Constitutional Origin and Norm Creation in Colombia: Discursive Institutionalism and the Empowerment of the Constitutional Court 1032134577, 9781032134574

Citation preview

Constitutional Origin and Norm Creation in Colombia

This book explains the growing empowerment of the Colombian Constitutional Court in the early years of the 21st century and develops the concept of the deliberative judge. Taking the case of the Colombian Constitutional Court and drawing on neo­ institutional theory to explain the relationship between political crisis and institutional reforms, the book challenges the notion of rational choice institutionalism that agents act strategically. It indicates the limits of path dependence and argues instead that discursive institutionalism is the most appropriate method for analyzing processes of institutional learning. Combining theoretical and empirical research, it builds the argument that judicial independence promotes the case for deliberative democracy over rational choice or strategic action approaches. Finally, the book suggests that by introducing communicative and cognitive variables in our understanding of key actors and processes, we are more capable of bridging institutional origin and legacy. The work will be a valuable resource for academics, researchers, and policy-makers in Constitutional Law, Constitutional Politics, and Constitutional History. Jan Boesten is a postdoctoral researcher at the Institute for Latin American Studies (LAI) at the Freie Universität Berlin, associated researcher with the CONPEACE Programme, as well as an Associate Member of Nuffield College at the University of Oxford. His work has appeared in the Latin American Research Review (LARR) and Colombia Internacional, Precedente, Revista Jurídica, and several edited volumes.

Comparative Constitutional Change Series editors: Xenophon Contiades is Professor of Public Law, Panteion University, Athens,

Greece and Managing Director, Centre for European Constitutional Law, Athens,

Greece.

Thomas Fleiner is Emeritus Professor of Law at the University of Fribourg, Swit­ zerland.

Alkmene Fotiadou is Research Associate at the Centre for European Constitu­ tional Law, Athens, Greece.

Richard Albert is the William Stamps Farish Professor in Law and Professor of

Government at the University of Texas at Austin, USA.

Comparative Constitutional Change has developed into a distinct field of constitu­ tional law. It encompasses the study of constitutions through the way they change

and covers a wide scope of topics and methodologies. Books in this series include

work on developments in the functions of the constitution, the organization of

powers and the protection of rights, as well as research that focuses on formal

amendment rules and the relation between constituent and constituted power. The

series includes comparative approaches along with books that focus on single jur­ isdictions, and brings together research monographs and edited collections which

allow the expression of different schools of thought. While the focus is primarily on

law, where relevant the series may also include political science, historical, philoso­ phical and empirical approaches that explore constitutional change.

Also in the series: The Law and Politics of Unconstitutional Constitutional Amendments in Asia Edited by Rehan Abeyratne & Ngoc Son Bui The Right to Political Participation A Study of the Judgments of the European and Inter-American Courts of Human Rights Edited by Gabriella Citroni, Irene Spigno, and Palmina Tanzarella Constitutional Origin and Norm Creation in Colombia Discursive Institutionalism and the Empowerment of the Constitutional Court Jan Boesten For more information about this series, please visit: www.routledge.com/Comparative-Constitutional-Change/book-series/ COMPCONST

Constitutional Origin and Norm Creation in Colombia Discursive Institutionalism and the Empowerment of the Constitutional Court

Jan Boesten

First published 2022 by Routledge 4 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Routledge 605 Third Avenue, New York, NY 10158 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2022 Jan Boesten The right of Jan Boesten to be identified as author of this work has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data Names: Boesten, Jan, author. Title: Constitutional origin and norm creation in Colombia : discursive institutionalism and the empowerment of the Constitutional Court / Jan Boesten. Description: Abingdon, Oxon ; New York, NY : Routledge, 2022. | Series: Comparative constitutional change | Based on author’s thesis (doctoral ­ University of British Columbia, 2016) issued under title: Between democratic security and democratic legality : discursive institutionalism and Colombia’s Constitutional Court. | Includes bibliographical references and index. | Identifiers: LCCN 2021054178 | ISBN 9781032134574 (hardback) | ISBN 9781032134581 (paperback) | ISBN 9781003229285 (ebook) Subjects: LCSH: Colombia. Corte Constitucional. | Constitutional courts-­ Political aspects--Colombia. | Political questions and judicial power--Colombia Classification: LCC KHH2548 .B64 2022 | DDC 347.861/04--dc23/eng/ 20220202 LC record available at https://lccn.loc.gov/2021054178 ISBN: 978-1-032-13457-4 (hbk) ISBN: 978-1-032-13458-1 (pbk) ISBN: 978-1-003-22928-5 (ebk) DOI: 10.4324/9781003229285 Typeset in Galliard by Taylor & Francis Books

For Oma and Opa, who did not live to see the end of this, and Mama and Papa, who always gave a helping hand!

Contents

List of illustrations Foreword by Laurence Whitehead Acknowledgments

viii

ix

xi

1

Introduction: Ethical Legal Discourse in Delegative Democracies

2

The Novelty of the 1991 Constitution: A Critical Juncture in

Colombia’s History

36

The Choque de Trenes between Colombia’s High Courts: Path

Dependence and Legal Argumentation

88

3 4 5 6

1

Discursive Institutionalism in Colombia’s Constitutional Court: The

Deliberative Judge and Norm Creation in a Politicized Context

160

Conclusion: The Pre-eminence of Democratic Legality over

Democratic Security and Discursive Institutionalism

218

Epilogue

241

Appendices Index

251

262

Illustrations

Figures 2.1 Ideal-type trajectory of sub-cases of institution building in Colombia (1905–1910; 1953–1957; and 1991) 2.2 Ideal-type trajectory of sub-cases of institution building in Colombia (1905–1910; 1953–1957; and 1991) 2.3 The 1991 constitutional moment as critical juncture 3.1 The courts and the branches of government in Colombia’s 1991 constitutional regime 5.1 Ideal-type trajectory of sub-cases of institution building in Colombia (1905–1910; 1953–1957; and 1991)

41 78 79 96 220

Tables 1.1 2.1 3.1 3.2 4.1 4.2 4.3 4.4 4.5 4.6 4.7 A.1

Democracy measurements for Colombia, 1985–2011 Colombia’s high courts in the new constitutional regime Phases of the proceso 8000 scandal Phases of the parapolítica scandal Strategic vs. deliberative action Uribe’s majorities in Congress, 2002–2010 Votes in Congress for Legislative Act 02 of 2004 and Law 1354 of 2009 Constitutional Court decisions between 2003 and 2010 and the substitution doctrine Formal and informal rules of deliberation inside the Constitutional Court Hübner Mendes’ theory of deliberative decision-making in constitutional courts Observable implications of strategic and deliberative action Quality of discourse in Semana during Uribe presidency, 2002–2010

