tary, a new but very weak military, several militias withunclear chains of command, foreign mercenaries, theremnants of former president Mobutus presidentialguard from neighboring Zaire, elements of the Rwan-dan Interhamwe, and Angolan soldiers. The fightingwas conductedprimarily with small arms(Yengo 2006).Likewise, the United Nations described the conflict inSomalia as a situation where armed clashes tended totake the form of wild, chaotic exchanges of fire, featur-ing front-lines which could shift fifty or one hundredkilometers in a day as lines of defense disintegrated andregrouped. Supply lines were ad hoc to nonexistent, re-lying mostly on looting (United Nations DevelopmentOffice for Somalia 1998, 75).

COLD WAR AND CIVIL WARS: THEEMERGENCE OF ROBUST INSURGENCY

Although the term guerrilla (small war) was origi-nally coined to describe the Spanish resistance againstthe Napoleonic armies that invaded Spain in 1807, ir-regular war is as old as human history. In fact, muchof what is described as guerrilla warfare before thetwentieth century consists of instances of indigenousresistance against imperial encroachment, frequentlytaking the form of a frontal clash between vastly un-equal armies and often ending in slaughter. Robustinsurgency, however, is intertwined with the Cold War.

What became known as peoples war or revolu-tionary guerrilla warfare first emerged in the 1930s.It was honed by Mao Zedong in interwar China, testdriven by communist resistance movements in Eu-rope and Asia during World War II, and reached itsapex during the Cold War throughout the develop-ing world. Although sharing the same moniker with

11 It is possible to reduce this threefold distinction into a dichotomybetween asymmetricand symmetric civilwar, with the lattercategorycontaining both conventional and SNC wars. However, we believethat the two categories of conventional and SNC war capture a realand important difference.

419


6/15


traditional guerrilla war, this was a very different kindof war (Leites and Wolf 1970), a fact fully recognizedby counterinsurgency theorists (Galula 1964; Trinquier1964). As Beckett (2001, viii) explains, traditional guer-rilla warfare was generally understood as a purely mil-itary form of fighting using classic tactics of hit andrun and employed by indigenous groups where a con-ventional army either had been defeated or had never

existed. Rarely, he argues, did its primarily unsophisti-catedpractitionersdisplay any wider comprehension ofthe potential of irregular models of conflict in the waythat became commonplace after 1945, when guerrillawarfare became revolutionary and was termed in-surgency. In fact, this shift coincideswith a remarkablereversal in the outcomes of irregular wars: althoughstates routinely defeated irregulararmies before WorldWar II, this pattern was reversed afterward, with insur-gents increasingly more likely to force a draw ordefeat their stronger foes (Arregun-Toft 2005; Lyalland Wilson 2009).

We argue that robust insurgency is linked to theCold War through three channels: material support,

revolutionary beliefs, and military doctrine. First,whereas traditional guerrilla warfare depended on themobilization of local resources with the occasional sup-port of a neighboring state, robust insurgency benefitedfrom extensive and multifaceted superpower support.A central aim of Soviet foreign policy was to trainandmotivate, directly or throughsurrogates, insurgentsthroughout the developing world (Mott 2001; Westad2005). The initial beneficiaries included the Chineseand Greek Communists and the Third World becamea foreign policy priority in the early 1950s (Kanet2006, 334). China, Cuba, Libya, and the PalestinianLiberation Organization (PLO), among others, played

an important role in providing training and support.12

The concept of proxy war is a poor description ofSoviet policy because it only stresses the mechanicalinfusion of material resources into rebel movements;often, it implies a purely instrumental relationship be-tween opportunistic rebels who pretended to believein socialism in order to receive Soviet weapons. Al-though opportunism was certainly present, it did notexhaust the range of motivations, and although mate-rial support typically included the supply of weaponsand advisers, it extended to multiple forms of assis-tance, including political training and indoctrination(Dzhirkvelov 1989, 271). Most important, assistanceand support were channeled through transnational so-

cial movements. Thousands of radical activists builtsupraregional and even global contacts and networkswhile training in Soviet-funded military camps and uni-versities, the most famous of which was the PatriceLumumba Friendship University in Moscow.13 Thekey role played by this transnational social movement

12 For example, the civil war in South Yemen entailed the partic-ipation of Cubans, Syrians, PLO units, and some personnel fromEastern Europe (Kirkpatrick 1989, 8).13 The universitys first vice-rector and a number of its staff wereKGB officers whose objective was to recruit revolutionaries fromthe student body (Andrew and Mitrokhin 2005, 432). The HigherParty Schoolfor foreignersalso playeda keyrole in educatingradical

clearly distinguishes robust insurgency fromtraditionalguerrilla warfare. Whereas the latter was based onthe mobilization of primarily conservative, local senti-ments and/or local patronage tribal and kin networks,the former mobilized transnational revolutionary net-works often composed of educated and cosmopolitanindividuals; these would link up with traditional ruralnetworks but assume the leadership.14

Transnational networks were fed from, and in turnpropagated, revolutionary beliefs that constitute thesecond critical component of robust insurgency. Af-ter all, the Cold War was an ideological contest on aglobal level (Stein andLobell 1997, 109), with cognitiveframes and ideologies that aroused passionate ideo-logical commitmentsamong combatants, bothdomesti-cally and internationally (Hironaka 2005, 123).15 Thepower of these beliefs was well understood by coun-terinsurgents (Kirkpatrick 1989, 7; Olson 1989, 19) andis worth stressingbecause recentresearchhas tendedtodisregard the ideological Che Guevaras in favor of thepredatory Charles Taylors (Collier 2007; Hirshleifer2001).16 Beliefs are relevant in three ways. First, the

broad availability of a credible counterhegemonicmodel of political and social organization captured theimagination of millions. Specific ideasand labels varied,but the faith in the possibility, indeed inevitability, ofradical political transformation did not. Second, beliefsare important as sources of motivation for the crucialfirst movers willing to undertake high levels of riskand suffer enormous deprivation for the cause of rev-olution. Last, beliefs matter because they shape per-ceptions about the feasibility of radical change via thespecific path of armed struggle: subordinate or weakactors could successfully take on stronger actors pro-vided they learned how to deploy the technology of ro-

bust insurgency. Wolin (1973, 354) remarked how themilitary mode of thinking has all but supplanted thepolitical mode in revolutionary circles. Wherever one

leaders from around the world, in programs ranging from 2 years to2 months (Dzhirkvelov 1989, 271).14 This point was elaborated by Carl Schmitt ([1963] 2007, 30) whodistinguishes between two ideal types of irregular fighters: the tra-ditional defensive-autochthonous defenders of home and the ag-gressive international revolutionary activist. Modern revolutionaryguerrilla war, he argues, reached its fullest expression when it con-nected these two.15 We are referringto the beliefs of rebelleaders, cadres, and activists.Rank-and-file fighters and sympathizers were typically motivated

by a variety of heterogeneous concerns, of which ideology was notnecessarily the most important (Kalyvas 2006). Also, although somerebel entrepreneurs may have been keen to disguise narrow or op-portunistic goals under the cloak of socialist revolution in order togain accessto external support, many otherswere genuinely inspiredand empowered by these beliefs. This was notably true of severalnational liberation movements that blended nationalismwith bothMarxist ideology and revolutionary guerrilla principlessuch as theEritrean Popular Liberation Front (EPLF) and the Partiya KarkerenKurdistan (PKK) or Kurdistan Workers Party.16 An ironic testament to the pervasiveness of radical beliefs duringthe 1960s is the fact that Paul Collier (2007, ix) himself was temptedduring his youth: I was a student at Oxford in 1968, he recalls.I remember joining something called the Oxford RevolutionarySocialist Students, a name now beyond parody. But it all seemedsimple then.

