Reading I NEED HELP ON MY HOMEWORK

Governance of Terror: New Institutionalism and the Evolution of Terrorist Organizations

Author(s): Scott Helfstein

Source: Public Administration Review , Jul. - Aug., 2009, Vol. 69, No. 4 (Jul. - Aug., 2009), pp. 727-739

Published by: Wiley on behalf of the American Society for Public Administration

Stable URL: https://www.jstor.org/stable/27697917

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms

Wiley and American Society for Public Administration are collaborating with JSTOR to digitize, preserve and extend access to Public Administration Review

This content downloaded from �������������128.111.180.8 on Wed, 15 Dec 2021 23:43:27 UTC�������������

All use subject to https://about.jstor.org/terms

Scott Helfste?n Combating Terrorism Center

United States Military Academy

Governance of Terror: New Institutionalism and the

Evolution of Terrorist Organizations

Fundamentally Rethinking the

War on Terror

Scott Helfstein ?s an assistant professor

in the Department of Social Sciences and

an associate at the Combating Terrorism

Center at the United States Military Acad

emy at West Point. He has provided terror

ism policy advice to the U.S. Department

of Defense and has been invited to brief

Special Operations Command on various

aspects of terrorism. His research addresses

grand strategy, terrorism, intelligence, and

research methodology.

E-mail: [email protected]

A preoccupation with network approaches in terrorism studies has inadvertently marginalized the

fact that terrorist groups are subject to many of the

same bureaucratic forces that impact all purposeful

organizations. Because typical organizations are subsumed in the concept of networks, it is curious that

scholars have been so quick to bypass more traditional

models of organizations and bureaucracy that may help us

understand network topology. This article relies on the new

institutional approach to organizational behavior. Using the Coase theorem to explain the costs and benefits associated with different organizational structures, it follows that

counterterrorism efforts may drive some groups toward greater autonomy while

compelling others to adopt common bureaucratic

processes, often referred to as isomorphism. By exploring

the different costs that terror groups face and examining

the characteristics of terrorists associated with different

groups, organizational theory can help explain a divergent trend in terrorism research: leaderless jihad and increased bureaucratization.

In The Management of Barbarism (2004), Abu Bakr Naji recommends that jihadis attend West ern business schools to improve their manage

ment and logistical skills. Naji identifies a major fault with previous jihadi movements: Groups have mismanaged resources, recruits, and violence. This explicit condemnation of prior campaigns high lights the fact that terrorists subject themselves to critical review, are interested in learning about better organization for better operation, confront internal and external constraints, and need to iden

tify the best way to achieve success. These problems and processes confront just about any purposeful organization, and it is important to examine how institutional and organizational forces impact ter rorist groups.

Naji's critique suggests that, in many ways, terrorist organizations may be subject to the same institutional and bureaucratic forces that influence any other

purposeful organization. Some may believe that this is self-evident, while others reject the idea that terrorist

organizations are subject to basic institutional forces, but there is a tremendous amount of explanatory

Viewing terrorist activity through a governance or

institutional lens provides a parsimonious explanation for

the evolution of terror since the

September 11 attacks and the subsequent war on terror.

leverage to be exploited by this simple concept. Viewing terror ist activity through a governance or institutional lens provides a parsimonious explanation for the evolution of terror since the

September 11 attacks and the subsequent war on terror. This is particularly relevant at a time

when terrorism scholars are ad

dressing divergent trends among terror groups and trying to understand the changing nature of the terrorist threat.

Scholars and experts have approached the study of terrorism from a variety of perspectives over the past

eight years. This work has relied on historical study of groups and movements, studies of individuals and their characteristics, game theoretic and statisti

cal treatments, and studies focusing on networks as a structural form. Comprehensive classification and summary of terrorism studies would require a whole article at a minimum (McCormick 2003), and is

not practical here. Instead, this paper fills a gap that lies between two rigorous approaches to the study of terrorism that are popular in the political science and public policy communities: rational choice models and network theory.

Rational choice, specifically game theory, has been used to model different features of terrorism by

focusing on the micro-level decisions of terrorists, governments, and populations. Some of the work has examined the use of terror attacks as a resource

signal (Overgaard 1994), optimal hostage negotia tion strategies (Sandier, Tschirhart, and Cauley 1983),

Governance of Terror: New Institutionalism and the Evolution of Terrorist Organizations 727

This content downloaded from �������������128.111.180.8 on Wed, 15 Dec 2021 23:43:27 UTC�������������

All use subject to https://about.jstor.org/terms

the risks associated with proactive counterterrorism policies (Rosendorff and Sandier 2004), the pos sibilities that terrorist factions negotiate with the government (Bueno de Mesquita 2005b), the way in which government and counterterrorism policies impact recruitment (Bueno de Mesquita 2005a), and the efficient allocation of defensive resources given a rational terrorist opponent (Powell 2007a, 2007b).

The underlying assumption in all of these works is that terrorists, like the governments they compete

with and the citizens they interact with, are economi cally rational in the most basic sense of the term. The

terrorists are actors with certain goals, and they adopt strategies that they perceive as optimal to achieve

those ends. By this logic, the timing or targeting of terror attacks, recruitment activities, and internal

feuds are rational decisions that are representative of motivated individuals and groups.

While the rational choice

literature often looks at strategic

decision making contingent on preferences and competitive constraints, network analysis has focused on the organizational aspects of terrorism. Instead of

deducing the terrorists' strate gies based on their preferences, network scholars examine the connections that link terrorists

and focus on ways of destabiliz

ing those networks (Arquilla and Ronfeldt 2001). Some of this work has traced the

network of the 9/11 hijackers (Krebs 2002), addressed the breaking points of terror cells (Farley 2003), compared collaborative networks to dark networks (Raab and Milward 2003), argued that interdiction is not sufficient to destabilize terror networks (Ken

ney 2003), examined steps needed to destabilize covert networks (Carley et al. 2003), assessed the vulnerabilities of large networks (Carley 2004), and addressed the resilience of terror financing networks

(Basile 2004). One unifying theme is that the network paradigm better characterizes terror groups than traditional organizational behavior paradigms, and counterterrorism will be most effective by focusing on

network-centric policies.

There is no doubt that networks offer a compelling and valuable way to think about terrorist groups, but

it is important to recognize that networks refer to a broad class of social structures, and there is little re

search that explains the variation in network topology.

Given that typical organizations are subsumed in the concept of networks, it is curious that many scholars

have been so fast to bypass more traditional models of institutions, organizations, and bureaucracy. Prior applications of organizational theory to terrorism have suggested that terror groups often act in ways that

are more consistent with organizational than rational behavior (Crenshaw 1987) and identify some of the organizational trade-offs that are unique to terrorists (Shapiro 2006). The network paradigm may seem particularly attractive, but the almost myopic focus on terrorist organizations as covert networks has margin alized the forces that influence organizational structur ing or network topology in the first place. The rational choice literature has assumed that actors are economi

cally rational, and the network literature has presup posed that actors adopt network structures, but there is an important body of literature that examines the

organizational and structural decisions of rational or bounded rational actors (DiMaggio and Powell 1991). There is, therefore, a disjunction or fissure between the rational choice models of strategic behavior and the network study of terror organizations. It would be a mistake to abandon institutional and bureaucratic

theory altogether based on the assumption that they

... this article suggests that

governance approaches to organizational structure offer a solution to one of the most

pressing problems facing terrorism experts today:

understanding the actual nature of the terrorist threat.

offer no analytical purchase.

