Harmony and Disharmony:
Exploiting Al-Qa'ida's Organizational Vulnerabilities

back to report home

The Challenges of Organizing Terror
A Theoretical Framework for Analysis[1]

  This section uses a combination of economic and sociological organization theory to identify where and under what conditions terrorist organizations have faced, and continue to face the greatest challenges in pursuing their goals. Evidence from declassified Harmony documents (the full text of which is provided in part 2 of this report) and open source material suggest al-Qa’ida faces a number of these same organizational trade-offs and operational constraints. This theoretical frame provides a way of thinking about groups that starkly highlights where we expect to see terrorist limitations and vulnerabilities along with corresponding opportunities governments have to make the terrorists’ goals more difficult to achieve.

Introduction

      Terrorist organizations face difficult tasks in a hostile operational setting. First, they must execute a controlled use of violence as a means to achieving their specified political ends. Doing too much can be just as damaging to the cause as doing too little.[2] Second, they must maintain this calibrated use of force in an environment where becoming known to government equals operational failure.[3]

      These challenges lead to several recurring themes in terrorists’ organizational writings. We see a consistent focus on how to achieve the appropriate use of violence when the rank and file often clamor for more violence than is useful, or seek to enrich themselves in the course of their duties. Groups also struggle with the problem of maintaining situational awareness while staying covert, so that members understand which actions will support the political goal, and which will be counterproductive. Finally, there is regular concern with balancing the need to control operational elements with the need to evade government attention and limit the consequences of any compromise.

      Problems of control in terrorist organizations first enter into the organizational writings of early Russian Marxist groups which had regular problems with local cells conducting revenge attacks that could not be justified by Marxist theory.[4] In like fashion, a 1977 “Staff Report” for the Provisional Irish Republican Army (PIRA) General Headquarters (GHQ) details reorganization plans intended to minimize security vulnerabilities while maintaining sufficient operational control.[5] Islamist groups are not immune from these concerns, though maintaining situational awareness seems more problematic for them. A lessons learned document from the Harmony database describing the failed jihad waged against the Assad regime in Syria from 1976-1982, includes a discussion of the problem of becoming detached from the masses because of the exigencies of maintaining security.[6] This same document contains a discussion of how to emulate the Italian Red Brigades’ to better compartmentalize information while maintaining operational effectiveness. Finally, captured letters between al-Qa’ida members released from Harmony discuss how planning and conducting too many attacks can become counterproductive, bringing unwanted government attention toward the group.[7]

      The key insight is that terrorist groups, and other covert organizations, face two fundamental trade-offs. The first is between operational security and financial efficiency. Here problems of trust and control—agency problems—create inefficiencies in resource allocation. Strategies to mitigate these problems all entail security costs. The second tradeoff is between operational security and tactical control. Here agency problems and other group dynamics lead to counterproductive violence. Strategies to mitigate these problems through greater control entail security costs for groups as a whole. We will explain the enduring importance of these tradeoffs for developing counter-terrorism strategies using theoretical insights developed from the study of firms, political parties, social movements, and traditional insurgencies. There are strong theoretical reasons to believe these problems are inescapable for all terrorist groups; evidence from the Harmony documents and open source accounts reinforce our assessment that al-Qa’ida struggles with similar trade-offs and challenges. Developing a better understanding of how to take advantage of these problems will help government more effectively degrade al-Qa’ida’s (and other terrorist groups’) lethality.

The Terrorists’ Challenge

      The terrorists’ challenge is simple to describe. For security reasons, political and ideological leaders, the principals, have to delegate certain duties—planning attacks, soliciting funds, recruiting, and the like—to middlemen or low-level operatives, their agents. Such delegation poses no problem if all the agents are perfectly committed to the cause and agree with leaders on how best to serve the cause. Under those conditions, the preferences of the principals and their agents will be perfectly aligned, and the agents will act exactly as the principals would like.