420


7/15


turns. . .one finds sophisticated discussions of tactic,firepower, guerrilla warfare, and combat techniques.Indeed, leftist guerrilla movements used to host hun-dreds of activists and journalists from across the world,thus socializing them in the ways of armed struggle.

The last component of robust insurgency was mil-itary doctrine. What became known as revolution-ary or peoples war was an innovation whose

global breakthrough came about with the success ofthe Cuban Revolution, which put the guerrilla strat-egy on the worlds front pages (Hobsbawm 1996,438). The writings of Mao Zedong, Che Guevara,R egis Debray, and Amilcar Cabral, among others,were widely disseminated and read by thousands ofactivists and sympathizers in the developing world, es-pecially among the educated urban youth. They pro-vided a model of revolution taking off in the ruralperiphery and waged by peasant armies.17 The ex-amples of China, Cuba, and Vietnam appeared toconfirm that, despite occasional setbacks, guerrilla war-fare correctly waged was both a feasible and successfulpath to political and social change. From this perspec-

tive, irregular war was never a mere military tactic,akin to insurgent special forces storming their wayto power. Instead, rebel entrepreneurs learned that thekey to success lay in the patient construction of a highlystructured political organization, typically a party infirm control of a disciplined armed wing whose objec-tive was to acquire (liberate) and govern territory.On the one hand, organization guaranteed discipline inthe absence of which rebels could never hope to standagainst the states military superiority. On the otherhand, territory constituted a key resource for armedstruggle. Effective administration, mass mobilization,and sustained indoctrination in liberated areas (Eck

2010) were essential foundations for the developmentof armed struggle under conditions of military inferi-ority. Revolutionary state building (Kalyvas 2006) wasabsent, of course, in traditional guerrilla warfare.

In sum, massive material supportcombined withthespread of revolutionary beliefs andthe military doc-trine of revolutionary war during the Cold Wartrans-formed irregular war into robust insurgency, a credi-ble and much more effective technology of rebellion.The application of the military doctrine required highlymotivated individuals, their beliefs were sustained anddisseminated by examples of successful wars that re-lied on this doctrine, and both the dissemination ofbeliefs and the implementation of the doctrinerequired

17 Around these principles grew a rich global discussion about thebest way to organize, fight, and win. Proponents of Che Guevaras

foco theory emphasized the voluntaristic action of a party vanguardthat wouldcatalyze popular discontent through highly visible actionsfrom the periphery, supporters of Carlos Marighelastheories arguedin favor of urban guerrilla that would strike directly at the center,and the readers of Vo NguyenGi ap pointed to the long-termprocessof building a proficient insurgent military force. It is, perhaps, only aslight exaggeration to say that the military doctrine of revolutionaryguerrilla warfare achieved in its temporal context an importancereminiscent of present-day corporate management theories, namely,a thriving intellectual enterprise on a global scale with its specializedschools, international conferences, summer camps, gurus, and globalbest-selling books.

material assistance. Although it was possible for eachfactor to operate alone (e.g., some leftist insurgenciessucceeded in theabsence of external support), thecom-bination of all three contributed to raise rebel qualityduring the Cold War. In contrast, although the UnitedStates assisted rebels challenging pro-Soviet regimes,it primarily directed its support toward governmentsprofessing anticommunism. U.S. military assistance to

friendly regimes boomed, and the United States in-vested in the development of a counterinsurgency doc-trine closely tailored to match revolutionary guerrillawar (Leites and Wolf 1970). Military personnel frommany countries trained in the United States, and theSchool of the Americas in Fort Benning, Georgia, ac-quired a notorious reputation.18

Because the Cold War raised the capacity of bothstates and rebels, the asymmetry characterizing therelation of governments and rebels remained funda-mentally in place: states were still militarily superiorto rebels. However, the Cold War benefited rebelsmore than governments because it turned their deepweakness, which had prevented many budding rebel-

lions from reaching the stage of civil war, into relativeweakness, which allowed them to mount an effectivemilitary challenge against a stronger state, via robustinsurgency.

THE IMPACT OF THE END OF THECOLD WAR

The end of the Cold War is associated with (1) thedecline of rebel capacity, (2) the decline of state ca-pacity, and (3) the emergence of new postcommuniststates. These processes had distinct implications for thetechnologies of rebellion used in civil wars.

First, the end of the Cold War hurt rebels in a deci-sive way; it put an end to superpower competition, theUSSR itself, and the abundant provision of materialsupport to rebel forces across the world. Byman et al.(2001) document a dramatic shift in the sources of in-surgent support during the postCold War era towarddiasporas, refugees, and neighboring states, most ofwhich were as poor as the states facing insurgenciesand contributed little trainingall in all, a poor substi-tute for superpower support. Among rebels, the end ofthe Cold War hurt disproportionally the revolutionarytypes because it destroyed the belief in radical politicalchange (Przeworski 1991, 100) and the transnationalsocial movement that sustained it. Radical activistsacross the developing world awoke in a new worldwith their ideology shattered. This development fa-vored states.

But, second, the end of the Cold War also hurt states.With the Soviet threat gone, the United States lostinterest in propping up client states in the developingworld and divested itself from many weak states, thusweakening them further (Hale and Kienle 1997, 5).Obviously, things became really bad for Soviet client

18 Conversely, the Soviet Union also helped allied states in placessuch as Angola, Nicaragua, or Afghanistan.

421


8/15


FIGURE 2. Impact of the Cold War on Civil Wars

A B C D E

Robust Peace Irregular War Peace SNC Conventional

War

State

State State State & Rebels

Rebels State

Rebels Rebels Rebels

Cold War effect

Post-Cold War effect

states that, on top of losing massive aid, also saw theirlegitimizing principles melt away (Kanet 2006, 343).19

With superpower support reduced or gone, states hadto rely primarily on their domestic capacity. This wasa serious problem for several states whose domesticcapacity was notoriously wanting and had requiredenormous efforts to prop up in the first place, especiallythose located in sub-Saharan Africa (Clapham 1966;Herbst 2000, 2004; Reno 1999; Stedman 1996).20

These low-capacity states faced daunting prospectsas they became vulnerable to equally low-capacityrebels who were able to challenge them by foregoingthe painstaking process of organization, mobilization,and state building required by robust insurgency.Thus, our analysis suggests that the level of domesticcapacity on the eve of the postCold War perioddifferentiates states that became less vulnerable tocivil war onset from those that were more likelyto experience civil war, particularly in the form ofsymmetric nonconventional civil war.

