1

AN ORGANIZATIONAL THEORY OF TRANSFORMATION

This book is motivated by a single empirical fact: at its core, rebel-to-party transformation is an organizational phenomenon. It involves taking a militant organization—a group with a dedicated set of roles, relations between them, norms and institutions guiding behavior, and goals—and transforming it into an organization of a fundamentally different type.1 Making the jump from rebel group to political party requires a different set of roles, different ways of relating, new (and discarded) behavioral norms, and very different goals. Success and failure in rebel-to-party transformation happen at the organizational level—and getting traction on this process requires a comprehensive organizational approach.

To be sure, conflict scholars often lean on organizational intuition to explain important outcomes—including this one. Consider this common and intuitive refrain from the rebel-to-party literature: To transform from a violent militant organization into a viable political party, rebels must “build party structures from scratch” on the heels of war.2 On multiple dimensions, this proposition takes organization seriously. It locates the phenomenon at the organizational level of analysis, it captures the salience of organizational structures, and it implies that transformation—the change in the group’s form—is the outcome in need of an explanation. The problem is not that we are ignoring organizations, it is that we are taking organizations as seriously as our existing toolkit allows.

Before I propose a corrective, an important and outstanding question remains: are rebels thinking about organization as much as I am? The answer is a resounding yes. Militant groups and their leaders are thinking about organizations broadly, deeply, and even humorously—and it would behoove us to follow suit. Indeed, my most unexpected archival find was reference to a running joke among Central American leftists in the 1970s: Q: What do you get when five Salvadorans meet? A: Three organizations.3 The fact of the matter—albeit counterintuitive to some—is that writing a good joke about something demands taking it very seriously. True to form, the origin story of the Farabundo Martí National Liberation Front (FMLN) is one of coalition building, infighting, splintering, and endless negotiations about structure and organizational protocols.

The decade preceding the Salvadoran Civil War (1970–79) was punctuated by a variety of revolutionary leftist groups springing up and splitting off across El Salvador. The regime’s blatant repression in this period exacerbated budding rifts in the underground Communist Party (PCS) over whether to take up arms.4 No longer convinced that peaceful engagement was plausible, a contingent calling themselves the Fuerzas Populares de Liberación (FPL) split off in 1970 to wage a prolonged popular war in the countryside. Two years later, some members within the FPL became frustrated at the lack of military progress and split off to form the Ejército Revolucionario del Pueblo (ERP) and wage a massive, conventional war against the regime. Two years after that, some members within the ERP decried others for being too militant and they split off to form the Resistencia Nacional (RN). In 1976, yet another faction split from the ERP: a group of radicalized students and faculty calling themselves the Partido Revolucionario de los Trabajadores Centroamericanos (PRTC) sought a more politically-oriented revolution across Central America. Finally, in 1978, following a massive voter-intimidation campaign that culminated in a state-perpetrated massacre of civilians, the PCS had a change of heart and formed an armed wing.

Each group was vying for both popular and international support to get the one true revolution off the ground. Then, in 1979, the Sandinistas’ victory against the Samoza regime in Nicaragua established precedent for a successful leftist insurrection while simultaneously creating an opportunity to exploit the military supply chain from Cuba to Central America. There was only one catch. While Fidel Castro indeed viewed the Nicaraguan victory as an opportunity to incite leftist revolutions throughout the region, he made Cuban support contingent on the five groups establishing a united front. Not without considerable drama, they agreed.

The five groups came together under the FMLN umbrella, yet each retained its own organizational structure, its own tactical approach, and its own finances.5 The FMLN was complex and multifaceted: a conglomerate of armies, mass organizations, international and diplomatic wings, and countless alliances with professional associations. In a document recounting the FMLN’s emergence, RN Secretary General Fermán Cienfuegos said the following of the organization: “The agencies we have abroad have correspondents, and the documentation centers have analysts in economics, financial affairs, the political situation in the U.S., as well as other groups of intellectuals and professionals who study other areas. It is a structure that even we do not know, for this, we would have to sit down and study it for a month. At this time, the FMLN does not have an official organizational chart.”6

To be fair, describing the organizational structure in the midst of war is perhaps a difficult and inappreciable task. The question is, how well equipped are we to do it now? The literature is replete with in-depth accounts of the FMLN and sophisticated analyses of the war it fought.7 Moreover, scholars of the Salvadoran Civil War are, overall, quite a harmonious bunch. New research tends to focus more on revealing new dimensions of the conflict than overturning what came before. However, when it comes to the organization at the center of these analyses, researchers’ descriptions of the FMLN sometimes diverge so sharply, one questions whether they are talking about the same group: The FMLN was “fragmented” with a “limited and decentralized structure.”8 Yet, it was “well organized [and] ideologically sophisticated.”9 It was an “unwieldy coalition of quarrelsome groups,” which nevertheless “achieved some cohesion.”10 Tommie Sue Montgomery tells us that RN had “a more formal structure than the other [component groups],” yet, according to Leigh Binford, the ERP “is the best organized.”11 The FMLN was “the strongest guerrilla group to ever emerge in Latin America,” and yet, according to one of its leaders, it was pegada con saliva—glued together with spit.12

The goal here is not to reveal which description is right and which is wrong. The spoiler is that they are all correct. Irrespective of whether they were all wrong, half wrong, or all right, their simultaneous existence points to the same problematic conclusion: political science lacks a systematic framework for describing and analyzing organizations. The problem is reminiscent of the blind men and the elephant: each description accurately portrays some aspects of the organization, but is incorrect or incomplete for others. The contradictions arise because our discipline has borrowed an incomplete handful of organizational concepts and we lack a framework for applying them. These shortcomings, in turn, compromise our ability to describe organizations fully or compare organizations to each other.

