Existing categorizations of rebel groups have difficulty classifying some of today's most vexing rebels–those, such as the Islamic State, that reject the Westphalian state system and depend on an almost entirely religious justification for their cause. Such rebel groups often have unlimited war aims and are unwilling to negotiate with the states whose sovereignty they challenge. In this essay, I present the new category of “religionist rebels.” I show that religionist rebels have been present throughout the history of the state system, and explore the particular challenges they pose in the civil war context. Religionist rebels are often brutal in their methods and prosecute wars that are especially difficult to end. But the nature of religionist rebellion also suggests natural limits. Thus, religionist rebels do not, ultimately, present a long-term threat to the state system.

Rebellion is a defining element of civil wars, in which armed opponents challenge the sovereign authority of the state. But the set of political aims sought by rebels is as diverse as rebel groups themselves. Archetypical civil wars include the U.S. Civil War, in which the Confederacy fought to secede from the Union and create a new, independent state, and the Spanish Civil War, in which the Nationalist rebels sought to overthrow and replace the governing Republican regime. In both of these cases, the rebels not only accepted, but premised their war aims on the continuance of the international state system. Even if they may have sought to change borders or redesign the seat of government, both sets of rebels also sought to control a state.

My aim in this essay is to investigate a type of rebel group that is not new, but may seem so: what I call religionist rebels. Religionist rebels reject the Westphalian notion of the state: that is, a political entity that governs a clearly delimited population and territory and interacts with like units in the international system. Religionist rebels do not seek to carve out a new, independent state that will receive international recognition and, today, the benefits of a seat at the United Nations. Nor do they seek to capture the capital for the purpose of reclaiming the reins of government from those accused of having betrayed the public's trust. For religionist rebels, sovereignty does not reside in the people or in the recognition provided by other states. Instead, sovereignty is given–or even lent–by the divine.

Understanding the particular challenges posed by religionist rebels is important for three main reasons. First, if the right to rule and represent can only emanate from the heavens, religionist rebels are likely to reject approaches from the state whose sovereignty their rebellion directly challenges, as well as from third parties whose interests may also be at stake. It will, in other words, be difficult to negotiate with a group whose members recognize a divine source of legitimacy. Second and related, classifying religionist rebels can help shed light on their war aims and, in particular, whether those war aims are limited. And third, religionist rebels may also conduct war differently from rebels with other types of war aims, because the prosecution of the war–as well as its goals–may be justified on religious grounds.

Today's headlines abound with examples of Islamic religionist rebels, such as the Islamic State and Boko Haram. But religionist rebellion is not an exclusively Islamic, or even modern, phenomenon.1 Armed rebellions in China and Brazil in the mid- to late nineteenth century and even to some extent that of the (Christian) Lord's Resistance Army in Northern Uganda today all fall into the category of rebel groups whose beliefs lead them to reject the modern state system in favor of a different theology of sovereignty.

These differences suggest that there is a fundamental mismatch between religionist rebels and the current state system in terms of how the sources of the right to rule translate into the commencement, conduct, and conclusion of civil war. This mismatch sets religionist rebels apart from other types of rebel groups. Secessionist rebel groups, for example, have been shown to be less likely to target civilians than other types of rebel groups, in part because they are trying to gain the favor of an international community that disapproves of civilian targeting.2 And while existing scholarship has done less to tie war aims to the mode of war termination, there is at least some evidence suggesting that power-sharing–a popular postwar governing proposal–may be more likely to succeed in wars in which the rebels seek to replace the existing government than in those in which the rebels are divided from the government by ethnic identity.3

Many rebellions have a religious focus. Not all of these would be considered religionist, however. Many religious (but not religionist) rebel groups have secessionist aims, and actively seek to join the international states system. The Moro in the Philippines, for example, have a clear Muslim identity, but have often sought autonomy if not secession. Others aim to take over the central government, but maintain the state's relations with its neighbors, once again accepting the state system as it currently operates. The early part of the civil war between North and South Sudan was driven at least in part by the fear that the North was trying to impose Sharia law on the South, but at the beginning of this decades-long conflict, the rebels still sought to work within the confines of the Sudanese state by reshaping the government in Khartoum. Religionist rebels not only give prominence to religious beliefs, but also explicitly reject a notion of statehood centered on limits to sovereignty and mutual recognition.