      However, preferences aren’t always aligned. When they are not, the covert nature of terrorist groups necessarily implies that agents can take advantage of delegation to act as they prefer, not as their principals would like. We see this type of problem repeatedly among members of al-Qa’ida who exhibit differing levels of commitment to the cause. Thus, L’Hussein Kherchtou, a member of al-Qa’ida’s operational cell in Nairobi during the 1998 U.S. Embassy bombings, testified for the prosecution because he disagreed with the spending priorities of the senior members of his team.[8] He saw their priorities as un-Islamic, essentially charging them with embezzlement. We see this same dynamic in al-Qa’ida’s affiliate organizations. For example, Jemaah Islamiyah (JI) has recurring problems with local leaders engaging in counter-productive violence. In October 2003, senior JI leaders had made a decision to develop Poso, in Central Sulawesi, as a safe haven and area for ideological outreach. However, local motivations led to a series of sectarian attacks against Christians, against the leadership’s preferences. These attacks attracted government attention to the area, destroying its value as a safe-haven.[9]

      Thus, terrorist leaders have a problem. Security concerns mean they cannot perfectly monitor what agents are doing.[10] Moreover, the nature of the operational environment means that it is hard to punish agents, even when leaders do catch them taking unauthorized actions. In terrorist groups, the agents hold an implicit threat over group principals—they can go to the government. For example, Jamal Ahmed Al Fadl, who testified in the Embassy bombing trial, stole money from al-Qa’ida, got caught, went on the run, and approached the U.S. government in an attempt to save himself and his family. Moreover, the agents are often operational elements that specialize in violence, so leaders cannot wield a credible threat against them. This problem plagued the leadership of the Ulster Volunteer Force (UVF) and Ulster Defense Alliance (UDA), protestant paramilitaries in Northern Ireland, who could not put a stop to the politically damaging mafia-style activities of their operational cells.[11]

      The problems outlined above fall into the larger category of “agency problems.” Such problems arise when three conditions exist: (1) a principal needs to delegate certain actions or decisions to an agent: (2) the principal can neither perfectly monitor the agent’s actions, nor punish him with certainty when a transgression is identified; and (3) the agent’s preferences are not aligned with the those of the principal’.[12] The framework of agency theory has been used with great success to explore a wide range of issues including corporate governance, pork-barrel spending in Congress, the behavior of regulatory agencies, civil-military relations, and problems of tactical control in insurgent organizations.[13] Here we use it to frame our discussion of vulnerabilities within the al-Qa’ida network, many of which are revealed by captured jihadi documents in the Harmony database. Understanding why groups face preference divergence, and when preference divergence creates operational challenges, facilitates government actions intended to exacerbate internal organizational problems of the terrorists. Doing so multiplies the impact of other counterterrorism efforts.

      The remainder of this section looks at two specific areas of conflict within terrorist groups: resource allocation and tactics. In the first area, agency problems create inefficiencies in how resources are allocated. Methods to mitigate these problems create operational vulnerabilities. Thus, groups face a security-efficiency tradeoff. In the second area, agency problems lead to cells undertaking politically suspect behaviors. Groups can mitigate this by exercising tighter control, which reduces group security. Thus, groups face a trade-off between security and control. In both cases, there are a number of actions that governments can take to make the trade-off harder for the terrorists.

Why Preference Divergence?

      Open-source analyses of terrorist organizations generally begin from the perspective that members of these groups are uniformly motivated by the cause, are equally willing to sacrifice for the cause, agree on what the cause is, and see eye-to-eye on the best tactics to achieve their strategic end.[14] However, substantial evidence has accumulated to indicate this is not the case. The Harmony documents reveal a surprising level of infighting and conflict, even within highly capable groups such as al-Qa’ida (Al Adl Letter, Harmony AFGP-2002-600080). Historically, terrorist groups have repeatedly splintered into different factions because of differences of opinion about how to conduct the struggle. For example, the Irish Republican Army has spawned at least 6 splinter groups since the mid-1970s, including the Provisional IRA, Official IRA, Real IRA, Continuity IRA, Irish National Liberation Army (INLA), and the Catholic Reaction Force (CRF). Several documents from our sample provide additional evidence to suggest that the cohesion of Islamist terrorist groups is similarly tenuous. The Harmony documents reveal strong evidence of significant disagreements over strategic focus and conflicts over arcane points of doctrine.[15]We next examine why there is such divergence in preferences over spending and tactics.