13 72 106 118 173 182 183 186 196 197 206 255

Foreword

Enacting and sustaining a democratic rule of law system is a noble objective. But it is also a highly demanding project even in the most favorable of settings. It requires the establishment of robust legal institutions staffed by a phalanx of skilled and committed professionals, with support from a tissue of supportive societal props and practices. These legal professionals should not only work together to a high standard of rigor and commitment, but also need to communicate effectively with the general population whose freedoms, rights, and responsibilities are in their charge. Like any other set of human arrangements, the judiciary will always be fallible, subject to conflicts and deviations and so will periodically require revi­ sion and updating. If this is a challenge in all circumstances it is all the more so in a country like today’s Colombia, where the gulf in understanding between highly trained experts and the people they serve is aggravated by extremes of social inequality and the traumas of insecurity and widespread criminal violence. Jan Boesten’s wide-ranging and innovative monograph addresses both the subtleties of rule of law discourse and practices in general and the specific com­ plexities that have arisen in Colombia in the three decades since the adoption of the pathbreaking Constitution of 1991. He brings a sophisticated European perspective to bear on a subject that has only recently attracted strong attention from comparativists, and where for the most part social science modeling has been filtered through the prism of US constitutional doctrine and experience. Boesten argues persuasively that court rulings on constitutional issues have to extend well beyond literalist applications. Indeed he uses the Colombian case to support the idea that the discursive evolution of judicial reasoning may need to include creative and adaptive development of foundational principles under the pressure of changing and initially unforeseen political realities. Certainly legal precedents matter (they are needed to stabilize and generalize the rule of law), but this study casts doubt on the linear determinism that has often characterized “path dependency” approaches to historical institutionalism. As new “legal facts” are established and the highest courts encounter unanticipated political chal­ lenges they may well “deliberate” in a creative manner, and even overturn prior judgments given the discursive force of “better” judicial reasoning. His evidence indicates that such developments cannot be modeled solely in terms of “strate­ gizing” in accordance with rational choice assumptions about the maximization

x

Foreword

of fixed preferences. Rather they may involve shifts of interpretation and the collective adoption of new understandings arising from dialogue between mutually respectful colleagues. Such shifts of perspective will be limited and controlled, but outcomes will always remain open to further review and modification. This carefully argued (and recognizably Germanic) approach to legal philosophy and judicial institutionalism acquires vivid topical relevance when applied to the fruits of Colombia’s 1991 “critical juncture” of constitutional reform, which opened the door to a more inclusively democratic rule of law regime. The new dispensation took time to emerge through a complex process of adaptation and learning. It was severely tested by a succession of major political crises (first illicit election funding; then paramilitary penetration of the Congress; then presidential attempts to intimidate the courts, culminating in the bugging of a judge’s cham­ bers; and finally turmoil over a polemical peace agreement to terminate half a century of guerrilla warfare). Boesten provides telling detail on all these dramatic episodes, drawing on an original archive of both written and oral evidence focused particularly on legal argumentation in the highest courts. His central puzzle con­ cerns the doctrinal shift explaining why President Uribe was granted permission to stand for an unprecedented second term of office, but was then refused permission for a third term. The Colombian case study is not only remarkable on its own terms, but is also highly relevant to the wider debates currently ongoing throughout the world’s constitutional democracies concerning courts as political actors and the judiciali­ zation of politics. In particular it invites reflection on what basis they may uphold the separation of powers in a climate characterized by citizen distrust of remote institutions, and executive encroachment on all countervailing sources of restraint on the powers of government. Laurence Whitehead Nuffield College, University of Oxford

Acknowledgments

Giants of the natural sciences, and polar opposites in their political conviction, Albert Einstein and Wernher von Braun agreed that research essentially is what scientists do when they do not know what they are doing. These endeavors of letting run free the speculative mind are impossible to conduct without the assistance and guidance of others. So it is now my privilege to thank those individuals and organizations that supported me in my modest effort to shed a little more light on the processes and procedures of political interaction that keep baffling our inquisitive minds. This book is the result of my dissertation research at the University of British Columbia (UBC), Vancouver. Thanks is therefore foremost due to my super­ visor, Maxwell A. Cameron, and committee members, Antje Ellermann and Pilar Riaño-Alcalá. All provided meticulous comments that, regardless of how painful the realization of one’s own imperfection as a consequence may be, always helped to significantly improve the content. Their patient support, willingness to read numerous drafts at various stages of my research, and donate time and expertise greatly sharpened questions and results. To all, I am indebted for greatly benefiting from their advice. At the University of Oxford, I have to thank Annette Idler. The work together at the CONPEACE Programme has not only been rewarding and deepening for my understanding of Colombian politics; the intellectual environment, generally, is both challenging and illuminating. In the same vein, I have to thank my mentors at Nuffield College, Laurence Whitehead and Ezequiel González Ocantos, whose support, amongst other things, helped to hold a workshop in Oxford that facilitated the continuation of the research scripted into these pages. At the Institute for Latin American Studies (LAI) at the Freie Universität Berlin, I want to thank Sérgio Costa and Marianne Braig, who not only were instrumental in procuring more funding for studying Colombia through the Oxford Berlin Research Partnership; their keen interest in all things Colombian has truly been astonishing. Their mentorship in Berlin has been outstanding! Of course, research is never an individual effort. Even if we are bound to our laptops on our own—how the ivory tower more realistically looks these days—we not only proverbially stand on the shoulders of giants (of whom there are way too many to provide even a shortlist); we always benefit from our peers’ thoughts, arguments, and challenges. The Political Science Department of the University of