Because religionist rebels are not interested in joining the international community as a newly independent state or in replacing the existing government's role, many recent strategies for war termination are unlikely to be effective when dealing with this type of rebellion. In this, religionist rebels may be somewhat similar to rebels whose main aim is plunder. But insofar as loot-seeking rebels are motivated by money, they can likely be paid off; the same is not true for religionist rebels. Recognizing the particular challenges that religionist rebels pose for war termination is critical to the resolution of religionist conflict. In this essay, I first present the conventional understanding of the possible array of rebel war aims. Next, I argue for adding a new type of rebel–religionist rebels–to existing categorizations. I lay out my logic via example, and delve into a series of cases of religionist rebels both to motivate and make my argument. I then examine briefly how past religionist rebellions were terminated, and to what extent the nature of their termination was driven by the religionist war aims of the rebels. I conclude with policy implications meant to apply to today's–and future–conflicts with religionist rebels.

Scholars have classified civil wars on a number of dimensions. I focus here on the classification of civil wars according to the political aims of the rebel group. The Uppsala Conflict Data Program (UCDP), an invaluable resource for civil war researchers, identifies two principal “incompatibilities” between governments and rebel groups fighting civil wars. Wars over the government are defined as “concerning the type of political system, the replacement of the central government or the change of its composition.”4 Wars over territory are defined as “concerning the status of a specified territory, e.g. the change of the state in control of a certain territory (interstate conflict), secession or autonomy (intrastate conflict).”5 An influential article that preceded a boom of research on civil wars distinguished “ethnic” from “nonethnic” civil wars, and described “nonethnic” civil wars as “revolutionary or other types of war.”6 In addition to wars over the control of the central government and ethnicity/territory, scholars have sometimes pointed to a third category: civil wars that are described as “resource wars,” in which the rebels' primary aim is the plunder and sale of natural resources.7

Many important insights regarding the conduct and termination of civil war have emerged from the analysis of variation in rebel war aims.8 A focus on the possibly distinctive behavior of groups that seek, at a minimum, autonomy and, at a maximum, a new independent state appears to be most common.9 And wars in which rebels are motivated primarily by profit are thought, for example, to generate indiscipline and extensive civilian targeting, in contrast to conflicts in which clear political aims govern rebels' behavior.10

Connecting rebel political aims to the commencement, conduct, and conclusion of civil war is a sensible strategy. One problem, however, with the execution of this strategy to date has been the great deal of heterogeneity in the reference category. In other words, when we distinguish between “ethnic” and “nonethnic” or “secessionist” and “nonsecessionist,” we push aside the significant variation in the “non” category. For example, another seminal article in the field identifies twenty ethnic civil wars, leaving a remainder of thirty-six nonethnic civil wars.11 The latter category includes a diverse set of conflicts, from Marxist uprisings such as the Cuban Revolution in 1958 to the attempt by Islamists to take over Algeria in 1992 to southern secessionism in Yemen in 1994. The UCDP data code wars as being over government, territory, or both; a system that can also lead to some confusing coding decisions. For example, a conflict between the United States and Al Qaeda (part of the so-called global war on terror) is coded as being over the type of government, presumably that of the United States.

Exploring this variation can be a risky proposition in that classification schemes can go too far. Overly detailed typologies are often too confusing to be useful. Bearing this caveat in mind, I nonetheless suggest adding at least one more type of rebel political aims: that of religionist rebels.

Before presenting specific examples of religionist rebels, I first want to clarify the claim that religionist rebellion is inconsistent with the Westphalian state system. I conceive of the Westphalian state system as one populated, at least ostensibly, by Weberian states whose governments possess (again, at least ostensibly) “a monopoly on the legitimate use of force within a given territory.”12 Limits are inherent to this definition of statehood. States control a limited amount of territory. Their right to rule is limited by two kinds of recognition, from the domestic population of the state and from the recognition of other states populating the system.