This chapter aims to remedy these shortcomings. By first specifying the conceptual and empirical boundary that rebels must traverse as they move from the battlefield to the campaign trail, I identify key gaps in our understanding of rebel and party organizations. Taking an organizational approach, I argue that when rebel groups diversify their wartime structures into domains like governance, political messaging, and social services, their organizational legacies set them up for a smoother and less-risky path into electoral politics than their more homogeneous counterparts.

Defining Rebels and Parties

To understand how rebel groups become functioning political parties on the heels of war—and why I approach this question in the way I do—we first need a clear picture of the actors and dynamics at the heart of this study: rebels, parties, and change. The rebel-to-party story is really about two types of change: an internal transformation in the form of the organization and an external shift in the environment in which it operates. Because the book engages two types of actors and the process of changing from one into the other, we need definitions of rebel groups and political parties general enough to obtain across contexts, yet specific enough to preserve the outcome. Some definitions of political parties, however—for example, “nongovernmental political institutions”—are so broad that they erase the conceptual line between rebels and parties entirely.13 Drawing on my previous work, I propose taking a functionalist conceptual approach to defining the actors at the center of the phenomenon. Motivated by Panebianco’s argument that political parties “can be distinguished [from other organizations] by referring to the specific environment in which they carry out a specific activity,” I define rebels and parties in terms of their functions and their environments.14 By centering (1) their internal functions and (2) their environments, this approach accounts for the start and end points of both types of change.

Rebel groups are armed organizations that operate outside the legal electoral environment by using violence as at least one strategy to influence the outcome of a political incompatibility within a given territory.15 This definition imposes two critical scope conditions for the purposes of studying rebel successor parties. Viable rebel-to-party contenders cannot simultaneously operate as a political party during the conflict.16 Broader understandings of militant organizations may well omit this scope condition, yet it is critical for preserving the outcome: transforming from one type of actor (and one environment) to the other.

In Giovanni Sartori’s words, political parties refer to “any political [organization] identified by an official label that presents at elections (free or not free), and is capable of placing through elections, candidates for public office.”17 This definition is optimal for a few reasons. It specifies the function and environment of party organizations, it takes a catholic approach to electoral openness (making it applicable to a wider range of regimes), and it engages the rebel-to-party literature on its own terms.18 Furthermore, by making office holding a criterion of party status, this definition more closely comports with the agreed-upon stakes of rebel-to-party transition: the long-term integration into legal politics.19

The Limits of Our Organizational Toolkit

Political science is clear on the point that both rebel groups and parties are organizations and need to be studied as such.20 Functionalist definitions provide important insight into the subdivisions undergirding rebels’ and parties’ defining activities; however, they are susceptible to a bias that is particularly consequential for this study. Since our goal in defining political organizations is to answer the question, what makes this type of organization distinct from others?, definitions will focus on what makes the actors unique, rather than what makes them complete. For example, defining rebellions as nonstate organizations that use violence to resolve political incompatibilities is both accurate and misleading. This definition may well cause us to—as Zachariah Mampilly warns against—“reduce militant organizations to their most gasp-inducing components.”21 Even describing rebel-to-party transformation means specifying the structures that change and the process by which change unfolds, yet we lack a consistent and complete set of tools for modeling rebel and party organizational structures beyond their defining features.

To illustrate precisely where current approaches fall short, I turn to Nidia Díaz. Díaz was a commander of the PRTC, one of the five politico-military organizations that make up the FMLN. Figure 1.1 is one of many illustrations of the PRTC’s structure from the journal Díaz kept during the war.22 This organizational chart is one quick sketch of one-fifth of one rebel organization. The question I pose here is, how well equipped are we in conflict studies—or political science, more broadly—to even describe this small slice of the group’s structure?

The answer is, not much. Holding our existing toolkit up to the PRTC’s structure is both bleak and illuminating. Table 1.1 lists the modal concepts used to describe organizations in both the rebellion and the political party literatures. These concepts allow scholars to get some traction on the arrangement of the organization (hierarchy, centralization), the direction and quality of channels through which orders travel (bureaucracy, cohesion, institutionalization), and the size and configuration of combat units (cell structure versus larger brigades). Yet, examining these concepts alongside the organizational chart of the PRTC (figure 1.1) reveals a pronounced gap between the structure of the group and our ability to describe it.

FIGURE 1.1. PRTC organization illustrated by Commander Nidia Díaz. Wartime Diary, Nidia Díaz Papers Collection, box 1, folder 6, Hoover Institution Library & Archives. Arrows, text boxes, and translations added by Sherry Zaks.

TABLE 1.1. Existing organizational toolkit

Two shortcomings highlight the need for a more integrated organizational approach. First, our conceptual inventory is systematically incomplete: most of our concepts focus on the organization’s arrangement to the exclusion of the organization’s content. However, describing the shape of the organization’s components without specifying the scope of tasks the group performs severely restricts how much we know about it. Again, consider figure 1.1: branching from the Central Committee (labeled “C.C.” in the chart) are nine subdivisions of the organization, most of which are dedicated to noncombat tasks such as propaganda, education, and financing. Even at the level of the rank-and-file, only some cells are active combat while others are affiliates of political mass organizations. As an organization, the PRTC exhibits a wide diversity in its roles and subdivisions—and we lack the vocabulary to even acknowledge them, let alone analyze their effects on the organization as a whole. To illustrate this problem conceptually, imagine a hierarchical organization with an entrenched bureaucracy, high levels of cohesion, and a strong commitment among the rank-and-file to the institutions guiding its behavior. Without specifying the different roles within the organization, it is unclear whether I am describing the US military, the Catholic church, or Harvard.

