How to Party? [PDF]

Vanger (1963), Real de Azúa (1964), Barrán and Nahum (1979), Panizza (1990), and Caetano (2011). The pervasiveness of

25 downloads 29 Views 2MB Size

Recommend Stories


HOw TO ORGANIzE A NEIGHBORHOOD BLOCK PARTY
Love only grows by sharing. You can only have more for yourself by giving it away to others. Brian

[PDF] How to Draw
We may have all come on different ships, but we're in the same boat now. M.L.King

PdF How to Rap
Ask yourself: Where am I making my life more complicated or difficult than it has to be? Next

[PDF] How to Draw
It always seems impossible until it is done. Nelson Mandela

Quick How-To PDF
Ask yourself: When was the last time I did something nice for myself? Next

Cocktail Party Menu (PDF)
We must be willing to let go of the life we have planned, so as to have the life that is waiting for

how to make money with a virtual party
Be who you needed when you were younger. Anonymous

Externship Party Agreement (PDF)
Life is not meant to be easy, my child; but take courage: it can be delightful. George Bernard Shaw

How to find us (PDF)
Ask yourself: How confident are you in your abilities to make decisions for yourself? Next

PDF How to Teach English
Ask yourself: What worries me most about the future? Next

Idea Transcript


PONTIFICIA CATÓLICA DE CHILE INSTITUTO DE CIENCIA POLÍTICA

How to Party? Static and Dynamic Party Survival in Latin American Consolidated Democracies

Por Fernando Rosenblatt

Tesis presentada al Instituto de Ciencia Política de la Pontificia Universidad Católica de Chile para optar al grado de Doctor en Ciencia Política

Profesores Guía: Juan Pablo Luna y Kurt Weyland Comité: David Altman, Valeria Palanza y Julieta Suárez Cao

Septiembre, 2013 Santiago, Chile ©2013, Fernando Rosenblatt

©2013, Fernando Rosenblatt Se autoriza la reproducción total o parcial, con fines académicos, por cualquier medio o procedimiento, incluyendo la cita bibliográfica que acredita al trabajo y a su autor.

30 de Septiembre de 2013

Ejército 333 Tel: +56-2-26768408 [email protected]

1

Para Yamhi

2

AGRADECIMIENTOS El proceso de elaboración de esta tesis contó con el invalorable apoyo de numerosas personas. En primer lugar, mis tutores Juan Pablo Luna y Kurt Weyland. Fue un honor y un placer haber contado con sus comentarios, sugerencias, críticas y orientación desde el momento en que se inició esta aventura. Esa orientación excedió ampliamente lo estrictamente relacionado con esta tesis. Ellos me guiaron en el largo proceso de formación y me brindaron sus generosos consejos para dar los primeros pasos en la vida académica. Son y serán mis padres intelectuales y siempre les estaré agradecido.

Durante estos cinco años del doctorado tuve la posibilidad de conocer a grandes amigos; Rafael Piñeiro, Felipe Monestier, Sergio Toro, Germán Bidegain y Ana María Farías. Con ellos compartimos grandes charlas sobre nuestros proyectos y el desafío de sacarlos adelante. Esta tesis y todo mi pasaje por el doctorado se vio enormemente enriquecido por la generosidad de cada uno de ellos. Pude construir una gran amistad y siempre conté con ellos incondicionalmente. Además, con muchos de ellos he tenido la oportunidad de emprender nuevos proyectos e investigaciones. Me siento privilegiado por contar con este grupo de amigos y colegas. Rafael Piñeiro me recibió desde los primeros días en su casa, ayudándome desde el inicio aún bajo la inesperada lluvia de Santiago. Largas charlas y proyectos fueron surgiendo. Además, me brindó grandes ideas que se transformaron en componentes esenciales de esta tesis. Jamás voy a olvidar toda su incondicional amistad.

Desde el primer día del doctorado en el Instituto de Ciencia Política de la Pontificia Universidad Católica me sentí muy a gusto por un clima de trabajo exigente pero al mismo tiempo muy humano. Estoy muy agradecido y feliz por haber sido parte de las primeras generaciones del doctorado en Ciencia Política. Especialmente, quisiera agradecer al profesor Alfredo Rehren, quien fuera el primer director del programa. Siempre tuvo la paciencia y la disposición para lidiar con mi ansiedad y me ayudó mucho a poder concretar, entre tantas cosas, la estadía en la University of Texas, Austin. También, les agradezco a los profesores Anthony Pezzola, David Altman, Valeria Palanza, Pierre Ostiguy y Rodrigo Mardones. A su vez, quisiera agradecer a Julieta Suárez Cao quien me hizo muchos comentarios y soportó varias lecturas de los capítulos

3

de esta tesis. Finalmente, estaré siempre muy agradecido con Sonia Silva, Yohanna Jepsen y Nora Farfan por toda su calidez y ayuda.

Además, agradezco la oportunidad de financiamiento de parte de mi doctorado a la beca otorgada por Conicyt en su programa de Formación de Capital Humano Avanzando.

Este proyecto se vio enriquecido por los comentarios que también realizaron muchos colegas y profesores. En especial, agradezco a los participantes del Seminario del Instituto de Ciencia Política de la Pontificia Universidad Católica de Chile. En las dos oportunidades en las que presenté este proyecto recibí invalorables comentarios y sugerencias que me permitieron mejorar la tesis. En este sentido, quiero agradecer a Antoine Maillet, Roody Reserve, Pierre Ostiguy por sus comentarios en dichas instancias y en posteriores encuentros. Durante el doctorado, tuve además la oportunidad de hacer una estancia en el Department of Government, University of Texas at Austin que me permitió trabajar de cerca con Kurt Weyland y conocer otros profesores y estudiantes de doctorado. Cuando me encontraba en la etapa inicial de mi investigación participé del Congreso de LASA de 2010. Allí también recibí muy valiosos aportes, especialmente de Mariano Torcal. Más adelante en el tiempo, tuve la posibilidad de presentar mi trabajo en la Maestría en Ciencia Política de la Universidad de Costa Rica, donde recibí muy interesantes aportes especialmente relacionados al caso Costarricense. Finalmente, fueron muy importantes las críticas y sugerencias que recibí en la research design session en el Institute for Qualitative and Multi-Method Research, Syracuse en 2012. Allí, tuve la oportunidad de recibir aportes y conversar con James Mahoney, Jason Seawright y Paula Muñoz.

El proceso intelectual de desarrollo de esta tesis fue desde un primer momento un ejercicio colectivo. Más allá de los aportes ya mencionados en instancias específicas, diversos colegas y profesores me hicieron comentarios que resultaron aportes muy valiosos desde el punto de vista teórico y metodológico para mejorar este producto. De este modo, quisiera agradecer profundamente a Daniel Chasquetti y Daniel Buquet. También, recibí muy buenas sugerencias y comentarios de Omar Sánchez, Gary Goertz, Kenneth Roberts, Alfredo Joignant y Patricio Navia.

4

El trabajo de campo fue toda una gran aventura. En cada país donde trabajé encontré la enorme generosidad y la ayuda profesional de muchas personas. Indudablemente, la buena disposición de los 148 entrevistados fue fundamental. Sus testimonios me permitieron desarrollar y mejorar la investigación y muchos de ellos también me ayudaron a concretar entrevistas con sus colegas. En Chile, el apoyo del CEP y CIEPLAN resultó de suma importancia. Las entrevistas fueron realizadas en el marco de una investigación que desarrollamos junto al profesor Juan Pablo Luna como parte del proyecto “Democracia con Partidos.” En este sentido, quisiera agradecer a Lucas Sierra y Francisco Javier Díaz por permitirme utilizar para esta tesis el trabajo de campo realizado para dicho proyecto. Además, Pilar Giannini y Germán Bidegain me ayudaron con la concreción y realización de algunas de las 53 entrevistas. En Uruguay, a pesar de ser mi país de origen, muchas personas también ayudaron en el trabajo de campo. El apoyo de Edmundo Canalda, Javier Castro, Emiliano Cotelo, Rafael Piñeiro y el trabajo de mi asistente de investigación Nicolás Portela hicieron posible la realización de 44 entrevistas en tiempo record.

El trabajo en Costa Rica merece un comentario aparte. El desafío allí era inmenso. No conocía el país, más allá de haber leído mucho sobre él. En este sentido, diversas personas me hicieron sentir en casa y me introdujeron con ese maravilloso país. Eugenia Zamora fue clave en todo el proceso. Largas conversaciones antes de viajar y toda su ayuda mientras estuvimos allí junto con mi compañera Yamhi, hicieron que la experiencia fuera muy disfrutable y que el trabajo saliera a la perfección. Además, encontré en Jorge Vargas Cullell un gran colega y amigo que también se preocupó en todo momento por nosotros. Ileana Aguilar, Sary Montero, Juliana Martínez y Gerardo Hernández y Kevin Casas también me brindaron su calidez y apoyo antes, durante y después de mi estadía en Costa Rica. Finalmente, el trabajo profesional de mi asistente Lhiam Vega fue de suma importancia para el éxito del trabajo allí.

Durante el período en el que escribí la tesis tuve la oportunidad de empezar a trabajar en la Escuela de Ciencia Política de la Universidad Diego Portales. Me siento privilegiado por la confianza depositada. Además, agradezco la posibilidad de terminar de escribir mi tesis desde allí y por las facilidades que se me brindaron para que esto fuera posible. Por ello, quisiera agradecer al decano de la Facultad de Ciencias Sociales e Historia, Manuel Vicuña y a la directora de la Escuela de Ciencia Política, Rossana Castiglioni. Gracias al aporte de la Facultad

5

de Ciencias Sociales e Historia pude contar con el trabajo profesional de Elaine Thomas quien me ayudó a perfeccionar la redacción en inglés de esta tesis.

A lo largo de los más de cinco años en los que estuve inmerso en el doctorado, la familia y los amigos fueron muy importantes. A mis padres, Julio y Ana que, por ejemplo, cargaron con decenas de kilogramos en libros, mis abuelos Erika y Friedel, mi hermana Beatriz, mi cuñado Javier y mi sobrino Maty, mis suegros Rosario y Ricardo y mi cuñado Federico que siempre me hicieron sentir un hijo o un hermano más; todos ellos me regalaron su amor y calidez y también hicieron posible que llegara hasta el final. Mis amigos de la vida Javier, Sergio, Naor, Jaime, Diego, Idan, Daniel y Fabián supieron acompañarme con su buena onda haciendo “el aguante” de siempre.

Para el final, Yamhi. Ella es la compañera y mujer que ni siquiera fui capaz de desear en mis mejores sueños. Su amor fue y es incondicional. Me acompañó en este sueño con su calidez y alegría, su comprensión, sus palabras justas, sus consejos, su maravillosa cocina que hacían las delicias de los finales de largos días de trabajo. Transcribió decenas de entrevistas, compartió conmigo la aventura de Austin, luego la de Costa Rica y construimos juntos nuestra vida en Santiago. Sufrió y se alegró conmigo en cada instante y me ayudó a crecer. Fue y es la luz de este hogar. Por todo esto y por lo que es imposible de expresar en palabras, a ella le dedico esta tesis.

6

Table of Contents Abstract....................................................................................................................................... 10 Introduction ............................................................................................................................... 11 Overview ................................................................................................................................ 11 The Puzzle .............................................................................................................................. 13 Defining the Dependent Variable ....................................................................................... 17 Operationalizing the Concept of Party Survival .......................................................... 18 The Parties Analyzed........................................................................................................ 21 The Theory ............................................................................................................................. 22 Research Design .................................................................................................................... 25 Why study these countries as instances of party survival? ........................................ 25 Differences among Equals?.............................................................................................. 27 Methods and Data ............................................................................................................. 28 Why Another Study of Party Systems?.............................................................................. 30 Outline .................................................................................................................................... 31 Chapter 1. Existing Arguments ............................................................................................... 32 Introduction ........................................................................................................................... 32 1.1. The Stability Literature: Revisiting the Concept of PSI and PI ................................ 32 Pictures without motion? ................................................................................................. 39 1.2. The Change Literature ................................................................................................... 41 1.3. Explanations from the “negative pole” ....................................................................... 48 Conclusion ............................................................................................................................. 50 Chapter 2. The Theory: Exit, Trauma, Purpose, and Ambition .......................................... 52 Introduction ........................................................................................................................... 52 2.1. A Dynamic Approach to Institutional Analysis ........................................................ 53 2.2. Introducing Conditions for Types of Survival of Political Parties .......................... 59 2.3. How are Exit, Ambition, Trauma, and Purpose Conceptualized ........................... 62 Exit....................................................................................................................................... 62 Trauma................................................................................................................................ 64 Purpose ............................................................................................................................... 66 Ambition............................................................................................................................. 73 2.4. How are these dimensions related?............................................................................. 76

7

Pairs of Interactions .......................................................................................................... 77 Illustration of Interactions of Causal Conditions ......................................................... 79 Conclusion ............................................................................................................................. 82 Chapter 3: A Guide for a Systematic Analysis ...................................................................... 83 Introduction ........................................................................................................................... 83 3.1. Codification of each Condition .................................................................................... 83 3.2. From Unit of Observation to Unit of Analysis ........................................................... 94 3.3. Observing the Outcome ................................................................................................ 97 Conclusion ........................................................................................................................... 102 Chapter 4. Chile ....................................................................................................................... 104 Introduction ......................................................................................................................... 104 4.1. General Overview ........................................................................................................ 105 4.2. Configuration within each Party ............................................................................... 109 PPD.................................................................................................................................... 109 RN...................................................................................................................................... 113 DC...................................................................................................................................... 116 PS ....................................................................................................................................... 124 UDI .................................................................................................................................... 134 Conclusion ........................................................................................................................... 145 Chapter 5. Costa Rica.............................................................................................................. 146 Introduction ......................................................................................................................... 146 5.1. General Overview ........................................................................................................ 147 5.2. Configuration within each Party ............................................................................... 154 PAC ................................................................................................................................... 154 PLN ................................................................................................................................... 161 PUSC ................................................................................................................................. 168 ML ..................................................................................................................................... 174 A brief comment on the left in Costa Rica ................................................................... 176 Conclusion ........................................................................................................................... 178 Chapter 6. Uruguay ................................................................................................................ 179 Introduction ......................................................................................................................... 179 6.1. General Overview ........................................................................................................ 180

8

6.2. Configuration within each Party ............................................................................... 187 PN ...................................................................................................................................... 196 PC ...................................................................................................................................... 201 FA ...................................................................................................................................... 205 Conclusion ........................................................................................................................... 216 Chapter 7. All Things Considered: Causal Conditions and Party Survival ................... 218 Introduction ......................................................................................................................... 218 7.1. The Routes to Dynamic Survival: Conditions in and over Time ............................ 218 7.2. Conjunctural Causation .............................................................................................. 223 Dynamic Survival............................................................................................................ 223 Relative Importance of Configurations Associated with Dynamic Survival ......... 227 Static Survival .................................................................................................................. 229 7.3. Discussion of Alternative Empirical Hypotheses .................................................... 232 Are worldwide trends responsible for lack of vibrancy? .......................................... 232 Does Presidentialism inhibit dynamic party survival? ............................................. 233 Is vibrancy lost with years in government? ................................................................ 234 Is government success or failure key to explaining vibrancy? ................................. 234 Is corruption a causal condition for lack of dynamic survival? ............................... 235 Party organization as such ............................................................................................. 235 Conclusions .......................................................................................................................... 236 Chapter 8. Conclusion: Towards Consolidating a Dynamic Perspective ....................... 237 8.1. Core Components of the Study .................................................................................. 237 Research Questions and Puzzle .................................................................................... 237 Deductive and Inductive Construction of the Argument ......................................... 239 Main Empirical Findings................................................................................................ 241 8.2. Empirical Reach of the Argument ............................................................................. 244 Two Cases of Party Survival.......................................................................................... 244 8.3. The Road Ahead ........................................................................................................... 250 8.4. Dynamic Party Survival and Democratic Representation ..................................... 251 Bringing Dilemmas Back In ........................................................................................... 253 Appendix 1. List of Interviewees .......................................................................................... 256 References ................................................................................................................................ 260

9

Abstract

How do political parties remain significant and vibrant organizations? This comparative study of political parties in Chile, Costa Rica, and Uruguay explains how parties’ vibrancy is maintained and reproduced over time. The study identifies the complex interaction between four causal conditions: Exit, Trauma, Purpose, and Ambition. These conditions are jointly sufficient for explaining the difference between two types of party survival, dynamic and static. Dynamic parties are more resilient. They are stable over time and, unlike statically surviving parties, also remain vibrant. When the four conditions are present, parties exhibit dynamic survival. Other combinations also explain this type of party survival, albeit to a lesser degree. The conjunction of Exit, Trauma, and Purpose is also sufficient for dynamic survival, as is Exit together with Ambition. These combinations, however, produce dynamic survival less reliably. Tracing these conditions historically over time in each party examined, the analysis also shows that Trauma and Purpose temporally precede Exit and Ambition. Collective organizations are more resilient and vibrant if they are first united by retrospective or prospective loyalty (respectively activated by Trauma and Purpose). However, Exit and Ambition need to then follow to facilitate a party’s dynamic survival over time.

10

Introduction Overview

Latin America is a region characterized by new, short-lived parties (Coppedge 1998a, Mustillo 2007). Some argue that the problem is that most party systems in the region have emerged since the Third Wave of democratization, an era of individualism and globalization that hinders party organization. Others claim that presidentialism, the predominant type of government in the region, personalizes political competition and precludes the consolidation of political parties. From this perspective, more than a problem of timing, institutions account for parties’ fragility. However, some parties and party systems in the region have had resistant organizations. In some cases parties were stable, provided myths and were true symbolic assets. They were also valued channels through which to pursue a political career. For many decades, this stylized description was valid in some Latin American countries. The story of the traditional parties in Uruguay (Nacional y Colorado), Partido de Liberación Nacional in Costa Rica, or the Socialist, Communist, Radical, and Christian Democratic parties in Chile are illustrative. They were stable and, to varying degrees, valued institutions. The same could be said of the Liberales and Conservadores in Colombia and AD and COPEI in Venezuela. However, some of these parties have collapsed, while others are still alive but have lost their significance as institutions. Finally, some are success stories of survival under changing circumstances.

In light of this regional variation, some studies have sought to account for differences in parties’ and party systems’ stability. A portion of the literature has focused on distinguishing stable and unstable party systems, using the concept of party system institutionalization (Mainwaring and Scully 1995, Mainwaring 1999, Jones 2005). However, some of the instances of institutionalization they identified have since collapsed, as in the case of Venezuela, while others have been suffering a gradual but 11

persistent process of erosion irreducible to the effects of exogenous shocks, as in the case of Chile. In this sense, the emphasis on stability has obscured profound differences among allegedly “stable” parties and party systems. Alone, this approach cannot explain why some of the so-called “institutionalized” parties and party systems remain vibrant organizations while others have declined, or why “stable” parties’ value as institutions of democratic representation shows significant signs of deterioration in some consolidated democracies but not others.

Another line of research has analyzed collapsed parties and party systems (Morgan 2009, 2011) or processes of “deinstitutionalization” (Boudon 2000, Dargent and Muñoz 2011). The analysis of these exceptional cases responds to one of the flaws of the Party Systems Institutionalization (PSI) literature: it left a vacuum when it came to explaining instances of collapse of institutionalized cases. These studies look at why some parties survive and others do not, but do not address the great differences among “surviving” parties.

Others have explained parties’ changes in light of exogenous shocks (e.g.Coppedge 1998b, Levitsky 2003, Roberts and Wibbels 1999, Roberts forthcoming). Those studies analyze the differential ability of Latin American parties to react to dramatic and changing circumstances. These studies’ use of the concept of adaptability is also not useful for understanding parties’ uneven ability to operate as channels of democratic representation over time and across historical junctures. Though they frame their research in terms of democratic representation, the authors dissociate adaptability from parties’ role as agents of democratic representation, and the concept of adaptability is operationalized as a strategic choice made by leaders in light of shocks (see Levitsky 2003 pp. 3 and 7-8). Moreover, it is always depicted as a sudden decision of one or a few leaders occurring without deliberation inside the organization (see Chapter 1 for a thorough discussion).

12

The focus on stability or on exceptional instances of change (collapse or adaptation) has thus obscured understanding of the evolution and survival of parties in different consolidated democracies in Latin America. Based on this approach, one could falsely conclude that nothing relevant has happened in the “stable” (i.e., institutionalized) party systems: their parties are stable and have not collapsed, and they have not been subject to urgent needs to adapt —or have been subject to such needs but often without experiencing the dramatic changes commonly analyzed by the literature. In short, less has been said about the process of survival in cases of stability.

In contexts of institutionalized partisan survival, beyond merely explaining stability, it is also relevant to understand what makes some “survivor” parties deteriorate while others remain vibrant. I aim to explain partisan ossification, in contrast to reproduction of a party’s vibrancy across time and historical junctures. The purpose of this dissertation is to closely examine an overlooked but theoretically relevant problem: what enables some stable and institutionalized parties to survive while remaining vibrant organizations that retain their value over time, while others do not?

My analysis identifies two types of survival that differ very clearly in terms of parties’ capacity to remain vibrant and valued institutions. With the help of this distinction, I provide a causal mechanism to explain how parties’ vibrancy and relevance is maintained and reproduced over time. To explain different types of survival which, in turn, help to account for different results in terms of democratic representation, I investigate parties in three seemingly similar Latin American democracies: Chile, Costa Rica, and Uruguay.

The Puzzle

The study of party and party system survival amid change has been addressed in the literature of comparative politics in general (Huntington 1968, Duverger 1954, Sartori 13

1976, Katz and Mair 1994, Dalton and Wattenberg 2000, Benoit and Laver 2006), and in the study of Latin American countries in particular (Dix 1992, Mainwaring and Scully 1995, Mainwaring 1999, Levitsky 2003, Hunter 2010). In the region, the topic received substantial attention following the transition from authoritarian regimes to democracy.

Party system institutionalization (hereafter, PSI) was conceived as an independent variable for explaining democratic consolidation. PSI has been mainly regarded as the stability of actors and rules of the game, which was thought necessary for achieving democratic consolidation (Mainwaring and Scully 1995, Mainwaring 1999). However, within a few years, some of the party systems Mainwaring and Scully had identified as “institutionalized” experienced major problems. For instance, in 1998 the Venezuelan party system collapsed (Morgan 2007, 2011), and Colombia has undergone a process of “deinstitutionalization” (Boudon 2000, Dargent and Muñoz 2011). In other cases, such as Chile, electoral stability has coexisted with declining legitimacy, decreased rooting of parties in society, and weakened partisan organizations (Altman and Luna 2011). Finally, in Costa Rica there has been a persistent process of electoral dealignment (Sánchez 2003, Lehoucq 2005, Alfaro Redondo 2006), where new parties have emerged and are diluting the traditional division of Liberacionistas and Anti-Liberacionistas that had been consolidated for almost two decades in a two-party system. The Uruguayan party system, by contrast, has not experienced these sorts of problems. Classification of these cases as members of a common “institutionalized” category now seems to provide little analytical leverage for understanding these subsequent, often divergent, transformations.

There have also been many problems at the party level within the systems classified as institutionalized. Some parties have almost collapsed, as in the case of the Costa Rican PUSC. Others have been unable to consolidate organizationally despite some successes, as in the cases of the PAC in Costa Rica and the PPD in Chile, both of which have had

14

basic organizational problems. Other parties in these systems have achieved stability and kept their vibrancy, as in the cases of the FA in Uruguay and the UDI in Chile.

The picture presented above suggests a theoretical puzzle. I argue that the concept of PSI as used by the comparative politics literature has little analytical power to explain party survival and to differentiate between types of party survival. Some have noted the PSI literature’s conflation of attributes, causes, and consequences of institutionalization (Altman and Luna 2011, Zucco 2010). Subsequent developments and critiques thus call into question both the theoretical clarity and explanatory utility of the concept of institutionalization as recently used in the Latin American PSI literature. Theoretically, my study seeks to respond to some of these difficulties.

The reliance on snapshots (e.g., public opinion polls of a specific year, or electoral volatility for a particular set of elections) and subsequent emphasis on stability have actually conflated very dissimilar cases. The concept of “institutionalized” parties may be descriptive, but it is certainly not helpful for explaining significantly different trajectories of apparently similarly institutionalized parties and party systems. Some institutionalized party systems have collapsed, while others have lost their ability to channel representation. Existing definitions and operationalizations of PSI have proven insufficient for explaining: a) different results in terms of the evolution of particular stable parties and party systems; and b) whether a particular party or party system remains vibrant and valued for channeling political representation. The latter is essential for better understanding the actual relationship between party system institutionalization and democratic representation. To account for the divergent empirical outcomes now apparent in this regard, one needs to address the dynamics feeding party survival and the party system as a whole, particularly the underlying mechanisms enabling parties, qua organizations, to persist and retain legitimacy and support among principals—their voters and militants.

15

Modernization and globalization increasingly account for parties’ difficulties to remain as relevant institutions (Dalton and Wattenberg 2000). One could therefore argue that the puzzle advanced above might no longer be relevant and that the causes of parties’ irrelevance are entirely separate from those of parties’ survival. Accordingly, one might argue that we should instead focus on other long-term processes affecting societies (Mainwaring and Zoco 2007, Inglehart and Welzel 2005, Dalton and Wattenberg 2000, Webb and White 2007a). At the same time paradoxically, it is still widely accepted that parties are needed for democratic representation (Hagopian and Mainwaring 2005). How might these apparently conflicting insights be reconciled? This is precisely one of the goals of this project: to improve descriptive and causal inference so as to understand how and why parties remain valued institutions across junctures in some, but not other, developing and unequal democracies. The Partido dos Trabalhadores in Brazil (Workers Party), Frente Amplio in Uruguay (Broad Front), Unión Demócrata Independiente in Chile (Independent Democratic Union), are all dense and vibrant parties which, to varying degrees, have succeeded in making themselves relevant organizations. Studying how parties can remain vibrant and legitimate organizations over time and across junctures therefore remains relevant despite the overarching transformations and challenges associated with modernization. Parties can remain significant and stable institutions in settings where existing theories would predict otherwise (Dalton and Wattenberg 2000, Dalton and Weldon 2007). To account for this phenomenon, I analyze major parties in consolidated democracies in Latin America.

More precisely, the study addresses two unanswered questions. First, how does a party regarded as institutionalized and stable avoid becoming ossified, rigid, and detached from society? Second, how do some stable (i.e. institutionalized) parties retain the necessary flexibility to adapt to new challenges, while keeping members and activists sufficiently involved and attached to the organization? In sum, how do some parties thereby remain stable and vibrant over time? These questions are interrelated and all relate to the fundamental theoretical problem of explaining survival and organizational 16

vibrancy amid change. As suggested above, these questions have not been adequately addressed. Responding to these important lacunae in our current understanding of parties and party systems, this dissertation analyzes a difference in outcomes that existing definitions of PSI fail to capture. My own causal analysis aims to shed new light on the actually divergent evolution of these instances of party survival. Based on this analysis, I advance a new explanation of types of party survival. The study thereby contributes to explaining both different varieties of party survival and their impact on parties’ value as channels of democratic representation.

Defining the Dependent Variable

To define the dependent variable I take advantage of current advances in concept formation and measurement in the social sciences. Conceptualization has gained increasing attention in political science. Since Sartori´s (1970) classic paper on concept misformation, there has been extensive new study of the theoretical and epistemological issues involved in concept formation as well as of specific tools for concept building (Goertz 2006).1

In this section, I provide a definition and operationalization of party survival. Based on Goertz (2006), I aim first to clearly define party survival. I define the secondary level by identifying the attributes that define the dependent variable.2 Turning to issues of operationalization, I then suggest ways in which each dimension of the secondary level may be measured. Having done so, I then make the relationship among dimensions of each level and across levels explicit. Why use the term survival and not stability? I use the term “survival” because it clearly highlights the daunting challenges to which parties—and democracies—are subject in 1

Munck (2009) on democracy is one recent example of thorough, practically-oriented conceptual work. The secondary level should present the “constitutive dimensions of the basic-level” (Goertz 2006 p. 6).

2

17

developing countries. It explicitly expresses the notion that political organization for pursuing public goods is often extremely difficult given actual structural conditions. Many daunting challenges can threaten the maintenance of the political regime and of political organizations as well. Survival also explicitly denotes a process while stability does not.

Party survival denotes remaining a political organization in a context of potentially unfavorable endogenous or exogenous challenges. However, this dissertation seeks to go beyond a simple account of literal survival alone. A party may be stable over time, but it may or may not remain a vibrant organization. For a party to remain vibrant over time, it is not enough for it just to survive in this narrow sense. Vibrancy is thus an attribute that qualifies how the process of survival occurs.

Survival, then, is comprised of two attributes: age and vibrancy. The age of a party clearly denotes survival over time while vibrancy refers to the vitality (dynamism) of the surviving organization.

Operationalizing the Concept of Party Survival

A first observable indicator of survival is the age of a given party. The more years that have elapsed since its foundation, the longer the party has survived. If a party has existed for many years, then it is fairly safe to assume—given the type of countries analyzed here— that it has successfully traversed numerous challenges. A second observable set of indicators of the quality of party survival, related to vibrancy, are: the extent to which militants and prospective leaders are involved in different types of partisan activities, the intensity and frequency of different types of meetings, and the territorial or functional

18

penetration of the party organization. This list is not exhaustive but is suggestive of some of the different ways party vibrancy can be detected.

Why is electoral success not considered in the definition? This dissertation is not an explanation of electoral success.3 Again, it is an analysis of types of party survival in consolidated and unequal democracies in Latin America. Moreover, democracy is not only about winners. Losers—parties of the opposition—are also relevant for democratic representation (Dahl 1971, Przeworski et al. 2000, among others). To repeatedly win elections, parties sometimes even resort to strategies conflicting with democratic representation, such as illegal governmental disbursement of private goods to buy voters’ loyalty. Why are parties’ particular forms of organization not considered relevant for determining their vibrancy? As will become increasingly apparent over the course of my discussion, vibrancy is an observed property of many different types of party organizations. Mass, social democratic, conservative, vertically organized, and other types of parties can all be vibrant. One therefore cannot determine whether a party is vibrant or not simply by looking at how it is organized.

Party survival can be either static or dynamic. Parties of the static type of survival are not vibrant. By contrast, a party surviving dynamically not only retains its label over time but also remains a vibrant organization. This definition captures two assumptions and two observable attributes. The first assumption is that survival is an ongoing process. This assumption captures the need for longitudinal analysis to capture parties’ evolution as agents of democratic representation.4 The second assumption is that 3

However, as presented below, some electoral relevance is taken into account for case selection. This requires being explicit about the scope of my analysis. Huntington´s definition of institutionalization was inspired by analysis of the process of modernization. This dissertation focuses on societies undergoing processes of development and modernization. Hence, the basic conflicts described by Huntington (and other scholars who have focused on modernization) are present. Thus, whether we are 4

19

survival can take different forms with different consequences. A party not only can cease to exist, but also can become ossified. We therefore need to know not only whether a party survives, but whether it does so statically or dynamically.

As Figure 1 illustrates, dynamic types of party survival are contained in the intersection of the two defining attributes of survival: age and vibrancy. Parties that show no Vibrancy but have Age, are surviving in the Static sense. I do not theorize what happens with parties that only show vibrancy but, for example, it could be a party mobilized by a charismatic leader. The light square delimits the universe of this study, which analyzes supposedly equally institutionalized parties. The empirical analysis in Chapters 4 to 7 unpacks and explains this universe.

Figure 1. Static and Dynamic Survival

Age

Static

Vibrancy

Dynamic

“Institutionalized” Area analyzing institutionalization or types of survival, parties´ roles in modernizing societies as potential agents of conflict reduction are regarded as vital. Determining whether this process of modernization ever actually ends and major disputes then disappear, or whether parties can ever be regarded as having finally and definitively achieved “institutionalized” status would be a matter for further study. 20

The Parties Analyzed

All of the parties analyzed as instances of survival in Chile, Costa Rica, and Uruguay meet the following conditions. First, the party must have participated with the same label in three general elections.5 This threshold is not trivial; it is the logical minimum number of electoral cycles for a political party to have experienced winning executive office, losing power, and winning again. For example: the party wins election 1, loses election 2, and again wins in election 3 (see Figure 2 below). Second, the party must have obtained national-level legislative representation in each of the three elections (see Figure 2).6 Strictly setting these criteria for case selection helps control for potential confounders. The only cases to which applying these criteria is potentially problematic are those in which parties win in their first election and keep winning thereafter.7 Those situations should be analyzed separately, but in principle they also must be considered.

In Latin America, presidential elections are normally held every four to five years. Depending on the particular electoral rules in force in its system, a Latin American party therefore needs to have survived at least 12-15 years to qualify. However, another condition has to be considered: if the founding leader still governs the party after this period, the analysis must take this fact into consideration, and conclusions regarding the case should be treated as extremely preliminary. This precaution is warranted because, in such situations, there is not yet evidence that the party is more than a personalistic vehicle, or that it can exist independently of the current leader. By themselves, the rules presented above will not filter out these cases.

5

Rose and Mackie (1988)and Sánchez (2009) use this criterion to define their cases. It is worth mentioning that the cases analyzed in this dissertation do not hold mid-term elections. In cases that do have mid-term elections, one should treat “national elections” those in which both the Executive and part of the legislature are elected. 7 The clearest example would be a political party born under an authoritarian regime that wins the first democratic election. 6

21

Figure 2. The Electoral Process of a Party Winning, Losing, and Regaining Executive Power

W e2

W e3

W

L

e1 L

L

References: e1,2,3: election 1,2,3 w: Win l: Lose

The Theory

To answer the research questions introduced above, I take a dynamic approach to institutional analysis. The purpose is to provide an analysis of institutions over time— particularly their gradual change or gradual decline.8 This dynamic perspective requires an approach to institutions that simultaneously understands them as sets of rules; recognizes their influence on actors’ incentives; and is attentive to history. Such an approach has to be sensitive to different historical settings accounting for institutional reproduction or decline. Several scholars have sought to pursue such combined emphasis on agency and historical processes (e.g. Greif and Laitin 2004, Mahoney and Thelen 9

2010). My work also bridges these perspectives, providing a causal account based on analysis of parties as organizations composed of ambitious politicians, born and embedded within specific institutional settings that undergo numerous historical

8

Dramatic changes are also a potential result. However, as discussed in Chapter 1, we already have welldeveloped analytic tools for understanding these instances of massive change. 9 Along similar lines, Hunter (2010) provides an interesting analysis of the evolution of the Partido dos Trabalhadores (Workers´ Party) that combines Rational Choice Institutionalism and Historical Institutionalism. 22

junctures. I develop this perspective in Chapter 2. Again, to be explicit about the scope of my argument, I focus on consolidated and developing democracies which have undergone a relatively smooth process of democratic consolidation.

Recalling the questions introduced above, what mechanisms have fostered (or hindered) party survival in Latin American consolidated democracies? How do parties remain stable and preserve their vigor over time? Finally, how do parties remain legitimate and relevant institutions of democratic representation?

Based on extensive fieldwork, I argue that it is possible to identify two types of party survival: static and dynamic. In dynamic types of survival, parties not only continue to exist as organizations but also retain their vibrancy as organizations. This dissertation thus examines how static and dynamic types can be explained.

I define causal conditions that explain self-reinforcing mechanisms of types of party survival over time.10 I claim that the type of party survival is conditioned by two sets of variables located at the organizational level: channels of elite promotion and ideas. They are also conditioned by two other variables: barriers to exiting the organization and the kind of political experience that the party organization has historically undergone. In different combinations, these conditions account for both static and dynamic forms of party survival.

The notable success of some parties in this regard, even in politically and economically developing countries in Latin America, shows that parties do have the potential to remain important channels of representation today. Their actual ability to do so, however, is variable, not a given, and processes should be considered in analyzing it. The success stories of Frente Amplio and UDI suggest a first, novel premise: no 10

As will become clear from the theoretical argument, those mechanisms might actually be selfundermining. 23

purpose, no dynamic party survival. If there is purpose, vibrant collective organization is more likely. I define purpose as the ability of the organization to coherently unite around a project, a set of ideas and intensively orient its action to it.

However, more than identifiable purposes alone are needed to ensure individuals’ continued participation in collective enterprises. As Aldrich (1995) has claimed, Ambition in a narrow sense has to be channeled for parties to remain meaningful for individuals and make them want to join the organization. If a politician can satisfy his ambition within the organization, then collective organization and individual ambition are in harmony.

Another condition for accounting for particular types of survival is actors’ perceptions of the costs of leaving the organization. Inspired by Hirschman (1970), this variable is here labeled as Exit. The more actors perceive high barriers to exiting from their party organizations, the more likely those parties are to persist. Finally, types of survival are conditioned by parties’ particular histories. I claim that parties with an identified purpose, that satisfy individuals´ ambitions, whose politicians perceive high costs of exit, and formed during or shortly after a traumatic experience, or that have suffered such an experience somewhere along their history, are more likely to survive dynamically over time.11

Together, these causal conditions determine how parties survive. Parties’ variable ability to foster collective prospective loyalty through a collective sense of purpose, and satisfying individual ambitions, in combination with politicians’ perceptions of the costs

11

As will be discussed in Chapter 2, the power of a traumatic experience as a powerful glue of collective organizations fades over time. Also, if that traumatic experience was suffered many years after the founding of the political party its imprint on the party´s identity is less powerful as a glue mechanism (see Chapters 4 to 7). 24

of exit, and parties’ historical experiences in relation to trauma, together explain their likelihood of static or dynamic survival.

As observed through the analysis of my cases, these four causal conditions (Purpose, Trauma, Exit, and Ambition) interact and together comprise the set of causal conditions collectively sufficient for explaining different types of party survival. These causal conditions explain why some parties have become vibrant and meaningful organizations and remain so even today whereas others have not. When these mechanisms are jointly present, they explain whether particular parties are able to survive dynamically, and thus remain vibrant.

Each type of party survival in turn has different consequences for the maintenance of individual parties’ and party systems’ value as vehicles of democratic representation. In the conclusions I also briefly assess their respective substantive political implications. Huntington defines institutionalization as “…the process by which organizations and procedures acquire value and stability” (1968, p. 12). It is often argued that parties are more likely to remain valuable as institutions of democratic representation when they, as agents, reduce transaction costs for their principals (in this case, voters or militants). To retain that value, they must also remain important agents of representation over time. Dynamic types of survival, those that show vibrancy, are thus more likely to retain their value as institutions of democratic representation and to be resilient in the face of exogenous shocks. Research Design Why study these countries as instances of party survival?

My cases are each individual party in these countries that comply with the rules defined above. Thus, political parties are the unit of analysis. The unit of observation is the politician from each of these political parties. 25

Studying instances of party survival is not an easy task. There are no clear spatial or temporal frontiers to demarcate the study (George and Bennett 2005, Gerring 2012). Studies of party or party-system collapse (Morgan 2011) have a clear dependent variable and straightforward questions: why has a party disappeared? Why have established party systems suddenly collapsed? My study takes on the apparently more mundane task of explaining regularity in established and comparatively healthy democracies. Establishing criteria for case selection is therefore both challenging and imperative.

This dissertation builds on a most-similar-systems design (Przeworski and Teune 1970), focusing on endogenous properties. Chile, Costa Rica, and Uruguay have been regarded as the most consolidated democracies in Latin America (Mainwaring and Scully 2010). Chile, Costa Rica, and Uruguay are systematically located at the upper end of various indices of democratic consolidation and governance. Whether it is in terms of development, transparency, or other such measures, the three are unanimously regarded as consolidated democracies (see, e.g., Mainwaring and Scully 2010). They are successful stories in terms of the very demanding challenges faced by developing societies. These three cases enable me to control for exogenous variables, thus providing an outstanding opportunity to focus analysis on the role of organizational and institutional variables in the evolution of their parties.

They are thus the cases where Mainwaring and Scully´s (1995) conclusions should hold best, and in which the party systems should have the most similar trajectories. Hence, these countries and their parties offer an excellent opportunity for conducting a combination of heuristic and crucial case studies(George and Bennett 2005 pp. 73-88).

As will be presented in Chapter 1, none of the available theoretical frameworks enables a clear understanding of the evolution of these countries and their parties. In theoretical terms, none of these countries have suffered from both of Morgan’s two criteria for collapse: death of at least one major party and party-system collapse. Chile, Costa Rica, 26

and Uruguay’s major parties are still competing in national elections, albeit with varying degrees of success and after having followed different trajectories. The PUSC in Costa Rica is the case that comes closest to meeting Morgan’s first criterion for collapse, but none of these party systems meets the other.

Following the stability literature, we cannot explain significant differences within these party systems or, more importantly, among parties within each of them. Parties in these three countries vary greatly in the levels of trust and legitimacy they enjoy. Parties in Costa Rica and Chile have become untrusted organizations; people there have become increasingly detached from parties, and party organizations have become largely empty vessels. However, even there, some individual parties have somehow remained vibrant organizations. In contrast, the major Uruguayan parties have all remained vibrant and significant organizations. The literature on party stability and PSI cannot explain the evolution of these cases and their present differences. Nor can the change literature. The conceptual problems in these existing literatures impede making precise descriptive and causal inferences about these cases and the variation they present. Conventional interpretations of adaptability as an instance of change across time are also inadequate for these purposes, if perhaps better suited for interpreting change in the wake of severe crises confronting individual parties.

Their parties and party systems have evolved in strikingly dissimilar ways. My project aims to explain the important differences observable among these cases in terms of their parties’ trajectories of evolution and resulting forms of survival. Although some implications are drawn at the system level, my unit of analysis is at the party level and the unit of observation at the politician-individual levels. Differences among Equals?

The three countries—Chile, Uruguay, and Costa Rica—have undergone significant transformations in the last decades. Chile and Uruguay have experienced a successful 27

transition from authoritarian rule to democracy. The years under authoritarian regimes left very deep imprints on politicians and forged a shared epic among those who fought for democracy or were persecuted. In the case of Costa Rica, the experience of transition is further in the past. For the last six decades, Costa Rica has been governed as a democracy.

The three countries have all lived through a harsh period of structural reforms, which in other countries led to significant clashes and destabilized democratic regimes. In the case of Costa Rica and Uruguay, those reforms have been implemented in a gradual fashion under democratic rule. In the case of Chile, the Pinochet dictatorship imposed a new economic model that was on solid ground by the time democracy was restored in 1990. Finally, Costa Rica and Uruguay suffered a harsh economic crisis under democracy while Chile has never undergone such a severe crisis since its transition to democracy.

These three main differences between the three cases will be taken into consideration when substantiating my theoretical argument in the empirical chapters. However, since all my cases—i.e., parties— have been subject to the same processes in each of these three countries, they do not affect my causal claims, which pertain to the party level in each country. Methods and Data

Building on classic accounts on the comparative method and recent advances in multimethods research, my case-oriented research strategy enables me to provide an argument sensitive to complex and conjunctural causation (Ragin 1987, 2000, 2008) and to deepen the existing literature on causal mechanisms (Goertz and Mahoney 2012). In this sense, my theory—advanced in Chapter 2—is sensitive to a complex relationship among conditions (Goertz 2006, Goertz and Mahoney 2012).

28

More specifically, from an in-depth study of my cases, I introduce a theory based on Hirschman’s (1970) trilogy of “Exit, voice, and loyalty” to explain how different combinations of my causal conditions account for two different types of party survival. In-depth analysis is needed to uncover and explicate the causal relationships stated in my theory. It derives logically from the aforementioned dynamic approach to institutional analysis. Though my focus will be primarily on the short-term, historical evaluation of the evolution of each party considered in my analysis is also fundamental to assessing the causal argument here introduced. The tools of comparative historical analysis will therefore be used in the course of the empirical chapters (Pierson 2004).

Given that my theoretical argument assumes complex and conjunctural causation, in this project I make use of QCA (Ragin 1987, Rihoux and Ragin 2009) for systematizing the information obtained throughout my fieldwork, as well as that derived from secondary sources. The aim of using this method is to summarize the abundant evidence and present it parsimoniously and systematically. This analytic tool is also well adapted to situations marked by conjunctural causation, a key component of my theoretical argument. The goal of the QCA is to report the interviewees’ perceptions on the four causal conditions defined in Chapter 2.

To gather the data necessary for analyzing each case, I conducted structured interviews with actors from each of the main political parties in Chile, Costa Rica, and Uruguay. I constructed a theoretical sample covering relevant cleavages inside each party: factions, generational cleavages, and ideological perspectives. With all these in mind, I conducted forty-four interviews in Uruguay, fifty-three in Chile, and fifty-one in Costa Rica (see Appendix 1 for the complete list of interviewees). This material is the primary source of data for my analysis. I also collected data on each political party and took advantage of studies conducted in each country by different scholars. Finally, I drew on secondary sources for reconstructing each party’s history—which I also discussed extensively with each interviewee. 29

Why Another Study of Party Systems?

A study that makes a contribution in both conceptual and theoretical terms promises to be significant for the study of parties and party systems in Latin America. The theoretical relevance of this study lies in the need to advance in developing a dynamic perspective on institutions with a balanced emphasis on agency and historical processes. As stated above, one of the main critiques of institutionalism is its inability to explain how institutions evolve (Thelen 2003, Mahoney and Thelen 2010). The study of types of party survival in developing democracies provides a great opportunity to go beyond a methodological discussion and actually apply a dynamic perspective to a specific research problem that integrates historical and rational choice institutionalist foundations.

The study also responds to the urgent need to bring theoretical order to the widely studied subject of party and party system stability and change. There are three main problems in the literature on PSI and party adaptability. First, it has been marked by poor conceptual definitions and operationalizations of PSI and adaptability, the two conventional interpretations of survival. Second, no explicit and systematic effort has been devoted to clarifying the relationship between attributes. And finally, this literature has often unwittingly conflated causes and effects. Chapter 1 will be focused on these issues, reviewing in depth existing arguments regarding parties’ survival.

Through conceptual analysis permitting improvement of descriptive inference and application of this kind of dynamic approach to institutional analysis, this study ultimately aims to improve our ability to explain the paths that different parties follow. In this sense, we can better specify the relationship between party stability or survival and the quality of democracy in a specific country.

30

Outline

Following upon this introduction, the dissertation is divided in eight chapters. The first chapter is dedicated to critically reviewing existing theories on parties and party systems. In that chapter, I mainly focus on existing arguments about party survival in developing democracies. Chapter 2 introduces my dynamic approach to institutional analysis and my theoretical argument. It also presents the four mechanisms identified as jointly sufficient for explaining dynamic party survival. Chapter 3 explains the basis for the QCA analysis. It shows how each condition was computed. It thus links the conceptualization and the empirical assessment of my theoretical argument. The next chapters proceed with the QCA analysis. Chapters 4, 5, and 6 are dedicated to analyzing parties on a country-by-country basis. For each country, I first present a general picture using secondary sources and interviews conducted there. I also analyze the configuration observed within each party, building upon the evidence collected in my fieldwork. Also, using secondary sources, I briefly sketch the historical evolution of the causal conditions and how they interacted with significant events to bring about the present configurations.12 Chapter 7 relates the four conditions, the observed configurations, and their relationship with the dependent variable (type of party survival). Finally, Chapter 8 presents the conclusions of this study.

12

The narrative of the historical evolution of my conditions in interaction with other factors is, at best, a tentative, hypothetical historical analysis. In this sense, the possible imbalance observed in the length of the analysis of the evolution of each condition for each party reflects differences in the density of their histories as well as the depth and number of available studies. 31

Chapter 1. Existing Arguments Introduction

This chapter discusses existing arguments on party and party-system survival, analyzing current explanations as alternative approaches to my own argument, which I fully introduce in Chapter 2.13 I mainly consider those approaches that survey developing countries. There are three different but closely related lines of research in the literature, all of which share a common concern: the relationship between party-system survival and parties’ ability to retain their value as channels of democratic representation.

A first line of work has focused on explaining party and party systems stability. Within this line of work, the party system institutionalization literature has provided a vast array of evidence. A second group of authors has looked at how the ability of specific parties to adapt to changing circumstances conditions their ability to survive. Though related to the stability literature, these works have emphasized exogenous shocks as triggers of party and party system change. These studies have thus shed light on the ability of specific parties within a party system to adapt and change in response to environmental changes. Hence, their focus has been on parties’ survival across critical junctures. Finally, a more recent line of research has explained party systems collapse, which is the negative pole of this study. This last line of research has sought to explain why some party systems have collapsed in a short period of time. 1.1. The Stability Literature: Revisiting the Concept of PSI and PI

Huntington argued that: “Institutionalization is the process by which organizations and procedures acquire value and stability. The level of institutionalization of any political system can be defined by the adaptability, complexity, autonomy, and coherence of its 13

I will consider more specific and empirical alternative hypotheses in Chapter 4. 32

organizations and procedures” (1968, p. 12). Through this definition the author made clear what institutionalization is and what defines the level of institutionalization existing in a specific political system. Huntington then applied this definition to party systems. In defining one of the four criteria for political institutionalization the author says: “The more adaptable an organization or procedure is, the more highly institutionalized it is; the less adaptable and more rigid it is, the lower its level of institutionalization. Adaptability is an acquired organizational characteristic.” (p.13). The notion of adaptability brings dynamism to the analysis of institutionalization. For decades, though, scholars have then associated institutionalization with rigidity. As will become clear in the following pages, subsequent analyses based on the concept of institutionalization have not paid enough attention to the ability of parties to adapt to changing circumstances. Huntington’s book deals with the disruptions caused by the process of modernization and how they can be reduced. The author claims that the process of modernization breeds conflict and increases political participation: “Political stability…depends upon the ratio of institutionalization to participation. As political participation increases, the complexity, autonomy, adaptability, and coherence of society’s political institutions must also increase if political stability is to be maintained” (p. 79). Thus, disruptions that ensue from increased participation are mitigated by solid and stable institutions: higher levels of political participation, an intrinsic characteristic of modernization, must be accompanied by strong, complex, autonomous institutions (p. 85). However, in recent studies little has been said about the traits that enable parties to remain solid and stable institutions over time.

Mainwaring and Scully directed a collaborative project analyzing party systems in Latin America. Their goal was to analyze PSI, and they essentially constructed their 33

introductory framework around Huntington’s assumption regarding the role of institutionalization. Their introductory chapter defines party system institutionalization in terms of four attributes.

For a democratic party system to be institutionalized, four conditions must obtain. First and most important is stability in the rules and the nature of inter-party competition. Patterns of party competition must manifest some regularity, which is not to suggest that these pattern become “frozen”…Second, the major parties must have somehow stable roots in society; otherwise they do not structure political preferences over time and there is limited regularity in how people vote…Third, in an institutionalized party system, the major political actors accord legitimacy to the electoral process and to parties…Finally, in an institutionalized party system, party organizations matter (1995 pp. 4-5).

Although Mainwaring and Scully (1995) aim to define PSI, two of the four defining characteristics refer to parties as units while the other two refer to the party system as a whole. Stockton (2001, p. 96) rightly notes that this flaw is repeated in conventional studies using PSI. Also, Mainwaring and Scully’s (1995) definition seems to treat the four dimensions as independent and additively related.14 Working from the same basic definition, Mainwaring (1999) slightly refines the relationship between the different dimensions defining PSI. He says: “These four dimensions of institutionalization need not go together, but they almost always do” (p. 27). Hence, we can infer from Mainwaring (1999) that each dimension tends to interact with the others.

The original definition became the standard approach to PSI analysis in subsequent studies (see, e.g., Coppedge 1998b, Bielasiak 2002, Sánchez 2009, Kuenzi and Lambright 2001, Lindberg 2007). These different studies do not diverge in their understandings of the substantive meaning of PSI. Differences among them reside in 14

This understanding is also reflected in the Index of Party Institutionalization (IPI) subsequently constructed by Jones (2005). The IPI is the sum of the four original dimensions (divided over four). 34

their operationalization of PSI and use of different indicators for measuring it (see Table 1.1). Thus, there have been a considerable number of studies applying the concept of PSI as Mainwaring and Scully (1995) defined it.

There are, however, underlying problems of conceptualization in subsequent analyses of PSI. First, half of the operationalization of the first dimension of PSI, i.e. “stability of rules”, was largely lost.15 For measuring this dimension, scholars have typically just used electoral volatility, which captures the stability of voting behavior rather than that of rules. Moreover, this dimension has been widely treated as a sufficient attribute of PSI.16 This shows the privileged emphasis frequently placed on stability and the corresponding disregard for the role adaptability. Second, in Mainwaring and Scully’s (1995) original version, the idea of “roots in society” implies that parties have stable connections with society (p. 9). In PSI studies, parties’ social rootedness has been operationalized in ways that are methodologically or conceptually problematic. For example, it has been operationalized as the congruence between legislative and executive vote, an operationalization that merely captures the relationship between the party elites and does not have a clear relationship with the dimension “roots in society”.

The third dimension of Mainwaring and Scully´s (1995) definition of PSI—legitimacy— has since been changed. In addressing this dimension, subsequent research has included the legitimacy of parties in society. There has thus been a shift of focus from whether elites concede legitimacy to parties and elections (as the original definition states) to whether citizens do. This shift has contributed to the pervasive lack of consistent conceptualization in the recent PSI literature. Moreover, parties’ legitimacy in society

15

Bielasiak (2002) is the exception in this regard. See Luna (forthcoming) for an analysis of the problems with using this indicator.

16

35

should be conceived as an effect of institutionalized parties. Current PSI definitions thus conflate attributes and effects.

36

Table 1.1: Institutionalization of Party Systems: Summary of Dimensions and Indicators Dimensions

Stability of rules of the game and competition

Roots in society

Legitimacy

Party Organizations

Indicators -Pedersen Index of Volatility of votes and/or seats (Mainwaring and Scully 1995, Schedler 1995, Coppedge 1998a, Mainwaring 1999, Kuenzi and Lambright 2001, Stockton 2001, Bielasiak 2002, Pedahzur and Brichta 2002, Jones 2005, Lindberg 2007, Sánchez 2009). -Effective Number of Parties (Bielasiak 2002) -Number of parties in the legislative and number and % of new parties in the legislative -Number and % of parties voted out -% of seats held by the largest party -% of seats held by the runner-up party (Lindberg 2007) -Continuous participation in national elections (Pedahzur and Brichta 2002) -Stability of rules (Bielasiak 2002) -Age of parties (Mainwaring and Scully 1995, Mainwaring 1999, Roberts and Wibbels 1999, Kuenzi and Lambright 2001, Stockton 2001) -Congruence in executive and legislative elections (Mainwaring and Scully 1995, Mainwaring, 1999, Stockton 2001) -Identification with parties (Jones 2005, Mainwaring 1999, and Mainwaring and Torcal 2006)). Mainwaring and Scully (1995) claim that having this data would be helpful. - Nonpartisan or antiparty candidates chances of electoral success (Mainwaring 1999) -% of legislators that believe that parties are not distant from society (Jones 2005)

-Conceded by elites (Mainwaring and Scully 1995, Bielasiak 2002). They also mention the possibility of seeing how the population perceives electoral processes and parties. -Measured through public opinion: legitimacy, trust, “parties are indispensable” (e.g., Jones 2005, Mainwaring 1999). Also, from public opinion, “elections are fair”, “elections offer real alternatives” (Jones 2005) -If the opposition boycotted the election, if losers accept results, and if international observers considered the electoral process free and fair (Kuenzi and Lambright 2001) -Party Resources, presence in districts, internal rules, legislative discipline, control of position distribution. Zucco (2010) summarizes the indicators used in other studies. -Age of parties. Legislators’ view of whether party is a permanent organization (Jones 2005)

37

Some authors have made significant contributions in analyzing the institutionalization of individual parties (Panebianco 1988, Dix 1992, Randall and Svåsand 2002, Pedahzur and

Brichta

2002).

The

main

problems

with

party-level

explanations

of

institutionalization are that they reproduce the emphasis on stability and mainly provide descriptive accounts. They also present problems of conceptualization.

In conclusion, the literature on PSI presents systematic problems in terms of conceptualization. There is no clarity regarding the relationship between attributes and how to aggregate measures at different conceptual levels (i.e., attributes and indicators). Some scholars have also called attention to the conflation of attributes, causes, and consequences of institutionalization (Zucco 2010, Altman and Luna 2011 Luna forthcoming). For example, the effectiveness and legitimacy of parties in representing interests—resulting in a better quality of democracy—is part of what has to be explained and should thus be seen as an effect of PSI. Nonetheless, it has instead been included as an indicator of PSI.

Such a poorly defined concept is inadequate for enabling us to understand its impacts. Some scholars have highlighted that the concept is not useful for explaining outcomes in terms of democratic representation (Altman and Luna 2011, Torcal 2011, Zucco 2010). The concept of PSI was originally introduced as an independent variable for explaining democratic consolidation. However, different studies have found diverging results in terms of the effects of PSI on democratic consolidation (Kuenzi and Lambright 2001, Stockton 2001, Thames and Robbins 2007, Luna forthcoming).

Although most analyses have focused on the relationship between party-system institutionalization and democratic consolidation, some have treated PSI as a dependent variable (see Remmer 1991, Roberts and Wibbels 1999, Tavits 2005, Malbrough 2008,

38

Kitschelt et al. 1999)17. While they provide interesting findings, the main problem with these studies is that they continue to link PSI to stability and mainly operationalize it as low volatility. They thus reproduce problems already noted with the operationalization of the concept.

Finally, both the original and subsequent definitions have relied on a normative preference for stability. The emphasis on stability has obscured the role of adaptability as a concept inherently tied to institutionalization. It has also diverted attention from seeking to understand the mechanisms explaining how institutionalization is actively perpetuated. In sum, there are no studies that explain how parties retain their value over time. Pictures without motion?

The picture presented above suggests both theoretical and empirical problems. The concept of PSI as used by the comparative politics literature has little analytical power. Discussing how the concept was built and assessing its explanatory power, Chasquetti (2008b) mentions a common problem in different definitions of PSI: current definitions allow one to take good “pictures”, but they do not provide a way to account for processes of transformation. The literature’s emphasis on stability as measured through snapshots has lumped together very dissimilar cases, and it cannot explain the evolution of a party or a party system.

The concept may be descriptive but it certainly cannot explain specific outcomes. Not only are some previously “institutionalized” party systems (IPS) no longer stable, but other IPSs have lost their ability to channel representation. Existing definitions and operationalizations are unable to explain: a) how likely a given party or party system is to remain stable; or b) whether a given party or party system is to remain a viable and 17

Their analysis does not use the concept of PSI, but their goal is to explain post-communist party systems’ stability. 39

legitimate channel of political representation and thus contribute positively to democratic consolidation over time. Although the concept of PSI was originally conceived as an independent variable for explaining democratic consolidation, we still do not know how party system institutionalization influences levels of democratic consolidation. Moreover, less has been said about the causal conditions accounting for parties’ ability to remain institutionalized. The PSI literature has not addressed the dynamics feeding institutionalization of parties and party systems. In particular, they have not addressed the ability of parties, qua organizations, to persist and retain legitimacy and support among citizens over time.

Three key questions thus remain unanswered or inadequately addressed: How does a party regarded as “institutionalized” avoid becoming ossified, despised, or socially detached? How does it retain the flexibility needed for adapting to new challenges, while keeping members and activists involved? Hence, how do parties and party systems remain stable and retain their value as democratic channels of representation over time? Recalling what I discussed above, how can we introduce a “dynamic” dimension to the analysis of party survival, thus drawing attention more explicitly to how and under what conditions parties can hope not only to survive but simultaneously remain over time as valued channels of democratic representation? To answer these questions, we first need better to understand the factors shaping party systems’ evolution and ongoing importance as channels of representation.

Hagopian (2005) gives us a hint in this regard. Concluding an edited volume about “advances and setbacks” in the third wave of democratization in Latin America, she says that “Where institutions of political representation are vibrant, and where parties are responsive and accountable to the preferences of citizens that can associate in the terrain of civil and political society, democracy is somewhat inoculated from setbacks, even when banking systems collapse and corruption scandals are brought to light.” (321). 40

Hence, where parties persist as dynamic institutions, they are more likely to remain effective as institutions of representation. Following Hagopian, one could hypothesize that what I call “dynamic survival” makes parties more resilient. Thus, the goal of this study is first and foremost to provide an explanation for types of party survival. 1.2. The Change Literature

The second major line of research dealing with parties and party systems survival is what I call the “change” literature. This literature explains organizational survival amid change. In this sense, it brings more dynamism to the study of parties and party systems.18 However, it does so by divorcing survival from the ability of parties to retain their vibrancy qua institutions of democratic representation. In this line of research it is possible to identify two related traditions.

The first tradition goes back to Duverger´s classic book (1954). It essentially provides a typological analysis of the evolution of party organizations over time in light of environmental changes (Collier, Laporte, and Seawright 2008). Katz and Mair’s different contributions on party-system change in developed societies have been significant (Katz (Katz and Mair 1994, Katz and Mair 1995)19. Despite their great contribution to descriptive inference, this work has not explained variation across time or within cases. This work has also been marked by an implicit nostalgia for the bygone mass-party era. In analyzing and evaluating the evolution of party systems, this line of study has therefore tended to equate

change from the mass-party model with decline

(Janda and Colman 1998, pp. 611-613).

18

In fact, Levitsky (2003) claims that “less” institutionalization might actually explain the greater capacity to adapt and survive 19 Kirchheimer (1966)’s introduction of the “catch-all” party concept was of course also a milestone in the tradition of mapping organizational changes. 41

Studies along these lines have also often emphasized the role of specific types of party organizations’ ability to reproduce across time regardless of major or minor environmental changes (Crotty 1971, Dix 1992, Mair 1997, Appleton and Ward 1997, Janda and Colman 1998, Kato 1998, Mustillo 2007). While specific forms of partisan organization (like the “organised disorganisation” of PJ in Argentina20) may play a role, my cases show that specific forms of party organization are neither necessary nor sufficient conditions for parties’ dynamic survival. A second major strand in the “change” literature is comprised of works explaining parties’ adaptability. The main difference with the first tradition is that, rather than providing a descriptive picture of the evolution of parties’ organizational traits, this line of research focuses on causal mechanisms accounting for the attributes of parties that foster or inhibit their adaptation to secular or dramatic environmental changes, i.e., to socioeconomic, structural challenges that force adaptation. These studies analyze why some parties are able to resist and adapt while others are not.