Many internationally recognized states fail to meet the Weberian ideal of statehood today. Somalia is often taken to be a prime example of this failure. The Somali government created by the constitution of 2012 certainly does not hold a monopoly on the use of force within its internationally recognized borders. But representatives of Somalia nonetheless hold a seat at the United Nations and appear to play by the rules of the international game. That Somali representatives retain these rights points to the fact that today's state system is Westphalian in a thin sense. States may interfere in each other's politics. They may even challenge each other's ownership to particular pieces of territory. But, with only a very few exceptions, they do not challenge each other's right to exist, at least because they understand that their own existence depends on the maintenance of the system and the sufferance of others.13

As a weak, even failed state, Somalia faces a series of domestic challenges. For the most part, however, these challenges are to the Somali state in particular and not to the international state system in general. Somaliland in the north seeks to create its own independent state and very much wants a seat at the UN. In the 1980s, the Somali National Movement and Somali Salvation Democratic Front sought to overthrow President Siad Barre. More recently, perhaps partly as a result of Somalia's prolonged state failure, the religionist group Al Shabaab has gained control over significant portions of Somalia, with the aim of establishing a global Islamic caliphate.

Three related characteristics distinguish religionist rebels from other types of rebel groups. First, religionist rebels are categorized by how they view the source of sovereignty, defined here as the right and fact of ruling a given population and territory. Both center-seeking and secessionist rebels seek to control a state, and view sovereignty as emanating from the people and/or recognition by other states in the international system. Rebels concerned principally with plunder may not seek sovereignty in any form, and may in fact prefer that some other party holds de jure sovereignty as long as they maintain control over resource-rich territory and access to markets through which they can sell their goods. Religionist rebels reject the notion that sovereignty is rooted in the people or in the international community, and their (ostensible) motives for any control of territory are theological rather than pecuniary.

Second, because religionist rebels rely on a divine source of sovereignty, they reject the legitimacy of other units in the world whose sovereignty claims rest on secular sources, such as popular support and/or recognition by other units in the system. If religionist rebels do not recognize the legitimacy of other units in the system, they are unlikely to engage in formal relations with them. This restricts trade, negotiation, and diplomacy, all of which are critical to the current states system. Religionist rebels seek to be a world apart while remaking the world in their image.

Third, religionist rebels do not accept territorial limits on their sovereignty claims, unless those limits have a divine justification. As a scholar of the Holy Spirit Movement described her subject: “The goal of the war, as Lakwena explained, was less the military conquest of foreign territory than the spreading of the Word of God throughout the world…. The Holy Spirit Movement had a supra-ethnic, pan-African, and finally universal mission.”14 If the mandate from heaven is to convert and spread the word of God, there can be no inherent limit to this task.

A historical example of a religionist rebel group is the Yellow Cliffs rebels in mid-nineteenth-century China. At that time, the reach of the Chinese state did not extend to its recognized–and claimed–borders. As a result, there were a number of challenges to Peking's hold on outlying areas; many of these rebellions exhibited a strong religious cast.15 For example, the Miao and Hui in the southwest, along with the Tungan and Xinjiang Muslims in the northwest, fought for autonomy if not independence from the Chinese state in the 1860s and 1870s.

Another challenge came from the Nien, bandits who were terrorizing China's northern countryside. Refugees from the Nien flocked to the Yellow Cliffs in the north, where they were gathered in and then ministered to by religious leader Zhang Jizhong. As time passed, supplicants arrived at the Yellow Cliffs not as refugees, but as pilgrims seeking to join Zhang's religious group.16

Zhang Jizhong subscribed to an alternative form of Confucianism known as the Taizhou school. The Taizhou school had been founded by Wang Gen, who had been inspired by “a dream in which he single-handedly prevented the heavens from imminent collapse and restored the sun, moon, and stars to working order.”17 Zhang's realization of this theology led him to create what was effectively a small theocracy in the Yellow Cliffs. Zhang controlled entrance to the community and provided safety from local bandits, education, food and shelter, and, of course, a system of worship. Sovereignty was effectively invested in Zhang, who took on the mantle of “high priest.”18

At their height, the Yellow Cliffs rebels fielded eight hundred soldiers.19 But these were primarily meant to protect the Yellow Cliffs community from bandits; they had issued no formal challenge to the government in Peking, effectively ignoring its claim to sovereignty over the Yellow Cliffs. The rebels were a group unto themselves, and only took up arms when their ability to self-govern was challenged.