The second issue concerns how we apply the concepts we have. Organizational attributes are generally treated as though they have a fixed valence. Fragmentation is bad for organizations; cohesion is good.23 Bureaucratization confers strength and durability.24 And the list goes on. Even when we acknowledge that a “negative” trait confers an organizational advantage, we are more likely to deem the case puzzling than to rethink how the concept works. For example, Paul Staniland notes that in some cases fragmented structures were crucial for organizational survival. Similarly, in the parties literature, Steven Levitsky observed—contrary to many expectations—that weak institutionalization enabled the Justicialist Party to succeed. What appears digressive, however, is actually quite predictable when we consider the traits in light of the outcome. Both authors were looking to explain organizational adaptation to environmental shifts. As an outcome, however, adaptation is more challenging in the presence of traits that give rise to rigidity.25 The pervasiveness of this problem highlights the importance of casting our outcomes in organizational terms before searching for explanations.

This fixed-valence problem is compounded by a tendency to apply traits “to the organization as a whole, when they clearly apply only to parts.”26 Once again, we find assessments taking the following forms: organizations are fragmented or cohesive, centralized or diffuse, strong or weak.27 This tendency reveals how scholars who study the same organization can alternately characterize it as both fragmented and cohesive, as we saw with the characterizations of the FMLN. We need an approach capable of parsing organizational structures in such a way that we can apply traits only to the pieces that com-port with that description. In the FMLN’s case, for example, fragmentation manifested almost exclusively within the General Command: each group (and thus, each leader) disagreed with the others over the best way to fight the war, leading to discord at the top. Yet, the different groups were remarkably cohesive in their own right. By contrast, Sierra Leone’s Revolutionary United Front (RUF) exhibited pervasive fragmentation: it lacked central control, it lacked horizontal coordination, and loose collections of foot soldiers often made their own decisions. These two different manifestations of fragmentation likely have very different implications for a variety of organizational outcomes. However, existing approaches lack the descriptive nuance to capture this distinction.

What are the implications of these shortcomings? To illustrate the consequences of an incomplete approach to studying organizations, consider the human face as an analogy. A face is at once a collection of individual pieces, a composite of those pieces, and a thing that changes over time in a way that is driven by both its innate features and the environment. Now, imagine a colleague asked me to describe a guest speaker they were picking up from the airport, and I said, “She has brown eyes.” This fact is not false, and it is not entirely useless, but it is incomplete. Eyes are an isolated feature, and color is only one trait of them. Alternately, if I were to just provide an aggregate trait and say, “Her face is very symmetrical,” that would also not be especially helpful. Symmetry—like hierarchy—tells us some information about the arrangement of individual facial features, but not the content of those features: eye color, eye shape, skin tone, nose width, and so on.

Like the human face, organizations exhibit traits at multiple levels: their individual structural components (roles, subdivisions, internal networks; eyes, nose, hair texture), how those structural components aggregate to create holistic features (cohesive, fragmented, hierarchical; symmetrical, feminine, masculine), and how they evolve over time (how an organization adapts to new environments; how a face ages). Without a holistic understanding of how individual facial characteristics vary and interact to render how someone looks, those traits—on their own—lose much of their meaning.

By reducing the concept of organizational structure to the top-down flow of military orders, we restrict ourselves to only examining those traits that describe the path and efficiency with which orders and discipline move from commanders to foot soldiers. As such, many descriptions of rebel groups (and, likely, other organizations) are inadvertently one-dimensional. The limitations of our concepts, in turn, narrow the scope of inferences we can draw, acting essentially as analytic blinders. Scholars interested in describing or comparing militant organizations end up doing so along a limited number of dimensions—because those are the concepts we have. Moreover, an incomplete and imprecise understanding of organizational traits has led scholars to misfile important insights about rebel organizations into insights about rebel behavior.28 Yet, “behaviors” like governance, civilian administration, and social service provision tell us as much about what rebel groups are (their structures, their priorities, their resource allocation) as they tell us about what rebel groups do.29 The takeaway is that current organizational approaches have a wealth of descriptors, but imprecision in what they are describing. In short, we have adjectives in need of nouns.

Building a New Organizational Approach

Though I generally believe the devil is a suspect figure for whom to advocate, I take his side for a moment to address an important question: Why do organizations deserve such special treatment? At the risk of being fanciful, the answer is that once they coalesce, organizations take on a life of their own. With that life comes unique traits, dynamics, goals, and challenges that transcend the individuals who make up the group at any given time. Organizations are actors and collections of actors. They are agents and constraints on agency. Understanding organizational dynamics—how they form, function, transform, and fail—demands an integrated approach that accounts for these unique properties. This approach has three analytic pillars, which I detail in the following pages.

Specify the Outcome in Organizational Terms

Taking an organizational approach first entails specifying the outcome in organizational terms. Rebel-to-party transformation involves taking an organization with structure optimized to work during war and turning it into an organization capable of running campaigns, winning elections, and governing once in office. In organizational terms, this outcome is fundamentally a puzzle about how groups execute a massive organizational change.