Third, the end of the Cold War resulted in the disso-lution of states such as the USSR and Yugoslavia; thisprocess was accompanied by the division of existing

19Data from U.S. Agency for International Development (2009)

shows a decrease in total U.S. military assistance to third countriessince the early 1980s. The Stockholm International Peace ResearchInstitute estimates that world military expenditures dropped from$1.1 trillion in the late 1980s to $740 billion in 1997. There was also adrastic reduction in international arms sales: from 1986 to 1995, theyplummeted 55% (Stockholm International Peace Research Institute2008).20 TheSoviet Unionbecame actively involved in sub-SaharanAfrica,which came second after the Middle East in the volume of Sovietassistance received;during the 195688period, it received $23 billion(Mott 2001, 52). In 1974, there were approximately 3,600 Sovietadvisers in Somalia alone (Andrew and Mitrokhin 2005, 449). Suchaid may have been militarily effective in the short term but did notstrengthen weak states in the long term. In fact, Clapham (1996)argues the opposite in the case of African states.

armies into competing factions that could fight againsteach other conventionally. Hence, our expectation ofa rise in conventional civil wars in new postcommu-nist states. Figure 2 illustrates how the three processesactivated by end of the Cold War are associated withdistinct outcomes.21

Consider three capacity thresholds affecting bothstates and rebels. The first threshold marks a su-perior level of state capacity (roughly correspondingto that of advanced industrial societies): above thislevel, civil peace is robust (outcome A). The end ofthe Cold War makes no difference for these countries.The second threshold marks the level above whichstates or rebels are able to field a conventional armywith heavy artillery and armor. Below that threshold,states or rebels cannot field such an army: rebels areorganized irregularly, and states are considered to havefailed. Last, marks the threshold of rebel capacitybelow which armed nonstate actors are too weak tomount a sustained military challenge against a state;usually, this is the realm of bandits, mafias, and ter-rorists.

Civil war only takes place in the area below thresh-old , but its form varies depending on the militarytechnologies available to states and rebels. As dis-

cussed previously, the Cold War lifted rebel capacityabove , thus raising the likelihood of irregular war(outcome B). In contrast, the end of the Cold Warpushed many potential rebels below threshold , thussheltering many states from insurgency that had beenpreviously vulnerable to it (outcome C). Note here thatour argument identifies similar capacity states facing ahigher likelihood of either civil peace or civil conflict(outcomes B and C): this variation is driven exclusivelyby rebel capacity. At the same time, the end of the Cold

21 We are grateful to Referee 1 for his or her suggestions in thatrespect.

422


9/15


TABLE 2. Civil War Onsets, by Technology of Rebellion (19442004)

Cold War PostCold War Both Periods

N % N % N %Technology of Rebellion (194490) (194490) (19912004) (19912004) (19442004) (19442004)

Conventional 28 27.72 22 47.83 50 34.01Irregular 67 66.34 12 26.09 79 53.74Symmetric nonconventional 6 5.94 12 26.09 18 12.24Total 101 100 46 100 147 100

War causedstates whose residual capacity waslow (andhad been previously propped up by superpower assis-tance) to fall below. Rebels located belowwere nowable to challenge these weakened states by means of aSNC war (outcome D). Last, the dissolution of somestates following the end of the Cold War allowed rebelsto move above threshold and made a conventionalchallenge possible (outcome E).

In short, our analysis producesa clear setof empiricalpredictions about the impact of the Cold War and itsend on civil war, while acknowledging multiple path-ways. Next, we incorporate these theoretical insightsinto a cross-national analysis.

EMPIRICS

Our first empirical task is to ascertain whether therehas been a significant shift in technologies of rebellionfollowing the end of the Cold War. In line with ourtheoretical conceptualization, we operationalize thesetechnologies with the type of weaponry used by thecontending armed actors during the first year of the

conflict.Weuseadatasetof147civilwarsbetween1944and 2004. We coded as conventional wars those con-flictswhere bothincumbents and insurgents usedheavyweaponry (artillery and armor); as irregular wars, theconflicts where incumbents deployed heavy weaponrybut insurgents relied on light weapons; and as SNCwars, the conflicts where neither incumbents nor insur-gents used heavy weaponry.22

In Table 2, we can observe that irregular war is thedominant technology of rebellion during the entire pe-riod 19442004, thus justifying the scholarly attentionithas received so far. However, it accounts for just morethan half of all civil wars (53.74%): conventional warsare much more common than generally thought (34%),and SNC wars account for slightly more than 12% ofthe total. When we partition the data in two distinctperiods, the Cold War (194490) and the postColdWar (19912004),23 we find that the end of the ColdWar is associated with an important shift: although

22 The coding protocol is described in the Appendix.23 We establish 1991 as the cutoff year because it corresponds tothe dissolution of the Soviet Union and the emergence of severalnew states. We have estimated our regression models with differentdummies for the end of the Cold War (exploring cutoff points from1985 to 2000) and found that the critical year for conventional andirregular war is 1991, and the critical year for SNC is 1989 (these are

irregularwar dominates the ColdWar period (66.34%),it is much less frequent after its end (26.09%). Conven-tional war becomes the dominant type of civil war after1990 (47.83%), and SNC wars experience considerablegrowth, rising to the level of irregular war (26.09%).24

The decline of irregular war following the end of theCold War is striking (Fig. 3), a trend that is robust tonormalization (i.e., onsets by year).

Geographic and temporal patterns appear consistentwith our argument. With the end of the Cold War, civilwars shifted away from Asia and Latin America andtoward Eurasia, sub-Saharan Africa, and, to a lesserdegree, the Middle East and North Africa (MENA).After 1990, irregular war declined steeply in both Asiaand Latin America.25 In contrast, Eurasia experienceda rise of conventional civil wars linked to processesof state dissolution and new state formation. In sub-Saharan Africa, the most remarkable postCold Wartrend is the explosion of SNC wars; the abrupt inter-ruption of superpower assistance to low-capacity statesdegraded their ability to deter even poorly organizedrebels.26 Last, the MENA region diverges from these

trends in that both conventional and irregular warsexperienced a rise in the postCold War period, de-spite the relatively high capacity of states there. Theresilience of irregular war in that region can be tracedto the emergence of militant Islamism and the U.S.invasions of Iraq and Afghanistan.