The topography of the Yellow Cliffs region both protected and doomed Zhang and his followers. The cliffs created a geographical refuge, but also enabled a siege by government forces. While the residents of the Yellow Cliffs were offered the opportunity to surrender, none took advantage. All ten thousand souls–with the exception of four hundred women and children who were spared but had not surrendered–perished in the assault, committed to Zhang and his religious teachings.20

Similar to Zhang's disciples in the Yellow Cliffs of China, the Canudos of Northeast Brazil rejected the sovereignty of the newly formed Brazilian republic in 1896 without intending to challenge it directly. Antonio Maciel–also known as Antonio Conselheiro–led a Catholic community of twenty thousand who viewed the Republic as the “anti-Christ.”21 While the Canudos believed in the divine right of monarchy, they did not support the prior monarchy of Dom Pedro; they believed that a Portuguese crusader king would rise from the dead to lead them, and referred to their community as the “New Jerusalem.”22 The Canudos were organized along the lines of the Church itself. Although he did not claim the status of divine messenger or prophet, Conselheiro was treated as such by his subjects, who referred to him as the messiah.23

Also like the Yellow Cliffs rebels, the main political aim of the Canudos appears to have been to be left alone. The Canudos neither claimed an independent state nor directly challenged the sovereignty of the Brazilian government.24 At the same time, they clearly rejected the rules and reign of the state. The Canudos created their own governance structures, based on Catholic teaching, which included a group of apostles to Conselheiro. The group held to a fairly austere ethos, likely driven as much by religious devotion as economic need, given that its membership was extremely poor.25 The implicit challenge to the government lay in the combination of the rejection of local landowners' authority and the sheer size of the Canudos revolt. At its height, the Canudos community consisted of fifteen to twenty thousand people and at least four thousand soldiers, and was the second largest community in the state of Bahía.26

The Canudos rebellion itself occurred only in response to government action. Neighboring communities became nervous about the Canudos, accusing them of primitivism and of being “superstitious zealots.”27 Four separate government expeditions were launched against the Canudos. The first three were repelled by Conselheiro's juaguncos, a fighting force whose members were recruited as much for their skill as for their willingness to die for the Canudos cause.28 The fourth and largest was successful, by starving out the Canudos, who never surrendered.29 Government forces ultimately killed nearly all the adult male Canudos and burned the community to the ground.

Two additional related cases–the Holy Spirit Movement Forces (HSMF) and the subsequent Lord's Resistance Army (LRA)–illustrate some of the challenges in identifying religionist rebels. There is no question that the Holy Spirit Movement, which fought the Ugandan government in the mid-1980s, was religiously motivated. The movement's founding moment is said to have occurred when a holy Christian spirit named Lakwena visited an Acholi woman, Alice Auma, in Northern Uganda. Alice Lakwena (previously Auma) then took on the mantle of prophetess, and called troops to her side in the name of God. While it is certainly the case that the religious beliefs of the HSMF led to poor tactical and operational decision-making on the battlefield and governed the war aims of the group, the HSMF accepted the confines of the state system in that it aimed to replace the central government in Kampala. Indeed, the HSMF was defeated when it launched a failed assault on the capital in 1987. By my coding, the HSMF would be considered religious, but not religionist.