Studies of adaptation to secular change have predominantly analyzed consolidated democracies (e.g., see Rose and Mackie 1988, Kitschelt 1994). However, for this dissertation those analyzing developing and third wave democracies are more directly relevant. These newly democratized polities have suffered numerous challenges risking both democratic consolidation and parties’ very survival. In Latin America, severe economic crisis presented the main challenge for parties. (Coppedge 2001) coined the term “political Darwinism” to describe Latin American parties’ capacities to adapt and survive such challenges. Levitsky (2003) explained how less formal institutionalization actually helped Peronism to adapt to these dire circumstances by giving Carlos Menem, Argentina’s president from 1989 to 1999 necessary room for maneuver.21 Similarly, Burgess and Levitsky (2003) argue that: 20

See Levitsky (2001) Hu and Heller (2010) analyze policy change with a formal model and then analyze the SKDL party in

21

42

…deep economic crises and absence of a left-wing electoral challenge create stronger incentives to adapt in a market oriented direction than lower levels of crisis and strong competition from the Left. …internally fluid party structures, subject to executive influence, possess greater adaptive capacity than highly bureaucratic ones that can block or override executive initiatives” (pp. 882-883).

Building on this literature, (Hunter 2010) combines historical and rational choice institutionalism to analyze how strategic programmatic moderation enabled Brazil’s Workers´ Party (Partido dos Trabalhadores, PT) to gain access to governmental power. Although the PT did not suffer from severe economic challenges, to triumph electorally the PT faced the challenge of adapting its programmatic appeals.

This literature has made a great contribution to explaining party change, a key factor for organizational survival and electoral success. However, this line of work—especially those studies focusing on Latin American cases—also suffers from two main problems. On the one hand, adaptability has been mainly equated with parties’ ability to converge to the neoliberal turn, not understood in terms of the party accompanying changes in voters’ preferences. In this sense, adaptability refers to one particular direction of change that every party must correctly adopt in order to survive and retain its relevance.22 This assumption was not problematic when Latin American parties confronted severe challenges to the ISI model in the 1990s. However, excessive focus on that historical juncture is not helpful for providing more generalizable theoretical claims.

On the other hand, the concept of adaptability lacks conceptual clarity because it has been associated with rapid programmatic change. Adaptability is a property of parties helping them to survive amid change (Huntington 1968). Thus, it can have two

Finland. Their conclusions are the opposite of what Levitsky asserts. 22 Hunter´s thorough analysis of the PT claims that the party needed to moderate its platform in order to attract more centrist voters, but we do not know why the PT also maintained some of its salient, much less centrist characteristics. 43

meanings. First, it can refer to key strategic choices leaders make to keep the party attractive to voters at elections—regardless of the party's previous programmatic orientations. Second, it can refer to politicians’ more general ability to sustain parties as agents of democratic representation (Schattschneider 1942, Adams 2001). For parties to maintain their

value

in

this

regard,

they

need

to effectively

reduce

transaction

costs, particularly when it comes to the provision of public goods. Parties (agents) are effective in reducing these transaction costs over time only when they succeed in keeping their programmatic appeals aligned with voters’ and militants’ (principals’) preferences.23 From historical institutionalist analyses explaining why positive feedback processes are intense in politics24 and recent advances in bounded rationality (see Jones 1999, Weyland 2007), we know that politics is not an arena of rapid rational learning and adaptation. Thus, it should be accepted that party leaders normally cannot very rapidly change their programmatic perspectives if they are to remain attuned with members’ and militants’ preferences. Let us think of an analogy. How do businesses survive over time? One of firms’ main tasks is to be attuned to environmental changes and adapt to them. Their main tool for doing so is innovation. Innovation is achieved through processes such as improving products to reflect technological transformations or modernizing their organizational structures. More importantly, however, firms have to react quickly. This is the main 23

Thus stated, this framing of the causal claim also includes the “law of curvilinear disparity” and its variations (May 1973, Kitschelt 1989a). 24 See Pierson (2004). His claims concerning the prevalence and intensiveness of positive feedback processes in politics are well-suited to supporting my claim. In assessing the central role of collective action, density of institutions, possibilities for using political authority to enhance asymmetries of power, and political power’s intrinsic complexity and opacity, Pierson argues, “Each of these features make positive feedback processes prevalent in politics” (30). He adds: “…the absence or weakness of efficiency-enhancing mechanisms of competition and learning, the shorter time horizons of political actors, and the strong status quo bias generally built into political institutions...makes positive feedback processes in politics particularly intense” (pp. 30-31). 44

difference with politics.25 In politics, adaptation cannot occur too fast. If it does, it will then constitute a “policy switch” (Stokes 1999), potentially jeopardizing the aforementioned relationship between parties and their militants or adherents.

From this perspective, adaptability as conceptualized by Levitsky (2003) represents a case of “policy switch” (Stokes 1999). Because programmatic preferences do not change rapidly (Carmines and Stimson 1989), such rapid change does not preserve the connection of preferences between recent voters and militants, on the one hand, and leaders, on the other.

Such dramatic policy changes are therefore at odds with

principals’ embrace of parties as solutions for collective action problems. Such “policy switches” thus conflict with the second, more substantive meaning of adaptability.

Levitsky (2003) frames his case study in line with this more substantive understanding of adaptability. He regards change as needed to maintain the value of a party as an institution of democratic representation. However, throughout his analysis, adaptability simply becomes the party leader’s strategic ability to react to dire economic circumstances by pursuing neoliberalism, thus pulling off a neoliberal “policy switch”. It is not the direction of policy change described by Levitsky that makes it an instance of policy switch, but rather its speed. Such rapid policy change hinders leaders’ ability to persuade militants of the need to redirect policy in favor of market reforms. This understanding of adaptability as a strategic choice to survive contradicts the notion of adaptability as a property related to democratic representation. In this case, adaptability inheres in leaders’ choices. Such choices theoretically risk rapidly increasing the distance between those leaders’ programmatic orientations and those of recent voters and the majority of militants and middle leaders.

25

Although Schlesinger (1994) makes the analogy between political and economic markets, he does mention that “Elections create an imbalanced market…while parties receive private benefits, they offer only collective benefits in exchange” (p. 13). 45

Ideological shifts signal different types of “adaptation” depending on the speed with which they occur. Adaptable parties update their ideological or programmatic positions only gradually because that is the only way for their leaders, militants, and voters to maintain a consistent relationship in their respective preferences and their inclinations towards different policies, values, etc. Menem campaigned in the midst of a dire economic situation. However, as Levitsky carefully reviews, he campaigned with the traditional PJ appeals. When he took power, he changed his policy. This cannot be labeled as adaptability. Instead, it is a perfect case of policy switch. As a corollary, depending on the pace of the change, electoral and strategic adaptation might conflict with democratic representation.

The PJ case analyzed by Levitsky is only in line with the notion of adaptation as a strategic choice that proves electorally successful, rather than a property of a political organization enabling it to retain its value as a channel of democratic representation. It could be argued that a party’s electoral success suffices to show that its transformation was accepted by party adherents and is thus not in conflict with democratic representation.26 However, there is no evidence to support that claim. Electoral success is not necessarily directly connected with programmatic representation; voters may support a candidate for a wide variety of reasons, such as clientelism.27

Figure 1 presents a hypothetical example with three political parties. Party A is located on the center-right at T0. However, it moves gradually towards the center, arriving closer to the center at T2. Party B follows a different trajectory. This party departs from a center-right position at T0 but at T2 is located at a left-of-center position. This is what Downs (1957) termed leapfrogging. Finally, Party C seems to undergo a process similar 26

Lupu (2011) shows that parties that pursue such transformations, which he labels “brand dilution”, lose adherents and become more sensitive to governmental performance. 27 From Weyland´s (2002) research, we do know that citizens valued the ability of the party to circumvent the harsh economic crisis of the late 1980s. Nevertheless, this is not directly related to programmatic appeals and preferences. 46

to that of Party A. However, Party C makes a drastic change in its ideological stance all in one shot. Even though it does not engage in leapfrogging, it moves significantly along the ideological continuum in one single jump. The lengthier the ideological distance the party has moved in a short period (for example, one electoral cycle), the less the probability that it will maintain a coherent relationship among militants’, voters’, and leaders’ preferences. Thus, party A is a potential case of adaptation understood as a matter of maintaining a party’s value in terms of democratic representation—regardless of whether or not strategic choices are involved—because its programmatic orientation changed at a slower rate. The pace of its transformation is thus crucial here. As Hunter (2010) shows for the PT and Garcé and Yaffé (2005) for the case of the FA, their transformations did not occur overnight. One would then need to determine whether those changes resulted from dialogue among different party members and factions.

In conclusion, the adaptability literature, has suffered from two main problems. First, those who have used the concept of adaptability have focused on critical junctures at which dramatic changes occur. Second, they have not established a clear relationship between adaptability of the kind they have identified and democratic representation. This literature is thus not well-tuned to account for parties’ variable relevance as agents of democratic representation amid change.

47

Figure 1. Hypothetical Example of Programmatic Transformation Across Time Time

T2

A T1

T1 T0 B

T0 T1

C

T0 L

C

R

1.3. Explanations from the “negative pole”

Scholars working with PSI as an independent or dependent variable are not the only ones who have delved into the theoretical problem of party or party-system survival within developing democracies. As previously reviewed, some have instead focused on adaptability to explain why some parties are able to withstand changing and even extreme circumstances. Meanwhile, many scholars have chosen to explain instances of decline, collapse, electoral dealignment, or the pervasiveness of inchoate parties in some party systems. I will focus on those arguments that explain the negative pole of my dependent variable: party system collapse, especially in Latin America.

48

Scholars working along these lines have focused on one of the most striking instances of collapse: Venezuela, for which they have offered a variety of explanations. Dietz and Myers (2007) claim that deteriorating economic conditions, increasing levels of corruption and insecurity combined with centrally and hierarchically organized parties, and the emergence of personalistic leadership explain the unexpected collapse of Venezuelan party system. Morgan (2009) compares instances of collapse and survival of systems that could have collapsed. She claims that collapse occurs when parties fail to link society with the state (p. 2). This deterioration emerges when there are “core threats and constraints on suitable adaptations” (p. 2). More specifically, they occur: “…when structural changes threaten the logic of a party system’s core linkage strategies and contextual constraints limit the major parties’ ability to adapt appropriately and respond to these foundational challenges” (p. 6). Morgan (2011) provides a more detailed account of the Venezuelan party system’s collapse.

While Morgan and Dietz and Myers focus on the supply side, Seawright (2012) provides a compelling argument concentrating on the demand side by investigating why voters defected from traditional parties and voted for outsiders.28 He concludes that, although economic performance played a role in the Venezuelan collapse, the best predictor for the collapse was increasing ideological divergence between voters’ preferences and traditional parties’ appeals. A line of research related to the literature of collapse has studied the dynamics of “party non-systems” (Sánchez 2009). Sánchez introduces this concept to capture polities in which previous party systems have collapsed or in which parties have failed to survive major “extra systemic” volatility, as evidenced by “…fundamental lack of inter-temporal continuity in the identity of the main parties of the polity” (Sánchez 2009, p. 514). After 28

Lupu (2012) seeks to explain the effects of party change on party identification among citizens. He claims that those changes “…diluted the brands of their parties, eroding voters’ attachments to these parties. Without the assured support of partisans, parties become more susceptible to negative retrospective evaluations and, when they performed poorly in office, broke down” (Lupu 2012, p. 3). 49

developing the concept of “party non-systems”, Sánchez briefly describes the party systems in Peru and Guatemala, comparing them with the “inchoate” party systems of Ecuador and Bolivia. He concludes that party non-systems have deleterious consequences in terms of democratic representation and governability.

None of my cases exhibit the attributes and dynamics that define a party non-system. All three have low extra systemic volatility. Regardless of differences in terms of “internal” volatility, moreover, their main parties show important degrees of inter-temporal continuity. As argued before, my cases are also lack the causal conditions identified as accounting for collapse. One could argue that the absence of these conditions is sufficient for explaining their survival. However, when one unpacks types of survival, a causal asymmetry (Goertz and Mahoney 2012) between collapse and survival actually becomes apparent. Although Chile, Costa Rica, and Uruguay’s party systems have all survived, their divergent trajectories raise the question of how to explain differences between types of survival. Moreover, differences among parties within each of those systems also call for explanation centered first at the party level. Providing such explanation is the major goal of this study. Conclusion

There are two major strands of arguments regarding party survival—stability and change literatures—and one recent line of research—the study of party collapse. The explanation of types of party survival developed in this study might not be sufficient to explain collapse because there is a causal asymmetry: what explains parties’ survival does not necessarily explain the collapse of parties.29

The

stability

literature

has

significantly

advanced

our

understanding

of

institutionalization. One of the main problems with this approach is conflation of 29

However, as discussed in Chapter 8, static survival may tend to precede actual collapse. 50

“institutionalization” (a process) and “institutionalized” (a status). Because it has been insufficiently attentive to the institutionalization process, this literature is unable to explain major differences between supposedly similarly institutionalized parties and party systems. Moreover, even assuming that a party can ever become fully “institutionalized”, the literature has not explained how a party then retains this status over time.

The change literature has given some dynamism to the picture presented above. Those analyzing parties’ adaptability have drawn attention to a crucial aspect of party politics: the environmental challenges parties face that threaten their continued relevance. However, studies on adaptability suffer from conceptual problems, and those focusing mainly on evolution of certain types of party and party systems have provided mainly descriptive accounts.

Given the dramatic junctures parties in developing societies face, how can and do some parties remain vibrant, legitimate and relevant institutions of interest aggregation and valuable channels of democratic representation? The next chapter proposes causal conditions to answer this question, which I will demonstrate using a dynamic approach to institutional analysis. My dynamic approach to institutional analysis responds to the critique of treating stability synonymous with parties’ retention of value across time.

I define necessary and sufficient attributes for explaining self-reinforcing or selfundermining mechanisms accounting for types of party survival across time and specific junctures. Those conditions are: ambition, purpose, trauma, and exit. Different combinations of these conditions explain static and dynamic party and party-system survival. In this sense, they explain parties’ and party systems’ differential resilience over time and across historical junctures. In turn, each type of survival has different consequences for the maintenance individual parties’ and the party system’s vigor and, in turn, as vehicles of democratic representation. 51

Chapter 2. The Theory: Exit, Trauma, Purpose, and Ambition Introduction

In democracies, parties’ ability to remain legitimate, valued, and active organizations is a never-ending process. This is so because these societies undergo numerous challenges—redistribution being the most problematic and salient. 30 The purpose of this dissertation is therefore to explain different types of party survival in developing but consolidated democracies in Latin America.

From its inception, the notion of institutionalization was conceived as a factor moderating the inherently conflictual process of modernization. More specifically, it was assumed that those societies with party systems showing significant degrees of institutionalization—defined as synonymous with stability—enjoyed greater chances of remaining consolidated democracies and remaining on a smooth path to development. However, the literature put excessive focus on stability, obscuring significant differences among stable parties and party systems. I refer to these differences among apparently similar stable parties and party systems as different types of survival across time.

There is no magic threshold beyond which parties are ensured eternal relevance. However stable a party is, its survival as a significant institution is a variable to be examined. Parties that survive may nonetheless become increasingly irrelevant or delegitimized.31 Most conceptualizations and empirical analyses have disregarded this possible outcome. The analysis of the evolution of parties in developing, unequal, but consolidated democracies must take this possibility into consideration.

30

Its salience is not solely determined by the intensity of its politicization but also by what happens when it is politicized. 31 As reviewed in Chapter 1, parties can also cease to exist. 52

This chapter will provide a theoretical framework for answering the following two questions. First, what mechanisms explain specific types of party survival? And second, what explains parties´ abilities to endure and reproduce themselves as valued institutions over time and across historical junctures? In this chapter, I outline my theoretical argument, with party survival figuring as the dependent variable. The chapter begins by describing a dynamic approach to institutional analysis. 32 This dynamic approach provides the necessary framework for generating a causal argument sensitive to a combination of historical legacies, rules, and individuals who act strategically to advance their own interests. I then present and operationalize each condition of my theoretical framework. The third section briefly sketches how these conditions interact. 2.1. A Dynamic Approach to Institutional Analysis

The dynamic approach to institutional analysis I rely on combines traits of Historical Institutionalism (HI) and Rational Choice Institutionalism (RCI). This approach aims to treat survival as a variable, regardless of exogenous shocks and whether or not change is needed and actually occurs.

The study of institutions has drastically changed over the years. Rational choice, historical, and sociological approaches to institutionalism have predominated (Hall and Taylor 1996, Mahoney and Thelen 2010). Several studies have used the analytical tools of one of these major approaches to explain institutional survival. As has been extensively highlighted, this has had both advantages and disadvantages. Just a small proportion of theory-driven works has combined them (Hunter 2010).

32

Laver and Sergenti (2012) also provide a dynamic analysis of party competition. However, their model is based on agency, and it focuses on both the supply and demand sides of the equation. My argument here isolates the supply and analyzes the dynamics of parties and their members. 53

Regarding the assessment of institutional survival, Hall and Taylor (1996) argue that RCI provides an elegant explanation for the survival of institutions. The concept of equilibrium, understood as a Pareto-optimal situation for agents, explains endurance. Nevertheless, RCI arguments regarding equilibrium prove unrealistic. First, the RCI approach assumes that institutions persist, i.e. there is equilibrium, when they comply with agents’ interests and does not consider the possibility that past choices might impose constraints to the fulfillment of agents´ interests. They analyze outcomes as a result of games between agents, neglecting the role of structure and processes in determining the actual feasibility of certain outcomes. Second, they analyze equilibrium situations, neglecting the fact that survival is an ongoing process, for which it might be relevant to understand how the moments of survival analyzed are historically interconnected. Some defenders of the RCI approach have themselves recently begun to acknowledge this fact. Katznelson and Weingast (2005) mention that RCI has turned to history: “…[to] better understand how actor preferences have been fashioned and how institutions have introduced biases or other distortions” (p. 6).33 Levi (2009) also notes that RC arguments have been paying increasing attention to history, contingency, and processes. Although RCI has begun to take context seriously, Katznelson and Weingast (2005) write: “Still, the dominant trend with RCI is for the focus of particular models to be short term, relatively presentist, and closely linked to particular strategic situations with particular structures of payoffs” (p. 11). Writing in support of the HI tradition, Pierson (2004) argues in favor of “placing politics in time” (p. 2).34 His main contribution related to the analysis of institutional persistence and change is emphasizing the study of “institutional development” as opposed to “institutional change”:

33

The analytic narratives framework is an example of such an endeavor (Bates et al. 1998). His argument rests on three pillars: the relevance of path dependence with its characteristics and consequences; the importance of temporal order and sequence; and the need to incorporate the “slowmoving dimensions of social life” (p.14) and therefore broaden our temporal sweep. 34

54

Exploring institutional development brings to the forefront a range of fundamental concerns that actor-centered functionalism typically ignores: the implications of widespread unanticipated consequences in institutional design; the consequences for institutions of ongoing, sometimes rapid and often unpredictable change in surrounding social environments; the capacity of learning or competitive pressures to generate institutional change; and the possible bases of institutional resilience…” (p. 105). Although it corrects one of RCI’s main flaws, his historical perspective misses the role of agents. Actors’ incentives are actually crucial to understanding the prospects of an organization like a political party surviving. In fact, conventional HI paid less attention to agents and their preferences and was instead focused on: “…holistic analyses that treated structures both as outcomes of large-scale historical processes and as causes that, alone and in combination, established fields of action within which human beings lived, cooperated and conflicted.” (Katznelson and Weingast 2005, p. 13). Thus, conventional HI analyses have paid little attention to agents’ incentives and preferences.

Regarding their understanding of processes of persistence, HI studies distinguish long periods of stability and instances of massive change. The gradual processes by which institutions reproduce and survive are not problematized. They are also inattentive to agents’ role within the process, or along the path, where survival occurs. Thelen (2003), for instance, notes that path dependence arguments tend to divorce moments of institutional innovation from instances of institutional reproduction. During the former, agency and contingency are relevant, while in the latter agency and choice play a minor role. She notes that the HI approach treats institutional change too much as a function of exogenous shocks. However, institutions may change or decline in a more subtle way (pp. 212-213).

Departing from an HI perspective, several studies have provided analytical tools for addressing these issues. Recent HI studies have opened the black box between critical junctures and paid more attention to preference formation. 55

Greif and Laitin (2004) have tried to bridge the RCI and HI perspectives. On the one hand, they try to overcome RCI’s problems addressing change. On the other hand, criticizing Pierson’s idea of “increasing returns”35, they try to define: “…the mechanisms by which increasing returns take place” (p. 636). They introduce the concept of “quasi parameters” to fix the gap in Pierson´s argument, which introduces the possibility of endogenous change as well as the potential presence of self-undermining mechanisms.36 Although Greif and Laitin’s quasi-parameters make endogenous explanation of institutional change possible, these quasi-parameters need to be refined to be applied to a specific study.

More recently, Mahoney and Thelen (2010) have suggested a general framework for studying gradual institutional change with the inclusion of agency.37 Regarding the former, and in line with Greif and Laitin, they raise the need to include endogenous sources of change in HI analyses: “…a dynamic component is built-in [to institutions]; where institutions represent compromises or relatively durable though still contested settlements based on specific coalitional dynamics, they are always vulnerable to shifts” (p. 8). With respect to the inclusion of agency, they highlight distribution of power as a central incentive for institutional change within the path, arguing that such change can occur: …if the over-time distributional effects of institutions trigger divisions among institutional power holders. Or change can occur if institutions disadvantage subordinate groups to the point that they organize and come to identify with one another, thereby increasing their power and capacity to break prevailing institutional arrangements” (pp. 9-10).

35

See Pierson (2004, pp. 17-53). One of the main critiques (see Mahoney and Thelen 2010) to their framework is that it is difficult to distinguish ex-ante parameters from quasi-parameters. 37 Sometimes they instead use the term “institutional evolution.” 36

56

Mahoney and Thelen´s call for studying power-distribution conflicts might initially seem to recall RCI explanations. However, in contrast to RCI approaches, they extend attention to the disputes over power and distribution even when institutions are established. Hence, in their framework, institutional reproduction almost never involves a smooth “best choice for all” situation. Their approach also differs from conventional HI. The latter conceives power disputes as located mainly at critical junctures. Mahoney and Thelen bring actors back to the scene, correcting most HI analyses’ inadequate attention to the role of agency in the process of institutional reproduction—i.e. between critical junctures. These positive contributions aside, Mahoney and Thelen’s emphasis on change neglects the problem of how institutional mechanisms can resolve ongoing distributional conflicts and persist despite incentives for change. Though contributing significantly to greater dialogue between HI and RCI, Mahoney and Thelen’s emphasis on change makes it difficult to generalize their model.

To account for different types of survival, as I seek to do in this study, one needs to explain institutional change as well as institutional persistence or decline. My focus is not on institutional change or incentives for change, but rather on institutional reproduction based on given traits that reinforce or undermine survival.

In summary, although recent advances in HI and RCI have made significant contributions to analyses of the relationship between agency and structure, they have problems addressing how institutions persist through change or identifying the built-in institutional mechanisms in a specific institutional setting that can lead to gradual (or drastic and sudden) institutional erosion. New approaches to institutional change have emphasized that, along with the possibility of drastic changes, gradual transformations might also occur (Thelen 2003, Pierson 2004, Mahoney and Thelen 2010, Greif and Laitin 2004). However, the causal mechanisms accounting for types of persistence have

57

yet to be clarified.38 My study aims to contribute to filling that gap. Leaving aside the effects of exogenous shocks, what endogenous factors contribute to different types of institutional persistence and what explains institutional resilience in the face of different environmental challenges? What is the role of agency, history, and structure in the process of different types of institutional survival? These are some of the remaining questions the dynamic approach to institutional analysis applied in my study aims to help address.

Based on this review of HI and RCI approaches to institutional evolution and my assessment of the literature on political parties survival presented in Chapter 1, it is clear that a dynamic approach to institutional analysis needs to explicitly incorporate the effect of time and, in turn, the impact of past decisions and structures. It also has to combine agency and structure within its framework. Finally, it has to clearly establish which endogenous conditions have self-enforcing or self-undermining effects on institutions.

The aim of my dynamic approach to institutional analysis lies in the identification of causal conditions which allow party reproduction over time, including through periods of rapid environmental transformation. It captures causal conditions that account for their chances of persistence across time and junctures and, more importantly, how they persist. The causal conditions that factor in my theoretical argument thus combine choice and structure. They also incorporate references to processes within their definition and, finally, are evaluated across time. As Strom (1990) noted, conventional analyses of political parties have tended to be static. My focus on types of party survival brings dynamism to the analysis of political parties. Hunter (2010) provides an interesting application of such a dynamic institutionalist perspective to the analysis of political parties. Her study of the evolution of the Partido dos Trabalhadores (Workers´ 38

Advances have been made with respect to institutional change as a dependent variable, neglecting the need to fill substantial gaps on persistence and survival. 58

Party) also combines Rational Choice Institutionalism and Historical Institutionalism. This dissertation builds on Hunter´s perspective.

My dynamic approach combines agency, history, and the traditional notion of institutions as a set of “…stable, valued, recurring patterns of behavior” (Huntington 1968, p. 12). I am attentive to how different combinations of conditions across time, and their evolution, affect the outcome. My purpose is to unpack how the process of survival occurs and its impact. Relatedly, the idea of increasing returns is thus problematized, and comes into play as a variable useful for identifying sources of self-reproducing or self-undermining mechanisms. I treat parties as endogenous institutions (Aldrich 1995).39 In this sense, I follow Mahoney and Thelen (2010)’s and Greif and Laitin (2004)’s stress on the relevance of endogenous properties explaining self-enforcing mechanisms. The causal conditions of survival I identify combine agency, structure and the endogenous mechanisms contained within parties that explain their self-reinforcing or self-undermining survival. Together, I argue, these factors account for each type of party survival. 2.2. Introducing Conditions for Types of Survival of Political Parties

My argument is mainly inspired by the mechanisms of Hirschman (1970): exit, voice, and loyalty. Hirschman explains organizations’ ability to avoid deterioration, which is closely related to the aim of this study: explaining types of party survival and, more specifically, how parties retain their vibrancy over time. Thus, the following causal conditions could be interpreted as factors enabling Hirschman’s mechanisms; i.e., as conditions fostering the operation of these mechanisms within political parties. As will be thoroughly analyzed in Chapters 4 to 7, different interrelations between the causal 39

Aldrich mentions that “Instead of statute their basis lies in the actions of ambitious politicians that created and maintain them. They are, in the parlance of the new institutionalism, edogenous institutions...” (Aldrich 1995, p. 19). 59

conditions account for organizations’ ability to retain value over time, i.e., to avoid deterioration.

My first dimension is Exit. Hirschman (1970) introduces the idea of Exit, which has been widely studied in the literature analyzing the effects of rules, such as those of electoral systems. I evaluate whether politicians perceive parties as stable organizations and their perceptions of the chances of forming a new organization or of changes occurring in the constellation of parties competing in the system. I thereby evaluate their perceived barriers to exit. Where the chances of exiting are lower because of higher perceived costs a party organization is more likely to persist over time. 40 In essence, the higher perceived barriers to exit, the likelier the collective—party—organization is to persist. Hirschman (1970)’s Loyalty is activated by two mechanisms: trauma and purpose. Loyalty to a party in developing and unequal but consolidated democracies is mainly built through these mechanisms. Trauma activates retrospective loyalty. If a party was born during or in response to a dramatic political experience, or the party underwent such an experience, individuals united around that party label will feel themselves to be members of a group or community with the same memories and history, and that feeling will be intense.41 The intensity of identification resulting from trauma gives the party a chance to survive with vibrancy over time. Such traumas are particularly relevant in the context of developing countries which, again, typically share numerous challenges. The experience of a traumatic past forges strong bonds among members of a political party. It thereby helps to mitigate inherent conflicts emerging as a result of competition for scarce resources, such as positions in the Executive, in Congress, or when discussing sensitive policies. In contrast, a party born as a product of a secular process or as a result

40

I later introduce some adjustments to this relationship. As will become clearer in Chapters 4 to 6, the closer the traumatic experience is to the party’s foundation as a political organization, the greater the imprint the trauma leaves among members. 41

60

of a non-programmatic fight between two party leaders will find it problematic to build an identity and thus to arouse a keen sense of loyalty among its members. For instance, a party born as the result of an internal fight between leaders disputing positions that lead to an internal split and creation of a new party label can hardly be expected to call forth the same feeling of loyalty as one resulting from a revolution or civil war.42 Nevertheless, the cooperation ensuing in the wake of trauma has a “discount parameter” (Axelrod 1984). For instance, younger leaders of the PLN in Costa Rica are less united than older ones by the traumatic experience of the country’s civil war. The same applies for the Uruguayan and Chilean left, and their experience under authoritarian regimes.

Purpose activates prospective loyalty. Party members are loyal to a collective project, an idea about the future. A party that was born with a definite purpose and maintains it has more chances of recruiting devoted adherents and ambitious politicians sharing a clear profile. A voluntary organization normally requires a significant number of devoted adherents to exist. This requirement is particularly relevant in developing democracies with daunting challenges. A party might define its purpose constantly throughout its trajectory as an organization, but it can also change it. However, as will be explored further below, the effects of these transformations depend on their pace and interaction with other causal conditions. Ambition activates Hirschman’s “voice” mechanism because of its role within the organization. Voice is activated through ambition because this condition ensures that different prospective leaders have the chance to promote their new ideas, even where they diverge from those of the established elite. The more ambition is satisfied through available organizational mechanisms, the greater the chances that new leaders—new 42

Relating my idea of Trauma to arguments regarding the relationship between the timing of democratization and global trends (Dalton and Wattenberg 2000 (Dalton and Wattenberg 2000, Mainwaring and Zoco 2007, Webb and White 2007a), the pervasiveness of individualism, globalization, and television becomes less relevant when a party, as a community of individuals, shares some traumatic experience. 61

voices—will refresh the organization when they are willing to do so. My idea of ambition relates still more directly to Aldrich´s (1995) theory answering the question “why parties” exist? Parties, he argues, provide resources for channeling collective action and individuals’ ambitions. I concur with this explanation. Aldrich’s axiom is: if parties qua organizations can satisfy individuals’ ambitions, the likelihood of vibrant reproduction of the organization increases. As will be extensively analyzed in Chapters 4 to 7, vibrant political parties are those in which politicians feel they have a chance to build their political career. It is thus a party that both young and established leaders see as a promising channel through which to pursue their political ambitions. The UDI in Chile has informal mechanisms of elite promotion, and new UDI leaders constantly emerge. In Uruguay, this also happens to varying degrees in two main political parties (PN and PC). Other parties, like the DC and PS in Chile, by contrast show symptoms of difficulty promoting new leaders, and their own politicians feel the party has become ossified. 2.3. How are Exit, Ambition, Trauma, and Purpose Conceptualized?

Exit

This dimension of my theory is drawn directly from Hirschman (1970). My main difference with Hirschman lies in the fact that he identified exit as pertaining to the economic realm, understanding it in terms of whether customers decide to buy a product from another company or not. In his model, customers use the market to defend their welfare or to protect their interests. From the firm´s perspective, Exit is thus a deteriorating mechanism. This mechanism of exit is, however, also clearly present in the political arena. Its use is more likely under a democratic regime, though as Hirschman (1970) argues, in politics it is not as simple to turn to it as it is in the economic market

62

where a customer simply needs to buy another brand.43 This mechanism fosters partisan organization when politicians from a political party perceive high barriers to leaving their organization or to pursuing a career as an independent.

One way to evaluate the relative opportunities for exit in different systems is to survey different systems’ electoral rules. These rules play an important role in shaping politicians’ decisions to remain in a given political party, join another established party, or simply form a new organization. Electoral rules also generate different incentives for politicians to cultivate personal reputations as opposed to more partisan ones (Carey and Shugart 1995, Samuels 1999, Crisp et al. 2004). In Cox (1997)’s terms, if the district magnitude increases and if the party’s list is closed, then the chances for collective, party strategies to outweigh individual reputation also increase. However, these kinds of formulations cannot account for the differences I observe in my sample among parties within a given system at a given point in time. To capture the effect of this condition at the party level, I sought politicians’ assessments of the stability of the system and the probability of new organizations emerging. Even though this condition might at first seem to operate at the system level, individuals from different parties might actually have diverging evaluations of the system’s stability and the costs of exit from or entry into it. If politicians from different parties differ in their evaluation of the possibility for new parties to emerge or for politicians to leave a party (and the actual probabilities are determined by the type of rules and the party system’s history), then perceived costs differ from what the systems’ rules alone would indicate. A gap of this kind is precisely what I observed in my fieldwork. To incorporate this gap I have extended the consideration of barriers to exit beyond the analysis of the effects of

43

This dissertation focuses on the supply side of the equation (as does Hirschman when referring to the political arena), but citizens also use this mechanism, for example, when they decide not to participate in an election. 63

rules. For example, in Chile politicians from different parties have different (and even opposing) assessments of the costs of party exit and entry.

My argument regarding Exit is as follows. Where perceived barriers to exit are low, parties’ chances of surviving over time will be reduced. Where perceived barriers to exit are high, the effect depends on its interaction with the other conditions I discuss. Basically, if politicians’ perceive high costs of exit or high stability of the party system, it is assumed that established parties will play an important role qua organizations. Politicians’ incentives to strike out as independents or foster the emergence of new parties will then be low. However, my main contribution with respect to this dimension lies in my consideration of how these perceptions interact with other causal conditions. As Chapters 4 to 7 will analyze, depending on the values of other causal conditions, high perceived barriers to exit can also hinder party survival. This sort of conditionality also applies to the effect of my other causal conditions; my theory is ultimately one operating in terms of conjunctural causation (Ragin 1987). In sum, the impact of Exit should be analyzed in conjunction with Ambition, Purpose, and Trauma. However, it could be said that Exit has a curvilinear effect on party survival. The empirical chapters examine this curvilinear effect more closely, but the basic idea is: rules of the game cannot raise costs of exit too high without risk of freezing political competition and thus ossifying political parties.44

Trauma

Trauma breeds retrospective loyalty among party members. When a party is born as a product of a significant political crisis or undergoes one somewhere during its history, 44

The curvilinear effect is thus given by rules, though Exit is operationalized in a more complex way (see Chapter 3). Schedler (1995, p. 17) discusses this effect through his analysis of electoral market openness and its impact on what he calls “over institutionalization.” 64

this experience generates Trauma as a bonding mechanism. The closer the trauma is to the party’s foundation, the greater its impact. I am restrictive regarding the presence of this condition, for which only extreme situations count: dictatorship45, civil war, persecution, or marked ethnic discrimination. Each of these situations undeniably affects a group, and its relevance does not depend on the group’s decision to politicize or mythologize it.

It could happen that the organization decides to neglect that experience as a political asset. However, the existence of this possibility does not contradict my definition. It is sufficient to think of trauma as a condition that brings people together and creates a sense of community as a consequence of a dramatic experience. It is not something that the organization necessarily uses to get votes or to create an identity. In this sense, even if the party is “in denial”, it does not affect the undeniable fact that party members underwent a traumatic experience, which is known to them and to other citizens. During my interviews, politicians from the left in both Uruguay and Chile made many references to their past during the authoritarian regimes and the epic of fighting against it, regardless of the electoral use of that past.

How does this condition operate in settings where self-interested individuals are pursuing their own ambitions? What does it add to the relationship among politicians competing for scant resources? Simply put, why does trauma matter for party survival? I argue that trauma gives the party symbolic resources and bolsters members’ sense of common identity. Moreover, it also stimulates cooperation among individuals. This is particularly relevant for the type of democracies analyzed in this dissertation. They are all democracies in developing countries, where structural conditions present latent—and sometimes explicit—challenges to the political regime. A shared traumatic experience

45

Authoritarian regimes have unfortunately been common in Latin America. However, they have not generated the same impact on all politicians. Some were especially persecuted and suffered exile or torture. Those are the ones who suffered during dictatorships. 65

softens potential disagreements within a party, facilitating collective organization. In some sense, it is the “weight” parameter presented in Axelrod (1984), but, operating in the opposite direction46; that is, individuals cooperate because of a past experience. However, as with Axelrod’s weight parameter, its positive impact on cooperation declines as time passes. A shared traumatic experience among party members creates bonds among individuals competing for political positions and other scant resources and harmonizes their relationship. Over time, it becomes less significant as that traumatic experience wanes in their memories and as the organization recruits new members who did not participate in the traumatic experience directly and partake of it only indirectly, through identification. My appreciation of the importance of Trauma as a causal condition emerged from my extensive fieldwork. In Chile, some of the PS and PPD politicians made constant references to the importance of Pinochet’s regime and repression in generating incentives for their being united. The same was true in Costa Rica with memories of the 1948 Civil War. For older politicians, that experience is still an incentive for continued unity. However, its distance in history is constantly mentioned by younger politicians, who often refer to it as “a thing of the past”, and consider it an impediment to finding new, more current reasons for being united. Purpose

Purpose breeds prospective loyalty. Purpose is observed when a party creates or coherently politicizes a project, a set of ideas, related to a relevant cleavage and intensely defends a position related to it. Hence, Purpose is the worldview, the set of ideas, the ideology, or project of the organization, the degree to which it is shared, and the intensity with which it is defended and projected. The organization shows Purpose when it creates common goals that supersede individual careers and are translated into a 46

Axelrod (1984)’s “weight” parameter refers to the duration of the relationship between two players. The further they are from the expected end of their interaction, the greater the weight parameter and the greater the chances they will cooperate. 66

coherent and identifiable position of the party as a whole.47 Purpose is therefore present when two conditions are met: when the party as a whole has some identifiable and coherent long-term goals which guide its actions; and when it strongly defends and insists on them. When those goals are in line with the main conflicts associated with modernization (particularly that of redistribution) or associated with some significant cleavage, it is more likely that Purpose will be intense, regardless of whether the cleavage is used instrumentally to capture votes.

Purpose is programmatic coherence within the party, and it may or may not result in programmatic linkages with voters (see below). In this sense, (Coppedge 1998a) says: “Sufficiently ideological parties are thus parties that take clear, widely understood positions on a conventionally interrelated set of issues” (p. 552). Benoit and Laver (2006) provide a very interesting analysis of political parties’ policy positions. They measure different dimensions of policy positions and the relative importance each has for a political party. They thereby measure the intensity of parties’ policy positions, but without considering their coherence, which is also critical.

Coherence can be regarded as a necessary but insufficient component of purpose. The same could be said for intensity of the identifiable issue or cleavage politicized; it is necessary but not sufficient. Both attributes (coherence and intensity) must obtain for purpose to be considered present.

Politics is the arena where public goods are produced and delivered. Even though club, private, and common pool resources are also produced and delivered, the main trait of this arena lies in the provision of public goods. Democracy is the regime where public goods are expected to be the principal goal of agents. Thus, evaluating the degree to

47

For this condition, it is not relevant whether the source of these goals is the impulse of an undisputed leader or collective deliberation. 67

which collective agents, i.e., parties, agree to and defend a specific view of different public goods seems logical.

No matter what kind of electoral strategy parties use, it is worth analyzing the degree to which the party collectively agrees on its ideas regarding public goods.

48

It is more

important that partisans and leaders agree strongly on distribution of public goods, which has huge implications, than on the distribution of private or particular goods. History is telling in this regard. The most important Latin American civil wars and conflicts occurred because of disagreements on decisive public goods. Examples include the 1948 Civil War in Costa Rica49, the successive nineteenth-century “revolutions” in Uruguay50, the “ruido de sables” episode in Chile in 1924. Therefore the greater the degree of coherence within a political organization concerning the production and provision of public goods, and the greater the social importance of the goods regarding which there is such agreement, the more likely the party is to endure over time.

48

For many decades, scholars have analyzed whether political parties appeal programmatically to voters, thus essentially evaluating the balance between private and public goods provided by different political parties and party systems (e.g., Miller and Stokes 1963, Kitschelt 2000, Laver 2005, Kitschelt and Wilkinson 2007, Kitschelt et al. 2010). However, Luna (forthcoming)’s detailed analysis of Chile and Uruguay effectively challenges conventional approaches to linkages between citizens and parties, showing that political parties in unequal democracies, even Latin America’s most consolidated and developed ones, typically segment linkages to capture socio-economically divergent constituencies. Thus, it is futile to assume that political parties only emphasize one of the two types of goods, that is, that they are either clientelistic, solely emphasizing private goods, or programmatic, solely emphasizing public ones. Other scholars have also advanced this line of thinking (e.g., Valenzuela 1977, Coppedge 1998a, Taylor-Robinson 2010). Moreover, Luna (forthcoming) demonstrates that the most successful political parties are those that segment their linkages with voters while also harmonizing their electoral strategies. Thus, the conventional assumption that a given party offers either a platform or private goods (but not both) is unrealistic. Parties need votes, and they hope to win elections. At least in unequal societies, they therefore use multiple strategies, often including both clientelism and programmatic appeals. Surveying party-voter linkages is beyond the scope of this dissertation, which focuses exclusively on the supply side of the equation. I am thus only concerned with whether a political party builds a distinctive identity or reputation and how this affects its organizational vigor in terms of whether it ensures the dedication of party militants and politicians. 49 The PLN was born right after that war. 50 In fact, the two traditional parties, Colorado and Nacional, both emerged as a consequence of the same major civil war (see chapter 6). 68

The idea of intensity—as a synonym for salience—penalizes parties that only embrace valence issues. Because intensity of commitment on certain, defining issues fortifies party organization, it is important for parties to have such intensity, and not only embrace valence issues, if they are to survive in the dynamic sense. Moreover, even assuming a normal distribution of preferences among voters, vibrant parties will not position themselves at the center as median voter approaches posit (cf. Downs 1957). Finally, vibrant parties are those that are able over time to maintain an identifiable identity in terms of ideas, gaining a reputation for consistency while it tries to win over an increasing proportion of voters.51 The idea of intensity is thus at odds with electorally successful parties that significantly and rapidly change their platforms to capture more votes. Though potentially successful in the short term, such a strategy can carry high long-term costs when it comes to party survival (see Lupu 2011).

Nonetheless, there is an intrinsic tension between democratic stability—which might very well be ensured by moderate parties—and intensity of Purpose. In dichotomous and schematic terms, developing democracies face the following dilemma: democracy there is either boring and stable or intense and unstable. There is no straightforward solution to this dilemma, and my definition of Purpose takes this pervasive—and widely demonstrated—reality into account. Not only extremist parties have Purpose, however. Some parties resolve the tension between stability and engagement by taking a clear position on significant cleavages while also embracing valence issues and moderating their position on other cleavages or issues. Those parties would be located within the upper-right corner of my Graph 2.1 (the Purpose area), but not at its extreme. Does it matter when a party’s Purpose emerges? How does Purpose evolve? As will be demonstrated through empirical contrasts in the following chapters, Purpose arising in

51

Since my definition of a vibrant surviving party does not entail its being electorally successful, these efforts need not prove compelling. 69

the first stages of a party’s development is stronger than purpose generated after the party label has consolidated.52 My theoretical argument also allows for potential gradual change in parties’ programmatic positions. As argued in Chapter 1, other scholars have identified adaptability as a crucial trait explaining parties’ relative ability to survive different junctures (Levitsky 2003). However, these studies have assumed a given direction of change (the neoliberal turn) to be synonymous with adaptation, and the speed of the “adaptation” they portray effectively divorces the idea of programmatic supply from that of democratic representation. I am not neglecting parties’ need for ideological renewal: adaptability as change of a party’s ideological stands is often crucial. Nevertheless, to remain consistent with voters’ preferences53, major programmatic changes have to be gradual (Carmines and Stimson 1989 provide an example of this idea).54

One can thus summarize the logical relation between purpose, coherence, and intensity as follows:

Hence, in order to measure a party’s Purpose I use the “weakest link” logic (Goertz 2006): the smallest value on each of the two necessary (but insufficient) terms (coherence or intensity) gives the value of purpose for a given party. Since political parties are heterogeneous organizations, it is desirable to use measures capturing such 52

This finding is in line with historical institutionalism’s basic principles of path dependence and increasing returns. The political party literature also shows that a party´s origin leaves a genetic imprint (Mustillo 2007 or , Panebianco 1988). 53 That is in fact the basic assumption underlying the Responsible Party Government thesis (Schattschneider 1942, Adams 2001). 54 The authors describe the evolution of a very divisive and contentious issue in American politics: race. Thus, their analysis is in line with the claim made above about the impact of substantive issues. 70

variability in levels of Purpose. Whatever indicators or data are chosen, it is theoretically relevant to report a range of values across members within a party. The minimum on Coherence and Intensity gives their respective value for the party over time and, again, the smaller of the two gives the party’s value on Purpose.

Graph 2.1 presents the hypothetical examples of Parties A, B, and C. Each party has different ranges on each dimension, resulting in different values on Purpose. This figure is simply illustrative. Given my qualitative approach, I just consider the “weakest link” area (P) for each party. That area represents the combination of the minimum value on Coherence AND Intensity. Had I preferred to use a specific value, rather than an area, I would have had to take the smallest value on the two dimensions as the value for Purpose. Clearly, Party C has the greatest Purpose, while Party B has the lowest Purpose. In fact, Party C is the only one located in what I call the “Purpose Quadrant.” Notice that each party´s combination of intensity and coherence is not necessarily a perfect circle. A party might be very coherent, but its members or politicians might show a wide range of perspectives in terms of intensity, resulting in a vertically elongated ellipse, as in the cases of Party A and Party C. Or, there may be little variation in intensity but a wide range in politicians´ coherence, resulting in a horizontally elongated ellipse as in the case of Party B. Thus, longer lines (horizontal for coherence and vertical for intensity) show lower levels of agreement between interviewees.

The four quadrants show: 1) an intense but incoherent party; 2) high coherence, high intensity; 3) high coherence, low intensity; 4) low coherence, low intensity. The upperright quadrant of each graph is the “Purpose Quadrant”. The closer a party is to this quadrant, the more Purpose the party concerned has. The closer to the upper-right part of that quadrant and the more compact the yolk (Tsebelis 1995), the more politicians regard their party as vibrant.

71

Graph 2.1. Hypothetical Examples of Purpose Purpose Quadrant

C

Intensity

A

P

P

B

P

Coherence

One straightforward measure of each dimension of Purpose is available through existing survey data. For example, PELA’s measure of politicians’ self-placement across the ideological continuum could yield a good proxy for coherence, and elites’ range of answers on specific issues is a good proxy for intensity. Kitschelt et al. (2010) also have databases on politicians useful for capturing my dimensions of purpose. However, for this dissertation I will instead codify my interviewees’ answers on different questions.

72

Ambition

“…Individuals must have their political ambitions excited as well as restrained.” (Schlesinger 1994, p. 35).

This condition is derived from conventional rational choice approaches to political parties. It is also in line with Downs (1957, p. 25) and Schlesinger´s (1994) definition of parties as organizations composed of office seekers (Schlesinger 1994, p. 6). More specifically, it builds on Aldrich (1995)’s claim that parties solve collective action problems for ambitious politicians.55

In this sense, there are theoretical accounts of the relationship between parties and the office seekers of which they are mainly comprised. Some scholars have analyzed the chances of someone going “through the ‘eye of the needle’ into the highest office of state” (Norris and Lovenduski 1995). In Latin America, there are unfortunately few works focusing on political careers and their causes (e.g., Botero 2008, Samuels 2003, Siavelis and Morgenstern 2008, Chasquetti 2010). Siavelis and Morgenstern’s framework focuses on party and institutional variables producing different types of candidates, which in turn affects political behavior (2008, p. 4). Their framework contributes significantly to the understanding of types of candidates. However, I am mainly interested in whether the party acts as a supplier of resources and as a promising channel for advancing a political career, not in what kinds of politicians are nominated and how this influences their behavior.56 I am not interested in surveying an individual’s actual chances for furthering his or her political career or in the actual stages he or she undergoes in doing so. I therefore relax the classic assumptions regarding discrete, static, and progressive ambitions (Schlesinger 1994, pp. 39-40) and their causes. For this 55

Strom (1990) also mentions the resources that parties provide (or need to provide) for politicians to become elected (pp. 574-575). 56 The role of the party in the advancement of an individual career is something that Botero (2008) and Chasquetti (2010) do consider. 73

dissertation, these different types of ambitions are irrelevant. The only question that is being answered is whether an individual politician thinks his party can help him progress in his career.57 In analyzing parties’ origins, the literature highlights the role of leadership (Pedahzur and Brichta 2002). My argument does not neglect the role of founding leaders, but I use Ambition as a more general term also capturing leadership reproduction across time. Again, the presence of mechanisms for satisfying individuals’ ambitions contributes positively to ensuring the ability of a party to survive with vibrancy over time, and to outlive its founders. Ambition in this sense encompasses a common trait of personalistic parties, in which individual leaders—or small groups of politicians— use a party label to promote their own political careers. Such parties are pervasive in Latin America. When considered in interaction with the other conditions I identify, however, it becomes clear that ambition as it figures in such personalistic parties is certainly not sufficient for predicting a party’s survival qua organization.

Others have analyzed mechanisms of candidate selection and their effect on democraticness or parties’ legitimacy with voters (Carey and Polga 2006) or on policy outcomes (Cross 2008). Hazan and Rahat (2010) provide a comprehensive appraisal of the political consequences of candidate selection mechanisms in terms of participation, representation, competition, and responsiveness (pp. 89-164). While taking advantage of these analyses, I claim that the substantive effect of Ambition is nevertheless determined by the interaction with the other causal conditions here presented.

This dissertation conceptualizes Ambition as the degree to which the political party is a potential and perceived channel for individuals’ ambitions. The availability of

57

While some analysts do focus on the effects of candidate selection and recruitment into parties qua organizations, they are not interested in party survival as a dependent variable (Freidenberg and López 2001, LeDuc 2001, Pennings and Hazan 2001, Hazan and aha 2010). 74

mechanisms for fulfilling a political career, thus satisfying individuals’ ambitions, partly explains the capacity of a party to survive with vibrancy over time. A party is able to adapt and thus survive over time if it shows potential for elite renovation. Renovation, or perceived potential for promotion of new leadership, signals adaptability in a broader sense because it reflects the ability of a party, qua organization, to remain attractive to new generations that, in turn, arrive with new ambitions and eventually new ideas. The mechanisms for channeling a politician’s ambition may be formal or informal. If formal mechanisms exist, they are usually part of the electoral laws (e.g., mandatory party primaries). Nevertheless, some parties might also have their own clear and respected rules of elite promotion. However, as has been extensively considered in previous analyses of political parties, internal rules are subject to constant changes (Aldrich 1995, Hazan and Rahat 2010, pp. 6-7). Their inconstancy complicates identifying and systematically analyzing parties’ mechanisms of promoting new leaders by solely focusing on internal written rules. If informal mechanisms are the only ones available, then promotion has to privilege insiders rather than outsiders in order to give value to the political organization. If a very small group of party leaders alone act as recruiting agency, party organization is meaningless and potential militants with intentions to pursue a political career will rapidly abandon the organization. Also, those informal rules, though not written, should be well-known within the organization. The party must have a reputation for promoting new leaders. Through extensive fieldwork, it is possible to capture whether a party offers prospective leaders the promise of pursuing a political career within the organization. One can also see if politicians are convinced that they can pursue their careers within the party and if they regard the party as a crucial enabler for channeling their ambition in that regard, regardless of specific written party rules.

In my interviews I asked both established, well-known leaders and young, prospective ones whether they considered that their party had informal or formal mechanisms of elite 75

promotion. I also asked established politicians to imagine beginning their careers again today, and contrasted their recounting of each step they thought they would need to accomplish with young politicians’ accounts of what they thought they needed to do. Their contrasting answers on these questions gave me a clear idea of each party as a valued channel of political ambitions. 2.4. How are these dimensions related?

As mentioned before, the theoretical argument advanced in this dissertation builds on the notion of “multiple conjunctural causation” (Berg-Schlosser et al. 2009, Ragin 1987, 2000). I will therefore briefly present how the aforementioned causal conditions interact. In the following paragraphs I will introduce the main interactions observed as a result of my fieldwork. Other logical relationships observed will be developed in the empirical chapters (Chapters 4 to 7). Finally, and as will become clear in those chapters, they form a set of sufficient conditions. That is, they are parts of a set of conditions that are jointly sufficient to explain types of party survival (Mackie 1980, Goertz and Mahoney 2012). Given that parties’ ability to remain legitimate, valued, and active organizations is here assumed to be a never-ending process, these causal conditions have to be evaluated across time, not on the basis of a static “snapshot” of the party at one single moment in time. This account of the evolution of a party, based on what I call a dynamic approach to institutional analysis, is needed to obtain better explanations of types of survival that can then be used as an independent variable. For that purpose, the definition and operationalization of my causal conditions include temporal references and I outline how these dimensions behave over time in the cases I analyze.

I will first analyze some pair of interactions and then briefly discuss each of these combinations of all four causal conditions.

76

Pairs of Interactions

Trauma reinforces Purpose. Retrospective loyalty increases the chances of prospective loyalty. It seems that a party that undergoes a traumatic political experience is more likely to have a strong purpose as an organization. Also, when both trauma and purpose are present, a party’s survival is more likely to be dynamic.

The Frente Amplio in Uruguay and UDI in Chile are cases in point here. For different reasons, both parties underwent significant traumatic experiences. Politicians from the Frente Amplio share memories of jail, exile, and in some cases torture during the authoritarian regime (1973-1984). Their experience in those years left a very strong imprint on their sense of collective identity and reduced divergence in their economic and social views. In fact, it could be argued that their intense and coherent programmatic inclinations were possible because of the sense of needing to remain united after their traumatic past experience. UDI´s leader, Jaime Guzmán, was assassinated on April 1,, 1991 while leaving the Pontificia Universidad Católica’s campus Oriente. He was the undisputed leader of Gremialismo, the student movement at the PUC forming the backbone of what later became the UDI party. His killing had a strong impact on his closest—young— collaborators. His collaborators’ sense of sharing a mission to transform Chilean politics and to preserve the legacy of the model established by the authoritarian regime, a concept he always insisted on, was powerfully reinforced following his assassination. Trauma and Purpose thereby strongly interacted in this case.

However, this interaction between Trauma and Purpose is not necessary. If it were, it would be pointless to separate these two mechanisms. A party might politicize a relevant conflict, but it might emerge at a period of relative normalcy, as in the case of the PAC in Costa Rica, for instance. Hence, the possibilities of building a collective identity are not determined by past suffering. However, the lack of shared traumatic experiences 77

reduces the chances for building a collective identity which, in turn, fosters establishing long-term organizational goals. A party also might have undergone a traumatic period but then neglect the sort of prospective loyalty activated by purpose by choosing to avoid conflict and promote “national unity” or other valence issues. Indeed, such is the case of some parties within the Concertación in Chile.

The presence of high barriers to exit (Exit) and Ambition also commonly appeared in my cases. Electoral rules, such as PR, high district magnitude, and blocked lists generate more competition between and coordination within parties. Also, concurrent and fused presidential and legislative elections force the maintenance of a smaller number of broader parties. This reality is generally perceived by politicians; they perceive high barriers to exit. This reduces perceived chances of personal voting and the chances of new parties emerging. Also, because of the effects of high district magnitude, PR, and blocked lists, Ambition is almost automatically satisfied. Almost no purposive and explicit action is required from partisan leaders to promote new entrants to the political career. Politicians from different generations have the sense that the party provides channels for satisfying their ambitions.

These connections notwithstanding, the presence of high barriers to exit (Exit) does not always ensure that Ambition will be satisfied. Electoral rules do not automatically ensure satisfaction of individuals’ ambitions. There is therefore variance among Uruguay’s political parties, despite the shared institutional conditions within which they operate. Although the country’s use of PR, blocked lists, and concurrent and fused elections automatically stimulates partisan organization and generates incentives for cultivating one’s political career through existing party channels, Uruguayan parties vary in how well they fulfill the condition of Ambition. Thus, Ambition can be isolated as a separate, organizationally dependent mechanism.

78

A similar pattern obtains for most of the Chilean parties analyzed. Barriers to exit are high, and electoral rules force the maintenance of a reduced number of political parties. More importantly, they also stimulate coalition building. However, parties organize their nominations in such a way that most of the interviewees felt that their party organizations hindered the careers of new leaders. In turn, the effect of Exit, given the value of Ambition, is the opposite from the situation described above. It forces politicians to remain within established parties, but, given parties’ failure to promote new leaders, younger members feel trapped within ossified organizations. This is another example of the relevance of taking into consideration interactions of conditions.

Again, this stylized description is thoroughly analyzed in Chapters 4 to 7. Illustration of Interactions of Causal Conditions

Different combinations of Exit, Ambition, Trauma, and Purpose account for parties’ membership in the two types of party survival. I now describe the most relevant combinations observed in my fieldwork. These combinations of conditions will be analyzed in-depth in Chapters 4 to 7.

The existence of high barriers to exit and having suffered a traumatic experience and purpose and mechanisms that continuously satisfy individuals’, especially new entrants’ ambitions, are together collectively sufficient conditions for explaining the persistence of a vibrant and valued party over time and across different historical junctures. In short, the existence of this combination of conditions explains a party’s dynamic type of survival. The combination EtPA would be mainly observed where trauma has waned, but perceived high barriers to exit still foster partisan organization, the party maintains a clear purpose, and it is able to channel individual ambitions.

As will be discussed in the empirical chapters, there are three different combinations of conditions sufficient to explain dynamic party survival. Thus, my argument allows for 79

equifinality. However, as will also be seen, these sufficient conditions result in different degrees of membership in the set of dynamic party survival: Degree of membership

Dynamic

Causal conditions

ETPA

1.0

ETPa

0.75

EtpA

0.50

Another observed combination of conditions is the absence of perceived high barriers to exit, and the presence of Ambition combined with the absence of trauma and purpose. The clearest example of a party experiencing this combination of conditions is that of the PAC in Costa Rica. The party emerged in a context in which politicians did not perceive that it was costly to exit established organizations.58 The party has always put emphasis on promoting new leaders, and their members feel that they can advance a career within the PAC. Hence, Ambition is observed. However, this party emerged without having experienced a traumatic past. It appeared as a consequence of the public protests and social mobilization against the privatization of the state company in charge of telecommunications and electricity, ICE (Instituto Costarricense de Electricidad). The protests were popularly known as Combo ICE, but the PAC’s members were not subjected to political persecution, imprisonment, exile, or other such traumatic forms of repression. This absence of Trauma inhibited the formation of a community with a shared sense of belonging. Memories of the protests are still evoked, but they do not play the role of a traumatic experience. Finally, the party is not coherently and intensely 58

My argument is not suited to explaining parties’ emergence, a phenomenon affected by numerous factors. Explaining that exceeds the aim of this dissertation. The PAC was born in the context of an emerging crisis of Costa Rican bipartism. The perceived costs of forming a new party were reduced in that situation. A window of opportunity opened, probably changing politicians’ assessments of the costs and benefits of forming a new party. 80

united around a Purpose. This combination of conditions explains why the party has not been able to continue to reproduce its vibrancy well beyond its founding.59

The existence of a traumatic experience, but without perceived high barriers to exit, and mechanisms of elite promotion (absence of Ambition) and absence of Purpose explains static party survival. In other words, loyalty solely sustained by a past traumatic experience, leads to static survival. As this finding implies, the effect of a shared trauma erodes over time, much like a discount factor (Axelrod 1984). The organization should not suffer erosion close to the time of the trauma. Nevertheless, as years go by, the epic of the heroic times of, say, fighting against persecution by an authoritarian regime or fighting on one side of a major civil war become increasingly irrelevant to keeping the collective together. Such is the case of the Socialist party in Chile two decades after the transition from authoritarian rule.

The extreme opposite of full membership in the dynamic type of survival set is given by the combination of etpa. This combination produces static party survival. For such parties, trauma either never occurred or faded over time, and politicians do not perceive high barriers to exit, whatever the actual rules of the game. Moreover, they do not identify Purpose in their parties, and they do not feel the party promotes elite renovation. The presence of just one of my causal conditions (and the absence of the rest) is also sufficient for causing static party survival. Conversely it is insufficient to produce dynamic party survival.

Throughout the empirical chapters (4 to 7) I will also analyze the evolution of each of my causal conditions in each party and will emphasize the interactions between these causal conditions. In this sense, for example, I will show how Trauma fades and its potential negative effect on Ambition. Those who have suffered the traumatic event are 59

In Chapter 5, I will thoroughly analyze the evolution of Purpose within this party. Here I am presenting a stylized description. 81

bearers of an epic which—though inadvertently—negatively affects the prospects of new leaders. Also, I will tentatively describe a hypothetical route followed over time by dynamically surviving parties. In this sense, I will show that Trauma and Purpose emerge first, but Exit and Ambition have to follow to ensure dynamic party survival over time.

It could be argued that other exogenous variables should also play a relevant role. However, I will demonstrate that these four conditions are sufficient to explain types of survival. The scope of my theoretical argument is Latin America and, more specifically, the more democratic regimes in the region. It is also worth insisting that my argument does not pretend either to explain how parties that will survive emerge or to identify necessary or sufficient conditions for understanding parties’ collapse. Conclusion

This chapter has introduced my dynamic approach to institutional analysis. The purpose has been to develop a theoretical framework sensitive to the analysis of political parties across time. Consistent with a theory combining RCI and HI, I emphasize how the causal factors I have identified are interrelated, and their respective role across time. I have introduced the foundations of my theoretical argument. In this sense, I have introduced each of my causal conditions: Exit, Trauma, Purpose, and Ambition. I have highlighted their role in party survival and introduced some interactions, a crucial aspect of my theory.

The following chapters will be dedicated to a detailed analysis of major political parties in Chile, Costa Rica, and Uruguay. I will analyze exit, ambition, trauma, and purpose for each party.

82

Chapter 3: A Guide for a Systematic Analysis Introduction

Some of the conditions I identify are novel or introduce significant theoretical departures from prior research. Discussion of these conditions is therefore warranted before proceeding with the QCA. This chapter provides the basis for the QCA analysis which comprises the bulk of Chapters 4, 5, 6, and 7. The goal is to show how each condition was computed. The chapter thus provides a link between the conceptualization and empirical assessment of my theoretical argument.