We now turn to the determinants of each of the threetechnologies of rebellion, using a multinomial logit es-timation to examine the likelihood that a specific tech-nology of rebellion is used in a civil war, given its onset.The dependent variable is categorical (Technology ofRebellion), taking the value of 1 for conventional wars,2 for irregular wars, and 3 for SNC wars. Our mainexplanatory variable is a dummy differentiating thetwo periods under consideration (Post 1990), which we

the years for which the dummy has the greatest substantive effect onthe occurrence of these types of wars).24 Setting the cutoff year in 1989 does not change the picture. Dur-ing the 194488 period, 66.30% of all civil wars were irregular,29.35% conventional, and 4.35% SNC. During the 19892004 pe-riod, 32.73% were irregular, 41.82% conventional, and 25.45% SNC.25 Castaneda (1993) indeed notes that in Latin America the era ofarmed politics and guerrilla insurgencies ended with the Cold War.26 Indeed, Stedman (1996, 236) points out that these conflicts areclosely connected with the end of the Cold War, which underminedthe external sources of support for Africas patrimonial regimes andleft some with no legs to stand on.

423


10/15


FIGURE 3. Irregular Civil Wars as a Proportion of All Civil Wars, by Decade

expect to be positively associated with conventionaland SNC civil wars, and negatively associated with ir-

regular civil wars.27 We also use two additional speci-fications of the explanatory variable to better capturethe mechanisms at work: the first is a dummy variablemarking new countries emerging from a former com-munist state (New Postcommunist);28 we expect it tobe positively associated with conventional civil warsbecause these processes of state partition tended tobequeath large military arsenals to rival factions. Thesecond one is a dummy variable indicating whether amajor rebel actor claimed to be guided by a Marxist-leaning agenda (Marxist Rebels); civil wars featuringsuch rebels were naturally much more common dur-ing the Cold War, but were far from the only ones in

that period.29

This variable should be associated withirregular war given our posited link between Marx-ist inclinations and robust insurgency. We use these

27 Post 1990 takes value 1 for all years between 1991 and 2004 (in-cluding both these years) and value 0 for the remaining years.28 We code as such all countries that emerged from a communiststate, not necessarily in 1991, but always after 1990. The corollaryhere is that these new states result from processes of state parti-tion that entail the partition of their armed forces. For the list ofcases, see Table A4 of the supplemental Appendix available onlineat http://www.journals.cambridge.org/psr2010002.29 Of all civil wars that took place before 1991 (a total of 101), 33(32.67%) involved Marxist rebels.

three variables in separate models to avoid collinearityissues.

We also include an additional set of independent andcontrol variables. Gross domestic product (GDP) perCapita isageneralproxyofstatecapacitythatshouldbepositively associated with conventional and irregularwars, and negatively associated with SNC wars.30 GDPper capita is a problematic proxy for state capacity(Collier et al. 2003; Hegre et al. 2001), but it is standardpractice to include it in civil war models. As a way tocorrect for this problem, we include Military Personnel,a much betterproxy forthe capacity of a states military,taken from COW 3.02 (Singer, Bremer, and Stuckey1972), in thousands, lagged 1 year. Again, we expectthis variable to be positively associated with conven-

tional and irregular wars, and negatively associatedwith SNC wars. We also include Rough Terrain, whichshould have a positive effecton irregular warcomparedto the other two types because mountainous terrainhas been claimed to favor this technology of warfare

30 In our main set of regressions, we use Fearon and Laitins (2003)lagged measure of GDP (log of GDP per capita in thousands of1985 U.S. dollars, in World Bank data), but we also run a set ofrobustness tests(availableon request)with twoalternativemeasures:lagged constant 2000 U.S. dollars, also from the World Bank (2007);and lagged value of current international dollars, from Penn WorldTable 6.1 (Heston, Summers, and Aten 2006).

424


11/15


TABLE 3. Multinomial Logit Regressions on Technology of Rebellion (19442004)

M1 M2 M3 M4 M5 M6

Conventional

Rough terrain 0.004 0.003 0.009 0.003 0.003 0.005(0.01) (0.00) (0.01) (0.00) (0.00) (0.01)

Ethnic war 0.172 0.123

0.417

0.099

0.044

0.553(0.45) (0.43) (0.53) (0.46) (0.45) (0.55)GDP per capita 0.039 0.020 0.235 0.001 0.014 0.147

(0.15) (0.17) (0.14) (0.16) (0.18) (0.16)Military personnel 0.000 0.000 0.000

(0.00) (0.00) (0.00)Post 1990 1.422 1.090

(0.49) (0.53)New postcommunist 2.579 1.313

(1.15) (1.39)Marxist rebels 1.631 1.396

(0.64) (0.67)Constant 1.111 0.771 0.408 0.763 0.607 0.125

(0.46) (0.42) (0.50) (0.44) (0.44) (0.53)

SNC

Rough terrain 0.025 0.034 0.030 0.026 0.028 0.025(0.02) (0.02) (0.02) (0.02) (0.01) (0.02)

Ethnic war 0.245 0.277 1.205 0.611 0.086 0.705(0.61) (0.56) (0.62) (0.86) (0.59) (0.65)

GDP per capita 0.468 0.281 0.167 0.420 0.080 0.256(0.25) (0.22) (0.22) (0.43) (0.31) (0.32)

Military personnel 0.031 0.017 0.018

(0.01) (0.01) (0.01)Post 1990 2.756 3.402

(0.68) (0.81)New postcommunist 31.012 29.747

(1.28) (1.80)Marxist rebels 36.304 41.206

(0.50) (0.54)Constant 1.218 0.359 0.530 1.886 0.076 0.579

(0.53) (0.65) (0.67) (0.99) (0.68) (0.65)

Observations (N) 137 137 137 124 124 124Wald chi2 22.73 4138.27 10004.37 24.59 1186.21 10047.54Pseudo R2 0.1306 0.0754 0.1245 0.1774 0.0897 0.1572

p< .05; p< .01; p< .001.

(Fearon and Laitin 2003).31 Ethnic War is included toaccount for potential differences in warfare driven byethnic motivations, following Kaufmanns (1996) argu-ment that ethnic civil wars are more likely to be foughtconventionally compared to nonethnic civil wars, andKaldors (1999) conjecture that postCold War civilwars are both more likely to be motivated by ethnicanimosities and display features characteristic of SNC

31 We include Fearon and Laitins (2003) measure of Rough Terrain,which is the log of estimated percent of mountainous terrain in acountry. It must be noted, however, that this variable captures theproportion of a country that is mountainous, yet insurgencies canemerge in swamps, jungles, and other geographic contexts (81).

wars. Our theoretical prior is that technologies of re-bellion should be independent of ethnic motivations.32

The results of the multinomial regressions are pre-

sented in Table 3, which displays the estimated coef-ficients for conventional and SNC wars; the referencecategory is irregular war. We also ran a second set of

32 We include ethnic war as a dummy variable: 1 if the civil war isethnic, 0 if not, as coded by Sambanis (2001).Thisauthor defineseth-nic war as that taking place between communities (ethnicities) whoare in conflict over the power relationship that exists between thosecommunities and the state. He codes as ethnic civil war episodes ofviolent conflict between governments and national, ethnic, religious,or other communal minorities (ethnic challengers) in which the chal-lengers seek major changes in their status. . ..Rioting and warfarebetween rival communal groupsis notcodedas ethnicwarfare unlessit involves conflict over political power or government policy (67).