Coding the Lord's Resistance Army is more complicated. Many accounts trace the lineage of the LRA to the demise of the Holy Spirit Movement Forces, suggesting the possibility of an alignment of war aims. But experts on this case suggest that the LRA's war aims are more difficult to discern than those of the HSMF. On one hand, the LRA did seem to accept the idea of the state system–or, at least, did not directly challenge it–on more than one occasion. For example, Joseph Kony, the infamous leader of the LRA, accepted aid from the government of Sudan. The LRA also negotiated directly with the Ugandan government.30 On the other hand, the LRA appeared to reject the notion of state sovereignty on several fronts. Kony's stated war aims in support of “the application of the Ten Commandments and the liberation of the people of Northern Uganda” do not necessarily hold the maintenance of the state system dear.31 According to Kevin Dunn, a scholar of the conflict: “One of the more pronounced features of the conflict is Kony's limited interest in communicating with the outside world.” Dunn further notes that the LRA was less interested in “seizing the reins of power” from the central government than in creating an “enclave …or personal fiefdom.”32 This description of the LRA bears a striking resemblance to the cases of the Yellow Cliffs rebels and the Canudos: the state system itself is deemed secondary to the desire to be left alone.

The LRA is a borderline case in coding religionist rebellion. Its war aims appear to have been somewhat fluid: it may have sought to overthrow the Ugandan government at some points and rejected the state system in its entirety at others. This rebellion appears to have petered out (at least in terms of battle deaths), but recent reports indicate an increase in child abductions that may mean that the LRA is attempting to rebuild via the use of child soldiers.33

Most recently, the Islamic State emerged from the wreckage of the 2003 Iraq War, initially to push the United States out of Iraq but, subsequently, to declare a caliphate on earth with no clear territorial borders. Attempts by the international community to engage with the Islamic State have been generally rebuffed, perhaps most clearly when governments have tried–unsuccessfully–to negotiate with the Islamic State to return hostages.34 Broadly speaking, humanitarian agencies have been frustrated in attempts to reach out to groups such as the Islamic State and its cousin, Boko Haram, although there have been some limited successes, such as when the Swiss government and International Committee of the Red Cross negotiated the return of twenty-one girls taken by Boko Haram.35 Because religionist rebels do not subscribe to Western and Westphalian notions of sovereignty, the international community has little in the way of leverage to pressure these groups to comply with humanitarian norms.

Note that religionist rebels often provide many of the services of a state. Both the Yellow Cliffs rebels and the Canudos provided security, education, food, and shelter to their residents. Part of the Islamic State's success has been attributed to its ability to step in and provide basic services, such as garbage collection, at a time when the Syrian and Iraqi states were failing to do so (and notwithstanding the fact that the provision of these services has often been selective and used for recruitment purposes).36 But essential to the Westphalian notion of statehood is the dual concept of international recognition of boundaries. This implies limits on sovereignty, and such limits are inconsistent with the sovereignty of the divine. It also implies living in a world with other recognized states, which is also inconsistent with the notion that there is one true path that all should follow.

Religionist rebels may also fight differently from center-seeking or secessionist groups. While all these groups fight to win, the extent to which religious belief permeates the daily life of religionist rebels can extend to the battlefield. A prime example of this phenomenon is the belief in bullet-proofing. Soldiers of the Holy Spirit Movement were forbidden to kill, relied on holy water and religious songs to protect them, and (confusingly) invoked the name of James Bond as they engaged enemy troops.37 Another manifestation of the relationship between religionism and warfighting occurs when the belief in religious ends is used to justify brutal means. The Islamic State's governor in Aleppo drew upon Koranic scripture to endorse killing, crucifixion, and the amputation of hands and feet as punishment for enemies of the Islamic State, and evidence suggests that the organization has engaged in all of these practices.38

This is not to vilify religion: many if not most religious groups are devotedly nonviolent; see, for example, the Quakers and Jains. But religionist groups that use violence may accept few restrictions on their behavior, unless restrictions emanate from a divine source. This suggests that, in addition to identifying religionist rebels, it may also be useful to make doctrinal distinctions among them to determine which are likely to be the worst-behaved. Allegiance to the sovereignty of the divine may permit, demand, or restrict violence in different cases.

