Insurgent organizational structure and the control of collective violence

  • Fall 2013
Alec Worsnop

Alec Worsnop is a PhD student in Political Science with interests in international relations, security studies, and comparative politics.

By Alec Worsnop
November 1, 2013

A CENTRAL MOMENT during the United States surge in Iraq came with the August 2007 stand down of Muqtada al–Sadr's Jaysh al–Mahdi (JAM). As a result, sectarian violence declined drastically as Baghdad became more stable.1 However, few analysts in academia, the government, or the private sphere expected the ceasefire to hold. Pointing to JAM's internal divisions and geographic over–reach as well as Iran's attempts to divide and rule Iraq's Shiites, they argued that Sadr did not have sufficient control of the organization to ensure that his fighters would put down their arms.2 As a professor in Baghdad told the International Crisis Group: "Everybody was surprised by the degree to which militants obeyed Muqtada al–Sadr. At first, I expected about half of the Mahdi Army members to ignore him."3 Instead, compliance was substantial and stuck despite intrusive operations by Coalition and Iraqi forces.

My research addresses this puzzle: how did a seemingly fragmented and disjointed organization adhere to a costly ceasefire? Resolving this question requires better specifying when organizations will fragment as well as how and when such fragmentation will limit insurgent organizations' ability to employ and calibrate the use of violence. Rather than treating the causes and consequences of fragmentation uniformly, I find that the impact of factors such as state tactics, internal disagreements, imbalances in power between and within groups, or geographic stretch is dependent on the institutional context in which organizations operate.

In particular, I seek to identify the organizational characteristics which determine whether the leaders of formal insurgent organizations can control when violence is employed. Such control includes ensuring insurgents fight when ordered and abide by agreements or orders to cease violence. My research indicates that the degree of this control within formal organizations is related to the interaction of two organizational characteristics: (1) leadership embeddedness, or the extent to which leaders are rooted in strong underlying communities and social structures; and (2) resource centralization, or the extent to which leaders directly, and exclusively, distribute both material and social resources. Embedded leaders must provide resources in a manner which leverages control of the community mechanisms needed to motivate and sustain collective violent behavior.

In the remainder of this piece, I will first lay out the basic tenets of this theory and underscore how it clarifies current approaches to fragmentation. Second I will briefly elaborate how this theory can explain JAM's ability to control when violence was employed from 2003 to 2008.



This theory is derived from the need for formal military organizations to actively motivate and gain the allegiance of their fighters. During civil wars, insurgent organizations often try to develop such relationships with their fighters by building on pre–war community structures and social linkages. Indeed, many authors have found that strong community structures are crucial in both starting and sustaining rebellion. Such communities are able to employ status rewards based on solidarity, enforce norms of fairness, ensure monitoring and concomitant sanctioning of undesired behavior, and share information leading up to and during rebellion.4

However, my research suggests that organizations embedded in such communities are not able to simply absorb and then employ these communal mechanisms to support the aims of the formal organization. Organizations face the unique challenge of co–opting such community mechanisms for their own use. This challenge is evidenced by the major dissimilarities between insurgent organizations built upon similar communities, for example: Islamic Jihad and Hamas in the Gaza Strip, Hezbollah and Amal in Lebanon, the Taliban and Hizb–i–Islami in Afghanistan, the Badr Organization and Jaysh al–Mahdi in Iraq, or the Viet Minh, the Dai Viet, and the Viet Nam Quoc Dan Dang in Vietnam. As Selznick observed over a half century ago with respect to Bolshevik insurgents in Russia, "[O]rganizations become infused with value as they come to symbolize the community's aspirations, its sense of identity." Only by capturing the social base can formal leaders shift fighters from simple participants to "deployable personnel."5

I find that another organizational characteristic—how leaders distribute social and material resources—determines whether or not leaders who are embedded in strong communities can leverage the bonds within those communities. Leaders will benefit from the provision of pay, food, clothing, or services when they are able to directly and exclusively control how such resources are distributed. In a simplistic sense, if leaders do not directly pay or punish their members, they cannot credibly threaten to withhold pay or to apply punishment if their fighters defect. However, even when insurgent organizations directly distribute resources, the utility they gain is limited when other members of either the organization, the broader rebellion movement, or the state can also supply access to those resources.

Provision of resources by the formal leaders captures underlying communal networks by both increasing the importance of the ties between formal leaders and the community and positioning these formal leaders as key actors within the informal social structure. This status allows leaders to leverage the community mechanisms, such as norms of reciprocity or solidary benefits such as a sense of community and camaraderie, needed to motivate and sustain collective violent behavior. Thus, both the social context in which resources are provided (i.e., the extent to which leaders are embedded in strong communities) and the manner in which those resources are provided (i.e., resource centralization) determine whether organizations are able to ensure that their members use and cease using violence when they are so ordered.

This focus on organizational characteristics helps to clarify the role of many other factors often linked with fragmentation or defection within insurgent groups. For example, while many studies theorize that increases in relative power will make insurgent groups more effective, my research identifies how increases in relative power such as capturing territory or gaining access to new weaponry can reduce the importance of centralized leaders in distributing resources or empower local leaders. Similarly, while external support may provide more relative power to an organization vis–à–vis other conflict actors, if the formal leadership does not control the resources being distributed, this change in relative power would lessen their capacity to control when violence is employed.