425


12/15


models that add Military Personnel to the first threemodels.33

The results of model 1 reveal a strong and signifi-cant robust effect of the end of the Cold War (proxiedwith Post 1990) on technologies of rebellion, in the ex-pected directionshowing that the descriptive patternsare robust to a multivariate regression specification.34

Accordingto this model, andholdingall other variables

at their sample mean, the probability that a civil waris fought conventionally increases from 30.17% duringthe Cold War to 51.13% after it; the probability that acivil war is fought irregularly decreases from 66.45%to 27.15%; and the probability that a civil war is SNCincreases from 3.38% to 21.72%.35 In model 2, andconsistent with our expectations, New Postcommunistis highly significant, substantively and statistically, inaccounting for the likelihood of conventional vis-a-visirregularcivilwars. Everything else equal, the probabil-ity of a civil war being fought conventionally is 87.63%if it takes place in a new postcommunist state; thisprobability is quite lower (32.57%) if the civil war takesplace in any other country. This variable also displays a

strong negative effect on the likelihood of SNC wars.36In model 5, this variable loses statistical significancebecause of missing cases.37 Last, in model 3, MarxistRebels displays a significant negative effect on bothconventional and SNC wars.38 With all other variablesat their sample mean, the probability of a civil warbeing irregular if the insurgents have a Marxist-leaningagenda is 84%, almost twice as high compared to in-surgents who lack a Marxist agenda (46%). This resultis consistent with our argument that a key mechanismexplaining the dominance of irregular wars during theCold War is the Marxist character of rebels.39

GDP per capita remains not statistically significant

across all specifications. Contrary to our expectations,irregular civil wars are no more likely in wealthiercountries. We have already mentioned the problemsassociated with using GDP per capita as proxy for

33 We decided to run a separate set of regressions because data onMilitary Personnel is missing in 18 cases.34 Note that we lose 10 observations due to missing data on the vari-able GDP per capita. Of these cases, 4 correspond to conventional,5 to irregular, and 1 to SNC civil wars; 2 are Cold War observationsand 8 are postCold War ones. We ran the analyses excluding GDPper capita (thus, with 147 cases in models 1, 2, and 3), and the coeffi-cients for the key variables remain consistent, although one variablebecomes significant for SNC: Rough Terrain, which takes a negativesign.35

These probabilities are consistent with the descriptive patterns inTable 3. We obtain similar results in model 4, which includes MilitaryPersonnel as an independent variable. Also, if we pool together thetwo symmetric types (SNC and conventional), we again observe thatthe end of the Cold War has a positive effect on symmetric civil warsvis-a-vis irregular wars.36 The large size of this coefficient is explained by the fact that ofall civil wars that take place in New Postcommunist states (a total of12), 10 are conventional, 2 are irregular, and none is SNC.37 Nine out of the 12 cases of new postcommunist states are missingin this regression because of lack of data on Military Personnel, onGDP per capita, or on both variables.38 The large size of the coefficient of this variable for SNC is ex-plained by the fact that there are no SNC wars fought by Marxistinsurgents.39 This result holds in model 6.

state capacity; moreover, given the robust correlationbetween GDP per capita and civil war onset, our dataset contains primarily poor states. However, we findinstead that Military Personnel has a negative andsignificant effect on SNC wars, confirming that thesewars entail states with lower military capacity. The non-significance of this variable for conventional civil warsis consistent with our conjecture that states fighting

conventional wars have a military capacity comparableto those fighting irregular wars. Finally, Rough Terrainand Ethnic War are not significant in any of the models.This suggests that terrain and ethnic conflict are notassociated with a particular technology of rebellion.

Overall, the empirical analysis supports our theo-retical expectations. The descriptive data show clearlythat the end of the Cold War was a key turning pointfor civil wars: it is then that irregular war ceased tobe the dominant technology of rebellion. In turn, thisshift was accompanied by a change in the geographicdistribution of civil war. Our interpretation stresses thedegree to which many states were able to withstandthe reduction of foreign assistance that accompanied

the end of the Cold War by drawing on their own re-sources. Sub-Saharan African states were clearly themost affected in this respect, as indicated by the rise ofSNC wars in that region. The overall significance of thecoefficients of our three measures capturing the shiftin the international system (Post 1990, New Postcom-munist, and Marxist Rebels) points to the pathwaysthrough which the end of the Cold War affected inter-nal conflict. The multivariate regression specificationallows us to confirm that the descriptive trends arerobust to the inclusion of control variables.40

Finally, although our primary focus is on civil waronsets, we also examined the evolution of all irregu-

lar civil wars that were going on when the Cold Warended. We find that most of these wars were affectedby the end of the Cold War. First, four irregular civilwars terminated as a direct result of the end of theCold War (the Marxist insurgencies in El Salvadorand Guatemala and the anti-Marxist insurgencies inNicaragua and Mozambique). With ideological utopiasover and foreign subsidies gone, both sides found itmore beneficial to bring these conflicts to a negotiatedending (Hironaka 2005, 126). Second, three conflictssurvived the end of the Cold War, but switched awayfrom irregular war and toward either conventional orSNC warfare (Afghanistan, Angola, and Somalia).41

Third, three irregular wars persisted. These were all

40 The results hold when GDP per capita and other nonsignificantvariables are dropped from the two sets of equations. Also, in a setof additional analyses, we use panel data and find that our resultsare robust to a country/year specification. Using Fearon and Laitins(2003) data set, we replicate their analysis and find that the postCold War dummy has no impact on civil war onset, as they report.However, once we disaggregate civil wars by technologies of rebel-lion, this variable is significantlyassociated with a decline of irregularcivil war onsets and a rise of conventional civil war onsets.41 In Afghanistan, the Islamic resistance switched to conventionalwar against the weakened pro-Soviet government; after that govern-ment collapsed, the resistance split and engaged in an internecineconventional war, before the Taliban launched their own, largelysuccessful, conventional assault. In Angola, the collapse of a peace

426


13/15


Marxist insurgencies that relied primarily on domesticresources, in Cambodia, Peru, and Colombia. Becauseof their autonomy from the international system, theserebels were able to withstand the shock of the end ofthe Cold War, at least for some time (they were even-tually defeated in Peru and Cambodia). The Colom-bian FARC is the exceptional case that corroboratesour theoretical argument: a unique Marxist group that

survived the end of the Cold War having been indepen-dent from it in the first place. Because of its autonomyfrom the international system, the FARC managed tomaintain its ideological identity, type of recruitment,and technology of rebellion. Despite having to resortto narco-trafficking for its financing, it still does notpay its soldiers a salary, relying instead on politicalindoctrination (Guti errez Sann 2004).