The first section of this chapter is dedicated to a synthetic description of the codification process. The second section presents the rules used to go from evaluating the units of observation to coding the unit of analysis. Finally, I then introduce the evidence and rules used to qualify the outcome (type of party survival).

3.1. Codification of each Condition

Politicians are the ones who maintain the organization (Aldrich 1995) and their testimony is therefore key to determining whether particular causal conditions are present or absent in a given party. The bulk of the evidence used for my analysis comes from in-depth interviews with 148 politicians in the three countries. I conducted fiftythree interviews in Chile, fifty-one in Costa Rica, and forty-four in Uruguay. (See Appendix 1 for more details.) The sample was designed to avoid bias and thus included: 1) politicians from different generations; 2) politicians at different stages of their careers; 3) relevant internal cleavages and factions within each party. To determine whether a politician could be regarded as an established leader, I consulted country specialists among scholars. I also considered whether interviewees had been in politics for long and had occupied relevant positions (senatorial seats, ministerial positions, the presidency, etc.). 83

Although unconventional, the computation of crisp-set values for each political party will thus be derived from qualitative evidence. Codifying qualitative evidence carries its own risks. It is also subject to potential flaws, misinterpretation of an interviewee’s claims being the main source of error. However, I believe it is worth coping with these types of problems rather than using unhelpful hard data, especially since those required for this study are mostly unavailable.

I will now briefly explain how I have computed the crisp-set value of each of my causal conditions for explaining types of party survival for each politician on each condition. Exit: High barriers to exit are a key mechanism for fostering partisan organization. As

reviewed in Chapter 2, institutionalist analyses have claimed that rules determine incentives and constraints for political agents (e.g., Hall and Taylor 1996). The main function of rules by this account is to channel incentives for seeking collective organization. Thus, their function in terms of type of party survival is to delimit the incentives of ambitious politicians, both within and between organizations. In this sense, electoral rules are understood as the main set of institutions affecting party organization (e.g., Duverger 1954, Sartori 1976, Cox 1997). In general terms, the more politicians perceive their individual career as tied to their party’s success at the ballot box, the higher the costs they perceive to forming a new political party or to pursuing a career as an independent. The wider a collective organization one needs to succeed, the more politicians hesitate to strike out on their own. In this sense, high barriers to exit set by rules stimulate and reinforce collective organization within a stable political party. More specifically, the more congressional and presidential elections are tied and the higher the district magnitude, the greater the advantages politicians perceive to staying within the party. This illustrates the conclusions of Carey and Shugart (1995) and Cox and McCubbins (2001): closed and blocked, concurrent, and fused elections reinforce barriers to exit and foster collective organization. As the institutionalist perspective

84

would lead one to expect, interviewees indeed made constant references to rules and their impact on partisan organization. However, other factors also contribute to determining politicians’ calculations of the costs of staying, exiting to form a new party, or migrating to another established party organization.60 In fact, even where there have been actual changes in the electoral system, different politicians have different perceptions of the costs they impose. In this sense, rules operate in a complex context. Rules’ causal effect should then be understood in light of different processes and junctures. This is what constituted the very definition of barriers to exit (Exit) in Chapter 2. Barriers to exit are also given by parties’ trajectory, present situation, and ability to resist shocks (durability).

I have included politicians´ consideration about the past. Given the weight of established institutions, I believe it is important to capture whether politicians perceive their parties and the party system as having a significant history. History and tradition gives strong weight to party organizations because reduces would-be independents’ intention to defect. It also generates advantages to remaining in the organization compared to personalistic adventures in terms of chances of success: party reputation comes into play, the organizational know-how and machines, among other assets. Thus, perceived barriers to exit are partly given and reinforced by partisan history. Politicians’ perceptions change in light of both shocks (e.g., economic crises) and experiences (e.g., success or failure while in government). I have therefore included prospective Exit as a component of Exit.61 This helps to account for differences both

60

For example Tavits (2008) introduces an explanation for the success of new parties entering a system. She combines the effects of institutional rules and existing parties’ reputations (given by years of democracy) as costs to entry and the benefits of gaining office as incentives to forming new political parties. 61 Perceptions on these components might be caused by many factors that are not the focus of this study. 85

between and within parties.62 Relatedly, I have surveyed parties’ opinions, under a hypothetical scenario (a harsh economic crisis), regarding the configuration of the party system. The aim was to locate politicians using hypothetical, drastic scenarios that would test high barriers to exit.63 This query proved to be a good complementary question for assessing perceived barriers to exit and, more specifically, prospective Exit. The question proved relevant because calculations of costs of exit are tied to expectations regarding possible future scenarios. Including those prospective and hypothetical scenarios, politicians’ spontaneous references to their parties’ history, as well as their interpretation of their respective parties’—and systems’—current situation helped construct a more complete picture of perceived costs of exiting. These perceived costs can only be fully analyzed through consideration of a combination of rules, processes, and junctures. To estimate politicians’ perceptions of barriers to exit, I thus considered their mentions of their parties’ history, and asked each interviewee a set of three questions: whether they regarded the party system’s configuration as stable, whether they thought its composition would remain the same, and, finally, whether the party or party system would be able to resist prospective shocks (i.e., severe economic crisis). These components together give a full picture of politicians’ assessments of the presence of barriers to exit.

The elements factored in for computing the crisp-set value for each politician on exit can be summarized as follows:

62

It is worth mentioning that I am not studying the causes for politicians´ perceptions of barriers to exit. Given that Costa Rica and Uruguay suffered from severe economic crises (Costa Rica between 1982 and 1983 and Uruguay in the early 2000s), I made references to that experience and the ability to the system to resist another such crisis. 63

86

A. Past significant history is mentioned64 B. Present stability is mentioned65 C. Prospective change of the composition of the system is perceived as impossible66 D. Ability to resist shocks (severe economic crisis) is perceived as positive.

Where all of these factors are present, perceived barriers to exit are clearly high. Thus,

As the logical relationship among components clearly shows, the rule presented considers the “weakest link”, i.e., the “minimum value of all components” (Goertz 2006, pp. 129-155): all components are regarded as necessary parts of politicians’ perceptions of high barriers to exit because each captures a different aspect of the perceived barriers to exit. Ambition: Ambition contributes to dynamic party survival. As argued in Chapter 2, my

idea of Ambition and its impact builds on previous studies highlighting the role of parties as organizations solving collective action problems for ambitious politicians. In this study, however, Ambition is treated as a variable rather than an axiom. The condition of Ambition as I define it is met insofar as a party fosters elite renovation.

64

As expected, given the countries analyzed here, this aspect of the perception of barriers to exit was the most common one among interviewees, though their perception of such a history was not unanimous. I did not ask them about past stability, however. 65 As for the previous component, I did not directly ask their opinion about this component. 66 When the politician says that his/her party will remain but the others will change (because of their fragility or for another reason), I have computed 1 on this component of the dimension: for the politician his/her party will be stable in the future and, unless he/she explicitly mentions otherwise, exit is thus not an option. 87

As stated by several scholars (e.g. Siavelis and Morgenstern 2008), some electoral rules facilitate satisfaction of politicians’ ambitions within parties, at least at the outset. For example, Uruguay and Costa Rica have rules allowing individuals to fulfill their ambition without contradicting partisan organization. The combination of PR, blocked and closed lists, and large average district magnitudes tend to give significant power to parties as organizations, and also ensure some degree of elite renewal. Under these conditions, effective partisan organization furthers satisfaction of individuals’ ambitions. Despite these differences between countries, written rules alone cannot tell us whether political parties actually promote individuals’ careers and function as channels for satisfying ambition. Nor can they explain differences between political parties operating under the same rules.

I considered Ambition present when politicians themselves considered that the party, qua organization, provided the necessary tools and mechanisms for satisfying their political goals. In essence, political parties solve collective action problems and satisfy individual ambitions of advancing a political career when there are known formal or informal rules and traditions that stimulate competition and elite renovation. When a party does so, its reproduction as a dynamic organization is more likely. To check for the presence of this condition, I looked at whether politicians thought their party promoted political careers.

The empirical evidence considered in order to determine whether a party promotes individuals’ ambitions consisted of a set of questions I asked every interviewee. I first asked them about their own career. Later in the interview I went back to their initial narratives and asked the following set of questions67:

67

This is a basic set of questions. Given that I conducted in-depth, semi-structured interviews I also asked other questions, depending on the country, the dialogue, and the interviewee in question. 88

Si tuviera que sintetizar aquello que explica su trayectoria personal en la interna de su partido y en la política, ¿a qué factores lo atribuiría? Hablando en general del proceso de crecimiento político, ¿cuáles son los factores que explican, en su criterio, las posibilidades de una persona de crecer políticamente? Es decir, ¿cómo se pasa de ser un militante anónimo, o sencillamente un ciudadano anónimo, a ser líder partidario? ¿Cuál es la trayectoria habitual de una persona que crece políticamente? En otras palabras, si Ud. hoy en día tuviera que iniciar nuevamente su carrera política, ¿cómo se imagina los primeros pasos? ¿Qué debería hacer para crecer políticamente? Más específicamente, pensando en su partido ¿cuáles son los mecanismos que existen hoy día en su partido para hacer una carrera política? ¿Considera que el partido favorece la competencia entre los líderes actuales y personas con potencial de liderazgo e interés de participar pero sin cargos formales en el partido? ¿Y su sector, favorece o no la competencia entre líderes y quienes quisieran serlo? ¿Qué factores explican, en su sector, la posición que ocupa en la lista un candidato determinado a la Cámara de Diputados? ¿Y cómo describiría dichos factores para el sistema de partidos en general?

These questions put the interviewee in different situations. On the one hand, I asked them to summarize the factors that explained their own political career. If they felt they owed their career mainly to family ties with established politicians, then I tentatively considered ambition not to be present. The second and third questions asked about general attributes that explaining politicians’ successful advancement in their careers. These questions aimed to see whether politicians would mention competitive channels or highlight that the party elite was looking for prospective leaders among its youth or militants. The probability that ambition was being satisfied for prospective politicians within the party is higher where politicians’ answers to these questions were positive. The fourth question located the individual in a hypothetical situation in which she had to start her career all over again. This question tried to survey whether the politician would be willing to go through the process again, whether she was able to identify the keys to succeeding in a context different from when she began her career, and if so, whether she thought it was still possible to be successful. These questions are mainly relevant for politicians who were middle-aged or old and who were mid-level or established. For 89

example, given the exceptional contexts in which many Latin American politicians initiated their careers, one needs to ask such questions to know whether those who fought authoritarian regimes believe that now, in normal times, it is still possible to pursue a political career within the party. The fifth and sixth questions aimed to determine whether politicians had a deep understanding of the actual process of promoting political careers and whether they believed the party fostered or hindered political competition and political careers. The last question—only applied in Costa Rica and Uruguay— sought to examine the criteria used to order candidates. Here I examined whether the process was competitive or whether it was arranged by leaders at their discretion and, in that case, which criteria the leaders used. I used the following rule to assess each party’s role as a channel for promoting individuals’ ambitions. If on every question the politicians’ answers were in line with the idea that the party was a channel for promoting a political career, then there is no doubt that Ambition is satisfied within the party according to that politician. In cases where not all answers were in line with the satisfaction of Ambition as defined in this study, I computed Ambition as present only if on 5 out of the 6 (or 7) questions the interviewee provided answers showing that the organization was able to channel individuals’ ambitions. This rule is valid because, for example, a politician’s career might be explained by his family ties (e.g., as the son of a founding leader) but he might recognize that, beyond his own trajectory, other politicians with different traits might pursue a career because there are known and available conditions to do so. Thus, it seems logical that Ambition could be considered present even in some cases where not all of the components considered were present.68

68

An interesting problem for future research would be to compare the characteristics of politicians who claim that Ambition is satisfied within their parties and those who actually advance their careers. Such a project could also analyze the individual differences between established and young or un-established politicians. 90

Trauma: Trauma helps explain party vibrancy because it forges or activates retrospective

loyalty. The relevance of this condition emerged inductively as a result of my fieldwork—as opposed to Ambition and Exit, which have decades of research supporting their relevance.

Some of the parties analyzed in this dissertation have suffered traumatic experiences. In Latin America, leftist parties, in particular, suffered repression and persecution before and during the wave of authoritarian regimes. Histories of repression are unfortunately widespread in the region. Chilean and Uruguayan leftist parties were among those targeted by authoritarian regimes. Repression and persecution of Allende supporters and their respective parties was paradigmatic in the case of Chile. Thousands went into exile, while others suffered harsh conditions. Although repression in Uruguay was milder than in Chile, it was equally traumatic and significant for those on the left. This experience was referred to frequently in my interviews. These experiences generated a mystic and proved a source of unification.

The memories of a shared harsh past or an event that put the organization in danger operates as a bonding mechanism and stimulates collective action via the creation of a community, reducing transaction costs. In the case of the parties I have analyzed in the Latin American context, it seems that the main effect of Trauma has been to inhibit the impact of conflictive divergence—of any nature. Trauma reduces infighting over substantive issues (redistribution, privatization, abortion); it helps to unite individuals with power ambitions who might otherwise be in conflict, and it reduces programmatic tensions, which are pervasive in the context of the challenges that developing democracies have to face. As mentioned in Chapter 2, I am not interested in the party or individuals’ use of their past; whether they politicize their traumatic experience or not would be a different subject in itself. For purposes of this study, it suffices to claim that traumatic past 91

experiences leave an imprint among partisans. Nevertheless, the effect of traumatic experiences on facilitating cohesiveness has a discount parameter, with the effect gradually fading as generations pass. Also, as the evidence will suggest, as time elapses it can produce a generational cleavage affecting Ambition.

To avoid potential biases, I did not ask any of my interviewees about potential traumatic past experiences. For example, I did not ask politicians from the Frente Amplio or from PPD or PS in Chile about their time in jail or how they thought it had affected their political parties.69 I computed Trauma as present whenever a politician made explicit references to that harsh past as defined on Chapter 2 or whenever secondary sources supported its undeniable role in that period. For example, in many interviews with leaders of the PPD and PS, the reference to the traumatic end of Allende´s government and the subsequent experience of torture and exile was pervasive. In interview after interview, politicians spontaneously narrated their personal experiences—and fights— with the authoritarian regime of Augusto Pinochet. They also narrated how those events—the fall of Allende and extended experiences with torture and exile—influenced their respective organizations. They also mentioned how it acted as an incentive to participate in politics, to be involved. Purpose: In contrast with Ambition, which refers to individual interest, Purpose is a

collective cause. Purpose activates prospective loyalty. It fosters parties’ vibrancy through the shared sense of a commitment to an idea, a project. As presented in Chapter 2, I consider Purpose a combination of coherence and intensity. When both attributes are present, a party has Purpose.

69

When politicians mentioned their experience in the fight against an authoritarian regime or during the years of transition to democracy, I did discuss it with them to better understand its role as a bonding mechanism. 92

I codified my interviewees’ answers on different questions70: their references to programmatic issues, and the degree to which there was a shared idea within the organization. For example, when I asked my interviewees the following question, they discussed issues related to Purpose:

En este país los partidos no son meras etiquetas ¿Cómo se articula el funcionamiento del partido tomando en cuenta la importancia de los individuos (líderes de sectores o de agrupaciones)?

However, discussions of issues related to Purpose also emerged in other sections of the interviews. As the question above illustrates, I decided not to ask politicians about Purpose directly because of the pervasive danger of receiving politically correct answers.

Whenever the politician mentioned programmatic agreements—whether in terms of specific issues or abstract ideological positions—as a source of bonding, I considered coherence to be present. When they made references to the emphasis and role of these ideas on their political actions and decisions, I computed intensity to be present. It is worth mentioning that Purpose was considered absent when: a) the politician explicitly mentioned problems in terms of coherence or intensity within the party or; b) when the politician did not make any direct reference to anything related to Purpose.71

70

Benoit and Laver (2006) provide an interesting review of different ways to measure political elites´ policy positions. Although I agree with the convenience of expert instead of elite surveys, both my approach, which explicitly contrasts internal accounts of different groups more than the specific programmatic inclinations of the party, and the qualitative nature of my conclusions avoid the problems they raise about using surveys of politicians to collect this data. 71 Such an omission occurred in only 13 of 148 interviews and was not concentrated in any specific party or country. 93

3.2. From Unit of Observation to Unit of Analysis

I have decided to compute the values on all my conditions based on the evidence collected at the individual (politician) level, my unit of observation. In the cases of Ambition and Exit, this decision was unproblematic: both conditions refer to individuals’ decisions and perceptions about parties’ operation. However, this approach is also valid for Purpose and Trauma. In the case of Purpose, campaign platforms or manifestos can give a misleading impression of how politicians perceive their party. Because they can only reflect certain leaders’ thinking, they also risk obscuring internal divergences. Looking at congressional voting to capture a party’s commitment to a program furthermore involves taking into consideration variables that do not pertain to the party qua organization. In the case of Trauma, considering individuals’ evaluations is valid because, for some politicians, past experiences might not forge bonds with copartisans.

I will now detail the rules that guided the codification of each condition at the level of the party, my unit of analysis.

My codification was guided by the following simple premise: the greater the degree of agreement among politicians of party x concerning the presence of a condition, the greater the overall presence of that condition in party x.

I thus relied on inter-subjective agreement to determine the presence or absence of each condition. The risk of reporting revealed preferences was diminished substantially by the fact that I had interviewed different tendencies and fractions as well as different generations of politicians within each party.

Given my use of crisp-set values, and unanimity on the presence or the absence of one condition was the exception, I applied these rules for resolving disagreements: 94



If more than half of the interviewees agreed on the presence of a condition, I computed it as present.

However, given that I am working with qualitative evidence and that in-depth analysis is an asset of this study, to ascertain the presence or absence of a given condition I used more than the mere rate of agreement. More specifically, in situations where the presence or absence of a condition was close to the threshold, I also considered the following, complementary rules: 

Where there were substantive disagreements, I asked: what is the main cleavage explaining the disagreement within the party? If established politicians systematically agreed on the presence or absence of Trauma and Purpose, that condition was said to be present or absent accordingly. Given that this chapter provides a static account of constellations of causal conditions, it seemed logical to give more weight to established and important leaders’ accounts because they are the ones primarily responsible for the imprint made by a party through its main political actions, decisions, and electoral speeches and manifestos. 

Unlike Purpose and Trauma, Ambition and Exit are both associated with individual perspectives. For those conditions, it was therefore relevant to know whether younger and less established politicians felt their party was able to promote their careers, and whether they believed one could defect or that the composition of the system—including their party—would change. The qualitative evidence was therefore analyzed to see whether there was divergence in such perceptions reflecting a generational cleavage. If so, the condition was said to be absent.

95



If no relevant cleavage (e.g., established versus non-established, or young versus old) explained why some politicians thought a given condition was present, then I considered that the condition was present.

Table 3.1 summarizes the evidence regarding the presence or absence of each condition by political party. Gray shading on a cell indicates closeness to threshold indicated above. For those cases the imputation of presence or absence of the condition follows the complementary rules introduced above.

96

Table 3.1 Summary of Each Condition by Party Party

Trauma

Purpose

Exit

Ambition

UDI

1 (5/9)

1 (9/9)

1 (5/9)

1 (9/9)

RN

0 (0/8)

0 (0/8)

0 (3/8)

0 (4/8)

DC

1 (5/8)

0 (2/8)

1 (5/8)

0 (3/8)

PPD

1 (5/8)

0 (1/8)

0 (3/8)

0 (2/8)

PS

1 (9/11)

0 (4/11)

0 (4/11)

0 (5/11)

FA

1 (14/20)

1 (14/20)

1 (17/20)

0 (9/20)

PC

0 (0/10)

0 (1/10)

1 (10/10)

1 (9/10)

PN

0 (0/13)

0 (4/13)

1 (10/13)

1 (12/13)

PAC

0 (0/12)

0 (5/12)

0 (1/12)

1 (9/12)

PLN

0 (5/17)

0 (5/17)

1 (10/17)

1 (15/17)

PUSC

0 (4/12)

0 (5/12)

0 (2/12)

0 (3/12)

ML

0 (0/4)

1 (3/4)

0 (0/4)

0 (1/4)

* Note: In parentheses, I show the rate of agreement on the presence of each condition. The sum of responses for the three countries is less than my total number of interviews because I also interviewed some politicians and former leaders from parties not studied in depth here and non-leaders but close to a specific party.

3.3. Observing the Outcome

As discussed in the Introduction, party survival has two components. The first component, age, is straightforward. The age of some of the parties is disputable, but I was not interested in the number of years per se but rather in determining the historical events the party had undergone. Age was thus relevant as a proxy for 97

“institutionalization”. All my cases met with the minimum conditions, as established in the Introduction, for qualifying as institutionalized. The second component, vibrancy, was more complicated to determine empirically. In the Introduction, I advanced a series of observable indicators of this variable: the extent to which militants, prospective leaders, or citizens were involved in different types of partisan activities, the intensity and frequency of different types of meetings, the party’s public appearances (of any form), and the territorial or functional penetration of the organization. I have combined available evidence on these points with interviewees’ considerations about the functioning of the organization. Again, since different tendencies and perspectives within each party were considered, I am confident that I was able to capture a balanced picture. The basic set of questions I asked my interviewees was the following:

Pensando ahora en el funcionamiento interno de su partido, ¿Cuáles son los organismos internos que funcionan con mayor asiduidad? ¿Quiénes participan de dichos organismos? ¿Cómo se articula el funcionamiento del partido tomando en cuenta la importancia de los individuos (líderes de sectores o de agrupaciones? En general, ¿considera que la organización partidaria funciona de manera frecuente? ¿Cómo describiría la relación entre los líderes partidarios y el resto de los integrantes del partido? Y específicamente, ¿cómo funciona la relación con los militantes del partido? (Después que respondió preguntar) ¿Tiene su partido una relación más o menos fluida con los militantes que la que poseen otros partidos con sus respectivos militantes? ¿Existen instancias de participación de base/militancia que funcionen regularmente, más allá de las elecciones nacionales o municipales?

The formula used to compute a politician’s perspective on his or her party’s vibrancy was the following:

98

A. The organization works and functions permanently B. It has a fluid relationship with militants, C. There are permanent instances for militants and adherents participate in the organization.

Some precisions to this basic rule should be mentioned. The first component (the organization works and functions permanently) encompasses the first three questions. There I am evaluating whether there is actually an organization, rather than a personalistic machine or sum of individuals merely using the same label. The second component is not considered present if the relationship depends on a specific leader and not on the party, or at least a sector or faction of it. I did not consider the third component (existence of permanent instances for militants or adherents’ participation) present when politicians claimed that those instances existed only if party leaders convened the meeting. Thus, I have only considered such participatory instances present if there were institutionalized meetings operating beyond leaders’ initiatives or, when informal, they were convened by the base or local organizations within the party structure.

The aggregation from the unit of observation (politician) to the unit of analysis (party) follows the same rules applied to each causal condition.72 Table 3.2 below summarizes the evidence on both components of party survival: age and vibrancy. The column “strict vibrancy” changes the relationship among components of vibrancy. In the strict vibrancy operationalization column, I use the logical AND (resulting in

), a

more demanding operationalization, as the connector between “fluid relationship with militants” and “permanent instances for militants or members’ participation.” This is 72

Here again the threshold could be regarded as too permissive. However, the least agreement on vibrancy was for the PN in Uruguay, with 60% agreement; the other vibrant parties showed an even greater degree of agreement, whereas the static cases were all far below the threshold. 99

relevant because, in all three cases, it has consequences for the determination of the outcome, signaling different degrees of membership in the set of dynamic parties. This will be thoroughly analyzed when discussing the degrees of membership in the set of dynamic parties. Finally, the numbers between parentheses indicate the proportion of interviewees that think their parties have the components defining vibrancy.

100

Table 3.2. Computing the Dependent Variable, Types of Survival73 Party Age Vibrancy Strict Vibrancy Static/Dynamic UDI

1

1 (9/9)

1 (8/9)

Dynamic

RN

1

0 (1/8)

0 (1/8)

Static

DC

1

0 (2/8)

0 (0/8)

Static

PPD

1

0 (0/8)

0 (0/8)

Static

PS

1

0 (2/11)

0 (1/11)

Static

FA

1

1 (17/20)

1 (12/20)

Dynamic

PC

1

1 (8/10)

0 (4/10)

Dynamic /Static

PN

1

1 (8/13)

0 (6/13)

Dynamic/Static

PAC

1

0* (4/12)

0 (2/12)

Static

PLN

1

1 (12/17)

0 (6/17)

Dynamic/Static

PUSC

1

0 (1/12)

0 (1/12)

Static

ML

1

0* (1/4)

0 (1/4)

Static

* The PAC and ML should be analyzed with extreme care. Even though they both surpass the basic threshold—i.e., they have both participated with the same label in more than three national elections—it is not yet true that they have outlived their founding leaders (Ottón Solís in the case of the PAC and Otto Guevara in that of the ML).

73

It could come as a surprise that I have not included two parties from Chile: the Social Democrat Radical Party (PRSD) and the Communist Party in Chile (PCCh). Both parties are very old and have a significant presence in the system´s history. For the PRSD, I conducted two interviews but declined to deepen the analysis because of the party’s minor role since transition to democracy in 1989. The PCCh failed to meet one of my selection criteria at the time of my field research because it did not have members elected to Congress from the transition until 2009. 101

Let me illustrate the process with one clear example. The dialogue with one established, older leader from the PLN in Costa Rica showed that, in her perception, the party was vibrant: ¿Cómo describiría la relación entre los líderes del partido y los que no son líderes, la militancia? Buena. A ver. Se la pregunto, le voy a explicar por dónde va la pregunta, porque buena no me sirve. No, claro, quiero ir un poco más allá. Hay relaciones, por ejemplo, en algunos partidos en el mundo que la relación entre la militancia y los líderes es servil. Líder es líder. No, en Liberación no. Bien. En otros partidos… ¿vio lo que pasó con el Tea Party y los republicanos? Están los militantes que casi se podría decir que odian a sus líderes. Esos vendrían a ser los dos extremos del tipo de relación. ¿Cómo ubicaría la relación líderes – militantes en este continuo imaginario? Una relación de respeto. Una relación dura, a veces, porque el militante… ellos son los dueños del partido. Se sienten así. Sí. “Nosotros somos el partido”, entonces, al dirigente lo señalan cuando no les parece. Cuando no les parece. ¿Y qué quiere decir “cuando no les parece”? ¿Cuándo los traiciona ideológicamente, cuando no les lleva las cosas prometidas? Las cosas prometidas…Cuando en el gobierno no se les cumple. Cuando, también, en la estructura del partido vienen las elecciones y no está bien el tribunal de elecciones internas del partido. Le falló el día de la convocatoria, no los convocaron. Son capaces de venirte a tocarte la puerta y decirte, “yo soy del partido y a usted qué le pasa. Para eso la puse ahí.” Así son.

This quote illustrates the party’s close relationship with its militants and their central role within the party. Militants are proactive in their relationship with the party elite. Thus, this leader perceives the party as vibrant. Conclusion

This chapter has focused on the empirical operation of my four causal conditions, which are theorized as explaining types of party survival in Latin American consolidated democracies. I have operationalized each condition and explained how presence of a condition within my units of observation (politicians) and units of analysis, cases

102

(parties) was computed. The next chapters will analyze my cases on a country-bycountry basis building on the procedures here introduced.

103

Chapter 4. Chile Introduction

This chapter empirically assesses the presence of my causal conditions in Chile’s principal political parties. Using evidence from my fieldwork, the first section presents systemic traits and politicians’ agreements on different intersections of conditions. I then analyze each case (party) in Chile. Building upon the evidence collected in my fieldwork, I analyze the configuration observed within each party. Using secondary sources, I briefly sketch the historical evolution of the causal conditions and how they have interacted with significant events to bring about the present configurations.

Chile has been regarded as the epitome of success in democratic governance. It has also been held up as a positive model for developing countries. It has been argued that the stability of its parties and programmatic structure of competition in the system as compared to others in Latin America explain part of its success. However, Chilean society has grown increasingly distant from its parties, as evinced by low levels of trust and party identification. Chile is therefore one of the best cases one can examine to analyze the theoretical puzzle presented in the introduction to this study.

104

4.1. General Overview Table 4.1. Chile74 Exit

Trauma

Purpose

Ambition

0

0

0

0

0

0

0

1

Units of Observation RN (Y, M, 3), DC (Y, E, 2), Ex DC (M, 1), PS (Y, 1), PPD (Y, M, 3), PRSD (E, 1), Ex PS (Y, 1) RN (M, E, 2)

1

0

0

0

Ex Concertación (M, 1), Ex PPD (Y, 1), RN (Y, 1)

0

0

1

0

0

0

1

0

0

Ex PS (E, 1), Ex PPD (M, 1), PS (E, 3)

1

0

0

1

RN (Y, E, 2), DC (M, 1)

0

0

1

1

UDI (Y, 2)

0

1

0

1

PS (M, E, 2), PRSD (M, 1), PPD (E, 1)

1

0

1

0

0

1

1

0

0

PPD (M, E, 2), PS (E, M, 2), DC (Y, 2)

1

0

1

0

0

0

1

1

0

PS (E, 1)

1

0

1

1

UDI (Y, M, 2), PS (Y, 1)

0

1

1

1

UDI (M, E, 2), PS (M, 1), DC (E, 1)

1

1

1

0

DC (M, 1)

1

1

0

1

DC (E, 1)

1

1

1

1

PPD (M, 1), PS (M, 1), UDI (E 3)

The intersection of conditions etpa, i.e., the absence of all my causal conditions, is the most populated of all rows and is where leaders from PPD and RN are found, though some former established politicians, mostly PS and DC, who had defected or abandoned

74

In the truth tables for each country, I have not computed the observed outcome by row because inference is not derived from each politician as an individual but as member of a political organization. Finally, to preserve confidentiality, I have included just the following information for each politician: party affiliation; whether the politician was established and renowned (E); whether she was a mid-level leader (M), and if she was a young leader (Y). These are the most relevant attributes given the goals of this study. The scant number of interviewees in each case, precluded including other theoretically relevant traits of each interviewee. Nevertheless, I do also refer to other relevant traits in the course of my analysis where it would not distinguish an individual from a group of others within the sample and thus compromise the condition of anonymity. 105

politics are also located there.75 Politicians located in this row had the harshest accounts of their respective parties and the party system. They felt their parties or party system as a whole did not channel representation, were not leading actors of Chilean democracy, and would probably change significantly in the near future. These impressions were in line with very recent interpretations of the Chilean party system (Altman and Luna 2011, Luna and Rosenblatt 2012).

Politicians located in the first three rows (etpa, etpA, Etpa) were predominantly from RN. However, these rows also include politicians from different generations belonging to different parties of the Concertación coalition and some who defected to new parties or abandoned politics. They are, in addition, very critical of the current situation in their parties (and the system as a whole). What is interesting about the groups of politicians located in these first three rows and their assessments of conditions indicative of their respective parties’ vibrancy, is that they show that the presence of A or E alone is insufficient for producing perspectives different from those where no causal conditions were deemed present. Thus, like those in row 1, those in rows 2-3 were blunt in their assessments of their parties, with none describing their party as having conditions indicative of vibrancy. In fact, analysis of the effect of the binominal electoral system in the Chilean party system, and the discussions of candidate nominations similarly show their detrimental effect on parties with respect to Ambition (Londregan 2000, Navia 2003, 2004, 2008, Siavelis 1999, Siavelis 2000b, Siavelis 2000a, Luna forthcoming).76 75

However, one also finds here one young leader from the PS and a leader from the PRSD, which is not studied here as a case. 76 In the 1925 Constitution, Chile adopted a D´Hondt electoral formula, PR, and open-list and semiconcurrent legislative and executive elections. epresentatives’ terms were four years, Senators’ eight, and Presidents’ six. Although the open-list rule could be said to have inhibited partisan organization, the use of the D´Hondt formula solved strategic coordination problems. Parties could present lists of candidates up to the size of the district magnitude. The formula thereby stimulated cooperation of lesser known, younger candidates with established leaders, facilitating partisan organization (Gamboa and Morales typescript). The commonly known binominal electoral system not only affected small parties. It also reduced incentives for strategic coordination between candidates from the same list (see Carey and Siavelis 2003, Navia 2003, Siavelis 1999, Siavelis 2000b, Siavelis 2000a) 106

The leaders in the eTpa row are all older, established, leaders from the PS or PPD who played key roles during the transition to democracy. Interestingly, those located in this row agreed that their parties were not vibrant. Where Ambition was also present (eTpA), only some established leaders from PPD and DC were present. These major leaders thus still had memories from a difficult past during the authoritarian regime, memories that activate retrospective loyalty. The presence of Trauma is marked by a generational cleavage. It played a significant role in the 1990s both in terms of organizations and also in terms of policy-making. These leaders also thought their parties were able to promote politicians careers. However, as will be seen in the analysis by party, the perception that their parties satisfied individuals’ ambitions was not shared by younger prospective leaders—or even by all established ones. Finally, they agreed that their parties were not vibrant.

The ETpa row contains some older, established politicians, as well as some middle-aged politicians with deep knowledge of their parties’ structure. Like those in the eTpa row, they recognized that their parties did not satisfy ambitions. They also agreed on the presence of T and its operation as a unifying factor. However, because of the heroic epic, T also prevented young leaders from defying the elite, accounting for the absence of Ambition. As in the previous cases, there was almost unanimous agreement among politicians in this row that their parties were not vibrant.

In rows where only 0-1 causal conditions were present, there were no members of the UDI. At the same time, as we move down the table, rows where more than two conditions are present are mostly populated by UDI leaders. The general picture indicates that the UDI was the party within which there was more agreement on the presence of the conditions identified as conducive to dynamic party survival. In other parties, it was at most the older and more established leaders (mostly from the DC and PS) who felt that their parties were and would remain stable, promoted individuals’ ambitions, or had a clear and intense Purpose. 107

The Chilean case highlights the value of looking beyond just the hypothesized effect of electoral rules in assessing barriers to exit. Chilean politicians systematically referred to the binomial electoral system. However, politicians’ accounts of how and why it mattered differed notably, not only across parties, but even within them. Again, rules in and of themselves do not fully capture barriers to exit, even when evaluating “present” situations. Politicians regarded electoral rules as the main reason for the closed quality of the contemporary Chilean party system. Some saw the electoral rules as keeping the system artificially and excessively stable and rigid.

In this sense, some politicians regarded the Chilean rules as having promoted stability in the past, as one would expect given the restrictive electoral system. However, they also mentioned that it was becoming increasingly less costly for politicians to exit, regardless of changes in rules. They felt there were growing chances for politicians to exit and pursue a political career as independents. The “discolo”77 phenomenon is an illustration of this sort.

Thus, the evidence suggests that the passage of time—during which substantive changes have occurred in Chilean society—has changed the perceived costs imposed by electoral rules. Moreover, their supposed prospective stabilizing effect—the main consideration of those promoting the system—has diminished. Only a more complex perspective on barriers to exit allows one to capture their curvilinear effect: the longer restrictive rules endure (and the more restrictive they are), the less they promote partisan organization.

77

This term is used to refer to a politician who does not follow the party´s line, votes differently in congress, presents himself as critical of the system, and eventually ends up exiting from his or her party. It originated as a term applied to the “Colorines” faction within the DC. Adolfo Zaldívar, a former DC Senator, was expelled from the party because he refused to vote in favor of a bill that sought to inject more resources into the transportation system in Santiago, Transantiago. In turn, a group of DC representatives followed him and exited the party. These politicians from the DC became known as the díscolos. 108

Finally, in terms of Purpose, the graph depicts the location of parties along the two dimensions defined by the attributes identified in Chapter 2. The only party in the “Purpose Quadrant” is the UDI. This finding will be extensively analyzed in the chapter.

Intensity 1

5

4 3 2

Coherence Key: 1: UDI 2: RN 3: DC 4: PPD 5: PS The length of the line signals degrees of agreement on each component of Purpose. The lengthier the line, the lesser the agreement.

4.2. Configuration within each Party

PPD: Although some leaders are located in the row in which all conditions are absent,

others’ responses indicated that the party does provide a viable channel for individuals’ ambitions or that trauma was present. However, the responses of more than half the interviewees indicated the presence only of the latter. More than half of the politicians made references to the authoritarian regime and its impact on the party. This pattern was expected because the party was actually an invention of Socialist leaders who wanted to campaign for the No in 1988 (Garretón 1995, Roberts 1998, Luna forthcoming). Though the initial inspiration was to create a party coalescing all opposition forces, this 109

institutionalization of the opposition did not prosper (interview with PS leader). Since the PS was banned, they had to register a new party to participate in the transition and, in its origin, the PPD only acted as a pragmatic solution to overcome that prohibition. This process was narrated as follows by one of the party’s founders:

Entonces yo propuse hacer un solo partido, Partido por la Democracia. Yo siempre he dicho que el PPD es el resultado del default. Porque la DC dijo “perdón, yo voy a inscribir”. “Yo soy socialista, la constitución me lo prohíbe ¿Cómo te vas a inscribir y yo no? Quedo afuera. No, no, acá un solo partido.”Pero se inscribieron. Si se inscribe la DC, se inscribieron los radicales, si se inscriben los radicales se inscribe otro partido, los social-demócratas. Entonces, yo me quedé con la idea del Partido por la Democracia y al final el Partido por la Democracia fue la decisión del PS. El PS lo presidía Ricardo Núñez… A esas alturas, como había estado preso y todo eso, ya era más o menos conocido a esas alturas. Me acuerdo que llegamos a finales del 87 sin decisión y yo dije “estamos jodidos”. “El plebiscito, les apuesto que va a ser el 11 de setiembre del 88”, decía yo. Por lo tanto, sólo los partidos inscriptos 4 meses antes tienen derecho a designar apoderados…Por lo tanto, el partido tiene que estar hecho en marzo. Por lo tanto tengo que empezar a inscribir, como última fecha, en diciembre. El PS dio el ok a hacer esta cuestión y yo llamé a todos. Vino una fracción disidente del Partido Radical, vino una fracción disidente del Partido Comunista…la derecha republicana de Armando Jaramillo, yo fui presidente de eso. Empezamos entonces la lucha por inscribir el partido. Había que reunir 35000 firmas a lo largo del país, tú sabes la ley electoral, no te la explico…llevábamos un mes entero y estábamos jodidos. Pleno verano, enero. Entonces, en la desesperación, los cabros de la juventud decidieron sacar el escritorio para la calle. Teníamos una oficinita de abogados ahí cerca del Paseo Ahumada. Entonces, trasladaron a pulso, por la calle el escritorio y (lo pusieron) en el Paseo Ahumada “A firmar aquí, a firmar aquí, crear el Partido por la Democracia y echamos a Pinochet”. Yo estaba haciendo una gira por La Serena cuando me llaman en la noche y me dicen “hemos juntado 400 firmas”. ¡Puta! Y al día siguiente, 700 firmas y “además, ahora, hay unas discusiones en la calle, ¡los que están en contra y los que están a favor! La gente firma y hay un rollo tremendo en la calle.”...ya está, inscribimos el partido. Tercer día, ¿qué pasó? “Nos jodieron”. “¿Cómo nos jodieron?” “Dicen que hay que tener un permiso municipal”. “Bueno, ¡exija! Somos un partido en formación tenemos derecho.” “Sí, pero no nos dan el permiso en la municipalidad porque dicen que obstaculizamos el desplazamiento con el escritorio.” Eso es divertido. Me acordé, en ese tiempo…vendían cigarrillos en una bandeja como en las películas. Entonces mandé a hacer bandejas y salieron los cabros y chiquillas, a las calles con bandejas y la gente firmaba en las bandejas…

Thus, an interesting combination of T and A was present at the party’s origin. As illustrated in the quote above, at its inception, the PPD drew leaders from diverse origins, and many “renovated leftists” were also originally part of the party (Garretón 1988 and Luna forthcoming). The PPD was not ideologically unified, thus P was not present. It was a pragmatic solution devised to circumvent the banning of the Socialist

110

party in 1989. In fact, some of the PPD’s founders later returned to the PS.78 Some, though, decided to remain in the PPD.

The party tried to forge a separate identity combining liberal and progressive principles. However, the organization was unable to develop a clear and intense Purpose.79 The pragmatic origin, successive inclusion of other leaders from very different backgrounds, and defection of many socialists left the party’s Purpose unresolved. Interestingly though, according to my interviewees, those who have given the party a clear progressive and liberal stance are the ones who have developed a stronger network, and it is in areas where this has occurred that the party operates regularly and is thus more vibrant.

The following excerpt from an interview with a PPD leader not only shows the situation of Purpose within the party but also illustrates the close relationship between Purpose, partisan organization, and vibrancy. Although this connection was not always explicitly mentioned in my interviews, in this instance it is.

78

One can see this by comparing the party affiliations of congress members in 1990 and 1994. See Luna, forthcoming for a thorough analysis of programmatic cleavages at the party-system level in Chile. 79

111

R:…A ver, el PPD es una intellechia –intellechia no es la palabra–, el PPD no tiene lo que tuvo hace 15 años o al principio de la democracia que era algo que sumaba. Gente bien distinta, de distintos lugares y se sentaron. Gente bien de izquierda y otros que eran de izquierda progresista pero más liberales…después se legalizó el PS y se nos fue la mitad de la gente, era natural. Pero lo que quiero decir es que ha perdido esa capacidad de amplitud. Era convocante. P: ¿Por qué la perdió? ¿Por los líderes o por escasa organización? R: Bueno, si tienes una institucionalidad mala tienes líderes que aprovechan esa institucionalidad mala. Pero yo creo que tiene que ver con liderazgos que fueron personalistas.” P: ¿Cómo describiría la relación entre los líderes partidarios y el resto de los integrantes del partido? Y específicamente, ¿cómo funciona la relación con las bases sociales/territoriales del partido? Me habías dado una pista relacionada al clientelismo… R: No, el clientelismo es brutal poh huevón.” P: ¿Qué significa eso de “clientelismo brutal”? R: Significa que soy sub-secretario de salud y te consigo la pega en un consultorio en no sé dónde a cambio de tu lealtad. No hay ni que explicitarlo y se colocan a operadores en eso y eso yo creo que perjudicó. P: ¿Y ese es el tipo de relación que tienen con los no-líderes del PPD? R: No, porque también hay militantes que todavía –varios–…todavía hay gente que cree que es posible hacer algo distinto, un país mejor, estoy hablando de militantes que se juntan un día jueves de 7 a 10 de la noche; o sea, no es un huevón de la calle, no es el estudiante, es un tipo que tiene una conciencia política, que no es dramático, pero que todavía cree en algo… P: ¿Podemos sacar la siguiente hipótesis: allí donde el PPD tiene este tipo de militantes es allí donde el PPD tiene una organización más sólida? R: Sí, es correcto. P: Y allí donde tenés una relación más clientelar… R: Allí no funciona para nada. Pero incluso, yo diría que en aquellos lugares donde no hay esa relación tan clientelar no siempre las cosas van a funcionar mucho mejor. Por tanto, no hay un liderazgo que las organice. El que aparezca una persona, un diputado…tú quieres alguien que esté encima en el sentido que ayude a canalizar, a algutinar, porque también intereses. Los partidos son máquinas de interés. P: Nuevamente entonces, militantes con ideales y valores es una condición necesaria pero necesitarás un liderazgo. R: Necesitas uno o varios líderes, sí. P: ¿Existen instancias de participación de base/militancia que funcionen regularmente, más allá de las elecciones nacionales o municipales? R: Está funcionando cada vez más como maquinaria electoral…No hablo con banderas del PPD atrás. 112 “El PPD piensa”, no…nunca es PPD. Yo mismo me doy cuenta.

Regarding Ambition, politicians described a situation in which the party has provided the label but each politician has been individually responsible for its success. This arrangement is common to most parties in Chile. In the PPD, however, the party is strategically unable to determine the candidates. It is not the party that promotes new leaders through purposive actions; it is individuals from the elite of the party, at most, that hinder or can promote the rise of new PPD politicians. Consequently, those who have been involved in politics since the transition to democracy in 1989 have a significant advantage whereas younger leaders from the party are incapable of making progress in pursuing a political career. Thus, there are significant transaction costs for those trying to pursue a political career in the PPD. Although new leaders emerge, they are not always militants or younger leaders who have followed a known path to political promotion. The new leaders who do emerge are promoted by individual leaders based on such considerations as individuals’ public fame (interviews with PPD leaders). It is not an organization with accepted formal or informal channels of elite promotion and renovation. The result has been a party organization in which Ambition is not organizationally satisfied.

Creo que hoy día los caminos tradicionales siguen siendo una posibilidad, ser dirigente estudiantil arrancar desde ahí. Pero creo que hoy día es muy posible hacer una carrera política vinculándote más a causas puntuales que adquieren notoriedad, eso le da visibilidad a las personas que están vinculadas a ellas. Actores ciudadanos muy puntuales hoy día es muy posible que hagan una carrera política desde ahí. Ni siquiera hoy los partidos intentan cooptar a nadie, tan sólo le ofrecen apoyarlo para que sean candidatos. Muchas de esas experiencias resultan bien complejas porque son personas que no tienen una gran trayectoria en el compromiso partidario. Tienen votos pero después es complejo meterlos en una dinámica colectiva.

Thus, following the rule presented above, according to which agreement between politicians determines absence or presence of a causal condition, the intersection of causal conditions for the PPD is: eTpa. RN: The party was founded during the authoritarian regime of Augusto Pinochet. In

1987, in light of the Ley Orgánica Constitucional de Partidos Políticos (March, 1987), 113

different organizations (Movimiento Gremial and its political expression, the UDI; Frente del Trabajo; and Unión Nacional) united under this label. 80 After the UDI was expelled from RN (see below), the RN continued as the more centrist party of the Chilean right. Its younger leaders—most prominently Andrés Allamand and Sebastián Piñera—tried to forge a renovated identity on the right based on liberal ideas (Scully 1995). From the outset, however, the party was divided. During the transition to democracy, a significant portion of RN leaders were not convinced about the return to democracy (Scully 1992, 1995). A democratic-authoritarian cleavage was present in the early days of the restored democracy. Though some leaders of RN encouraged recognition of the loss of the October, 5, 1988 plebiscite and forged political agreements with the Concertación in preparation for the July, 30, 1989 plebiscite facilitating democratic elections, others were supporters of Pinochet’s leadership and legacy (Garretón 1995, 2000b, pp. 61-2). 80

Siavelis (1997 and 2000) suggests that the right has a history of unstable partisan organization. However, I agree with Garretón, who claims that “…organizational expressions [of the political right] have, moreover, been both clearly identifiable and stable over time” (2000, p.53). The author does claim, however, that “[t]he ight´s civil society is much stronger than its party expression, and its social universe is poorly articulated in its partisan organizations. Much more important in this regard are conservative forces´ autonomous economic base (entrepreneurs are genetically rightist), private primary and secondary schools and universities, and communications media.” (Garretón 2000, p. 55). The Conservative Party was a relevant and constant political force, governing the country or exerting a strong influence while in opposition from the nineteenth century until the 1950s (Gil 1966, Scully 1992). In democratic elections, the Conservative Party supported the successful presidential candidacy of Jorge Alessandri in 1958. The party was mainly composed of members from the upper-class, and was close to the Church´s doctrine ideologically. The Conservative Party proved a vibrant organization in times of restricted electoral participation (see Gil 1966, Scully 1992, Navia 2004). In 1966, the Liberal Party and the Conservative Party united and formed the National Party. Describing the National Party´s ideas, Garretón says: “[t]hese elements of anti-system criticism—nationalist in tone and more authoritarian in substance because they blamed economic stagnation on populist policies and Chile´s form of democracy—established the basis for the formulation of, for the first time since the 1930s, a more aggressive right-wing political program” (2000, p. 57). The National Party strongly opposed Allende´s government, and was voluntarily dissolved right after the 1973 coup (Valenzuela 1978, Scully 1992). The RN was in various respects continuous with previous rightist experiences: its liberalism can be traced back to the traditional Liberal Party, while its conservative stances have resembled those of the Conservative Party. Some of its leaders have, in fact, had family ties with both Liberals and Conservative Party leaders. 114

Tensions in terms of the party’s relationship with the authoritarian regime remained during the 1990s and are illustrated, for example, by Sergio Onofre Jarpa´s decision to exit the party in 1997.81 Thus, from its origin and up to the present—though now for different reasons more associated with a moral cleavage (Luna forthcoming)—the party is strongly divided internally in terms of Purpose.

In this sense, the evidence collected in my fieldwork shows that no politician from RN thought his party had P. Regardless of their ideological position there is an important agreement with respect to the party´s inability to promote an intense Purpose given the lack of agreement on major issues. Then, there is a significant division in two major areas: on the one hand, its stance on the democracy-authoritarian cleavage that has been pervasive in Chilean politics since 1989—though with fading relevance among the public. Some within the party were close to the Pinochet authoritarian regime while others—including President Piñera—voted No in the crucial 1988 referendum. Also, there is a conservative-liberal cleavage on moral issues, which are gaining substantive attention in Chile (Luna forthcoming).

Me gusta que RN es un partido poco ideológico. No es un partido donde las definiciones ideológicas sean tan tajantes y tan importantes. Yo concibo que la política tiene que divorciarse un poco de la ideología. Aquí hay poca ideología y esa es una fortaleza. Una debilidad, a la inversa, es que no tenemos un mínimo común denominador doctrinal, cosa distinta. No tenemos un mínimo común denominador doctrinal lo suficientemente fuerte. Entonces lo que es una virtud, por déficit, pasa a ser un defecto.

Politicians from this party are more uniformly distributed than those from the PPD. Some are located in the etpa row and others in the row for etpA. There is one in the EtpA row, who thus shares A with the ones in etpA. Some within the party feel that it is possible to pursue a political career, and not all are established leaders. However, such views within the RN do not surpass the threshold for considering Ambition present. In 81

Jarpa, an established leader from the National Party, was appointed Interior Minister by Pinochet in 1983 and was RN´s first president. 115

fact, those who regarded it as absent were emphatic regarding the party’s inability to recruit and promote new leaders, stressing that such promotion instead depended upon individual leaders themselves. In terms of barriers to exit, there is substantial agreement that it is increasingly possible to defect. In fact, looking at the list of candidates for the House of Representatives from 1989 until 2009, RN is the party that suffered most defections. Thus, the configuration of this party is etpa.

In conclusion, in Chile, both the party founded a few years before the democratic transition (RN) and the one founded right during that period (PPD) show an absence of Purpose. There is no perceived prospective loyalty activated by Purpose. Also, politicians from these parties, especially those from RN, did not think there were high barriers to exit, and most believed that the system was likely to change in the foreseeable future.82 DC: In August 1935, a group of young leaders founded the National Movement of the

Conservative Youth.83 Scully (1992) mentions that their main goal was political expression of the Church’s answer to the “social question.”84 In this sense, he says: “[t]he leaders of the Conservative Youth were determined, in an almost messianic

82

I analyzed prospective exit separately. These young leaders’ origin can be traced back to their participation in the Asociación de Estudios Católicos (ANEC), in which they learned the social doctrine of the Church during the Ibáñez del Campo dictatorship (1927-1931). They thus began their political activities during a period of severe repression that clearly marked their political socialization (Grayson 1968). Both Eduardo Frei Montalva, later founder of the DC and President of Chile, and Radomiro Tomic—the DC candidate who competed against Allende in the 1970 presidential election—participated in ANEC, as did Bernardo Leighton, and Manuel Garretón. Father Oscar Larson and Conservative Party leader Rafael Gumucio deeply influenced the ideological formation of these future leaders (Grayson 1968, Grayson 1969). See Grayson (1968) for a thorough description of the incorporation of these young leaders into the Conservative Party and their contribution to the party’s vibrancy. 84 The encyclical Quadragesimo Anno of 1931 stimulated Catholic Action, which made inroads with some priests. The Chilean Church hierarchy, however, did not become directly committed to this cause until the late 1950s, following the Rerum Novarum’s treatment of the social question. Throughout the thirties and forties, the Church in Chile was more concerned with the advance of communism (see Walker 1999). 83

116

interpretation of the importance of their mission, to take up the challenge of the political and social commitment asked for by the Church in its evolving social mission” (Scully 1992, p. 116). Very shortly thereafter, in 1938-39 following a controversy over the presidential candidate, the group left the Conservative Party, and the Falange Nacional became a separate political party.85 It increasingly developed a defined Purpose. Among other significant programmatic definitions contained in the Veinticuatro Puntos Fundamentales were the party’s social christianism, corporatism, and anti-communism and anti-fascism (Grayson 1968, pp. 168-171, Walker 1999, pp. 101-08, Nieva 2000).86 For twenty years, from 1941 to 1958, the party played only a very small role in national elections, never receiving more than 4% of the vote. Such electoral weakness, however, did not debilitate the organization. In fact, the resilience of electorally irrelevant groups—regardless of their later success—could be considered a useful proxy for Purpose; only determined organizations can resist irrelevance in the absence of notable electoral support.87 Their determination, a strong Purpose and the semi-traumatic distance from the Conservative88 leaders all might have bolstered the stability of the Falange Nacional. In the Popular Front years (1938-1948), the Falange Nacional

85

The name was adopted a few months before, in October 1937. It is worth mentioning that tensions with the Conservative Party elite first became evident when members of the Conservative Youth gained access to Congress and began differing with their elder colleagues there (see Nieva 2000). The split with the Conservative Party also complicated the group´s relations with the Church. They were tense throughout the 1940s and part of the 1950s. Their opposition to the banning of the Communist party (until 1958) by the Law for the Defense of Democracy (Ley Maldita, enacted in 1948) supported by the Church, illustrated this tension (see Scully 1992, Grayson 1968). 86 The influence of and relationship with the Falange in Spain and Primo de ivera’s proto-fascist movement is a matter of some dispute. However, admirers of Mussolini and Fascism rapidly became a minority within the group (Gil 1966, Grayson 1968, and Nieva 2000 review the intellectual dispute over the influence of fascism in the Falange Nacional). 87 The Communist Party in post-authoritarian Chile is another political organization with Purpose and consistent electoral irrelevance. They in fact rejected the terms of the transition, despite most likely anticipating the future poor electoral performance this decision would imply. Thus, again, Purpose can be sufficiently influential as to lead a party even to neglect or postpone Ambition, though the two need not necessarily work at cross purposes. 88 See Grayson (1968). 117

oscillated between supporting the alliance candidates and those of the Conservative Party.89

In the 1950s the party started to debate its organizational and programmatic identity. In 1953, they started calling for unification of social Christian forces. Programmatically, the party reinforced its concern for social issues, building on the increasingly more progressive social doctrine of the Church. Its other main ideals were state interventionism, change without revolution, democracy, and communitarianism—as opposed to liberal capitalist doctrine (Gil 1966). Changes in the Church consolidated during the papacy of John XXIII, the rise of a more progressive view within the Chilean Church,90 the fall of fascist regimes in Europe, and the increasing influence of the Cold War all might have influenced this development of Purpose.

Scully summarizes the relationship between the church and the Falange (later the DC) and the role this link played in the formation of new party leaders this way:

Over a period of few years, the roles and structures of the church and the National Falage overlapped and reinforced one another. At the same time the church was strengthening its organizational links with hitherto pastorally neglected sectors of the population…the Falange was busy forming leadership cadres among the same groups, drawing party militants from church-related programs” (1992, p. 131).

89

For example, in 1946 they supported the Conservative Party presidential candidate Eduardo Cruz-Coke. Scully says: “[i]n the 1950s, the battle within the Chilean church was resolved in favor of those who were inclined toward a progressive interpretation of the social question and, consequently, in favor of the National Falange...” (1992, p. 117). One of the most significant changes was the appointment, in 1958, of Raúl Silva Henríquez as Archbishop of Santiago. Also see Moulian´s (1993) analysis of the process of change within the Chilean Church (pp. 126-129). Walker (1999) attributes those changes to the influence of the social Christian groups, whose leaders apparently counter-balanced the dominant influence wielded by the Church´s hierarchy in the early years (1999, p. 67-69). 90

118

The Falange Nacional remained organizationally independent of the Chilean Church hierarchy (Walker 1999).91 This organizational independence would become increasingly marked (this was clearer during the Frente Popular years, 1938-1948). The greatest organizational independence was then achieved with the creation of the Christian Democratic Party.

Finally, in 1957 the Christian Democratic Party was founded, resulting in the coalition of social Christian groups, the Falange Nacional, and the Partido Conservador Social Cristiano (Grayson 1968, Walker 1999, Nieva 2000).92 When the DC was finally created, it took on the role of consolidating a centrist alternative between capitalism and Marxism (Valenzuela 1978, Scully 1992). Scully splendidly summarizes the DC’s Purpose: …the Christian Democrats appeared as a dynamic party with a fresh perspective, untainted by earlier political compromises. Under the leadership of Eduardo Frei and Jaime Castillo, the Christian Democrats developed a program that carefully reflected the aspirations of a crucial political ally, the Chilean episcopacy. Combining the developmental ideology of the…CEPAL, with the social promotion theories of the Jesuits…the Christian Democratic party articulated a new, Christianinspired vision for resolving the sharpened social conflict in the Chilean society” (1992, pp. 147-8). As this account suggests, the DC emerged with a strong Purpose93, albeit one that may have been less coherent than that of its predecessor, the Falange Nacional. The DC’s 91

The author says: “[v]eremos que es imposible comprender los orígenes y el desarrollo de la democracia cristiana sin referencia al catolicismo, pero que es perfectamente erróneo considerar este movimiento o partido simplemente como un subproducto de la Iglesia Católica, esto es, como un partido instrumental” (Walker 1999, p. 59). This point is also highlighted by Huneeus (2003). 92 For a thorough analysis of the evolution of the Falange, its relationship with the Church, and its political participation, see Scully (1992, chapter 4) and Grayson (1968, 1969). 93 For a detailed description of the changes in the center of the ideological spectrum, see Scully (1992), Moulian (1993), and Valenzuela (1978). They present different perspectives on the role of the DC in the context of polarization, and ultimately as an ideological party, to explain the 1973 democratic breakdown, but they do agree in that the DC had a strong Purpose. 119

promotion of new young leaders also signaled the presence of Ambition (interview with party leader, Loveman 1976, Scully 1992).94

The party kept gaining influence, and in 1964 it won the presidential election. Eduardo Frei’s “Revolution in Liberty” is perhaps one of the clearest manifestations of the relationship between Purpose and party vibrancy. The electoral campaign drew on different social organizations, different party institutions, and militants countrywide (Gil 1966, Grayson 1968, 1969, Valenzuela 1978, Scully 1992, Moulian 1993). Aided by the solidifying of mechanisms ensuring clean and competitive elections95, the DC had become a vibrant party, one with a strong Purpose. It was also capable of effectively promoting new leaders, as evidenced by the low average age of Frei’s ministers. Luna, Monestier, and Rosenblatt (2013) review the DC’s ideological trajectory. The party was formed as a progressive, religiously oriented party (Ozzano 2013). As polarization increased in Chile near the 1970 election, the balance between conservatives and progressives within the party increasingly tilted in favor of the former. Party leaders consequently split, forming two groups: Movimiento de Acción Popular Unitaria (MAPU, 1969) and Izquierda Cristiana (1971), which joined Allende’s Unidad Popular coalition (see Valenzuela 1976 and 1978). The polarization persisted and 94

Together with changes within the Church, the 1958 change in electoral rules protecting the secrecy of the vote in rural settings facilitated the DC’s strategy of aiming at lower sectors of the Chilean urban and rural society (Loveman 1976, Valenzuela 1985, Scully 1992). It would be interesting further to analyze the type of leaders it promoted, their characteristics, background, and ability to develop a political career— though the events of 1973 interrupted this process. In this sense, Moulian (1993) argues that all parties in these years (1932-1973) had an “oligarchic character”, with party elites enjoying unrivaled preeminence. 95 Valenzuela (1978, pp. 9-10), claims that the system prevented cooperation between parties’ candidates because voters cast their votes for just one candidate and parties therefore competed among one other for votes. Valenzuela (1994) also provides a detailed analysis of the failure of existing rules in Chile, especially the 1925 Constitution, and their unfortunate evolution before the 1973 coup. However, Gamboa and Morales (typescript) argue that the system introduced with the 1925 Constitution, which combined PR, a d´Hondt electoral formula, party lists, and greater district magnitude facilitated more strategic cooperation within party lists than the Cumulative Voting System previously in effect. 120

radicalized during the Allende government, putting significant pressure on the DC and increasingly placing it in the opposition (see below and Valenzuela 1978, Valenzuela and Valenzuela 1976, Scully 1992). The salience of exogenous anti-Communist rhetoric and international zeitgeist of the Cold War both encouraged the party’s conservative turn. Organizationally, the DC’s increasing “partido único stand” (Valenzuela 1978, p. 38)96 and the nature of its doctrine (socially close to the left and morally to the right) also contributed to its pendular swings in Purpose.

During the authoritarian regime, the party was formally suspended, but its closeness to the Church helped keep it going (Scully 1992). The party gradually moved into the opposition and, united with sectors of the left, increasingly organized a political front against Pinochet (Interviews with DC, PPD, and PS leaders, Garretón 1995, Scully 1995). The DC´s pre-Pinochet cleavage, i.e., that between Allende and the opposition, was still lively.97 As democracy was consolidated, the party was put in an increasingly uncomfortable situation: the Church was becoming more conservative whereas their allies in the Concertación were increasingly emphasizing progressive perspectives (Luna, Monestier, and Rosenblatt 2013). This, as well as the pre-1973 cleavage affected the party’s Purpose: its most defining identity was not resolved and remains disputed even today.98

96

This is disputed by Scully (1992) and Huneeus (2003). This cleavage is illustrated by the internal disputes over Aylwin´s presidential candidacy (Angell and Pollack 1990, Scully 1992, p. 198) and by internal debates over the desirability of the “own path” strategy (Garretón 1995). However, Aylwin showed great ability to smooth tensions between rivals of the past (Rehren 1993). 98 As reviewed above, the conservative identity is more associated with the party’s origins while the progressive identity developed from the 1950s until Allende´s government. 97

121

The DC was a major factor contributing to the smoothness of the Concertación coalition.99 The DC was the backbone of the alliance, and the most important party in the system, both electorally and organizationally, through the 1990s (Luna forthcoming).