Recasting rebel-to-party transformation in terms of organizational change provides a concrete path forward for theory building and analysis. It situates what may appear to be an idiosyncratic phenomenon within a familiar class of outcomes. While rebel-to-party transformations are relatively recent phenomena with a small—albeit growing—literature, organizational change is not. Exploring the process and prospects of change has been core to organizational sociology for decades. This step thus directs us to a vast and instructive literature that identifies the scope of questions a theory of organizational change must address as well as a set of possible answers. What changes? How does change unfold? How does the context factor into the process?30 To explain why some rebel groups succeed while others fail, we must also ask what traits (or other factors) make survival more likely? What provides stability amid the upheaval?

The organizational change literature yields two key insights that help get traction on these questions. First, transformation is less destabilizing to the extent that organizations have preexisting skills that are relevant to their new (electoral) needs. Second, organizations will benefit from a legacy of flexibility enabling them to absorb shocks and quickly reorient these skills into the budding party. Put simply, rebel-to-party transformation will be easier if some parts of the organization are equally useful for being parties as they were for being rebels. Yet, those structures will only be useful if the organization has a history of self-reflection and adaptation.

Recasting rebel-to-party transformation in terms of organizational change proves to be both fruitful and thorny. On the one hand, taking a comprehensive organizational approach ensures that our theories are constructed to explain the full range of questions that the phenomenon raises. On the other hand, the literature points us toward answers that lie outside the scope of our existing organizational toolkit. Consequently, building an organizational theory of rebel-to-party transformation first demands that we take a step back and expand our conceptual vocabulary.

Identify Relevant Structures

Think back to the refrain that motivated this chapter: to become parties, militant groups must build party structures on the heels of war.31 To be clear, I am not arguing that this transformation mechanism is entirely inaccurate. I am, however, arguing that to test whether it is, we need to know what rebel structures are, what party structures are, and what it takes to build, change, and destroy them. If we cannot describe a literal snapshot of an organization’s structure, explaining rebel-to-party transformation—what those structures are and how they change over time—is unthinkable.

Henry Mintzberg begins his foundational text on organizational structure with one of the more intuitive definitions of organizations I’ve yet found: “Every organized human activity . . . gives rise to two fundamental requirements: the division of labor into various tasks . . . and the coordination of these tasks to accomplish the activity.”32 Thus, organizational structure refers to the collection of roles (the different types of labor) within an organization and the types of relations coordinating interaction among those roles. This insight throws the shortcomings of existing characterizations of militant groups into sharp relief: most address only a single type of relation (the downward flow of orders), and most omit discussions of roles entirely.33 To boot, what few roles we regularly acknowledge—leaders, commanders, foot soldiers—paint a combat-centric picture of what rebellions are made of.

ROLES AND SUBDIVISIONS

When it comes to understanding rebel-to-party transformation, roles are the most critical, yet underspecified, component of structure. Roles, according to Parkinson and Zaks, are individual or collective “positions defined by the skills and practices used, the tasks assigned to them, the objectives associated with them, and their relationships to other roles.”34 Thus, roles like foot soldier, platoon leader, or smuggler are defined in part by what individuals in those roles do (fight, lead, smuggle) and in part by the relationships they have with other roles (e.g., a leader is not a leader unless they have people who report to them). Zooming out, individuals are often part of collective roles or subdivisions within the organization, like a platoon, a radio station, a press wing, or a hospital.35 A subdivision refers to a group of people within an organization who consistently work together performing related tasks in service of a specific goal. As we will see later, the diversity of subdivisions holds the key to explaining successful transformation.

Regardless of whether rebels indeed build party structures from the ground up or come to the table with preexisting competencies, modeling rebel-to-party transformation requires a comprehensive description of the structures parties need and the skills (and structures) rebels have at their disposal. By delineating the various types and divisions of labor within organizations, subdivisions directly capture institutionalized skill sets. To foreshadow the theory, this inventory of rebel and party subdivisions will allow scholars to evaluate whether some wartime structures have postwar utility.

Before moving on to the analytic implications, I want to forestall an important question: Why are subdivisions uniquely valuable over individual skills? In other words, is there a meaningful difference between an organization with a dedicated press wing and an organization without a press wing, but whose membership counts a number of good writers among its ranks? The answer is yes, and it boils down to visibility and experience. On the first point, rebel leaders looking ahead to electoral politics must take stock of their organizational resources: What do they have at their disposal? What do they no longer need? Leaders, however, are not omniscient. Organizational subdivisions (say, a wartime press wing) are easily visible from the top; individual skills are not. Even if members in unrelated roles happen to be especially adept at a skill the new party needs, those skills are private information. In short, combatants who also have a way with words are at risk of being unwittingly demobilized.

The second advantage of subdivisions comes in the form of experience. Members of a political-messaging division have spent the war practicing and refining skills with direct application to party politics. As such, they aren’t just good writers, they are good at communicating the group’s message to the people they need to convert from supporters to voters. Extant wings also confer organizational experience: the structure is already integrated into the organization and its members have spent years working together and ironing out their division of labor and norms around communication.36 Thus, membership in an electorally relevant subdivision provides as much experience with logistics and coordination as it does with the skills required to produce content communicating the party line.