Preference Divergence over Spending

      The primary cause of preference divergence over spending is a natural selection process that occurs over the course of terrorists’ career paths. Within the population of new terrorist recruits, there is a distribution of commitment to the cause.[16] Even though all may seem quite committed to us, some are always more willing to sacrifice than others.[17] Over the course of many years in the jihad, the most committed members are the most likely to volunteer for risky or inherently fatal assignments. As members of a cohort move into finance and logistics oriented positions, the proportion of less committed members will increase because those more committed remain in comparatively more dangerous assignments and are more prone to be selected out of the population. Note that terrorist organizations typically use individuals who have been around for some time to handle logistical and management tasks.[18] What this career progression means is that, on average, those handling financial and logistical tasks will be more risk-averse and less committed than the leadership or rank-and-file.

      These selection dynamics are exacerbated by the fact that participants in terrorist support networks face dramatically lower levels of risk than tactical operatives. Beyond not being asked to participate in risky or inherently fatal ventures, they are less likely to be dealt with by government forces. When they are dealt with, support personnel are less likely to be killed. And when arrested, they face more lenient treatment. Using biographical and network data on 366 members of al-Qa’ida and affiliated groups, we found that financiers’ have rarely been killed and that their survival rate has been consistently better than that of operators.[19] The capture rate for financiers also tends to be lower than that faced by tactical operatives.

      When governments do succeed in capturing logisticians and other support network members, they face dramatically lower consequences than operators. Only one of the 32 financiers and logisticians removed from the global Salafi jihad between January 2001 and December 2003 was killed. A particularly telling example is the Jemmaah Islamiyah cell which was broken up in Singapore in late 2001. The cell provided fund-raising services to JI and was engaged in making logistical arrangements for an al-Qa’ida attack in Singapore. Of the 30-plus people arrested, the 13 engaged in direct logistical support each received two years in prison.[20] Those engaged in fundraising activities were released but not permitted to leave the country.[21] This risk differential exacerbates the selection effects, as those who take operational jobs because they are extremely committed are more likely to be removed. This process makes it even more likely that those tasked with managing funds and distributing them to operational cells will have different preferences than the leadership.

      Even without this adverse selection process, there is reason to expect preference divergence. The lenient treatment observed for support network members means that the threshold level of risk acceptance and commitment required for participation in support activities is much lower than for participation in tactical roles. Thus, individuals with a given level of commitment might participate in support activities while balking at other roles within the organization. Seeking to maximize operational capability, terrorist groups would concentrate such individuals in support roles, freeing up the true believers for riskier operational duties. These personnel decisions would then lead to consistent variance between levels of the organization.

      Harmony documents suggest al-Qa’ida has formally encouraged such preference divergence within their ranks early on in the accessions and recruitment process. For example, one captured document describes the roles and responsibilities of the various committees al-Qa’ida members can serve on. In this document the Military Committee lists the following goals: Preparation of freedom fighting young men, their training, and organizing them for combat; Organization and supervision for combat participation on the battlefield; Preparation of programs and military procedures; Offering what is needed of military mechanics for combat. Compare this with the four listed goals of the Administrative and Financial Committee included in the same document: Offering the best of administrative services for all the group members and their families; Undertake the work of hospitality for the guests of different kinds in the most generous possible manner of hospitality; Undertaking the work of accounting, keeping books on the front, which safeguards the group’s general funds; Undertaking the financial work for the group.[22] Clearly, the preferences of individuals attracted to supporting requirements such as “providing hospitality” and “keeping books” are likely to differ substantively from those who select into positions responsible for organization and supervision for combat participation on the battlefield.