xii Acknowledgments British Columbia provides a collegial and thought-provoking environment that never lets a question go untested. I am thankful to my peers and professors—or simply stated: friends—who have continuously inspired deep debates. Beyond my committee, I particularly owe thanks to Alan Tupper for his interest in this project, and my friends Sam Reed, Edana Beauvais, Jan Lüdert, Katrina Chapelas, Agustín Goenaga, Stewart Prest, Rebecca Monnerat, Jen Allen, Sean Gray, Andrea Nüsser, Conrad King, Matt Gravelle, Matt Lesch, Charles Breton, Serbulent Turan, Bryan Peeler, Josephine Calazan, Priya Bala-Miller, Carla Winston, Jason Tockmann, Lydia Rhyne, Robertson McIlwain, and Kurt Hübner. In Colombia, I want to thank, above all, Hans Valero, who always helped me maneuver through when my Spanish was not that great yet; Luis Ricaurte, with whom I (prematurely) celebrated peace in Colombia in 2016, and Mario Cajas, who not only invited me to present my research at ICESI University in Calí but has always been someone to ask about anything per­ taining to any courts in Colombia. His expertise is staggering. At the University of Oxford, I want to thank my esteemed colleagues and friends Katerˇina Tkácˇová, Daíre McGill, Marcus Hochmüller, Élise Rouméas, and Tristin Beckmann for always having the time to examine controversial arguments—or football results—over a pint at the closest pub. Allessandro Corda, now at Queen’s University Belfast, also always took an interest in Colombia’s courts. Throughout my research I benefited from opportunities to spend time at distinguished institutions to develop my framework, collect data, and interpret results. I am grateful to Mattias Kumm at the Wissenschaftszentrum Berlin für Sozialforschung (WZB) for inviting me to his Rule of Law and Global Con­ stitutionalism Group—not once, not twice, but thrice! I could present my first findings from Colombia at colloquia and develop key components of my argu­ ments. Not least, I met Conrado Hübner Mendes, whose conceptualization of deliberative constitutional courts plays a crucial role in the argumentation that will follow. Christoph Möllers welcomed me as a DAAD researcher to his Chair at the Humboldt University. It was there and then that I commenced the writing of the manuscript for this book. The Universidad de los Andes in Bogotá, Colombia, allowed me to benefit from their first-class research facilities and helped me pro­ cure contacts to conduct interviews. I want to specifically thank Felipe Botero, Juan-Carlos Rodríguez, Angelika Rettberg, and Victoria Gómez Segura. Christian Steiner from the Konrad-Adenauer-Stiftung in Bogotá spent time and effort with me to organize interviews and co-ordinate them with student workshop held at their facilities. Marie-Christine Fuchs, Christian’s successor at Konrad-Adenauer­ Stiftung’s rule of law section, has become an equally supportive collaborator. I want to thank them both! Finally, at numerous conferences and workshop, col­ leagues have provided valuable comments. I want to mention David Landau, Gabriel Negretto, and Todd Eisenstadt at the 2016 Annual LASA Congress 2016; Julio Ríos-Figueroa, Catalina Botero, and Rudolfo Arango at various workshops at ICESI University in Calí. At the University of Oxford, Fernanda Farina organized a meeting with Denis Galligan at Wolfson College that invited Roberto Gargarella. I am thankful for the thoughtful discussion that occurred at the occasion. At Routledge, I want to thank Alison Kirk for her help to get this book project over

Acknowledgments

xiii

the finish line, as well as the three anonymous reviewers, whose comments greatly benefited the final version of the manuscript. Funding for my research has been provided through various grants from the Political Science Department at UBC. At CONPEACE in Oxford, support came through the Global Challenges Research Fund (GCRF), the Fritz-Thyssen-Stiftung, and the Oxford–Berlin Research Partnership. At the LAI in Berlin, funding has been provided by the Fritz-Thyssen-Stiftung as well as the DFG. For their support, I am, and remain, very thankful. Notwithstanding the importance of all of the aforementioned, the greatest debt goes to my wife Marlen, without whom it would have been difficult to live through the occasional anguish that comes with writing, and our little son Henri Egon Michael who gives us endless joy. In the end, I am only left to give credit to the var­ ious interviewees, who remain anonymous, but deserve utmost appreciation, because they often sat with me for hours to debate and discuss the evolution of Colombian constitutional politics. Their responses are the backbone of the analysis that follows in the next 300-plus pages. Of course, any errors of facts and interpretation are my own. Jan Boesten Berlin

1

Introduction Ethical Legal Discourse in Delegative Democracies

Democratic Legality Triumphs over Democratic Security The Colombian state is a social state of law (estado social de derecho), defined thusly in the first Article of the 1991 Constitution.1 The question of constitutional reform to allow presidential re-election, benefiting an extremely popular and powerful President, Álvaro Uribe, put the meaning of this clause to the test. Colombia’s Constitutional Court had to decide how far the legislature can go in amending the political charter before its basic principles are unrecognizably altered. In 2005, the Constitutional Court affirmed a law that sought to change the Constitution and allow one consecutive presidential re-election (C-1040/05).2 In 2010, the Court defected and disallowed a law that called upon Congress to ask Colombians in a referendum to change the Constitution to allow a potential second and consecutive re-election of the President (C-141/10). The Court found that a third term fundamentally changed the institutionalization of the separation of powers, which, as the Court argued, was an axiomatic principle implicit in the social state of law and consecrated in the Articles outlining the organic structure of the Colombian state (Art. 113). The Constitutional Court in Colombia, through its decision not to grant the seal of constitutionality to that reform, asserted its judicial power by drawing on the “substitution doctrine”. This doctrine increased the Court’s authority by means of a reinterpretation of constitutional norms enshrined in the 1991 Constitution (Cepeda Espinosa and Landau 2017). In the same period between 2005 and 2010, Colombia’s Supreme Court also disclosed scandals and revealed a web of nefarious relations between corrupt Members of Con­ gress (from the President’s coalition) and paramilitaries associated with the international narcotics trade. Together, the courts essentially stopped a corrupt and abusive executive in its tracks—an assertion of judicial power unprece­ dented in Colombia, and perhaps in Latin America. Since many constitutional reform efforts elsewhere in the Latin American region before and since have resulted in executive encroachment and in the erosion of judicial indepen­ dence, the surprising outcome in Colombia has profound implications for our understanding of how judicial empowerment can buttress the separation of powers and reinforce the rule of law. DOI: 10.4324/9781003229285-1

2 Introduction This book seeks to explain why the Court decided differently in 2010 from 2005, countered very evidently delegative trends in Colombia’s democracy, and reinforced the separation of powers and rule of law by buttressing horizontal accountability functions of a constitutional democracy. In the end, President Álvaro Uribe, who enjoyed significant formal powers assigned to the executive in the Constitution, and who could count on extremely high levels of popularity throughout his entire time in the presidency, complied with the verdict. This is not a narrative of institutional consolidation through interest alignment, as is the explanation given in neo-institutionalist accounts of the consolidation of the separation of powers (North and Weingast 1989; North 1990; Helmke 2004; Finkel 2008; for Colombia see Hartlyn 1988); nor does it follow a path-dependent logic of institution building (Ackerman 1991; Pierson 2000; Mahoney 2000; Page 2006). The consolidation of the separation of powers in Colombia as a consequence of the empowerment of the courts in general, and the Constitutional Court in particular, followed a deliberative and communicative logic that evolved around discursive patterns. Central for the significance of this case is the so-called substitution doctrine, which weighs if a constitutional reform exceeds the actors’ competence to implement such a reform. It is judge made law, giving dogmatic expression to the principle of pro­ portionality in constitutional adjudication (Bernal 2013). It is not only judge made law that does not appear in the written text of the 1991 Constitution (see, for example, Sierra Porto’s dissent in C-1040/05); proponents of Uribe’s reform argued that it explicitly contradicts the stipulation that the Constitutional Court can review constitutional reforms for procedural defects only (see Article 241, 1–3) by introdu­ cing substantive parameters for the review of constitutional reforms.3 For these rea­ sons, this doctrine is central for understanding the importance of this case for institutional theory, because by virtue of its creation and evolution it defies a strategic account of institutional development. It is therefore only prudent to quickly review the two decisions central to this book before explaining the contributions and designing the structure of this analysis. Legislative Act 02 of 2004: “By which some Articles of the Constitution are Reformed and Other Dispositions are Introduced” (C-1040/05) The 1991 Constitution only allowed one term in the highest office of the Republic. Álvaro Uribe, contrary to most of his predecessors, was hugely popular in Colombia, and had achieved considerable successes with his Democratic Security policy, which, as the name suggests, rested on the notion that good democratic governance arises in a secure environment. Due to the perceived fragility of these successes, and the key role that Uribe was said to play in implementing his policy, calls for amending the Constitution started surfacing early into his first term. In September 2003, Congress discussed the issue for the first time in plenum. The Colombian Constitution provides three ways for amending the charter and its norms: by legislative initiative through Congress, a popular referendum (that passes through Congress), and a constituent assembly. In 2004, Congress initiated