Although irregular war is no longer dominant, it hasnot disappeared. An examination of postCold Warirregular conflicts suggests that they come in two ma-jor types. The first one consists of minor, peripheralwars, which Fearon (2004) describes as sons of thesoil insurgencies (e.g., Aceh in Indonesia, southern

Thailand, Cabinda in Angola); these rebellions do notthreaten power at the center and seem to be containedor resolved without major international repercussions.The end of the Cold War has certainly affected theability of insurgent entrepreneurs to escalate themby linking them to global politics. The second typeconsists of insurgencies with a radical Islamist out-look (e.g., Egypt, Algeria, Iraq, Chechnya, Tajikistan,Afghanistan), which cluster in the Middle East, Cen-tral Asia, and North Africa. These insurgencies displaysome interesting parallels with the Marxist insurgen-cies of the Cold War that require further exploration.Both these types of irregular war, however, fail to fill

the gap left by the decline of robust insurgency.

CONCLUSION

Our analysis demonstrates that the end of the ColdWar has had a transformative impact on the way civilwars are fought. We show that despite being domesticconflicts, civil wars are shaped in nonobvious, yet de-cisive ways by the international system. In focusing onhow civil wars are fought, we also illustrate the impor-tance of moving beyond a single-minded focus on civilwar onset and taking the logic of warfare seriously. Wepoint to a source of systematic heterogeneity in civilwars, specify three distinct technologies of rebellion,and identify a striking decline in one of them, irregularwar, following the end of the Cold War. Possibly, thisimplies the future prevalence of civil wars that are notonly shorter, but also more intense and more openended in terms of their outcome.

Civil wars are military contests where each sidesmilitary capacity shapes the type of military interaction

agreement emboldened the rebels to launch an initially successfulconventional assault against a government that had lost its Sovietpatronage, conquering several cities. Last, in Somalia, the nationalarmycollapsedfollowingthe terminationof U.S. funding,and the warwas transformed into a factional conflict between several militias.

and, therefore, the nature of the conflict. Both insur-gent and counterinsurgent strategies vary accordingly,and yet their lessons are conditional on the prevail-ing technology of rebellion. For example, the combinedexperience of Iraq andAfghanistan hasled theU.S. mil-itary to focus single mindedly on irregular war. How-ever, the lessons from Afghanistan are not necessarilytransferable to an SNC conflict such as the Somali one.

Our analysis also implies that, as they consider peace-keeping and peace building operations, policy makersmust be aware of the variation in technologies of rebel-lion, as well as the transformation of internal conflictafter the end of the Cold War. For instance, neitherconventional nor SNC civil wars correspond to thepopular image of quagmire associated with irregularwars, which have deterred international interventionin the past.

This article helps bridge the current gap in the civilwar literature between two distinct research programs:one focusing on the macro, cross-national level, andanother one privileging the micro, subnational level.We show how insights generated at the micro-level

can fertilize cross-national, macro-level models. Ourfindings reinforce the call for theoretical and empiricaldisaggregation (Wimmer, Cederman, and Min 2009),place scope conditions on existing theories of rebellionthat emphasize exclusively state capacity or general-ize insights drawn from a single technology of warfare(most commonly Sub-Saharan SNC wars), challengethe equation of civil war with insurgencya centralassumption in a major theory of civil war onsetandredirect thetheoreticalfocus on therole of multidimen-sional external support (as opposed to mere financing),beliefs, and military doctrine and practice. We suggestthat a fuller understanding of both civil war onset and

dynamics calls for a more refined theorizing of warfare,including the relationship between state and rebel ca-pacity.

Finally, we stress the importance of the internationalsystem, suggesting that just as domestic economies areclosely interacting with global processes, a similar logicholds for domestic security dynamics. Although re-search on civil wars has recently turned its sight to theinternational dimension of civil wars, including the roleof neighborhood contagion (Buhaug and Gleditsch2008; Hegre and Sambanis 2006), refugee movements(Salehyan and Gleditsch 2006), and transnational dias-poras (Collier and Hoeffler 2004), it has been surpris-ingly neglectful of the international system. By spec-

ifying three distinct technologies of rebellion and byidentifying a major and overlooked process of trans-formation of civil wars, we are able to theorize the linkbetween system polarity, the Cold War, and internalconflict, as well as provide empirical support for it.The way in which civil wars are waged turns out to beclearly related to the international system in ways thatare more obvious (e.g., superpower interference) orless (e.g., the decline of irregular war). The prevalenceof irregular war as a means of waging civil wars turnsout to be a phenomenon closely associated with theCold War. Conversely, SNC wars are associated withprocesses of superpower withdrawal from weak states

427


14/15


following the end of the Cold War, whereas conven-tional war are associated with processes of imperialcollapse and state formation.

Overall, our article stresses the need to connectthe complex conflict processes taking place at thesubnational, national, transnational, and international-systemic level. Students of internal conflict can prof-itably recognize that just because they are domestic

conflicts, civil wars are no less immune to the effects ofthe international system than interstate wars.

REFERENCES

Andrew, Christopher, and Vassili Mitrokhin. 2005. The World WasGoing OurWay: TheKGB andthe Battlefor the Third World. NewYork: Basic Books.

Arregun-Toft, Ivan. 2005. How the Weak Win Wars: A Theory ofAsymmetric Conflict. Cambridge: Cambridge University Press.

Ayres, R. William.2000. A WorldFlying Apart? Violent NationalistConflict and the End of the Cold War. Journal of Peace Research37 (1): 10517.

Balcells, Laia. 2010. Behind the Frontlines: Identity, Competition, andViolence in Civil Wars. Ph.D. diss. Yale University.

Beckett, Ian F. W. 2001. Modern Insurgencies and Counter-Insurgencies: Guerrillas and Their Opponents since 1750. NewYork: Routledge.

Berschinski, Robert, Josh Bradley, John Frick, and Bryan Groves.2007. Back from Battle: Student Veterans Perspectives on theIraq War. Yale Journal of International Affairs 2 (2): 13442.

Brown, Michael E. 1996. Introduction. In The International Di-mensions of Internal Conflict, ed. Michael E. Brown. Cambridge,MA: MIT Press, 131.

Buhaug, Halvard, and Kristian Skrede Gleditsch. 2008. Contagionor Confusion? WhyConflictsCluster in Space.InternationalStud-ies Quarterly 52 (2): 21533.

Byman, Daniel, Peter Chalk, Bruce Hoffman, William Rosenau, andDavid Brannan. 2001. Trends in Outside Support for InsurgentMovements. Santa Monica, CA: Rand.

Castaneda, Jorge. 1993. Utopia Unarmed: The Latin American Leftafter the Cold War. New York: Knopf.