      Groups may even refrain from centrally-directed personnel movements because they create connections between cells, yet still suffer preference divergence. Because of security considerations, some terrorist organizations recruit directly into specific positions with little opportunity for movement. Terrorist organizations often fill positions using a strategy of recruitment through existing social ties.[23] Any member tasked with the recruitment and early ideological training of potential members will have access to a limited population. From this population, he will need to fill various spots. Commitment to a group’s ideology follows a bell-shaped distribution—with the purely ideological or purely venal individuals being rarer than those who place moderate weight on the cause—meaning that it will be harder for the recruiter to find potential tactical operatives than logisticians. Unless the recruiter knows a huge number of potential members, he will place individuals in the riskiest position they will accept. Thus, individuals will rarely be more ideologically motivated than is necessary given the risk level of their occupation, leading to preference divergence by position.

Preference Divergence Over Tactics

      The causes of preference divergence over tactics in terrorist organizations are somewhat more straightforward, stemming from the very nature of terrorist operations. The first cause is that people who are good at violence, who make ideal recruits as far as their ability to conduct operations, often seek more violence than is politically desirable.[24] For example, al-Qa’ida in Mesopotamia’s campaign of beheadings in 2004 significantly damaged the foreign insurgents’ reputation among Iraqi Sunnis.[25] Early Marxist militants were the first to document this problem. Lenin and others noted repeatedly that those recruited for their ability to conduct military operations often pushed for such activities even when not politically advantageous.[26] In like fashion, the PIRA suffered repeated problems with Active Service Units (ASU), made up of combat specialists, pushing for violence when the organization as a whole wanted to lay low. There is something of a trap here for organizations that adopt limited non-terrorist uses of violence in response to government pressure. Such organizations are often forced to adopt more violent tactics than the strategic situation demanded in order to retain the allegiance of their most radical cells.[27] This trap also affected Italian and German left wing militants. These groups’ dependence on violent factions for survival, given the tactics of state police, pushed them into higher levels of violence against civilians, even when such violence was not politically ideal.[28]

      The second cause of preference divergence over tactics is that the cognitive dynamics of an underground organization—isolation from the outside, negative physical incentives to external contacts, excessively strong affective ties, and the like—mean that operational cells often become divorced from reality, seeking to do more violence than those removed from the situation would like.[29] Islamist groups suffered deeply because of this problem in Syria, where Harmony documents reveal local cells repeatedly made attacks that the outside leadership opposed.[30] The Italian Red Brigades suffered this problem as well. Over time the group had to devote an ever-larger portion of its energies to attacks that appealed only to the membership, attacks that were costly in terms of outside support.[31] In larger groups, where the leadership is organizationally isolated from operational cells, or where it is geographically separated from them, these cognitive dynamics will lead to preference divergence between leaders and operators.[32]

      Finally, competition for prominence within a movement often leads to more violence then political leaders would like. Here, the best example is the Israeli-Palestinian conflict. Hamas began promoting terrorist events largely due to competition with the Palestinian Islamic Jihad (PIJ), even though such actions ran counter to the preferences of the larger Muslim Brotherhood movement.[33] Before the second intifada, suicide missions were intended to undermine not only the peace process but also the legitimacy of Palestinian Authority.[34] After November 2000, in the second intifada, Hamas and other radical organizations relied on the success of suicide missions as a key to gaining popular support. Fatah then invested in suicide attacks to stem the growing popularity on the street of Hamas rather than to win concessions from Israel. It was therefore the dynamics of the factionalized internal politics within the Palestinian movement that prompted suicide missions.[35] In essence, the Palestinian Authority could only reestablish its authority by showing the population that it could play the suicide mission game. However, as the subsequent destruction of Palestinian Authority resources by Israel shows, this was tremendously counter-productive with respect to larger political goals.

      So we see three internal dynamics leading to preference divergence over tactics: (1) individuals recruited because of their skills in violence will tend to seek more action than leaders would prefer; (2) cognitive dynamics of underground organizations will lead operational units to see the world differently than their leaders; and (3) competition for prominence within the movement will lead factions to engage in politically unnecessary actions. All three result in agency problems.