Interviewee after interviewee concurred with the idea that governmental responsibilities negatively affected the party’s vibrancy.100 The findings of my fieldwork also suggest two other factors antecedent to the sense of tedium provoked by the Concertacion’s twenty-year hold on government. First, as seen above, the party lost its uniting sense of Purpose, which was increasingly disputed internally from the transition years onwards. Some interviewees felt the party needed to discuss its programmatic identity, more to improve its intensity of Purpose than its coherence. In terms of coherence, variance in the DC’s programmatic positions does not distinguish it from other political parties. Its problem is not so much one of internal divergences as lack of resolution regarding a defining set of ideas or project to which at least most of its leaders are committed. The abandonment of the “partido único” posture within the DC (Huneeus 2003) together with broader programmatic transformations (i.e., moderation within the PS, stronger bonds between the PS and DC, and emphasis on political stability), may have contributed to the DC’s neglect of intensity of Purpose.101

Second, party elites did not promote new leaders. Those who were leaders in 1973 retained their positions and exerted strong influence for the next two decades after democratic transition. The need to accomplish equilibrium in the distribution of power 99

For a thorough review of the Concertación years, see Weyland (1997), Drake and Jaksic ed. (1999), Muñoz (2007), and Ffrench Davis (2008). 100 This impact, among other considerations, is also highlighted by Walker (Walker 1999, 2003). See Walker (1999, p. 89 and pp. 129-92) for an interpretation of the electoral decline of the DC and the challenges it then faced. 101 Even though the attainment of democratic consolidation might be considered a source of Purpose, given the process that opponents underwent during the Pinochet authoritarian regime, I think it is more associated with Trauma. The need to build solid grounds for preventing a reversion to authoritarianism was their main motto—and the threat was actually pervasive as illustrated by the Operación de Enlace, in 1990 and Boinazo, in 1993 (Loveman 1995). 122

positions between parties of the Concertación contributed to reducing the DC’s ability to promote new leaders.

It is very complicated to get a clear perspective on the DC from the truth table because interviewees are extremely dispersed. It is also intriguing that their most consolidated and older leaders are located at the extremes.102 However, a slightly clearer picture emerges when one turns to agreement on individual conditions and reviews the interviews in depth. The most significant point of agreement—though more tenuous than those noted for other parties—concerns their memories of the difficult years under Pinochet, and their epic of the years leading to the democratic transition in 1990.103

One could be tempted to conclude from the final coding of results that there is equal agreement within the DC regarding the existence of high barriers to exit. However, although the rate of agreement on Exit is similar to that for Trauma, in-depth analysis of the evidence reveals that the agreement on Exit is less emphatic. Nevertheless, it complies with the rules presented above, and I thus computed it as present.

In contrast, there was considerable division and no clear pattern of agreement regarding Ambition. The rules for resolving disagreements among interviewees regarding the presence of a condition do not support concluding that the party meets the Ambition condition. Less than half of my interviewees agreed that the party satisfied Ambition, and there was an intense cleavage between generations and between established leaders and prospective ones on this point. The case of the DC illustrates the importance of 102

When analyzing other potential traits, no other relevant divide is apparent. One could argue that this party played a definite role in the weeks prior to the coup by not agreeing with the Allende´s government on finding a solution to the polarizing climate (see Valenzuela 1977, Scully 1992). Their role before the breakdown of democracy thus raises some doubts about their feelings about the regime’s breakdown. Here, however, I am interested in what politicians themselves report. In historical terms, it is also true that the party—through its young militants—played a significant role in the Vicariate of Solidarity and in the opposition movements of the early 1980s (see Garretón 1989, Luna, Monestier, and Rosenblatt 2013). 103

123

analyzing both formal rules and subjective accounts in evaluating whether the condition of Ambition is met. The DC has a formal structure and channels for new leaders’ pursuit of political careers. There are also informal rules and traditions on how to succeed within the party. However, those traditions and the formal structures have become increasingly unused by the party’s elite.

¿Cómo se pasa de ser un militante anónimo hoy en día a ser un candidato en la DC? El único método es trabajando con los líderes establecidos o habiendo conseguido dirigir una organización social importante, o sea tener una presencia pública reconocida o el trabajo interno con líderes. ¿Cuáles son los mecanismos que existen hoy día en su partido para hacer una carrera política? En otras palabras, si Ud. hoy en día tuviera que iniciar nuevamente su carrera política, ¿cómo se imagina los primeros pasos? ¿Qué debería hacer para crecer políticamente? Tú me decías que hay una generación muy marcada por la dictadura por la transición, entonces los que hoy en día los de 30-35 años son opacados por esa épica, entonces cómo se pasa a ser candidato teniendo a esa gente con una épica tan importante, ¿es así o no? ¿hay un opacamiento o no? No, es que los partidos se desintegran porque no quieren renovarse, pregúntale a los socialistas si alguien quiere entrar en el PS se va a encontrar con un montón de barreras, inscribirse en la DC era muy difícil, si eres calificado, peor, porque eres una amenaza entonces no te dejan entrar. En la DC lo que vamos a hacer es establecer la ficha electrónica, cualquiera se va a poder inscribir si quiere y eso va a perder la mitad del poder todas las maquinarias habidas y por haber en la DC, no lo van a poder evitar y ya va a entrar mucha gente y quiere mucha gente entrar.

A new generation of leaders has, nonetheless, started to take control of the party, and has tried to restore the party’s structural and organizational assets. Those new leader have shared more than two decades of political experience in the party with the older, established elite. This might facilitate the inter-generational transition and reduce the chances of drastic disruption.

The observed configuration for the party is thus ETpa. PS: The period from 1924 to 1932 was one of political instability and conflict in Chile

(Scully 1992). The history of the Socialist Party (PS) is associated with this turbulent 124

period. It was formed in April 1933, a few months after the brief but surprisingly significant juncture of the República Socialista.

Led by Marmaduke Grove, the

República Socialista set the foundations of modern labor union organizations and the PS. With its creation, a leftist axis was configured. 104 The PS united previous small socialist groups (Gil 1966, p. 283, Drake 1978, De Ramón 2003).105 It never attained the electoral significance of the Radicals or, later, the Christian Democrats. It was, however, responsible for forming the Frente Popular coalition, which governed until 1941 under Radical Party President Pedro Aguirre Cerda. Other Radical presidents then governed until 1952.106 The PS’s alliance with the Radicals—and other groups—provoked intense internal conflicts which led to successive splits and formation of a splinter party—the Socialist Workers Party formed in 1942 (Gil 1966, p. 284, Faúndez 1988, Moulian 1993, p. 38).107 Thus, the party was born with an interesting combination of Trauma associated

104

Grove was the party´s first presidential candidate. See Zemelman (1971) for a description of Grove´s adventure, the República Socialista that lasted just twelve days and its—otherwise surprising—influence. For a detailed description of the evolution of leftist, Marxist parties in Chile until 1973, see Drake (1978) and Faúndez (1988). Also, see S. Valenzuela (1976), and Zemelman (1971) for an analysis of the evolution of the labor movement and labor parties in Chile. Zemelman focuses on the 1930s. 105 Roberts summarizes the heterogeneous composition of the newly created PS as follows: “Thus the Socialist Party was born—and remained—an open, inclusive, and heterogeneous organization that rather awkwardly aggregated left-wing freemasons, populists, and democratic socialists under the same flag as revolutionary Marxists, Trotskyists, and anarcho-syndicalists.” (1998, p.87). One of its main differences from the Communist Party was its refusal to join the Socialist International and its strong “Latinoamericanismo.” (Faúndez 1988, p. 27, Drake 1978, Moulian 1993). 106 The party formally withdrew from the coalition in 1941, but it kept influence retaining ministers in the cabinet and sometimes through support of executive bills in Congress. 107 Faúndez summarize this period of the leftist parties as follows: “From 1932 to the end of the 1940s the Communist and Socialist parties were prominent participants in the political process. In many respects, their behavior was not unlike that of any other non-marxist party. For their actions seemed to be largely determined by the overriding objective of participating, albeit as junior partners, in coalition governments led by the Radical Party. To secure this objective, they campaigned vigorously to increase their electoral appeal, formed parliamentary alliances with non-marxist parties, subordinated the interests of the labour movement to their political ambitions, and occasionally even made use of their influence in government to attack and undermine each other” (Faúndez 1988, p. 77). 125

with Grove’s failed República Socialista and its violent antecedent events and suppression—and from the first strong internal divergences in terms of Purpose.108

The electoral system authorized the formation of electoral pacts. However, it was not until 1956, with the formation of the Frente de Acción Popular (FRAP), that leftist parties (most importantly the Communist and the Socialist Party’s two factions, PS and Socialist Workers Party) constituted a single unified pact, lasting until 1969 (Gil 1966, Moulian 1993). This also facilitated the re-unification of the Socialist Party in 1957. Purpose gets more unified and coherent in these years. Gil (1966)—who writes his book before Allende´s government—summarizes PS ideology at the time as follows: “In general terms the Socialist Party labels itself as Marxist, revolutionary, anti-capitalist, anti-imperialist, classist, and “Americanist” (meaning advocating the unity of all the Latin American peoples).” (p. 286).109

In the 1960s, the PS progressively developed more extremist ideological stances, i.e. increased the intensity of Purpose (Valenzuela 1978, Drake 1978, Roberts 1998).110 At a 1967 congress, the PS declared that “violent revolution was inevitable” (Valenzuela 1978, p.38, Scully 1992, p. 140). This declaration occurred in an increasingly polarized context with centrifugal tendencies (see Valenzuela and Valenzuela 1976, Valenzuela 1978, Valenzuela and Scully 1997, Scully 1992, Moulian 1993). In 1970, the Socialist candidate, Salvador Allende, was elected president. Under his Unidad Popular (UP)

108

Gil (1966, p. 285) mentions that it took thirteen years before the party developed a clear party ideology. 109 Moulian (1993) highlights that the main difference between the Communist Party and the PS resided in their international references; the PS was critical of Stalinism and the USSR (Moulian 1993). 110 It would be interesting to study the progressive radicalization of the PS. At the end of the 1930s and in the early 1940s it was a moderate party that negotiated with the centrist Radicals. However, a few decades later, the party radicalized its postures. This process is carefully reviewed in Faúndez (1988). The PS debated between the electoral strategy and revolution. The defeat at the 1964 presidential elections left the party in the control of the more radical wings—those who viewed revolution as inevitable and were inspired by the Cuban revolution (See Faúndez 1988, pp. 164-71). 126

government,111 radical militarized leftist organizations gained importance.112 An increasingly dangerous, polarized scenario ensued (Valenzuela 1978, Moulian 1993, Roberts 1998).113 Valenzuela (1978) and Roberts (1998) argue that the PS was very critical and active in challenging Allende’s belief in the possibility of founding a socialist order under democratic rules.114

The Socialist Party shares the presence of trauma (T) with the DC. This is unsurprising given that the party was the one that suffered the most during the Pinochet era, beginning with Allende’s overthrow and forced suicide. From the very inception of the coup, leftist leaders and militants were the main targets of repression. 115 The Pinochet dictatorship was among the harshest in the region. It gained full control of society through intense repression, and deeply penetrated Chilean culture through a

111

UP was the coalition of parties that supported Salvador Allende’s presidential candidacy and later presented joint lists for the congressional elections in March, 1973. The PS, Communist Party, IC, and MAPU, among other political groups and parties formed the UP. 112 The most important was the Movimiento de Izquierda Revolucionaria (MIR) founded by students of the Universidad de Concepción who left the PS. 113 The Cuban revolution exerted strong influence on leftist parties and organizations throughout the region. The Tupamaros in Uruguay are another example. 114 Valenzuela (1978, pp. 34-39) claims that increasing ideological polarization, combined with an increasingly fractionalized system, an ideologically divergent coalition, a minority president and—for him most important of all—the rupture during Frei´s government of the “system of compromises” explained by the DC´s ideologically centrist nature and partido único stance) all explain the breakdown of democracy in Chile. 115 oberts summarizes the degree of the violent unrest against the left: “Factories, universities, and shantytowns were occupied by security forces, who killed thousands of left-wing activists who resisted the coup or were detained in custody. Within six months they [the military] had arrested 8,000 persons, and 160,000 suffered politically motivated job dismissals. An estimated 200,000 persons, including political refugees and their family members, went into exile in the global diaspora of the Chilean left…Both parties [the PS and the Communists] had their leadership decapitated three times during the first three years of the military dictatorship, and their ranks were decimated by repression, exile, and political withdrawal.” (1988, p. 94-95). 127

transformative project.116 Among its most defined and targeted enemies were Socialist and Communist Party leaders and militants.

The following quotes from two leaders incorporate Trauma as part of their analysis. The first summarizes his experience during the authoritarian regime and the moments after the transition; the other highlights Trauma’s impact on the organization’s endurance:

El 11 mismo fui incorporado en una lista de requeridos y fueron a allanar mi casa y a los días siguientes la casa de familiares así que tuve una clandestinidad inmediata. Luego, el secretario general de la JS, que hoy es detenido desaparecido, me ordenó a asilarme y el padre Rafael Maroto me asiló en la embajada de Austria. Estuve 6 meses en Austria. En el año 74 estuve un año en Moscú. Después, del 75 al 78 estuve en Cuba, en La Habana. Después estuve del 78 al comienzo del 82 en Berlín Oriental. En el año 78 fui nombrado encargado de exterior de la JS, era encargado para el exterior del trabajo de la JS y era miembro del Comité Central del partido por esa razón. Tenía 23 años, a fines del 78. En el año 82 me vine clandestinamente a Chile. Fui el jefe de la operación de retorno del PS que encabezaba Clodomiro Almeyda. Tuve 6 años de clandestinidad. En el año 84 entré a la comisión política de ese partido…en las condiciones de la época, de severa clandestinidad y yo era encargado interno del partido durante 4 años, del 84 al 88…En el 89 fui candidato a diputado y salí electo…

Yo creo que si porque nosotros no hemos vivido ese tipo de crisis tan profundamente a pesar que la crisis de Frei fue brutal, yo creo que pueden sobrevivir por el sustrato histórico porque o sea nosotros sobrevivimos a una dictadura cuando te exiliaron a toda la gente y te mataron a la otra mitad (…) y aquí estamos como decían las primeras marchas que se hicieron contra la dictadura donde aparecieron partidos los socialistas gritaban y que juegue que juegue que estamos otra vez

The next quote illustrates how Trauma creates an epic featuring those involved in the experience, influencing their role as leaders. It also shows how this effect wanes over time.

116

Though it tried to craft Chilean politics through the introduction of a corporatist model, which included the elimination of parties, it was not successful. Its greater “success” was in laying the foundations of a new economic model. For detailed descriptions on the authoritarian regime, including its economic and political project, see Garretón (1983, 1995), Valenzuela (1995), Silva (1996), Huneeus (2000a, b). 128

P: ¿Qué fortalezas y debilidades le encuentra al sistema de partidos chileno hoy en día, en comparación con otros de la región? R: Las fortalezas son que los partidos chilenos tienen una tradición que viene de muy atrás. Las debilidades son las propias de una constitución que hace que si saca 60% elige uno y si yo saco 40 elijo el otro. En consecuencia, cuando en una elección se elige uno y uno, eso significa a la larga que es empate. Si yo sé que van a empatar ¿para qué me meto en política? Por lo tanto, los partidos dejan de tener relevancia. No quiero decir que es exclusivamente por el sistema electoral, el sistema electoral es responsable pero también los partidos políticos. Usted habla de una épica, claro, es re fácil golpear la puerta de una casa y decir “señora vengo a invitarla que vote que no, este señor lleva 17 años y mire cómo estamos, estamos jodidos bla bla bla, vote que no y listo.” Pero es muy distinto golpear la puerta de una casa y decir “señora vengo a invitarla a que vote por mí” “¿Por usted?” “Sí, por mí, yo sé lo que hay que hacer…” Bueno, para hacer eso más te vale tener ideas y creértelas porque es mucho más fácil invitar a votar que no. P: ¿La épica se terminó? R: …Esa épica corresponde a un momento histórico del país que no es reproducible. Lo que ocurre es que la política deviene en una actividad muy profesional en el buen sentido. No en el sentido de que para vivir tienes que vivir de la política, no. Profesional en el sentido de que se necesitan conocimientos, etc. Pero también, como me dijeron en una reunión que tuve hoy a la mañana, “señor, muchas políticas públicas y poca política” y eso es malo.

Unlike other Latin American parties suffering Trauma, the PS was brutally deposed from office. This difference, and perhaps also the PS’s role in governing the country immediately after democracy was restored, is crucial for understanding the history of Trauma and its relationship with Purpose within the PS (see chapter 7).

Roberts (1998) carefully analyzes the process of transformation undergone by the PS (and the Communists) during Pinochet’s dictatorship. The PS reproduced its characteristic organizational fragmentation during the authoritarian regime: “The party´s congenital factionalism led to the emergence of competing centers of authority in clandestinity” (p. 95).117 Thus, repression and factionalization (both inside Chile and in 117

Those “competing centers” disputed the interpretation of the fall of Allende and ideology. Later, ideological debates were complemented by a debate on the best strategy for confronting Pinochet. The two main factions were the Núñez-Socialists and the Almeyda-Socialists; the former represented the “renovated” socialists while the latter had a more radical ideological stance and approach to the fight against Pinochet. Ricardo Núñez and Clodomiro Almeyda were the secretary generals of each faction. For a detailed review of this internal conflict and its resolution, see Roberts (1998, pp. 94-110). 129

exile) provoked serious organizational disputes within the PS. In the early 1980s, the later victorious faction was known as the Núñez-Socialists, and the first of a series of ideological transformations was the revalorization of democratic politics (interviews with PPD and PS leaders, Garretón 1995, Roberts 1998). This group forged agreements with the DC, starting in September 1983 with the Democratic Alliance. 118 Other attempts to unite the opposition followed, the most important of which were: an alliance in February 1988 of fourteen parties united first under the Concertación de Partidos por el No coalition to defeat Pinochet in the October 5 plebiscite and later under the Concertación Por la Democracia pact for the 1989 elections.119 Thus, in the mid-1980s PS leaders, together with leaders from the DC, emphasized democratic consolidation and programmatic moderation.120 Valenzuela and Scully (1997) highlight that Chile’s post-authoritarian party system did not show the “same sharp ideological polarity” (p. 515) as in the 1960s, and that the greatest change was within the PS. In my fieldwork, most politicians did not identify Purpose as present, and moreover explicitly referred to problems of coherence and/or intensity.

118

That same year, more radical socialists and communists formed the Movimiento Democrático Popular (Garretón 1995, Roberts 1998). 119 These two coalitions did not include significant participation from center-right groups. AlmeydaSocialists reunited with the Núñez-Socialists, Izquierda Cristiana, and MAPU, shortly after the 1989 elections. See Garretón (1983, 1995) for an analysis of the process of transformation and consolidation of the opposition to Pinochet. 120 The failure of streets protests (mainly in 1983), which proved ineffective for debilitating Pinochet and the dangerous violence emanating from the Frente Patriótico Manuel Rodríguez (FPMR), combined with the increasing economic success of the regime and its level of institutionalization (Roberts 1998, Huneeus 2003, Garretón 1995) might also explain the emphasis on moderation and democratic consolidation above all other claims and strategies. This emphasis included stressing the need to accept the dictatorship´s terms of transition, which would prove very influential for later developments. Roberts (1998) identifies 1986 as the critical year in the definition of a political-electoral strategy for defeating Pinochet. The FPMR arms arsenals, which were discovered by the regime, and the failed assassination attempt against Pinochet proved ineffective, augmented repression, and jeopordized the chances of defeating Pinochet. 130

The absence of Purpose could be mainly explained by the harsh memories resulting from the failed Allende government. In this sense, Trauma may not have activated Purpose because the traumatic experience resulted from failure while in government. Former party ideas were thus cast into doubt, as illustrated by the profound debates within the PS during Pinochet regime (see Roberts 1998), limiting the party’s ability to generate an intense sense of Purpose. Also, many PS leaders were exiled in East Germany and East European communist countries and were disappointed with their experiences there. This experience of disillusion affected their programmatic beliefs. The PS’s intensity of Purpose was further undermined by the fall of the Berlin Wall and the collapse of the Soviet Union (interviews with PS leaders, Scully 1995, Roberts 1998). 121 Finally, as seen above, the party decided to privilege democratic transition and consolidation over other ideals. A long process of discussion and organizational rearrangement culminated in a predominantly moderate, ideologically “renovated”, PS party.

As Roberts (1998) and others highlight, the PS may have had no other option than to moderate its intensity in terms of Purpose. Concertación members’ main task was to convince the country they could govern. They therefore had to embrace pragmatism (Weyland 1997). If democratic transition was to be achieved, allying with the centrist DC was probably the only chance they had. It could be argued that this moderation was a key factor in ensuring democratic stability after the transition. Be that as it may, here again we see that, as suggested in Chapter 2, in unequal societies like those in Latin America, democracy is either boring and stable or intense and unstable.

It is now clear that the authoritarian regime and democratic transition left its marks. Leaders, as they themselves acknowledge, also exerted self-restraint in the creation of a new defined Purpose (interview with PS leaders, Roberts 1998, p. 144). Moreover, years

121

oberts’ (1998) analysis seems to imply that the PS’s abandonment of its former clear Purpose was mainly elite-driven. Disputes at the elite level in a context of exile, clandestinity, and repression may also have made maintaining a clear sense of Purpose impossible. 131

of success in this regard also eroded any passion for change or determination to restore an intense sense of purpose (interview with PS leader).

The non-definition in terms of Purpose—most of all in terms of intensity—has affected party vibrancy, above all through an intentional distancing of the party from social movements and organizations. Interviewees, informal conversation with PS militants, and Roberts (1998) all agree that internal party disputes have increasingly centered on personal issues, reinforcing the organization’s absence of Purpose. In the following quote, an interviewee analyzes this process.

"Creo que nosotros no hemos logrado instalar un relato a partir de todo lo que hicimos, porque es el único país del mundo donde se vende el curel como salmón y se vende porque en Chile el dominio cultural, político, económico de la clase alta es brutal, es un país extremadamente heterogéneo pero hay 25 grupos que son los dueños de los medios de comunicación, de las empresas, de los bancos, son los que sustentan a la Iglesia, a las fuerzas armadas. Ese grupo sustentó a la dictadura y dijo que todos los que nos dedicábamos a la política éramos terroristas. Esa cuestión se acabó en los 90 y se transformó ese discurso en que éramos todos unos delincuentes, mentirosos. Cuando tienes que pelear contra una hegemonía que antes era armada y ahora es cultural y política...Durante 20 años cuando La Tercera decía que un ministro había robado todos decían que sí, todos los titulares eran robos, eso era el ambiente. Y asumió Piñera y eso cambió radicalmente. Nosotros tenemos un lío de que tenemos una capa dirigencial que esta comprada, otros que están arrendados, que quieren ser como el otro desde el punto de vista de la moda, de lo que vive, de lo que hace, es típico colonial. Esa capa está colonizada por la derecha que se dice de izquierda pero habla como la gente de derecha; quiere vestirse como los de derecha, van a los mismos seminarios que van los de derecha y eso se traduce en una cosa bien sintomática que es producto de la dictadura y de lo que ha pasado después. Antiguamente en Chile en el mundo cultural y político el tipo que se sentía sofisticado intercalaba en sus frases palabras en latín hoy en día lo hace en inglés. Claro, porque toda la clase alta chilena terminó yendo a estudiar en los últimos 37 años a EEUU se considera de lujo hablar en inglés, eso nos ha jodido. El colonialismo político, considerar que un gallo de derecha tiene buen olor y en realidad tiene mal olor. Si tú escuchas a los políticos hablar los tipos ni usan conceptos políticos, usan conceptos técnicos que es propia de los políticos de derecha. La racionalidad que se ha instalado en nuestra gente es la racionalidad de derecha, por ejemplo el uso de parámetros que es una cultura gringa, todo se traduce en índices y no en las personas. Nuestras principales peleas con la gente del aparato público de la Concertación es estar preocupado por las cosas y no por la gente. Eso fue el gran cambio que intentó Bachelet; pensar más en la gente y no en las cosas. Esa racionalidad es muy consonante con la racionalidad de los parámetros; nosotros en educación no discutimos si lo que se hizo es bueno o malo, sino que nos trasladamos hacia la racionalidad de la derecha. Tu no haces épica hablando en parámetros. Usas la lógica americana para decir lo que es bueno o malo, terminas hablando de parámetros y no de gente y ahí no construyes épica, ese es el problema que tenemos, esa es la firme. Si uno tuviera que dividir a la clase política ya no en general sino a las clases políticas internas de los partidos hay unos que están preocupados de los parámetros y hay otros que están preocupados por la gente. Y esa es la discusión que hemos tenido durante los últimos 20 años."

132

The fact of being in government at the time of collective Trauma, the party’s subsequent emphasis on stability, and the epic that surrounded those who fought successfully for the transition to democracy seem to have reduced the party’s ability or willingness to promote new leaders, i.e., Ambition. However, it is worth pointing out that most Chilean parties have never had a very democratic structure permitting vertical advancement (perhaps the greatest incorporation of new political elites may have come about with the emergence of leftist parties, including the PS, in the early twentieth century, though further analysis of this conjecture would be required). This trait is highlighted by Moulian (1993) and other scholars who have analyzed Congress and parties since the transition (e.g. Navia 2000)) and it also emerged during my fieldwork. Certainly, a preliminary analysis of historical accounts of Chilean parties’ history does reveal an impressive repetition of surnames.122 One of the favorable traits of the party in terms of Ambition is that its history gives a more organizational imprint to those who actually perceive it as possible to purse a political career within the party—and through the party.

In striking contrast to the DC, politicians from the PS disagreed significantly about barriers to exit. The party’s history operates as a barrier to exit and the label is valued. Interestingly, established leaders from the PS almost unanimously claimed that the configuration of parties might change in the future or that the system as a whole might not be able to resist a severe exogenous shock. Thus, their perception of low barriers to exit can be traced to their answers about prospective and hypothetical scenarios. In this sense, they do not think that the binominal electoral rules have exerted the same influence in recent years as they did previously. Precisely among those who considered that the party system would not be stable in the future (i.e., that substantive changes would occur), many mentioned the waning effect of the binomial electoral system on reducing and stabilizing the number of parties. In fact, many mentioned the electoral 122

This pattern is also common in Costa Rica and Uruguay. However, two qualifications have to be made. First, given their comparatively small populations, it is striking how they have been able to develop wider political elites. Second, in the case of Uruguay and the PLN in Costa Rica, new political elites from different backgrounds did emerge at different points in time. 133

system’s contribution to a personalization of politics which, combined with an increasing ossification of parties, might generate the emergence of new organizations. As one young PS leader stated:

“Lo que pasa es que yo creo que el sistema institucional heredado de la dictadura establece tal niveles de restricciones que pareciera que da lo mismo quién gane que todo va a seguir igual. Eso no es así, hay diferencias pero esta confusión está claramente en la sociedad. El propio sistema electoral binominal que establece un virtual empate en todas las circunscripciones y distritos a lo largo de Chile da la impresión que da lo mismo por quien votes que todo va a seguir igual.”

The resulting configuration for the PS is thus eTpa. UDI: The Movimiento Gremial (Gremialismo) was a political movement formed in 1965

at the Catholic University of Chile by Jaime Guzmán, a professor of constitutional law there. It gained control over the university’s student federation in 1968—and in every student election there until 1973—and gradually penetrated other universities as well (Huneeus 2001). It was composed of “…young rightists whose political orientation was essentially organicist and authoritarian” (Garretón 2000a, p. 60).

In the early days after the coup, Gremialismo received preferential treatment as a political organization (Huneeus 2000a, p. 340-2). During Pinochet’s dictatorship, in which the movement was very influential (at least until 1983), it concentrated its activities on youth; through the Secretaría de Juventud within the Secretaría General de Gobierno, and through the Frente Juvenil de Unidad Nacional, created by the Junta Militar. Distancing themselves from prior rightist political organizations and taking advantage of the non-competitive setting and the 1976 Organic Law of Municipalities, they also gradually penetrated poor sectors through careful work at the local level (Huneeus 2000a, b, Luna forthcoming). Finally, in the process of economic reforms, they developed close ties with the Chicago Boys (Silva 1996).123 Thus, the movement

123

Angell (2001) highlights that this was a strange relationship given their ideological differences. However, Miguel Kast, an economist who was a leader of Gremialismo at the PUC, headed the Oficina de 134

showed vibrancy, though the term in this case seems inappropriate given the authoritarian context within which the organization grew and consolidated.

From the very beginning, Jaime Guzmán had a very strong ideological and political influence on the Junta Militar, which he maintained throughout the Pinochet dictatorial regime. On September 13, 1973—two days after the coup—he was, at age 27, designated by the Junta to look into a new constitution for Chile (Acta Junta N°1 in Cristi 2000, p. 121-23).

Guzmán was strongly influenced by the conservative Catholic priest Osvaldo Lira and admired Franquismo in Spain (Cristi 2000, pp. 24-5, Huneeus 2000a, pp. 332-336). According to Cristi (2000), Jaime Guzmán built his ideas out of the traditions of conservatism (especially Catholicism, corporatism, and nationalism) and liberalism (Hayek’s neoliberalism and the Chicago School). The author adds: “Las nociones de autoridad y libertad son los pilares sobre los que se asienta el pensamiento político de Guzmán. Constituyen las ‘palancas armoniosas y equilibradas’ que le permiten conjugar un Estado autoritario y una economía libre mercado.” (Cristi 2000, p. 23).124 These ideas, developed gradually and in stages over time, formed the backbone of Gremialismo. Most important of all for the later democratic regime was his idea of democracia tutelada. Speaking in 1977, Jaime Guzmán explained the idea this way:

Sí, creo que la democracia que intentamos construir es una democracia de participación social efectiva, una democracia tecnocrática y, como usted bien lo dijo, una democracia autoritaria y protegida. Yo sé que estas palabras a veces no se comprenden bien, se dice “una democracia autoritaria”, ¡pero cómo! ¡es la antítesis! Es que el concepto de “autoridad” en Chile tiene un sentido muy preciso. Es el sentido de una autoridad fuerte, con poder para conducir el país a través de la justicia y Planificación Nacional (ODEPLAN) and built a close relationship with the Chicago Boys—mostly a group of outsiders (Huneeus 2000a, pp 395-405). 124 For a detailed review and documents on the evolution of Guzmán´s ideas, see Cristi (2000, pp. 23-44) and Fontaine (1991). 135

la ley. Diría que es una democracia enérgica, no una democracia débil. (Jaime Guzmán, as quoted in the documentary, “Chile: Order, Work, and Obedience.”) The UDI was finally formed in 1983 as the political organization of the Gremialismo, and it was led by Jaime Guzmán. In terms of ideas, the party was guided by the ideas of Jaime Guzmán (Huneeus 2001, Cristi 2000, Morales and Burgueño 2001, Pollack 1999). The party’s ideology thus synthesized a conservative identity125 with economic neoliberalism— thereby consolidating its ties with upper sectors (see Luna forthcoming, Morales and Bugueño 2001). As Jaime Guzmán clearly summarized the basic ideas of the UDI:

Hemos definido a la UDI como un partido que reúne tres características: ser un partido popular, de inspiración cristiana y partidario de una economía social de mercado. Lógicamente que estas tres características, como toda descripción o perfil de un partido no son las únicas, que tiene nuestra colectividad ni son las únicas con las que se podría caracterizar a la UDI. ¿Por qué las escogemos? Porque las consideramos sumamente relevantes en su contenido. Y, en segundo lugar, porque ofrecen, lo que siempre se busca al describir a una entidad, que es procurar, no solamente señalar lo que ella es, sino también diferenciarla de otras entidades parecidas; como son otros partidos políticos que existen…todo partido político por definición es organizado, entonces [esto] no aclara ni dice mayormente nada… Un partido popular porque busca enraizarse en los sectores más modestos del país, en los sectores medios-modestos y en los sectores más pobres y más desvalidos de la población. Esta fue una idea que nosotros adoptamos cuando se formó la UDI…Por una parte creemos que es indispensable compartir la realidad de la pobreza con 125

Luna, Monestier, and Rosenblatt (2013) review the party´s conservative stance. They claim that the party emerged during a conservative turn in the Catholic Church under John Paul II. The party became the representative of conservative groups within Chilean society, as illustrated by the influence of Opus Dei and the Legionaries of Christ in the UDI’s leadership and rank-and-file. However, the authors add: “Strikingly, in spite of being firmly rooted in traditional Catholicism…the UDI avoided from its beginning any doctrinaire or practical subordination to religious authorities…even if the party does espouse conservative values that mesh with the conservatism of the Church, the UDI emphasizes those values from a more ecumenical standpoint than that of the PDC.” (Luna, Monestier, and osenblatt 2013, p. 925) 136

quienes la sufren para poder conocerla mejor y para poder, desde ahí, impulsar a esas personas a trabajar por la solución de sus propios problemas…Esta tarea implicó entrar en las poblaciones del gran Santiago y posteriormente en otras grandes regiones del país…de mayor densidad demográfica. Y ahí empezamos a penetrar en aquellos lugares que se consideraban bastiones de la izquierda, bastiones tradicionales del comunismo o del marxismo en su versión socialista y comenzamos a hacer un lento trabajo bastante arduo, bastante perseverante. Precisamente en el ánimo de poder llegar a cada una de las personas y poder comprometerlas con nuestro partido; buscando aquellos dirigentes, aquellas personas que tuvieran una sintonía, una predisposición más favorable. Y eso se tradujo en que en la última elección…casi todos los parlamentarios que la UDI eligió fueron precisamente electos en distritos o en circunscripciones de votación de estratos medios o modestos…De inspiración cristiana. Defendemos y postulamos los valores y principios de la civilización cristiana a la cual pertenecemos y también de un enfoque y un sentido de la vida que esté marcado por un fuerte acento en el destino trascendente del Hombre; en el carácter espiritual del Hombre…No somos un partido ligado a ninguna confesión religiosa, no somos un partido clerical tampoco…A lo que apuntamos con la inspiración cristiana es a poner en la raíz de toda acción política los principios y los valores morales fundamentales que enseñó Cristo…sin ello todo lo demás pierde su sentido más profundo y su norte. La tercera característica es la de ser un partido que defiende una economía social de mercado, basada fundamentalmente en una economía abierta hacia el exterior, competitiva por lo tanto…y basada en la iniciativa creadora de las personas como el elemento principal para promover el desarrollo económico y social. El Estado, en una economía social de mercado juega un rol subsidiario. (Jaime Guzmán, at an UDI activity after the 1989 national elections).126

The party emerged and developed a coherent Purpose. The UDI also became the party most strongly defending the political and economic legacy of Pinochet’s regime (see e.g. Visconti 2011, Novoa 2013, Luna forthcoming).127 The UDI is the most coherent and 126

Video uploaded by the Fundación Jaime Guzmán, n.d. The speech is available at: http://www.youtube.com/watch?v=9WX4yErgsJU 127 Jaime Guzmán and the UDI defended the position that Pinochet should be the regime´s candidate in the 1988 plebiscite. This generated increasing tensions with other members of RN. The situation escalated, and over the course of a few months the supreme tribunal of RN decided to expel Jaime Guzmán, provoking the resignation of all UDI members from RN (for further details see Huneeus 2001, 137

intense party in the Chilean party system. It has a definite and compact conservative identity, and its leaders put great emphasis on it (Luna forthcoming; Luna, Monestier, and Rosenblatt 2013). This fact has not only emerged from the interviews but has also been evident in recent debates on several issues, such as giving legal recognition to same sex couples or when some aspects of the “Chilean model” might potentially be subject to change.

In terms of the intensity of its Purpose, Guzmán left a deep imprint on the party. Luna (2010, forthcoming) describes the “sense of a mission” and “commitment to public service” ingrained in the UDI’s militants and leaders, who operate through a disciplined and hierarchical organization (Joignant and Navia 2003), both influencing the party’s intensity of Purpose. This made other rightist groups that initially composed RN label the UDI as a sect. It also resulted in the construction of very dense network of social activists, especially in poor neighborhoods (Soto 2001, Luna forthcoming), just as had been the case for Gremialismo.128 It promotes young leaders attuned to the party’s identity. The party also promotes independent candidates for congressional and municipal elections. Some leaders have therefore raised concerns about the problem of growing electorally while preserving their party’s distinctive programmatic ideas. However, they never indicated that the party had lost coherence or intensity in programmatic orientation. Instead, they referred to decisions to avert that potential danger.

pp. 32-33). The main tension between Guzmán—and Gremialismo—and the dictatorship was the conflictual relationship between Guzmán and Manuel Contreras, head of the repressive intelligence police (DINA). Though Guzmán did not raise his voice against human rights violations, he did consider that the uncontrolled repressive apparatus was detrimental to the regime´s interest. Cristi (2000) mentions that Guzmán considered human right violations unjustified for the period after 1978, when the “situation of exception” was over. Guzmán then tried to emphasize this dispute with Contreras prior to the Congressional elections in 1989 (Huneeus 2000a, 2001). 128 To achieve this goal, its closeness with the authoritarian regime and economic interests was functional for gaining access to necessary resources (see Huneeus 2000a, 2001, Soto 2001, Luna forthcoming). 138

“Durante mucho tiempo nos hacían bromas de que éramos el último partido leninista que quedaba en el mundo porque éramos todos iguales y pensábamos todos igual. Eso se ha ido relajando un poco pero es un partido con sello; es un partido bastante doctrinario, bastante homogéneo y con una situación de tener un perfil muy marcado.”

However, the most striking aspect of the party was its leader’s neglect of partisan organization, anti-politics rhetoric, and emphasis on a non-ideological approach to politics (Joignant and Navia 2003, Luna forthcoming, Morales and Bugueño 2001). This could seem to suggest a neglect of Purpose. However, its emphasis on the founding and evolving ideas, promotion of new leaders (albeit most from the elite), and rapid consolidation of a partisan organization show that this rhetoric did not translate into practice. They also highlight the difference between my concept of Purpose and that of “ideology” or “programmatic appeals.”

The case of the UDI illustrates another possible source of Trauma. Their main leader was assassinated a year after democracy was restored in Chile. The Frente Patriótico Manuel Rodríguez took responsibility for the attack. I consider Guzmán’s assassination as an instance of Trauma not only because established leaders mentioned this event throughout my interviews but also because of Guzmán’s crucial role as the main leader and ideologue in the origins and evolution of Gremialismo and the UDI.

Jaime

Guzman´s assassination marked a whole generation of UDI militants. Much like the left’s experience of persecution and repression, Guzmán’s assassination unified those whom he had introduced to politics and reinforced the sense of duty Guzmán himself had always emphasized. As the following quote illustrates, Guzmán´s assassination ignited a fervent process of organizational reinvigoration.

“La [militancia] más fuerte la hicimos, precisamente, cuando asesinaron a Jaime. Creo que fueron 100 actos en todo Chile, hicimos actos para, de alguna forma, llenar el vacío de Jaime…Nos permitió estructurar el partido en muchas zonas donde había gremialismo.”

139

UDI militants meet at the memorial built in his memory, and dozens of books about his ideas and assassination have been published. It is notable, however, that explicit references to the impact of his killing appeared mostly in the interviews with established and older leaders. There is thus a prominent generational cleavage when it comes to Trauma. This established leader mentions both this generational cleavage and the role of Guzmán’s assassination:

“Yo te decía que la característica de los cuerpos sociales, lo que le da la esencia, lo que su manera de ser es la forma de relación de los miembros. Nosotros podemos tener dentro de la UDI personas que piensan de una forma determinada en materia valórica bastante distinta a lo que podemos pensar otros, tenemos en la UDI personas que vienen del mundo popular en circunstancia que la UDI nació, llamémosle, de la elite universitaria, de la elite de la elite porque era de la UC…pero la forma de relación nuestra era en torno a un proyecto común, en torno a la defensa de ciertos valores en que tratábamos de evitar las competencias internas, tratábamos siempre de buscar el consenso y eso le dio una característica. Entonces, la gente que entra a la UDI sabe a lo que está entrando. Lo segundo, aunque pueda parecer un poco paradojal, el hecho que hayan asesinado a Jaime Guzmán que fue nuestro fundador, dejó como fijo en el tiempo las virtudes y las características que Jaime quería que tuviera la UDI. Entonces, nosotros invocamos mucho esa línea, esa misión y visión. Entonces, eso es como una especie de columna vertebral que nadie la puede cambiar.” ¿Sigue vigente eso hoy? Hay muchos que dicen que el corte en la UDI no es tanto ideológico sino que generacional… “No, yo creo que obviamente hay una especie de diferenciación entre los que estaban vivos y conocieron a Jaime y los que no. Pero eso no produce, no es una diferencia que sea tensionante. Incluso los más jóvenes son más “guzmanistas” que uno mismo [se ríe]; porque son más idealistas a veces.”

Younger leaders concurred that the party had PA but lacked T, thus they are mainly found at EtPA. This stemmed largely from the fact that, because they were too young, they had not met Jaime Guzmán and they were not involved in politics when he was assassinated, which is the exact opposite from more established elder leaders. It still complies with the rule because, as previously argued in Chapter 3, for this condition this cleavage is irrelevant (i.e. the identity of a party is mostly given by its established leaders). Guzman’s legacy plays a strong role in these younger leaders’ sense of the UDI as a party with Purpose. They thus referred to his thinking as a legacy for the party which they sought to follow. 140

UDI leaders perceive high barriers to exit. This is expected given that the rules of the game—most of all the binominal electoral system and the borders of electoral districts— were designed to preserve Pinochet’s political legacy, favoring rightist political parties (Rahat and Sznajder 1998, Londregan 2000). However, UDI leaders were not unanimous in their perceptions of barriers to exit. Some believed that the party system would change and a new arrangement of political parties would eventually emerge. There was, however, no cleavage in this regard, responses were sufficiently positive to consider Exit present.

In terms of the evolution of Ambition, Gremialismo— like the DC and later the UDI— was mainly formed by young people. Thus, it seems that is the emergence of a new generation of leaders breaking with past political experiences that provides Ambition to parties in Chile (see Luna, Monestier, and Rosenblatt 2013). In contrast with the DC, though, UDI was able to reproduce this trait. Interviewees from different sectors of the party and at different stages of their political careers concurred that the organization provided clear opportunities for those seeking a political career. Such opportunities resulted from a combination of promotion of new leaders by party elites and a clear set of informal rules and stages for advancing one’s political career—regardless of the age at which one becomes involved—establishing a viable path of promotion within the organization. They even have programs for adolescents (Vívelo program). This process was narrated by an established leader in the following exchange:

141

Hoy en día en la UDI, ¿cómo se pasa de ser un militante anónimo a ser líder partidario? “Creo que tiene que tener claro qué es lo que tiene que hacer…Cuando tú vas preparando a ese joven en el “Vívelo”, en los trabajos de invierno, en los de verano, tú vas viendo que tiene algo la persona, cuando van tomando las banderas del liderazgo ahí mismo en los grupos; vas conformando grupos de 10-15 y sabes quien la lleva en ese grupo.” Es fácil ser dirigente de 15 y que la dirigencia de la UDI te reconozca tu liderazgo, ahora ir creciendo para terminar siendo diputado, hay un salto importante. ¿Cómo se va dando el salto entre las etapas? “En política hay dos opciones: tú haces el camino largo, que es el camino real que hace la gran mayoría de la gente, que empieza como dirigente juvenil para pasar a ser dirigente más adulto, pasa a ser concejal, alcalde, pasa a ser diputado y senador; el camino de la política, que es la lógica. Pero en la vida, no sólo en la política, a veces hay que estar en el lugar, en la hora, uno no tiene por qué –y eso es parte de la formación– decir: “mira, yo voy a hacer esto” y tiene que ser tan rígido. La vida no es rígida, la vida te da oportunidades. Lo importante es aprovecharla y estar preparado para aprovecharla.” En la UDI ¿qué significa estar preparado para aprovecharla? “Que todos estos jóvenes, su meta no es ser dirigente mañana. Es ser alguien que el día de mañana lo pueden nombrar y que esté preparado para el cargo que lo puedan nombrar.” ¿Dirías entonces que en la UDI el camino que predomina más es el largo? “No, yo creo que está más en la que nosotros estamos formando gente que esté preparada para asumir en el momento de asumir en el día que se lo nombra…hay gente que está haciendo la carrera…creo que está en la gente que está preparada para servir con los pobres.” ¿Cuál es la diferencia entre una y otra? “La larga es la de los partidos tradicionales, que tú te metes en la política y haces tu vida en función de la política…Aquí, lo que nosotros tratamos de capacitar a la gente es que la política no te da para vivir, nosotros, con nuestros trabajos, podemos convivir con la política. Cuando llega el momento y la oportunidad ahí tienes que dedicarte 100% a la política. Pero, mientras tanto tienes que ser el mejor ingeniero, el mejor abogado, el mejor profesor, el mejor profesional…” Pero entonces, estos “mejores” profesionales, cuando van a la política, ¿no pasaron por las etapas? “Muchos de ellos no.” Pero entonces, el joven que sí viene de esta carrera larga, ¿no siente cierta frustración? “No, el que está adentro tiene más posibilidades, siempre, pero no significa que tenga la exclusividad. Siempre puede haber alguien mejor que esté por fuera. Eso es real.” Entonces, estos jóvenes, ¿llegan? “Sí, las puertas a nadie se le cierran, ellos mismos pueden cerrar y abrir puertas.” 142

One of the most valued rules for attaining greater rewards later in one’s career is that one first must “ir a perder”, running as a candidate in unfavorable districts. Doing so allows one to demonstrate commitment and thereby contribute to reproducing the organizational vibrancy.

Interviews with UDI leaders show that the party elite thinks about incorporating young leaders under an electoral system otherwise favoring incumbents and hindering renovation. This illustrates that, although electoral rules influence how parties organize and survive, party organizations can forge different strategies improving their prospects of dynamic survival.

The party has also promoted independents and thus outsiders as candidates. However, this does not contradict the idea of a clear career path. In fact, many of these independents later become UDI militants (a revision the list of candidates for the House since 1989 confirms this impression).

The consequences of satisfying Ambition through purposive strategies is very straightforward in the case of the UDI: it renovates the party’s elite, projects new leaders, and also renews the party machine. The result is a political party with the youngest cohort of representatives in Congress. The strategy described here, combined with substantial access to material resources for financing campaigns, has been key to UDI’s ongoing electoral growth (Luna forthcoming).

Nevertheless, some tensions have emerged. Recently, there have been some tensions between established party leaders on the one hand and a group of representatives seeking re-election and long-time militants who want to run for Congress on the other (Cristián Letelier, Carlos Recondo, and Carlos Alarcón, among others). These militants have been displaced by established leaders’ decision to give their slots to sons and daughters of the party elite (Juan Antonio Coloma Álamos, Joaquín Lavín, Ignacia Golborne, y José 143

Labbé). Those unable to compete because their slots have been given to relatives of established leaders feel that the party is generating a dynasty that reduces the heterogeneity of the party and does not prize those who have been working with the party for many years. This tension is new for the party, and risks weakening Ambition. However, the party has still been able to avoid a generational cleavage or major conflicts in terms of Ambition. Joignant and Navia (2003) attribute the UDI’s electoral success to the cultural and social homogeneity of its congress members, which has given the party its capacity for unity of action. Though relevant, I am more interested in first explaining the UDI’s organizational evolution and vibrancy. In this sense, the combination of Purpose, Trauma, Ambition, and Exit together explain the ability to develop a vibrant party that later eventually proved electorally successful. Moreover, there is no clear causal relationship between social and cultural homogeneity and the impressive development of the political party organization.129 In this sense, Luna (2010, forthcoming) describes how first Gremialismo and later the UDI were successful in penetrating lower sectors of Chilean society, traditionally niches of the left or the DC. His thorough description and analysis show how my causal conditions were relevant in generating a vibrant partisan structure. In fact, Joignant and Navia indirectly acknowledge the relevance of the “historical” relationship between UDI leaders. They were all formed and politically socialized by Jaime Guzmán in the 1960s; they were thus all indoctrinated with a strong Purpose, suffered Guzmán’s assassination, and developed political careers through the organization. As Navia and Joignant themselves note: …la importancia de esta relación es también histórica y política, en la medida en que los orígenes de la UDI se remontan a la década del 60 precisamente en la PUC, período en el cual el líder fundador del “gremialismo”…y de la UDI, Jaime Guzmán, le proporciona fundamentos doctrinarios…metas políticas, un horizonte histórico y una 129

The PAC in Costa Rica shows the non-necessary relationship between these two conditions, as shown below. 144

identidad específica a un pequeño grupo de dirigentes que poco a poco se institucionalizará, transitando desde la fisonomía incierta del “movimiento” a la forma cada vez más objetivada de un “partido”(Joignant and Navia 2003, p. 158).

Thus, the configuration for this party is ETPA. It is also the only party within which politicians agree that their organization is vibrant and is thus an instance of the dynamic type of party survival.

Conclusion

Throughout this chapter, I have analyzed the configuration of causal conditions in Chile’s political parties. As I have shown, the only dynamically surviving party is the UDI, whose configuration is ETPA.

The other parties are not cases of dynamic survival, but are theoretically instructive in several respects. First, analysis of the PS and DC shows the potentially conflictual relationship between Trauma and Ambition. Second, other parties underline the importance of both coherence and intensity as components of Purpose, and lack of Purpose’s negative impact on party vibrancy. Third, as will become still clearer in the next chapter, these cases illustrate the waning of Trauma’s positive organizational effects over time. Fourth, although Purpose can be reinforced by Trauma occurring after the party has been in government, it can also have the opposite effect. Fifth, that the Chilean cases show that Exit and Ambition are organizationally and not just systemically dependent; these conditions therefore vary across parties even within the same system. Sixth, these cases show that rules have curvilinear effects on barriers to exit. Finally, that the Chilean cases prove that no causal condition, taken individually, is sufficient for dynamic party survival

145

Chapter 5. Costa Rica Introduction

This chapter analyzes my cases in Costa Rica. I analyze the configuration observed within each party, building upon the evidence collected in my fieldwork. Using secondary sources, I briefly sketch the historical evolution of the causal conditions and how they interacted with significant events to bring about the present configurations.

Costa Rica is considered a relative success case of democratic governance and provides another opportunity to assess differences in parties’ survival. It has undergone a process similar to that of Chile, the difference being that Costa Rican parties have been unstable for the last decade. I will focus my analysis on four parties that comply with my rules for case selection: the PLN, PUSC, and the PAC, and (though more briefly) the ML.

146

5.1. General Overview Table 5.1. Costa Rica130

0

Exit

Trauma 0

0

Purpose 0

Ambition

0 1 0

0 0 0

0 0 1

1 0 0

0 1 0

1 0 0

0 0 1

0 1 1

0 1 1 1 0 1 0 1 1 1

1 0 1 0 1 0 1 1 1 1

0 1 0 1 1 1 1 1 0 1

1 0 0 0 0 1 1 0 1 1

Units of Observation PAC (Y, 1), PLN (M, 1), PUSC (Y, M, E, 3), PASE (M, 1)Ex ML (M, 1) PLN (M, E, 2), PAC (Y, M, 5), Ex PLN (E, 1), PUSC (E, 1) PAC (E, 1) PUSC (M, 2), Frente Amplio (Y, 1), ML (Y, M, 2), PAC (M, 1) PUSC (E, 1), PLN (E, 1), Ex PUSC (Y, 1) PLN (M, Y, 5) PAC (E, M, Y, 4), PUSC (M, Y, 2), ML (E, 1), Yunta Escazuceña (M, 1) Ex PLN (E, 1) 0 PUSC (Y, 1), Ex Fuerza Democrática (E, 1) 0 0 PLN (E, M, 3), Frente Amplio (Y, 1) Eva Carazo (FA) PLN (E, 1) PUSC (Y, 1) PLN (E, 1) PLN (E, 1)

It is notable that, in Costa Rica, there is only one politician at the ETPA intersection. Thus, almost no politician sees his or her party as having all the causal conditions jointly sufficient for explaining dynamic party survival and which, as will be commented later, together explain the strongest membership in the set of dynamic party survival. Given the time that has elapsed since the 1948 civil war and the absence of other major

130

In the truth tables for each country, I have not computed the observed outcome by row because inference is not derived from each politician as an individual but as member of a political organization. Finally, to preserve confidentiality, I have included just the following information for each politician: party affiliation; whether the politician was established and renowned (E); whether she was a mid-level leader (M), and if she was a young leader (Y). These are the most relevant attributes given the goals of this study. The scant number of interviewees in each case, precluded including other theoretically relevant traits of each interviewee. Nevertheless, I do also refer to other relevant traits in the course of my analysis where it would not distinguish an individual from a group of others within the sample and thus compromise the condition of anonymity. 147

political conflicts since that crucial event, it is not surprising that few identify T.131 Younger members of the PLN and PUSC and leaders from the PAC generally do not mention the civil war. When they do mention it, the conflict figures as something that the country needs to transcend in order to attend to more relevant issues. Thus, the civil war in Costa Rica is the clearest example of the waning effect of Trauma in all the countries studied here. A leader from the PUSC summarized the waning impact of the 1948 civil war and its impact on party loyalty as follows:

“…hay una generación que ha venido desapareciendo la gente que estuvo vinculada con la guerra civil de 48 era gente que se mantenía más aferrado al concepto de partido, se mantenía más aferrado al concepto de lealtad, elementos que son fundamentales en una sociedad, que estás buscando metas, no como algunos que lo quieren es utilizar a los partidos para ser alguien.”

The next excerpt, from a PLN leader, also illustrates the role of Trauma as a bonding mechanism, while highlighting its waning effect.

131

Finding a good and systematic narration of this critical event is hard. The division permeated all strands of society, including scholars and thinkers. However, the most respected accounts of the events are Bulgarelli (2004), Lehoucq and Molina (2002) and Rojas Bolaños (1979). Although the political environment had already been polarized since the presidency of Calderón Guardia, the events of 1948 were precipitated by the electoral results of the February election. Lehoucq and Molina (2002, pp. 218222) offer a thorough and detailed analysis of that critical election. Although the Tribunal Nacional Electoral declared Otilio Ulate, a journalist and presidential candidate from the opposition, the winner, Calderón demanded Congress annul the election. As Lehoucq and Molina (2002) then explain, ”The decision to annul presidential election results triggered preparations for war even as it accelerated efforts to mediate the increasingly bitter conflict between pro- and antigovernment forces…While P N [Calderón´s party] and PUN [Ulate´s party] leaders negotiated to resolve the political impasse, word reached San José on 12 March that hardliners led by Figueres had attacked government troops sent to discover whether rumors were true that he was assembling an army on his finca (farm).” (pp. 223-224). Thus, the disputed 1948 electoral results and previously exacerbated passions led to the civil war. Opposing Calderón and his alliance with the Church and the Communists were the Centro, José Figueres, and both conservative and nationalist groups of the elite. After six weeks and with two thousand dead, the civil war ended on April, 19, 1948 with the Pacto de la Embajada, an agreement signed by all parties at the Mexican embassy (see Aguilar Bulgarelli´s interview with Benjamín Núñez Vargas in Bulagrelli 2004, pp. 371-90). 148

En la literatura sobre el sistema de partidos de Costa Rica se habla de un proceso de “desalineamiento electoral”. Esta noción refiere a la incapacidad de los partidos de mantener la lealtad de sus votantes y, al mismo tiempo, refiere al alejamiento de los partidos costarricenses respecto a la sociedad. ¿A qué atribuye usted este fenómeno? A mí me parece que es un fenómeno demasiado complejo para darle una respuesta simple, hay muchas categorías, hay mucha gente que se ha visto desplazada porque no conectó con los líderes que tenía en el poder, hay otra gente que se ha desplazado por su visión ideológica que se han ido más a la izquierda, Liberación Nacional ha segregado partidos muchas veces, el mismo PAC está segregado de Liberación y eso hace que se arrastre a la gente en otras direcciones, el menú se amplió. Por otra parte hay un fenómeno histórico que fue decisivo que fue la revolución del 48, fue un acontecimiento tan traumático y tan rico también en posibilidades para el país que provocó unas adhesiones extraordinarias en un país pacífico, en que toda la población se involucre, donde se crearon enemistades enormes, es decir la pasión tan tremenda que se desató duró varias décadas con adhesiones que permitían a cada partido tomarse las libertades más grandes que se pudieran imaginar. Eso se fue atenuando y se acabó la gasolina en algún momento.

Finally, the next quote from an ML leader also highlights both the effect of Trauma as a bonding mechanism and its waning effect.

En un momento alguien te decía muy orgullosamente que militaba en un partido x y con el paso del tiempo, con el envejecimiento de la política con el rol protagónico que van jugando las nuevas generaciones, aquella época donde se nacía marcado por uno y otro partido, en la influencia del 48 con el quiebre institucional, en aquella época habían grandes pasiones, inclusive había consignas de no le compre no le venda porque es de tal partido y la gente de alguna manera lo seguía, se guardaban muchos resentimientos unos y otros porque alguien murió en ese proceso, porque a alguien lo encarcelaron, porque a alguien lo exilaron. Eso que marca parte de la historia de los sistemas de partidos políticos de la Costa Rica contemporánea, desde 1950 hasta mediados de los 80 las familias venían marcadas de nacimiento. Conforme las nuevas generaciones empezaron a sustituir a los abuelos ya se votaba más por la persona, ya no se nacía con ese sello y se veían los liderazgos más personales.

Thus, one would expect the dynamic type of survival in Costa Rica to be constituted at most by EtPA. However, only three politicians are located even in that row: one from the Frente Amplio132 and two from the PLN.

132

Although I conducted some interviews with politicians from the Frente Amplio, it was not included as a case because it did not comply with the case selection rule. 149

As in Chile, there is a substantial presence of politicians (16/51) in the first two rows, i.e., etpa, and etpA. Hence, a significant portion of politicians consider that their parties only function, at best, as electoral machines satisfying individuals’ ambitions. In this sense, they do not see their parties as resting on retrospective or prospective loyalty— Trauma and Purpose—and they do not see high barriers to exit, a perception which could logically facilitate defection. With the exception of the PLN, descriptive accounts of the current Costa Rican party system present an increasingly unstable constellation of parties, and there is almost systemic agreement on the absence of high barriers to exit (with the exception of leaders from the PLN).

This general picture coincides with multiple descriptions of the progressive erosion of the Costa Rican party system, which depict increasing levels of fragmentation, volatility, and electoral abstention compared to the 1983-1998 two-party system era (Alfaro 2006 and 2010, 2010b, Hernández Naranjo 2006,2007, Ramírez Moreira 2010).133 Several authors have described the erosion of the two-party system, which first arose from the competition between PLN and different coalitions of Anti Liberacionistas, later consolidated under a single party, the PUSC (see Rovira Mas 2001, Mora Alfaro 2001, Sánchez 2003, Alfaro Redondo 2006, Hernández Naranjo 2007, 2009).

The rules of the game are more permissive than in the Chilean system for forming political parties. However, Carey (1996) says that the quasi-plurality rule and concurrent elections limit the centrifugal incentives created by PR: “…in systems where presidential and Assembly elections are held on the same date, the parties presenting the top two presidential candidates benefit in Assembly elections as well, despite proportional representation” (Carey 1996, p. 30).134 Moreover, rules have not changed dramatically during this period. New electoral rules do not explain the increasing erosion of party stability and the emergence of new parties. Nor has increasing defection been 133

The same is also evident at the citizen level. See Lapop 2010, 2012. See Also Sánchez (2001).

134

150

simply a consequence of the imprisonment of two former presidents from the PUSC (see below). The two most successful new parties, PAC and ML, both emerged before this scandal erupted. As described by some politicians, there has instead been a gradual process of decay, which I will discuss more thoroughly below. Still, more than explaining the values on each condition, the interest for this study is that those located in these first rows almost unanimously agree that their parties are not vibrant (12/16).

However, there are also a relevant number of politicians at another intersection of conditions, etPA. Thus, prospective loyalty activated through Purpose combined with Ambition is also present. Half of those located in that row are from the PAC.135 However, the absence of perceived high barriers to exit increases dangers of defection. In this case, there is also agreement that their parties are not vibrant. The same happens for those located in the etPa row.

Regarding Purpose, the graph below summarizes the situation of the parties analyzed here.

135

This does not mean that this is the party´s configuration of conditions. 151

Intensity

1

2

3

Coherence Key: 1: PLN 2: PUSC 3: PAC

The length of the line signals degrees of agreement on each component of Purpose. The lengthier the line, the lesser the agreement.

As the graph depicts, no party has Purpose. The PAC is the most coherent, though not significantly so, while the PLN is the most intense, but again at a low level.

A thorough review of La Nación, the most important newspaper in the country, signals that no significant programmatic cleavage divides the Ticos. In fact, when asked about the future dividing line in Costa Rican politics, interviewees’ arguments were not homogeneous. In fact, some claimed that it was likely that the political landscape would be divided between Liberacionistas and Anti-Liberacionistas, though deprived of the substantive content that characterized this cleavage in the past when it divided social democrats from liberals or social Christians. In fact, the only thing uniting AntiLiberacionistas, in their own words, was their desire to remove the PLN from power. Some from the PUSC, Frente Amplio (not included as a case), and PLN are found in combinations: eTPA, ETPa, EtPA. In the EtPA row, there is total absence of leaders from the PUSC or PAC, whether young or established.

152

Most politicians are thus found in the rows etpa, etpA, and etPA. With the exception of the latter configuration, politicians from parties studied here (i.e., excluding the Frente Amplio) do not identify either past or prospective loyalty. As expected, only older, established leaders mentioned T, citing the legacy of the 1948 civil war as a bonding mechanism. In this sense, as T wanes, the organization increasingly becomes an empty vessel unless P emerges. Only the older interviewees who identified T have a sense of unity within the party. If P does not emerge, prospective loyalty is not ensured, and long-term dynamic survival is unlikely. In terms of individual conditions, 30 out of 46136 agreed that their parties satisfied Ambition. PR partially explains this. Costa Rica seems epitomize elite renovation because of term limits (Carey 1996). The prohibition on National Assembly members’ running for immediate reelection brings new faces to politics and forces political parties constantly to look for new candidates. The historical rate of reelection—of course after one term out of the congress—is just 10% (Carey 1996, PELA). However, how are individuals’ ambitions satisfied when there are term limits? Elite renovation in this case could actually run against satisfaction of ambitions.