Chernick, Marc W. 1996. Peacemakingand Violence in Latin Amer-ica. In The International Dimensions of Internal Conflict, ed.Michael E. Brown. Cambridge, MA: MIT Press, 267307.

Chester, Crocker A. 1992. High Noon in Southern Africa: MakingPeace in a Rough Neighborhood. New York: W.W. Norton.

Clapham, Christopher, S. 1996. Africa and the International System:The Politics of State Survival. Cambridge: Cambridge UniversityPress.

Collier, Paul. 2007. The Bottom Billion: Why the Poorest CountriesAre Failing and What Can Be Done About It. Oxford: OxfordUniversity Press.

Collier, Paul, V. L. Elliott, Havard Hegre, Anke Hoeffler, MartaReynal-Querol, and Nicholas Sambanis. 2003. Breaking the Con-

flict Trap: Civil War and Development Policy. Washington, DC:World Bank and Oxford University Press.

Collier, Paul, and Anke Hoeffler. 2004. Greed and Grievance inCivil War. Oxford Economic Papers 56 (4): 56395.

Dzhirkvelov, Ilya. 1989. Political and Psychological Operations inthe Soviet Promotion of National Liberation Movements. InGuerrilla Warfare and Counterinsurgency: U.S.Soviet Policy inthe Third World, eds. Richard H. Shultz, Jr., Robert L. Pfaltzgraff,Jr., Uri Raanan, William J. Olson, and Igor Lukes. Lexington,MA: Lexington Books, 26975.

Earle, Timothy. 1997.HowChiefs Cometo Power:The Political Econ-omy in Prehistory. Stanford, CA: Stanford University Press.

Eck, Kristine. 2010. Raising Rebels: Participation and Recruitment inCivil War. Ph.D. diss. Uppsala University.

Ellingsen, Tanja. 2000. Colorful Community or Ethnic WitchesBrew? Multiethnicity and Domestic Conflict during and after theCold War. Journal of Conflict Resolution 44 (2): 22849.

Evangelista, Matthew. 1996. Historical Legacies and the Politics ofIntervention in the FormerSoviet Union. In The InternationalDi-

mensions of Internal Conflict, ed. Michael E. Brown. Cambridge,MA: MIT University Press, 10740.

Fearon, James. 2004. Why Do Some Civil Wars Last So MuchLonger Than Others? Journal of Peace Research 41 (3): 275302.

Fearon, James, and David Laitin. 2003. Ethnicity, Insurgency andCivil War. American Political Science Review 97 (1): 7586.

Findlay, Trevor. 1996. Turning the Corner in Southeast Asia. InThe International Dimensions of Internal Conflict, ed. Michael E.Brown. Cambridge, MA: MIT University Press, 173204.

Gaddis, John Lewis. 1997. We Now Know: Rethinking Cold War

History. Oxford: Clarendon.Galula, David. 1964. Counterinsurgency Warfare; Theory and Prac-

tice. New York: Praeger.Goldgeier, James M., and Michael McFaul. 1992. A Tale of Two

Worlds: Core and Periphery in the PostCold War Era. Interna-tional Organization 46 (2): 46791.

Guti errez Sann, Francisco. 2004. Criminal Rebels? A Discussionof Civil War and Criminality from the Colombian Experience.Politics & Society 32 (2): 25785.

Hale, William, and Eberhard Kienle. 1997. Introduction. In Af-ter the Cold War: Security and Democracy in Africa and Asia,eds. William Hale and Eberhard Kienle. London: I.B.Tauris,112.

Harbom, Lotta,and PeterWallensteen.2009. ArmedConflicts, 19462008. Journal of Peace Research 46 (4): 57787.

Harkavy, Robert E., and Stephanie G. Neuman. 2001. Warfare andthe Third World. New York: Palgrave.

Hegre, Havard, Tanja Ellingsen, Scott Gates, and Nils P. Gleditsch.2001. Toward a Democratic Civil Peace? Democracy, PoliticalChange, and Civil War, 18161992. American Political ScienceReview 95 (1): 3348.

Hegre, Havard, and Nicholas Sambanis. 2006. Sensitivity Analysisof Empirical Results on Civil War Onset. Journal of ConflictResolution 50 (4): 50835.

Herbst, Jeffrey. 2000. States and Power in Africa: ComparativeLessons in Authority and Control. Princeton, NJ: Princeton Uni-versity Press.

Herbst, Jeffrey.2004. African Militariesand Rebellion:The PoliticalEconomy of Threat and Combat Effectiveness. Journal of PeaceResearch 41 (3): 35769.

Heston, Alan, RobertSummers, andBettina Aten. 2006. Penn WorldTable Version 6.2. Center for International Comparisons of Pro-duction, Income and Prices at the University of Pennsylvania (CI-

CUP). September 2006. http://pwt.econ.upenn.edu (Accessed July19, 2010).

Hironaka, Ann. 2005. Neverending Wars: The International Commu-nity, Weak States, and the Perpetuation of Civil War. Cambridge,MA: Harvard University Press.

Hirshleifer, Jack. 2001. The Dark Side of the Force: Economic Foun-dations of Conflict Theory. Cambridge: Cambridge UniversityPress.

Hobsbawm, Eric. 1996. The Age of the Extremes: A History of theThird World, 19141991. New York: Vintage.

Human Security Centre, The University of British Columbia,Canada. 2005. Human Security Report 2005: War and Peacein the 21st Century. New York: Oxford University Press.www.humansecurityreport.info (Accessed July 19, 2010).

Huntington, Samuel P. 1993. If Not Civilizations, What? Paradigmsof the PostCold War World. Foreign Affairs 72 (5): 18694.

Jervis, Robert. 1994. Leadership, PostCold War Politics, and Psy-chology. Political Psychology 15 (4): 76977.

Jung, Dietrich, Klaus Schlichte, and Jens Seigelberg. 2000. After-word: Warfare in 1995. In Warfare since the Second World War,eds. Klaus Jurgen Gantzel and Torsten Schwinghammer. NewBrunswick, NJ: Transaction, 16979.

Kaldor, Mary. 1999. New and Old Wars: Organized Violence in aGlobal Era. Cambridge: Polity.

Kalyvas, Stathis N. 2005. Warfare in Civil Wars. In Rethinkingthe Nature of War, eds. Isabelle Duyvesteyn and Jan Angstrom.Abingdon: Frank Cass, 88108.

Kalyvas, Stathis N. 2006. The Logic of Violence in Civil War. NewYork: Cambridge University Press.

Kalyvas, Stathis N., and Laia Balcells. 2009. Consequences of War-fare in Civil Wars: An Empirical Analysis. Presented at the

428


15/15


Annual Meeting of the American Political Science Association,Toronto.

Kanet,RogerE. 2006. The Superpower Quest forEmpire:The ColdWar and Soviet Support for Wars of National Liberation. ColdWar History 6 (3): 33152.