How Groups Respond to Preference Divergence

     This analysis reveals a number of terrorist group strategies for responding to agency problems, all of which create specific security vulnerabilities—the type of tradeoff we highlight throughout this report. One strategy applies primarily to the handling of funds, while other strategies apply more generally.

      Providing funds only on a need-to-have basis is a very effective way in which principals can prevent less-committed individuals from taking advantage of their control over funds. The Embassy bombings in Africa, the Bali bombings, and the 9/11 attacks were funded in this fashion, with operators receiving a certain amount of funds, burning through it, and having to request more from the central leaders. By increasing the frequency of transfers and reducing their size, leaders build up better knowledge about the nature of the relationship between what they spend and the success rate they observe. This reduces the scope of what the agents can get away with. However, because each additional transfer entails communications and financial transactions, there is a security cost to this strategy.

      Auditing strategies are another option for developing better information about what agents are doing. This entails requiring agents to provide periodic, detailed reports on their activities, as al-Qa’ida used to do.[36] These reports effectively make it easier for leaders to know when their agents are behaving differently than they would like. However, this additional efficiency comes at the cost of additional communications traffic, which entails an increased risk of compromise. To the extent that groups believe they have secure communications channels, these strategies will be more likely to be employed.

      Punishing agents who do not behave as the principals would like depends on both identifying such behavior and being able to wield a threat over agents. There are both efficiency and security costs to using punishment. The efficiency cost is that engaging in punishment diverts resources from the struggle. For example, as Hezbollah moved into Southern Lebanon in the early 1980s, it encountered efforts by Israeli proxies to penetrate the organization, and had to create a security bureaucracy to police the organization. While this provided better security, it diverted significant resources from the struggle with Israel. The security challenge is that punished agents can always follow the example of Jamal Ahmed Al Fadl and go to the government. This security cost is especially hard for transnational groups who have agents in areas where they lack operational capabilities, and hence the means to violently punish their agents.

      Another common way to deal with agency problems is to encourage members to enter into trust-inducing relationships such as marriage within their group.[37]Those who enter into marriages within the movement face a larger cost if they are caught behaving against their leaders’ wishes. Not only do they lose a future income stream, but familial and community connections as well. Such a strategy is central to the success of the hawala funds transfer system.[38] Of course, if a member embedded in a dense network of strong ties is captured, myriad opportunities for compromise are created. A second problem for terrorists groups whose members live far away is that they will enter into close relationships outside the network which can dilute commitment, as competing social costs become important. This dynamic was observed in the Syrian case, as the foreign jihadis who married locals often lost their motivation and left the struggle.[39] Thus, government counterterrorism efforts naturally involve scrutiny and attention toward a terrorist’s personal and social network—attention which creates security vulnerabilities for the group. For example, Hamas had to cut ties with a generation of trusted, experienced operational leaders because they were easily traced.[40]

      Terrorist leaders may also reduce preference divergence by requiring initiation rights that either prove their members’ commitment or make it hard to leave the group. The Japanese Red Army followed this strategy, making prospective members commit violent crimes.[41] Some accounts suggest that the training program in Afghanistan served as such a screening process for al-Qa’ida.[42] The problem for groups is that these strategies create predicate offenses that can identify individuals to the government. Former trainees in Afghanistan have received significant scrutiny around the world, with those who evince even tangential ties being arrested in the United States.[43] The lengthy ideological debates that are a critical part of the recruiting process in European Islamic expatriate communities fulfill this function. Indeed, lengthy ideological discussions are an old screening tactic of militant organizations, one practiced by GSPC, the Red Brigades, ETA, and others. However, this strategy may weed out people with useful skills who do not have the patience for lengthy doctrinal debates.

      Overall, strategies to reduce agency problems entail security costs. It is vital for government counterterrorism officials to identify ways for raising theses costs.

The Difficult Challenge of Balancing Security with Efficiency and Control

      Organizations configure themselves and operate in ways that maximize their utility. For businesses, this utility is normally measured by profit. For terrorist organizations, it is most accurately determined by political impact. The maximum political impact a terrorist group can have is constrained by the security environment in which it operates, the efficiency with which it disburses its resources and the degree to which it can control its members.