Contradicting that belief, this article suggests that governance

approaches to organizational structure offer a solution to one

of the most pressing problems facing terrorism experts today:

understanding the actual nature of the terrorist threat. The war on terror initiated after

the September 11 attacks has put great pressure on terrorist

organizations (Hoffman 2003). This has led some ex perts to argue that most senior members of al-Qaeda have been captured or killed, and that the organiza tion no longer exists. These scholars point to a new and dangerous trend, often referred to as "leaderless jihad" (Sageman 2008). This describes individuals and small groups that are unaffiliated with the major terrorist outlets and the utilization of information

technology such the Internet to learn about jihad and connect with others. These groups receive no training or support from established groups, but are intent on carrying out terrorist acts.

A second group of scholars voice concern about the leaderless jihad, but argue that the remnants of al-Qaeda have reorganized, replenished themselves, and are still very much a threat (Riedel 2007). In

fact, they suggest that these organized groups are the primary terrorist threat facing the United States and its allies. They point to the increasing violence and Taliban resurgence in the Federally Adminis tered Tribal Areas (FATA) of Pakistan (Tellis 2008).

Evidence suggests that some Taliban and al-Qaeda leaders have found sanctuary in the FATA and

that they are using this opportunity to reorganize. The presence of al-Qaeda in Iraq may pose serious risks to the United States in light of current troop

728 Public Administration Review July| August 2009

This content downloaded from �������������128.111.180.8 on Wed, 15 Dec 2021 23:43:27 UTC�������������

All use subject to https://about.jstor.org/terms

withdrawals. These groups display a high level of or ganization and bureaucratization, evidenced by seized documents with detailed registries of foreign fighters (Personnel Database for 394 Individuals), resources

(ISI Border Sector Income, Expense, Equipment and Personnel Report), and operating requirements (Template for Weekly Border Activity Report).

At first glance, there is little to explain this seemingly divergent evolutionary pattern. Terrorists are increas ingly unaffiliated Web surfers, at the same time that groups in the Middle East and Asia are morphing into highly effective bureaucracies. While these pat terns seem unrelated, a governance analysis that treats

terrorism like any other rational enterprise provides a surprisingly coherent explanation for this latest evolution.

The point of departure is the Coase theorem, which explains why actors choose to eschew the efficiency of markets in favor

of organizational or nonmar ket structuring (Moe 1984). After showing how market and nonmarket paradigms may apply to terrorism, we examine the costs associated with each

structure and the impact that counterterrorism efforts have on those costs. Counterterrorism

policy may act to increase the transaction costs facing terror

groups, thereby making non market or organizational forms more appealing. Alternatively, organizations often have a more visible footprint because there are more

targets and more connections linking them. Terrorists seek to avoid paying counterterrorism costs and to select structures cognizant of their own strengths and

weaknesses. These features, particularly the similarity among the terrorists themselves, help dictate potential optimal organizing principles.

The coevolution of leaderless jihad and increased bureaucratization matches well with two concepts in organizational literature: autonomy and isomor phism. Autonomy refers to institutions that shed their prior links with supporters in order to develop their own goals and procedures (Carpenter 2001). This is the essence of the leaderless jihad movement, in which small groups of autonomous actors are

self-radicalizing with the help of information tech nology. The nonmarket groups are going through isomorphism, which is the tendency to adopt similar (and perhaps best) practices from similar organiza tions (DiMaggio and Powell 1991). These groups are struggling to develop the right mix of organizational control with bureaucratic operating procedures

aimed at minimizing their exposure to counterter rorism.

Network approaches to terrorist structure have and will continue to offer benefits, and this article is not

suggesting that these efforts be abandoned. It is argu ing, however, that rational approaches to bureaucracy and organizational behavior also have some important things to add to the terrorism discussion and should not be dismissed as outdated and inapplicable. The next section provides the cornerstone for the institu tional arguments by examining the Coase theorem and its application to terror networks. Then the paper offers an examination of the terrorist groups?specifi cally, the level of homogeneity among the actors. The final section shows how the different groups are likely

to respond to the pressures of

The point of departure is the Coase theorem, which explains why actors choose to eschew

the efficiency of markets in

favor of organizational or nonmarket structuring....

After showing how market and nonmarket paradigms may

apply to terrorism, we examine the costs associated with each

structure and the impact that counterterrorism efforts have on

those costs.

counterterrorism, and in so

doing, explains the complex patterns that have led to much debate among terrorism experts.

A Coase Theorem for Terrorism Ronald Coase observed the world around him and asked an

astonishingly practical question: If markets are supposed to be the most efficient means of eco

nomic transaction, why do peo ple constantly choose to adopt nonmarket or organizational structures that are supposed to be economically suboptimal? In short, he wanted to know why

organizations are so prevalent given their inefficien cies relative to supposedly efficient market outcomes (Coase 1960; Moe 1984). Coase's question is the a priori issue of rational organizational behavior. Before studying networks or any other particular structures, it is important to ask why actors would choose the inefficiencies of organizing in the first place.

The point of departure for Coase was the economic efficiency of free markets. Conventional microeco nomic analysis shows that actors benefit from trading at equilibrium levels, and any market distortions,

such as regulation, create deadweight losses that are inherently inefficient. When individuals choose to organize, there is a formal or informal contract that

involves fixed compensation for agreed-upon output. Organizing and contracting establishes set patterns of behavior and requirements outside of the efficient market, which essentially has the effect of distort

ing markets and manipulating equilibria, much like regulation. By bypassing the free market, the organiza tion is theoretically functioning at a suboptimal level relative to the outcomes it would expect if purchasing goods and services directly through the market.

Governance of Terror: New Institutionalism and the Evolution of Terrorist Organizations 729

This content downloaded from �������������128.111.180.8 on Wed, 15 Dec 2021 23:43:27 UTC�������������

All use subject to https://about.jstor.org/terms

According to Coase, the reason individuals choose to adopt nonmarket structures, essentially contracting goods and services, lies in the existence of transaction costs (Coase 1960; Moe 1984). When the costs of accessing markets or completing transactions rise too high, nonmarket structures may become economically optimal. If employers or parts purchasers had trouble locating reliable suppliers, the transactions costs may

make contracting an optimal approach to securing the desired goods. Suppliers, for their part, are happy con tracting, because it ensures them a fixed rate, thereby

eliminating the risks associated with market pricing. When transaction costs are sufficiently high, both demanders and suppliers can derive benefits from nonmarket structure.

Nonmarket structure and contracting, however, are not without their own costs. Contracted employees or suppliers have to be monitored in order to ensure that they are delivering the products or services of agreed upon quality (Radner 1981; Keck and Sikkink 1998). This gives rise to the commonly cited agency problem, in which an employee can shirk his or her responsi bility. Employers, for their part, are left to observe outcomes and question whether poor outcomes are the product of market forces or employee shirking. In

many formulations of the principal-agent problem, monitoring is further complicated by the fact that the agent has more information than the principal and can blame poor performance on externalities when his or her shirking was actually the culprit. The market versus nonmarket decision is a complicated balance of transaction costs associated with accessing the efficient

markets, and the monitoring and contracting costs associated with nonmarket structures. Most exchange happens somewhere on a spectrum between pure mar ket and complete bureaucracy. Many bureaucratized firms incorporate markets within the organization (Baker, Gibbons and Murphy 2001), just as many

markets incorporate institutional principles such as property rights or price systems (Bates 1995).