Kaplan, Robert D. 1994. The Coming Anarchy: How Scarcity,Crime, Overpopulation, and Disease Are Rapidly Destroying theSocial Fabric of Our Planet. Atlantic Monthly 273 (2): 4476.

Kaufmann, Chaim. 1996. Intervention in Ethnic and IdeologicalCivil Wars: Why One CanBe Done and the Other Cant. Security

Studies 6 (1): 62100.Kirkpatrick, Jeanne. 1989. Protracted Conflict and U.S. Policy. In

Guerrilla Warfare and Counterinsurgency.U.S.Soviet Policy in theThird World, eds. Richard H. Shultz, Jr., Robert L. Pfaltzgraff, Jr.,Uri Raanan, William J. Olson, and Igor Lukes. Lexington, MA:Lexington Books, 511.

Laidi, Zaki. 1994. Rethinking PostCold War. Economic and Po-litical Weekly 29 (32): 206769.

Lake, David A., and Donald Rothchild. 1996. Containing Fear:The Origins and Management of Ethnic Conflict. InternationalSecurity 21 (2): 4175.

Lebow, Richard Ned. 1994. The Long Peace, the End of the ColdWar, and the Failure of Realism. International Organization48 (2): 24977.

Leites, Nathan, and Charles Wolf, Jr. 1970. Rebellion and Authority:An Analytic Essay on Insurgent Conflicts. Chicago: Markham.

Lockyer, Adam. 2008. Foreign Intervention and Warfare in CivilWars: TheEffect of Exogenous Resourceson the Course andNatureof the Angolan and Afghan Conflicts. Ph.D. diss. University ofSydney.

Lyall, Jason,and Isaiah Wilson III.2009. Rage against the Machines:Explaining Outcomes in Counterinsurgency Wars. InternationalOrganization 63: 67106.

Mearsheimer, John J. 1990. Backto the Future: Instabilityin Europeafter the Cold War. International Security 15 (1): 556.

Mott, William H., IV. 2001. Soviet Military Assistance: An EmpiricalPerspective. Westport, CT: Greenwood.

Mueller, John. 2004. The Remnants of War. Ithaca, NY: Cornell Uni-versity Press.

Munkler, Herfried. 2005. The New Wars. Cambridge: Polity.Olson, William. 1989. U.S. Objectives and Constraints: An

Overview. In Guerrilla Warfare and Counterinsurgency. U.S.Soviet Policyin the Third World, eds. Richard H. Shultz, Jr., Robert

L. Pfaltzgraff, Jr., Uri Raanan, William J. Olson, and Igor Lukes.Lexington, MA: Lexington Books, 1342.

Prunier, G erard. 2009. Africas World War: Congo, the RwandanGenocide, and the Making of a Continental Catastrophe. Oxford:Oxford University Press.

Przeworski, Adam. 1991. Democracy and the Market: Political andEconomic Reforms in Eastern Europe and Latin America. NewYork: Cambridge University Press.

Reno, William. 1999. Warlord Politics and African States. Boulder,CO: Lynne Rienner.

Salehyan, Idean, and Kristian Skrede Gleditsch. 2006. Refugeesand the Spread of Civil War. International Organization 60 (2):33566.

Sambanis, Nicholas. 2001. Do Ethnic and Nonethnic Civil WarsHave the Same Causes? A Theoretical and Empirical Inquiry(Part 1). Journal of Conflict Resolution 45 (3): 25982.

Sambanis, Nicholas. 2004. What Is Civil War? Journal of ConflictResolution 48 (6): 81458.

Schmitt, Carl. [1963] 2007. Theory of the Partisan. New York: Telos.Simons, Anna. 1999. War: Back to the Future. Annual Reviews of

Anthropology 28: 73108.Singer, J. David, Stuart Bremer, and John Stuckey. 1972. Capability

Distribution, Uncertainty, and Major Power War, 18201965. InPeace, War, and Numbers, ed. Bruce Russett. Beverly Hills, CA:Sage, 1948.

Stedman, Stephen John. 1996. Conflict and Conciliation in Sub-Saharan Africa. In The International Dimensions of Internal Con-

flict, ed. Michael E. Brown. Cambridge, MA: MIT Press, 235

65.Stein, Arthur A., and Steven E. Lobell. 1997. Geostructuralism

and International Politics: The End of the Cold War and theRegionalization of International Security. In Regional Orders:Building Security in a New World, eds. David A. Lake and PatrickM. Morgan. University Park: Pennsylvania State University Press,10122.

Stockholm International Peace Research Institute (SIPRI). 2008 .SIPRI Yearbook 2008: Armaments, Disarmament, and Interna-tional Security. Oxford: Oxford University Press.

Trinquier, Roger. 1964. Modern Warfare: A French View of Coun-terinsurgency. New York: Praeger.

United Nations Development Office for Somalia. 1998. Study onGovernance in Gedo Region. Nairobi: United Nations Develop-ment Program.

U.S. Agency for International Development (USAID). 2009.Overseas Loans and Grants: Obligations and Loan Autho-rizations, July 1, 1945September 30, 2008. CONG-R-01015.http://qesdb.usaid.gov/gbk (Accessed July 19, 2010).

U.S. Central Intelligence Agency. 2002. Balkan Battlegrounds: AMilitary History of the Yugoslav Conflict, 19901995. Washington,DC: Central Intelligence Agency, Office of Russian and EuropeanAnalysis.

Wallensteen, Peter, and Karin Axell. 1993. Armed Conflict at theEnd of the Cold War, 198992. Journal of Peace Research 30 (3):33146.

Westad, Odd Arne. 1992. Rethinking Revolutions: The ColdWar in the Third World. Journal of Peace Research 29 (4):45564.

Westad, Odd Arne. 2005. The Global Cold War: Third World Inter-ventions and the Making of Our Times. Cambridge: CambridgeUniversity Press.

Wimmer, Andreas,Lars-Erik Cederman, and Brian Min. 2009. Eth-

nic Politics and Armed Conflict: A Configurational Analysis of aNew Global Dataset. American Sociological Review 74 (2): 31637.

Wolin, Sheldon. 1973. The Politics of the Study of Revolution.Comparative Politics 5 (3): 34358.

World Bank, International Economics Department, DevelopmentData Group. 2007. World Development Indicators 2006 [CD-ROM]. Washington, DC: World Bank.

Yengo, Patrice. 2006. La guerre civile du Congo-Brazzaville 19932002. Paris: Karthala.

Yergin, Daniel. 1977. Shattered Peace: The Origins of the Cold Warand the National Security State. Boston: Houghton Mifflin.

Young, Crawford. 2006. The Heart of the African Conflict Zone:Democratization, Ethnicity, Civil Conflict, and the Great LakeCrisis. Annual Reviews of Political Science 9: 30128.

Zurcher, Christoph. 2007. The Post-Soviet Wars: Rebellion, EthnicConflict, and Nationhood in the Caucasus. New York: New YorkUniversity Press.

international system and technologies of rebellion

Documents