The central power of party elites resides in their ability to promote careers through other means (Taylor 1992, Carey 1996), putting a brake on term limits’ disruptive effects. If politicians “serve their party well” (p. 85), they enjoy the prospect of being nominated to other offices: “Despite term limits, and the near absence of Assembly careers, Costa Rican legislators look and think like political professionals, not amateurs” (Carey 1996, p. 85). Their fortunes are thus closely tied to the success of their party’s presidential candidate.137 Moreover, presidential candidates strongly influence the otherwise

136

5 had N/A. Carey (1996) is unable to show systematic evidence for this claim (p. 96). However, my qualitative evidence and PELA´s various surveys do support it. 137

153

decentralized system of selecting candidates to form the party lists: “The means by which party leaders are able to retain control over the nomination process…range from the use of patronage, procedural tricks and manipulation, to outright fraud” (Carey 1996, p. 89).138 As shown below, this is most present in the PLN. A complex mechanism of candidate selection established by the electoral code would also help a candidate to pursue a political career (see Alfaro Salas 2003). As previously stated, A is insufficient for dynamic survival. 5.2. Configuration within each Party

PAC: The Partido de Acción Ciudadana (Party of Citizen Action) was founded in the

year 2000. The PAC emerged as a consequence of the protests against liberalization and privatization efforts in the electric and telecommunications sectors (popularly known as the “Combo ICE”).139 The party was first formed and led by Ottón Solís, a former Liberacionista and minister of National Planning and Economic Policy during Oscar Arias’ first presidential term (1986-1990). Just fourteen months later, it obtained an impressive electoral result in the 2002 election, which forced the first second round ever for the presidential election. In the words of Ottón Solís, the PAC presented itself as: “…a new option for all citizens tired of traditional parties and politicians” (In Sánchez 2003, p. 124).140 An established PAC leader described the political environment that facilitated the emergence of the party this way:

138

A more detailed analysis for each party is presented below. The press covered the protests, which were the largest in the country since the 1970s. For example, see: http://wvw.nacion.com/ln_ee/2000/marzo/29/ultima1.html. For a thorough analysis of this process see: VII Informe del Estado de la Nación (Estado de la Nación 2001, pp. 213-220). 140 Translation by the author. 139

154

“Esta, se dice, es una de las democracias más sólidas que dice el imaginario colectivo y puede ser que haya algo de verdad en ello (…) Todo iba muy bien hasta la década de los 80. En la década de los 80 con toda esa reacción conservadora en el mundo con Consenso de Washington; con Reagan en los EEUU y la Thatcher en Inglaterra, y todo el ambiente que se vivía en el ambiente centroamericano. Y una ideología que penetró a los partidos hizo que cada vez el partido Unidad Social Cristiana y Liberación Nacional se unieran y se conformara una sola cúpula de los dos partidos; al punto que los medios de comunicación empezaron a hablar del PLUSC y que ambos se movieran hacia la derecha dejando el social cristianismo y la social democracia. Entonces, si vos me decís ¿qué los caracteriza a esos dos partidos desde el 80 para acá? Es que ideológicamente prácticamente no se diferenciaban. Pero sí se parecieron en cosas terribles que no hubiésemos deseado que se parecieran: en el clientelismo político pero sobre todo en la corrupción. Entonces, los casos sonadísimos de corrupción, la incoherencia entre lo que se dice y lo que se hace. Predicaban como si fueran socialdemócratas o socialcristianos y se comportaban como neoliberales puros ,es lo que ha llevado a que este país esté en la decepción tremenda, en la desesperanza más absoluta y a que se abrieran las posibilidades para el surgimiento de una tercera fuerza política como es el PAC.”

Although it emerged after the protests against the Combo ICE, the founders’ development of a program was coherent but not intense. The reasons for this lack of intensity may have been associated with the historical context (i.e., the relative economic success of the neoliberal turn). Importantly, no epic accompanied the party’s foundation, at least none so pervasive as to constitute a “project” or “model” for the country. The foundation of the PAC was more a reaction to the PLN’s neoliberal turn and the dismantling of the “social democratic matrix”—despite some former PUSC members also joining the PAC—and to increasing accusations of corruption within the two major parties.

The spirit with which the party was founded and the conjuncture in which this occurred are very well depicted in the following quote from an established leader:

155

¿Cómo me involucré en esto de la política? Yo era una más de las apáticas universitarias hacia la política, no que la considerara sucia pero me sentía en mi charco en la universidad estaba muy cómoda, mi hijo me decía que estaba en una zona de confort ¿cómo me salí? Empujada por Don Ottón Solís. Una vez estábamos en el centro de investigaciones de desarrollo sostenible yo era miembro asesor de ese centro de investigación y el director me dijo arriba está Ottón Solís por si queres ir a saludarlo, él sabía que yo lo conocía y subí hablar con Ottón. Y en ese momento él me confesó que se iba a salir del partido Liberación Nacional. Yo me asombré muchísimo porque yo pensaba que él iba a ser candidato dentro de Liberación Nacional. Y me dijo Ottón “pero no me voy a salir para quedarme sin hacer nada, me voy a salir para fundar un nuevo partido político”, y me agarró así de los hombros y me dijo, “¿vos me ayudás?” Y yo retrocedí, casi di un salto atrás, le dije “no Ottón, yo no te puedo ayudar, Ottón, lo mío es la academia.” Entonces él me dijo “porque la gente honesta no se mete a la política, el espacio lo están ocupando quienes lo están ocupando.” Para mí fue como un banderillazo y le dije “bueno, tenés toda la razón, yo te ayudo pero eso si me dejás ayudarte en lo que yo sé hacer; te ayudo con la parte programática.” Vieras que éxito. Bajé al primer piso donde estaba mi reunión y les dije, “dice Ottón que se va a salir de Liberación Nacional para fundar un nuevo partido.” Y ahí de inmediato tres personalidades de la academia de este país dijeron “nos vamos con él.”

This quote reflects the sociocultural origin of the founders of the PAC. The reference to “apathy” is not incidental. It illustrates the pervasive feeling, which was neither one of “commitment” nor one of “anger.” In truth, the country was not undergoing a severe and dramatic crisis, and no major political conflict (associated with my definition of Trauma) predated the organization. Thus, no trauma preceded its formation.

The PAC led the social movement opposing the CAFTA and led the No campaign in the 2007 referendum.141 However, this conflict did not translate into intensity in terms of Purpose. Skeptics agreed that the referendum campaign failed to generate an enduring Purpose, perhaps because they lost the referendum. My interviews revealed no clear agreement concerning Purpose’s coherence. Ethics and transparency in public office was the aspect of programmatic identity most repeated across the interviews, and has also been the one most repeated in the press. This emphasis has gone along with a certain degree of disdain toward “politics as usual.” Such emphasis would at first seem to signal that the party is coherent in terms of Purpose. The PAC remains a group of leaders and militants that is compact in terms of 141

For a detailed description of the referendum, see Vargas Cullell (2008). 156

their socioeconomic and cultural backgrounds. However, it has a very marked, but somehow neglected, internal political cleavage between conservatives and progressives. Throughout the interviews there was also marked disagreement concerning the rapid programmatic transformation of the party, its undefined identity and, in some cases, the neglect of left-of-center ideas by Ottón Solís.142 Leaders who mentioned the party’s original intention to maintain the traditional development model in Costa Rica also highlighted is quick erosion as one of the party’s main ideas. The following quote testifies to the PAC’s problems where Purpose is concerned.

¿Qué es lo que funciona mejor y lo que funciona peor en el PAC, comparado con el resto de los colegas, con el resto de los partidos políticos de Costa Rica, claro? En el PAC tenemos un gran problema, que puede ser un gran beneficio, pero a la hora de llegar o llevar adelante el manejo de una fracción es muy complejo, y es que todos los que estamos ahí, todos muy analíticos, con sus propios ideales, y no tenemos líneas de partido trazadas, entonces cuando se llega a una discusión de tomar posición con respecto a temas… tomar posición es difícil. ¿No hay una identidad PAC, digamos? Es que la identidad del PAC está plasmada en un documento que se llama Convocado de la Ciudadanía, que fue nuestro compromiso de campaña, pero hay temas que ahí no están. De los temas que ahí están, ahí podemos tratar de encasillarnos y, sin embargo, nosotros prometimos una reforma fiscal y hay compañeros que se salen en el momento del acuerdo y dicen sí, pero después dicen no. Pero, digamos, hay temas que no están claros que permiten la libertad… Podríamos decir que el PAC no está claro si es de centro, centroderecha o centroizquierda. El PAC es un partido que no está claro y que tiene, incluso, dentro de su haber ciudadanos de todas denominaciones. Tenemos gente de centro, gente de centroderecha, gente de centroizquierda y tenemos gente de izquierda. Entonces, al haber esa gama tan grande y esa pluralidad dentro del partido, es muy difícil a veces que se nos pueda identificar de forma clara. Entonces depende de las posiciones así la ciudadanía nos ve. Por ejemplo, cuando viene el TLC con Estados Unidos, trataron de ligarnos a la izquierda, porque pareciera ser que es lo más sencillo. Pero digamos ahora que estamos con reforma fiscal, estuvimos con reforma fiscal… una parte mayoritaria de la fracción nos tratan de acercar más a un centroderecha. Entonces, depende de la posición que tomamos nos van a mover hacia afuera según la óptica de quien esté haciendo el análisis.

142

I asked several interviewees whether the ideological transformation of the Liberación Nacional made them the heirs of the “Tico development model,” but their perspectives differed. 157

We cannot conclude from this testimony that Purpose within the party is low. Many parties have different tendencies, ranging from the center to the left (as with the Frente Amplio in Uruguay), or from the center to the right (as with the Partido Colorado in Uruguay or Renovación Nacional in Chile). These internal variations commonly influence parties’ coherence and intensity. However, the PAC’s internal differences go beyond the common tendencies seen in many other political parties. In the case of the PAC, party leaders explicitly and spontaneously said that the party lacked a defined and agreed upon programmatic identity that it coherently and intensely defended. This lack, they said, had become increasingly problematic. Interviewees explicitly maintained that there was no agreement even when it came to defining the party as one committed to defending “ethics in public office”. More progressive politicians believed this was an empty slogan insufficient to unite the party around a cause and enable it to govern. They perceived this appeal as an opposition stance at best. They felt that the PAC was not ready to govern because it was still defining itself negatively in relation to the PLN rather than in more positive and affirmative programmatic terms. The next two quotations shed more light on the PAC’s absence of Purpose and its impact on the party’s vibrancy.

Tenemos un nivel muy bajo de fidelidad somos el partido de todos el que tiene menor fidelidad en las distintas votaciones presidencia, diputados, regidores. Creo que yo que hay un tema de identidad partidaria de un partido que está entrando en la adolescencia pues que creció muy rápido pero todavía ese tema identitario ideológico en algunos casos no se ha terminado de resolver. La misma conformación del partido incide en eso, porque el inicio del partido fue una reacción muy clara a la corrupción que se venía dando, a las políticas neoliberales y se recoge esa sensibilidad y se resume en esos grandes ejes pero hay grandes diferencias en temas polémicos, por ejemplo, en temas donde entra el tema religioso de por medio. Tenemos desde los sectores más progresistas hasta los sectores más conservadores dentro del partido y son temas que cada vez están más en la agenda política. Recogimos una sensibilidad muy heterogénea y eso ha hecho que estemos en los dos congresos ideológicos que hemos tenido en fuertes pugnas por temas que no fueron la génesis misma del partido y nos enfrentamos a un Liberación Nacional que es un partido histórico, una Unidad Social Cristiana disminuida pero es un partido histórico con una identidad muy clara y a un Movimiento Libertario que es un poquito más viejo que nosotros y que tiene una ideología que se ha ido desdibujando pero que en principio era muy clara también.

158

O sea la ciudadanía es Liberación y anti Liberación A mí me parece que sí, que en general es así y además hay que decir una cosa que cuesta entender pero los partidos en Costa Rica son poco ideológicos. Pero el momento de gloria del PAC es el momento más ideológico Absolutamente pero resulta que el líder fundador, esto si tiene que ver con perspectivas de liderazgo, le dio un pánico escénico cuando el partido se convirtió en un partido demasiado progresista de lo que él creía que podía ser. Porque, en realidad, yo creo que su planteamiento político ideológico era un planteamiento de hacer bien lo que Liberación Nacional hacía mal o sea no era un planteamiento de una nueva propuesta ideológica. Era más bien reactivo más que proactivo Exactamente, por eso es que las banderas fundamentales son contra banderas de Liberación Nacional, aunque una las pueda contextualizar como banderas súper modernas efectivamente en su perspectiva son banderas anti Liberación. Pero resulta que quienes llegaron al PAC fue gente muy interesante entonces al final la construcción ideológica que se dio en ese momento ya no sólo tenía la perspectiva de don Ottón, todos aportábamos un poco, las caras de la diversidad aportábamos. (…) “Pero yo siento que como ya nos igualamos para el común de la gente ya no representamos la esperanza, el cambio, la posibilidad de transformar el país, eso quedó muy desdibujado.”

Finally, and more strikingly, the absence of Purpose is also clearly evident in the contemporary discussions of forming political alliances to defeat the PLN.143 Some in the party agree that they should coalesce with the ML, a rightist party often associated with business interests that also explicitly supports privatizations and reduction of state involvement in the economy.

143

See La Nación 11 March 2013 for an analysis of opposition parties’ inability to build a clear opposition to the PLN. 159

Pero, el escenario que vamos ahora ¿no es similar a Liberación vs anti-Liberación? Es que es así la historia. Desgraciadamente, por el momento, mientras no logremos quienes pensamos que Liberación le está haciendo un flaco favor a la democracia, romper estructuralmente esa estructura política partidaria para generar otras, porque lo único que si podemos decir es que a partir del año 2000 Costa Rica tiene una lógica, comillas, más multipartidista en términos de composición, sin embargo en términos absolutos es una lógica bipartidista porque es un partido contra la oposición que puede ser coalición o no. ¿Tú dirías entonces que esa alianza por Costa Rica es un mal necesario? Tiene que pasar, es una cosa que es imprescindible. ¿El PAC por sí solo no podría? Para las elecciones del 14 no podría Pero suena extraño Movimiento Libertario y PAC ¡Ah! No, ¡es absurdo, es absolutamente indigerible! pero para que veas la fuerza de Liberación Nacional cuando una fuerza política puede lograr que fuerzas tan disimiles…Es una locura en términos ideológicos es una locura.

Thus, analyzing the interviews in depth reveals extreme and explicit disagreements among PAC politicians regarding both components of Purpose (i.e., both coherence and intensity). This disagreement, interviewees highlighted, has also been increasing rapidly over the years. The PAC shows how the absence of Trauma curtails the duration of Purpose.

The combination where it is possible to find more members of this party is either in the etpA or etPA row. Thus, in the PAC, there is substantial agreement that the party is able to satisfy individuals’ ambitions. Electoral rules foster the almost automatic emergence of new leaders. Also, the formally decentralized system of candidate nomination and the aggregation of preferences jointly give power to militants and local leaders while stimulating partisan coordination. In the case of the PAC, however, there is also a purposive intention of seeking new leaders. Recently, its founding leader, Ottón Solis, decided not to run for the presidency, generating renewal within the party’s elite. The 160

greatest problem is that the party has not been able to transcend the elite level (middle to high socioeconomic and cultural background) in terms of either votes or their leaders. At the electoral level, this was highlighted by some interviewees and is also analyzed by Alfaro Redondo, who highlights how the PAC only captures votes from the core of the Great Metropolitan Area (Alfaro Redondo 2006, 2010).

There is also substantial agreement regarding the absence of high barriers to exit, which is a point of systemic agreement in Costa Rica. Concurrent elections help reduce the centrifugal effects of PR. However, elections are not fused (as in Uruguay), and the creation of new parties has been a systemic trait for the last decade. Though not the goal of this study, an explanation for low perceived barriers to exit could be given by the absence of Trauma and Purpose, and currently permissive electoral rules might partially explain the evolution of perceptions of barriers to exit, perceptions which actually helped precipitate the creation of the PAC itself.

Thus, the configuration for the PAC is: etpA. PLN: This party has been the most stable in the Costa Rican party system. It has

competed in every election since its foundation in 1951, winning the presidency in nine of the fifteen presidential elections since 1953 (i.e., the 15 held under the reformed constitution of 1949). The decision to form a political party organized by part of the winning coalition of the 1948 civil war proved to be critical to later developments.

The increasingly conflictual 1940s created multi-class coalitions that were ideologically broad. Yashar synthetizes this perfectly: “This period of alliance building led to ideologically opposed political bedfellows.” (1997, p. 81).144 This extended to the civil 144

Increasing tensions and successive coalitions, as well as the increasing polarization in the two opposing camps marked the Calderón Guardia (1940-1944) and Teodoro Picado (1944-1948) governments. Despite the electoral reform of 1946 that created the first Electoral Code, and which sought to defuse tensions in light of pervasive regime crises (Carey 1996, Lehoucq 2002), conflict systematically increased. Tensions 161

war and its aftermath. However, right after the war ended, the Movimiento de Liberación Nacional—which united the opposition against Calderón Guardia—quickly became more ideologically compact.145 The resulting Junta that governed until November 1949—when Otilio Ulate, the alleged winner of the 1948 election, took power— incrementally alienated the conservatives within the Movimiento. The organized economic (mostly coffee) elite quickly distanced itself from social-democratic groups emphasizing social reforms they opposed—as they did during the Calderón government (Bell 1981, Paige 1998).146

Though some intense discussions occurred at the time, the decision to form a political party was settled on September 15, 1950, when Daniel Oduber Quirós, Francisco Orlich, Luis Alberto Monge and José Figueres Ferrer met in Geneva, Switzerland 147 to discuss the foundation of a political party. Oduber Quirós and Monge’s closeness with the

were partially explained by three factors. First, Calderón grew increasingly independent from the rest of the coffee elite. Second, Calderón allied with the Communist Party, which experienced systematic and significant electoral gains during the 40s, generating more concern among the elite (Torres Rivas 2001). And finally, the elite opposed Calderón´s social reforms and the alleged political clientelism channeled through his recently created social security institutions. Immediately before and during the civil war, each coalition of the two opposing camps was ideologically diverse in economic terms. In fact the social democrats’ factions of the opposition did not oppose the economic reforms of Calderón Gaurdia. See Lehoucq (1991) Lehoucq and Molina (2002), Carey (1996), and Yashar (1997). 145 Conservatives coalesced with Calderonistas (politically expressed at the Partido Republicano Calderonista). Thus, they coalesced with Calderon’s previous rival; again, during Calderón Guardia´s government they felt increasingly alienated from his politics. 146 See Paige (1998) for a detailed description of the evolution of the coffee elites´ thinking and influence in modern Costa Rica. 147 They were the four main political leaders of the early years of the PLN. Each eventually became President of Costa Rica (Figueres also led the Junta Fundadora de la Segunda República, which governed for eighteen months after the civil war, from May 1948 to November 1949). José Figueres Ferrer was the caudillo who led the civil insurrection against Calderonismo (see Ameringer 1982, Paige 1998). Francisco Orlich, José Figueres, and Luis Alberto Monge were involved in Acción Demócrata and Partido Social Demócrata in the 1940s. Daniel Oduber Quirós, also participated in those political organizations but, together with Rodrigo Facio, was also a member of the Centro para el Estudio de los Problemas Nacionales. 162

emergence of social democratic parties in post-World War II in Europe148; the previous experience of Acción Demócrata and its influence on the Partido Demócrata; the foundation of the Partido Social Demócrata in 1945; and, most of all, the experience at the Centro para el Estudio de Problemas Nacionales (heretofore, Centro) think tank all influenced the creation of the PLN as a stable, institutionalized party.

Those political organizations, composed of leaders who later formed the PLN, provided the party’s Purpose (interview with PLN leader, Rojas Bolaños 1979, Oduber Quirós 1985, Rovira Mas 1982). Purpose was thus present from the very beginning (Rovira Mas 1982). In this sense, Daniel Oduber Quirós, founder of PLN and President of Costa Rica (1974-1978) summarizes the ideas around which those who formed the Centro in 1945 had gathered:

Lo que fue para los estudiosos de entonces la Revolución Mexicana; lo que fue el "Nuevo Trato" del Presidente Roosevelt; lo que fue el liberalismo colombiano de Enrique Olaya Herrera, de Alfonso López Pumarejo y de Eduardo Santos; lo que fue la gran figura de Víctor Raúl Haya de la Torre, campeando por toda América Latina desde 1928; lo que fue el pensamiento que recogió el Repertorio Americano de don Joaquín García Monge; lo que fue el pensamiento de conferencistas que venían a Costa Rica a exponer sus tesis en los congresos de estudiantes; lo que en los años veinte y treinta significó el pensamiento de Augusto César Sandino para toda el alma nacional... todo esto, más el liberalismo y el laborismo inglés, sumándole el pensamiento social cristiano más el pensamiento social francés y los libros y artículos de socialistas españoles durante la Guerra Civil en España, todo ello se tradujo y se ligó a la gran herencia que Eugenio Rodríguez, con gran sabiduría, asigna a aquel grupo de costarricenses, que sintetizó todas esas corrientes de ideas y que las hicieron desembocar, primero, en la fundación del Centro 148

Daniel Oduber Quirós was living in Paris and Luis Alberto Monge in Geneva. Oduber acknowledged the influence of the European social democratic experience: “Nosotros estudiamos a fondo las democracias europeas. Estábamos en la Segunda Guerra Mundial y nos dábamos cuenta de que los partidos democráticos, permanentes, con alto grado de responsabilidad, tenían una gran importancia para el mejoramiento paulatino del sistema democrático. Y era mucho más lógico pensar en ese tipo de partidos que en los ocasionales. Estas ideas las debatimos, de igual a igual, entre el Centro para el Estudio de Problemas Nacionales y Acción Demócrata, el grupo pensante del cortesismo.” (1985, p. 194). 163

para el Estudio de Problemas Nacionales, después, en el Grupo de Acción Demócrata, más tarde, a partir de 1945, en el Partido Social Demócrata, el cual desde 1951, se expandió hacia lo que es hoy el Partido Liberación Nacional. (1985, p. 17)

This quote reflects the political and philosophical currents that initially informed the PLN’s social-democratic emphasis. The most influential thinker and leader of the social democratic spirit of those days was Rodrigo Facio. His 1941 thesis in law laid the groundwork for the Centro and the social- democratic doctrine (as well as the cooperatism and other major foundations) of Costa Rica’s later social-democratic political organizations, including the PLN (Oduber Quirós 1985).

Rodrigo Facio described the type of political organization that was needed to change the country this way:

...no revolucionaria o de acción directa […] partido ideológico y permanente, tanto por su composición orgánica como por sus objetivos concretos de reforma[…]A su vez, resulta claro que semejante partido constituiría el instrumento adecuado para la adaptación realista y sin violencias al país, de la evolución socialista internacional que está iniciándose en estos momentos. (1943 [quoted in Oduber Quirós 1985, p. 187]) The PLN maintained its social-democratic inspiration for three decades.149 Nevertheless, in the early 1980s the party leaders started to promote economic liberalization and privatization, a long-standing project of the Anti-Liberacionistas (see below for a description of these groups). The process was primarily elite driven, and no process of deliberation accompanied it. It originated in the Monge government (1982-1986). Wilson (1998) describes it as a process that could be considered a case of “policy switch.” Monge campaigned with the PLN´s traditional platform, but while in office 149

For detailed descriptions of the social-democratic era of PLN (and of the Costa Rican development model) see Rovira Mas (1982) and Wilson (1998). 164

implemented austerity measures and structural reforms, which then continued in the following governments.150 The severity of the economic situation forced the dramatic change.151 The geopolitical context of war in the isthmus also played an important role in determining the political and policy outcomes.152 Finally, the recommendations emanating from the International Financial Institutions (IFI´s) also influenced the neoliberal turn.153 However, the actual decision to change has to be traced back to the internal dynamics of the PLN government and party (Wilson 1994, 1998). In fact, in discussing issues related to the change in Purpose, some interviewees quoted a famous 150

In this sense, Wilson says: “The Unidad administration (1978-1982), although committed to significant economic reforms, ultimately found itself expanding rather than reducing the role of the state in response to the worsening economic conditions. The irony of the 1980s was that the initiator and facilitator of the process of dismantling the social democratic development model was the social democratic PLN rather than the more conservative PUSC, the political party that had promised repeatedly to bring about the necessary changes.” (1998, pp. 113-14). 151 Like the rest of the region, Costa Rica suffered severely from the 1973 oil shock. Increasing fiscal deficits sustained the impact of the crisis: “Costa ica financed this burgeoning fiscal deficit by borrowing heavily from abroad. Foreign loans permitted the government to keep up its own consumption and to subsidize state enterprises, the private sector, and consumers.” (Booth 1998, p. 160). 1982 was the worst of those years for the Costa Rcan economy. Debt reached 147% of the GDP, the inflation rate (measured through annual change in consumer prices) was 90% in 1982. GDP per capita declined during the 1980s (see Booth 1998, pp. 158-66 and Seligson and Muller 1987)). It is impossible to know whether such a change would have occurred had no severe crisis afflicted the country. 152 The Reagan administration was exerting a strong influence to overthrow the Sandinistas in Nicaragua. The US rescued the Costa Rican economy with substantial aid in exchange for its help in that endeavor (Paige 1998, Wilson, 1998, Booth 1998, Lehoucq 2012). 153 The country signed three Programas de Ajuste Estructural (PAE) with the IMF, World Bank and other lenders (Booth, 1998, p. 163-64). For a detailed analysis of each PAE, see Raventós (2001). Pérez Brignoli and Baires Martínez summarize the stages of this transformation in the development model: “La primera, entre 1982 y 1990…estuvo centrada en la recuperación del equilibrio macroeconómico, vía el restablecimiento de relaciones con los organismos financieros internacionales…y la disponibilidad de una cuantiosa ayuda externa proporcionada por el gobierno de los Estados Unidos, en el contexto de la crisis centroamericana. Lograda la estabilidad en el corto plazo (disminución de inflación y control del déficit fiscal), se pasó enseguida a renegociar la deuda externa, a liquidar las empresas públicas deficitarias pertenecientes a la Corporación Costarricense de Desarrollo S.A. (CODESA), y a eliminar subsidios a muchos productor agrícolas e industriales. Institucionalmente se preparó el camino para la eliminación del monopolio estatal de los bancos de depósito y se abrieron amplios incentivos para la promoción de las exportaciones ‘no tradicionales.’ La segunda etapa, entre 1990 y el momento actual (2000), continuó con la apertura comercial…incorporación de Costa ica al…GATT, liberó el mercado de cambios y completó la apertura bancaria…” (2001, p. 146). 165

phrase attributed to Daniel Oduber Quirós, who took the social democratic parties in Europe as a model: “The PLN is little more than an electoral machine and little less than a political party” (Primer Seminario Ideológico del PLN, 1967). 154 Together with the pervasiveness of the crisis, this view of the PLN might help to explain its dramatic change in Purpose in such a short period of time. The sudden change in Purpose generated tensions (Cortés Ramos 2001) and a series of defections, the most significant of which occurred only in the 1990s. Since this juncture, the PLN has suffered internal conflicts of Purpose. However, the neoliberal group has predominated (Booth 1998, Lehoucq 2012).155

During my fieldwork some established—and very well-known—politicians agreed that the PLN had Purpose. However, they explicitly diverged quite significantly in their views as to whether the party still had a clear identity. Some said the party lacked a clear and coherent identity, with some saying the party lost it when it promoted structural reforms in the 1990s. Others disagreed, maintaining that the party still had an identifiable set of ideas and worldview, albeit different from those around which the party was formed in 1951.

The main problem in terms of Purpose is not that the party changed its programmatic identity but that it did so without agreement in support of such change on the part of a broad internal majority. Instead, the process was driven by a few elite leaders in the 1980s. It was also a sudden change, albeit one later defended by those believing the party needed to adhere to the structural reforms associated with the so-called Washington consensus. The PLN’s neoliberal turn was thus a divisive issue on which very identifiable center-right and center-left groups, nowadays associated with the

154

Translation by the author. The Arismo fraction (led by Oscar Arias) has dominated the party. However, Mongismo, a more social democratic group, has succeeded in nominating one of its own members (Johnny Araya Monge, mayor of San José for 20 years) as the party´s presidential candidate for the 2014 national elections. 155

166

“Monge” and the “Arias” factions, were opposed. Its lasting consequences included increasing distance between unions and the party (interviews with PLN leaders). Given the party’s origin, it is clear that it was born with Trauma. However, because more than sixty years have elapsed since the party’s foundation, only a few of the older interviewees still made references to the 1948 civil war as a unifying factor. Memories of the civil war do still permeate the party and influence the party’s identity, but not in terms of Trauma as defined in this study.

All but two of the PLN politicians indicated that Ambition was present. The sense that the party acts as an efficient channel for ambition can be traced back to the origin of the party. Ameringer (1982) suggests, and I concur, that the fact that power was distributed between four leaders (Monge, Figueres, Orlich, and Oduber Quirós) instituted the idea of power dispersion and the possibility of accessing the elite through different “promoters.”156 As one would normally expect, the party machinery became increasingly dense and institutionalized over time. This solidification went with increasing recognition of the presence of those channels and thus of the condition of the party’s capacity for effectively furthering individuals’ political ambitions. Interviewees of different generations and different sectors of the party identified the same process that someone has to undergo to pursue a successful political career. Moreover, they agreed that it was possible to be successful, while one’s relationship with the party elite was relevant, one’s success was not determined only by a particular leader or small group of leaders.

156

This was also highlighted by established, older politicians with deep knowledge of the functioning of the party in its early days. 167

Más específicamente, pensando en su partido ¿cuáles son los mecanismos que existen hoy día en su partido para hacer una carrera política? Creo que concurriendo a las actividades propias del partido para lo cual hay que estar informando. Dos, decidiendo “quiero.” Tres haciendo lo correspondiente para que otros entiendan, asuman y compartan ese quiero. Se puede hacer desde la base comunal o metiéndose en la estructura partidaria. Se puede optar por la base comunal es el camino más largo, más difícil, más estable porque como somos un partido que empieza en la base geográfica tiene esa posibilidad de subir esa escalera. Es una selección política dura muy sufrida pero se puede hacer.

They also agreed that there were high barriers to exit. However, this agreement was not as marked as that concerning ambition. As in the case of the UDI in Chile, that of the PLN illustrates the importance of deepening the analysis of perceived barriers to Exit. Politicians from the PLN saw the system very differently from leaders of other parties. In this case, leaders perceived the party system to be undergoing a period of increasing instability. They believed their own party would remain stable, but were less sanguine about the future of other parties. Moreover, regardless of the possible occurrence of a shock as a consequence of a severe economic crisis, they believed their party would remain. The main reason party leaders gave for the PLN’s present and expected future stability and the costs associated with defection were its history and the rules of the game, particularly closed lists. These findings for the PLN thus illustrate the interactive nature of the history of an organization and rules as components of Exit, i.e. in accounting for perceptions of to the costs of exit as compared to benefits of defection.

Thus, the configuration for this party is: EtpA. This case shows that this configuration is sufficient for dynamic party survival. However, as I will analyze in Chapters 6 and 7, it has the lowest degree of membership in the set of dynamically surviving parties. PUSC: The PUSC is the most organizationally complex party of those in the post-civil

war opposition camp that successively formed since 1953. It grouped the war’s various losers: Calderonistas, and a group of conservatives distancing themselves from the

168

social-democrats during Figueres’ eighteen months as head of the Junta.157 Prior to the PUSC’s formation, these forces were mainly grouped and known as AntiLiberacionistas. A change in rules facilitated this group’s more institutionalized expression as a party. The PUSC was founded after a political agreement between Rafael Ángel Calderón Fournier and President Monge.

The formalization and institutionalization of the coalition was possible thanks to a modification of articles 54 and 62 of the Electoral Code. These changes permitted parties forming the PUSC to receive funds based on their respective electoral results in the previous election.158 Thus, a change in the rules of the game to facilitate partisan organization helped channel the incentives of a group of leaders willing to dispute the

157

This complex grouping of coalitions was more easily resolved in the Liberación movement (the coalition that fought against Calderón). The losers (Calderonistas), and the conservatives that later joined them, had a hard time transcending their Anti-Liberación stance. It was not until the 1962 presidential election that all relevant political organizations competed. Calderón Guardia was exiled in Mexico until 1958, and he competed only in the 1962 election. The inability to unite opposition forces against the PLN in the 1962 election led to the first coalition attempt in 1966, the Unificación Nacional. It was mainly an alliance between the Partido Unión Nacional and the Partido Republicano Calderonista. In subsequent elections, other organizations joined this coalition, though each party kept its organizational autonomy and identity (Rovira Mas 2001). The coalition then became Coalición Unidad for the 1978 election (See Hernández 1998). Ameringer (1982, pp. 64-5) describes the lack of formal organization of the Coalición Unidad, which included other small parties but most importantly, again, the Partido Unión Nacional and Republicano Calderonista. Rodrigo Carazo Odio (a former PLN member) was nominated as presidential candidate for the 1978 election. This election was the first time the PLN also lost the majority in the National Assembly (Hernández Naranjo 2001). The dramatic failure of Odio’s administration provided the final necessary incentive for pursuing a different type of organization. 158 See Salazar (1990) for a description of the evolution of the electoral rules. See Rovira Mas (1987, 1990, 2001) for a description of the specifics of the negotiations and the change in rules. As in 1946, when some of the elites foreseeing increasing systemic risk adopted reforms of the electoral code (see above), in 1982 they again agreed to change the rules (Cortés Ramos 2001). They were motivated by a context of severe economic crisis and regional war. In reaction to these twin crises, leaders consolidated a twoparty system. This benefited the Anti Liberación groups. In return, Monge obtained their support for the structural reform agreements with the IFIs (Calderón Fournier 2005, Rovira Mas 1987, 2001). This might have prevented increasing systemic problems, mostly in terms of Purpose—together with the waning of Trauma—which increasingly risked diminishing the vibrancy of those two parties, especially the PUSC. 169

PLN’s power and confront it with a different set of ideas: a mix of liberalism and social christianism.

Calderón Fournier was the son of the 1940s caudillo Calderón Guardia. He led and represented Calderonismo. The social reforms of his father’s government, which were not dismantled by the PLN, had left a lasting imprint on the country. He was the leader that made possible the formal unification of divergent groups. Calderón Fournier was certainly marked by Trauma. However, given the very different origins of the various groups initially founding the party, Trauma did not serve as a bonding mechanism. Trauma could have helped to unite the Calderonistas, but not the other groups which, for the 1948 civil war, were in the opposition to Calderonismo. As with previous coalition attempts, the PUSC was first and foremost united through its shared opposition to the PLN.

The other components of the party, with origins in the economic elite and liberal groups, also influenced the direction of the party. The liberal ideas gradually became dominant, and the PUSC thus became the most coherent of all the forces of Anti-Liberacionistas (Fernández 1996, Furlong 2001). However, the social ideas of Calderonismo also present (Wilson 1998 and Furlong 2001), and authors agree that the party never had a strong Purpose. Again, it was changes in Exit and the presence of a strong caudillo, not Trauma or Purpose, that facilitated the party’s unification.

Los partidos grandes que han sido Liberación y la Unidad, han tenido de todo. Han tenido desde extrema derecha a extrema, casi izquierda, con sindicatos, sindicalistas, porque han representado las diversas visiones del país dentro del partido curiosamente. Es una realidad. Me acuerdo cuando estábamos haciendo la Unidad Social Cristiana, alguien decía, un connotado decía, el partido Unidad Social Cristiana será grande cuando detrás de una huelga de sindicalistas vayan metiendo un Mercedes Benz apoyando. Es así. Así es. Liberación Nacional tiene de todo. La Unidad Social Cristiana tenía de todo.

Rojas Bolaños and Sojo (1995), Fernández (1996), Rovira Mas (2001), Cortés Ramos (2001) argue precisely that there has been a process of ideological “mimesis” between 170

the PLN and the PUSC. This was a consequence more of the changes that occurred in the PLN (see above) than in the PUSC. Moreover, Fernandez highlights the “decreasing relevance of their doctrinal references” (p. 155).159 The PLN moved to the right, occupying the PUSC’s ideological space. Rovira Mas provides two reasons: first, the centripetal tradition of Costa Rican society and, second, its reinforcement by the changes in rules that favored the creation PUSC’s creation. The Pacto Figueres-Calderón of 1995, in which the sons of the two caudillos agreed on a series of reforms, was one of the clearest illustrations of the ideological proximity of the two parties and the PLN’s change in Purpose (Figueres Olsen campaigned on modifying the pattern of neoliberal structural reforms underway in the country).

From my fieldwork, the PUSC emerges as the party with the greatest internal disagreements. This is to be expected, because this party has almost collapsed. After having been one of the two principal parties of the Costa Rican bipartisan era (19831998), its electoral relevance was dramatically reduced to almost a testimonial presence. The 2004 imprisonment of its two most important leaders of the last decades, former Presidents Rafael Ángel Calderón and Miguel Ángel Rodríguez, both of whom were accused of corruption (Lehoucq 2005), first and foremost affected its internal cohesion. In fact, the only area of almost unanimous agreement, as was to be expected given the current situation of the party, concerned the absence of high barriers to exit. Calderón was the leader who enabled the cohesion of the four parties that united under the PUSC. Calderón´s demise was very influential in the atomization of the party, which had never been united by either Trauma or Purpose. Some leaders within the party considered that Rafael Ángel Calderón had used and co-opted the party for his own purposes, mostly to protect himself politically (interview with party leader), and had in the process undermined the party’s ability to satisfy ambition through renovation and promotion of

159

Translation by the author. 171

new leaders.160 Although rules in Costa Rica make parties more likely to satisfy my condition of Ambition, the organizational decisions of leaders also influence its satisfaction. PUSC leaders depicted a party in which the two most important leaders had exerted extreme influence on other members’ chances of using the party as a vehicle for successfully pursuing a political career.

Moreover, non-established leaders mentioned that education, family ties and economic resources explained the chances of an individual succeeding in his political career. Not partisan organization but individual assets thus explained the chances of growing one’s political career. The following quote from a PUSC leader illustrates this perception. Although he mentioned that, given his party’s present situation, almost anyone could try to advance his or her political career, individual assets were nonetheless ultimately necessary for succeeding. Theoretically, he thus distinguished between the chances of young leaders advancing careers within the party and the party’s contribution to such advancement. For this study, only the latter is regarded as a mechanism stimulating party survival over time.

160

As throughout the study, for the current analysis absence or presence of conditions was determined by leaders´ perceptions. 172

Si tuviera que sintetizar aquello que explica su trayectoria personal en la interna de su partido y en la política, ¿a qué factores lo atribuiría? Para ser sincero si mi partido estuviera bien yo nunca hubiera tenido esa oportunidad. Lo primero es no soy de una familia multimillonaria, no tengo un apellido que te dice nada, tengo una formación académica normal, no tengo un doctorado en Harvard ni Stanford, la circunstancia me permitió estar en donde estoy, la perseverancia, la formación y el deseo de querer cambiar las cosas. Hablando en general del proceso de crecimiento político, ¿cuáles son los factores que explican, en su criterio, las posibilidades de una persona de crecer políticamente? Es decir, ¿cómo se pasa de ser un militante anónimo, o sencillamente un ciudadano anónimo, a ser líder partidario? Lo primero es el deseo de hacerlo, lo segundo es creérsela, lo tercero es tener cierta estabilidad económica para hacer que las cosas pasen, si tienes el deseo, te la crees y tienes estabilidad para poder invertirle sin duda las cosas van a pasar. ¿Cuál es la trayectoria habitual de una persona que crece políticamente? Es muy variable, en los últimos años la situación del país y de la vida política ha sido tan terrible no sé si antes lo era pero aquí lamentablemente no necesariamente llegan los que tienen que llegar. Y ¿por qué no llegan? Hay mucha gente que ha hecho esfuerzos enormes y ha luchado y no lo ha logrado, por las circunstancias y porque no tenían un padrino. Y hoy no, hoy es diferente estamos en el otro extremo, hoy llega cualquiera.

Paradoxically, the only condition whose presence my interviewees disputed was that of Purpose. Most of those who considered that the party did have Purpose were young or middle aged. This is an interesting and quite surprising finding. The fact that young and middle-aged leaders felt that the party did have a defined Purpose, despite the scandals involving its main leaders, suggests that it might be possible for the party to overcome these recent setbacks. More specifically, the ideas they associate with the Purpose of the PUSC are more associated with social-christianism than neoliberalism. Still, disagreements among interviewees regarding the party’s Purpose were pervasive. Not only it does not comply with the basic rule for considering this condition present, but those who did not indicate that purpose was present were emphatic concerning its absence, and they mentioned explicit problems in terms of either coherence or intensity. This dispersion of responses could be traceable to the diverse origin of its founding 173

members, which still marks the organization. A similar pattern emerged for the rest of the conditions, with the exception of agreement on the absence of barriers to exit. Those who did not indicate that a given condition was present tended to be very emphatic about its absence. For instance, in the case of Ambition, they explicitly highlighted problems of elite renovation (see above).

The following quote associates lack of vibrancy with the absence of Purpose and Ambition.161

“A mí me parece que en Costa Rica los partidos políticos mayoritarios, cuando teníamos el bipartidismo, se convirtieron en cúpulas nacionales. El alejamiento de lo que fue, tanto en Liberación como de la Unidad de la participación de sectores, creo, pasó a un segundo y tercer plano. Todos los movimientos juveniles, incluso universitarios, incluso sindicales, solidaristas, todos los movimientos empresariales, cooperativistas. Cuando teníamos partidos con estructuras muchísimo más participativas, yo creo que ese desarraigo no se sentía tanto. En el momento en que los partidos dejan esos procesos de participación y empiezan a convertirse en cúpulas donde dos o tres tomaban las decisiones o incluso, entre ellos se ponían de acuerdo para tomar las decisiones, me imagino que te han hablado por allí del llamado PLUSC, que después dio como surgimiento un partido como el PAC, que más bien el PAC ingresa con fuerza porque empieza a darle participación a sus sectores; el problema que no lo continuó. Me parece que esos procesos de participación; llámele grupos de presión, o grupos organizados de la sociedad civil pasaron a un tercer plano.”

Thus, the configuration is etpa. This case highlights that my argument is one of sufficiency. My conditions do not provide an exhaustive explanation of a given outcome—in this case lack of dynamic survival. Other variables can also come into play, in this case corruption scandals, which have undermined the party and which could be seen as a sign of absence of Purpose (see Chapter 7). ML: The party was founded in 1997. It was formed by a group of liberals previously

close to Miguel Ángel Rodríguez in the PUSC. The ML first participated in national elections in 1998, obtaining one seat in the National Assembly. Thereafter, it quickly gained greater electoral relevance. It was the first of several new parties breaking the bipartisan dynamic between the PLN and PUSC (or former Anti-Liberación coalitions). 161

This same interviewee previously referred to the waning of Trauma. 174

The ML has now succeeded in surviving for a decade and a half. Its leaders acknowledge, however, that the party is extremely weak and has not succeeded in building an organization transcending electoral cycles.

I interviewed only a small number of politicians from the ML. The content of those interviews clearly suggested that the party has an identified Purpose. This finding is illustrated in the following quotation from an ML leader.

Vos podés preguntarle a cualquier partido político y a cualquier analista político de Costa Rica y en términos generales te dirán que el Movimiento Libertario es un partido que se ha mantenido consistente a lo largo del tiempo, es un partido que siempre te va a poner peros cuando alguien impulsa impuestos. Es un partido que procura simplificar trámites, desregular la economía, es un partido que defiende libertades individuales, defendemos que cada quien tenga sus preferencias sexuales, credo religioso, el tema de la libertad de expresión etc. Somos un partido que promueve la competencia en todas las actividades todas las actividades económicas, luchamos a muerte contra el monopolio público, contra la carterización de la economía, somos el partido que apoya el comercio libre y entonces desde ese punto de vista buscamos que a nivel de políticas públicas nuestro país se inserte en la economía global, promovemos la libre movilidad de los factores de producción, somos un partido liberal, defendemos una sociedad de emprendedores y un Estado mucho más chico fuerte en las cosas que tiene que ser fuerte como en el tema de la administración de justicia, una justicia pronta, imparcial, técnica, eficiente, barata. Luego por otro lado en el tema policial indudablemente garantizar el Estado de Derecho y luego empezamos a meter algunos conceptos un poco más social cristianos, si se quiere, como el principio de la acción subsidiaria en las acciones o actividades del Estado, lo que pueda hacer el individuo lo haga el individuo y si no se hace solo en ese caso se justificaría algún tipo de participación del Estado pero siempre promoviendo que el sector privado y los individuos se metan en eso.

Nevertheless, this Purpose rests on one leader, Otto Guevara, who gave the party its libertarian identity. It is very difficult to assess whether this Purpose will outlast him.

Estamos convocando en junio una asamblea nacional donde pretendemos aprobar en el artículo 3 de nuestros estatutos la parte ideológica programática porque de alguna forma desde la existencia del partido hasta hoy la ideología o las propuestas del partido es lo que Otto piense, y ha permeado a un grupo de personas, pero no estaba consignado en ningún documento formal del partido esa identidad.

The case of the ML is thus relevant because it shows that Purpose, taken individually, is insufficient to explain party survival. As this case illustrates, Purpose can actually rest narrowly on the well-defined ideas of a small group of leaders—or, more extremely, a 175

single leader. The following quote from an ML leader accordingly shows the absence of party vibrancy and its excessive reliance on the leader.

¿Dónde dirías que funciona el partido de manera cotidiana más allá de aquí [sede partidaria]? Más allá de aquí [sede del partido] y de la representación parlamentaria que tenemos, es en reunión en bancada y el presidente del comité ejecutivo nacional, aparte de eso no hay nada más. ¿Cómo se articula el funcionamiento del partido tomando en cuenta la importancia de los individuos? …hay un mea culpa en el sentido de que esto no está funcionando como debiera funcionar porque no hay una cantidad de personas que estén dispuestas a asumir responsabilidades independientemente de que Otto esté o no esté. A nivel de organización partidaria esto gira mucho en torno de Otto Guevara. (…) En general, ¿considera que la organización partidaria funciona de manera frecuente? No.

Regarding Ambition, leaders mentioned that, as for the party´s Purpose, it relies too much on its founding leader, upon whom elite promotion has been entirely dependent. Thus, the party complies with my case selection rules, but it is still too reliant on one leader. Hence, the preliminary configuration for this party is: etPa.

A brief comment on the left in Costa Rica

I interviewed some leftist leaders in Costa Rica and spoke with individuals from different generations, traditions, and political roles. Costa Rica is one of the few countries in Latin America where there is no significant party or organization that could be labeled “leftist.” I discussed this fact with my interviewees. No clear explanations for this peculiarity of Costa Rica emerged. However, their interpretation was not far from some available scholarly reflections. On the one hand, a review of the literature indicates that the Communist party from the very beginning generated a “minimal” program, and

176

it quickly agreed with the Church and Calderón, subsequently renouncing to its Communist label (Paige 1998, pp. 133-140).

Molina Jiménez (2007) carefully analyzes the inroads of anti-communism in Costa Rica. In this sense, Orlich (1962) once said:

A la Democracia la tiene paralizada el temor al comunismo, porque los demócratas no nos hemos resuelto a ensayarla en todos sus aspectos. Según un filósofo francés son tres los únicos métodos posibles de acabar con el comunismo. Por la violencia, encarcelando a todos los comunistas. Por la convicción, ilustrando sus mentes en tal forma que comprendan, sus errores. Por la superación, eliminando la injusticia. Ahora bien, lo primero no es humano. Lo segundo no es corrientemente posible. Queda pues sólo el tercer método, sin desmayar un momento, desde luego, en la vigilancia sobre quienes en alguna forma quieren destruir la libertad y las instituciones democráticas de nuestro país. (quoted in Rovira Mas 1982, p. 92).

With approval of the 1949 constitutional reform, the communist party was proscribed. Article 98 of the Constitution prohibited the existence of political organizations defying the democratic order. Although leftist parties participated in elections, this prohibition greatly limited the left. This article was only reformed in 1975 (Rovira Mas 2001). Moreover, in terms of quality of procedures and policies, the PLN and the “Tico model” of democracy were successful. The successive failure or conflictual nature of other leftist experiences in Central America also helped create a sense of needing to protect the peace of the country, which many regarded as a paradise of democracy in a politically convulsed region (interview with leaders from the left and Rojas Bolaños and Sojo 1995). Thus, a self-reinforcing process generated enormous barriers to establishing a politically competitive, leftist project.

177

Conclusion

In this chapter I have analyzed parties in Costa Rica. There is almost systemic agreement on the absence of high barriers to exit. The only exception is the PLN. Variation among Costa Rican parties in this regard again attests to the theoretical importance of looking beyond rules alone in assessing Exit. Given that more than sixty years have elapsed since the occurrence of the civil war, Costa Rica also provides the clearest example of the waning effect of Trauma.

The Costa Rican cases of the ML and PAC also show the insufficiency of individual conditions taken alone for ensuring dynamic party survival. The PAC case of also highlights the relevance of surveying both coherence and intensity as components of Purpose and the probable role of absence of Trauma in reducing Purpose’s duration. The PLN is Costa Rica’s only dynamically surviving party. This case shows the waning effect of Trauma and how loss of Purpose saps party vibrancy. It also suggests that, once lost, Purpose may never be recovered—a subject of further analysis in the next two chapters. Nevertheless, I have shown how the causal combination EtpA is a sufficient for permitting dynamic party survival. As further discusses Chapters 6 and 7, though, EtpA is the weakest of all the configurations accounting for dynamic survival.

178

Chapter 6. Uruguay Introduction

Uruguay differs from Chile and Costa Rica in that its parties are all dynamically surviving. There are, however, interesting differences between the Partido Colorado and Partido Nacional, on the one hand, and the Frente Amplio, on the other.

As with Chile and Costa Rica, I first present an overview, analyzing systemic agreements and those among politicians (my units of observation). I then analyze the configuration observed within each party, building upon the evidence collected in my fieldwork. Using secondary sources, I also sketch the evolution of the causal conditions and how they have interacted with significant events to bring about the present configurations.

The analysis of Uruguayan parties will enrich the analysis because the three cases are dynamically surviving. Also, given that the Partido Colorado and Partido Nacional have more than 170 years, they allow for a broader analysis of the evolution of my causal conditions. Finally, the case of the Frente Amplio introduces another sufficient configuration of conditions sufficient for dynamic party survival.

179

6.1. General Overview Table 6.1. Uruguay162 Exit 0

Trauma 0

Purpose

Ambition

Units of Observation

0

0

0

0

0

0

1

PN (Y,1)

1

0

0

0

PN (E,1), PC (E,1)

0

0

1

0

0

0

1

0

0

FA (M, 2)

1

0

0

1

PC (Y, M, E, 8), PN (Y, M, E, 7)

0

0

1

1

PN (E, M, 2)

0

1

0

1

0

1

0

1

0

FA (Y, 2)

1

1

0

0

FA (E, 1)

1

0

1

0

FA (M, 1)

0

1

1

0

FA (M, 1)

1

0

1

1

FA (Y, M, 2), PN (Y, M, 2), PC (Y, 1)

0

1

1

1

0

1

1

1

0

FA (M, E, 4)

1

1

0

1

FA (Y, M, E, 3)

1

1

1

1

FA (M, E, 3)

In Uruguay, the situation is the opposite of what I found in Costa Rica and, to some degree, of that observed in Chile. First, the etpa row is empty, and only one politician is located in the etpA row.

162

In the truth tables for each country, I have not computed the observed outcome by row because inference is not derived from each politician as an individual but as member of a political organization. Finally, to preserve confidentiality, I have included just the following information for each politician: party affiliation; whether the politician was established and renowned (E); whether she was a mid-level leader (M), and if she was a young leader (Y). These are the most relevant attributes given the goals of this study. The scant number of interviewees in each case, precluded including other theoretically relevant traits of each interviewee. Nevertheless, I do also refer to other relevant traits in the course of my analysis where it would not distinguish an individual from a group of others within the sample and thus compromise the condition of anonymity. 180

The Uruguayan party system has the oldest parties in the region, the Partido Colorado and the Partido Nacional.163 For more than a century, the country had a stable two-party system (Caetano and Rilla 1984, Caetano, Rilla, and Pérez 1987, Pérez Antón 1984, 1988, Buquet 2003). Both parties, and later also the Frente Amplio, are essential components of Uruguayan politics. Not only have they lasted, but they have also remained vibrant organizations (though some variation might be observed if deeper analysis were conducted on their respective historical evolution). The combination of survival and vibrancy, i.e., the dynamic nature of the PC and PN, is observed through the combination of E and A, a configuration sufficient to ensure dynamic survival.

Stability has been ensured by a combination of rules: PR (since 1918), closed lists, concurrent executive and legislative elections (since 1934), and the double simultaneous vote (since 1910).164 The relevance of the Uruguayan set of rules in determining the presence of Exit (and Ambition, see below) is undisputed among scholars and my interviews. As González (1995) incisively explains these rules and their effects as follows:

The DSV means that voters vote at the same time for a party and for a specific set of candidates within the party, though they have to select lists–which cannot be modified—from among those presented by the rival factions within each party. Once they choose the party, they have as many options as there are lists presented by the party. An obvious consequence of voting for a party is that split tickets are not allowed. The party that obtains a plurality of votes is the winner; the winner of a plurality within that party becomes president. Since 1934, finally, all elections occur at the same time. This simultaneity obviously reinforced the effects of the ban on split tickets. In such a system, whatever the importance of other elected posts, the most important is the presidency… Simultaneous elections block PR’s central effect on the party system, for the ban on split tickets imposed by the DSV assures that the plurality rule 163

For this study, I use the terms Partido Nacional and Blanco indiscriminately. The DVS allows intraparty preference. Piñeiro (2007) and Buquet et al (1998) summarize the logic and incentives generated by this set of rules. Piñeiro (2007) also describes the change in incentives introduced with the 1996 Constitutional reform. 164

181

in the top race—the presidency—dominates electoral logic…the DSV provides the stimulus and is reinforced by PR, which assures that it is not necessary to get the first prize to stay in the race; a modest electoral success is enough to provide some important posts for the top leadership of the faction or subfaction, which in turn may “trickle down” to their lesser associates. As a result, this system continuously creates new challengers to the established leaders who, if moderately successful, will become leaders of new factions or subfactions. In organizational terms the crucial point is that the system allows the would-be leader to skip one step in the party hierarchy, opening a parallel, competitive candidacy or set of candidacies” (González 1995, pp.146-7). In terms of Cox and McCubbins’ (2001) distinction between strong and feeble electoral systems, the Uruguayan electoral rules correspond to the strong type. PR and the DSV alone would lead one to classify it as feeble. Uruguayan parties have the singularity that each has institutionalized factions (mostly established since 1934), and the number of electoral lists is therefore immense (see Monestier 1999 for a thorough study of the increasing number of lists at the representative level). Some scholars have argued that given the pervasiveness of party factions, their political relevance, and the apparent nonexistence of organization at the party level, Uruguay historically had a multi-party system with a two-party façade (Lindahl 1962, Sartori 1976). However, most Uruguayan scholars have provided more compelling evidence to the contrary (see, e.g., González 1995, Mieres 1992, Buquet and Chasquetti 2004b).165 Buquet et al. (1998) show that the DSV has never produced an atomization of factions, as otherwise expected. Through the analysis of the effective number of Senate lists166, the authors show that the effective

165

The exceptions are Solari (1991) who speaks of a “bipartidismo aparente y un pluripartidismo real” (p. 122) and ial, who sometimes refers to each party as a “federation of parties” or “confederation of factions.” For instance, ial says: “Se trata de dos grandes partidos, el blanco y el Colorado, que en realidad constituyen confederaciones de fracciones partidarias casi totalmente autónomas, unidas para la acción electoral, únicamente” (Rial 1984, p. 32). 166 They analyze Senate, rather than House, lists because the former have national circumscriptions, and politics in Uruguay are national. Moreover, the Senate lists cut parties vertically (Buquet et al. 1998, p. 41). 182

number of factions has traditionally been between two and three (p. 44). 167 Moreover, the effects of these features of the Uruguayan system are offset by its closed lists and concurrent and fused executive and legislative elections.168 Thanks to these features, the electoral system is not feeble but strong. These rules generate incentives to concentrate votes among few parties, thereby reducing the centrifugal effects of PR. This is why Uruguay should be classified as a strong electoral system.

The following quote reflects the important role of the specific institutional design in explaining the trajectory of partisan stability in Uruguay.

167

Moreover, given that presidential and legislative elections are fused and concurrent, the former had an “ordering effect” on the number of Senate lists. 168 The introduction of ballotage in the 1996 constitutional reform did, however, introduce some variations. 183

“Primero, el PC y el PN, llamados partidos fundacionales, creo que son los más antiguos de América, de los que aún existen. Son partidos que por su origen como que cortan la sociedad y cortan las ideologías…Los partidos del Uruguay, yo creo que primero todo el andamiaje electoral y demás está armado sobre los partidos, en un sólido y cerrado sistema de lema, para usar el término constitucional, muy cerrado que reivindica y fortalece a los partidos, lo cual se puede ver en Chile, la Unión Cívica Radical en Argentina, donde se pueden ver partidos que tratan de estructurarse como tales, más allá de sus momentos mejores y peores.” Pero, ¿qué es lo que explicaría a su criterio que esa historia permanezca viva, que se pueda reproducir, más allá de esta raigambre histórica? “Hay razones sociológicas pero fundamentalmente hay razones también del esquema electoral y partidario. El esquema constitucional y legal no facilita el desmembramiento, no facilita las coaliciones ocasionales, cosa que por ejemplo en Italia hasta la última reforma era una cosa maravillosa. ¿Por qué razón? Todavía no está estudiado del todo los efectos que ha generado la reforma del ballotage, que ha generado, está generando y seguirá generando cambios hacia dentro de la sociedad y de los partidarios. Hay detalles bastante técnicos, por ejemplo: los lemas permanentes pueden acumular por sublemas, y los no permanentes, no. Eso que parece casi griego para el resto de la población, es una de las llaves del sistema. Vamos a usar un ejemplo histórico: el FA, cuando se crea, le pidió prestado el lema al Partido Demócrata Cristiano porque sino no podía tener distintas listas al senado, lo cual hacía incompatible; porque el FA desde que se crea va al mismo sistema de los partidos tradicionales, es decir una oferta policlasista, polideológica, que es como un supermercado y parte del problema que tienen hoy los partidos tradicionales es que han dejado de ser supermercados y el FA ha ampliado sus estanterías. El sistema de partidos le da estabilidad a la democracia, combate los personalismos, que es uno de los grandes problemas de América Latina. El sistema de partidos fortalece a la democracia, con todos sus vicios pero la fortalece. Por lo tanto, esa es mi manera de ver la esencia. Si tú ves las democracias maduras, la madre de todas, Inglaterra, tiene un sistema de dos partidos y medio desde mucho antes de que se fundara Montevideo. Pero también es su sistema electoral lo que le da estabilidad. El sistema por circunscripción desalienta el desmembramiento, porque si tú te desmiembras yo te gano. (…) Entonces acota y le da seriedad a la discusión (…) Lo mejor contra el populismo generalmente es un sistema sólido de partidos. En el fondo, por más que haya grandes líderes, despersonaliza un poco, rebaja un poco el peso individual y fortalece el colectivo, y eso lo que a mi juicio (…) tiene que ver con la estabilidad del sistema, tiene que ver con las libertades.”

Until 1971 the two traditional parties together obtained more than 90% of the votes in every election.169 But since the 1970s, the Frente Amplio has also become a historically established actor (see Yaffé 2003). Since 1971 the Frente Amplio has broken the traditional two-party system opposing the PC and PN, giving rise to a moderate multiparty system. The system’s stability and perceived barriers to exit go hand in hand. They have, moreover, long been reinforced by parties´ history (see below). This history thus helps explain Uruguayan interviewees´ almost unanimous perception of high barriers to 169

This stability also translated to the system level; Uruguay was a democratic country from 1919 until 1973, with the exception of the authoritarian Terra regime of 1934-42 (Jacob 1983). 184

exit; a dense history reduces intentions to defect. This history and tradition gives strong weight to party organizations: party reputation comes into play, the organizational know-how and machines, among other assets. Thus, in Uruguay perceived barriers to exit have been reinforced partly by party history.

In terms of Ambition, PR and the DSV at different levels allow internal differences to be resolved within the party. If a political leader has problems with a senior leader, he can opt to open his own list without leaving the party. As González highlights (see above), the DSV regulates internal competition and solves internal disputes (see also Buquet 2003). These rules foster competition within parties, partially accounting for party leaders’ perception of satisfaction of Ambition.170 Along these lines, Lanzaro (1998) highlights: “…un régimen electoral ingenioso…que sirvió durante muchas décadas para asegurar una representatividad amplia y ventilar la competencia entre las elites...” (p. 200). The last constitutional reform, in 1997, introduced mandatory and simultaneous primary elections.171 This change in rules institutionalized intra-party competition and moved it to an early stage in the electoral process.

Moraes (2008), however, rightly notes that candidate selection is not strictly determined by rules. Parties have their methods and, thus, operate as “accelerants or brakes on the general tendencies exerted by the institutional framework” (p. 167). In this sense, although DSV and PR generate incentives for the proliferation of lists at the representative level, because elections are concurrent and fused, the “vertical integration” from the presidential and senate level reduces the effective number of factions. As Moraes (2008) and Buquet (2005) show, Ambition has thus been controlled 170

See Chasquetti (2010) for a thorough study of legislative careers in Uruguay. He shows that parties play a crucial role in satisfying individuals´ legislative career ambitions. Also, see Altman and Chasquetti (2005). 171 For further details of the changes introduced by this reform, see Chasquetti (2003) and Piñeiro (2007). The most important modifications were introducing party’s selection of unique candidates through open, simultaneous and mandatory primaries, and the prohibition of vote accumulation by sub-lemas at the representative level. 185

by party elites, above all leaders of senate lists and of factions operating at the national level. The existence of primaries, first enacted in 1999, has not severely changed this. Sometimes it has introduced a clear criterion for settling competition between prospective candidates for the House of Representatives. Traditional parties are now using votes cast by different local-level lists in the primaries to order the list for the national election. Nonetheless, national faction leaders competing for the Senate still have tremendous nomination power.