The logic of Coase's argument relies on the assump tion that actors are economically rational, which is simply to assert that actors choose among behaviors that are most likely to accomplish reasonably well defined goals. If terrorists are the rational agents that game theoretic scholars assert, then these compet ing forces should be taken into account, just as they

would among any other set of rational individuals. In other words, we cannot assume that terrorists are

rational agents and then eschew the rational theories

that provide the foundation for organizational or bu reaucratic behavior. Admittedly, applying the Coase theorem and rational choice explanations of organi zational behavior to terrorism may seem inappropri ate to some who prefer psychological or sociological rationales. Those favoring a sociological approach

would argue that terrorism is a social phenomenon

based on collective goals. This is certainly an impor tant element of terrorism; however, these approaches do not preclude an economic or rational approach to these problems. Elinor Ostrom (1990) examined how institutions overcome individual incentives to

counter collective action problems, an examination of collective goals that explicitly incorporates rational action. Mark Granovetter (1978) relied on individual

incentives as important causal mechanisms in his study of collective action, specifically addressing riot behavior and collective violence. A careful analysis shows the value of applying economic theories of rational institutions to terrorism, and that terrorist

groups may be constrained by some common market forces.

A necessary precondition for applying the Coase theorem to any situation is the existence of a market.

The absence of a market would render the subsequent logic about market and nonmarket structuring ir relevant, as there would be no market alternative. This

begs the question, what does a terrorism marketplace look like? Rational choice treatments of terrorism have

implicitly envisaged the terror market as analogous to the employment market. For example, Ethan Bueno de Mesquita (2005b) examined an interac tion whereby government policies may drive citizens toward terror groups who are interested in recruiting the best of those available. In this structure, there is

some demand for terrorist acts, represented by both the terrorist group and the discontent among the

general population that drives citizens to seek employ ment as terrorists. The terrorist group is the engine of

market demand, though it may be responding to an underlying demand in the population as a whole, and the members of the population represent the supply of people available for employment?or more accurately, deployment?as terrorists.

This represents a supply and demand construct that has the familiar look of a labor market, in which

output is the completion of terror attacks. Employ ers have some need for laborers and prefer to hire as

many people for as low a rate as possible. The number of units demanded is a decreasing function of price. The population represents the pool of people avail able for recruitment, or the supply. Suppliers want to sell as many products as possible for as high a rate as possible, and they are willing to offer greater supply as the price increases. The number of units supplied is an increasing function of the price. In an unencumbered free market, the intersection of supply and demand represents the equilibrium specifying the quantity of units traded at a specified price. Terrorists, therefore, offer some form of compensation to those who are willing to carry out terror attacks. This compensation may take the form of financial support for families,

training, or simply respect and reverence. More people should volunteer to participate in terrorism when

730 Public Administration Review July| August 2009

This content downloaded from �������������128.111.180.8 on Wed, 15 Dec 2021 23:43:27 UTC�������������

All use subject to https://about.jstor.org/terms

this compensation looks attractive relative to other opportunities.

The labor market model of supply and demand in the terrorist context is the most commonly applied and the easiest to envision, but there is an alternative that

is worth mentioning briefly. Traditional notions of

recruitment may be poorly suited to some transna tional terror groups (Sageman 2004). An alternative approach to a terrorism marketplace could involve competition for resources among individuals who are interested in carrying out terrorist acts, making the terrorist the demander. As the resources available for

an operation decline, fewer demanders are willing to accept the terms. The supply of resources is based on the support made available from global jihad organi zations. In this case, organizations increase the supply of resources as the number of qualified candidates capable of carrying out significant attacks increases. This concept flips the supply and demand dynamic of the labor market upside down.

Despite the plausibility of this structure, this analysis will follow others by treating the ter

rorism marketplace as analogous to the labor market, although the actual market structure does

not play a significant role in the ensuing analysis.

The Cost of Doing Business: Market versus Nonmarket Markets, irrespective of the players, have some com mon features. In theory, all that is necessary for a mar

ket system is three actors, where one actor is trying to transact and can choose between alternatives offered

by the other two actors. While pure market systems

require no organizational construct because outcomes should be self-enforcing, there are often institutional

elements such as property rights or price systems to make exchange easier. In a market, any demander can meet a supplier and complete a transaction. Terrorism markets, in their pure form, pose quite the dilemma for counterterrorism officials. There are a few tools that states and counterterrorism officials can use to combat market-based terrorism. The most basic

counterterrorism policy, though the most difficult to enact effectively, would require finding all of the

individual, unrelated market players or constricting all forms of communication between market actors.

Alternatively, states could act in ways that reduce the incentive to participate in terror markets in the first

place by increasing the value of outside options, creat ing a shortage that would impede market function. In short, counterterrorism in a market structure requires

increasing the transaction costs so that markets become inefficient mechanisms or inhospitable places for terrorists to inhabit.

Transaction costs are common in economic dialogue and usually involve costs embedded in a market or necessary to complete economic exchange. Often, these costs involve behaviors such as information

search, locating buyers or sellers, shipping or delivery, and bargaining. This is certainly not an exhaustive list of transaction costs, but it serves as a guide for

application to the terrorist market framework. Some of these issues present a greater challenge for terrorists than others. For example, information search is likely to be more costly than bargaining. The illicit nature of terrorism, coupled with the need for secrecy, makes information search, and communication more gener

ally, relatively expensive. The information needed to complete transactions is rarely readily available, mean ing that actors have to do more research requiring

higher transaction costs. Locating transaction partners may also present problems that are contingent on the operational environment. In Western or well-policed

In Western or well-policed societies, terrorists cannot really go out to the market and openly represent themselves as suppliers

or demanders. On the other

side of the spectrum, terrorists

operating in lawless areas may be less constrained.

societies, terrorists cannot really

go out to the market and openly represent themselves as suppli ers or demanders. On the other

side of the spectrum, terrorists

operating in lawless areas may be less constrained.

Counterterrorism, in essence, tries to increase these trans

action costs in the hopes of discouraging involvement in the terror market. Typical micro

level counterterrorism efforts are aimed at identifying those who are interested in terrorism and remov

ing them from the marketplace. Counterterrorism officials may track people's information searches or plant false information, making information gather ing either overly risky or worthless (Ressler 2006).

Finding transaction partners is also complicated by counterterrorism efforts, as officials may use spies in

the hopes of identifying those who are participating in the market. Even bargaining costs and shipping costs can be manipulated by counterterrorism policy. The former is made more costly by using signals intercepts to reduce communication, and the latter by imposing better border and transit security measures that make

the movement of people or dangerous substances more difficult.

In a marketplace in which transaction costs are already high, counterterrorism can have a large dis torting effect. The terrorists' problem is best expressed in a probabilistic or expected value context, which is appropriate if they are rational actors (Morrow 1994). Transaction costs are increasing in the probability of getting identified. Costs would also be increasing in the actual penalty exacted for exposure or capture. In the expected utility context of microeconomic trans actions, the probability of capture and the associated

Governance of Terror: New Institutionalism and the Evolution of Terrorist Organizations 731

This content downloaded from �������������128.111.180.8 on Wed, 15 Dec 2021 23:43:27 UTC�������������

All use subject to https://about.jstor.org/terms

penalty must be absorbed in the transaction. If actors know that they have to absorb this cost in order to participate in the market, then it might be in their best interest to organize outside the market in the hope of bypassing such costs.