The greatest challenge posed by religionist rebels may be war termination. The historical examples described above offer limited comfort, in that many of these groups have tended to create strongholds in remote locations. While they have gained thousands of supporters, their popular appeal has been limited, and they have typically been defeated by government forces. Indeed, the likelihood of their defeat may be a function of their religionist nature; to the extent that their actions were governed by theology rather than strategy–particularly on the battlefield–they may have, in effect, been “selected out” when facing nonreligionist adversaries.39

A logic of internal containment may help explain the fate of groups like the Yellow Cliffs rebels and the Canudos. But from a distance of decades, these groups may appear small and weak, especially in hindsight. The Taiping Rebellion–fought in China just a few years prior to the outbreak of the Yellow Cliffs Rebellion–may therefore serve as a more compelling comparison to contemporary examples. The Taipings, also known as the God-Worshippers, fought in Eastern China from 1850–1864, marching on Nanking and Shanghai among other cities. The leader of the Taiping Rebellion, which took over twenty million lives, claimed to be the younger brother of Jesus Christ. While the Taipings certainly contrasted themselves with what they saw as a corrupt Manchu regime, they viewed their own mandate as coming from the heavens, and held that “the whole empire is the universal family of our Heavenly Father, the Supreme Lord and Great God.”40 What is more, they rejected overtures from diplomats from abroad, unless such delegations swore allegiance to “the Heavenly King, the head of the movement” as well as to the Taiping religion itself.41 Even though it was orders of magnitude stronger than the Yellow Cliffs and Canudos rebellions, the Taiping Rebellion was ultimately (and soundly) defeated by Western-trained imperial government forces. Similarly, at this writing, the reach of the Islamic State has contracted dramatically, with significant losses in territory, population, and financing and the fall of core strongholds, such as the Syrian city of Raqqa.

Conflict resolution with religionist rebels is difficult because they eschew negotiation. Faith and divine responsibility cannot be negotiated away, and other actors are viewed as illegitimate negotiating partners if they do not subscribe to the same beliefs. This leaves two unsavory options on the table. The first option is to fight to the end, a strategy employed by the Chinese and Brazilian governments when negotiations with religionist rebels in their territory failed. Although ultimately effective, this strategy was also extremely costly, with tens of thousands of government and rebel soldiers–as well as civilians–dying in each war. The second option is to accept a hybrid system in which religionist rebels coexist alongside the Westphalian state system. Some version of this hybrid has existed since the emergence of the modern state system. This particular type of hybridity is problematic not (just) because it upsets the Westphalian apple cart, but more so because of the lack of limits inherent to the evangelism of religionist rebels. It is not only their existence, but also their rejection of alternative justifications and systems of rule, that challenges nearby states. At the same time, however, it is precisely this lack of limits in aspiration that likely places a natural bound on the expansionist tendencies of religionist rebels. Because their attitude toward the state system undercuts third-party support for religionist rebels, they may be especially likely to turn to other, often criminal, sources of financing. The resort to criminality in turn undermines discipline within the organization as well as the credibility of its ideological appeals and governance efforts.42 At this writing, for example, it appears that both the Islamic State and Boko Haram have begun to fall back, after bumping into limits to expansion, as well as pushback from the international community.43 An uneasy truce may therefore be possible, albeit neither pleasant nor likely.

Religionist rebels' plans may appear to resemble the claims of past empires relying on justifications of sovereignty that invoked ideology or the divine right of monarchs.44 But there are important differences from these historical analogs. Not only was the Soviet Union a member of the United Nations, it participated in its founding, and even argued for the admission of Soviet Socialist Republics that lacked the independence necessary to make their own foreign policy decisions. Similarly, the European monarchs of old certainly had an insatiable thirst for expansion, but also treated frequently with each other, and recognized that there were probably natural limits to the scope of their empires.

Religionist rebels are not new to international politics. And they are likely to continue to emerge and persist in one form or another. The bad news is that their aims are often without limit, their means are frequently brutal, and attempts at negotiation may be futile. The good news is that religionist rebels do appear, historically, to have bumped up against natural limits, precisely because of the claims they make and practices they engage in during the wars they fight. And while the bottom line has been an especially bloody one, as belief in the sovereignty of the divine makes religionist rebellions particularly difficult to end, the key takeaway is the necessity of distinguishing religion from religionism in identifying these especially challenging rebellions.