According to my fieldwork, most politicians from the foundational parties (PN and PC) are located in the EtpA row. They believe that rules of the game foster party organization via high barriers to exit and that the history of their parties also set high costs for defecting. They do not think the constellation of parties will change, and almost unanimously agree that both their parties and the system will survive a severe economic crisis. At the same time, they think their parties are viable channels for pursuing a political career, and thus A is observed. In terms of vibrancy, of the 15 politicians found in this row, 11 described their parties as vibrant ones.

In terms of Purpose, the graph below summarizes the evidence analyzed in the following pages. The only party in the “Purpose Quadrant” is the Frente Amplio172. The next section thoroughly analyzes each party’s situation in terms of Purpose.

172

For an analysis of ideological placement of Uruguayan parties and factions, see Altman (2001). 186

Intensity Purpose Quadrant

1

2

3

Coherence Key: 1: FA 2: PN 3: PC The length of the line signals degrees of agreement on each component of Purpose. The lengthier the line, the lesser the agreement.

Finally, the rows where at least three of the conditions are present are almost entirely populated by politicians from the FA. There are only two politicians from the foundational parties in these rows, and almost all of the FA’s established leaders are located there. 6.2. Configuration within each Party173

Origins, Early Evolution, and Long-Term Traits of the Foundational (Traditional) Parties. In contrast with the relatively volatile party systems of Chile and Costa Rica, 173

Given that party factions are institutionalized in Uruguay, I will consider the possibility that faction membership may account for areas of internal party disagreement. 187

two of Uruguay’s oldest parties, the Partido Colorado and Partido Nacional (or Blanco), have been among the world’s most enduring political organizations. These parties are thus excellent cases through which to examine how parties in developing countries can achieve vibrancy and have it endure over time.174 Both are said to have been born at the same time, and their evolution has been closely intertwined (Pérez Antón 1988). The origins of both the PN and PC can be traced back to the 1836 “Batalla de Carpintería”, in which the two camps used distinctive divisas (Blanca and Colorada) to differentiate themselves from one another (see Luna 2006).175 The birth of Uruguay as an independent state was a diplomatic invention, and no defined national identity existed at the time. However, such was the initial condition of all countries in the region. What was historically singular about Uruguay was that durable partisan identities quickly emerged immediately after independence, consolidated over time, and continue today to capture a significant portion of the votes cast in every election.

The early identity of each party first developed over the course of a series of violent conflicts. The Guerra Grande (1839-1851) constituted a major event in the country, though it involved a regional conflict, and even France and Britain, as well.176 Each party had its martyrs, and an epic was imbued on each side. Thus, the evolution of each partisan identity has been associated with traumatic events. After the war ended, different authoritarian presidents with ties with the Colorados tried to appease passions

174

The cases of the Partido Liberal and Partido Conservador in Colombia would have provided another possibility. Comparing the divergent trajectories of parties in Colombia and Uruguay could be theoretically rewarding. 175 Though originating in the first half of the nineteenth century, it is also clear that they would not become parties, in the strict sense of the term, until the twentieth century. However, as Caetano and Rilla (e.g., 2003) suggest, though they were not institutionalized parties, these parties´ identities started to develop beginning in the first half of the nineteenth century. For a thorough analysis of the origin and evolution of the Uruguayan traditional parties, see Pivel Devoto (1942). 176 For a detailed description of the Guerra Grande, see Barrán (1990). 188

and abandon partisan divisions.177 However, the series of conflicts persisted, illustrating the relevance even then of each camp’s identity.

The ending of the Revolución de las Lanzas (1870-72) with the Paz de Abril was crucial. For the first time, it generated a mutual recognition, and a power distribution agreement between Blancos and Colorados (known as the coparticipación) (Pérez Antón 1988).178 However, the weakness of the army—and all state institutions—allowed for regular eruption of conflicts. This scenario started to change in 1876, with Lorenzo Latorre’s military government. Under the successive authoritarian governments, state institutions would consolidate, and the state would gain the monopoly of the use of force. Therefore, the potential success of insurrections decreased. Their shared evolution, though conflictual, thus evolved into a tradition of “…partisan pacts and cooperation agreements between both camps.” (Luna 2006, p. 130). However, Blancos still fought 177

In 1855, Oribe and Flores, army leaders from each side signed the Pacto de la Unión, renounced seeking the presidency of the Republic. At the time, there were different attempts by some members of the elite tried to transcend partisan divisions. In this regard, Buquet and Chasquetti summarize both those attempts and the effects of the pervasiveness of partisan identities. They state: “La idea de que los viejos partidos no tenían razón de ser por no servir de conducto de diferencias ideológicas ni de proyectos políticos se convirtió en el argumento central de la intelectualidad ilustrada de la época. Las ideas de ‘fusión de los partidos’ en torno al bien común de la sociedad, eran solamente sostenidas por el elemento culto –los ‘doctores’– de las antiguas colectividades y no pudieron mantener su predominio durante mucho tiempo. Enfrentados a ellos, los ‘caudillos’ –jefes naturales de los sectores populares– hicieron prevalecer los sentimientos partidistas. Los “fusionistas” de los años 50, que se transformarían luego en los ‘principistas’ del último cuarto del siglo XIX, seguirían insistiendo en la cuestión ideológica como única razón para unirse a los partidos. Sin embargo, la fortaleza de las identidades partidarias primitivas resistiría estos embates en cierto modo hasta el presente, demostrando ser desde el punto de vista sociopolítico los agentes primarios de referencia y adhesión de la ciudadanía. La fortaleza y permanencia de los partidos tradicionales, superando los proyectos ‘fusionistas’ y ‘principistas’, puso en el orden del día la necesidad de establecer reglas para su coexistencia pacífica” (2004a, pp. 224-5). 178 Barrán (1990) says that the Revolución de las Lanzas was the first “Uruguayan” insurrection, because the camps (still the Blancos and Colorados) were only then not aligned with regional (and international) actors. The coparticipación refers to the geographical distribution of power, which was a major demand of the Blanco side: four of the thirteen departamentos (Uruguay’s subnational administrative units) would go to the Blancos. The Blancos—led by Timoteo Aparicio—also demanded fair elections. However, neither issue was resolved once and for all at the time. This lack of resolution explains the later insurrections led by Aparicio Saravia. 189

for inclusion, with two more major conflicts in 1897 and 1904, partly over state access and fair elections. Both were led by the Blanco caudillo Aparicio Saravia. The picture presented so far depicts two parties born right after the country’s independence that fought one another for state power and access, leaving an epic imprint on their identity (Pivel Devoto 1942, Barrán 1990). These parties were closely associated with the founding of the country´s national identity, each representing different perspectives that permeated society in different arenas. 179 They developed their identities in light of conflict. Trauma was thus present. These two parties eventually started to recognize each other. Peace was finally achieved through the consolidation of state institutions around a series of institutional arrangements leading to the consolidation of the system of “coparticipación.” Finally, they developed an agreement on electoral rules ensuring free and fair elections, which were finally realized through the 1918 Constitution, and the later guarantees developed in the Laws of 1924, 1925, 1938, and 1939 (see Buquet 2003). Mutual recognition and acceptance, agreements, and competitive rules thus together enabled the development of stable, modern parties in Uruguay (Pérez Antón 1984, Caetano, Rilla, and Pérez 1987).

These parties had endured through the non-competitive nineteenth century, and then led the modern transition to competitive elections and full participation. 180 In the process, each camp gradually became more institutionalized and assumed the traits of modern political parties.181 The process ended with a set of rules of the game that satisfied E and created incentives for parties to satisfy A.

179

For example, it is possible to identify two different historiographies. The construction of the national identity was built through these lively debates. 180 Both properties were effectively achieved at roughly the same time as the Constitution of 1918 (Buquet and Castellano 1995). They were thus attained in Uruguay long before they were in Costa Rica and Chile (in 1953 and 1958, respectively). 181 For an analysis of the electoral evolution of the two parties, see Buquet et al (1998). 190

For Sartori (1976), the PC was a predominant party from 1865 until 1959.182 González (1995) disputes this assertion for two reasons. First, it was not until 1918 that the system was polyarchic, and the PC thus cannot be characterized as predominant prior to that. Second, the PC did not control the majority in the House for three successive elections, as Sartori´s (1976) definition of a “predominant” party requires.183

In terms of Purpose, the existence of institutionalized factions did have an impact. For the non-competitive, pre-democratic nineteenth century, González (1995) says that the Colorado party evolved as the party of the city and was closer to a typical Liberal party of the time. The Blanco Party was meanwhile more rural and closer to being a Conservative party. However, even this characterization seems problematic given that, from the very beginning, each camp constituted a complex coalition, with personalistic rather than ideological loyalties uniting each group (Rial 1984, Luna 2006). By the early twentieth century, different ideological perspectives within each party had begun to coexist,184 though loyalty towards a specific leader within the party also remained relevant. This later consolidated, with the enactment of the DSV and other contemporary rules of the electoral system leading to institutionalized factions. It therefore seems clear that having institutionalized factions limited the Purpose of Uruguay’s two traditional parties185. Each party came to be composed of a very complex array of programmatic preferences.186

182

In the 1958 elections, the Blancos won the majority of the National Council Government, a collegiate (colegiado) government established in the 1952 Constitution. According to the rules set forth in that Constitution, six members were from the majority and three from the minority party—in this case the PC, which lost control of government for the first time in nine decades. 183 He does acknowledge that the system could be classified as having a predominant party, the PC, from 1942 until 1958 (p. 142). See above for other references defending the two-party system categorization. 184 For example, the first cleavage within the PC was that between Batllistas and Anti-Batllistas, or socalled “Colorados no Batllistas” (Caetano and Rilla 2003). 185 Buquet, Chasquetti, and Moraes (1998) mention that most of the time this did not provoke party indiscipline. 186 The programmatic differences within each party have generated a lively debate about their nature,a debate mostly focused on party-voter linkages, which are not the subject of this study. González claims 191

Although neither party has ever been extremely coherent, they had moments of intensity when directed with a definite project of development by one of their constituting factions. The Batllismo project in the early twentieth century is the clearest example.187 Other examples can be found throughout their history. In the PN, the popularly known Herrerismo faction, led by Luis Alberto de Herrera, could be regarded as an example of intensity in terms of Purpose. The Herreristas’ electoral preeminence facilitated coherence and thus an eventual presence of Purpose within the PN. Rial (1984) claims that Uruguay’s traditional parties were merely electoral machines. He bases his argument on the fact that parties increasingly relied on distributing particularistic goods through the state apparatus to keep voters’ loyalty. Rama (1971) describes the increasingly clientelistic nature of the Club Político, the local-level partisan organization.188 The author claims that the quality of their role degenerated over time.

Nevertheless, I concur with Pérez Antón (1984) and Caetano, Rilla, and Pérez (1988)’s description of the traditional parties. They argue that, the clientelistic nature of much of the party-voter linkage and importance of the parties’ factions in the first half of the

that both have always been catch-all parties (1995, p.143). Luna provides some reasons for the lack of ideological mobilization: “The lack of ethnic cleavages, the historical weakness of the Church and conservative forces in the country, and the rapid suffrage expansion at the turn of the century pushing dominant parties to rapidly incorporate the mass electorate through the expansion of the state apparatus (given the small size of the internal market and an economy based on low-labor intensive agriculture and commerce), the enactment of social legislation, and the creation of powerful clientelistic machines, all contribute to an explanation of the weakness of ideological mobilization in the system” (2006, p. 114). 187 In fact, this period of Uruguayan political history has been thoroughly studied. For example, see Vanger (1963), Real de Azúa (1964), Barrán and Nahum (1979), Panizza (1990), and Caetano (2011). The pervasiveness of scholarly works on this period signals its significance for the country´s history and, in turn, for the Colorado party. For an analysis of the consequences and the development of Uruguay democracy, see Rama (1987), Real de Azúa (1971) and Panizza (1990). 188 For a contemporary assessment of the workings of partisan organization at the local level, especially in relationship with voters, see Luna (2004) and Luna (forthcoming). 192

twentieth century aside, parties were also vibrantly organized, with multiple activities, roles, and structures.189 Thus, the clientelistic nature of parties’ main linkage to voters was not in contradiction with pervasive partisan organization and considerable party activism.190 Similarly, other secondary sources suggest that, starting very early, the two traditional parties developed vibrant political organizations, which they sustained throughout their history (Pivel Devoto 1942, González 1991).

The eventual absence of Purpose thus did not prevent vibrant organizations from developing. The impact of Trauma, which forged partisan identities based on retrospective loyalty and opposition to the other together with Exit and Ambition, favored by the complex set of rules presented above, help to explain the high degree of discipline in Congress, the nonexistence of inter-party alliances for most of the twentieth century191, and the fact that only one party faction ever switched sides. Lack of Purpose aside, Trauma, Exit, and Ambition were sufficient to generate party vibrancy.

When Purpose was lacking, whether because of problems of coherence or intensity, the parties were left with the promise of particularistic goods as their sole tool, leading to the process described by Rama (1971, 1987). This was particularly grave for the period starting in 1955. Interviewees and scholars have highlighted the impact of the successive economic crises (starting in 1955) as a reason for the debility of the foundational parties, arguing that this debility arose from their dependence on state resources to deliver private goods. This sequence shows the fragility of the combination of E and A. It seems that this combination, though sufficient for vibrancy, is less resilient to shocks than when P is also present. 189

Moreover, in terms of party-voter linkages, as Luna (forthcoming) highlights, clientelistic and programmatic linkages can coexist, and actually have. 190 As previously stated, I am here presenting a tentative narrative of the evolution of my causal conditions and my dependent variable, but my main focus lies on the observed contemporary configurations and outcomes, for which this narrative is essential. A more systematic analysis using primary sources would be needed for more rigorous tests. 191 Chasquetti (1998, 2008a) analyzes the PN and PC coalitions that emerged in the 1990s. 193

During the authoritarian period (1973-1985), some groups within both parties supported the regime. However, in contrast to Chile, the major factions were in the opposition and campaigned against the constitutional reform proposed by the Junta for establishing the foundations of a new regime. The proposal was subjected to a plebiscite in 1980, and the Junta lost (Rial 1984, Caetano and Rilla 1987). In terms of Trauma, although they were in the opposition and fought for the restoration of democracy, leaders of Uruguay’s traditional parties were not as pervasively persecuted and widely forced into exile as Uruguayan militants and leaders of the left. In my interviews, leaders from the PN and PC referred to the 1980s as a period of political involvement and reorganization of political parties.192 Almost no references emerged that could lead one to consider Trauma as having been present. (See below for a more detailed analysis of the PN.) The factions opposing the regime obtained clear majorities in the internal elections of 1982.

The post-authoritarian period was marked by some similar trends. In terms of Purpose, both the PC and PN lost their defining identities.193 The emergence and growth of the Frente Amplio (FA) forced traditional parties to place themselves more clearly on the center-right of the ideological spectrum. This could have generated more coherence within both the PN and PC.194 Nevertheless, most of the interviewees mentioned their parties having significant problems with coherence, a crucial component of Purpose. Nor did any widely shared project emerge to inspire intensity (interviews with PN and PC leaders). Moreover, the establishment of different governmental coalitions in the period since 1990 (and more clearly since 1995) until 2002195 led the two parties to neglect their individual programmatic identities. Instead, the logic of the party system became

192

For a detailed analysis of the negotiated transition to democracy in Uruguay, see Gillespie (1992, 2006). 193 However, the faction of Luis Alberto Lacalle (President 1990-1995) forged something close to a defined Purpose for a brief period. 194 In fact, the evidence introduced by Luna (forthcoming) supports this claim. 195 See Chasquetti (2008a). 194

one of bipolar competition between two blocs: the traditional parties versus the FA (Lanzaro 2000, Moreira 2004, Chasquetti and Garcé 2005). This change could become further consolidated; the two foundational parties are currently discussing the possibility of presenting a unified lema for the next (2015) municipal election. If that were to happen, it could constitute a significant milestone in their evolution.

It appears, though hypothetically, that when parties have had a dominant faction they have been able to conjure a strong Purpose, which has also influenced their vibrancy. Also, in the period before 1971, the electoral predominance of one faction did not imply a neglect of Ambition, mostly because of the existence of the DSV. For the period after the authoritarian regime, the growing electoral relevance of the FA, the crisis of the public sector196, and (for the last three national elections) establishing the rule of one candidate per party for the presidential election all seem to have created a tension between Purpose and Ambition.

The evidence from my fieldwork supports this narrative and shows, for example, that most politicians believe their parties have no Purpose. This is also supported by the coalition of factions that supported, for example, Sanguinetti’s second bid for the presidency (which he won in 1994). A disparate coalition of left-of-center Partido por el Gobierno del Pueblo (led by Hugo Batalla), the rightist Cruzada 94 (led by Pablo Millor) and his own faction, Foro Batllista, won the national election. Moreover, during his presidency a coalition with the PN was forged, debilitating the PC’s Purpose. Still, unlike in the PC, some within the PN did identify Purpose. Interestingly, these leaders were, with only one exception, middle aged and not fully consolidated as established leaders. Thus, both parties have been, and still are, mainly characterized by the configuration EtpA.

196

For a complete review of twentieth-century political economy in Uruguay, see Astori (2001). 195

In sum, we have two foundational parties with trauma similarly present at their origin, absence of Purpose for most of their history, and presence of Exit and Ambition. Despite these similarities, the next pages will be dedicated to separately analyzing each. I will mostly highlight their instances of identifiable Purpose and their present configuration as determined based on my fieldwork. PN: The PN has traditionally been considered more uniformly conservative than the PC.

This has been so since the turn of the twentieth century. However, the party has had leftist groups and an “insurrectionary spirit” that some leaders even identify as still present within the party (interview with PN leaders). Thus, coherence in terms of Purpose is not observed for portions of its history. However, it is worth highlighting that, as did the PC, the PN had a period of near hegemony of one faction, under the leadership of Luis Alberto de Herrera. This period began in 1920 when he first headed the party, and lasted until his death in 1959. Herrera tried to bridge the previous differences, rooted in the nineteenth century, between caudillos and doctores (see Caetano and Rilla 2003). He developed a series of ideological definitions defended with intensity, an ideological tendency that persisted within the party for the rest of the century. Its preeminence gave coherence to the party for much of the time Herrera led it. Caetano and Rilla summarize these ideas as follows: …una propuesta anti estatista, claramente defensora del rol primordial del Mercado y de la iniciativa privada; una desconfianza visceral a los impuestos (más de una vez el caudillo nacionalista convocó a la “huelga de los bolsillos cerrados”) y al crecimiento del Estado; un ruralismo radical y popular; una militancia política de adhesión democrática y liberal (con las sombras de sus simpatías por el fascismo y el franquismo…), cimentado en una visión popular de cómo hacer la política, y en un anti doctrinarismo acendrado; una afirmación de la primacía de la sociedad y de sus agentes; un recelo ante el aluvión inmigratorio…y ante las políticas educativas y sociales de perfil universalista… (Caetano and Rilla 2003, pp. 43-44).

196

The pervasiveness of Herrerismo197 generated, as typically happened in Uruguay’s traditional parties, an internal opposition. However, unlike those in the PC, the differences and tensions within the PN led to a split. The Partido Nacional Independiente was formed and competed as a separate party (lema) from 1944 until 1958.198 After unification, the PN won two successive elections (1958 and 1962), though with different and divergent factions199, bringing the party back to power after nine decades. The PN’s reunification with the PNI, and later the death of Herrera, generated a more balanced dispute between the herreristas and those opposed to them.

The party underwent other programmatic transformations associated with the emergence of a new faction, Por la Patria of Wilson Ferreira Aldunate, who led the party from the end of the 1960s until his death in 1988. Ferreira Aldunate gave the party a centrist perspective in years of polarization and was influenced by the desarrollismo economic perspective in vogue at the ECLAC. He later became a fervent opponent of the authoritarian regime, denouncing it in international forums. He was in exile and not authorized to run for the presidency in the 1984 elections, the first after the authoritarian regime. However, the PN did not suffer from persecution as pervasive as experienced on the left. In the PN’s case, repression was more focused on individual leaders like Ferreira Aldunate. Moreover, his alignment with Julio María Sanguinetti’s PC-led government against the referendum campaign that sought to repeal the amnesty Ley de Caducidad implied a clear rejection of the role of Trauma within his organization.

197

Although Herrerismo and Batllismo signal the importance of leaders (caudillos) in determining the identity of a party´s faction—and its opposition—neither became the tool of a single leader. Their history and the internal disputes ever-present may have prevented this total personalization of party politics. 198 Doing an in-depth study with primary sources to shed more light on the configuration of my causal conditions for explaining party vibrancy would be necessary to further understand this period. 199 Mostly between a more conservative faction, Herrerismo (which forged an alliance for the 1958 elections with an outside group known as Ruralismo led by Benito Nardone) and a more progressive one called Unión Blanca Democrática (which resulted from the unification of Nacionalistas Independientes and emerging, progressive leaders like Daniel Fernández Crespo). 197

Moreover, the rest of the party (the minority of herreristas) did not suffer under the authoritarian regime. With the restoration of democracy in 1985 and after Ferreira Aldunate’s death and the demise of Por la Patria, two major groups alternated in controlling the party. One, led by Luis Alberto Lacalle (President of the Republic from 1990 until 1995), was more conservative and followed Herrera’s ideas. The other, now controlled by Jorge Larrañaga, was a more centrist group, with different factions.200

My interviewees agreed almost unanimously that the PN had Ambition. In fact, some highlighted this trait as both a virtue and a structural problem of the organization (interviews with PN leader).201 This is accomplished through a series of recognized requirements and established processes. First and foremost, one has to be conscious that a political career takes time and requires constructing a network based on mutual trust. One also has to show that one can gather votes for the party, normally proving one’s ability in this respect through primary or internal elections. Finally, one has to be able to deal with the party elite. According to interviewees, it is politicians’ differential ability to meet these requirements that accounts for their relative likelihood of pursuing a successful political career. These factors are widely mentioned by politicians from the traditional parties (Colorado and Nacional). They represent a mix of formal and informal but known rules and traditions that individuals can follow to pursue their ambitions through the party organization. These mechanisms for satisfying ambition are intertwined with reproducing the partisan organization, its identity, and a sense of belonging. The following dialogue with a young leader from the PN synthetizes how Ambition has operated within the party. 200

Although leaders from these centrist factions claim to be heirs of Ferreira Aldunate, established leaders agree that, with his death, the Purpose he tried to forge within the party almost disappeared (interviews with PN leaders). 201 One illustration is the fratricide electoral competition in the primary election in 1999 (Chasquetti and Garcé 2005). 198

Si tuviera que sintetizar aquello que explica su trayectoria personal en la interna de su partido, ¿a qué factores lo atribuiría? “Eso es el laburo político, no tiene misterio. El laburo político es algo que es mucho más jodido y desgastante de lo que los comentaristas creen o creemos. Los dirigentes políticos tanto departamentales como nacionales precisan de los dirigentes más chicos digamos (el barrial, el de la ciudad, lo que fuere) para el apoyo de los votos y ese apoyo y esa relación se genera en el plazo de cierto tiempo, que implica una relación de confianza, que implica una relación en donde efectivamente se le mide el aceite al dirigente y se lo percibe como alguien válido para hacer esa actividad, para conseguir cosas, para lo que fuere que sea la evaluación que hacen los dirigentes chicos de por qué vamos a apoyar a este. Hay también todo lo que hace a la complejidad de los apoyos en gente más veterana que es las trayectorias en lugares paralelos. Vos si fuiste dirigente gremial del agro durante mucho tiempo y te manejaste con cierto criterio tu salto a la política es más fácil en la medida que ya sos conocido. (…) En fin, son múltiples los canales de legitimación, ahora lo que sí está claro es que nadie te regala nada (…) y que el lugar lo ocupás por la actividad específica en la política. Capaz que la puerta se te abre (…) por lo que se te ocurra, pero esa es una puerta que te permite cierta notoriedad, en donde se te puede hacer más fácil conocer gente. Pero no alcanza ni cerca con eso para después pasar a votos lo que después tengas. Los votos es todo lo que es la tarea proselitista clásica: el puerta a puerta, el trabajo, la recorrida, el conocer a la gente, el que te vayan probando en cosas que podés conseguir, en cosas que efectivamente hacés”. ¿Cuáles son los mecanismos que existen hoy día en su partido para hacer una carrera política? En otras palabras, si Ud. hoy en día tuviera que iniciar nuevamente su carrera política, ¿cómo se imagina los primeros pasos? ¿Qué debería hacer para crecer políticamente? “Las elecciones. Vos te presentás a una elección y si te va bien empezás a avanzar. Después tenés que enfrentarte a los otros dirigentes que te van a querer matar porque te fue bien.” ¿Considera que el partido favorece la competencia entre los líderes actuales y personas con potencial de liderazgo e interés de participar pero sin cargos formales en el partido? ¿Y su sector, favorece o no la competencia entre líderes y quienes quisieran serlo? “El partido lo hizo muy claramente con las elecciones de juventud. Las elecciones de juventud del partido fueron un almácigo que generó dirigentes de veintipico que hoy en día son ediles en muchos departamentos gracias a esa proyección previa. En 2007 armaron sus amistades, sus cuestiones, se descubrieron liderazgos, vieron cómo había que hacer las cosas, y en el 2010 no hay departamento creo yo en donde el partido haya ganado por lo menos en donde no haya uno o dos ediles de veintipico que surgieron de las elecciones juveniles”.

Leaders from this party claim that the almost anarchically uncoordinated stances of PN leaders, while it facilitates the presence of Ambition, negatively affects the party’s

199

Purpose—and even its strategic-electoral decisions.202 The following two quotes summarize the party’s Purpose and the relationship between Purpose and Ambition:

“Yo creo que una de las cosas que funcionan bien es que es un partido que es realmente libre y abierto, todo el mundo opina lo que quiere, todo el mundo dice lo que quiere, todo el mundo hace prácticamente lo que quiere. En segundo lugar, están todas las calles abiertas. Si yo quiero ser Presidente de la República en el PN, puedo postularme; y si quiero ser diputado, puedo postularme; y si quiero ser Intendente, puedo postularme. Es decir, no tengo que pedir permiso, tengo que lograr que la gente me acompañe, punto, nada más. O sea yo puedo sacar mi lista a diputado y al que no le gusta que se embrome, que no me vote, pero si soy más que el de al lado le voy a ganar, el diputado voy a ser yo y no va a ser él. Y lo mismo a nivel presidencial, cualquiera de los que está hoy en el tapete, o que ni siquiera imaginamos que pueda estar, puede aparecer en el escenario y decir “yo soy blanco, quiero hacer algo diferente, quiero que la gente me acompañe”, pone su nombre arriba de la mesa y no hay manera de decirle que no, y entonces la que decide es la gente. Es un partido en eso muy libre, muy abierto. Eso mismo, obviamente tiene su contracara, como toda moneda. Es un partido al que le cuesta mucho lo orgánico, le cuesta mucho la disciplina, le cuesta mucho a veces la coherencia, porque es muy abierto. Pero yo me quedo con que hay movilidad adentro y hay una estructura que permite la renovación.”

El PN es un partido liberal por excelencia, porque es el partido que ha defendido históricamente desde su fundación el valor de la libertad. Es un partido que ha levantado revoluciones por la libertad. Y a no ser por ese pacto que hace Herrera con Terra en el 33 para justificar el golpe de Estado, (…) el partido siempre ha estado signado por la vigencia y la defensa del valor de la libertad como valor fundamental en la convivencia dentro de la sociedad. Ese me parece que es un elemento que le da mucha fuerza al partido, y además (…) por esa misma razón (…) siempre se ha dicho que los blancos son un poco anárquicos, porque siempre son reivindicadores de ese valor de la libertad y son más auténticos en hacerlo. Por eso fue el partido también de la revolución. Wilson Ferreira Aldunate decía en determinado momento (cuando en esa época el FA era muy novato, era una fuerza todavía muy pequeña), “el único partido político que existe en el Uruguay es el PN, porque el PC es el nombre que se le pone en el Uruguay al poder”. El poder siempre estuvo asociado con el PC históricamente. Entonces, yo diría que la autenticidad en ese aspecto es una de las virtudes que tiene el partido. De los defectos, la contracara de esa autenticidad y de ese sentimiento libertario: el no saber muchas veces entender o adaptarse a una concepción más orgánica, más disciplinada, y más estructurada, que es lo que requiere el funcionamiento de un partido político. Este ha sido uno de los grandes problemas históricos y ha sido causa de más de uno de los líos y de los problemas internos que ha tenido el partido, e incluso en los momentos en que le tocó gobernar fue un serio factor, un grave problema, sobre todo en el período 1959-1967.”

202

Although it exceeds the goals of this study, it would be interesting to further explore the inherent tensions of parties that, on the one hand, have to promote individuals´ careers but, on the other, must preserve and protect a collective identity. See below for a similar discussion of the relationship between vibrancy and years in office. 200

PC: The Party Colorado, traditionally the PN’s main rival, historically had at least two

groups. José Batlle y Ordóñez’s presidencies in 1903-1907 and 1911-1915 forged a strong Purpose that marked the party, and the country.203 However, during Batlle’s years, the party also had an internal “Anti-Batllista” opposition comprised of former Batllistas. This internal opposition consolidated mainly in 1913, during Battle’s second presidency, in the context of discussions of the colegiado executive (Barrán and Nahum 1979). However, the main focus of the internal opposition, which was mainly composed of conservative groups, was the direction of Battle’s reforms. These conservative groups later founded the faction Partido Colorado General Rivera, and came to be known as the riveristas.204 Power rotated within the party. Feliciano Viera was elected in 1915 and put an “alto” (Alto de Viera), i.e. a stop to Batlle´s social and economic reforms, though they were not reversed (Nahum 1986). After the death of Batlle in 1919, the division between Batllistas and Colorados no Batllistas consolidated (Caetano 1992). Still, the influence of Batlle and his reforms was pervasive within the party and gave it a defined Purpose lasting for decades (Barrán and Nahum 1979, Caetano 2011). Batlle’s nephew Luis Batlle Bérres, who served as President of the Republic 1947-1951 and President of the National Council of Government (the colegiado) from 1955-1956, is considered the leader of a second Batllismo, known as Neobatllismo (Panizza 1990). He was an influential leader within the party until his death in 1964. During this period, the party nonetheless once again had rival factions, basically reproducing the disputes of the earlier Batllistas and Colorados no Batllistas. Moreover, starting in 1950, two factions disputed the Batllismo tradition and legacy, “the 15” and “the 14.”205 In the 203

Some of his own major ideas (e.g., rationalism and cosmopolitanism) were already part of the party´s patrimony though (Pérez Anton 1984). For a thorough review of this significant period of Uruguayan politics, see Barrán and Nahum (1979). 204 The terms refer to the attempt to rescue the traditional values of the founding period of the protoparty, led by General Fructuoso Rivera (Caetano 1992, 2011). 205 The numbers correspond to those of their Lists. The 15 was led by Luis Batlle Bérres while the 14, more conservative, was led by César and Lorenzo Batlle Pacheco. This rivalry, in addition to the nonbatllistas factions, would last for decades, forging the main identities of the contemporary PC. 201

years preceding the authoritarian regime, the party was mainly captured by the more rightist, more conservative sector led by Óscar Gestido and Jorge Pacheco Areco (Panizza 1990).206

For purposes of this study, it is sufficient to claim that significant disputes in terms of Purpose became increasingly significant within the party, thus diluting the intensity and coherence of its Purpose. With the restoration of democracy in 1985, the PC also favored policies in line with structural reform and liberalization of the economy, albeit not as intensely as Lacalle’s PN government (Lanzaro 2000). The clearest long-term consequence within the party has been gradual erosion of its left-of-center, Batllistas factions (Moreira 2004).

The PC differs from the PN in that almost no interviewees mentioned the PC having Purpose. The following quote identifies the loss of Purpose within the PC.

Empieza a perder un poco el perfil [batllista], también son los tiempos, fue la época del Consenso de Washington, estamos en la década de los noventa. Eso fue desde el punto de vista sociológico y político, ir abandonando un público, un sector de la sociedad, que empezó a emigrar a quien le hacía esa oferta (a pesar que tenía cosas totalmente incompatibles) que era el FA.

In fact, some actually referred explicitly to its increasing problem of programmatic and ideological identity. This might be explained by the description presented above and, more recently, the electoral collapse of 2004 after the economic meltdown of 2002.207 Examining the reasons for the PC’s downfall and comeback would be a subject in itself.

206

From 1954 onward, the faction “14”, more conservative though Batllista, started to ally with the Colorados no Batllistas, consolidating a new conservative group within the party (Buquet 2003). In the 1966 presidential election, the faction, which had since 1962 been called Unión Colorada y Batllista since 1962, won the presidential election. 207 See Rosenblatt (2006) for an analysis of the political management of the economic crisis. For a detailed description of the critical years of 2001 and 2002, see Paolillo (2004). 202

The evidence collected in my fieldwork suggests that it resulted from a conjunction of Exit (rules and history) and Ambition. The PC has recently been almost homogeneously governed by one faction, located more to the right of the ideological spectrum (Vamos Uruguay led by Pedro Bordaberry). Paradoxically enough, this coherent and intense faction, which has constituted the majority of the party, has not led to the presence of Purpose. This apparent anomaly can be traced to the existence of another group within the party: Propuesta Batillista or ProBa. While not as electorally successful as Vamos Uruguay, ProBa is organizationally relevant; it is comprised of more centrist Batllistas, who are self-proclaimed heirs of that tradition. Interviewees not belonging to Vamos Uruguay therefore think that the organization has lost its programmatic identity. Moreover, those interviewed who were from Vamos Uruguay—with the exception of one young, prospective leader— acknowledged that the party has problems in terms of coherence and intensity and that their faction has been unable to promote a defined Purpose for the party as a whole.

As already highlighted, interviewees from the two traditional parties almost unanimously agreed on the presence of Ambition, not only because of the rules of the game (like mandatory primaries) but also thanks to organizational decisions to adopt primaries as a formal mechanism for ranking candidates on its national electoral list.208 This next quote highlights the presence of Ambition. The interviewee attributed its presence to the 1997 Constitutional reform, which established mandatory primary elections, explicitly highlighting the difference the reform had made in this regard. Like other leaders, he also signaled that Ambition had been most notably absent in the 1990s when Julio María Sanguinetti and Jorge Batlle had served as the main leaders, even though they ultimately conducted the party with the rules of mandatory primary elections.209 This interviewee and other leaders therefore also presented the change in

208

The FA, however, does not do so, as discussed below. Still, the reasons for the eventual absence of Ambition during a specific period of the party’s recent history far exceeds the goals of the present study. 209

203

leadership after the electoral collapse of 2004 as having facilitated Ambition’s reemergence. “Yo creo que es esencial la relación con la dirigencia de base, esencial. Sin eso no hay carrera política. Pero, ¿cuáles son los mecanismos que existen? Porque quizás vos hacés eso y el partido te dice “no”. No, es que el partido no me va a decir que no, el partido le dice “sí” a todo el mundo. Existe esa posibilidad. Antes vos lo que tenías que hacer era ser amigo de Jorge Batlle o de Sanguinetti. Tenías que ponerte las rodilleras. Esa era la forma de ascender, esa es la verdad, porque eran ellos los que armaban la lista. (…) Era el líder el que te identificaba y decía “vos, dale vení vas acá, vos vas segundo en la lista, vos vas tercero, vos vas cuarto” porque sí, si no había forma de decir quién era mejor, era el olfato de él para ver quién servía más o quién era más funcional a él o al partido. La elección interna cambió todo. Entonces la elección interna es la forma de surgir en política. Ahora, para eso hay que manejar estructura, y no hay otra, porque ningún tipo desconocido va a hacer una declaración y una campaña televisiva para que lo conozca todo el mundo, y que la opinión diga “hay que votar a este fenómeno”. Entonces, ¿dirías que hoy se favorece o no la competencia entre líderes actuales y quienes quieren serlo? Sí, sin dudas. En realidad el Batllismo siempre fue partidario de las internas. Me parece que la interna da las oportunidades, y si yo quiero mantenerme donde estoy voy a tener que trabajar.”

In the words of one young leader from the PC, the party has overcome a period of ossification:

¿Cuáles son los mecanismos que existen hoy día en su partido para hacer una carrera política? En otras palabras, si Ud. hoy en día tuviera que iniciar nuevamente su carrera política, ¿cómo se imagina los primeros pasos? ¿Qué debería hacer para crecer políticamente? “Yo creo que lo mismo, exactamente lo mismo. Hoy hay una chance muy interesante para el que empiece a militar en el PC porque lo peor aparentemente pasó y aparentemente lo que espera al PC es subida, por ahora todo subida, no sabemos hasta dónde pero subida. Eso acompañado a que en el PC hay un ambiente favorable, hay un ambiente de alguna manera un poco desintoxicado de corporativismos o de clanes en el sentido etáreo y en el sentido de gente que está hace mucho tiempo en determinado lado y entonces de alguna manera se cuida entre ella. Como hay una ausencia de corporativismos. Si uno entra al PC y se quiere llevar el mundo por delante a mérito propio tiene todo para ganar.”

In terms of the evidence collected in my fieldwork, the PC’s overall configuration is similar to the PN’s: EtpA. 204

FA: Leftist parties, both Socialist and Communist, have existed in Uruguay since the

beginning of the twentieth century.210 However, they never achieved electoral relevance. In the 1960s, the Socialist Party and the Communist Party built separate coalitions with other small parties,211 ventures that proved unsuccessful. In 1971, the Frente Amplio (FA) broke into the two-party system. It emerged in a context of economic crisis, political polarization, violence, and repression. The country was experiencing increasingly severe economic conditions. The years of “happy Uruguay” as a prosperous Latin American country were long gone (Cancela and Melgar 1985, Astori 2001, Garcé and Yaffé 2005).212

The traditional parties (PC and PN) lost their Purpose, and increasing conflicts in terms of coherence emerged (see above). The combination of absence of Purpose and economic crisis was fatal for their clientelistic structure.213 Society, as in the rest of the region, was increasingly polarized, and analyses at the time, showed it was increasingly distant from the foundational parties. In fact, though existing rules set high barriers to exit, defection nonetheless occurred, and groups from traditional parties 214, in coalition with leftist parties, formed the FA.

210

The Socialist Party was founded in 1910 and the Communist Party, which split off from it, in 1921. For a review of the pre-FA leftist parties, see Lanzaro (2004). 211 In 1962, the PS formed the electorally disastrous Unión Popular, and the PC formed FIDEL. 212 In fact, it was a time of profound intellectual production, with very critical assessments of the sociopolitical situation of the country. The work of Rama (1971) or the weekly Marcha are examples of this zeitgeist among intellectuals. 213 As already mentioned, in extreme contexts of crisis, Exit and Ambition can prove insufficient for party vibrancy. 214 Among the founders of the FA were Francisco Rodríguez Camusso from the PN and Zelmar Michelini from the PC. 205

The FA originated mostly as a coalition of the Communist Party, the Socialist Party, the Christian Democratic Party215, and factions from the traditional parties. Thus, the FA was a coalition of previously existing political organizations, which intended to mobilize Uruguayan society. The term coalition could appear to mean that in 1971 the FA was not a party; even today some refer to the FA as a coalition. However,—the partners’ participation in national elections under the same lema (the legal term for party) and the complex structure created at the FA’s origin and consolidated over time signal that it has always operated as a party, albeit with the institutionalized factions characteristic of Uruguayan political parties. Moreover, in its Declaración Constitutiva the newly created organization committed to operating permanently. This commitment was then manifested in the Reglamento de Organización approved in March 1971. Like every party, it underwent a process of institutionalization, through which the FA gained importance relative to each component.216 Lanzaro summarizes the original organizational nature of the FA as follows:

Desde un principio el compuesto fue más que la suma de sus partes, dando lugar a un corpus propiamente frenteamplista, en términos de identidad y en términos de estructura…a la vera de los grupos que lo forman, el FA monta su red peculiar de organizaciones de base…que operan como núcleos de integración y de militancia, forjando su activismo en la campaña electoral y en las movilizaciones de masa. El aparato del FA refleja su dualidad constitucional. Sus instancias directivas se integran con los representantes de los partidos miembros. Pero al mismo tiempo incluyen una cuota de representantes directos de las bases frenteamplistas…con un régimen de mayoría calificada para la resolución de algunos asuntos (2004, pp. 38-9).

215

In fact, for the 1971 elections, the official name of the FA was Partido Demócrata Cristiano (PDC). The PDC was a splinter from the traditional, though electorally irrelevant, Unión Cívica founded in 1911, representing Catholic interests and values. 216 For a detailed analysis of that process, see Yaffé (2005, pp. 101-25). The author also describes the complex internal structure (pp. 109-16). 206

As in the case of the PC and PN, after the FA’s establishment, rules fostered partisan organization and granted a certain degree of Ambition. Regarding the latter, the union of different political organizations under the Frente Amplio constituted a great challenge. The satisfaction of different ambitions—as well as the programmatic, ideological compromise (see below)—was thus something that leaders took care of. The condition of Ambition was initially reflected in the FA’s foundational structure, which granted a role to all its constituent units, and its use of DSV to determine legislative seats; by using DSV, the FA habilitated the existence of multiple lists.217

In terms of Purpose, despite the divergent origins of its components, from its start the FA put forth a leftist platform, as illustrated by its Declaración Constitutiva of February 5, 1971:

Un programa de contenido democrático y antiimperialista que establezca el control y la dirección planificada y nacionalizada de los puntos claves del sistema económico para sacar al país de su estancamiento, redistribuir de modo equitativo el ingreso, aniquilar el predominio de la oligarquía de intermediarios, banqueros y latifundistas y realizar una política de efectiva libertad y bienestar, basada en el esfuerzo productivo de todos los habitantes de la República.” (Frente Amplio, Declaración Constitutiva)218

The reasons parties are able to develop Purpose are manifold and exceed the parameters of this study. However, the harsh times that the country was suffering, leaders’ emphasis on consenso between the components of the FA to prevent any group from overwhelming the rest, and agreement on few major ideological and programmatic

217

However, to preserve the unity of the new organization, the FA decided to present one presidential candidate (although the DSV habilitated the existence of a plurality of presidential candidates within the same lema). This trait maintained until the 1999 electoral cycle in which Tabaré Vázquez defeated Danilo Astori in the presidential primary with a wide margin (82% of the votes cast at the FA´s primary went to Vázquez´s candidacy). 218 Available at: http://www.frenteamplio.org.uy/sites/frenteamplio.org.uy/files/DOCUMENTO%201.pdf 207

issues understood as “basic agreements” could well help to explain why the FA was able to do so (Garcé and Yaffé 2005 and Yaffé 2005).

The authoritarian regime (1973-1985) meant exile, prison, and torture for most Frente Amplio leaders.219 This process was epitomized by the imprisonment of the FA’s first presidential candidate and then leader until 1996, Líber Seregni.220 Thus, the party emerged with a coherent and intense Purpose, and the authoritarian regime inflicted a traumatic experience on its militants and leaders.

After the authoritarian regime, when the PC and PN tried to pursue structural reforms and liberalization of the economy, the FA also became the defender of the ISI model and the Uruguayan welfare state. In this sense, Luna mentions: Frente Amplio gave political expression to a “veto-coalition” of ISI beneficiaries while at the same time, drawing on a reinterpretation of batllismo to attract votes from the sectors that became increasingly alienated from the traditional system in the wake the “happy Uruguay” crisis. Ideologically, this coalition put forth a statist platform advocating and enacting legislation (through the use of direct democracy mechanisms to roll-back reformist legislation in favor of the status-quo (strong state intervention in the domestic economy)” (2006, p. 116).

Garcé and Yaffé (2005) and Yaffé (2005) detail the process of ideological and programmatic moderation of the FA after the authoritarian period, identifying three stages: the building of frenteamplismo until 1984221, a period of transition from 1985 to 1994, and a progresista era starting in 1995. The latter was characterized by progress in moderation. Their studies show how the party gradually adopted more centrist

219

For historical reviews see Bermeo (2003) for the process leading to the coup and Caetano and Rilla (1987) for the authoritarian regime. 220 Seregni resigned from the FA´s presidency as a consequence of disputes over the proposed constitutional reform. 221 Yaffé (2005) labels this the period of institutionalization. 208

perspectives compared to the clearly leftist program of 1971. This process is also described by the authors as having materialized in lively debates within subsequent party congresses and documents.222 This gradual and mostly deliberative process signals a true process of party adaptation (see Chapter 1). Yaffé summarizes this process, which he analyzes thoroughly through party documents, as follows: “…y ese cambio no debe interpretarse como una mera operación de maquillaje electoral, o un giro oportunista de último momento, pues ha sido el resultado de un largo y complejo proceso de renovación ideológica y programática.”(2005, p. 97).223 This case illustrates that, although the party’s Purpose changed, as was to be expected given the changes in historical context, neither coherence nor intensity have been affected (see Lanzaro 2004). Thus, the FA shows how the existence of Purpose over time does not imply the neglect of changes. Also the case of the FA illustrates that internal divergences do not contradict the presence of Purpose. Today this divergence is expressed in the existence of two identifiable blocks. One of the blocks is composed of the Communist Party, the Movimiento de Participación Popular (MPP)224 and other minor factions. José Mujica (a former Tupamaro225 and currently President of the Republic) is the most important leader of this sector. The other major sector is composed 222

It also meant the splitting off of some moderates groups in 1989. The Partido Demócrata Cristiano and Movimiento por el Gobierno del Pueblo left the FA. Together with the Unión Cívica, they then founded the Nuevo Espacio (NE). The end of the 1980s and early 1990s were years of realignment within the party and the formation of new factions that continue to exist today (e.g., Vertiente Artiguista and Asamblea Uruguay). In 1988, the political faction of former Tupamaros (the MPP) also integrating other leaders was formally accepted in the FA. The NE aligned with the FA for the 2004 presidential elections and later rejoined the FA. 223 See also Lanzaro (2004). The evolution of the party contradicts their conclusions regarding the emergence of a Progresista identity. In fact, the coalition Encuentro Progresista, which sought to include other groups outside the FA, disappeared. The Frente Amplio label prevailed, however, and influenced the ideological and programmatic debates within the party. Though the party did moderate, this moderation did not end up influencing the party´s core ideas and identity. 224 For a detailed analysis of the ideological evolution of the post-authoritarian regime Tupamaros and later MPP, see Garcé (2006). For an analysis of its organizational strategies and party-voter linkage strategies, see Luna (forthcoming). 225 For a detailed history of the Tupamaros, as well as its later evolution, see Labrousse and Klang (2009). 209

of various moderate left-of-center factions, of which Danilo Astori (currently vicepresident) is the most important leader. What is important is how the organization resolves internal disputes; coherence does not necessarily means unanimity but agreement on a project and some major ideas concerning the role of the party and its goals which are intensely defended.

The evidence from my fieldwork shows agreement about the existence of Purpose. However, it is important to mention that some members of left-of-center factions made no references to Purpose, thus reinforcing the divide mentioned above. This fact did not lead me to compute the absence of Purpose (p) for two reasons. First, a majority explicitly referred to the presence of Purpose. Second, the rest did not explicitly refer to problems related to coherence and/or intensity. In fact, only four of the twenty interviewees claimed that there were problems specifically related to coherence, and most of those were not established leaders. Perhaps the role of Trauma or the fact that the party has not yet lost control of government has helped the organization in terms of coherence. A waning of Trauma or loss of power could perhaps generate greater conflicts in the future.

The following statement illustrates that, although important internal tensions and discussions have occurred within the FA, the party has had an ideological and programmatic identity.

210

Yo te voy a decir que el FA no es el Frente Grande y no es una cuestión semántica. Yo creo que el FA tiene definiciones en su constitución que obligan a determinado posicionamiento político, y porque no existe “el imperio” a la usanza de la década de los 60s o de los 70s, no quiere decir que la primera potencia en el mundo política, económica y militar, no siga ejerciendo determinada actitud imperial. Entonces, a mi juicio las definiciones siguen estando vigentes, yo creo que hay sectores que no tienen con el mismo grado de decisión esas definiciones. Entonces, cuando hay otras propuestas o existieron otras propuestas, como por ejemplo el Frente Grande, yo creo que hay en ello una voluntad más laxa que la que pudo constituirse en determinado momento como Nueva Mayoría. Por algo nosotros logramos sin ningún tipo de crisis, y con mucha gente que todavía seguía y sigue al día de hoy usando “Encuentro Progresista” y algunos registros institucionales que dicen “FAEP-NM”, el que quedáramos como FA. O bueno, vos me dirás es la conveniencia de una identidad, de una grifa, de un sello. No alcanza sólo con eso, es también un marco de definiciones, que no te obliga a no ser amplio, pero que sí te marca un sendero. Y yo digo, el tipo de debate que se está dando hoy, no es sólo porque Uruguay está en una coyuntura de crecimiento que vos te proponés “discutamos la redistribución de la riqueza”; podríamos pasar a la chilena, concentrando en un determinado tipo de matriz productiva la riqueza y con políticas públicas levantando a los pobres y sacando a los indigentes. Podríamos seguir así, entonces lo agro exportador va a seguir concentrando la riqueza y podemos tener un MIDES más eficiente que siga haciendo determinadas políticas sociales compensatorias. Y sin embargo, no, estamos discutiendo efectivamente en qué punto estamos: si hemos crecido hasta ahora y si vamos a seguir creciendo por un período no menor, sigamos viendo si a la reducción de la pobreza, a la reducción de la indigencia, le hemos mejorado los índices de igualdad de oportunidades. Pero me parece que esto se plantea en una organización política donde hay determinada definición ideológica. Y me fui al carajo que me olvidé lo que me habías preguntado.

As already mentioned, the FA has used the singular electoral system, which facilitates the resolution of disputes over power positions.226 However, as Moraes (2008) rightfully suggests, rules do not determine the end result in terms of career trajectory; organizational mechanisms are also relevant. My fieldwork confirms this perspective and reveals problems of elite promotion and renovation. Yaffé (2005) describes a process of elite renovation in the 1990s, manifested for example in the emergence of Tabaré Vázquez. However, in-depth interviews with leaders showed that the process stopped in the first years of the 2000s. Thoroughly reviewing the transcripts reveals significant disagreement about Ambition, disagreement which cannot be explained in terms of interviewees’ generational or relative positions. In fact, problems associated with elite promotion and renovation have been sources of public discord and 226

In this sense, Piñeiro and Yaffé (2004) and Yaffé (2005) review the evolution of the effective number of factions in the FA. See also Moraes (2008). 211

preoccupation among party elites. Those who raised this concern also mentioned that the party might have problems revitalizing its leadership in the near future. The following quote from a young prospective leader illustrate this point:

¿Considera que el partido favorece la competencia entre los líderes actuales y personas con potencial de liderazgo e interés de participar pero sin cargos formales en el partido? ¿Y su sector, favorece o no la competencia entre líderes y quienes quisieran serlo? “No, en general no, hay como líderes que están instaladísimos. Es natural, no es lo más sano. (…) Hay liderazgos que son tan fuertes, como el de Tabaré, el del Pepe, Danilo, hay características personales que tampoco podemos inventar, si tenemos candidatos que son buenos, que tienen legitimidad. Me gustaría que hubiera gente más joven, pero tampoco por poner un tipo de 50 años no vas a dejar de reconocer. (…) Yo creo que la juventud del FA debería competir electoralmente, creo que ahí los partidos tradicionales hicieron bien, compitieron y crecieron. (…) Lo que tienen que hacer los jóvenes es competir y ver qué pasa. (…)

As observed in the discussion of Chile, and as the next quote illustrates, most interviewees attesting to the presence of Trauma (T) also claimed that the party suffered from the absence of Ambition (a).

¿Considera que el partido favorece la competencia entre los líderes actuales y personas con potencial de liderazgo e interés de participar pero sin cargos formales en el partido? ¿Y su sector, favorece o no la competencia entre líderes y quienes quisieran serlo? “No, no favorece.” ¿No favorece la estructura o no favorece “el peso de la historia” como conversábamos antes? “El peso de la historia, inequívocamente.” ¿Y la estructura te permite? Sí, la estructura te lo permitiría. No es la estructura lo que está trancando, lo que está trancando son los elefantes blancos, se sentaron arriba de la historia.

Future research could examine the relationship between Trauma and Ambition in greater depth. At first sight, as shown in the PS in Chile, it seems that Trauma solves collective action problems but also hinders elite renovation through the consolidation of a marked 212

generational cleavage. Leaders who suffered repression or persecution gain a heroic profile, becoming figures in a very significant epic which may be hard for younger leaders wanting to advance their political careers to dispute. Again, as in Chile, this problem was raised in several interviews in the context of discussions of the role of trauma. This relationship between Trauma and partisan organization was also acknowledged by one young FA leader. This leader also mentioned the differences between generations in terms of their relationship to those harsh years and their role:

“Hay una generación que es aquellos que hacían política antes de que exista el FA y que empiezan a militar con el nacimiento del FA (hay una generación etaria bastante importante). Otra generación que es la que está a la salida de la dictadura, que se forma en la propia dictadura, que tiene que ver con espacios de clandestinidad, por espacios sociales, un debate muy limitado, que va desde esos años hasta los primeros años de los 90s. Y luego una generación que se forma en democracia, con una izquierda internacional que pierde pisada y que pierde utopías, y que por lo tanto vive una función bastante nihilista, individualista, que es la que llega como jóvenes hasta que el FA gana. No vive una derrota, no vive la tortura, no vive el momento fundacional con éxtasis, no vive la primavera del 85…Y la pregunta es qué mierda va a aportar esta generación que vivió en los 90s.

Another potential explanation for differing perceptions regarding Ambition could lie in the complex structure within the FA. This structure has dissociated factions’ electoral success from their representation within the party (interviews with FA leaders, Yaffé 2005). Finally, for some interviewees accessing government has required the party to rely on experienced and well known leaders, preventing the emergence of new ones. Nonetheless, though imperfect, the partisan structure does provide various known channels for the emergence of new leaders (including open internal elections).227 The following excerpt presents an interesting variance: there are known—though informal— mechanisms to advance a political career. However, this leader feels that younger generations are not fighting enough for their place:

227

The last internal election, which decided part of the internal structure, was held in 2012 and gathered more than 170,000 voters. The first open internal election was held in 1997. 213

¿Cuáles son los mecanismos que existen hoy día en su partido para hacer una carrera política? En otras palabras, si Ud. hoy en día tuviera que iniciar nuevamente su carrera política, ¿cómo se imagina los primeros pasos? ¿Qué debería hacer para crecer políticamente? “No hay un mecanismo. (…) El que yo diría que es esencial para que vos te hagas un lugar en la izquierda uruguaya, es ser de la casa, que te reconozcan. Yo fui guardaespaldas de Seregni, salí del FA, volví a él. Yo soy de la casa, yo siempre fui uno de nosotros, eso es fundamental. Si sos un extraño es muy difícil. Yo creo que el mecanismo promocional en la mitad de la tabla para abajo y apenas para arriba es ser de la casa. El advenedizo no tiene chances, el recién llegado está muerto antes de empezar a correr en la izquierda uruguaya, no puede dar sus primeros pasos, no llega” ¿Considera que el partido favorece la competencia entre los líderes actuales y personas con potencial de liderazgo e interés de participar pero sin cargos formales en el partido? ¿Y su sector, favorece o no la competencia entre líderes y quienes quisieran serlo? “Ese no es un problema del FA. Es un problema más bien de que el único mecanismo de sustitución generacional y de apertura a los cambios es que alguien luche por los cambios. Cuando veas algo que no tiene mecanismos promocionales no es por culpa de que los taponean sino por culpa de los que no pelean lo suficiente.”

The electoral rules (DSV and PR, in particular) also provide a basic structure through which to pursue a political career. Thus, compared to those of the PS, DC, or PPD in Chile, the FA’s prospective leaders have better chances of defying the established elite.

As for Exit, my fieldwork revealed broad agreement on the presence of high barriers to exit. Some interviewees even mentioned that, based on the meager results of groups that split off (moderate Nuevo Espacio, then the Partido Independiente and the more radical party Asamblea Popular) leaders have learned there is no future outside the FA. Not only by their assessment of the existing rules, but also their appraisal of the history of the party and its future inform their views regarding Exit. This is in line with scholarly descriptions of the process of tradicionalición of the FA, a process through which it became a traditional party, with its own history, symbols, and myths (Queirolo 1999, Moreira 2004, Lanzaro 2004, Yaffé 2005).228

228

The concept is also extended to its format (institutionalized factions). 214

There was wide agreement on the presence of Trauma, with many politicians referring to the harsh experience under the authoritarian regime. The terrible experiences during the authoritarian regime still deeply affect the FA’s leaders and militants. For those who formed the backbone of the Tupamaros, the extended periods in isolation suffering torture, and the killing of many members is present in their speeches and actions (interview with MPP-FA leaders). Although at the foundation of the FA in 1971 they were not formally members and some factions within the FA (from the PDC to the Communist Party) even opposed their strategy and actions they eventually incorporated into the FA and reevaluated their experience in terms of the importance of forging unity on the left.

The party fought for the failed 1986 Referendum to repeal the Law De Caducidad de la Pretensión Punitiva del Estado extending amnesty from possible prosecution to military and police officers. Although civil organizations claimed that the FA was not fully committed to the second attempt to repeal the Law in 2009, this effort does show that issue’s is continued presence within the party. Also, every 20th of May the “march of silence” takes place in the streets of downtown Montevideo. On the May 20, 1976, Héctor Gutierrez Ruiz and Zelmar Michelini (a founder of the FA) were killed in Buenos Aires. As emblematic commemoration of those disappeared during the authoritarian regime, civil society organizations centered around the “organization of family of detained and disappeared” repeats the march every year. Even though no partisan signs accompany the march, FA militants and leaders attend the march and occupy a relevant place in the procession. This is an example of how Trauma still operates as a factor of bonding within the Uruguayan left. Since the Vázquez government, the FA has also implemented policies seeking to investigate those assassinations and disappearances. These deliberate actions of the FA government also show the relevance of this traumatic period for the party.

215

One established FA leader talked about the role of this experience in party formation and consolidation:

“Y teníamos un agregado, que también se dio en el caso de algunos dirigentes del FA: el prestigio se Seregni, de Licandro, y de esos militares que estuvieron presos que se podían haber ido entre su primera detención y la segunda y haber militado desde el exilio, y se quedaron y sufrieron la cárcel. La gente, aunque no lo sepa expresar así o no lo exprese así, reconoce que se jugaron con la gente, porque no se podía ir todo el mundo del país. Y a nosotros también nos reconocen lo mismo. Y esto fue una discusión muy fuerte que se dio en el MLN. Nosotros le pedíamos a Sendic que se fuera, no queríamos que cayera preso, y él nos decía “no puedo, porque si a mí me agarran yéndome por la frontera, ¿qué va a pensar la gente? ¿Que los estoy abandonando? Entonces yo me tengo que quedar”. Entonces esa decisión de quedarse, que se hizo extensiva a toda la dirección, les dio un prestigio a esos compañeros. Entonces después eso juega afectivamente, emotivamente, en la vinculación política con la gente.”

The party gained access to government in 2005. For the first time in Uruguayan history, the left also won the presidency,229 with Tabaré Vázquez leading the first government. In 2009, the FA again won the election, with José Mujica becoming president. It would be interesting to examine how the party would be affected if it were to lose control of government. Thus, the configuration of causal conditions for the FA is: ETPa. Conclusion

This chapter has provided new evidence on the effects of my causal conditions and reinforced some previous findings. EtpA is sufficient for dynamic survival, as the cases of the PC and PN illustrate. Their more than 170-year history has also helped to show how such history reinforces perceived barriers to exit. At the same time, as in the case of the PLN, once lost, their Purpose was never recovered. Finally, these parties’ histories also illustrate the waning effect of Trauma as a bonding mechanism activating retrospective loyalty. As the brief historical review shows, the configuration of causal conditions EtpA show [is the weakest configuration with which to face dramatic

229

For thorough analyses of the electoral growth and eventual success of the FA, see Buquet and De Armas (2005), Moreira (2004), Lanzaro (2004), Garcé and Yaffé (2005), and Yaffé (2005). 216

situations; retrospective or prospective loyalty thus ensure more solid dynamic survival, a point to which I will return in the next chapter.

The case of the FA reinforces the potential conflict between Trauma and Ambition. As shown above, Trauma facilitates partisan organization through the activation of retrospective loyalty but, as time elapses, it can consolidate a strong generational cleavage. The difference between the FA and the PS or DC in Chile lies in that the FA has a more developed structure with known channels and operates under more competitive rules, facilitating the emergence of new leaders. The configuration observed in the FA, ETPa, is also sufficient for dynamic party survival. However, for its reproduction it will be necessary that individual ambitions be satisfied and elite renovation be promoted.

217

Chapter 7. All Things Considered: Causal Conditions and Party Survival Introduction

Building on the evidence presented throughout Chapters 4, 5, and 6 the following sections are dedicated to analyzing the configurations for the two types of party survival. The first section will summarize some general lessons derived from the analysis of the evolution of my causal conditions for each case (party). For simplicity, I will only highlight the main causal relationships for dynamic party survival, while making some reflections on static survival. The next section presents a classic conjunctural causation analysis mostly based on my fieldwork.230 Finally, I discuss some alternative empirical hypotheses. 7.1. The Routes to Dynamic Survival: Conditions in and over Time

This section will take advantage of the narrative presented regarding the evolution of my causal conditions within each party. The following should be considered as tentative causal relationships between the evolution of my causal conditions and dynamic party survival across time.

All the parties identified as dynamic (PC, PN, FA, UDI, and PLN) had Trauma at their origin, either prior to their institutionalization as political organizations or immediately after their consolidation. However, three precisions should be made. First, the Trauma suffered by the PN and PC was in the distant past (120-170 years ago). Thus, I am not claiming that this foundational Trauma—the conflicts between them in the nineteenth and early twentieth century—still influences their type of survival. However, it did influence their dynamic survival for decades in the twentieth century as reviewed by 230

However, as argued earlier in this study, the type of conditions integrated in the theoretical argument, though analyzed today, incorporate the dynamic perspective. 218

several scholars (see Chapter 6), and today those events are part of their identity (interview with PC and PC leaders).