Participation in a marketplace, irrespective of struc ture, must be measured against nonmarket alterna tives. Nonmarket alternatives are selected when

transaction costs get sufficiently high that markets become inefficient exchange mechanisms (Moe 1984). These high costs justify the nonmarket structure and organization that constrain actors' behavior. Unlike market activity, nonmarket activity requires some form of centralization and commitment. It is these

two elements that allow individual agents to avoid the transaction costs that make markets a poor alterna

tive. Centralization may have functions of planning, resource allocation, training, procurement, or recruit

ment. Centralization, or organization more broadly, requires that agents commit to fulfilling their respon sibilities so that the organization maintains an accept able level of output. This requires a formal or informal

contracting mechanism.

The easiest way to conceive of a terrorism market is an environment in which individu

als respond to some demand for terror acts without active

logistical support. In the market context, individuals respond to a community grievance by acquiring the necessary skills or resources (or self-organizing, if operating as cell) needed to carry out an attack. In a

nonmarket structure, the actors carrying out the terror

attacks have the advantage of accessing resources made available by some form of organization or preexist ing resource network. Situations where individuals are vetted, drafted, trained, and supplied by groups show institutionalization, specialization, and hierarchy reflective of nonmarket structures.

Evidence shows that global and regional jihad move ments have traditionally relied on a nonmarket or organizational structure (Byman 2003). Al-Qaeda had a central planning staff that managed resources,

organized recruiting, and coordinated isolated cells. Regional groups, such as Hamas, often have com mand or logistical units outside their primary conflict zone that are charged with providing resources to the in-theater units that actually carry out attacks (Levitt

2004). Many of these groups have organizational structures that would look like almost any other en

terprise or bureaucracy staffed with financial officers, operations officers, strategy officers, and even public

relations officers. Adopting a nonmarket structure

or organizational features is not surprising, as coun terterrorism policies theoretically attempt to increase transaction costs, but nonmarket structures are not

without their own costs.

There are two costs that are particularly vexing for these illicit nonmarket groups: contracting costs and secrecy (or what we call footprint) costs. Contracting involves fixing some agreed level of output for a nego tiated price, allowing the parties to bypass the market place. There are inherent inefficiencies associated with these commitments. The major costs associated with contracting, both products of different preferences or incentives among the actors, are monitoring and enforcement problems. Monitoring involves paying costs to ensure that the contracted parties deliver as

promised, and the difficulty of accurate monitoring can vary with the specific situation (Fudenberg, Le vine, and Maskin 1994). Enforcement involves impos ing punishments on parties who shirk or deviate from their obligations (Fearon 2003; Koremenos, Lipson, and Snidal 2001). These problems are common in analyses focusing on principal-agent problems, but as noted here, this is just one set of problems associated with market/nonmarket decisions and clandestine

groups.

Contracting has some unique dynamics when applied to

terrorism, but forms recovered

from Iraq show that fighters and suicide bombers do sign

contracts for attacks, separation, and temporary leave.

Contracting has some unique dynamics when applied to terrorism, but forms recovered

from Iraq show that fighters and suicide bombers do sign contracts for attacks, separa tion, and temporary leave (ISI

Template for Suicide Op eration Pledge; AQI Template for Leaving Iraq or the AQI

Organization; Template for Overseas Medical Treat ment Authorization). Unlike firms in the economic

marketplace that can rely on writ of law to enforce contracts, there are no external enforcement mecha

nisms available to terrorist organizations. As a result,

illicit groups should have particularly high costs for monitoring and enforcement. These costs can vary sig nificantly, and therefore have a significant impact on organizational behavior. Monitoring becomes increas ingly important as the terrorist leaders and subordi nates have different priorities. Terrorists can overcome these costs by increasing reporting requirements and creating middle management that improves oversight, as al-Qaeda in Iraq has attempted to do (Template for

Weekly Border Activity Report). All of these activities create organizational or operational costs. Enforce ment is also a costly activity because punishments are rarely cost free to those being punished or those imposing punishment.

One might be tempted to ask how important moni toring really is. After all, terrorism is about creating

732 Public Administration Review July | August 2009

This content downloaded from �������������128.111.180.8 on Wed, 15 Dec 2021 23:43:27 UTC�������������

All use subject to https://about.jstor.org/terms

a sense of insecurity and anarchy in the hope of gaining political concessions (see definitions from the State Department, Defense Department, or Federal Bureau of Investigation). The importance of monitor ing and enforcement, however, is directly tied to the risks associated with shirking or defection. There are numerous ways in which terrorists may defect from

obligations. One option is that they choose not to carry out the attack. This defection may be relatively modest if the group has made little investment in the individual, but the defection of well-placed or highly trained individuals may be costly. Enforcement mechanisms may involve threats to family and friends.

A second form of shirking is disregarding leadership wishes and carrying out unsanctioned attacks. Groups may respond by not playing martyr videos or with holding remuneration guaranteed to family members. The monitoring costs may be modest when it comes to attacks, as these are widely observable events, but enforcement is likely to be a serious bureaucratic problem.

Another type of defection is the most dangerous and severe for terror organizations. This is the actual defec tion of terrorists to the government or counterterror

ism authorities. Actors serving as informants or spies

could impose significant costs on a terrorist organiza tion. Given the illicit nature of these organizations, a betrayal of this magnitude would have to be met with harsh punishment possibly involving torture or execu tion. Even if enforcement penalties are set particularly high, a well-placed spy may be able to avoid detec tion for some time, thereby making monitoring and counterintelligence functions central to any terrorist organization (Arquilla and Ronfeldt 2001). This type of monitoring can be particularly costly, and jihad groups have tried to circumvent these costs by relying on kinship or friendship networks when recruiting and planning (Sageman 2004).

This is closely tied to secrecy costs, or what might better be referred to as footprint costs. Nonmarket

structures, such as organizations, have larger foot prints than individuals. Individuals who engage in illicit activity, especially those whose participation is brief, can be difficult to identify or stop. Organi zations, on the other hand, have regularized com munication among certain people, as evidenced by the seized documents cited earlier. Communications

and the connections between people or groups leave a record of interaction that can be used to track

the group (Carley 2004). Regular communication, whether voice or data, can be intercepted. Meetings can be photographed and taped. Because communi cation in organizations is a regular occurrence, the existence of patterns increases the susceptibility of future correspondence and the likelihood of intercep tion. Basically, organizations leave a larger footprint than individuals, and counterterrorism officials excel

at exploiting these connections.

Terror organizations, therefore, are confronted with a

complicated mixture of incentives and costs. Non market structure helps to avoid the transaction costs associated with entering the market, but also involves contracting and footprint costs. By contrast, terror ists can rely on the marketplace, but the illicit nature of the activity makes for an opaque market with high transaction costs. The features of market and nonmar

ket structures, and the associated counterterrorism

policies, are summarized in Table 1. Ultimately, ter rorist groups may settle on models between these two extremes in order to optimize their utility across the

different risks, but these hybrid structures are meant to balance these basic forces.

The incentives of market and nonmarket structures

are further distorted by counterterrorism efforts. Counterterrorism, in its purest form, tries to increase the transaction costs associated with accessing the terror market. Officials can pose as demand

ers or suppliers, restrict money transfers, disrupt connections, intercept communications, and use incentives to discourage participation. The costs of being caught are severe. As discussed earlier, how ever, nonmarket structures are not immune from

counterterrorism pressures. While these groups face lower transaction costs, counterterrorism increases

the contracting and footprint costs. Terrorist groups have to spend more time monitoring members. The footprint costs are particularly problematic. They are fundamental to the functioning of any group, but they also provide counterterrorism officials with their best source of information. In order to limit these

footprint costs, some jihad movements have tried to bureaucratize their secrecy procedures or focus on horizontal rather than hierarchical structuring (Jenkins 2002).