I'll briefly discuss some empirical findings with respect to JAM from 2003–2008 to illustrate the utility of the theory in explaining their compliance with the 2007 ceasefire in Baghdad. JAM is rooted in the movement of Muqtada al–Sadr's father, Muhammad Sadiq al–Sadr, and distant uncle, Muhammad Baqir al–Sadr. In stark contrast to the majority of the Shiite clergy in Iraq, they espoused a distinctly non–quietist approach calling for the removal of Saddam Hussein and establishment of an Islamic government in the mold of their Iranian neighbors. In particular, Sadiq sought to formally connect the religious establishment with Iraq's large and underprivileged Shiite community. His efforts were largely successful as he built a strong persona, deep patronage networks, and garnered vast public support. Muqtada al–Sadr was embedded in this strong Shiite community when he assumed a leadership position in Jaysh al–Mahdi following the United States' 2003 invasion of Iraq.

However, Sadr's ability to exclusively provide material and social goods, such as mosque sermons, local charity, payment, weapons, and spiritual guidance, was limited. Beyond competing with many more accomplished religious leaders within the Sadrist trend, Sadr lacked the ability to dispense patronage due to the relative poverty of his followers, his lack of access to outside resources, and competition from other religious figures. Illustratively, during the 2004 spring and summer conflicts, JAM members were not paid and were forced to buy their own weapons and provide their own transportation.6 As such, given that JAM was not resource centralized, the theory correctly predicts that during this time period, JAM was unable to maintain a number of ceasefires. As a result, JAM suffered severe military setbacks and was unable to fully take up arms again until early 2006.

By 2006, Sadr was able to capitalize on being embedded by centralizing control of resources. He became a main provider of the religious and social resources of order, guidance, and Islamic governance. Sadr created the Mahdist Institute to teach basic principles of faith and established a code of conduct enforced by a Judgment Committee which disciplined those violating the rules. More importantly, by joining the government, Sadr took control of a number of ministries and was able to provide rents to Shiites, particularly in Sadr city where his support was seen as strongest. Members of JAM became cabinet ministers in health, human services, transportation, etc., allowing for the capture and provision of government resources. These services were directly distributed by Sadrist neighborhood offices (Maktab al–Sayyid al–Shahi) where citizens had to go to get access to the services.7

Thus, while there were many factions within JAM by 2007, the theory predicts that the organization had the necessary characteristics to ensure compliance with ceasefires. Obeying the ceasefire was not an easy choice for these foot soldiers. Foot soldiers and mid–level commanders lost significant territory and prestige. Beyond simply losing explicit control, many JAM members were forced to flee Baghdad fearing retribution from other Shiite and Sunni militants. Nonetheless, there were numerous reports in the Arabic and U.S. news media of continued allegiance to Sadr. Indeed, the International Crisis Group interviewed a number of followers abiding by the ceasefire who expressed impatience, but commitment to Sadr. One follower told them that we "are impatiently waiting for Muqtada al–Sadr to announce a resumption of the Mahdi Army's activities. You'll see what we'll do with those…The only reason we are not reacting now to Badr's attacks is that we respect Muqtada al-Sadr's decision."8



My research underscores the importance of studying how processes of fragmentation are influenced by organizational characteristics. In particular, the extent to which formal leaders are embedded in strong communities and can directly distribute resources determines their capacity to control when violence is collectively employed. Nonetheless, different organizational characteristics should be associated different causal processes. Indeed, organizations that are able to control when violence is employed may still be quite weak militarily and easily destroyed. My dissertation builds on this logic by researching how additional organizational factors such as formal hierarchy, centralized training, and command and control procedures are needed to explain insurgent military effectiveness.



1 Anthony Cordesman and Emma R. Davies, Iraq's Insurgency and the Road to Civil Conflict (Prager Security International and Center for Strategic and International Studies, 2008); Ann Scott Tyson, "Petraeus Says Cleric Helped Curb Violence,"Washington Post, December 7, 2007.

2 For example, see Sabrina Tavernise, "Cleric Said to Lose Reins of Parts of Iraqi Militia," New York Times, September 28, 2006; Marisa Cochrane, The Fragmentation of the Sadrist Movement, Iraq Report (The Institute for the Study of War, 2009); Anthony Cordesman, "Iraq's Sectarian and Ethnic Violence and Its Evolving Insurgency," Center for Strategic and International Studies 2 (2007); Paul Staniland, "Explaining Cohesion, Fragmentation, and Control in Insurgent Groups" (Ph.D. Dissertation, Massachusetts Institute of Technology, 2010), 40.

3 International Crisis Group, Iraq's Civil War, the Sadrists and the Surge, Middle East Report, February 7, 2008, 17. 

4 For example, see James Scott, The Moral Economy of the Peasant: Rebellion and Subsistence in Southeast Asia (New Haven: Yale University Press, 1977); Charles Tilly, Doug McAdam, and Sidney Tarrow, Dynamics of Contention (New York: Cambridge University Press, 2001); and Roger Petersen, Resistance and Rebellion: Lessons from Eastern Europe (Cambridge, U.K.: Cambridge University Press, 2001).

5 Philip Selznick, Leadership in Administration: A Sociological Interpretation(University of California Press, 1957), 19; Philip Selznick, The Organizational Weapon: A Study of Bolshevik Strategy and Tactics (Glencoe, Ill.: The Free Press, 1960), 20.

6 International Crisis Group, Iraq's Muqtada Al–Sadr: Spoiler or Stabiliser?, Middle East Report (International Crisis Group, July 11, 2006); Patrick Cockburn, Muqtada Al–Sadr and the Battle for the Future of Iraq (Simon and Schuster, 2008). 

7 International Crisis Group, Iraq's Muqtada Al–Sadr: Spoiler or Stabiliser?, 13, 20. 

8 International Crisis Group, Iraq's Civil War, the Sadrists and the Surge, 14.