RELATIONS

The second component of an organization’s structure is its relations: the social linkages between roles. As “the backbone of organizational structure,” relations specify how different roles and subdivisions interact and where roles are positioned within the organization.37 Thus, relations determine the organization’s shape and internal dynamics. If the key insight driving us to reconsider roles is that not all rebels carry guns, then the key insight driving us to reconsider relations is that not all interactions involve giving or receiving orders. Relations are both deeper and broader than previously acknowledged.38 Most important for the present study, militant organizations exhibit a wide variety of extraorganizational relations. External relations take the form of alliances or rivalries between other militant groups; affiliations with other organizations such as political parties, nongovernmental organizations, or labor unions; and linkages with local populations. Again, with few exceptions, conventional portrayals of the interaction between rebels and “civilians” focus on the production and type of violence or the lack thereof.39 On the ground, however, relations between rebels and local populations range from coercive to familial, from extractive to mutually supportive, from authoritarian to democratic. Since the long-term survival of rebel successor parties depends in part on their ability to mobilize voters, the variety and extent of external relations have critical implications for organizational functioning both during and after the conflict.

Turning to the analytic implications, scholars acknowledge that the skills acquired, practiced, and refined during wartime have consequences for how the war is fought and what rebel organizations do when the war is over.40 A nuanced understanding of organizational structure provides a framework for specifying the scope of relations and skills different militant organizations have at their disposal and how the corresponding organizational differences affect a variety of outcomes. As a result, while we implicitly acknowledge that rebels do more than just kill—they govern, provide social services, write and distribute propaganda—this framework allows us to systematically translate observations of rebel behavior into insights about the group’s structure.

Identify Relevant Organizational Traits

The previous section laid out the building blocks of organizational structure: roles, subdivisions, and relations. To explain how militant organizations survive the massive overhaul that rebel-to-party transformation entails, we need more than just an inventory of useful (and less useful) structures. The first thing we learned when viewing rebel-to-party transformation through an organizational lens was that we do not just have to explain how rebels do it, we have to explain how rebels survive it. What traits promote resilience beyond rebellion? The nature of the outcome requires both resilience (an organization’s capacity to endure shocks and continue functioning in spite of them) and adaptability (an organization’s capacity to function in a new way in response to new demands).41 Once again, our existing toolkit is found wanting. Indeed, many of the traits we invoke most often (bureaucratization, centralization, hierarchy) tend to foster rigidity, inhibiting precisely the type of flexibility rebel-to-party transformation demands.

The organizational literature sheds light on problems, but it does not leave us in the lurch for solutions. In addition to preexisting competencies, organizations are more likely to survive major overhauls if they have traits and practices that confer flexibility: diversification, deference to expertise, and a culture of self-reflection. Stemming from the preceding structural insights, the first trait is diversification. Diversification asks a simple question: Does an organization specialize in one type of task or many? While massive change is always difficult, organizations with a wide scope of expertise on which to draw are more likely to find creative solutions quickly and from within.42 To foreshadow the theory, all types of diversification are conducive to flexibility, but aspiring successor parties will likely benefit from diversification into some domains more than others.

The second trait that facilitates quick adaptation to a changing environment is a culture of deference to expertise. This trait works in tandem with preexisting competencies. Militant organizations may have the skills needed to pivot efficiently into electoral politics, but the leaders must be willing to tap into relevant expertise irrespective of where it falls in the wartime hierarchy. Finally, organizations that practice critical self-reflection (particularly if it is routinized) tend to be more resilient to shocks than their less mindful counterparts. Much like individuals, organizations that view failures as an opportunity for growth and keep close tabs on environmental risk will develop the skills to see and respond to threats quickly. When put into practice, this cycle of failure-reflection-adaptation creates what Debra Minkoff calls a “repertoire of prior flexibility,” which matters because one of the best predictors of future resilience is past resilience.43

An Organizational Theory of Rebel-to-Party Transformation

In the preceding pages, I listed the key components of an organizational theory of change and the two traits that make change easier for some groups than others. Explaining any organizational transformation means accounting for what changes, how change unfolds, and whether the environment shifts alongside the organization. Surviving even the smallest change is more likely when groups have preexisting competencies and a legacy of flexibility. The problem (and reason for the interlude between that paragraph and this one) was that our existing tools were inadequate to even list the structural differences between rebels and parties—let alone to systematically identify the structures and traits that would make transformation possible. Expanding our conceptual toolkit allows us to bridge the disjuncture between the intuition that rebels must build party structures and our inability to specify what those structures are and what structures rebels had to work with. Here, I build a three-part organizational theory of rebel successor party formation, which corresponds to the driving questions in the organizational change literature: What changes? How does change unfold? How does context factor into the process?

The Content of Change

The most basic account of organizational change must specify how the group differs between the beginning and the end of the transformation. What about the structure is altered? What remains the same? According to conventional accounts of building rebel successor parties, the answer to “what changes?” is “everything.” Taking an organizational approach to the question reveals two facts that disrupt this assertion: (1) rebel groups are often much more structurally diverse than previous depictions imply, and (2) change is easier when groups have preexisting competencies. The first part of the organizational theory of transformation synthesizes these two insights. After disaggregating political party organizations into their core subdivisions, I use this blueprint of what parties need to evaluate the postwar utility of the different structures rebels have.