Table 1 Summary of Market and Nonmarket Attributes

Market Nonmarket

Construct Exchange Contract Primary Cost Transaction Monitoring

Relationship Bargained (horizontal) Specified (hierarchical) Counterterrorism Goal Disrupt market functioning Disrupt organization/command and control

Primary Tool Increase transaction costs Increase operating (monitoring) costs Impact Drive terrorists toward nonmarket Drive terrorists toward looser connections/market

Governance of Terror: New Institutionalise! and the Evolution of Terrorist Organizations 733

This content downloaded from �������������128.111.180.8 on Wed, 15 Dec 2021 23:43:27 UTC�������������

All use subject to https://about.jstor.org/terms

Stepped-up counterterrorism since the September 11 attacks has made these choices particularly difficult for terrorists. On the one hand, transaction costs have

increased, decreasing the appeal of loose or market oriented structures. Alternatively, counterterrorism has also increased contracting and footprint costs, decreasing the appeal of the nonmarket structure (al-Hakaymah 2006). While terrorists are caught in a dilemma, terror experts have debated how their foes have reacted to these dual pressures. Without

explicitly considering the rational pressures of markets and organizations, there can be little hope of produc ing an internally consistent explanation of seem ingly divergent trends: leaderless jihad or increased bureaucratization. These questions are perhaps best addressed by thinking about the market and non market issues in combination with an analysis of the terrorists themselves.

Terrorist Characteristics: All Alike? Terrorists try to escape the costs of counterterror

ism, but their means of doing so is contingent on the characteristics of the terrorists themselves. The

incentive structure and capabilities of actors play a crucial role in rational institutionalism (DiMaggio and Powell 1991; Koremenos, Lipson, and Snidal 2003), and it would be a mistake to assume that

these factors do not play a role in terrorism and its

organizing principles. The institutional and organi zational behavior literature address these issues from

a number of angles, such as individual incentives, function-based structure, and organizational culture (Moe 1984; Vaughan 1996; Weiss 1982). Each of these issues, in different ways, addresses the degree of

homogeneity between actors involved in an organiza tion. Homogeneity, a generic term used to represent a degree of similarity between entities, can refer to interests, skills, or other pertinent variables. This interests bureaucratic scholars because of the organi zational problems discussed earlier, and particularly the principal-agent problem.

Despite some rigorous studies of terrorism before September 11 (Hoffman 1998; Lesser 1999; Reich 1998), common conceptions of terrorists, particularly suicide terrorists, portrayed them as psychologically deficient individuals hell-bent on reeking havoc indis criminately (Atran 2004; Sageman 2004). While this characterization is a gross oversimplification, this type

of portrayal still appears in popular media accounts and political speeches. Recent research into terrorist organizations and terrorists themselves reveals a much more complicated picture (Kimhi and Even 2004; Sageman 2004). Islamic extremists who opt to use terrorism may or may not be poor, may or may not be well educated, and may or may not have been edu

cated in madrassas. Despite this mix of characteristics, there are empirical patterns among terrorists and the types of terrorist groups they are associated with.

Marc Sageman's (2004) first canonical study of 170 transnational terrorists, as opposed to those associ ated with regional groups, provided some startling revelations. With few systemic exceptions, these individuals were not poor, and most were actually from

middle-class or wealthy families. Those who were poor tended to hail from poor neighborhoods in Morocco or Western Europe, a group that Sageman calls the

Maghreb Arabs. It also turns out that the vast majority of individuals in the sample were highly educated and were often doctors, engineers, or scientists. By contrast,

the Maghreb Arabs were relatively uneducated. Most in the sample were not educated in madrassas and did not come from religious families. Most were radical ized outside their home countries, often after exposure

to Western society. This profile of transnational terror

groups such as al-Qaeda paints a picture of highly educated, highly motivated, rational, and relatively homogenous individuals coming together to influence political decisions by deliberate targeting of civilians.

Sageman (2008) has recently deviated from his initial opinion that jihadis are likely to be the best and the brightest. Over the past few of years, he has identified a new trend among terrorists. They are unremarkable,

average people. Initially, he reasoned that the proc ess of recruitment, a requirement in any nonmarket

structure, allowed the leadership of groups such as al-Qaeda to screen candidates and only maintain association with the best. As the leaders of al-Qaeda and similar groups have been driven underground, those interested in jihad have to rely on Web sites and local connections to people with similar interests. The lack of formal recruitment, training, or leadership has

changed the profile of terrorists. Groups no longer train, recruit, and screen, meaning that terrorists

are more likely to be average or unremarkable. The emergence of leaderless jihad, Sageman argues, has essentially eliminated any terrorist profile.

Both of these stories about terrorists and their

personal characteristics stand in stark contrast to the image of poor, unemployed, uneducated youths that resonates through the media. It would be wrong, however, to assume that all terrorists are well-off and educated. While terrorist leaders maintain that suicide

bombings are a costly signal because martyrs are often the best and brightest from the population (Atran 2004), regional groups such as Hamas clearly distin guish between leadership and martyr brigades (Kimhi and Even 2004; Levitt 2004). This differentiation is

particularly evident when examining al-Qaeda in Iraq. There, al-Qaeda recruited largely poor and uneducated Sunnis from Saudi Arabia and brought them into Iraq to carry out suicide attacks (Parker 2007). The leader

ship tries to keep count of those recruited to carry out attacks and the resources that they bring with them (ISI Foreign Fighter Registry December 2006-January 2007). This is the type of heterogeneity that gives rise

734 Public Administration Review July| August 2009

This content downloaded from �������������128.111.180.8 on Wed, 15 Dec 2021 23:43:27 UTC�������������

All use subject to https://about.jstor.org/terms

to principal-agent problems in institutional theories of organization.

This means there are two different narratives about

terrorists and their characteristics. On one hand,

terrorists may be educated or average people who connect over a common desire to resort to violence.

On the other hand, bureaucratized groups with well-defined leadership recruit lesser individuals to further the violent mission of the organization. If both of these stories are correct, then there appear to

be two generic types of terror groups: groups with

high homogeneity and groups with high hetero geneity. Groups of seemingly regular individuals that self-organize are likely to have a high degree of homogeneity of interests, and may have a good deal in common by virtue of their interpersonal connec tions. Contrarily, organizations that recruit poor and

uneducated individuals are likely to exhibit higher levels of heterogeneity. Leaders want to further the

cause and strengthen the organization, while a host of issues that include religion, discontent, or financial remittance for family members may motivate their

martyr recruits.

At first glance, the discussion of terrorist character istics seems semantic. Does it really matter whether the participants in these groups have similar interests?

Does the profile of these people help to understand the broader trends in terrorism today? The next sec

tion will argue that it does. In fact, the complicated trends in terrorism can be explained by the forces of

rational institutionalism offered by Coase, the degree of heterogeneity within a group, and the pressures of counterterrorism.

A Tale of Two Trends: Isomorphism and Autonomy Today, and since the September 11 attacks, terrorists find themselves between a rock and hard place. Coun terterrorism efforts increase the transaction costs asso

ciated with a terror marketplace at the same time they increase the footprint and contracting costs associated

with nonmarket structures. The natural reaction to

the pressures of counterterrorism is to seek escape,

and in this case, that involves minimizing exposure to these costs. The best way for terrorists to minimize the costs associated with counterterrorism is conditioned

on their characteristics and heterogeneity.