Introduction

3

the re-election reform with Law 02 of 2004 via the ordinary procedure through Congress. The reform project to allow a second immediate term in the presidential office encompassed four different Articles: Article 1, legislating the provisions for the incumbent in order to level the playing field; Article 2, altering Article 197 of the charter to the wording that any individual cannot be president for more than two consecutive terms; Article 3, specifying the role of the vice-president; and Article 4, guaranteeing the rights of the opposition in a statutory law. The last Article included a transitory clause, which gave the Consejo del Estado (State Council), Colombia’s highest administrative court, measures to legislate in case the law did not pass Congress prior to expiration of the legislative term, or if the Constitu­ tional Court declared parts of the legislation unconstitutional.4 Article 241 of the Constitution stipulates that the Constitutional Court has the mandate to review constitutional reform projects. The norm differentiates between the ordinary way through Congress, in which case the Constitutional Court reviews those complaints submitted by citizens, and the call of a referendum or constituent assembly, in which case the review by the Constitutional Court is not limited to the individual complaints by citizens, but is all-encompassing. Regardless of the way the constitu­ tional reform is initiated, the Court’s review is procedural only, and not substantive or material (“solo por vicios de procedimiento en su formación”; Article 241). The con­ stituents in the assembly in 1991 wanted a flexible constitution that is open to reform. In the case of the ordinary route, plaintiffs submit demands (demandas) against a law or reform. In 2005, citizens presented ten complaints against the con­ stitutionality of Uribe’s reform, arguing that it incurred procedural defects in the formation of the legislative act and exceeded the competence of the actors that introduced the reform (“substitution doctrine”).5 To begin with the formal com­ plaints, plaintiffs lamented that the authors of the reform had disowned the prin­ ciple of the separation of powers by having the executive assist in legislative debates; the required first reading in the first committee of the Senate had not met constitutional standards with the absence of the Vice-President of the Senate; an open debate in Congress was suppressed when the President of the Chamber of Representatives had ordered the vote to go forward on 17 June 2004, when opposition members protested with a walk-out; the exact text of the reform was not published 24 hours prior to debate in the Congressional Gazette as is demanded by the Constitution, nor were citizens’ interventions in the debates published in the Gazette; and complaints by fellow representative Germán Navas Talero against Yidis Medina against her last-minute change of opinion were not given sufficient voice in the process (C-1040/05). The most contentious part of the submissions argued that the reform constituted a substitution of the Constitution. They held that Congress did not have the competence to perform such a reform—only the primary constituent, namely the sovereign people, could invoke original constituent power to draft a reform of such substantive degree. They argued that Law 02 violated this doctrine, because of the absolute prohibition of re-election in Article 197 of the Constitution. In addition, they viewed the principle of equality violated by granting the President the possibility of re-election without according governors and mayors the same

4 Introduction prerogative. Finally, plaintiffs argued against a statutory law that empowered the Consejo del Estado to expedite norms in a transitory and supplementary way. The complainants held that Congress holds that power and cannot confer this power to any other institution—much less a judicial institution without powers to legislate. They submitted that this, too, constituted a violation of the doctrine of competence and constituted a substitution of the political charter for which Congress lacked authority.6 The Constitutional Court followed the plaintiffs only on the point that the statu­ tory law authorizing the Consejo del Estado with new powers amounted to a partial substitution of the Constitution, while the law enabling a second re-election in itself did not amount to such overreach of competence under the condition that the equality of chances for other candidates would be addressed in a statutory law. The Constitutional Court wanted to have the unfair advantage of a sitting president over his/her opponents minimized. In its entirety, the Court argued: The essential elements that define a social and democratic system based on the rule of law, [and] on human dignity, were not replaced in this reform. The sovereign people will decide whom to elect to the presidency, the institutions with supervisory or overseer roles in electoral matters completely preserve their powers, the checks and balances system is still operating, the indepen­ dence of the government branches is granted, the executive branch does not receive new powers, the reform contains rules to reduce the inequality in the electoral competition, which will be enforced by independent entities, and their decisions will continue to be subject to judicial review to protect the rule of law. It is not enough to make historical references suggesting that the drafters of the Constitution had the intention to limit presidential powers, and that therefore an amendment that contradicts that purpose is unacceptable. It is not enough to remark that the reason that may have inspired the drafters of the Constitution to prohibit a presidential reelection is today a valid standard, by which to conclude that the elimination of such a prohibition amounts to a substitution of the Constitution. The historical analysis takes us to the exact opposite conclusion. The Constitution has clauses [and] mechanisms to update the institutional design when the social and political reality requires it. Some people could argue that there is not enough political maturity in Colombia to adopt an immediate reelection scheme; that the reform could take us to scenarios of violent confrontation or institutional instability; or that the powers of the presidency could be used in the reelection project. Such opinions, to the extent that they are not an expression of an objective sub­ stitution or destruction of the institutional design, belong to the sphere of political assessments, of timing and convenience, and cannot be a matter to the constitutional judge. (C-1040/05) The reform went ahead, and Álvaro Uribe was re-elected President in a landslide victory. In addition, in the legislative elections held two months prior to the