In an investigation of the changing strength of parties, Richard S. Katz and Peter Mair lay out a framework that sheds light on parties’ organizational diversity. They identify—albeit without organizational parlance—three core subdivisions of the party: the party in public office, the party on the ground, and the party in central office.44 The party in public office refers to the governing apparatus.45 The party on the ground refers to the subdivision that forges linkages with society, including party congresses and ancillary organizations (e.g., youth wings, women’s wings, and student organizations). Finally, the party in central office refers to the subdivision that coordinates campaigns, oversees media and press, and acts as a conduit between the other two wings. Of course, party organizations will vary in the composition of and diversification beyond these structures. The goal here is not to suggest that political parties are uniform, but to identify the structures that support their most central needs. In doing so, this framework strikes an important balance: it is specific enough to identify the tasks associated with each major subdivision of the party, but general enough to obtain widely across party organizations.

This blueprint of party organizations allows us to specify the types of structures that rebels need in order to function as a political party. Thus, at the very least, we can more concretely specify what “party structures” refers to when we are told that rebels must build them on the heels of war. Notwithstanding the value of this newfound clarity, the organizational framework allows us to take the analysis a step further. Rather than simply asking what types of structures rebels need to build when war ends, this insight into party organizations raises a more important question: are there some rebel groups that don’t need to build much at all?

Looking beyond combat and logistics wings, I identify a collection of wartime subdivisions that functionally mirror the core elements of party organizations. Wings dedicated to governance, social service provision, and political messaging imbue rebel groups with structures, routines, and skills that are critical for functioning in electoral politics. Since—in addition to their wartime utility—these subdivisions have direct electoral utility, I refer to them as proto-party structures (irrespective of whether future party building was part of the organizational plan). Proto-party structures play a critical role in the formation and performance of rebel successor parties. Because organizations specialize in “doing the same things in the same way, over and over,” rebel groups with party-relevant structures will have less to learn and less to disrupt as they pivot from the battlefield to the campaign trail.46 As such, these proto-party structures represent the preexisting competencies that facilitate smoother transformations.

The Process of Change

Since—as the empirical record suggests—wartime organizational structures vary dramatically in their postwar relevance, the next logical question is how different organizational legacies affect the process of party building. As such, the second part of my theory unpacks the mechanism of transformation. Following a postwar audit in which leaders assess the gap between what they need and what they have, the process of building a rebel successor party will effectively unfold in one of two ways: reconstruction or repurposing. If the audit reveals a sizable gap between the organizational needs of a party and the organizational resources of the rebellion, then the party-building process will unfold as the conventional wisdom suggests. Groups that end the war with more homogeneous, combat-centric organizations will have no choice but to dismantle their militias and build the governing, messaging, and outreach subdivisions from the ground up. This process will often involve recruiting specialized talent from outside the organization to staff new wings while simultaneously integrating the new subdivisions into the organization. Consequently, the literature is not incorrect in identifying reconstruction as a path to party formation; the problem is that this transformation mechanism is inherently unstable. Building even a single wing from scratch means hiring skilled personnel, sorting the division of labor, and establishing new routines for both working together and performing whatever task they were hired for (be it writing or governing or community organizing). This process is not impossible, but it is inefficient, it compromises organizational legitimacy, and it leaves a lot of room for error.

If, alternately, the postwar audit turns up a collection of subdivisions whose skills and functions match the needs of the party, then the transformation process can unfold by repurposing existing wings and only dismantling what is no longer needed. Herein lies the value of proto-party structures. In the first place, repurposing wings staffed with members who were part of the armed struggle helps preserve organizational legitimacy. When familiar pieces of the organization remain intact during change, organizations reduce the likelihood of internal conflicts that arise when longtime loyalists are passed over in favor of outside talent. Second, repurposing existing subdivisions is a more efficient and less risky party-building mechanism.47 When the process of change unfolds by building on what is already there, resources and information can flow through existing channels to an existing subdivision to be managed by people who have experience working together. Thus, the extent to which rebel groups can repurpose existing subdivisions whole cloth into the new party is the extent to which the organization benefits from the skills, relations, and routines that come with it.

The reality of postwar party building lies somewhere between these two extremes. Rarely will a group need to build absolutely everything from the ground up and just as rarely will they have a fully functional inbuilt party ready to hit the campaign trail. Yet, the more proto-party structures the rebel organization has, the less risky transformation will be. Each structure acts as an anchor point, stabilizing the organization amid an otherwise turbulent time.

The Context of Change

One question remains: if it looks like a party and it quacks like a party, can it win like one? If the stakes of rebel successor party formation hinge on whether former belligerents have access to legal (and peaceful) channels for dissent, then the final part of the theory must address whether becoming a party in form actually affects whether groups become parties in function as well. This step in the theory shifts focus from transformation (the internal organizational changes required to build a party out of the pieces of rebels) to transition (the external pivot from operating on the battlefield to operating in electoral politics). What affects whether successor parties can mobilize voters, win seats, and hold onto those seats once they’ve won them?

Whether rebel successor parties adapt to electoral politics is a function of three factors: their capacity to perform electoral and governing tasks, their capacity to mobilize voters, and the institutional constraints of the new party system. The party’s wartime organizational legacy shapes the first two factors in crucial ways. Performing electorally relevant tasks is the most straightforward application of proto-party structures. Groups with experience in governing and political communication with the masses will hit the campaign trail with at least some of the necessary skills built in.