Second, the reverse is not true; i.e., not all parties with Trauma are dynamic, meaning Trauma is not sufficient for dynamic party survival. The PS in Chile had Trauma, but it is not a dynamic surviving party. The fact that the PS governed the country after its Trauma seems not to be theoretically relevant in terms of affecting the role of Trauma as a causal condition explaining dynamic party survival; the PLN also governed Costa Rica after the civil war, and it was dynamic. What does seem crucial is that the PS was deposed from government by a coup in 1973. Hence, its control of government preceded Trauma, which was not the case for the PLN or the FA.

Third, the case of the DC in Chile, in its foundation or early years when no Trauma was apparent, raises some doubts about the ultimate necessity of Trauma as a causal condition for dynamic survival (before 1973). Nevertheless, the Falange (antecedent of the DC) had a very tense relationship with the political establishment; though not a Trauma, these problems may have influenced the organization. Further research on this process of influence could help unpack Trauma’s variants and their influence on party organization.

There are two interesting relationships between the duration of Purpose and the occurrence of Trauma. First, if the organization had Purpose when it suffered Trauma, or the party developed Purpose close to Trauma, Purpose seems to persist. Thus, the effect of Trauma is to reinforce Purpose, and both Trauma and Purpose then influence dynamic party survival. The notable exception to this rule is the PS in Chile. However, again, if the traumatic event occurred after the party was in government, this seems to break the reinforcing relationship.231 In any case, though Trauma might reinforce Purpose, it is not 231

Because the traumatic event was related with a failed experience in government, the organization´s willingness to emphasize the intensity component of Purpose was negatively affected. 219

sufficient as a causal condition for Purpose’s presence. However, it is necessary: no Trauma, no enduring Purpose. Though this claim is bold, my data clearly show a necessary relationship. If Trauma does not precede Purpose (or the two emerge close to the same time), the latter is very unlikely to endure. The case of the PAC and PUSC in Costa Rica are clear examples of this sort of trajectory; neither had Trauma, and their Purpose did not last.

Historical review of each party also suggests that when a party loses its Purpose— regardless of the inevitability of this change— it is very unlikely to recover this condition. The PLN and PUSC in Costa Rica, PC and PN in Uruguay, and the PS and DC in Chile all had, but then lost, Purpose, and none of these parties was then able to recover this condition.232

There is also an interesting relationship between Trauma and Ambition. The PS and PPD in Chile, as well as the FA in Uruguay, show that Trauma can have a negative effect on Ambition: the pervasive role of leaders that suffered these parties’ trauma has hindered the emergence of new leaders who have no such epic and glorious past, thereby producing a situation in which a generational cleavage consolidates. Younger leaders have a hard time contending for leadership positions held by leaders with heroic histories, and this seems to delay elite renovation and the emergence of new faces within the party elite. However, this is insufficient for explaining new leaders’ possibilities for developing political careers below the highest levels. For this, known channels of elite promotion within the party become crucial. In short, retrospective loyalty activated by Trauma can enhance prospective loyalty, but it can hinder elite renovation.

232

However, it should be mentioned that each party is differently distanced to that change—from presence to absence—in Purpose. In some cases many decades while in others just a few years have elapsed. Perhaps it is too bold to claim the PAC will not be able to develop Purpose. 220

The parties that do exhibit dynamic survival forged political organizations that were based on the premise of satisfaction of ambitions under rules favoring partisan organization.233 Thus, to sustain dynamic survival over time it seems important that parties satisfy Ambition and that Exit be present.

However, as the case of the FA shows, Ambition need not always be present. The party emerged and consolidated with Ambition present, but leaders suggested that in the last few years the party has had some problems channeling ambitions and projecting new leaders, perhaps as a side-effect of its focus on successfully capturing and retaining control of government and the consolidation of a generational cleavage. As this shows, dynamic analysis is important for understanding party vibrancy, and no condition taken individually is necessary. Nevertheless, as Trauma wanes, Ambition becomes increasingly crucial for dynamic survival.

EtpA is the weakest configuration of conditions for dynamic party survival. This is illustrated by the 1960s PC and PN. Both parties lost Purpose, and it is safe to assume that Trauma had already faded. Although these conditions do not fully explain the democratic breakdown of 1973 or the political polarization of the time, it does tentatively seem sufficient for explaining their weakening qua party organizations in the turbulent 1960s. This pattern is also illustrated by the case of the PLN in Costa Rica. The party lost Purpose at the end of the 1980s. This loss of Purpose increasingly led to defections, splits, and debility of the party. The parties mentioned here also signal a final lesson to be drawn from the historical analysis: the oldest dynamic parties in each party system have this configuration: EtpA. This could lead us to a very significant conclusion, which should be furthered investigated: Trauma fades and thus has decreasing returns over time as a bonding mechanism and, if Purpose is almost never recovered as suggested above, parties have a life-cycle, and dynamic survival is sustained over a tired —i.e. diminished—vibrancy. 233

The UDI, again, first developed in the context of an authoritarian regime. 221

In sum, the basic route to the consolidation of dynamic survival would be as follows: Purpose need not precede Trauma, but when Trauma occurs and the organization has Purpose, Trauma reinforces Purpose and both reinforce dynamic party survival. This seems logical: collective organizations are more resilient if they are first and foremost united by retrospective or prospective loyalty (respectively activated by Trauma and Purpose). However, Exit and Ambition have to follow to ensure dynamic survival over time. As the evidence presented shows, achieving this developmental sequence is not so simple, because Trauma can negatively influence Ambition. If Ambition is achieved, dynamic survival is then more likely. The ideal-typical order for the presence of all the conditions for dynamic party survival could be summarized as follows: T/P—E/A234

Moreover, the evidence suggests that, after a process of consolidation, dynamically surviving parties are born out of ETPA. However, as time elapses and Trauma fades into the past, dynamic survival increasingly rests on Purpose, Exit and Ambition. Sustaining Purpose over time is a difficult endeavor, all the more so in unequal and developing societies with daunting challenges and sometimes (as illustrated by my cases) without much room for maneuver. If Purpose disappears, but Exit and Ambition persist, dynamic survival still obtains.235 However, this is the weakest configuration for dynamic survival (as also evinced through simple conjunctural causation below), and it is very unlikely that Purpose will reemerge. The FA in Uruguay shows a configuration that falls before the weakening of T, which is ETPa. As previously argued, this derives from the apparent consolidation of a generational cleavage. The ideal-typical route of dynamically surviving parties is:

234 235

The / symbol indicates the order in which the conditions emerge is irrelevant. Notwithstanding, Purpose can be reproduced over time if Ambition is granted. 222

ETPA—[ETPa]—EtPA—EtpA236

In conclusion, it is important to stress that, although there are some observable regular routes to dynamic survival, it is impossible to ascertain whether dynamic (or static) survival has ever been obtained once and for all. The above synthesis just presents some tentative conclusions about sufficient conditions for periods of dynamic survival over time, and some conclusions about a route that has accounted for durable dynamism. 7.2. Conjunctural Causation

The picture presented in this section constitutes a classic complex, conjunctural causation analysis. As such, it also admits equifinality (Ragin 1987). As in the previous section, I will mainly focus my analysis of dynamic survival on that issue. Table 7.1. Truth Table of Boolean Configurations Cases

Exit

Trauma

Purpose

Ambition

UDI FA PC, PN, PLN DC PAC PPD,PS ML PUSC, RN

1 1 1 1 0 0 0 0

1 1 0 1 0 1 0 0

1 1 0 0 0 0 1 0

1 0 1 0 1 0 0 0

Dynamic Survival 1 1 1 0 0 0 0 0

Dynamic Survival

No condition taken individually explains membership in the set of dynamic party survival. None of my theoretical conditions is sufficient to ensure such dynamic 236

It is worth stressing that the EtPA link has not been observed but is expected to occur in the FA and UDI. It also coincides with the analyzed trajectory of the PLN. The analysis of the time elapsed between the absence of T and the later absence in P could be a fascinating subject in itself. 223

survival. This section examines the observed interactions that proved jointly sufficient for explaining dynamic survival. The dynamic type of party survival is observed in parties with the following combinations of conditions: ETPA, ETPa, EtpA.237 These findings could be summarized as follows:

Dynamic Party Survival= E(A+TP+TPA)

Every combination accounting for dynamic party survival has E, high barriers to exit. However, for dynamic survival, perceived high barriers to exit have to interact at least with A. Also, proceeding with the analysis of dynamic party survival, Trauma and Purpose interact. In fact, as already highlighted in the previous section, no Trauma, no Purpose238; but the presence of Trauma does not automatically lead to Purpose (i.e., Trauma seems to be necessary but insufficient for Purpose).239 Together, Trauma and Purpose are necessary components for dynamic party survival in the absence of Ambition. Nevertheless, two important considerations must be made here:

a) T and P have to interact with E for the dynamic type of survival to be observed. Hence, TP is not sufficient for dynamic party survival; b) T and P are not necessary if and only if Exit interacts with Ambition.

Does this mean that TP are not necessary conditions for dynamic survival? First, if we apply a stricter operationalization of vibrancy240 the combination EtpA is not sufficient for dynamic survival. Thus, for a more demanding operationalization of vibrancy (see Table 4.2 above), Trauma and Purpose are both necessary.

237

As explained below, each of these combinations of factors explains a different degree of membership in the set of dynamic party survival. 238 The Trauma-less ML´s Purpose rests on one individual, its main leader. 239 Though the over-time analysis raised some doubts on its necessity, it proved to be necessary for Purpose´s endurance. 240 Instead of OR, in the second term we use AND. (See Table 3.2 above.) 224

Second, sticking with the original operationalization of vibrancy, if dynamic survival is converted into a multi-value category (in a fuzzy-set logic), we would find degrees of membership in the set of dynamic survival explained by different combinations of conditions. In this sense, more intense vibrancy—more robust dynamic party survival— is explained by the presence of T and P.

What about Ambition? Interacting with E, T, P explains the greatest degree of membership in the set of dynamic party survival and, in turn, makes it more likely that a party will reproduce its vibrancy over time.241 Still, when A only interacts with E, although it may ensure vibrancy and thus dynamic survival, this is the causally weakest combination that can account for dynamic party survival.

I am now going to relate the observed intersection of conditions with dynamic party survival as observed in my cases. The FA’s combination of conditions is ETPa. Also, I codified it as a dynamic party because it has vibrancy. Why does this intersection of conditions suffice to explain dynamic survival? As presented in Chapter 2, high barriers to Exit provide the basic precondition fostering partisan organization, facilitating collective action by reducing incentives to defect. Exit is determined in each case not only by the configuration of rules but through consideration of politicians' views of their parties’ histories, the history of the system, and prospective and hypothetical considerations of the configuration of the party system.

Trauma activates a retrospective sense of loyalty. It thereby stimulates a sense of belonging and boosts the perceived need to militate for that collective, to “fight” for the organization. Apparent divergences between members of the FA—in ideological terms as well as in socioeconomic and cultural background—are reduced by that sense of a shared traumatic past, which politicians regarded as a powerful unifying tool. In turn, 241

A alone only ensures individuals’ real prospects for success, but not necessarily through a vibrant party. 225

Purpose—together with T and E—energizes the collective and provides meaning to that belonging. Both conditions T and P in turn impact on the vibrancy of the party by endowing political action with a sense of “mission.” Purpose gives an idea about the future, a prospective reason for the collective action that stimulates partisan organization. This sense of prospective mission constantly emerged in my interviews with UDI and FA leaders. The difference between the UDI and FA is that the former better satisfies individuals’ ambitions, according to interviewees. It is also the party to which most interviewees attributed vibrancy. It is thus the party with the highest degree of membership in the set of dynamic parties.242 However, as noted by some established leaders, the UDI is not free of challenges. The party has been growing electorally in the last decade (Luna forthcoming). One of the favored tools for achieving growth at the ballot box has been promoting independents as UDI candidates. Some therefore raised concerns that satisfying independent politicians’ ambition (A) might conflict with militants or insiders waiting in line to advance both their careers and the party’s Purpose. However, all of them made it clear that the party explicitly privileged insiders. They also indicated that the UDI was fully aware of that tension and intended to control it by ensuring that independents closely adhered to the UDI’s programmatic preferences. The promotion of independent candidates is not only used by the UDI—though it is one of the parties most often having recourse to it. Interestingly, many independent candidates backed by the UDI end up joining it, thus signaling that the party manages to harmonize Ambition with Purpose. To fully survey UDI’s vibrancy as well as my causal conditions, future study should include a deeper analysis of young and prospective politicians as well as militants’ perspectives. It should be acknowledged that I am here providing a partial and tentative source of surveying the theorized causal conditions as well as the outcome, i.e., type of 242

Parties that only satisfy Ambition (PPD, RN, and PAC) do not show vibrancy. 226

party survival. The size of my sample and the agreement between my analysis of vibrancy and other measures gives initial strength to the analysis I am here developing. Nevertheless, in methodological terms, my comparative case studies should be regarded as an effort at “theory building” (George and Bennett 2005). Relative Importance of Configurations Associated with Dynamic Survival

As the preceding section suggests, there are different configurations of conditions sufficient to explain dynamic party survival. In this sense, the observed outcomes constitute a classic case of equifinality: different combinations of conditions that account for the same outcome (Ragin 1987). However, each sufficient combination is related to a different degree of membership in the set of dynamic party survival. To analyze the relative importance of sufficient configurations, they should be treated, as sets, as sufficient conditions. The more they are present in the outcome, the more important (Goertz and Mahoney 2012). To order the relevance of each configuration, I ordered from those for which there was the most to those for which there was the least degree of agreement on vibrancy. I am cautious, though, regarding generalization of the argument because conclusions are circumscribed to my cases, all of which are consolidated democracies in Latin America.

The case that most politicians regarded as vibrant is the UDI. As shown above, all the UDI interviewees claimed that their party was vibrant.243 ETPA, the configuration for the UDI, is thus the most powerful configuration and corresponds to full membership in the set of dynamic party survival.

The configuration ETPa follows, and it is observed in the Frente Amplio in Uruguay. Parties with this configuration should be considered as having .75 membership in the set of dynamic party survival. Not all interviewees considered that the party had vibrancy. A 243

The same holds true if one considers the stricter operationalization of vibrancy (see Table 4.1). 227

clear majority did, and the party meets the stricter operationalization of dynamic survival. In fact, had only hard data been used to compute vibrancy, there would have been no doubts as to the degree of vibrancy within this party. The Frente Amplio has a vast number of party locales all over the country (more than 125 244); it has a number of established instances of participation of militants and party members, of which the Comités de Base are the most traditional.245 However, consideration of interviewees’ perceptions enables one to get a deeper perspective. Some interviewees actually lamented the evolution of the party’s participation structure, claiming that, even though the Frente Amplio was founded as a movement and a coalition, it had becoming increasingly less active. Was this explained by the FA’s access to government? I discussed this issue with interviewees from the Frente Amplio, and they did not believe that being in government had affected the functioning of the party in terms of vibrancy. They did recognize that there was a tension, because, in their words, the party since had operated “at a different pace” and the government has set the agenda. The main problem, they claimed and agreed, was renewing and transforming the available instances for participation, including the available participatory structures (i.e., the Comités de Base).246

The final configuration of conditions found in cases of dynamic survival is EtpA. This combination is observed in the PLN in Costa Rica and in both the PC and PN in

244

There are also a vast number of locales belonging to the party’s different fractions. According to party leaders, the number could be well above 500. 245 As Líber Seregni, the FA’s founding leader and presidential candidate until 1994 said: “Otra vez de Nuevo: nuestra fuerza, el pueblo; nuestro campo, la calle; nuestro método, la movilización; nuestra herramienta, los comités de base.” (Seregni, 1971 in Caetano 2005, p. 230). The Comités de Base have traditionally been the structure through which the party´s militants have participated. They were created as part of the founding structure of the party in 1971. 246 However, the recent explosion of the Redes Frenteamplistas (Frente Amplio´s Networks), organized by militants show the ability of the Frente Amplio bases to create different and innovative structures to participate. The Redes were initially neglected by militants and leaders within the established structure but their impressive power to reach adherents and militants in the 2009 national election gave impulse to the demand of building new mean of participation (Interviews with Redes FA leaders). 228

Uruguay. In each of these parties, there was less agreement on vibrancy—though more than half of the interviewees did agree vibrancy was present. Actually, if we apply the stricter operationalization of vibrancy, they all fall below the threshold set for determining membership in the dynamic type of party survival—though the results for these parties were still far from the cases labeled as static. This is not surprising because these parties lacked prospective or retrospective loyalty activated through T or P. Given its dependence on how vibrancy is operationalized, parties with this configuration are considered as having a .5 membership in the set of dynamic party survival.

How can a party enjoy dynamic survival without Purpose or Trauma? This is crucial given that these three parties, in which EtpA is observed, are the most traditional parties in Costa Rica (PLN) and Uruguay (PC and PN). In the previous section, I highlighted that the oldest dynamic parties had this combination and these parties had Trauma and Purpose in the past. Static Survival

The following should be taken with greater caution than the analysis of dynamic survival. This is so because of the pervasiveness of different configurations and the divergent type of parties grouped within this type of survival. However, some conclusions can be specified. The observed configurations of causal conditions for static types of survival are: etpA, ETpa, eTpa, etpa, etPa. As already highlighted, each causal condition is individually insufficient to explain dynamic survival. This should not be surprising. For example, a party might have Purpose, but it could well be the result of a small group of leaders (or even a single individual) that does not have much effect on the rest of the organization, in close relationship with a transnational network or think tank that builds and gives the programmatic identity to the party. This can serve the purposes of a leader unable or unwilling to promote new leaders and who cannot draw others to identify with and commit to the party’s Purpose. This stylized description approximates the current situation of the ML in Costa Rica, a situation that is actually 229

recognized by its main leaders. The party is closely related to a neoconservative network, the United States CATO Institue, and its leader has repeatedly argued in favor of transforming the Costa Rican development model implemented after 1948. The rest of the party elite agree with its leaders but, in their view, the organization has not been able to combine this perspective with a dense organization, and the party mainly rests on its founding leader (Otto Guevara) and a few other members.

The presence of both E and T is also insufficient for enabling dynamic party survival and explains membership in the set of static survival. In this sense, as time passes, Trauma is still significant for understanding the past but becomes increasingly irrelevant as a unifying condition and less accepted as a justification for avoiding elite renovation. Also, party members might start feeling trapped by high barriers to Exit set by rules, and they can feel that no other options are or will be available. This scenario is close to the current situation of the DC in Chile. In this sense, rules do foster partisan organization by punishing independents and by establishing high barriers to the formation of new parties. But in the absence of Ambition or Purpose, and with Trauma waning, politicians’ perspectives on the purpose of rules setting high barriers to exit changes from stimulating collective organization to preventing massive defections. The outcome is the same: no Exit, but its meaning is different given its interaction with other conditions. Nevertheless, the party is undergoing an interesting process in terms of satisfying individual ambition by promoting new leaders (at least at the highest levels).

It could come as a surprise that only one configuration of conditions accounting for static survival contains E. One would initially tend to assume that static survival and thus ossified party organizations should occur in contexts with high barriers to exit. However, I have not analyzed “objective” high barriers to exit through a survey of the set of rules governing parties and party competition. I have considered perceived barriers to exit from the point of view of each politician, and I have operationalized it in a complex way. In this sense, it might be that politicians operating under rules setting high 230

costs to exit (for example, to forming a new party) perceive their situation otherwise, for any of a vast array of reasons. I have indeed observed this paradoxical phenomenon in my fieldwork: dissociation between “objective rules” and their theorized effects on the one hand, and politicians’ perceptions of exit barriers on the other. PS and PPD leaders, for example, mention that rules set a high cost to Exit, but they nonetheless strongly believe that the constellation of parties will change in the near future or would not be able to resist a severe economic crisis. More and more politicians have in fact been defecting from these parties in the last years. Marco Enríquez-Ominami, a PS representative from 2006-2010 who exited the PS and formed a new party, PRO, is the clearest example.

In operationalizing Exit, I related all its components with the logical AND. Thus, the value I derived for Exit was given by the “weakest link” (the minimum value of all components).247 In this sense, e is explained by politicians’ consideration of the inability of their parties or the system as a whole to remain stable, to remain with the same configuration and the same organizations, or to resist exogenous shocks. In this sense, in many cases this is given by their consideration of a party/party system that is ossified. In other cases e is given by the consideration that their parties or the system is becoming dangerously inchoate (this is mainly the case in Costa Rica and some parties in Chile). In conclusion, most e are the result of lumping together, on the one hand, what could be considered as parties in danger of becoming inchoate—unable to retain their members and leaders—and, on the other hand, ossified parties unable to remain alive in the future or to resist severe shocks. This illustrates the curvilinear effect of barriers to exit. Differences in each case would be interesting for other studies but, for this study they mean the same: parties are or will not be stable and other organizations could easily emerge, regardless of the “objective” effect of electoral rules.

247

See Goertz (2006, pp. 129-155). 231

In many cases I have observed only one of the theorized conditions. The only reason why these cases have survived is that objective rules—which in the three cases either reduce incentives to form new parties—operate to keep the status quo. Let us recall that, in the end, I am analyzing three instances of supposedly similar “institutionalized” party systems. Thus, stability, explained through conventional measures, is given.

Finally, as the observed configurations that account for static survival suggest, the absence of ambition is individually sufficient for static survival, though its presence is not sufficient for dynamic, suggesting an asymmetric causal relationship. 7.3. Discussion of Alternative Empirical Hypotheses

In this section, I will briefly discuss these alternative hypotheses that inductively emerged in my fieldwork and which are related to prior research. Are worldwide trends responsible for lack of vibrancy? One of the classic accounts of

party systems’ instability in developing democracies is a logical extension of classic modernization arguments. By this account, low levels of development structurally constrain the stability of party competition and of party systems in general (Toka 1998 (Toka 1998, Kitschelt and Wilkinson 2007). Some scholars have been more precise regarding this relationship and have highlighted the effect of timing. The pervasiveness of globalization, television and individualism, affect the role of parties as political agents (Dalton and Wattenberg 2000, Mainwaring and Zoco 2007, Sánchez 2008). The effects of these worldwide trends affect both developed and third-wave democracies (Dalton and Weldon 2007, Webb and White 2007b, p. 360). However, party systems in the latter were not consolidated prior to the emergence of these worldwide trends. Consequently, these thinkers argue, new democracies will never have strong party systems because they entered democratic politics under conditions unfavorable to parties.

232

I directly asked my interviewees to reflect on this issue. Specifically, I asked whether the current apathy, distance, or lack of trust of citizens towards their parties was explained by global trends or whether it was possible to find an explanation at the party or partysystem level. Interestingly, those that agreed that globalization, the eruption of mass media communications, and individualism were to blame for the lack of vibrancy, were members of static parties. In contrast, leaders from dynamic parties (i.e., UDI, FA, PN, PC, and PLN) did not systematically agree with this vision. Does Presidentialism inhibit dynamic party survival? The effect of presidentialism is

also often cited as a variable explaining party systems’ instability in Latin America. The relationship

between

presidentialism

and

democratic

instability

has

become

conventional wisdom for the comparative politics literature (Linz 1994, Mainwaring and Shugart 1997, Przeworski et al. 2000), and this general relationship has also now been extended for understanding instability in party systems (Webb and White 2007b, p. 362, Samuels and Shugart 2010). According to this argument, this type of regime personalizes political competition, thus endangering party organization. Samuels and Shugart (2010) provide an impressive analysis of the relationship between separation of origin and survival under presidentialism and their effect on party organization and behavior. In a nutshell, this argument highlights that fact that under presidentialism there are far more “policy switches” than under parliamentarism. Thus, for the authors it is the regime that explains pervasive party fragility across presidential democracies. In my fieldwork some interviewees raised this issue and claimed that Presidentialism is a real problem for partisan organization over time. In this sense, there are three crucial points to make. First, if this were the case, I would not have observed significant variance between countries. Second, this point was mainly raised in Chile, which Siavelis (2002) labeled as having an “exaggerated presidentialism.” Third, the pervasiveness of presidentialism was not even mentioned by leaders from the UDI, the dynamic party in Chile. Thus, even within countries I observe divergent trajectories between parties operating under the same rules. 233

Is vibrancy lost with years in government? The loss of vibrancy of one party could be

explained by the years it has been in government. Thus, the longer a party controls the Executive, the greater the chances of its losing its vibrancy. This hypothesis might explain why none of the left-of-center parties in Chile—which formed the Concertación coalition governing Chile from 1990 until 2010—are vibrant. As previously discussed, being in government generated a kind of paradox for Concertación parties: winning elections is parties’ primary goal, but winning many elections can actually hinder partisan organization. Accessing office is a major party goal. However, the more time elapses in government, the more motivation and the creation of an alternative project fade. A party has to defend its work, which is different and less appealing than having a dream or proposing an alternative. Thus, time in government might also reinforce observed static party survival.

Some Concertación leaders claimed that the best militants and politicians were consumed by their time in office, meanwhile neglecting their partisan activities. This alternative hypothesis could help to explain what happened with those parties. However, most of these parties explicitly suppressed Purpose from the very outset of democratic transition and showed signs of difficulty satisfying individuals’ ambitions. In terms of vibrancy, they abandoned and disconnected themselves from their militants and bases very early on. Moreover, politicians also argued that they had mistakenly used government as a direct substitute for their parties. Although my causal conditions precede the potential exhaustion associated with being in government, it should be acknowledged that being in government does have an impact on the organization. My theory does not contradict this possibility because I am here identifying conditions jointly sufficient for dynamic and static survival, not necessary conditions. Is government success or failure key to explaining vibrancy? A hypothesis based on

government failure or success could eventually explain the votes a party receives in one election. However, it is neither necessary nor sufficient to explain vibrancy and, in turn, 234

a party’s degrees of membership to the dynamic type of survival. One example that supports this claim is the PC in Uruguay. In 2001 and 2002, the country faced the most severe crisis of its modern history (Porto 2002, Paolillo 2004). The PC suffered electorally in the 2004 national election, in which it received the lowest level of support in its history. However, the party did not collapse as an organization. It underwent a process of organizational and leadership renovation, mainly by serving individuals’ ambitions. New leaders, particularly Pedro Bordaberry, won control of the party and transformed its leadership. Is corruption a causal condition for lack of dynamic survival? Many scholars have

been concerned by increasing corruption scandals in Costa Rica248 and their negative effect on the parties and party system (e.g., Furlong 2001, Raventós 2005). It is indisputable that this problem is indeed significant. However, all such cases of corruption have been in parties that had lost Purpose—and within which Trauma had waned. Thus, corruption could be associated with the weakening of party vibrancy as a result of the absence of Purpose and Trauma. Although not demonstrated in this study— given the preliminary assessment of of the effects of different sequences of presence/absence of conditions and combinations of conditions, it appears plausible that the absence of Ambition may also typically precede the eruption of massive cases of corruption. Party organization as such: I have analyzed cases in which different types of parties

with different types of organization experienced the same outcome. The pair of parties most interesting in this regard is that of the UDI and FA. They are very different organizationally but both regarded here as dynamic parties. The UDI has a vertical organization, with a small group of leaders controlling the party since the assassination of Jaime Guzmán. They are known as the “coroneles” and have a strong influence on 248

Other scholars have analyzed the negative effects of corruption in party systems´ stability in Latin America. See e.g. Mainwaring, Bejarano, and Pizarro (2006). 235

determining the direction of the party. On the other hand, the FA has a very complex structure of power composed of a number of instances privileging consensus and giving a strong role to its militants. Thus, there is no observed necessary or sufficient relationship between parties’ vibrancy and their specific form of organization. Conclusions

This chapter has offered some general insights derived from the empirical chapters of this study. First, I have presented a tentative route followed by dynamically surviving parties. After the four of my causal conditions consolidate, and ETPA thus becomes consolidated, a sequence that amounts to a life cycle follows. Parties first lose Trauma as a bonding mechanism, and later—though hypothetically—Purpose becomes absent. Thus, dynamic survival in the oldest parties rests on Exit and Ambition.

I have also provided a classic complex and conjunctural analysis. In this sense, different combinations of configurations are sufficient to explain dynamic party survival, leading to this minimization formula: E(A+TP+TPA). However, I have also shown that each sufficient configuration of conditions shows different degrees of membership in the set of dynamic party survival, with presence of all my conditions being the strongest configuration. The EtpA configuration is the weakest of all, and it is observed in the oldest dynamically surviving parties, reinforcing the notion of an apparent life-cycle. Finally, as the analysis of the static type of party survival shows, none of my causal conditions is individually sufficient to explain dynamic party survival.

236

Chapter 8. Conclusion: Towards Consolidating a Dynamic Perspective

The following conclusions underscore the value of a dynamic perspective in comparative political analysis. Such a perspective enhances understanding of political parties’ survival over time, and also has analytical payoffs for other studies. I will first summarize the core components of my study. Second, I will extend the empirical validation of my argument. The third section will discuss some tasks for the road ahead in this line of research. Fourth, I close this study with broader theoretical implications. I will discuss the impact of dynamic party survival on democratic representation. 8.1. Core Components of the Study Research Questions and Puzzle

How do political parties remain significant and vibrant organizations? Why are some parties able to resist the numerous challenges that affect developing countries? What explains their ability to retain a vibrant organization over time? This study has provided an explanation for the resilience of dynamic parties in Latin American consolidated democracies.

Several theoretical explanations have been provided to shed light on what I have labeled party survival. As reviewed in Chapter 1, some researchers have focused on explaining stability. Referring to the problems of “over institutionalization”, Andreas Schedler has claimed that: “Hyperstable party systems are kingdoms of boredom.” (1995, p. 16) He was the first to formally distinguish between types of institutionalization and highlight the potential analytical hazards of focusing excessively on stability. Another major strand in the literature has explained parties’ ability to resist dramatic instances of change through adaptation. But the emphasis on shocks as incentives to adapt has neglected the seemingly more boring, though ultimately also significant, aspect of change occurring in times of relative stability, a process crucial in explaining parties’ 237

survival over time. Others concerned with change have sought to describe the evolution of partisan organization over time. Their descriptions, however, have not provided theoretical explanations for survival amid change.

To build a theory of party survival, I have analyzed political parties in Chile, Costa Rica, and Uruguay. The three countries all rank highly within the region in terms of democratic stability, governance, and socioeconomic success. However, their parties have evolved very differently, signaling different outcomes despite their numerous initial similarities.

Organizationally, parties in Costa Rica and Chile have become increasingly weak. They have lost their former role between electoral cycles, and they have been constantly under attack or losing influence in public debate. Moreover, new—though mostly “flash” or “flat”-- political parties have emerged in Costa Rica (Mustillo 2007). Chile has meanwhile seen an increasing number of independent candidates or new political groups, most not yet formally established as political parties. Some are former leaders of established parties like Alejandro Navarro, who left the PS, and his Movimiento Amplio Social or, most successfully, Marco Enríquez Ominami, who left the PS to form the Partido Progresista. Others are young new entrants like Giorgio Jackson and Revolución Democrática, Francisco Figueroa, Gabriel Boric, and Daniela López from Izquierda Autónoma or outsiders like Franco Parisi. Although there have been a number of independent candidates ever since the 1989 election, in most cases they were not part of new organizations or were supported by one of the major coalitions (Concertación or Alianza). Today, in contrast, an increasing number of independents are not supported by either of the major electoral coalitions. In Chile and Costa Rica, parties have suffered a constant decline in citizens’ appreciation. Specifically, party identification, trust in parties, and, the stability of citizens’ declared voting preferences have shown a gradual but systematic drop 238

(LAPOP). Neither country has ever had an outstanding score on these indicators, but their scores have been observably declining over the last decade.

By contrast, Uruguayan parties have remained vibrant political organizations. They are still central and valued players within the political arena—at least in comparison to those of Chile and Costa Rica. Uruguayan major parties organizations function permanently, and no significant outsider or new parties have emerged.

This dissertation has focused on the supply side—parties qua political organizations— and their ability to survive amid change and across different types of junctures. Thus, I have focused on explaining different types of party survival. Deductive and Inductive Construction of the Argument

I have used the term survival to label the dependent variable, and vibrancy to describe an attribute of parties surviving dynamically. As argued in Chapter 2, the idea of vibrancy connotes an organization that is alive beyond electoral cycles. Dynamic parties are those that combine stability over time (observed through their age) and the ability to remain vibrant.

I have conducted a comparative case study applying a dynamic approach to institutional analysis. For this goal, I have combined rational choice and historical institutionalism’s main claims regarding choice, on the one hand, and structure and historical events, on the other. These approaches constitute the foundations of my causal conditions explaining party survival, inherently a process that unfolds over time with both ambitious, established leaders and prospective ones playing key parts.

Neither tradition of institutional analysis has alone proven fully conducive to precise analysis of the process of survival in and of itself. Recently, some HI and RCI scholars have introduced more dynamism into the study of institutions and tried to better 239

integrate choice and structure in order to capture processes occurring over time (e.g., Thelen 2003, Katznelson and Weingast 2005, Mahoney and Thelen 2010).

To still better integrate choice and structure and thereby capture survival, I have focused on parties qua organizations (see Chapter 2). Political organizations are an optimal focus of dynamic analysis explaining survival. Political organizations unmistakably integrate choice and structure. Given my consequent emphasis on political organizations, I have incorporated Hirschman´s (1970) organizational theory. Part of my theory explaining organizational survival derives from Hirschman’s (1970) analysis. His idea of exit, related to conventional institutionalist analyses, is incorporated directly into my own causal model. If barriers to exit are low, I follow Hirschman in arguing, political organizations deteriorate or never consolidate. Meanwhile, Hirschman’s concept of voice is reflected in my emphasis on Ambition as a causal factor. In line with conventional RCI claims, I consider that satisfaction of individuals’ ambitions within parties marks a realization of voice, since it is the most effective way of ensuring their voices are heard within their partisan organizations. In applying Hirschman’s argument on voice to political parties, I thus build upon the political party literature founded on rational choice.

However, although parties are indeed formed by ambitious individuals, there are two peculiarities of political parties warranting greater emphasis than typically granted in RCI analyses. First, Ambition is a condition involving collective dynamics; in particular, it often involves intergenerational conflicts, or conflicts between party elites and other members. Organizational dynamics resulting from such interactions are analytically significant, and are independent of the actions of any particular individual. Unlike conventional RCI analyses, mine therefore treats Ambition as a variable rather than an axiom, while focusing on collective organizations rather than individuals. 240

Although important parts of my analytic framework were thus deductive in origin, inductive processes also contributed to its formation. My extensive fieldwork allowed me to develop the original formulation of my argument inductively and to improve my application of Hirschman´s ideas to explaining party survival. First I found that Hirschman´s third category, loyalty, could be analytically divided into retrospective and prospective loyalty, respectively reflecting what I call Trauma and Purpose. In this sense, the analytic tools of HI led me to pay close attention to the influence of past events on the consolidation of a political organization. Fieldwork and extensive reading about each country also contributed to refining my claims about how my causal conditions really interact. Main Empirical Findings

I have identified two types of party survival: static and dynamic. Their difference lies in the presence or absence of a vital attribute: vibrancy. Dynamic parties are not only established organizations—observable through their age—but also vibrant ones.

The explanation for dynamic party survival advanced throughout the study has emphasized the complex interaction between four causal conditions: Exit, Trauma, Purpose, and Ambition. These conditions are jointly sufficient for explaining the difference between dynamic and static party survival. When the four conditions are present, dynamic survival is observed. However, other combinations also produce this type of party survival yet to a lesser degree. The conjunction of Exit, Trauma, and Purpose is also sufficient for dynamic party survival, as is the presence of both Exit and Ambition. I have also concluded that each sufficient configuration of conditions has a different degree of membership in the set of dynamic party survival. The presence of all the conditions has full membership while presence of Exit and Ambition alone has the weakest, and perhaps debatable249, membership in this set. 249

As shown in Chapter 7, the determination of membership in the dynamic party survival set is too dependent on the operationalization of one attribute of vibrancy. 241

A preliminary analysis of the historical evolution of causal conditions in my cases has also suggested a basic route to and trajectory of dynamic party survival. Trauma and Purpose precede Exit and Ambition. Collective organizations are more resilient if they are first united by retrospective or prospective loyalty (respectively activated by Trauma and Purpose). However, Exit and Ambition need to then follow to facilitate a party’s dynamic survival over time.

This sequence is not simple to achieve; the evidence has shown that Trauma can negatively influence Ambition. If the satisfaction of individuals’ ambitions is achieved, dynamic survival is more likely over time. As time elapses, Trauma fades as a bonding mechanism, and dynamic survival increasingly rests on Purpose, Exit and Ambition. Sustaining Purpose has also proven difficult; parties’ ability to do so seems to be related to the daunting challenges they must confront in developing countries (e.g., low degrees of maneuver, lack of resources, and weak state capacity). If Purpose disappears, but Exit and Ambition persist, dynamic survival may still be observed. However, this is the weakest configuration for dynamic survival, and as shown throughout my empirical chapters, it is very unlikely that Purpose will then reemerge. The case of the FA introduces a potential deviation to this hypothetical historical path. The party’s observed configuration is ETPa, i.e., no Ambition is observed. Trauma will eventually wane, and EtPa is not sufficient for dynamic survival. However, rules of the game and the erosion of T´s negative influence on the renovation of elites will likely enable A increasingly to be satisfied. To put it bluntly, A can be recovered once the generation shaped by T steps aside.

In sum, dynamically surviving parties seem to undergo the following cycle. When they emerge and consolidate, they are shaped by all four causal conditions included in my theoretical model. This configuration of conditions (ETPA) is sufficient for the greatest degree of membership in the set of dynamically surviving parties. Of the parties studied in this project, the one in which the four conditions are all observed is the UDI. Then, as 242

Trauma fades, the observed configuration is EtPA. This configuration could be hypothesized as characterizing the PLN before the 1980s neoliberal turn and will probably characterize the UDI and FA in the near future. Finally, the configuration may narrow to EtpA, as in the cases of the PLN in Costa Rica, and the PC and PN in Uruguay. The analysis has shown that the oldest dynamically surviving parties exhibit this last configuration. This finding corroborates my theoretical expectation that Trauma should exhibit decreasing returns, and that Purpose is a condition difficult to maintain over time, and very rarely recovered once lost. Analysis of my cases also suggests that dynamic party survival positively affects parties’ resilience. The dynamically surviving parties analyzed in this study have overcome challenges of different sorts, e.g. economic crisis, cycles of political contention, and long periods of electoral defeat, among other potential disruptions for the organization.

The PLN survived the critical years of the 1980s, and it is now the least affected by citizens’ apathy and mistrust of parties. This relative immunity cannot be explained simply by the party’s control of government. The crisis of representation in Costa Rica can be traced back to 1998—and may have been silently developing even before (Seligson 2002, Lehoucq 2005). Since 1998, the PLN has been both in government and in the opposition. The party´s relative immunity to decline rests on its dynamic survival of many previous critical junctures. The PC, PN, and FA in Uruguay have survived different challenging junctures: the turbulent 1960s, the authoritarian regime, and the harshest economic crisis of Uruguay’s modern times. Finally, the UDI emerged as the party closely associated with Pinochet’s authoritarian regime, and it then had to wait almost twenty years before its electoral coalition (with RN) won the first Presidential election since the transition to democracy in 1990. Thus, the political scenario was challenging for the party. However, it has survived dynamically over the years. Nevertheless, since it has not undergone the type of major challenge that the PC, PN,

243

and Uruguayan FA have faced, it is not possible to claim that it has yet acquired the same degree of resiliency for facing future crises.

However, that resiliency might, as suggested above, reach a point of exhaustion—as given by the combination of conditions explaining a lesser degree of membership in the set of dynamic party survival. The dynamic type of survival is not a condition that, once achieved, can never be lost. A once vibrant party can eventually stagnate and become ossified; this is a danger it must continuously overcome to continue surviving in the dynamic sense. The life-cycle notion introduced in Chapter 7, driven by the gradual loss of Trauma and Purpose indicates that parties’ resiliency may eventually be prone to exhaustion. Also, as analyzed in Chapters 3 through 7, different sufficient configurations for dynamic party survival account for what could be labeled different levels of resiliency. 8.2. Empirical Reach of the Argument Two Cases of Party Survival

Assessing whether my theory has broader applicability would require a wide-ranging analysis beyond the bounds of this study. However, the present section offers some illustrative evidence regarding the external validity of my theoretical argument. For that purpose, I focus on two cases: Acción Democrática (AD) in Venezuela and the Partido dos Trabalhadores (PT) in Brazil. AD illustrates the evolution of a party that gradually became statically surviving and eventually collapsed. PT represents a case of dynamic survival in the context of an inchoate party system. Both cases thus represent interesting paradoxes that can be addressed by applying my theory. In the case of the AD, the obvious question is that of how it decayed—and then collapsed—despite many years of strong institutionalization. In the case of the PT, the question is how that vibrant organization emerged out of Brazil's inchoate party system.

244

Acción Democrática in Venezuela: The Collapse of the Epitome of Institutionalized Parties

Prior to 1998, Venezuelan parties were among the most stable in the region and were considered the epitome of institutionalized parties. Acción Democrática (AD), like the whole Venezuelan party system, was described as a successfully institutionalized party shedding light on the challenges of consolidating democracy in the region (Kornblith and Levine 1995). However, it soon showed signs of decline, and with the start of the Chavismo era in 1998 the AD almost disappeared from the political scene, as did Venezuela’s other institutionalized party, the COPEI.

Existing arguments (e.g., Morgan 2011) explain the actual collapse of the AD and the Venezuelan party system. My theory complements this work by helping to explain the initial vulnerability of the AD that preceded and contributed to precipitating its collapse. The AD was a statically surviving party. It was therefore not particularly resilient when it came to confronting the dramatic junctures of the 1980s and 1990s.

Due to the process that preceded consolidation of democracy in Venezuela in 1958 (Coppedge 1994, p.9), the AD emerged with Trauma. Before AD was formally founded and legally recognized in 1941, its leaders, organized in proto parties, suffered persecution and had to operate underground.250 Once founded, the AD quickly developed a complex organization that would last for decades. AD was the social democratic party in Venezuela, and thus emerged with a defined Purpose. However, the Pacto de Punto Fijo signed in October 1958 by AD, COPEI, and the URD as well as other agreements signed afterwards (especially the Programa Mínimo de Gobierno) ultimately contributed to undermining each pacting party’s Purpose—mainly in terms of intensity (Coppedge 1994). In the late 1980s, under the second presidency of Carlos Andrés Pérez (1989-1993), the AD—especially the President himself carried out a 250

Even though it was the most important political organization in Venezuela by the beginning of the trienio in 1945, a coup led by Marcos Pérez Jiménez in 1948 outlawed the AD. 245

dramatic policy switch by adhering to neoliberal reforms (Weyland 2002)251, and definitely lost its Purpose altogether. This loss of Purpose culminated with the ideological convergence of COPEI and AD in the late 1980s and 1990s (Morgan 2011).

The system of consensus ensuing from the Pacto de Punto Fijo created high barriers to exit, mainly through the exclusion of both the left and the right from political agreements. Although the electoral rules granted small parties the possibility of representation, informal institutions set forth by these agreements increasingly generated incentives for voters to cast their ballots for AD and COPEI. Coppedge (1994) and Crisp (2000) emphasize institutional and organizational rigidities helping to explain the consequent inability to adapt to changing—and dire—circumstances. Two political reforms (direct gubernatorial elections in 1989 and electoral system reform in 1993) that sought to lower exit barriers were implemented. However, they later proved relatively ineffective at re-engaging citizens with established parties.

Regarding Ambition, it is very hard to assess its value over the course of history without an in-depth study.252 However, scholars studying the golden era of Venezuelan institutionalized parties (until the 1980s) highlight the pervasive role of the party´s leadership in determining its decisions, and its increasing problems in elite renovation (e.g., Martz 1992, pp. 101-2). The literature suggests that 1990s demands for reforming mechanisms for selecting candidates may have reflected such problems within AD, and also COPEI.

As this brief narrative suggests, AD became a case of static survival. Its stability was not accompanied with continued vibrancy. The increasing lack of Ambition and Purpose, the waning of Trauma, and high barriers to exit (which were actually lowered in the 1990s)

251

However, this policy switch generated tensions within the party (see Morgan 2011). For AD´s early years, Martz (1966) describes an interesting generational cleavage in the party’s elite, additional evidence of the potentially conflictive relationship between Trauma and Ambition. 252

246

help explain the transformation of AD. As the party became a case of static survival, it proved increasingly unable to cope with successive shocks which, for reasons not included in my theory, eventually ended in the party’s collapse in 1998 (see Morgan 2011). As in Costa Rica or Chile, the loss of vibrancy affected the party’s relationship with society, first through increasing levels of electoral abstention and distance between citizens and the party (see Morgan 2011 for a thorough review).

The collapse of AD marked the dramatic end of a stable party previously regarded as institutionalized. This outcome does not imply that the same will necessarily happen in Chile or Costa Rica—also countries with static party survival. However, it is striking that the same process of diminishing vibrancy of party organization—explained through my causal conditions—has in Chile and Costa Rica translated into the same results in terms of democratic representation, i.e., an increasing distance between citizens and parties. The PT in Brazil. When one party is not Enough

Brazil has typically been depicted as having an inchoate party system (Mainwaring and Scully 1995, Mainwaring 1999). However, the Partido dos Trabalhadores (PT), founded in 1980, has recently been described as a successful case of positive institutionalization, i.e., as a party that has been able to consolidate as a stable organization (Zucco 2010)253. Hunter (2010) summarizes the groups that originally coalesced around this leftist party in Brazil: Christian base communities, intellectuals, and most of all, labor organizers (p. 22). The party emerged in the context of a repressive authoritarian regime that lasted until 1985. Major leaders and grassroots members from the foundational groups suffered political persecution. Thus, like other leftist parties in the Southern Cone, the organization was born with Trauma.

253

The author does describe a “non-institutionalized stability,” but referring mainly to the party´s relationship with voters and society in general. 247

Samuels (2004) and Hunter (2007, 2010) describe the ideological orientation of the party during its early years and its gradual transformation. In the early years of the party, both T and P were present. As the party grew stronger electorally, it moderated its program and made strategic decisions. Although the PT moderated, especially since 2002 (Power and Zucco Jr 2009), it did so gradually and without policy switch or gradual “leapfrogging” transformation into a rightist party.254 Nevertheless, in-depth analysis of the sort provided in my study is needed to better assess moderation’s impact on Purpose within the PT.

Hunter uses rational-choice foundations to analyze the PT´s strategies in terms of its relationship with society. Useful though this analysis is, it is very difficult to capture Ambition as conceptualized in my study. However, her evidence and that of Samuels (2006) and the analysis in Zucco (2010) appear to suggest that the party has been able to promote new leaders.

In terms of barriers to exit, the authoritarian setting set the highest possible barriers both to entry and exit. In terms of entry, in the context of Brazil’s National Security doctrine, the PT´s leftist imprint was not at all to the taste of the military junta. Hunter says: “Working against the odds of opposition party formation during the time, activists went to great lengths to have the party officially registered and recognized” (p. 22). In terms of Exit, once organized, it seems clear that only by staying together could this novel organization survive—much as shown for the FA in Uruguay. With its federal system and open lists, among other traits, the electoral system in Brazil since democracy was restored could be considered an extreme case of low barriers to exit).255 However,

254

Moreover, Zucco (2010) rightly says: “By the late 1990s, radicals who remained in the PT’s leadership could not claim that moderation would amount to a betrayal of the party’s supporters” (p. 11). 255 See Samuels (1999, p. 495) for a summary of electoral rules for legislative elections in Brazil. The introduction of electronic voting has promoted a more partisan vote and lowered percentages of invalid and null votes (Hidalgo 2012). 248

Samuels (1999) analyzes label votes256 in legislative elections in Brazil and demonstrates that the PT was the party regularly garnering the highest percentage of them (p. 513). The author identifies a set of mechanisms the PT has developed that have set high barriers to exit: “…PT politicians have distinguished themselves from others…by making the costs of individualistic behavior too high. Several sanctioning mechanisms, including controlling access to the label and credibly threatening expulsion, achieve this result” (p. 513).

Thus, the PT would tentatively appear to have the following configuration: ETPA. In terms of vibrancy, available research suggests that the PT is the most vibrant party in the Brazilian system. For example, Samuels (2006) highlights that the PT is the only party with a dense organization that works permanently and mobilizes its militants. More importantly, it is “the only party with relatively widespread member involvement in its governance” (Samuels 2006, p. 22). Thus, the PT could be considered dynamically surviving.

The recent wave of massive protests in Brazil (see Folha de Sao Paulo, June-July 2013) might appear to contradict dynamic party survival’s positive impact on democratic representation. However, because the PT is the sole dynamically surviving party in Brazil, the party system there cannot be considered as dynamically surviving. A party system cannot be considered dynamically surviving without having at least “two to tango” (see below). This deficiency could help explain the party system’s inability to channel citizens’ demands through institutionalized mechanisms—and the protesters’ unwillingness to appeal to parties and to participate in party politics.

256

Citizens can cast a vote for a politician or simply for a party label. 249

8.3. The Road Ahead

The dissertation has tried to deal with multiple and challenging tasks: concept formation, theory building, and methodological innovation. In this sense, it is evident that there are a number of pending tasks for the future.

First, better measurements of vibrancy are needed. I have here mostly relied on secondary sources (for the historical analysis) and in-depth interviews to compute the presence of this attribute. The conclusions, though, are in line with different sources of evidence. For example, in the last internal election of the Frente Amplio, more than 170,000 militants participated in the election of the President of the party. In comparison, in the recent open convention for the selection of a presidential candidate of the PAC for the 2014 national elections in Costa Rica, only 23,000 voted (La Nación July, 23, 2013). In Chile, internal elections of the DC or PS usually involve some 20,000 to 30,000 people (Interview with DC and PS leaders).257 Still, measurements that could bring more reliability and precision to systematically assessing parties’ vibrancy need to be designed.

A potential test of vibrancy of a party would be case studies in least-likely settings. In this sense, assessing parties’ vibrancy in contexts of extreme deprivation and in towns distant from the center of the countries in which they operate could work as plausibility tests for my theory. Given the socio-territorial inequality and weak state capacity in developing countries, both approaches would imply decisive tests for party vibrancy. It is parties with continuous presence in such settings that are vibrant organizations.

257

Even though recent primary elections elicited more participation in Chile, they decided the election of the presidential candidate and were thus mostly personalized. In Uruguayan primary elections, a greater percentage of voters participate in comparable instances, suggesting that turnout was not that high in comparative perspective 250

Although I have advanced an analysis of the evolution of my causal conditions and its theoretical consequences, a more systematic and thorough approach is warranted. I am currently developing an analytic tool combining QCA with comparative historical analysis’s main foundations—particularly path dependence. This tool would help make dynamic, over-time analysis of a research problem like the one in this study more parsimonious. Such an analytic tool would consider configuration of conditions’ evolution over time and the effects of that evolution.

Also, more systematic and precise measurements are warranted, for example of Purpose. Similarly, a wider sample of politicians within each case is needed to more precisely and reliably assess Ambition. In particular, future research should focus more on younger leaders. Analyzing this group in greater detail could also shed light on the generational cleavage in terms of Trauma, and its decreasing returns as a bonding mechanism.

In this study I have laid out the foundations of a theory of party survival in Latin American consolidated democracies. The road ahead demands improvements in terms of measurement; more refinement in terms of methods to pursue a diachronic analysis of an argument that rests on complex causation; and, finally, the validation of this argument in different contexts with a view to establishing the theory’s external validity. 8.4. Dynamic Party Survival and Democratic Representation

Although the core of this study is centered at the party level and type of party survival is the dependent variable, I advance a simple rule for computing types of party system survival: two are needed to tango. If a party system has at least two electorally relevant, dynamic parties, the system as a whole is considered a dynamically surviving party system. As a minimum, two dynamic parties are needed to ensure a minimal set of options for the electorate. Where there are at least two dynamic parties, they should also

251

normally be expected to reinforce each other’s traits through competition, thereby reducing the chances of either becoming ossified.

What is the impact of a dynamic party system? A dynamic party system positively affects democratic representation through a relationship between citizens and parties. Dynamic parties are vibrant organizations. This kind of parties can thus be expected to fulfill the theoretical role attributed to parties in terms of democratic representation: they reduce transaction costs. Given their stability, their “brand label” is known. Moreover, dynamic parties are vibrant, i.e., they are active organizations that work permanently, have a constant presence in the territory, have mechanisms of participation, and build ties with civil society. It is thus expected that when these two properties are both observed over time in at least two parties, citizens will have a close relationship with political parties and identify with them. Such a party system will thus be effective in promoting democratic representation.

My preliminary evidence supports this claim. In Uruguay, where the three major parties are dynamically surviving, citizens show the highest degrees of identification with parties, most trust in political parties, and lowest degrees of indecisiveness in terms of their electoral preferences (LAPOP 2010, 2012)—which brings the necessary degree of predictability to electoral contests. Moreover, the evidence from successive LAPOP surveys suggests that Costa Rica and Chile258 have experienced a process of gradual decay in the relationship between citizens and parties (see LAPOP 2006, 2008, 2010, and 2012). Thus, rethinking the notion of institutionalized parties and party systems and identifying two types of party survival provides better analytical power when it comes to explaining the impact of parties on democratic representation.

258

In Chile, only one party (the UDI) is dynamic. As in Costa Rica, the party system therefore does not qualify as a vibrant one. 252

Bringing Dilemmas Back In

The preliminary assessment presented above suggests that the combination of stability and vibrancy (i.e., dynamic party survival) is positively related with democratic representation. However, good things do not always come together. Two of the theorized conditions that have been highlighted as components of sufficient combinations of conditions for dynamic survival contradict the notion of democratic consolidation. For vibrant party survival to develop, traumatic events (e.g., civil wars, coups) have to occur to bolster retrospective loyalty. Such conflicts run deep within societies and create divisions that persist for decades, weakening the prospects for good democratic governance. Moreover, a coherent and intense Purpose on the part of two parties can contribute not only to a dynamically surviving party system but also to systemic polarization, endangering the prospects for democracy. Nevertheless, Trauma fades, and Purpose’s intensity is never constant. Indeed, parties tend to lower the intensity of their Purpose when their electoral prospects look promising. In this sense, by putting pictures in motion, it is possible to understand how these potential contradictions evolve and lose their disruptive effect in terms of democratic consolidation.

Indeed, all the current dynamically surviving parties in Chile, Costa Rica, and Uruguay suffered Trauma at some point during their consolidation qua political organizations, thus activating retrospective loyalty and reinforcing Purpose. 259 As democracies in Latin America become more consolidated, it is very likely that this condition that facilitates retrospective loyalty may not be reproduced. The emergence of new stable and vibrant parties therefore will be hard to achieve.

259

Moreover, some parties now regarded as static also have a traumatic past and were once dynamic parties. 253

This pessimistic account is also related to a dilemma mentioned earlier in this study: democracies in developing countries are boring and stable or intense and unstable. As democracy becomes consolidated and the civil wars or dramatic conflicts of the past cease to be observed, it will become very hard for new dynamic parties to form and survive.260 In this sense, consolidated democracies without dynamically surviving parties are unlikely candidates for generating additional such partisan organizations in the future.

Conditions for successful party building have, however, not been the emphasis of this study. I have instead focused on party survival and therefore on parties that are already institutionalized. Focusing on the conditions for building a party—regardless of its ultimate success at surviving—in the context of contemporary democratic consolidation could therefore contribute to better evaluating and refining this preliminary, pessimistic account.261

Finally, in Chile and Costa Rica politicians are well aware of the crisis of representation and legitimacy in their respective societies. Indeed, various expert committees and bills have been proposed for reforming institutions precisely with a view to reinvigorating representative democracy. In Chile, for example, proposals to reform the peculiar binominal electoral system or introduce congressional term limits have abounded. In Costa Rica, a number of reforms have been studied in the last years; these have recently 260

The success story of the Die Grünen (The Greens) in Germany might change this pessimistic note:, leading one to conclude that, if a sufficient level of socioeconomic development is achieved, societies can forge vibrant and stable political organization without the need for Trauma as a bonding mechanism. Given the different context in which the Green party operates, other causal conditions may need to be considered to explain its successful survival, but both Purpose and Exit are certainly present. 261 For classic accounts of party and party system formation, see Lipset and Rokkan (1967), Duverger (1954), Sartori (1976), and Cox (1997). For studies of the rise of parties in developed countries, see, e.g., Kitschelt (1989b)or Kalyvas (1996), and for recent studies on the rise of new parties and party systems in developing countries, see, e.g., Mustillo (2007), Hicken (2009), and Levitsky et al. (forthcoming). There are also a number of studies focusing on the determinants of national party-system formation. See, e.g., Caramani (2004) and Chhibber and Kollman (2004). 254

been summarized in the document Propuestas para Fortalecer la funcionalidad y calidad de la democracia costarricense (Pacheco coord. 2013). These efforts are remarkable. However, it is worth highlighting that traditional parties in Venezuela also tried to react by re-designing political institutions, yet proved unsuccessful. I believe that institutional reforms are not sufficient to reinvigorate citizens’ relationship to politics. Reinforcing vibrant political organizations may also be necessary. This endeavor is difficult and does not depend on mere institutional adjustments. It is closely tied to complex historical processes—like the occurrence of a traumatic event.