Terrorists have found two ways they can hope to minimize counterterrorism costs: autonomy and iso morphism. Both of these concepts play a significant role in new institutional approaches to organization al behavior. All organizations seek autonomy?the ability to operate independently, with little or no control from outside or central authorities?over

their missions (Emmerich 1950). Just because they seek it, that does not necessarily mean it is achieved.

Systemic study of autonomous and nonautonomous bureaucratic units within the U.S. government revealed that organizations are most likely to achieve autonomy when mid-level operators within the bureaucracy have good relationships with leaders or outside authorities (Carpenter 2001). In these instances, authority figures believe that mid-level

management is both competent and best suited to structure organizational initiatives. When mid-level

managers have weak relationships with the relevant authorities, it is becomes less likely that organiza tions gain autonomy.

Isomorphism refers to the organizational tendency to take on the structure and operational patterns of other (perhaps successful) organizations in a similar field (DiMaggio and Powell 1991). Isomorphism may be competitive or instrumental. Competitive isomor phism reflects the tendency to adopt the best possible practices, emulating successful organizations. Institu tional isomorphism refers to organizational changes that are intended to accommodate the outside world

that may not reflect competitive pressures. There are

three types of institutional isomorphism: coercive,

mimetic, and normative. Coercive isomorphism is similarity bred by pressure from outside forces, be

it suppliers, financiers or supporters. The more an

organization is dependent on outside actors, the more likely it is to experience this form of isomorphism.

Mimetic isomorphism, the tendency to copy visible organizational models, tends to happen in fields in which there are few organizations to copy or there is

high uncertainty. In environments with high uncer tainty, adopting a familiar organizational form may allay some concerns rooted in uncertainty from opera tions. Finally, normative isomorphism is driven by social ties such professional training.

These two organizational phenomena, autonomy and isomorphism, represent two very different behavioral

patterns that are driven by different forces. Because au

tonomy refers to the severing or loosening of organiza tional ties, at its most basic level, it represents a move

from nonmarket to market structure. Alternatively, isomorphism refers to the adoption of organizational best practices, which inherently reinforces nonmarket

structuring. These two forces have had a significant impact on the evolution of terror organization.

The relative power of these forces is conditional on

the capacities and nature of group. Homogenous groups?those with similar interests?benefit from adopting greater autonomy and essentially adopt ing as close to a market structure as possible. When homogeneity is high, there is less worry that actors will attempt to cheat or exploit, and they need less direction from outside. These actors are less concerned

with the possibility that market partners will not carry out their duties or inform on them because they have

Governance of Terror: New Institutionalism and the Evolution of Terrorist Organizations 735

This content downloaded from �������������128.111.180.8 on Wed, 15 Dec 2021 23:43:27 UTC�������������

All use subject to https://about.jstor.org/terms

similar interests. Greater autonomy, fewer connec

tions, and less regular bureaucratic routine allow these groups to minimize the costs of nonmarket structure,

while minimizing the exposure to transaction costs associated with the market alternative. Homogeneity among participants acts as a natural hedge against the increased transaction costs associated with counterter

rorism policies.

The traditional logic of markets is that they offer the most efficient outcome given actors with different

interests. Here, heterogeneous groups have less incen tive to adopt a market structure because actors with different interests are less reliable partners, and there

is a premium placed on reliability. Heterogeneous groups that adopt more autonomous structures run the risk that subordinates will act in ways that are con

trary to the interests of the leadership. Some terrorist groups, such as al-Qaeda, have relied on a degree of autonomy whereby small units or cells often function for long periods of time with little direction (Byman 2003). Other groups, such Hamas and al-Qaeda in Iraq, have limited the autonomy of subordinate units (Levitt 2004). Simply put, market-oriented structures are not efficient options for groups of heterogeneous actors and probably do little to minimize exposure to counterterrorism costs.

Because principal-agent prob lems are endemic to heterogene ous groups and make market

structure inhospitable, these groups have to adopt the most efficient nonmarket or bureau cratic structures available. This

is isomorphism in the truest sense. These groups cannot necessarily trust their subordi nates, so they must contract, monitor, and enforce.

Adopting the best practices from other organizations is an optimal strategy for avoiding counterterrorism costs. It would be limiting to think that terror groups only experience competitive isomorphism, when they might also undergo institutional isomorphism as defined earlier. Organizations are more likely to experience institutional isomorphism when there is greater reliance on outside actors, greater uncertainty,

fewer organizational models, and common profes sional training. It is interesting to note that the groups experiencing increased bureaucratization are often tied to external actors (such as al-Qaeda central), and their

members often go through some sort of training.

Isomorphism reinforces the bureaucratic structure as groups adopt operational procedures that help mini mize the contracting and footprint costs. Evidence of increased bureaucracy in terror groups is evident from computers and hard drives seized by counterterrorism officials (some of this data has already been cited).

Many of these documents are bureaucratic forms much like the standard paperwork or inventories of any organization. There are also forms outlining standard operating procedures, which is particularly important because these groups need to adopt the best counterintelligence and communication policies if they are to reduce the footprint costs associated with counterterrorism efforts.

An institutional approach to organizational behavior or governance, therefore, provides a parsimonious explanation for the divergent trends of leaderless jihad and increased bureaucratization. Terrorists who are

well positioned to minimize their organizational fea tures, by moving toward market structure and greater

autonomy, have chosen this method for countering the costs of counterterrorism efforts, resulting in the

leaderless jihad movement. Terror groups that require a nonmarket structure have adopted common (and perhaps better) procedures in the hopes of minimizing the costs involved with bureaucracy and confounding counterterrorism officials. There is evidence of these

patterns across groups and functions. Both Hezbollah (Harik 2005) and Hamas (Levitt 2004) have ostensi bly benefited from organizing themselves into military,

political, and social wings that perpetuate terrorist acts, participate in government, and

Because principal-agent problems are endemic to

heterogeneous groups and make market structure inhospitable, these groups have to adopt the most efficient nonmarket or

bureaucratic structures available.

provide social services. There is also convergence in transna tional groups, substantiated by bureaucratization among al-Qaeda in Iraq to emulate al Qaeda who have been keeping records and demanding pledges since the organization's incep tion (Wright 2006). The recent alliance between al-Qaeda and the Salafist Group for Preach

ing and Combat in Algeria, resulting in the creation of al-Qaeda in the Islamic Maghreb, may offer further evidence of common organizational practice.

Given these organizational trends, it seems pertinent to ask, what do autonomy and isomorphism mean for terrorism and counterterrorism? Both forces are

double-edged weapons for both terrorists and coun terterrorism officials. Autonomy reduces the organi zational footprint, making it more difficult to track or even identify members of autonomous groups. As recent terror arrests of autonomous groups in Toronto

and Amsterdam show, it is possible (Sageman 2008). While terrorists benefit from a reduced footprint, the

limitations of minimal structure may also be harmful.

Research on the operational capabilities of centralized versus decentralized groups shows that decentralized groups should have superior problem-solving skills when tasks are simple, but centralized structures perform better as tasks get increasingly complex (Kollman, Miller, and Page 2000). This means that

736 Public Administration Review July| August 2009

This content downloaded from �������������128.111.180.8 on Wed, 15 Dec 2021 23:43:27 UTC�������������

All use subject to https://about.jstor.org/terms

decentralized or autonomous groups may be more effective in carrying out smaller terror attacks, but will find it difficult to carry out complex attacks requiring

significant resources or coordination.