In his contribution to this project, Hendrik Spruyt outlines various schools of Islamic legal thought and shows that some view Islamic law as compatible with the Westphalian state system, while others do not. Hendrik Spruyt, “Civil Wars as Challenges to the Modern International System,” Dædalus 146 (4) (Fall 2017). For a more extensive analysis of Islamic law in this regard, see Emilia Justyna Powell, “Islamic Law States and the Peaceful Resolution of Territorial Disputes,” International Organization 69 (4) (2015).


Tanisha M. Fazal, Wars of Law: Unintended Consequences in the Regulation of Armed Conflict (Ithaca, N.Y.: Cornell University Press, forthcoming 2018), ch. 7; Virginia Page Fortna, “Do Terrorists Win? Rebels' Use of Terrorism and Civil War Outcomes,” International Organization 69 (3) (Summer 2015); and Hyeran Jo, Compliant Rebels (Cambridge: Cambridge University Press, 2015).


Caroline Hartzell and Matthew Hoddie, “Institutionalizing Peace: Power Sharing and Post-Civil War Conflict Management,” American Journal of Political Science 47 (2) (2003): 328; and Thomas Chapman and Philip G. Roeder, “Partition as a Solution to Wars of Nationalism: The Importance of Institutions,” American Political Science Review 101 (4) (2007). For a challenge to this finding suggesting that ethnic conflict is not a good predictor of the success of power-sharing arrangements, see Jason Michael Quinn, T. David Mason, and Mehmet Gurses, “Sustaining the Peace: Determinants of Civil War Recurrence,” International Interactions 33 (2) (2007).


Uppsala Conflict Data Program, Department of Peace and Conflict Research, Uppsala Universitet, “Definitions,” http://www.pcr.uu.se/research/ucdp/definitions/.




Nicholas Sambanis, “Do Ethnic and Nonethnic Civil Wars Have the Same Causes? A Theoretical and Empirical Inquiry (Part 1),” Journal of Conflict Resolution 45 (3) (2001): 260.


Michael L. Ross, “How Do Natural Resources Influence Civil War? Evidence from Thirteen Cases,” International Organization 58 (1) (Winter 2004).


Analysts (including this author) who connect the political aims of rebel groups to their behavior do not necessarily assume that all members of the group are committed to the same aims. Even if motivations of individual members are heterogeneous, the assumption is instead that there is a critical mass of commitment to the stated aims of the group.


Kathleen Gallagher Cunningham, Inside the Politics of Self-Determination (Oxford: Oxford University Press, 2014); Trace Lasley and Clayton Thyne, “Secession, Legitimacy and the Use of Child Soldiers,” Conflict Management and Peace Science 32 (3) (2014); Barbara F. Walter, Reputation and Civil War: Why Separatist Conflicts are So Violent (Cambridge and New York: Cambridge University Press, 2009); and Monica Duffy Toft, The Geography of Ethnic Violence: Identity, Interests, and the Indivisibility of Territory (Princeton, N.J.: Princeton University Press, 2003).


Jeremy M. Weinstein, Inside Rebellion: The Politics of Insurgent Violence (Cambridge: Cambridge University Press, 2007).


James D. Fearon and David D. Laitin, “Ethnicity, Insurgency, and Civil War,” American Political Science Review 97 (1) (2003).


Max Weber, “Politics as a Vocation,” in From Max Weber: Essays in Sociology, ed. Hans Heinrich Gerth and C. Wright Mills (New York: Oxford University Press, 1946).


This view lines up with Stephen Krasner's description of international legal sovereignty. Stephen D. Krasner, Sovereignty: Organized Hypocrisy (Princeton, N.J.: Princeton University Press, 1999), 14–20.


Heike Behrend, Alice Lakwena and the Holy Spirits, trans. Mitch Cohen (Oxford: Oxford University Press, 1999), 56–57.


Elizabeth Perry and Tom Chang, “The Mystery of Yellow Cliff: A Controversial ‘Rebellion’ in Late Qing,” Modern China 6 (2) (1980): 123.