255

Appendix 1. List of Interviewees

Country

Name

Party

Chile Chile Chile

Alberto Cardemil Alberto Espina Alejandro Bahamondes

RN RN PPD

Position (current at the time of the interview or last position held) Representative Senator Ex Secretary General of PPD

Chile

Alejandro Foxley

Ex DC

Former Minster of Finance

Chile Chile Chile

Álvaro Elizalde Ana Lagos Andrés Chadwick

PS PS UDI

Vice-President of the PS Party Finances Senator

Chile

Andrés Velasco

N/A

Former Minster of Finance

Chile Chile Chile Chile Chile Chile Chile Chile

Andrés Zaldívar Camilo Escalona Carlos Larrain Carlos Ominami Carolina Tohá César Valenzuela Mass Claudio Orrego Cristián Monckeberg

Senator Senator President of RN Former Senator President of PPD President of the PS Youth Mayor of Peñalolén Representative

Chile

Eduardo Vergara

DC PS RN Ex PS PPD PS DC RN Ex PPDPRO

Chile

Ena Von Baer

UDI

Chile Chile

Ernesto Velasco Eugenio Tironi

PRSD Ex PPD

Chile

Francisco Aleuy

PS

Chile

Gonzalo Müller

UDI

Chile Chile

Guido Girardi Ignacio Walker

PPD DC

Chile

Jaime Bellolio

UDI

Chile

Jaime Mulet

Chile Chile

Joaquín Godoy Jorge Manzano

Ex DC– Ex PRI RN UDI

Chile

José Antonio Gómez

PRSD

Chile

José Antonio Kast

UDI

Chile

José Antonio Viera Gallo

PS

Chile

Jovino Novoa

UDI

Chile

Juan Antonio Coloma

UDI

Chile Chile

Juan Carlos Latorre Manuel José Ossandón

Chile

Director Fundación Progresa Minster of the General Secretary of Government Secretary General of PRSD -----Member of the Central Committee of PS Advisor Ministry of Secretary General of Government President of the Senate President of DC and Senator Member of Political Commission of UDI

Date of Interview April, 27, 2011 April, 7, 2011 April, 4, 2011 November, 4, 2010 March, 30, 2011 December, 7, 2010 May, 2, 2011 November, 9, 2010 April, 12, 2011 April, 21, 2011 March, 29, 2011 April, 12, 2011 March, 18, 2011 April, 15, 2011 April, 15, 2011 March, 21, 2011 November, 5, 2010 April, 26, 2011 April, 19, 2011 April, 1, 2011 April, 20, 2011 May, 24, 2011 April, 8, 2011 June, 9, 2011 March, 23, 2011

Former Representative

May, 25, 2011 April, 18, 2011 April, 5, 2011

DC RN

Representative Executive Secretary UDI President of PRSD and Senator Representative Former Minster Secretary General of the Presidency Senator Senator and President of UDI Representative Mayor of Puente Alto

Marcelo Díaz

PS

Representative

Chile Chile Chile

Marcelo Schilling Marco Enríquez-Ominami Nicolás Monckeberg

PS Ex PS-PRO RN

Chile

Osvaldo Andrade

PS

Chile Chile

Pablo Longueira Pedro Browne

UDI RN

Representative President of PRO Representative Representative and President of PS Senator Representative

May, 4, 2011 April, 25, 2011 November, 10, 2010 May, 11, 2011 April, 4, 2011 April,18, 2011 March., 23, 2011 November, 29, 2010 April, 11, 2011 April, 1, 2011 May, 4, 2011 March, 28, 2011 March, 21, 2011 256 May, 13, 2011

Chile

Pepe Auth

PPD

Chile

René Jofré

PPD

Chile Chile Chile Chile

Ricardo Lagos Weber Ricardo Lagos Ricardo Núñez Ricardo Solari

PPD PPD-PS PS PS

Chile

Sergio Bitar

PPD

Chile Chile Chile Costa Rica Costa Rica Costa Rica

Soledad Alvear Víctor Maldonado Ximena Rincón Alicia Fournier Ana Helena Chacón Antonio Álvarez Desanti

Costa Rica

Arnoldo Barahona

Costa Rica Costa Rica Costa Rica Costa Rica Costa Rica

Bernal Jimenez Monge Carlos Araya Guillén Carlos Ricardo Benvides Claudio Monge Mario Zamora

DC DC DC PLN Ex PUSC PLN Yunta Escazuceña PLN PUSC PLN PAC PLN

Costa Rica

Francisco Antonio Pacheco

PLN

Costa Rica

Johnny Araya Monge

PLN

Costa Rica

Marco Antonio Vargas Díaz

PLN

Costa Rica

Luis Alberto Monge Alvarez

PLN

Costa Rica

María Lidia Sánchez

PLN

Costa Rica

Luis Gerardo Villanueva

PLN

Costa Rica

Rolando González

PLN

Costa Rica

Fabio Molina

PLN

Costa Rica

Roberto Gallardo

PLN

Costa Rica

Jorge Rojas Segura

PLN

Costa Rica

Ottón Solís

PAC

Costa Rica Costa Rica Costa Rica Costa Rica Costa Rica Costa Rica Costa Rica

Elizabeth Fonseca Margarita Bolaños Juan Carlos Mendoza García Gustavo Arias Navarro Epsy Campbell Román Macaya Hayes Sadie Bravo Pérez

PAC PAC PAC PAC PAC PAC PAC

Costa Rica

Sonia Marta Mora

PAC

Costa Rica Costa Rica Costa Rica Costa Rica

Jeannette Ruíz Rafael Ángel Calderón Miguel Ángel Rodríguez Luis Fishman

PAC PUSC PUSC PUSC

Costa Rica

Humberto Vargas Corrales

PUSC

Costa Rica

Gerardo Vargas

PUSC

Costa Rica

Marvin Herrera

PUSC

Representative Member of Political Commission of PPD Senator Ex-President of the Republic Former Senator Former Minster of Labor Former Minster of Education Senator National Secretary of DC Senator Representative Former Representative Former Representative

December, 3, 2010

Mayor of Escazú

May, 2, 2012

President of PLN Former Representative Minister of the Presidency Representative Minister of Security President of the National Board of the Banco Popular Mayor of San José Ex Minister of the Presidency Ex-President of the Republic Vice President of the National Board of the Banco Popular Representative President of the Agrarian Development Institute Representative Minister of Planning and Economic Policy Representative and President of PLN Youth Former Presidential candidate, PAC President of the PAC Secretary General of the PAC Representative Representative Representative ----Ex Representative Former Provost of Universidad Nacional Representative Ex-President of the Republic Ex-President of the Republic Representative Former Vice Presidential Candidate President of PUSC President of Convergencia Calderonista

May, 11, 2012 May, 7, 2012 May, 16, 2012 May, 8, 2012 May, 3, 2012

March, 31, 2011 April, 15, 2011 March, 29, 2011 March, 30, 2011 May, 9, 2011 November, 26, 2010 July, 1, 2011 March, 24, 2011 May, 9, 2011 May, 11, 2012 April, 24, 2012 May, 4, 2012

April, 27, 2012 May, 9, 2012 April, 26, 2012 May, 9, 2012 April, 30, 2012 May, 3, 2012 May, 15, 2012 May, 3, 2012 May, 10, 2012 May,14, 2012 April, 16,2012 April, 25, 2012 April, 30, 2012 May, 8, 2012 May, 9, 2012 April, 17, 2012 May, 4, 2012 April, 23, 2012 May, 10, 2012 May, 15, 2012 April, 26, 2012 April, 17, 2012 May, 7, 2012 May, 14, 2012 May, 4, 2012 May, 14, 2012

257

Costa Rica

Rodolfo Sotomayor

PUSC

Costa Rica

Guillermo Alvarado Herrera

PUSC

Costa Rica Costa Rica Costa Rica Costa Rica Costa Rica

Jorge Eduardo Sánchez Otto Guevara Guth Víctor Danilo Cubero Mireya Zamora Víctor Granados

Costa Rica

José Merino del Río

Costa Rica

José María Villalta

Costa Rica

Eva Carazo

Costa Rica

José Miguel Corrales

PUSC ML ML ML PASE Frente Amplio Frente Amplio Frente Amplio Ex PLN

Costa Rica

Vladimir de la Cruz

Ex FD

Costa Rica

Sergio Alfaro

Ex PAC

Costa Rica

Rolando Araya Monge

PLN

Costa Rica

Roberto Tovar Faja

Ex PUSC

Costa Rica Uruguay

Federico Malavassi Alberto Couriel

Ex ML FA

Uruguay

Alberto Iglesias

PC

Uruguay

Alejandro Atchugarry

PC

Uruguay Uruguay Uruguay Uruguay Uruguay

Álvaro Delgado Beatriz Argimón Carmelo Vidalín Daniel Arbulo Daniel Martínez

PN PN PN FA FA

Uruguay

Danilo Astori

FA

Uruguay Uruguay Uruguay Uruguay Uruguay

Eber Da Rosa Eduardo Bonomi Eduardo Fernández Eduardo Lorier Eduardo Vaz

PN FA FA FA FA

Uruguay

Esteban Valenti

FA

Uruguay

Fernando Amado

PC

Uruguay

Fernando Lorenzo

PC

Uruguay Uruguay Uruguay Uruguay Uruguay Uruguay Uruguay Uruguay Uruguay Uruguay Uruguay Uruguay

Fitzgerald Cantero Francisco Faig Jorge Batlle Jorge Brovetto Jorge Gandini Jorge Larrañaga José Amorín Batlle José López Juan Martín Posadas Lucía Topolansky Luis Alberto Heber Luis Alberto Lacalle

PC PN PC FA PN PN PC FA PN FA PN PN

Representative Former Member of the National Directorate of PUSC Ex Representative President of ML Representative Representative Representative

May, 8, 2012

Ex Represenative

May, 16, 2012

Representative

April,27, 2012

Youth leader

April, 20, 2012

Former Representative Former Presidential Candidate Former Representative Former Presidential Candidate Former Minister of Foreign Affairs Former Representative Senator Member of the National Executive Committee of the PC Former Minister of Economy and Finance Representative Representative Representative Councilor in Montevideo Senator Vice-President of the Republic Senator Minister of the Interior Secretary General of the PS Senator Militant Former Head of Tabaré Vázquez campaign Represenative Minister of Economy and Finances Representative Director Study Center PN Ex-President of the Republic President of the FA Representative Senator Senator Militant Former Senator Senator Senator Senator and Ex-President of

April, 23, 2012

May, 2, 2012 May, 15, 2012 April, 24, 2012 May, 8, 2012 April, 19, 2012 April, 19, 2012

April, 16, 2012 April, 18, 2012 April, 25, 2012 April, 23, 2012 April, 16, 2012 February. 9, 2011 February, 21, 2011 January, 26, 2011 February, 1, 2011 January, 31, 2011 February, 9, 2011 February, 4, 2011 January, 24, 2011 February, 11, 2011 February, 2, 2011 February, 4, 2011 February, 8, 2011 January, 26, 2011 February, 22, 2011 February, 2, 2011 January, 21, 2011 February, 10, 2011 January, 25, 2011 February, 8, 2011 January, 27, 2011 February, 25, 2011 January, 31, 2011 February, 23, 2011 January, 28, 2011 February, 15, 2011 February, 7, 2011 February, 7, 2011 February, 3, 2011 April, 5, 2011

258

Uruguay

Luis Hierro López

PC

Uruguay

Luis Lacalle Pou

PN

Uruguay

Martín Aguirrezabala

PC

Uruguay

Matías Rodríguez

FA

Uruguay

Mónica Xavier

FA

Uruguay

Pablo Álvarez

FA

Uruguay

Pablo Iturralde

PN

Uruguay

Pablo Mieres

PI

Uruguay

Pedro Bordaberry

PC

Uruguay

Rafael Michelini

FA

Uruguay

Rodrigo Goñi

PN

Uruguay Uruguay

Rosario Alaluf Tabaré Viera

FA PC

Uruguay

Víctor Rossi

FA

Uruguay

Victoria Schumacher

FA

the Republic Former Vice President of the Republic Representative Member of the National Executive Committee of the PC Member of the Executive of the PDC-FA Senator Director of Secretary of the Minister of Culture and Education Represenative President of the Partido Independiente Senator and Secretary General of the PC Senator Director at the Corporation for National Development Militant Senator Former Minister of Transportation and Public Works Militant

February, 2, 2011 January, 25, 2011 January, 28, 2011 February, 17, 2011 February, 9, 2011 February, 4, 2011 February, 2, 2011 February, 3, 2011 February, 21, 2011 February, 7, 2011 January, 27, 2011 February, 9, 2011 February, 15, 2011 January, 24, 2011 February, 14, 2011

259

References Adams, James. 2001. Party Competition and Responsible Party Government. Ann Arbor: Michigan University Press. Aldrich, John Herbert. 1995. Why parties? : the origin and transformation of political parties in America, American politics and political economy series. Chicago: The University of Chicago Press. Alfaro Redondo, Ronald. 2006. "Elecciones Nacionales 2006 en Costa Rica y la Recomposición del Sistema de Partidos Políticos." Revista de Ciencia Política no. 26 (1):125-137. Alfaro Redondo, Ronald 2010. "Elecciones nacionales 2010 en Costa Rica: Una mujer al mando sesenta años después." Revista de Ciencias Sociales no. 130:101-115. Alfaro Salas, Sergio Iván. 2003. "Costa Rica." In Partidos Políticos en América Latina. Centroamérica, México y República Dominicana, 31-105. México: FCE-IFE. Altman, David. 2001. "Percepciones ideológicas de lemas y fracciones: Un mapa del sistema de partidos uruguayo (1986-1997)." Cuadernos del Claeh no. 85:89-110. Altman, David, and Daniel Chasquetti. 2005. "Re-Election and Political Career Paths in the Uruguayan Congress: 1985-1999." The Journal of Legislative Studies no. 11 (2):235-253. Altman, David, and Juan Pablo Luna. 2011. "Uprooted but Stable: Chilean Parties and the Concept of Party System Institutionalization." Latin American Politics & Society no. 53 (2):1-28. Ameringer, Charles D. 1982. Democracy in Costa Rica, Politics in Latin America. New York: Praeger-Hoover Institution Press. Angell, Alan, and Benny Pollack. 1990. "The Chilean Elections of 1989 and the Politics of the Transition to Democracy." Bulletin of Latin American Research no. 9 (1):1-23. Appleton, Andrew M, and Daniel S Ward. 1997. "Party Response to Environmental Change A Model of Organizational Innovation." Party Politics no. 3 (3):341-362. Astori, Danilo. 2001. "Estancamiento, desequilibrios y ruptura, 1955-1972." In El Uruguay del siglo XX: La economía, edited by Benjamín Nahum. Montevideo: Instituto de Economía-Ediciones de la Banda Oriental. Axelrod, Robert M. 1984. The evolution of cooperation. New York: Basic Books. Barrán, José Pedro. 1990. Apogeo y crisis del Uruguay pastoril y caudillesco. 8 vols. Vol. 4, Historia Uruguaya. Montevideo: Ediciones de la Banda Oriental. Barrán, José Pedro, and Benjamín Nahum. 1979. Batlle, los estancieros y el Imperio Británico. 8 vols. Montevideo: Ediciones de la Banda Oriental. Bates, Robert H., Avner Greif, Margaret Levi, Barry R. Weingast, and Jean-Laurent Rosenthal. 1998. Analytic Narratives Princeton: Princeton University Press. Bell, John Patrick. 1981. Guerra civil en Costa Rica: los sucesos políticos de 1948: Editorial Universitaria Centro Americana (EDUCA). Benoit, Kenneth, and Michael Laver. 2006. Party policy in modern democracies, Routledge research in comparative politics. Milton Park, Abingdon, Oxon ; New York: Routledge. 260

Berg-Schlosser, Dirk , Giséle De Meur, Benoît Rihoux, and Charles C. Ragin. 2009. "Qualitative Comparative Analysis (QCA) as an approach." In Configurational Comparative Methods: Qualitative Comparative Analysis (QCA) and Related Techniques, edited by Charles C. Ragin Benoît Rihoux, 1-18. Thousand Oaks, CA: SAGE. Bermeo, Nancy Gina. 2003. Ordinary people in extraordinary times : the citizenry and the breakdown of democracy. Princeton, N.J.: Princeton University Press. Bielasiak, Jack. 2002. "The institutionalization of electoral and party systems in postcommunist states." Comparative Politics:189-210. Booth, John A. 1998. Costa Rica: quest for democracy, Nations of the modern world Latin America. Boulder, Colo.: Westview Press. Botero, Felipe. 2008. Ambitious career-seekers: An analysis of career decisions and duration in Latin America. Ph.D., The University of Arizona, Arizona. Boudon, Lawrence. 2000. "Party System Deinstitutionalization: The 1997–98 Colombian Elections in Historical Perspective." Journal of Interamerican Studies and World Affairs no. 42 (3):33-57. Bulgarelli, Oscar R Aguilar. 2004. Costa Rica y sus hechos políticos de 1948 Problemática de una década. San José: EUNED. Buquet, Daniel. 2003. "Elecciones y sistema electoral." In El Uruguay del siglo XX: La Política, edited by Benjamín Nahum, 137-171. Montevideo: Ediciones de la Banda Oriental. Buquet, Daniel. 2005. "Elecciones uruguayas 2004-2005: De la vieja oposición a la nueva mayoría." In Las Claves del Cambio: Ciclo electoral y nuevo gobierno 2004/2005, edited by Daniel Buquet, 11-26. Montevideo: Ediciones de la Banda Oriental Instituto de Ciencia Política. Buquet, Daniel, and Ernesto Castellano. 1995. "Representación Proporcional y Democracia en Uruguay." Revista Uruguaya de Ciencia Política no. 8:107-123. Buquet, Daniel, and Daniel Chasquetti. 2004a. "La democracia en Uruguay: una partidocracia de consenso." Política no. 42 (Otoño):221-247. Buquet, Daniel, and Daniel Chasquetti. 2004b. Presidential Candidate Selection in Uruguay (1942-1999). Paper read at Pathways to Power: Political Recruitment and Democracy in Latin America, April 3 and 4, at Graylyn International Conference Center, Wake Forest University, Winston-Salem, NC. Buquet, Daniel, Daniel Chasquetti, and Juan Andrés Moraes. 1998. Fragmentación Política y Gobierno en Uruguay: ¿Un Enfermo Imaginario? Montevideo: Facultad de Ciencias Sociales. Buquet, Daniel, and Gustavo De Armas. 2005. "La evolución electoral de la izquierda: crecimiento demográfico y moderación ideológica." In La Izquierda Uruguaya: Entra la Oposición y el Gobierno, edited by Jorge Lanzaro, 109-138. Montevideo: Editorial Fin de Siglo-Instituto de Ciencia Política. Caetano, G., J. Rilla, and R. Pérez. 1988. "La partidocracia uruguaya." Cuadernos del Claeh (44):37-62. Caetano, Gerardo. 1992. La república conservadora, 1916-1929. 2 vols. Montevideo: Editorial Fin de Siglo. 261

Caetano, Gerardo. 2005. La fundación del frente amplio y las elecciones de 1971. Edited by Gerardo Caetano, Líber Seregni. Montevideo: Taurus. Caetano, Gerardo. 2011. La república batllista. Montevideo: Ediciones de la Banda Oriental. Caetano, Gerardo, and José Rilla. 1984. "El sistema de partidos: raices y permanencias." Cuadernos del CLAEH no. 31:81-98. Caetano, Gerardo, and José Rilla. 1987. Breve historia de la dictadura. Montevideo: CLAEH-EBO. Caetano, Gerardo, and José Rilla. 2003. "Los partidos políticos uruguayos en el siglo XX." In El Uruguay del siglo XX: La política, edited by Benjamin Nahum, 15-64. Montevideo: Ediciones de la Banda Oriental. Caetano, Gerardo, José Rilla, and Roméo Pérez. 1987. "La partidocracia uruguaya. Historia y teoría de la centralidad de los partidos políticos." Cuadernos del CLAEH no. 44 (4):37-62. Calderón Fournier, Rafael Ángel. 2005. Raíces del Partido Unidad Social Cristiana. San José. Cancela, Walter, and Alicia Melgar. 1985. El desarrollo frustrado. 30 años de economía uruguaya 1955-1985, Serie Argumentos Montevideo: CLAEH-Ediciones de la Banda Oriental. Caramani, Daniele. 2004. The nationalization of politics : the formation of national electorates and party systems in Western Europe, Cambridge studies in comparative politics. Cambridge, UK ; New York: Cambridge University Press. Carey, John M. 1996. Term limits and legislative representation. Cambridge: Cambridge University Press. Carey, John M. , and John Polga. 2006. "Primary Elections and Candidate Strength in Latin America." Journal of Politics no. 68 (3):530-543. Carey, John M., and Peter Siavelis. 2003. "El "seguro" para los subcampeones electorales y la sobrevivencia de la Concertación." Estudios Públicos no. 90 (Otoño):5-27. Carey, John, and Matthew S. Shugart. 1995. "Incentives to Cultivate a Personal Vote: A Rank Ordering of Electoral Formulas." Electoral Studies no. 14 (4):417-439. Carmines, Edward G., and James A. Stimson. 1989. Issue evolution : race and the transformation of American politics. Princeton, N.J.: Princeton University Press. Collier, David, Jody Laporte, and Jason Seawright. 2008. "Typologies: Forming concepts and creating categorical variables." In The Oxford Handbook of Political Methodology, edited by J. Box-Steffensmeier, H. Brady y D. Collier 152-173. New York: Oxford University Press. Coppedge, Michael. 1994. Strong Parties and Lame Ducks: Presidential Partyarchy and Factionalism in Venezuela. Stanford: Stanford University Press. Coppedge, Michael. 1998a. "The Dynamic Diversity of Latin American Party Systems." Party Politics no. 4 (4):547-568. Coppedge, Michael. 1998b. "The Evolution of Latin American Party Systems." In Politics, Society, and Democracy: Latin America, edited by Scott Mainwaring and Arturo Valenzuela, 171-206. Boulder: Westview Press.

262

Coppedge, Michael. 2001. "Political Darwinism in Latin America’s lost decade." Political parties and democracy:173-205. Cortés Ramos, Alberto. 2001. "Cultura política y sistema de partidos en Costa Rica: ¿nuevas tendencias en el 2002?" In La democracia de Costa Rica ante el siglo XXI, edited by Jorge Rovira Mas, 233-253. San José: Editorial de la Universidad de Costa Rica. Cox, Gary , and Matthew McCubbins. 2001. "Introduction: Political Institutions and the Determinants of Public Policy." In Presidents, parliaments, and policy, edited by Stephan Haggard and Matthew D McCubbins, 1-20. Cambridge: Cambridge University Press. Cox, Gary W. 1997. Making Votes Count: Strategic Coordination in the World's Electoral Systems, Political Economy of Institutions and Decisions. Cambridge: Cambridge University Press. Crisp, Brian F. . 2000. Democratic Institutional Design: The Powers and Incentives of Venezuelan Politicians and Interest Groups. Stanford: Stanford University Press. Crisp, Brian F., Maria C. Escobar-Lemmon, Bradford S. Jones, Mark P. Jones, and Michelle M. Taylor-Robinson. 2004. "Vote-Seeking Incentives and Legislative Representation in Six Presidential Democracies." Journal of Politics no. 66 (3):823846. Cristi, Renato. 2000. El pensamiento político de Jaime Guzmán. Santiago: LOM Ediciones. Cross, William. 2008. "Democratic norms and party candidate selection." Party Politics no. 14 (5):596-619. Crotty, William J. 1971. "Party effort and its impact on the vote." The American Political Science Review no. 65 (2):439-450. Chasquetti, Daniel. 1998. "Compartiendo el gobierno: Multipartidismo y coaliciones en el Uruguay 1972-1997." Revista Uruguaya de Ciencia Política no. 10,:25-45. Chasquetti, Daniel. 2003. "El proceso constitucional en el Uruguay del siglo XX." In El Uruguay del siglo XX: La Política, edited by Benjamín Nahum, 65-93. Montevideo: Ediciones de la Banda Oriental. Chasquetti, Daniel. 2008a. Democracia, presidencialismo y partidos políticos en América Latina: Evaluando la "difícil combinación". Montevideo: CAUCE. Chasquetti, Daniel. 2008b. Retornar a las reglas: Examen de la institucionalización de partidos y parlamentos en Chile, Brasil y Uruguay. In Institucionalización de los Sistemas de Partidos en América Latina. Barcelona: CIDOB. Chasquetti, Daniel. 2010. Parlamento y carreras legislativas en Uruguay: un estudio sobre reglas, partidos y legisladores en las cámaras, Institutito de Ciencia Política, Universidad de la República., Montevideo, Uruguay. Chasquetti, Daniel, and Adolfo Garcé. 2005. "Unidos por la historia: Desempeño electoral y perspectivas de colorados y blancos como bloque político." In Las Claves del Cambio: Ciclo electoral y nuevo gobierno 2004/2005, edited by Daniel Buquet, 123-148. Montevideo: Banda Oriental - Instituto de Ciencia Política. Chhibber, Pradeep, and Ken Kollman. 2004. The Formation of National Party Systems: Federalism and Party Competition in Canada, Great Britain, India, and the United States. Princeton: NJ: Princeton University Press. 263

Dahl, Robert. 1971. Polyarchy: Participation and Opposition. New Haven: Yale University Press. Dalton, Russell J, and Martin P Wattenberg, eds. 2000. Parties without partisans: political change in advanced industrial democracies. Oxford University Press Oxford. Dalton, Russell J, and Steven Weldon. 2007. "Partisanship and party system institutionalization." Party Politics no. 13 (2):179-196. Dargent, Eduardo, and Paula Muñoz. 2011. "Democracy Against Parties? Party System Deinstitutionalization in Colombia " Journal of Politics in Latin America no. 3 (2):43-71. De Ramón, Armando. 2003. Historia de Chile: desde la invasión incaica hasta nuestros días (1500-2000). Vol. 983: Libreria Catalonia Ltda. Dietz, Henry A., and David J. Myers. 2007. "From Thaw to Deluge: Party System Collapse in Venezuela and Peru." Latin American Politics & Society no. 49 (2):5986. Dix, Robert H. 1992. "Democratization and the institutionalization of Latin American political parties." Comparative Political Studies no. 24 (4):488-511. Downs, Anthony. 1957. An Economic Theory of Democracy. New York: Harper and Row. Drake, Paul, and Iván Jaskic, eds. 1999. El modelo chileno. Democracia y desarrollo en los noventa. First ed, Colección Sin Norte. Santiago de Chile: LOM Ediciones. Drake, Paul W. 1978. Socialism and populism in Chile, 1932-52. Urbana: University of Illinois Press. Duverger, Maurice. 1954. Political Parties. London: Methuen. Faúndez, Julio. 1988. Marxism and democracy in Chile: from 1932 to the fall of Allende. New Haven: Yale University Press. Fernández, Oscar. 1996. "Los partidos políticos: su interrelación y sus rasgos centrales en la sociedad costarricense." Anuario de Estudios Centroamericanos no. 22 (2):147166. Ffrench-Davis, Ricardo. 2008. Entre el neoliberalismo y el crecimiento con equidad. Tres décadas de políticas económicas en Chile. Santiago: Dolmen Ediciones. Fontaine, Arturo. 1991. "El miedo y otros escritos: El pensamiento de Jaime Guzmán." Estudios Públicos no. 42:251-570. Freidenberg, Flavia , and Francisco Sánchez López. 2001. Partidos políticos y métodos de selección de candidatos en América Latina: Una discusión sobre reglas y prácticas. Paper read at Latin American Studies Association, 6-8 September, at Washington, D.C. Furlong, William L. 2001. "Política costarricense en transición." Anuario de Estudios Centroamericanos no. 27 (1):101-132. Garcé, Adolfo. 2006. Donde Hubo Fuego. El Proceso de Adaptación del MLN-Tupamaros a la Legalidad y a la Competencia Electoral (1985-2004). Montevideo: Fin de Siglo. Garcé, Adolfo, and Jaime Yaffé. 2005. La Era Progresista Montevideo: Editorial Fin de Siglo. Garretón, Manuel Antonio. 1983. El proceso político chileno. Santiago, Chile: Facultad Latinoamericana de Ciencias Sociales (FLASCO).

264

Garretón, Manuel Antonio. 1988. El Plebiscito y la Democracia. Santiago de Chile: FLACSO. Garretón, Manuel Antonio. 1989. The Chilean Political Process. Winchester MA: UNWIN HYMAN. Garretón, Manuel Antonio. 1995. Hacia una nueva era política: estudio sobre las democratizaciones: Fondo de Cultura Económica. Garretón, Manuel Antonio. 2000a. "Atavism and democratic ambiguity in the Chilean right." In Conservative parties, the right, and democracy in Latin America, edited by Kevin Middlebrook, 53-79. Baltimore: The Johns Hopkins University Press. Garretón, Manuel Antonio. 2000b. La sociedad en que vivi(re)mos. Introducción sociológica al cambio de siglo. Santiago de Chile: LOM Ediciones. George, Alexander L., and Andrew Bennett. 2005. Case studies and theory development in the social sciences, BCSIA studies in international security. Cambridge, Mass.: MIT Press. Gerring, John. 2012. Social science methodology : a unified framework. 2nd ed, Strategies for social inquiry. Cambridge ; New York: Cambridge University Press. Gil, Federico Guillermo. 1966. The political system of Chile. Boston,: Houghton Mifflin. Goertz, Gary. 2006. Social science concepts : a user's guide. Princeton: Princeton University Press. Goertz, Gary, and James Mahoney. 2012. A tale of two cultures : qualitative and quantitative research in the social sciences. Princeton, N.J.: Princeton University Press. González, Luis E. 1995. "Continuity and Change in the Uruguayan Party System." In Building Democratic Institutions: Party Systems in Latin America, edited by Scott Mainwaring and Timothy Scully, 138-163. Stanford: Stanford University Press. González, Luis Eduardo. 1991. Political Structures and Democracy in Uruguay. Notre Dame: University of Notre Dame Press. Grayson, George. 1969. "Chile's Christian Democratic Party: Power, Factions, and Ideology." The Review of Politics no. 31 (2):147-171. Grayson, George W. 1968. El Partido Demócrata Cristiano chileno. 2. ed, Biblioteca Francisco de Aguirre, 8. Colección: Ciencias políticas y sociales. Buenos Aires,: Editorial F. de Aguirre. Greif, Avner, and David D. Laitin. 2004. "A Theory of Endogenous Institutional Change." American Political Science Review no. 98 (4):633-652. Hagopian, Frances. 2005. "Conclusions: Government Performance, Political Representation, and Public Perceptions of Contemporary Democracy in Latin America." In The Third Wave of Democratization in Latin America: Advances and Setbacks, edited by Frances Hagopian and Scott Mainwaring, 319-361. Cambridge: Cambridge University Press. Hagopian, Frances, and Scott Mainwaring. 2005. "Introduction: The Third Wave of Democratization in Latin America." In The Third Wave of Democratization in Latin America: Advances and Setbacks, edited by Frances Hagopian and Scott Mainwaring, 1-13. Cambridge: Cambridge University Press. Hall, Peter A., and Rosemary C. R. Taylor. 1996. "Political Science and the Three New Institutionalisms." Political Studies no. 44 (5):936-957. 265

a an,

euven, and ideon ahat. 2010. Democracy within parties : candidate selection methods and their political consequences, Comparative politics. Oxford ; New York: Oxford University Press. Hernández Naranjo, Gerardo. 2001. "Tendencias electorales y sistema de partidos en Costa Rica." In La democracia de Costa Rica ante el siglo XXI, edited by Jorge Rovira Mas. San José: Editorial de la Universidad de Costa Rica. Hernández Naranjo, Gerardo. 2009. El Sistema de Partidos de Costa Rica 1953-2006. In Atlas Electoral de Costa Rica 1953-2006, edited by Universidad de Costa Rica Instituto de Investigaciones Sociales. San José. Hernández Naranjo, Gerardo 2007. Dinámicas del sistema de partidos políticos y del cambio institucional en el régimen electoral de Costa Rica 1952-2002, Centro de Estudios Sociológicos, El Colegio de México, México, D.F. Hicken, Allen. 2009. Building Party Systems in Developing Democracies. New York: Cambridge University Press. Hidalgo, F. Daniel. 2012. Digital Democratization: Suffrage Expansion and the Decline of Political Machines in Brazil. MIT. Hirschman, Albert O. 1970. Exit, voice, and loyalty; responses to decline in firms, organizations, and states. Cambridge, Mass.: Harvard University Press. Hu, Weiwei, and William B Heller. 2010. Investing in Agreement: Party Organization, Leadership Change and Policy Positions. Paper read at APSA 2010 Annual Meeting Paper. Huneeus, Carlos. 2000a. El Régimen de Pinochet. Santiago: Editorial Sudamericana. Huneeus, Carlos. 2000b. "Technocrats and Politicians in an Authoritarian Regime. The ‘ODEPLAN Boys´ and the ´ remialists´in Pinochet´s Chile." Journal of Latin American Studies no. 32 (2):461-501. Huneeus, Carlos. 2001. La Derecha en el Chile Después de Pinochet: El Caso de la Unión Demócrata Independiente. In The Kellogg Institute Working Papers. South Bend, In. Huneeus, Carlos. 2003. "A Highly Institutionalized Political Party: Christian Democracy in Chile." In Christian Democracy in Latin America: electoral competition and regime conflicts, edited by Scott Mainwaring and Timothy Scully, 121-161. Stanford: Stanford University Press. Hunter, Wendy. 2007. "The Normalization of an Anomaly." World Politics no. 59:440-75. Hunter, Wendy. 2010. The transformation of the Workers' Party in Brazil, 1989-2009. New York: Cambridge University Press. Huntington, Samuel. 1968. Political Order in Changing Societies. New Haven: Yale University Press. Inglehart, Ronald, and Christian Welzel. 2005. Modernization, Cultural Change, and Democracy: The Human Development Sequence. New York: Cambridge University Press. Jacob, Raúl. 1983. El Uruguay de Terra, 1931-1938: una crónica del terrismo. Montevideo: Ediciones de la Banda Oriental. Janda, Kenneth, and Tyler Colman. 1998. "Effects of party organization on performance during the `golden age' of parties." Political Studies no. 46 (3):611. 266

Joignant, Alfredo , and Patricio Navia. 2003. "De la Política de Individuos a los Hombres del Partido. Socialización, Competencia Política y Penetración Electoral de la UDI (1989-2001)." Estudios Públicos no. 89 (Verano):129 - 171. Jones, Bryan D. 1999. "Bounded Rationality." Annual Review of Political Science no. 2 (1):297. Jones, Mark. 2005. The Role of Parties and Party Systems in the Policymaking Process. In State Reform, Public Policies, and Policymaking Processes. IADB-Washington D.C. Kalyvas, Stathis N. 1996. The rise of Christian Democracy in Europe, The Wilder House series in politics, history, and culture. Ithaca, N.Y.: Cornell University Press. Kato, Junko. 1998. "When the party breaks up: Exit and voice among Japanese legislators." American Political Science Review no. 92 (4):857. Katz, Richard S, and Peter Mair, eds. 1994. How parties organize: change and adaptation in party organizations in Western democracies: Sage. Katz, Richard S., and Peter Mair. 1995. "Changing Models of Party Organization and Party Democracy: The Emergence of the Cartel Party." Party Politics no. 1 (1):528. Katznelson, Ira, and Barry Weingast, eds. 2005. Intersections between historical and rational choice institutionalism. Edited by Ira Katznelson and Barry Weingast, Preferences and situations : points of intersection between historical and rational choice institutionalism. New York: Russell Sage Foundation. Kirchheimer, Otto. 1966. "The Transformation of the Western European Party System." In Political Parties and Political Development, edited by Joseph LaPalombara and Myron Weiner, 177-200. Princeton: Princeton University Press. Kitschelt, Herbert. 1989a. "The Internal Politics of Parties: The Law of Curvilinear Disparity Revisited." Political Studies no. 37 (3):400-421. Kitschelt, Herbert. 1989b. The Logics of Party Formation: Ecological Politics in Belgium and West Germany. Ithaca: Cornell University Press. Kitschelt, Herbert. 1994. The Transformation of European Social Democracy. Cambridge: Cambridge University Press. Kitschelt, Herbert. 2000. "Linkages between Citizens and Politicians in Democratic Polities." Comparative Political Studies no. 33 (6-7):845-879. Kitschelt, Herbert, Kirk Hawkins, Juan Pablo Luna, Guillermo Rosas, and Elizabeth Zechmeister. 2010. Latin American Party Systems, Comparative Politics. New York: Cambridge University Press. Kitschelt, Herbert, Zdenka Mansfeldova, Radoslaw Markowski, and Gábor Tóka. 1999. Post-Communist Party Systems. Competition, Representation, and Inter-Party Cooperation. Cambridge: Cambridge University Press. Kitschelt, Herbert, and Steven Wilkinson, eds. 2007. Patrons, Clients and Policies. Patterns of Democratic Accountability and Political Competition. New York: Cambridge University Press. Kornblith, Miriam , and Daniel Levine. 1995. "Venezueal: The Life and Times of the Party System." In Building Democratic Institutions: Party Systems in Latin America, edited by Scott Mainwaring and Timothy Scully, 37-71. Stanford: Stanford University Press. 267

Kuenzi, Michelle, and Gina Lambright. 2001. "Party system institutionalization in 30 African countries." Party Politics no. 7 (4):437-468. Labrousse, Alain, and Laura Graciela Klang. 2009. Una historia de los tupamaros: De Sendic a Mujica. Montevideo: Fin de Siglo Editorial. Lanzaro, Jorge. 1998. "Uruguay: Las Alternativas de un Presidecialismo Pluralista." Revista Mexicana de Sociología no. 60 (2):187-215. Lanzaro, Jorge. 2000. "El Presidencialismo Pluralista en la "Segunda" Transición (19851996)." In La "Segunda" Transición en el Uruguay: Gobierno y Partidos en un Tiempo de Reforma, edited by Jorge Lanzaro, 19-196. Montevideo: Instituto de Ciencia Política - Fundación de Cultura Universitaria. Lanzaro, Jorge. 2004. "La izquierda se acerca a los uruguayos y los uruguayos se acercan a la izquierda. Claves de desarrollo del Frente Amplio." In La izquierda uruguaya entre la oposición y el gobierno, edited by Jorge Lanzaro, 13-107. Montevideo: Fin de Siglo-FESUR. Laver, Michael. 2005. "Policy and the Dynamics of Political Competition." American Political Science Review no. 99 (2 ):263-281. Laver, Michael, and Ernest Sergenti. 2012. Party Competition: An Agent-Based Model. Princeton: Princeton University Press. LeDuc, Lawrence. 2001. "Democratizing party leadership selection." Party Politics no. 7 (3):323. Lehoucq, Fabrice. 2005. "Costa Rica: Paradise in Doubt." Journal of Democracy no. 16 (3):140-154. Lehoucq, Fabrice. 2012. The politics of modern central america: civil war, democratization, and underdevelopment. Cambridge: Cambridge University Press. Lehoucq, Fabrice Edouard. 1991. "Class Conflict, Political Crisis and the Breakdown of Democratic Practices in Costa Rica: Reassessing the Origins of the 1948 Civil War." Journal of Latin American Studies no. 23 (1):37. Lehouc , abrice Edouard, and van Molina Jimenez. 2002. Stuffing the ballot box : fraud, electoral reform, and democratization in Costa Rica, Cambridge studies in comparative politics. New York: Cambridge University Press. Levitsky, Steven. 2001. "An Organised Disorganisation: Informal Organisation and the Persistence of Local Party Structures in Argentine Peronism." Journal of Latin American Studies no. 33 (1):29-66. Levitsky, Steven. 2003. Transforming labor-based parties in Latin America : Argentine Peronism in comparative perspective. Cambridge: Cambridge University Press. Lindahl, Göran. 1962. Uruguay's New Path. Stockholm: Library and Institute of IberoAmerican Studies. Lindberg, Staffan I. 2007. "Institutionalization of party systems? Stability and fluidity among legislative parties in Africa's democracies." Government and Opposition no. 42 (2):215-241. Linz, Juan. 1994. "Democracy, Presidential or Parliamentary: Does It Make a Difference?" In The Failure of Presidential Democracy: The Case of Latin America., edited by Juan Linz and Arturo Valenzuela, 3-87. Baltimore: John Hopkins University Press. 268

Lipset, Seymour M. , and Stein Rokkan. 1967. "Cleveage Structures, Party Systems, and Voter Aligments: An Introduction." In Party Systems and Voter Aligments: CrossNational Perspectives, edited by Seymour Martin Lipset and Stein Rokkan, 1-64. New York: Free Press. Londregan, John B. 2000. Legislative Institutions and Ideology in Chile. Cambridge: Cambridge University Press. Loveman, Brian. 1976. "The Transformation of the Chilean Countryside." In Chile: Politics and Society, edited by Arturo Valenzuela and J. Samuel Valenzuela, 23896. New Brunswick, N.J.: Transaction Books. Loveman, Brian. 1995. "The Transition to Civilian Government in Chile, 1990–1994." In The struggle for democracy in Chile, edited by Paul W. Drake and Ivan Jaksic, 305337. Lincoln: University of Nebraska Press. Luna, Juan Pablo. 2004. La Política desde el LLano: Conversaciones con Militantes Barriales. Montevideo: Ediciones de la Banda Oriental-Instituto de Ciencia Política Luna, Juan Pablo. 2006. Party-Voter Linkages in Two Institutionalized Party Systems: Chile and Uruguay in Comparative Perspective, Department of Political Science, University of North Carolina at Chapel Hill, Chapel Hill. Luna, Juan Pablo. 2010. "Segmented Party Voter Linkages in Latin America: The Case of the UDI." Journal of Latin American Studies no. 42 (2):325-356. Luna, Juan Pablo, Felipe Monestier, and Fernando Rosenblatt. 2013. "Religious Parties in Chile: The Christian Democratic Party and the Independent Democratic Union." Democratization no. 20 (5):917-938. Luna, Juan Pablo, and Fernando Rosenblatt. 2012. "¿Notas para una autopsia? Los partidos políticos en el Chile actual." In Democracia con Partidos. Informe para la reforma de los partidos políticos en Chile, edited by Francisco J. Díaz; Lucas Sierra, 115-268. Santiago, Chile: CEP-CIEPLAN. Lupu, Noam. 2011. Party brands in crisis: Partisanship, brand dilution, and the breakdown of political parties in Latin America. Ph.D., Princeton University, New Jersey. Mahoney, James, and Kathleen Thelen. 2010. "A theory of gradual institutional change." In Explaining institutional change: ambiguity, agency, and power, edited by James Mahoney and Kathleen Thelen, 1-37. Cambridge: Cambridge University Press. Mainwaring, Scott. 1999. Rethinking Party Systems in the Third Wave of Democratization: The Case of Brazil. Stanford: Stanford University Press. Mainwaring, Scott, Ana María Bejarano, and Eduardo Pizarro. 2006. The Crisis of Democratic Representation in the Andes. Stanford: Stanford University Press. Mainwaring, Scott, and Timothy Scully, eds. 1995. Building Democratic Institutions: Party Systems in Latin America. Stanford: Stanford University Press. Mainwaring, Scott, and Timothy Scully, eds. 2010. Democratic governance in Latin America. Stanford, Calif.: Stanford University Press. Mainwaring, Scott, and Matthew Shugart, eds. 1997. Presidentialism and Democracy in Latin America. Cambridge: Cambridge University Press. Mainwaring, Scott, and Mariano Torcal. 2006. "Party System Institutionalization and Party System Theory After the Third Wave of Democratization." In Handbook of

269

Party Politics, edited by Richard S. Katz and William Crotty, 204-227. London: Sage. Mainwaring, Scott, and Edurne Zoco. 2007. "Political Sequences and the Stabilization of Interparty Competition." Party Politics no. 13 (2):155-178. Mair, Peter. 1997. Party system change : approaches and interpretations. Oxford New York: Clarendon Press ; Oxford University Press. Malbrough, Monica. 2008. "Political History and Party System Institutionalization in Sub-Saharan Africa." Conference Papers -- American Political Science Association:127. Martz, John D. 1966. Accion Democratica: Evolution of a Modern Political Party in Venezuela. Princeton, N.J.,. Martz, John D. 1992. "Party elites and leadership in Colombia and Venezuela." Journal of Latin American Studies no. 24 (1):87. May, John D. 1973. "Opinion Structure of Political Parties: The Special Law of Curvilinear Disparity." Political Studies no. XXI:135-151. Mieres, Pablo. 1992. "Acerca de los cambios del sistema de partido uruguayo." Cuadernos del CLAEH no. 62:65-75. Miller, Warren, and Donald Stokes. 1963. "Constituency Influence in Congress." American Political Science Review no. 57 (March):45-57. Molina Jiménez, Iván. 2007. Anticomunismo Reformista. Competencia electoral y cuestión social en Costa Rica (1931-1948). San José: Editorial Costa Rica. Monestier, Felipe. 1999. "Partidos por dentro: la fraccionalización de los partidos políticos en el Uruguay (1954-1994)." In Los partidos políticos uruguayos en tiempos de cambio, edited by Luis Eduardo González, 19-86. Montevideo: Fundación Cultura Universitaria. Mora Alfaro, Jorge. 2001. "Costa Rica: deterioro de la calidad de la democracia y búsqueda de un nuevo camino democrático." In La democracia de Costa Rica ante el siglo XXI, edited by Jorge Rovira Mas, 121-141. San José: Editorial de la Universidad de Costa Rica. Moraes, Juan A. 2008. "Why Factions? Candidate Selection and Legislative Politics in Uruguay." In Pathways to Power: Political Recruitment and Candidate Selection in Latin America, edited by Peter M. Siavelis and Scott Morgenstern, 164-185. University Park: Pennsylvania State University Press. Morales, Mauricio, and Rodrigo Burgueño. 2001. "La UDI como expresión de la Nueva Derecha en Chile." Estudios Sociales no. 107:215-248. Moreira, Constanza. 2004. Final de juego: Del bipartidismo tradicional al triunfo de la izquierda en Uruguay. Montevideo: Ediciones Trilce. Morgan, Jana. 2007. "Partisanship During Collapse of the Venezuelan Party System." Latin American Research Review no. 42 (1):78-98. Morgan, Jana. 2009. Party System Stability Versus Collapse. In American Political Science Association annual meeting. Toronto, Ontario, Canada. Morgan, Jana. 2011. Bankrupt Representation and Party System Collapse: Penn State Press.

270

Moulian, Tomas. 1993. La Forja de Ilusiones: El Sistema de Partidos 1932-1973. Santiago de Chile: Universidad de Artes y Ciencias Sociales (ARCIS) y Facultad Latinoamericana de Ciencias Sociales (FLACSO). Munck, Gerardo L. 2009. Measuring Democracy. A Bridge between Scholarship and Politics. Baltimore: The Johns Hopkins University Press. Mustillo, Thomas J. 2007. Entrants in the political arena: New party trajectories during the third wave of democracy in Latin America. Ph.D., The University of North Carolina at Chapel Hill, Ann Arbor. Nahum, Benjamín. 1986. La Epoca Batllista: 1905-1929. 8 vols. Vol. 6, Historia Uruguaya. Montevideo: Ediciones de la Banda Oriental. Navia, Patricio. 2000. Incumbency in the Chilean Parliament: Continuities and Change. Paper read at Latin American Studies Association, March 16-18, at Miami, Florida. Navia, Patricio. 2003. Three’s Company: Old and New Cleavages in Chile’s Party System. Paper read at Latin American Studies Association, March 27-29, at Dallas, Texas. Navia, Patricio. 2004. "Participación electoral en Chile 1988-2001." Revista de Ciencia Política no. XXIV (1):81-103. Navia, Patricio. 2008. "Legislative Candidate Selection in Chile." In Pathways to Power in Latin America, edited by Peter Siavelis and Scott Morgenstern, 92-118. University Park: Penn State University Press. Nieva, José Díaz. 2000. Chile: de la Falange Nacional a la democracia cristiana. Spain: Universidad Nacional de Educación a Distancia, UNED. Norris, Pippa, and Joni Lovenduski. 1995. Political Recruitment: Gender, Race and Class in the British Parliament. Cambridge: Cambridge University Press. Novoa, Jovino. 2013. Con la Fuerza de la Libertad. Santiago, Chile: Planeta. Oduber Quirós, Daniel. 1985. Raíces del Partido Liberación Nacional. Notas para una evaluación Histórica. San José: CEDAL. Ozzano, Luca. 2013. "The many faces of the political god: a typology of religiously oriented parties." Democratization no. 20 (5):807-830. Paige, Jeffery M. 1998. Coffee and power : revolution and the rise of democracy in Central America. Cambridge, Mass.: Harvard University Press. Panebianco, Angelo. 1988. Political parties : organization and power, Cambridge studies in modern political economies. Cambridge, England ; New York: Cambridge University Press. Panizza, Francisco. 1990. Uruguay, Batllismo y Después: Pacheco, Militares y Tupamaros en la Crisis del Uruguay Batllista. Montevideo: Ediciones de la Banda Oriental. Paolillo, Claudio. 2004. Con los días contados. Montevideo: Editorial Fin de Siglo. Pedahzur, Ami, and Avraham Brichta. 2002. "The institutionalization of extreme rightwing charismatic parties: a paradox?" Party Politics no. 8 (1):31-49. Pennings, Paul, and Reuven Y. Hazan. 2001. "Democratizing Candidate Selection: Causes and Consequences." Party Politics no. 7 (3):267–275. Pérez Antón, Romeo. 1984. "Los partidos en el Uruguay moderno." Cuadernos del CLAEH no. 31,:63-79. 271

Pérez Antón, Romeo. 1988. "Cuatro Antagonismos sucesivos: La concreta instauración de la democracia uruguaya." Revista de Ciencia Política (2). Pérez Brignoli, Héctor , and Yolanda Baires Martínez. 2001. "Costa Rica en el año 2000: Una crisis política en ciernes." In La democracia de Costa Rica ante el siglo XXI, edited by Jorge Rovira Mas, 143-154. San José: Editorial de la Universidad de Costa Rica. Pierson, Paul. 2004. Politics in time: history, institutions, and social analysis. Princeton: Princeton University Press. Piñeiro, Rafael. 2007. "El sueño de la lista propia: los dilemas de coordinación electoral post-reforma de 1997." Revista Uruguaya de Ciencia Política no. 16:51-71. Piñeiro, Rafael, and Jaime Yaffé. 2004. "El Frente Amplio por dentro. Las fracciones frenteamplistas 1971-1999." In La izquierda uruguaya entre la oposición y el gobierno, edited by Jorge Lanzaro, 297-319. Montevideo: Fin de Siglo/FESUR. Pivel Devoto, Juan. 1942. Historia de los Partidos Políticos en el Uruguay. Montevideo: Tipografía Atlantida. Pollack, Marcelo. 1999. The New Right in Chile 1973-1997. London: MacMillan Press Ltd. Porto, Luis. 2002. Uruguay 2002: para entender la crisis bancaria. Montevideo: Ediciones de la Banda Oriental. Power, Timothy J., and Cesar Zucco Jr. 2009. "Estimating ideology of Brazilian legislative parties, 1990-2005." Latin American Research Review no. 44 (1):218-246. Przeworski, Adam, Michael E. Alvarez, José Antonio Cheibub, and Fernando Limongi. 2000. Democracy and Development: Political Institutions and Well-Being in the World, 1950-1990. Cambridge: Cambridge University Press. Przeworski, Adam, and Henry Teune. 1970. The logic of comparative social inquiry, Comparative studies in behavioral science. New York,: Wiley-Interscience. Queirolo, Rosario. 1999. "La "Tradicionalización" del Frente Amplio: la conflictividad del proceso de cambio." In Los partidos políticos uruguayos en tiempos de cambio: , edited by Luis E. González, 86-127. Montevideo: Fundación de Cultura Universitaria-UCUDAL. Ragin, Charles C. 1987. The comparative method : moving beyond qualitative and quantitative strategies. Berkeley: University of California Press. Ragin, Charles C. 2000. Fuzzy-set social science. Chicago: University of Chicago Press. Ragin, Charles C. 2008. Redesigning social inquiry : fuzzy sets and beyond. Chicago: University of Chicago Press. Rahat, Gideon, and Mario Sznajder. 1998. "Electoral engineering in Chile: the electoral system and limited democracy." Electoral Studies no. 17 (4):429-442. Rama, Germán. 1971. El Club Político. Montevideo: Arca. Rama, Germán W. 1987. La Democracia en Uruguay: Una Perspectiva de Interpretación. Buenos Aires: Grupo Editor Latinoamericano. Randall, Vicky, and Lars Svåsand. 2002. "Party Institutionalization in New Democracies." Party Politics no. 8 (1):5-29. Raventós, Ciska. 2001. "Democracia y proceso de aprobación de las políticas de ajuste en Costa Rica (1980-1995)." In La democracia de Costa Rica ante el siglo XXI, edited

272

by Jorge Rovira Mas, 367-380. San José: Editorial de la Universidad de Costa Rica. Raventós, Ciska. 2005. "Más allá del escándalo: bases políticas e institucionales de la corrupción en Costa Rica." Revista Centroamericana de Ciencias Sociales no. 2 (3):45-69. Real de Azúa, Carlos. 1964. El Impulso y su Freno: Tres Décadas de Batllismo y las Raíces de la Crisis Uruguaya. Montevideo: Ediciones de la Banda Oriental. Real de Azúa, Carlos. 1971. Uruguay, ¿una sociedad amortiguadora? Montevideo: Banda Oriental. Rehren, Alfredo. 1993. "La Presidencia en el Gobierno de la Concertación." Estudios Sociales no. 75:15-38. Remmer, Karen. 1991. "The Political Impact of Economic Crisis in Latin America in the 1980's." American Political Science Review no. 81 (3):777-800. Rial, Juan. 1984. Partidos políticos, democracia y autoritarismo. Montevideo, Uruguay: Centro de Informaciones y Estudios del Uruguay- Ediciones de la Banda Oriental. Rihoux, Benoit, and Charles C. Ragin. 2009. Configurational comparative methods : qualitative comparative analysis (QCA) and related techniques, Applied social research methods series. Thousand Oaks: Sage. Roberts, Kenneth. 1998. Deepening Democracy? The Modern Left and Social Movements in Chile and Peru. Stanford: Stanford University Press. Roberts, Kenneth. forthcoming. Political Representation in Latin America's Neoliberal "Critical Juncture". Roberts, Kenneth, and Erik Wibbels. 1999. "Party Systems and Electoral Volatility in Latin America: A Test of Economic, Institutional, and Structural Explanations." American Political Science Review no. 93 (3):575-590. Rojas Bolaños, Manuel. 1979. Lucha social y guerra civil en Costa Rica 1940-1948. San José: Editorial Porvenir. Rojas Bolaños, Manuel, and Carlos Sojo. 1995. "El malestar con la política." San José, Costa Rica: Flacso-Programa Costa Rica. Rose, Richard, and Thomas T Mackie. 1988. "Do parties persist or fail? The big trade-off facing organizations." In When parties fail: Emerging alternative organizations, edited by Kay Lawson and Peter H. Merkl, 533-58. Princeton, NJ: Princeton University Press. Rosenblatt, Fernando. 2006. "El dispar desenlace de la crisis económica en Argentina y Uruguay (2001-2002): una explicación desde la teoría de las prospectivas." Revista de Ciencia Política no. 26 (2):97-119. Rovira Mas, Jorge. 1982. Estado y política económica en Costa Rica: 1948-1970. 2 ed. San José: Editorial Universidad de Costa Rica. Rovira Mas, Jorge. 1987. Costa Rica en los años 80. San José: Porvenir-ICADIS-CRIES. Rovira Mas, Jorge. 1990. "Costa Rica: Elecciones, partidos políticos y régimen democrático." Polémica no. 11:44-60.

273

Rovira Mas, Jorge. 2001. "¿Se debilita el bipartidismo?" In La Democracia de Costa Rica ante el siglo XXI, edited by Jorge Rovira Mas, 195-232. San José: Editorial de la Universidad de Costa Rica. Salazar, Jorge Mario. 1990. "El modelo político-electoral de la democracia costarricense 1920-1980." Revista de Ciencias Sociales. Universidad de Costa Rica (48):17-30. Samuels, David. 1999. "Incentives to Cultivate a Party Vote in Candidate-centric Electoral Systems: Evidence from Brazil." Comparative Political Studies no. 32 (4):487-518. Samuels, David. 2003. Ambition, Federalism, and Legislative Politics in Brazil. New York: Cambridge University Press. Samuels, David. 2004. "From Socialism to Social Democracy: Party Organization and the Transformation of the Workers’ Party in Bra il." Comparative Political Studies no. 37 (9):999-1024. Samuels, David. 2006. "Sources of Mass Partisanship in Brazil." Latin American Politics & Society no. 48 (2):1-27. Samuels, David J, and Matthew S Shugart. 2010. Presidents, parties, and prime ministers: how the separation of powers affects party organization and behavior: Cambridge University Press. Sánchez, Fernando. 2001. "Sistema electoral y partidos políticos: incentivos hacia el bipartidismo en Costa Rica." Anuario de Estudios Centroamericanos no. 27 (1):133168. Sánchez, Fernando. 2003. "Cambio en la dinámica electoral en Costa Rica: un caso de desalineamiento." América latina Hoy (35):115-146. Sánchez, Omar. 2008. "Transformation and Decay: the de-institutionalisation of party systems in South America." Third World Quarterly no. 29 (2):315-337. Sánchez, Omar. 2009. "Party Non-Systems A Conceptual Innovation." Party Politics no. 15 (4):487-520. Sartori, Giovanni. 1970. "Concept misinformation in comparative politics." American Political Science Review no. 64 (4):1033-1053. Sartori, Giovanni. 1976. Parties and Party Systems: a framework for analysis. New York: Cambridge University Press. Scully, Timothy. 1995. "Reconstituting Party Politics in Chile." In Building Democratic Institutions: Party Systems in Latin America, edited by Scott Mainwaring and Timothy Scully, 100-137. Stanford: Stanford University Press. Scully, Timothy R. 1992. Rethinking the Center: Party Politics in Nineteenth and TwentiethCentury Chile. Standford: Standford University Press. Schattschneider, Elmer. 1942. Party Government. NY:Holt: Rinehart and Winston. Schedler, Andreas. 1995. Under-and overinstitutionalization: some ideal typical propositions concerning new and old party systems: Helen Kellogg Institute for International Studies. Schlesinger, Joseph A. 1994. Political parties and the winning of office. Ann Arbor: University of Michigan Press. Seawright, Jason. 2012. Party-system collapse: the roots of crisis in Peru and Venezuela. Stanford, California: Stanford University Press. 274

Seligson, Mitchell A, and Edward N Muller. 1987. "Democratic Stability and Economic Crisis: Costa Rica, 1978-1983." International Studies Quarterly:301-326. Siavelis, Peter. 1999. "Continuidad y transformación del sistema de partidos en una transición "Modelo"." In El modelo chileno. Democracia y desarrollo en los noventa., edited by Paul Drake and Iván Jaskic, 223-257. Santiago de Chile: LOM Ediciones. Siavelis, Peter. 2000a. The President and Congress in Post-Authoritarian Chile: Institutional Constraints to Democratic Consolidation. University Park: The Pennsylvaria University Press. Siavelis, Peter. 2002. "Exaggerated Presidentialism and Moderate Presidents: ExecutiveLegislative Relations in Chile." In Legislative Politics in Latin America, edited by Scott Morgenstern and Benito Nacif, 79-113. Cambridge: Cambridge University Press. Siavelis, Peter 2000b. "Continuidad y cambio en el sistema partidista chileno: sobre los efectos de la transformación de una reforma electoral." Revista de Ciencia Política no. XX (2):82-101. Siavelis, Peter, and Scott Morgenstern. 2008. "Political recruitment and candidate selection in Latin America: A framework for analysis." In Pathways to Power: Political Recruitment and Candidate Selection in Latin America, edited by Peter Siavelis and Scott Morgenstern, 3-37. University Park, PA: The Pennsylvania State University Press. Silva, Eduardo. 1996. The State and Capital in Chile: Business Elites, Technocrats, and Market Economics. Boulder: Westview Press. Solari, Aldo E. 1991. Uruguay, partidos políticos y sistema electoral. 2. ed. Montevideo, Uruguay: Fundacion de Cultura Universitaria. Soto, Angel. 2001. La irrupción de la UDI en las poblaciones, 1983-1987. Paper read at Latin American Studies Association, at Washington D.C. Stockton, Hans. 2001. "Political Parties, Party Systems, and Democracy in East Asia Lessons from Latin America." Comparative Political Studies no. 34 (1):94-119. Stokes, Susan. 1999. "What do Policy Switches Tell Us about Democracy?" In Democracy, Accountability and Representation., edited by Adam Przeworski, Susan Stokes and Bernard Manin, 98-130. Cambridge: Cambridge University Press. Strom, Kaare. 1990. "A Behavioral Theory of Competitive Political Parties." American Journal of Political Science no. 34 (2):565. Tavits, Margit. 2005. "The Development of Stable Party Support: Electoral Dynamics in Post-Communist Europe." American Journal of Political Science no. 49 (2):283-298. Tavits, Margit. 2008. "Party Systems in the Making: The Emergence and Success of New Parties in New Democracies." British Journal of Political Science no. 38 (1):113-133. Taylor-Robinson, Michelle M. 2010. Do the poor count? : democratic institutions and accountability in a context of poverty. University Park, Pa.: Pennsylvania State University Press. Taylor, Michelle M. 1992. "Formal versus informal incentive structures and legislator behavior: Evidence from Costa Rica." Journal of Politics no. 54 (4):1055.

275

Thames, Frank C., and Joseph W. Robbins. 2007. "Party System Institutionalization and the Level of Democracy." Conference Papers -- American Political Science Association:1-26. Thelen, Kathleen. 2003. "How institutions evolve. Insights from comparativae historical analysis." In Comparative Historical Analysis in the Social Sciences, edited by James Mahoney and Dietrich Rueschemeyer, 208-240. New York: Cambridge University Press. Toka, Gabor. 1998. "Party appeals and voter loyalty in new democracies." Political Studies no. 46 (3):589. Torcal, Mariano. 2011. La institucionalizción del sistema de partidos en America Latina: Medición, Causas y consecuencias. Madrid: CIS-CIDOB. Torres Rivas, Edelberto. 2001. "Contrapunto entre reforma y revolución: la democracia en Costa Rica y Guatemala." In La democracia de Costa Rica ante el siglo XXI, edited by Jorge Rovira Mas, 21-40. San José: Editorial de la Universidad de Costa Rica. Tsebelis, George. 1995. "Decision Making in Political Systems: Veto Players in Presidentialism, Parliamentarism, Multicameralism and Multipartyism." British Journal of Political Science no. 25 (3):289-325. Valenzuela, Arturo. 1977. Political Brokers in Chile: Local Government in a Centralized Polity. Durham, NC.: Duke University Press. Valenzuela, Arturo. 1978. The breakdown of democratic regimes : Chile. Baltimore: Johns Hopkins University Press. Valenzuela, Arturo. 1995. "The Military in Power: The Consolidation of One-Man Rule." In The struggle for democracy in Chile, edited by Paul W. Drake and Ivan Jaksic, 21-72. Lincoln: University of Nebraska Press. Valenzuela, Arturo, and J. Samuel Valenzuela. 1976. Chile : politics and society. New Brunswick, N.J.: Transaction Books. Valenzuela, Julio Samuel. 1985. Democratización vía reforma: la expansión del sufragio en Chile. Vol. 6. Buenos Aires, Argentina: Ediciones del IDES. Valenzuela, Samuel J., and Timothy R. Scully. 1997. "Electoral Choices and the Party System in Chile: Continuities and Changes at the Recovery of Democracy." Comparative Politics no. 29 (4):511-527. Vanger, Milton. 1963. José Batlle y Ordoñez of Uruguay: The Creator of his Times, 1902-1907. Cambridge: Harvard University Press. Vargas Cullell, Jorge. 2008. "Costa Rica: Una Decisión Estratégica en Tiempos Inciertos." Revista de Ciencia Política no. 28 (1):147-169. Visconti, Giancarlo. 2011. "Comportamiento diacrónico del congreso en Chile." Revista de Ciencia Política no. 31 (1):91-115. Walker, Ignacio. 1999. El futuro de la democracia cristiana: Ediciones B Grupo Zeta. Walker, Ignacio. 2003. "The Future of Chilean Christian Democracy." In Christian Democracy in Latin America: electoral competition and regime conflicts, edited by Scott Mainwaring and Timothy Scully, 162-195. Stanford: Stanford University Press.

276

Webb, Paul, and Stephen White, eds. 2007a. Party politics in new democracies. Oxford: Oxford University Press Webb, Paul, and Stephen White. 2007b. "Political Parties in New Democracies: Trajectories of Development and Implications for Democracy " In Party politics in new democracies, edited by Paul Webb and Stephen White, 345-369. Oxford: Oxford University Press Weyland, Kurt. 1997. ""Growth with equity" in Chile's new democracy?" Latin American Research Review no. 32 (1):37-67. Weyland, Kurt. 2002. The Politics of Market Reform in Fragile Democracies: Argentina, Brazil, Peru, and Venezuela. Princeton: Princeton University Press. Weyland, Kurt. 2007. Bounded rationality and policy diffusion : social sector reform in Latin America. Princeton, N.J.: Princeton University Press. Wilson, Bruce M. 1994. "When social democrats choose neoliberal economic policies." Comparative Politics no. 26 (2):149. Wilson, Bruce M. 1998. Costa Rica: politics, economics, and democracy. Boulder: Lynne Rienner Publishers. Yaffé, Jaime. 2003. "La tradicionalización del Frente Amplio y el nacimiento de la tercera divisa." Cuadernos del CLAEH no. 86. Yaffé, Jaime. 2005. Al centro y adentro: la renovación de la izquierda y el triunfo del Frente Amplio en Uruguay. Montevideo: Linardi y Risso. Yashar, Deborah. 1997. Demanding Democracy: Reform and Reaction in Costa Rica and Guatemala, 1870s-1950s. Palo Alto: Stanford University Press. Zemelman, Hugo. 1971. Génesis histórica del proceso político chileno, edited by Enzo Faletto, Eduardo Ruiz and Hugo Zemelman, 33-118. Santiago, Chile: Quimantú. Zucco, Cesar. 2010. Stability Without Roots: Party System Institutionalization in Brazil. In Working Paper Series, edited by SSRN. Available at SSRN: http://ssrn.com/abstract=2002359 or http://dx.doi.org/10.2139/ssrn.2002359.

277

Smile Life

When life gives you a hundred reasons to cry, show life that you have a thousand reasons to smile

Get in touch

© Copyright 2015 - 2024 PDFFOX.COM - All rights reserved.