Introduction

5

presidential elections, Uribe’s coalition increased its representative share in both houses of Congress, while the Liberal Party witnessed a crushing defeat in the elections to the Senate of the Republic. Uribe’s “coalition” in the Senate gained control of 61 out of 100 seats. Law 1354 of 2009: “By which to Convey a Referendum and Submit a Constitutional Reform to the People” (C-141/10) Uribe’s second term in office was a turbulent one. The President remained popular throughout the term and enjoyed spectacular approval ratings of up to 85%, not least due to equally spectacular military successes against the FARC (Spanish acronym for Revolutionary Armed Forces of Colombia), such as the hollywoodesque liberation of Ingrid Betancourt and the killing of some of its members of the high secretariat. However, at the same time, the second administration was repeatedly embroiled in different political and human rights scandals, tainting the successes of the military campaign against the guerrilla. Most damaging for the executive was the parapolítica scandal befalling Congress and threatening Uribe’s majorities in both chambers. This scandal, involving legislators (mostly from Uribe’s camp) and armed groups at the margin of legality, not only displayed that Colombia’s democratic regime was in real danger of parainstitutionalization of its representative institutions (the co-optation by non-state armed groups); Uribe’s response and confrontation with the investigating court openly showed the dangers of excessive (and delegative) presidentialism. As a consequence, even though Uribe’s personal popularity remained unaffected and high, the situation had shifted somewhat when the Court deliberated on the second constitutional reform to alter the norm guiding the election of the executive. Arguably, in 2005 denying a popular president another term carried a higher risk for the institutional stability of the country, while in 2009 excessive presidentialism very evidently posed a greater risk to Colombia’s democratic institutions. The law only included one Article that stated the referendum question of whether an individual, who had already been elected to the presidency, should be eligible to be elected once more. Congress had to pass the law that would pose that question to the people. Even though the referendum route differed from the constitutional reform in 2005, the juridical questions posed to the Constitutional Court were the same: did legislators violate formal aspects of procedure in the creation of the reform, and did the constitutional reform exceed legislators’ competence? If a constitutional reform takes the referendum route, the charter subjects it to an all-encompassing constitutionality test that must examine the reform against all precedents set in its jurisprudence. In this particular case, it again entailed that the Court had to investigate whether procedural violations had taken place as well as a substitution of axiomatic principles implicit in the political charter of the nation. The Court first affirmed its competence, arguing that its jurisdictional reach covered all procedures leading to a constitutional reform that pass through Con­ gress, even when they originate outside the formal institutions and are plebiscitary in nature. It then moved to address the five complaints litigators had submitted to the Court in the litigation period. When a referendum for constitutional reform is

6 Introduction convened, its committee of promoters and supporters must inscribe with the National Register of the Civil State (Registraduría Nacional del Estado Civil), and receive the support of 5% of eligible voters (C-141/10; 11). The promoters raised the funds necessary, but did so with little respect to rules governing such campaigns. As Botero et al. write, “the referendum records [were] murky and plagued with irregularities, such as self-loans between organizers that deliberately attempted to obscure the way in which the referendum was financed. Promoters spent six times more than the spending cap permitted”, with contributions by individual donors sometimes 30 times as high as allowed. In addition, Congress violated the requirement to have the votes of support confirmed by the National Registrar prior to vote on the bill and went ahead without approval. The next violation concerned the wording of the referendum. The text placed before the electorate actually read that Uribe would run again in 2014. Promoters of the law assuaged this when the bill was already in Con­ gress. The Court argued that Congress extra-limited itself by changing the wording this late in the process (2010). The final two procedural flaws involved the domiciliary right of the Colombian Congress. Inside the Court, the majority backing reform in Congress had five con­ gresistas (Members of Congress) among their ranks, who had joined Uribe’s coali­ tion from a party that had parted with Uribe (Cambio Radical). However, the laws against transfuguismo, the practice of changing party affiliation, disallowed such maneuvering between party caucuses, and Cambio Radical had sanctioned them, essentially invalidating their votes. Without their votes, the reform coalition did not have the required majority in Congress. Finally, the bill was struck down, because it was not passed within the time limit of the ordinary session that expired on 16 December 2008, at midnight. At that time, Uribe’s coalition had called in an extra­ ordinary session that began at 00:05 of 17 December 2008. However, since extra­ ordinary sessions had to be published in the Diario Oficial 24 hours in advance, that session, too, was invalid and the ordinary session long expired (Botero et al. 2010). These formal violations sufficed to end Uribe’s aspiration for another term in 2010. Inside the Court, the magistrates voted 7:2 to declare the law for a referendum unconstitutional on procedural grounds. The Court went further, though, and per­ manently closed the path through Congress for constitutional reforms of such magni­ tude. Five magistrates on the Court insisted during the deliberation with their colleagues that the Court had to apply the substitution test, because the Constitution stipulated that judicial review is all-encompassing for constitutional reforms taking the referendum route. The substitution test required the Court to investigate whether the proposed reform introduces a new element or whether it replaces an element from the original text. If it found that the proposition contradicts the original norm, it must declare the reform unconstitutional for extra-limiting the competence of its author. The Court argued that the separation of powers is an axiomatic principle of the social state of law that is inscribed in the first Article of the Constitution as well as in the democratic principle of the Colombian Constitution (Article 137). In addition, it specified that the separation of powers in the 1991 Constitution was institutionalized in what is referred to as the periodization of offices. Since pre­ sidents in Colombia’s constitutional design wield a large amount of power,

Introduction

7

constituents in 1991 had sought to constrain their authority by limiting the overlap of their time in office with appointees in offices that have accountability functions. For example, the president can present the shortlist of candidates for three positions (out of nine) on the Constitutional Court. Under the original scheme, only every second President, towards the end of his/her term, had the opportunity to get three of his/her choices elected by the Senate. With one re-election this had changed so that every President could have the opportunity to affect the composition of the Court, and enjoy a much longer time in office with a faction on the bench that the president him/herself had selected. The five magistrates argued that another term would further dilute these pesos y contrapesos (checks and balances), not only mod­ ifying an axiomatic principle of the Constitution, but substituting it for another. For that, Congress and the President lacked the competence, because only a constituent assembly wields original constituent power legitimating it to draft reforms of such degree. Thus, in a closer vote (5:4), those five magistrates voted the reform down on competence grounds (C-141/10). When President Uribe received the decision, he declared that the estado de opinión (state of opinion), the ideological figure in his rhetoric justifying the constitutional reform, was not a contradiction to the estado de derecho (state of law or rule of law), but an expression thereof; an assertion that he had not made unequivocally clear prior to the Court’s decision. Now he exclaimed that the estado de opinión must respect the law, because citizens’ participation, its key element, cannot function without con­ fidence in the law and the Constitution. Therefore, he complied with the Constitu­ tional Court’s decision, and conceded that there are limits to constitutional reforms.7 The importance of this decision arises not least because it was illustrative of an evo­ lution of institutional robustness of the Colombian judiciary in general, and the Constitutional Court in particular. In various fields of constitutional rights jur­ isprudence, the Court enforced a progressive and guaranteerist portfolio that encompassed private, social, and collective cultural rights. What is more, the decision and adherence by President Uribe and his followers to the verdict showed that Colombia was a constitutional democracy, in which the constitutional guardian’s verdict is respected as a guarantee for the “happiness of some and the tranquility of the rest”.8 The Question of Judicial Power in Colombia This book evolves around the question of why and how the courts’ authority increased between 2005 and 2010. Judicial power has two conceptual components. On the one hand, it is defined as the degree of judges’ institutional independence and, on the other hand, as the authority they command in their exercise of judicial review. Courts’ authority breaks down into vertical control involving rights adjudi­ cation, and horizontal control involving inter-branch adjudication (Helmke and Ríos-Figueroa 2011; Ríos-Figueroa 2011; Kneip 2011). To fully appreciate the significance of the Constitutional Court’s decisions, we need to emphasize three key observations: 1) President Álvaro Uribe was an extremely popular and powerful president, who had few scruples to utilize his transformative capacities to initiate a