Isomorphism also has upsides and downsides for all involved. Best practices should improve or at least

regularize communication, secrecy, logistics, planning, and financial procedures. Best practices improve or ganizational efficiency and performance, making these terror groups more difficult to counter. The counter

terrorism benefits to isomorphism lie in the fact that

groups adopt similar procedures. This provides a set of clues or cues that counterterrorism officials can use to

identify terror groups and suspected terrorists. Behav

ioral similarities?individual and organizational?im prove the effectiveness of profiling and other means to uncover these clandestine groups. Adopting similar practices also means that counterterrorism strategies that are effective against one group may be effective against others. Once officials find the chink in the

armor, they can exploit that weakness across groups.

Despite the weaknesses of each structure, both trends are daunting for those involved in counterterrorism efforts. Recognizing that your enemy is increasingly

stealthy at the same time that others are becoming increasingly efficient seems to offer little reason for

optimism. This is further aggravated by the fact that the skills needed to tackle increasingly autonomous groups might be very different from those needed to counter increasingly efficient groups. Different wars, or at least theaters of battle, usually require different

weapons and tactics. It would appear that the key is to recognize these divergent trends, acknowledge the fact that a one-size-fits-all counterterrorism policy will not

achieve its objectives, and work toward exploiting the weaknesses inherent in both trends.

Conclusion This essay has argued that organizational theory based on the economic logic of market and nonmarket structuring can help shed light on a major debate

within terrorism studies. It is important to note that

other organizational theories, such as those focus ing on sociology or political mobilization, may offer different perspectives on this phenomenon. Just as this article has argued that organizational frameworks other than network theory can still offer valuable

insights, it is also possible noneconomic theories of organization may be applicable to the issues addressed here. Perhaps further research will focus on the com parative merits of different approaches to organiza tional behavior within terrorism. One of the purposes

of this piece is to show how one traditional approach could shed light on contemporary issues.

Over the next few years, scholars will continue to

produce rational choice models of terrorist activity,

and network research will push on and provide valu able tools in the fight against terrorists. While tradi tional approaches to organizational behavior are likely to be relegated to the shadows of terror research,

we must be careful not to ignore them altogether. Lessons learned from more traditional organizations and bureaucracies may not appear to explain terror organizations, but, as this article has shown, they can be applied and may even have a good deal of explana tory power. Here, we argued that complex divergent trends in terrorism can be explained by a careful ap plication of these theories.

What about these trends? Who is right? Are terrorist groups increasingly decentralized and autonomous, as suggested by advocates of the leaderless jihad vi sion, or are terror groups increasingly organized and bureaucratic? The answer, inconvenient as it might be, is both. These trends are rational responses to the

pressures of stepped-up counterterrorism efforts since the September 11 attacks. One need not create caveats or carve-outs to explain these seemingly unique or divergent trends. Both patterns are reflective of

rational actors responding to organizational pressures in the hopes of minimizing operating costs. While these two patterns appear distinct, they are in fact

opposite sides of the same coin and can be explained by governance theories based on new institutional literature.

Perhaps most important, these divergent trends, and the explanation offered here, reflect evolution ary change implicit in dynamic processes. The terror threat has, like any well-functioning industry or or ganization, evolved over time. Evolution, as slow as it may be, is a central component of organizational and social behavior. Just as post-September 11 counter terrorism efforts helped to trigger the changes among terrorist groups that are now becoming evident, current counterterrorism efforts will spur a new wave

of change. As difficult as it has been to discern the divergent patterns of leaderless jihad and increased bureaucratization, the next stage of evolution may be even more perplexing. Both of these groups will surely evolve.

While the best-case scenario is that the two patterns converge toward a single model, the reality is that these two patterns may fracture even further with new

structures and operating procedures. Counterterror ism experts would be wise to recognize that we will probably need every tool in the social science arsenal to predict, analyze, and understand the next evolu tionary cycle. If the past has any predictive ability, we

will find that terror groups and networks are subject to the same organizational or bureaucratic headaches as everyone else. Perhaps we can take comfort in that knowledge as we retrench for the next wave of this long war.

Governance of Terror: New Institutionalism and the Evolution of Terrorist Organizations 737

This content downloaded from �������������128.111.180.8 on Wed, 15 Dec 2021 23:43:27 UTC�������������

All use subject to https://about.jstor.org/terms

Acknowledgments The author would like to thank Ann Lin, Brian

Fishman, Dominick Wright, panel participants at an Annual Meeting of the Southern Political Science Association, and the anonymous reviewers for helpful comments and suggestions. All errors contained in the paper are those of the author.

The views expressed herein are those of the author and do not reflect the position of the United States

Military Academy, the Department of the Army, or the Department of Defense.

References Abu Bakr Naji. 2004. The Management of Barbarism.

http://www.ctc.usma.edu/naji.asp.

Al-Hakaymah, Muhammad. 2006. Toward a New

Strategy in Resisting the Occupier. http://www.

opensource.gov.

AQI Template for Leaving Iraq or the AQI Organiza

tion. Harmony ID: NMEC-2007-657700. http://

www.ctc.usma.edu/harmony/Sinjar2.asp.

Arquilla, John, and David E Ronfeldt. 2001. Networks

and Netwars: The Future of Terror, Crime, and Mili

tancy. Santa Monica, CA: RAND.

Atran, Scott. 2004. Mishandling Suicide Terrorism.

Washington Quarterly 27'(3): 67-90.

Baker, George, Robert Gibbons, and Kevin J. Murphy.

2001. Bringing the Market Inside the Firm. Ameri

can Economic Review 91 (2) : 212-18.

Basile, Mark. 2004. Going to the Source: Why Al

Qaeda's Financial Network Is Likely to Withstand

the Current War on Terrorist Financing. Studies in

Conflict and Terrorism 27(3): 169-85.

Bates, Robert H. 1995. Social Dilemmas and Rational

Individuals. In The New Institutional Economics and

Third World Development, edited by John Harriss,

Janet Hunter, Colin M. Lewis, 27-48. London:

Routledge.

Bueno de Mesquita, Ethan. 2005a. The Quality of

Terror. American Journal of Political Science 49(3):

515-30.

-. 2005b. The Terrorist Endgame: A Model with Moral Hazard and Learning. Journal of Conflict

Resolution 49(2): 237-58.

Byman, Daniel L. 2003. Al-Qaeda as an Adversary: Do

We Understand Our Enemy? World Politics 56(1):

139-63.

Carley, Kathleen M. 2004. Estimating Vulnerabilities

in Large Covert Networks. Paper presented at the

2004 International Symposium on Command and

Control Research and Technology, San Diego, CA.

Carley, Kathleen M., Matthew Dombroski, Max

Tsvetovat, Jeffrey Reminga, and Natasha Kamneva.

2003. Destabilizing Dynamic Covert Networks. Pa

per presented at the Eighth International Command

and Control Research and Technology Symposium,

Washington, DC.

Carpenter, Daniel P. 2001. The Forging of Bureaucratic

Autonomy: Reputations, Networks, and Policy Innova

tion in Executive Agencies, 1862-1928. Princeton,

NJ: Princeton University Press.

Coase, Ronald H. 1960. The Problem of Social Cost.

Journal of Law and Economics 3(1): 1 -44.

Crenshaw, Martha. 1987. Theories of Terrorism: Instru

mental and Organizational Approaches. Journal of

Strategic Studies 4(10): 13-31.

DiMaggio, Paul J., and Walter W. Powell. 1991. The

Iron Cage Revisited: Institutional Isomorphism and

Collective Rationality. In The New Institutionalism in

Organizational Analysis, edited by Walter W Powell

and Paul J. DiMaggio, 1-38. Chicago: University of

Chicago Press.