Ibid., 132–133.


Ibid., 127.


Ibid., 136.


Ibid., 138.


Ibid., 147.


Lori Madden, “The Canudos War in History,” Luso-Brazilian Review 30 (2) (1993): 6.


Micheal Clodfelter, Warfare and Armed Conflicts: A Statistical Encyclopedia of Casualty and Other Figures, 1494–2007, 3rd ed. (Jefferson, N.C.: McFarland and Company, Inc., 2008), 351; and Bradford E. Burns, A History of Brazil (New York: Columbia University Press, 1993), 248–249.


Robert M. Levine, “‘Mud-Hut Jerusalem’: Canudos Revisited,” The Hispanic American Historical Review 68 (3) (1988): 533.


Teresa A. Meade, A Brief History of Brazil, 2nd ed. (New York: Facts on File, 2010), 99.


Robert M. Levine and Penny Roberts, The Massacre in History (New York: Berghahn Books, 1999), 195.


Levine, “‘Mud-Hut Jerusalem,’” 525, 527; and Burns, A History of Brazil, 251–252.


Susanna Hecht, The Scramble for the Amazon and the “Lost Paradise” of Euclides da Cunha (Chicago: University of Chicago Press, 2013), 10.


Levine and Roberts, The Massacre in History, 192.


Robert M. Levine, Vale of Tears: Revisiting the Canudos Massacre in Northeastern Brazil, 1893–1897 (Berkeley: University of California Press, 1992), 183–185.


Sam Farmar, “Uganda Rebel Leader Breaks Silence,” BBC, June 28, 2006.


Kevin C. Dunn, “The Lord's Resistance Army and African International Relations,” African Security 3 (1) (2010): 51.


Ibid., 51, 57.


Rick Gladstone, “U.S. Broadens Sanctions on Joseph Kony and His Group,” The New York Times, March 8, 2016.


“Japan Grapples for Dialogue with ISIS over Hostages,” CBS News, January 22, 2015; and Steve Almasy, “ISIS: Japanese Hostage Beheaded,” CNN, February 3, 2015.


Stephanie Busari, Jason Hanna, and Faith Karimi, “Boko Haram Releases 21 Chibok Girls to Nigerian Government,” CNN, October 13, 2016.


Yochi Dreazen, “From Electricity to Sewage, U.S. Intelligence Says the Islamic State is Fast Learning How to Run a Country,” Foreign Policy, August 19, 2014; and Megan Stewart, “Why the Islamic State Is so Bad at Being a State,” Political Violence @ a Glance, December 22, 2015.


Behrend, Alice Lakwena and the Holy Spirits, 58–59.


Islamic State in Iraq and al-Sham, “Specimen 7F: Statement from the Wali (Governor) of Aleppo Province: October 2013,” http://www.aymennjawad.org/2015/01/archive-of-islamic-state-administrative-documents (accessed October 10, 2016). It is important to note that similarly brutal prescriptions can be found in Judeo-Christian scripture. See, for example, the Book of Samuel, in which King Saul is told to commit genocide.


Jason Sharman, “War, Selection, and the State-as-Firm Analogy: A Critique of Evolutionary Theory in International Relations” (unpublished article, Griffiths University).


Franz Michael, “Military Organization and Power Structure of China during the Taiping Rebellion,” Pacific Historical Review 18 (4) (1949).


Eugene P. Boardman, “Christian Influence Upon the Ideology of the Taiping Rebellion,” The Far Eastern Quarterly 10 (2) (1951).


I thank Karl Eikenberry for suggesting this logic. See also Weinstein, Inside Rebellion.


Nathaniel Allen, “Charting Boko Haram's Rapid Decline,” War on the Rocks, September 22, 2016.


For a discussion of the possible similarities between Marxist and Islamist rebels, see Stathis N. Kalyvas, “Jihadi Rebels in Civil War,” Dædalus 147 (1) (Winter 2018).

This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) license, which permits copying and redistributing the material in any medium or format for noncommercial purposes only. For a full description of the license, please visit https://creativecommons.org/licenses/by-nc/4.0/legalcode.