8 Introduction far-reaching reform agenda that served his own political agenda; 2) the Court’s authority appreciably increased between 2005 and 2010; and 3) the jurisprudence of the Court involved a doctrine that is not explicitly mentioned in the Constitu­ tion, but a reinterpretation of the norms outlining judicial review of constitutional reforms. Consequently, an answer to the question that asks why the Court decided differently in 2010 from 2005 must account for these three caveats of this case. Institutionalist inquiries study the development of structures and agency therein sequentially (Sanders 2008), and I intend to do the same. To appreciate Uribe’s peculiar position in Colombia’s history—he commanded popular power like no other President before him—and the novelty of the 1991 Constitution that established the Constitutional Court, we need to pay close attention to macro historical processes of institutional development. Furthermore, the Constitutional Court’s increase of judi­ cial power must be viewed in the context of its 20-year existence at the time, making a structured post-genesis investigation of Colombia’s institutions indispensable. This will provide valuable insights for institutional development and theories of institution building. Finally, a close inspection of the processes taking place in the chambers of the Court during the deliberation of the specific re-election cases will help to under­ stand the discursive nature of decision-making processes in constitutional courts. Applying the vocabulary of comparative politics to these caveats, we are left with three research questions that will guide each chapter: � � �

Does the 1991 Constitution amount to a critical juncture in Colombia’s political history? What did the post-genesis evolution of Colombia’s constitutional jurisprudence look like? What motivated judges to develop and apply the substitution doctrine, which struck down Uribe’s reform to extend the number of terms in the presidential office?

This analysis seeks to demonstrate the relevance and importance of the Constitutional Court’s decisions for institutional theory by proposing a three-level inquiry that moves from macro to micro processes. The arguments that I propose in response to the three research questions are likewise threefold: firstly, the Constitutional Court’s moment of genesis is atypical in Colombia’s political history, because the constituent process that drafted the 1991 Constitution fundamentally differed from previous instances of institutional engineering. New actors participated in the drafting of the Constitution and procedures were much more transparent, public, and inclusive than before. It is a critical juncture in Colombia’s political history. Secondly, post-genesis evolution is difficult to grasp with the conventional or classical interpretation of path dependence that traces outcomes to early decisions from which they linearly develop, because legal decisions remain open to contestation. In Colombia, incremental change evolved from within the institution and implied course corrections. Thirdly, the institutionalization of deliberation inside the Constitutional Court centered on the differentiation between political and legal reasoning. Embracing the latter, delib­ eration inside the Court constrained magistrates’ individual preferences (be they

Introduction

9

ideological or material) and imposed an ethics of reason upheld by formal and infor­ mal rules governing deliberation. The empowerment of the Constitutional Court, viewed through the micro-institutional lens of legal decision-making, is therefore the consequence of deliberative rather than strategic action. In sum, this analysis will show an instance in which institutions ought not to be understood as structured expectations in a utility-maximizing game or regularized patterns of conduct, but as discursive structures. As such, the appreciation of judicial authority, manifest in the 2010 decision to curtail presidential terms, is an example of discursive institutionalism and empowerment—a discursive institutionalism that was evident in the constituent moment creating the Constitution, in post-genesis trajectories of institutional devel­ opment, and micro-institutional processes of judicial decision-making.

The Peculiar Case of Álvaro Uribe There are two aspects that make this case important for comparative politics: 1) Colombia sets itself apart from its Southern Cone neighbors, having a long history of democratic rule paired with high levels of violence; 2) Álvaro Uribe was an atypical leader in the nation’s presidential office, who commanded far-reaching de jure and de facto (popular) power that he used to move his agenda forward. It is this aspect of his style of governance that makes the 2010 decision so unique and important for our understanding of institutional development. To appreciate both phenomena, the propensity for violence with relatively democratic institutions as well as the particularity of Álvaro Uribe, it is equally important to place Colom­ bia’s political institutions in the context of long-established clientele relations. They set incentives for relatively democratic institutions resilient to fundamental change (and thereby also harbor a propensity for break-down and partisan cla­ shes). Conventionally, they also incentivized technocratic national leaders with an ability for brokerage amongst different regional factions, but with little charisma to inspire broad popular support to gratify majoritarian inclinations (Archer and Shugart 1997). Uribe was different, as he did inspire popular support, and used that popularity to transform Colombia’s institutional framework. His style of government fits well with elements of delegative rule that O’Donnell (1994) identified in defective democracies in the Southern Cone. Together, this makes Colombia an illuminating case to explore the empowerment of its highest court. The most remarkable trait of Colombian politics is, of course, that despite the tendency for violent conflict, democratic institutions continued to operate. Between 1849 and Uribe’s election in 2002, Colombian public politics was structured through the affiliation to one of the two political parties, the Liberal and Conservative Parties of Colombia, making it one of the longest lasting and most institutionalized party systems of the hemisphere (Mainwaring and Scully 1995). It survived prolonged internal conflicts as well as brief periods of military rule, and ensured the institutionalization of relatively competitive elections with a fairly generalized electorate.9 In its more than 200-year history, only twice was Colombia ruled by military men, but, contrary to military rulers in the Southern Cone, General Rafael Reyes (in power from 1904 until 1909) and General Rojas