Emmerich, Herbert. 1950. Essays on Federal Reorganiza

tion. Tuscaloosa: University of Alabama Press.

Farley, Jonathan David. 2003. Breaking AI Qaeda

Cells: A Mathematical Analysis of Counterterrorism

Operations. Studies in Conflict and Terrorism 26(6): 399-411.

Fearon, James D. 2003. Bargaining, Enforcement, and

International Cooperation. International Organiza

tion 52(3): 269-305.

Fudenberg, Drew, David Levine, and Eric Maskin.

1994. The Folk Theorem with Imperfect Public

Information. Econometrica 62(5): 997-1039.

Granovetter, Mark. 1978. Threshold Models of Collec

tive Behavior. American Journal of Sociology 83(6):

1420-43.

Harik, Judith Palmer. 2005. Hezbollah: The Changing

Face of Terrorism. London: I. B.Tauris.

Hoffman, Bruce. 1998. Inside Terrorism. New York:

Columbia University Press.

-. 2003. AI Qaeda, Trends in Terrorism, and Future Potentialities: An Assessment. Paper pre sented at the RAND Center for Middle East Public

Policy.

ISI Border Sector Income, Expense, Equipment and Per

sonnel Report. Harmony ID: NMEC-2007-657676.

http://www.ctc.usma.edu/harmony/Sinjar2.asp.

ISI Foreign Fighter Registry December 2006-January

2007. Harmony ID: NMEC-2007-657683. http://

www.ctc.usma.edu/harmony/Sinjar2.asp.

ISI Template for Suicide Operation Pledge. Harmony

ID: NMEC-2007-657925. http://www.ctc.usma.

edu/harmony/Sinjar2.asp.

Jenkins, Brian Michael. 2002. Countering al Qaeda:

An Appreciation of the Situation and Suggestions for

Strategy. Santa Monica, CA: RAND.

Keck, Margaret E., and Kathryn Sikkink. 1998. Activists

beyond Borders: Advocacy Networks in International

Politics. Ithaca, NY: Cornell University Press.

Kenney, Michael. 2003. From Pablo to Osama:

Counter-Terrorism Lessons from the War on Drugs.

Survival 45(3): 187-206. Kimhi, Shaul, and Shemual Even. 2004. Who Are the

Palestinian Suicide Bombers? Terrorism and Political

Violence 16(4): 815-40.

738 Public Administration Review July | August 2009

This content downloaded from �������������128.111.180.8 on Wed, 15 Dec 2021 23:43:27 UTC�������������

All use subject to https://about.jstor.org/terms

Kollman, Ken, John H. Miller, and Scott E. Page.

2000. Decentralization and the Search for Policy

Solutions. Journal of Law Economics and Organiza

tion 16(1): 102-28.

Koremenos, Barbara, Charles Lipson, and Duncan

Snidal. 2001. The Rational Design of International

institutions. International Organization 55(4): 761-99.

Krebs, Valdis. 2002. Uncloaking Terrorist Networks.

First Monday 7(4). http://firstmonday.org/htbin/cgi

wrap/bin/ojs/index.php/fm/article/view/941 /863

[accessed April 3, 2009].

Lesser, Ian O. 1999. Countering the New Terrorism.

Santa Monica, CA: RAND.

Levitt, Matthew. 2004. Hamas from Cradle to Grave.

Middle East Quarterly 11(1): 3? 15 - McCormick, Gordon H. 2003. Terrorist Decision

Making. Annual Review of Political Science 6: 473-507.

Moe, Terry M. 1984. The New Economics of Organi

zation. American Journal of Political Science 28(4):

739-77.

Morrow, James D. 1994. Game Theory for Political

Scientists. Princeton, NJ: Princeton University Press.

Ostrom, Elinor. 1990. Governing the Commons: The

Evolution of Institutions for Collective Action. New

York: Cambridge University Press.

Overgaard, Per Baltzer. 1994. The Scale of Terrorist

Attacks as a Signal ofResources. Journal of Conflict

Resolution 38(3): 452-78.

Parker, Ned. 2007. The Conflict in Iraq: Saudi Arabia

Role in Insurgency. New York Times, July 15.

Personnel Database for 394 Individuals. Harmony

ID: NMEC-2007-657674. http://www.ctc.usma.

edu/harmony/Sinjar2.asp.

Powell, Robert. 2007a. Allocating Defensive Re sources with Private Information about Vulner

ability. American Political Science Review 101(4): 799-809.

-. 2007b. Defending against Terrorist Attacks with Limited Resources. American Political Science

Review 101(3): 527-41.

Raab, J?rg, and H. Brinton Milward. 2003. Dark Net

works as Problems. Journal of Public Administration

Research and Theory 13(4): 413-39.

Radner, Roy. 1981. Monitoring Cooperative Agree

ments in a Repeated Principal-Agent Relationship.

Econometrica 49 (5): 1127-48.

Reich, Walter. 1998. Origins of Terrorism: Psychologies,

Ideologies, Theologies, States of Mind. Washington, DC: Woodrow Wilson Center Press.

Ressler, Steve. 2006. Social Network Analysis as an

Approach to Combat Terrorism: Past, Present, and

Future Research. Homeland Security Affairs 2(2).

http://www.hsaj.org/pages/volume2/issue2/pdfs/

2.2.8.pdf [accessed April 3, 2009].

Riedel, Bruce. 2007. AI Qaeda Strikes Back. Foreign

Affairs 86(3): 24-40. Rosendorff, B. Peter, and Todd Sandier. 2004. Too

Much of a Good Thing? The Proactive Response Di

lemma. Journal of Conflict Resolution 48(5) : 657-71.

Sageman, Marc. 2004. Understanding Terror Networks.

Philadelphia: University of Pennsylvania Press.

-. 2008. The Next Generation of Terror. Foreign

Policy, March/April, 37-42.

Sandier, Todd, John T. Tschirhart, and Jon Cauley.

1983. A Theoretical Analysis of Transnational

Terrorism. American Political Science Review 77(1):

36-54.

Shapiro, Jacob. 2006. Terrorist Organizations Inef

ficiencies and Vulnerabilities. Paper presented at

the Annual Meeting of the International Studies

Association, San Diego, CA.

Tellis, Ashley J. 2008. Pakistan's Record on Terrorism:

Conflicted Goals, Compromised Performance.

Washington Quarterly 31(2): 7-32.

Template for Overseas Medical Treatment Authoriza

tion. Harmony ID: NMEC-2007-657669. http://

www.ctc.usma.edu/harmony/Sinjar2.asp.

Template for Weekly Border Activity Report. Harmony

ID: NMEC-2007-658008. http://www.ctc.usma.

edu/harmony/Sinjar2.asp.

Vaughan, Diane. 1996. The Challenger Launch Decision:

Risky Technology, Culture, and Deviance at NASA.

Chicago: University of Chicago Press.

Weiss, Thomas G. 1982. International Bureaucracy:

The Myth and Reality of the International Civil

Service. International Affairs 58(2): 287-306.

Wright, Lawrence. 2006. The Looming Tower: Al-Qaeda

and the Road to 9/11. New York: Alfred A. Knopf.

Governance of Terror: New Institutionalise! and the Evolution of Terrorist Organizations 739

This content downloaded from �������������128.111.180.8 on Wed, 15 Dec 2021 23:43:27 UTC�������������

All use subject to https://about.jstor.org/terms