The real question, however, is how one-time rebels go from winning hearts and minds to winning votes and seats. When it comes to electoral performance, voter mobilization is so important that it borders on obvious. As such, the literature is rich with explanations: territorial control, administrative capacity, populist ideology, political wings, and networks of ethnic support all operate through this mechanism. The resources associated with these wartime experiences should translate into an ability to convert wartime supporters into postwar voters.48 To be sure, popular support during the war is likely a good predictor of whether citizens will want to vote for the rebel successor party. Wartime support, however, does not explain the long path from political alignment to casting a ballot. Voter mobilization requires registration (which itself requires official forms of identification and basic literacy), advance knowledge of when the election is, transportation to the polls, and a belief in the baseline safety of showing up and voting for the party you want in office. Without explaining how the party acquires the capacity to assist in the logistical undertakings required to turn supporters into voters, demand-side theories that focus on wartime support will only get as far as explaining voter galvanization, but not mobilization. Truly explaining the latter means accounting for the new party’s organizational capacity to channel excitement into action.

Once again, taking an organizational approach reveals the otherwise elusive mechanism connecting wartime support to postwar votes. Beyond the advantages they bring to organizational change, many proto-party structures maintain consistent relations with locals as well as other organizations (such as trade unions, religious institutions, and student groups). Crucially, these relations transcend violence and extraction. By regularly interacting with citizens in administrative and, indeed, mobilizational capacities, groups with proto-party legacies can leverage existing relations and experience to help manage the logistics of voter mobilization.

Where Do Organizational Structures Come From?

The book’s original (and considerably duller) subtitle was “How Wartime Organizational Structures Affect Rebel-to-Party Transformation.” While the new one is a fair bit catchier, it also obscures the heart of the theory: that variation in wartime structures influences postwar party success. To wit, structure is so integral to the theory that we required a new vocabulary for describing the many ways it varies. Yet, the centrality of rebels’ wartime structures—to say nothing of the ink spilled on a framework for describing them—raises an important question: where do these structures come from?49 Why do some groups diversify into a host of sociopolitical domains like governance and mass mobilization while others remain more homogeneous and combat-centric? More consequentially still, how do we know whether organizational structures are independently salient or whether they are by-products of an antecedent (and simpler) cause that also predicts rebel-to-party success?

These questions are not trivial. From a practical standpoint, if a simpler explanation accounts for both organizational structure and successful transformations, then it obviates the approach laid out earlier and the theory that follows. And the implications of that finding would not be limited to this study. Any evidence that structures are epiphenomenal to organizational change would disrupt a decades-old literature in sociology. What would this evidence look like? For one, we would need to identify a variable that could reasonable determine both organizational form and rebel-to-party success. Then, we would need evidence that groups with comparable prewar traits built comparable organizations and experienced similar outcomes. Finally, we would need evidence that the mechanism of transformation is not a function of organizational form, but of the initial trait that also happens to shape structure.

I turn to the rebellion literature to identify the factors most likely to account for wartime organizations (and, potentially, postconflict outcomes). Although few scholars examine wartime structure as their main outcome of interest, and fewer still theorize the origins of noncombat subdivisions, the literature reveals four explanations of how rebel organizations coalesce and why they take the forms they do.50 Broadly, existing explanations attribute variation in militant structure to one of four factors: ideology, state context, resource endowments, and prewar social networks.51 These factors, though important, lack a direct mechanism by which they translate into a consistent organizational form (let alone the successful postwar transformation of that form into a political party). Take ideology, for example. Though ideology has been shown to affect patterns of violence, recruitment practices, quotidian interactions and socialization, claim making, and the institutions the group adopts, groups with similar ideological stances often build very different organizations.52 In other words, the process by which ideology translates from the red (or green or any other color) book into the organization’s form remains elusive. As I demonstrate in the following chapter, even rebel groups that emerge within the same ideological, social, environmental, and resource contexts exhibit fundamental differences in their structures.

Existing explanations nevertheless provide a useful starting point for theorizing structural origins. Moving beyond this starting point demands a deeper understanding of the unique features of the early organizational context and an explication of how each variable works in an organizational medium. In their embryonic stages, organizations lack the inertia and institutional constraints we normally associate with them. As such, this context is akin to a critical juncture, except rather than deviating from an existing path, leaders are forging a path for the first time. Absent these constraints, leaders will have considerably more agency over major organizational decisions as the group coalesces.53 Of course, agency is not without limits. As Staniland rightly notes, leaders “do not have the freedom to make whatever kind of organization they want.”54 What, therefore, shapes and constrains their decisions about the kind of organization to build?

I argue that the nature and extent of structural diversification in militant groups is a function of the group’s organizational ideology, which, in turn, mediates how they use their resource endowments and the prewar social networks from which the group emerges. This explanation builds on insights from the literature by accounting for how these variables operate in an organizational context, which is where existing explanations fall short. In reality, ideologies need actors to harbor them. Resources need actors to wield them. Social networks need actors to comprise and reproduce them. And the properties of those actors matter.55

Organizational Ideology

Organizational ideology refers to the principles and priorities that emerge when a broad ideological or military doctrine (like Maoism or Conservatism) is filtered through a decision-making body in a specific context.56 Thus, for rebel groups, organizational ideology will define how the struggle should look on the ground: what type of war they fight (e.g., conventional war, prolonged popular war, acts of terrorism and sabotage), what tasks are part of the struggle (e.g., violence against the government, education of the people, international messaging), and what their objectives are (e.g., regime change, Communist revolution, power sharing). The implications for structure are straightforward. Leaders who espouse ideologies emphasizing the importance of political education and mass mobilization will likely aim to build an organization with diverse and specialized subdivisions capable of performing these tasks. Thus, where ideology generally refers to an established political doctrine that exists independently of the actors who espouse it and the context in which they operate, organizational ideology captures the context-sensitive interpretation of that doctrine and its corresponding implications for rebel group structures and institutions.