10 Introduction Pinilla (1953–1957) did not violently seize power nor institutionalize bureaucratic authoritarian dictatorships (O’Donnell 1973). Rather, these brief intermezzos with military regimes helped civilian elites in either party to strike power-sharing deals to overcome impasses that had resulted in violent bloodshed. In essence, these periods had the effect of further embedding the party system, and creating legacies of bipartisan consociationalism (Hartlyn 1988). The long trajectory of the party system was rooted in its formidable capability to absorb social cleavages without actually addressing the grievances at the root of social polarization. It structured the exercise of political power in the regions of Colombia, and brokered the interaction between different regions in the creation of a national fabric. Beneath that stability was a web of symbiotic relations between patron and clients (Archer 1989; Eaton 2006), characterized by inequality in status, wealth and influence, uneven reciprocity, and affectionate, non-formal interaction between patron and client (Powell 1970; Dix 1967 and 1980; Francisco 1984; Mainwaring 1991; Martz 1997). Fully appreciating the effects of such clientele relations, in particular their ability to provide consocia­ tional appeasement while at the same time inhibiting institutional renewal, will be key for understanding processes of institutional engineering that are explored in the next chapter. In addition, and somewhat paradoxically, clientelism also safe­ guarded democratic institutions. Historically, and despite the far-reaching de jure powers accorded to the executive, Presidents in Colombia never realized the potential for detrimental presidentialism: the President relied on informal networks in the party system that essentially functioned like horizontal controls on his power (Archer and Shugart 1997). As we will see, Uribe’s popularity, and the authority he derived from that support, rendered this informal check on executive power moot. In the last three decades, the Colombian conundrum must be viewed in the context of armed clientelism, which pitted various guerrilla groups, paramilitary groups, and the state’s armed forces against one another. In the course of this conflict, the state’s already tenuous capacities further contracted, because it could not eliminate or neutralize its internal enemies and thereby fulfill the four state building tasks.10 As a consequence, Bejarano and Pizarro contend, “Colombia’s regime is a democracy whose faults are not to be located at the level of the typical dimensions of polyarchy (i.e. participation and opposition)”, but on the extrainstitutional playing field that is populated by armed entrepreneurs. It was, and remains, a “besieged democracy”, which separates into the electoral field, the rules that are largely respected by legally recognized actors, and the extra-institutional field, where the rules of war apply rather than the rules of democracy (2005, 237). It is in this context of a highly contested monopoly of violence, and unequal social relations beneath the formal institutional level, that Álvaro Uribe appeared on the political scene in Colombia. Coming from a landowning family and having gained some political recognition as Governor of his home Department of Anti­ oquia for defending a tough-hand internal security policy approach, he spun his political mythology around the (alleged) assassination of his father by FARC guerrillas. In 2002, after peace negotiations with the FARC had failed, and the

Introduction

11

AUC exercised a tremendous amount of power (the paramilitaries of the Autodefensas Unitarias de Colombia, United Self-Defense Forces of Colombia, hereafter AUC), some believed Colombia to be on brink of becoming a failed state (McLean 2002).11 This context helped Uribe to win an extraordinary victory in the 2002 presidential elections; for the first time under the new Constitution he did so in a land-slide victory that did not require a run-off election. With his heritage as a landowning rancher, Uribe could hardly claim to represent the popular masses. Yet, his tough-hand security approach (and successes) enabled him to amass popular support. In addition, he ran outside the two traditional parties and could thereby position himself outside of traditional politics, although he bene­ fited from his links with the Liberal Party. He used his cloud of popularity and maverick posture to cultivate a strong following, which in addition to internal security focused on creating natural bonds between rulers and ruled as part of a Democratic Security. It is the second point, creating tightly knit bonds between rulers and ruled, that later gave rise to observations that Colombia also exhibited signs of a delegative democracy. Uribe’s popularity, in the service of his sense of mission to expand his power, led the Council of Hemispheric Affairs to criticize Uribe in 2007 for political practices associated with delegative democracy (Kline 2009).12 Foundational to the concept of delegative democracy is O’Donnell’s observa­ tion that the newly democratized nations in South America matched Dahl’s defi­ nition of polyarchy without becoming fully institutionalized democracies, while at the same time they were not really in danger of reverting to authoritarian rule (O’Donnell 1994, 56). Most importantly, he noticed that presidents seemed to govern as they saw fit. O’Donnell understood that in ideal democracies each institution fulfills a crucial role in mediating between structural factors and the aggregation of individual interests and identities (59). However, research on pre­ sidential regimes showed that “the president is a one-person executive, [in which] the members of presidential cabinets are mere advisers and subordinates of the president” (Lijphart 1994, 93). Linz and Valenzuela, too, argued that presidential democracies almost inevitably place the president above the pragmatic politics of mediation between interests. They held that the framework of being directly elected for fixed terms seems to suggest to presidents that they have a personal mandate (1994). O’Donnell coalesced these observations into his model of dele­ gative democracy, in which “whoever wins (reasonably free and fair) elections to the presidency is thereby entitled to govern as he or she sees fit, constrained only by the hard facts of existing power relations and by constitutionally limited term of office” (1994, 59–60). A critical component of delegative democracies is their plebiscitary features, which are instrumentally utilized to push aside key horizontal control institutions and undermine the rule of law. As O’Donnell explains, strategies often involve directly appealing to the people and bypassing the legislatures, parties, as well as organized civil society groups (O’Donnell 1994). He argues that representation may exist alongside delegation, but never achieves the degree of fixedness as in constituted representative democracies due to the lack of horizontal account­ ability. Of course, in systems of separation of powers, the legislature and judiciary

12 Introduction enforce horizontal accountability. When these are pushed aside as nuisances to the rebirth of the nation, the republican dimension of democracy that enforces the differentiation between the private and public exercise of power is also pushed aside, undermining the effective exercise of the rule of law. As Cameron et al. showed: if presidentialism tends to erode regime stability, due to its rigidness, immobilism, and aggressive majoritarianism, coalesced into a one-man executive, it also tends to result in a vicious circle in which electoral outcomes may be respected, but legal norms are treated with disdain (2006). In order to measure delegative democracies comparatively, González (2014) devised an eight-component index of delegativeness: � � � � � � � �

The President is taken to be the embodiment of the nation. The policies of his government need bear no resemblance to the promises of his campaign. The President’s political base is a political movement; the President is above both political parties and organized interest. Other institutions, courts, and legislatures are considered impediments to the exercise of power. Exercise of power is non-institutionalized. The President nominates isolated and shielded técnicos to office. There is extremely weak or non-existent horizontal accountability. There is swift policy-making leading to the possibility of gross mistakes and hazardous implementation.

In order to visualize Uribe’s effects on Colombia’s institutions, it is useful to reproduce his results for Colombia together with the scores from other democracy and freedom measurement indices (Table 1.1). Freedom House measures civil and political rights on a scale from 1 to 7, while polity measures the state of national elections for competitiveness and openness, the nature of political participation in general, and the extent of checks on executive authority. The table shows that measures of delegative trends in political regimes, at least in the case of Colombia between 2002 and 2010, pick up tensions within a democratic regime that other measures of democracy do not detect. The values in the delegative column make readily apparent that Colombian democracy—or more precisely its democratic institutions—came under pressure beginning in 2006. It was the year that Uribe was re-elected and the parapolítica scandal involving nefarious relations between Members of Congress and paramilitaries were earnestly investigated and prosecuted by the Supreme Court. Senators and Representatives from the Uribista coalition—including Uribe’s cousin, Mario Uribe—were particularly prominent amongst the indicted congresistas. As a con­ sequence, Uribe and his advisors commenced a campaign against the Supreme Court that did not shy away from personal attacks. Also noteworthy is that the delegative trends climaxed in the last two years of Uribe’s presidency. This was the period when the Uribista camp prepared the campaign for another constitutional reform process to extend periods in the presidential office. Both these processes