      As introduced earlier, terrorist organizations face two tradeoffs that create internal discord. The security-efficiency tradeoff creates conflicts over spending when three conditions exist: (1) preferences over spending are not perfectly aligned;[44] (2) principals cannot perfectly monitor their agents’ uses of money or cannot credibly punish them; and (3) resources are constrained so that leaders won’t just accept the financial inefficiencies created by agency problems.

      The security-control tradeoff creates conflict over tactics when three similar conditions are present: (1) preferences over tactics are not perfectly aligned, so that some agents want to attack different targets or want to conduct more or fewer attacks than leaders want; (2) principals cannot perfectly monitor their agents’ tactical planning and cannot wield a threat of violence over them; and (3) political goals are being placed at risk by the freelancing of operational elements. Under either of these sets of conditions, terrorist organizations will face significant internal tension.

      These tradeoffs are illustrated in Figure 1 below. This figure places the level of security on the y-axis, and the level of efficiency or control on the x-axis.[45] Given the level of government security pressure, and the level of preference divergence within a group, we can define a line expressing the feasible security-control tradeoffs. This line, labeled C, starts at the maximum level of security, achieved when the leadership has no control over operations and is decreasing to the maximum level of control or efficiency where a group is operating openly with no security or without any concern for fiscal accountability respectively.

Figure 1: The Security-Efficiency Tradeoff

      There is a set of tradeoffs between security and efficiency as well as security and control that are acceptable given the level of discrimination in the use of violence that groups must exercise to achieve their political goals. These tradeoffs define a series of indifference curves where the nature of this tradeoff is represented by the shape of the dashed, solid, dotted curves (U_1, U_2, U₃) presented on the chart.[46] Higher curves—those moving “up” and “right” in this graph—mean greater utility measured in terms of political impact.[47] The terrorists, seeking greater political impact, will prefer to increase their security, control and efficiency—move towards U_3, the dotted line indifference curve as shown here. Government efforts intended to degrade terrorist capacity can do so by reducing security, control and/or efficiency—moving the group “down” to U_1, the dashed line indifference curve, and its corresponding reduction in potential political impact.

      Government agencies can impact both ends of C. Interventions that reduce the level of security for the terrorist group will force them to accept lesser utility given their preferences for maintaining control and financial efficiency. Figure 2 below depicts this in graphic form. Here the government has taken actions that reduce the maximum feasible level of security from S to S’, and the group has had to respond by moving to a lower indifference curve—U’.

Figure 2: The Impact of Constraining Organizational Security

      Alternatively, a government can encourage agency problems that reduce terrorist groups’ ability to control its operatives or finance its operations. Reducing either or both of these shifts down the maximum possible level of political impact for the terrorist group. This dynamic is represented in Figure 3 below. In this case, the line C represents the possible combinations of security and control, or security and efficiency that a terrorist group can choose from before government intervention.

      Here, the line C¢ represents the case where government has taken actions to reduce the level of control its leaders can exercise, such as by increasing internal dissension within the group. As demonstrated in this illustration, the best feasible tradeoff for the group has shifted down from U to U’, meaning the group has less capability to achieve political impact. Now the group has to accept much less control than before in order to maximize political impact. A similar tradeoff occurs when the government degrades terrorist groups’ financial efficiency and forces them to accept lesser utility measured in terms of political impact.

Figure 3: The Impact of Constraining Organizational Control

      The key intuition represented in these examples is that government actions to make the security environment harder reduce the feasible level of political impact for terrorists. Government actions and environmental changes that increase preference divergence and challenge the control and financial efficiency within terrorist groups have a similar effect. Exactly how these changes will alter groups’ optimal tradeoff between security and control, or between security and financial efficiency, will depend on the exact shape of their indifference curves. These will vary across groups for numerous reasons such as  . In the concluding section of this analysis, we will use this framework to suggest a number of actions to make the terrorists’ organizational challenges harder. Doing so will reduce their ability to conduct attacks and achieve their desired political impact.

continue
back to report home