In this sense, we can think of organizational ideology as the rebel group equivalent of denominationalism. For example, Christianity, writ large, is a religion founded on the teachings of Jesus Christ and the core beliefs that he (1) is the son of God and the messiah, (2) died for our sins, and (3) will return. Within Christianity, however, are a variety of denominations with logically consistent practices. Different churches cleave to (and reject) different principles, organize and worship differently, and embrace distinct ways of engaging both their congregants and the masses beyond them. As such, there are some instances where saying “Bob is a Christian” can tell us a lot of information (say, on the question of polytheism). Yet, there are other instances where saying “Bob is a Christian” tells us very little (say, on the question of same-sex marriage). If we want a distinctly Christian heuristic for the latter, we would make much more reliable guesses about Bob’s politics if we knew whether he belonged to an Evangelical church or a Unitarian one. Such is the utility of organizational ideology. Though—for now, at least—we lack such clean denominational labels in rebellion, organizational ideology captures the logically consistent differences even between groups espousing the same dogmas.

Crucially, organizational ideology can also help get analytic traction on groups whose political ideologies are elusive. After all, some leaders’ preferences are shaped by factors other than “published ideological doctrine.”57 Nevertheless, these preferences are often as consistent as they are observable. For example, while exiled intellectuals formed the initial core of Sierra Leone’s RUF—setting the stage for a robust political education apparatus grounded in a clear ideology—the leadership was distrustful of the intellectual elite and made early efforts to marginalize their role in the organization.58 As a former combatant recalled: “[Leader] Foday Sankoh promoted the semi-literate because they were more loyal to him and were less likely to take over the movement. He did not like the educated ones.”59

Members of the elite core were barred from promotions within the organization and many were exiled entirely. Thus, in a sharp departure from their roots, the RUF was notoriously considered ideologically bereft and its organization comprised almost exclusively combat units and networks dedicated to illicit mining.60 However, the absence of political doctrine does not mean the absence of a cohesive set of ideas dictating the organization’s structure and objectives. Organizational ideology allows us to capture the organizing principles of war irrespective of whether they are attached to a theoretical doctrine. Without it, analysts can easily fall into the trap of characterizing organizations as ideologically bereft even though they operate according to consistent organizing principles.

Organizational ideologies, in turn, shape how groups will use and allocate resources and how they leverage the social bases from which they emerged. Both additional factors represent key opportunities and constraints for the nascent rebellion’s structure. Yet, they do not constrain or endow all rebellions equally. If organizational ideology allows us to infer the type of organization rebels want to build, the resources and personnel to which they have access will determine just how close they can get. By shedding light on the group’s wartime priorities, organizational ideology can reveal not only what types of resources rebel groups will seek, but also how resources will be allocated within the organization once they are acquired. This frame can thus provide insight into the types of structures that may emerge during war if and when resource endowments allow for it. Prewar social networks, likewise, affect wartime structures in more ways than we have previously documented. Beyond assisting in recruitment and contributing to group cohesion, prewar and wartime networks affect structures by determining the range of skills emerging organizations have at their disposal.61 However, whether and how those skills will translate into corresponding subdivisions is also a function of organizational ideology. Thus, two groups that emerge with strong links to teachers unions, say, may still build very different organizations depending on whether their leadership espouses an ideology that views political education as central to the struggle.

Rebels’ wartime organizations are irreducible to a single prewar trait or condition, but that does not mean they are unpredictable. Rebel leaders spill a great deal of ink adapting their political ideologies to their organizational contexts. Taking this more granular approach to ideology allows us to trace how these views give rise to distinct organizational forms. In the following chapter, I delve into more detail about how to identify organizational ideology and I test whether this approach adequately predicts the organizational development of the FMLN.

Bringing an Organizational Approach to the Micropolitics of Rebellion

In Jeremy Weinstein’s foundational book on insurgent organizations, he argues that “a clear understanding of the micropolitics of rebellion is achieved by focusing on how groups organize violence.”62 Of course, studying how groups organize violence—their strategies for combatant recruitment and training, their technologies of war, and their pattern of violence against civilians—is a crucial area of conflict research. How groups organize and produce violence is an important dimension along which rebel organizations vary. However, it is only one dimension. The exclusive focus on the production of violence deceptively casts rebel groups in a unidimensional light. As a result, many analyses of militant group dynamics are built on untested assumptions about the actors at the center of their theories. Rebel organizational structure refers to more than just the arrangement of combat units or the chain of command that connects them. Moreover, as Mampilly, Parkinson, and others have observed, even understanding how rebels produce violence is better achieved by acknowledging the myriad wings of the organization that do not brandish guns.63

Doing so, however, required taking a step back and reexamining the very nature of organizations in the first place. Insights from organizational sociology, when synthesized into a cohesive approach, help open the black box of intraorganizational phenomena: how organizations emerge, how structural components and ideologies combine to imbue organizations with different traits, and how they change over time. Taking a more comprehensive approach to the study of rebel-to-party transformation—and militant groups more broadly—has three central implications, which together facilitate new avenues for research within and beyond the scope of this book. In short, a clear understanding of the micropolitics of rebellion is best achieved by focusing on how groups organize.