Introduction

13

Table 1.1 Democracy measurements for Colombia, 1985–2011 Year

Freedom House Scores (1–7; 1 being most free)

Polity IV (Out of 10)

Delegative Democracy Indicator (Out of 8 Elements)

Freedom Rating

Civil Liberties

Political Rights

1985

2.50

3.00

2.00

8

2

1986

2.50

3.00

2.00

8

5

1987

2.50

3.00

2.00

8

0

1988

2.50

3.00

2.00

8

0

1989

2.50

3.00

2.00

8

1

1990

3.50

4.00

3.00

8

4

1991

3.50

4.00

3.00

9

1

1992

3.00

4.00

2.00

9

0

1993

3.00

4.00

2.00

9

0

1994

3.00

4.00

2.00

9

1

1995

3.50

4.00

3.00

7

5

1996

4.00

4.00

4.00

7

4

1997

4.00

4.00

4.00

7

5

1998

3.50

4.00

3.00

7

5

1999

4.00

4.00

4.00

7

2

2000

4.00

4.00

4.00

7

1

2001

4.00

4.00

4.00

7

3

2002

4.00

4.00

4.00

7

4

2003

4.00

4.00

4.00

7

1

2004

4.00

4.00

4.00

7

1

2005

4.00

4.00

4.00

7

3

2006

3.00

3.00

3.00

7

4

2007

3.00

3.00

3.00

7

5

2008

3.00

3.00

3.00

7

4

2009

3.50

4.00

3.00

7

6

2010

3.50

4.00

3.00

7

6

combined made evident that there was indeed a very obvious trend towards pushing aside horizontal control institutions as nuisances; and that the Constitu­ tional Court constrained executive and legislative powers to reform the Constitu­ tion, precisely at a moment when the executive attempted to expand its powers of constituted powers at the expense of horizontal constitutional controls. In the end, it should not come as a surprise that the Uribe presidency turned more delegative than previous presidencies in Colombia. His ideal of social

14 Introduction cohesion between rulers and ruled is built around notions of natural leadership, communitarian ideals of citizenship, and authoritarian trust in political leadership. His sense of mission and agenda to reform the constitutional order not only emulates these ideals, but further developed them into practical doctrines. In the first term he mostly spoke of the communitarian state, and in the second he used the figure of the estado de opinión (state of opinion as supposed to the state of law). It builds on rousseauesque notions of the General Will, claims for itself to precede constituted power, and thereby defies any control other than public opi­ nion (hence, estado de opinión). Neither is it surprising that the “maximalist” interpretation of the estado de opinión can be used as a means to treat horizontal controls with contempt.13 What is surprising is that both the Supreme Court and the Constitutional Court of Colombia withstood these pressures and earnestly followed their tasks of independently and autonomously applying the law of the land, regardless of the popularity and power of the President and the means uti­ lized by the executive against their mandate. These characteristics of Colombian politics, the uniqueness of Uribe, as well as the audacity of Colombia’s courts signify the importance of this case for the study of comparative politics.

Contributions: The Primacy of Democratic Legality over Democratic Security and Its Significance for Discursive Institutionalism The Constitutional Court’s decision to prevent a potential third Uribe term in the presidential office essentially meant that Democratic Legality, the legal authoriza­ tion of political power, prevailed over the imperatives of Uribe’s Democratic Security policy program and its communitarian and plebiscitary appeal to circum­ vent horizontal control institutions. This analysis will show that this outcome has profound ramifications for our understanding of micro and macro institutional processes, as well as post-genesis transformations of institutions. Each chapter will engage discursive explanans for the phenomena that constitute the explananda of this study—the rights-expanding nature of the 1991 Constitution, post-genesis relations within the judiciary and its effects on the institutional balance between the branches of government, and the judicial control of executive encroachment. The study adds important nuances to the vocabulary that explains institutional origins as critical junctures by highlighting the communicative origin of the 1991 Constitution, post-genesis trajectories as path-dependence by contending that judicial institutions remain open to contestation thereby defying irreversible lockin mechanisms, and judicial behavior in high courts by asserting that discursive action flourishes, and has an effect on outcomes in carefully administered dis­ cursive spaces. Key is that each concept—and therefore each chapter—is con­ nected through the predictions of institutional theory. This compendious organization of the book, combining macro and micro settings, is best suitable to explain the particulars of a single decision-making process within the context of broader institutional fault lines in Colombia. The book aims to add to a discursive understanding of institutional analysis, particularly judicial institutions. The contributions are two-fold: first, it challenges

Introduction

15

aspects of the new institutionalisms by productively utilizing the critical juncture framework, path-dependent conceptualizations of institutional trajectories, and strategic actor–focused explanations of judges’ behavior. While not redressing these concepts, the study posits to include discursive variables in the equation. Second, it also adds the caveat of procedure to our understanding of discursive institutionalism. In its more conventional application, discursive institutionalism focuses on the evolution of ideas, and how they travel to explain policy outcomes (see, for example, Schmidt 2008). In this study, I conclude that specific proce­ dures—their inclusiveness and transparency in the creation of the Constitution, and the rigid protection of rules of deliberation in the 2010 decision—were important, even essential, factors. Moreover, the book details that jurisprudential trajectories follow characteristics that are not found in conventional conceptualizations of path dependence, and require a discursive conceptualization of path dependence for judicial institutions (that remains more open to contestation, and therefore less deterministic). To be sure, while relatively novel in the school of institutionalism, there have been studies to engage discursive explanans for institutional outcomes—the force of ideas to propel (progressive) judicial independence in Colombia (Nunes 2010), and paths of endogenous growth of judicial power in Colombia’s Supreme Court (Bakiner 2020). What is missing in these, and other evocations of discursive institutions, is the centrality of procedures that facilitate open-ended and the force-of-the-better-argu­ ment-prevailing discursive practices. I work with Hübner Mendes’ definition of deliberation in constitutional courts—a type of practical reasoning in collective deci­ sion-making processes. The defining feature is that participants are able and willing “to transform their preferences in the light of well-articulated arguments” (2013, 18). The centrality of the ability to change preferences points to the distinctive importance of procedures while coming to collectively binding decisions that can result in dis­ cursive transformations of institutions. This is not a coincidence: Habermas’s dis­ course ethics shifts the normative nucleus of political legitimation from preconceived moral assertions, rooted in metaphysical or substantive notions of the good, to pro­ cedures of rational argumentation (1996). It carries the normative foundation of deliberation to exclude every kind of coercion other than that of the better argument, so that all motives except that of the cooperative search for truth are neutralized (Alexy 1989). This importance of process has not fully gained entrance into comparative analysis of judicial institutions. Yet, discourse and delib­ eration enabling proceduralism was evident at the constituent assembly in 1991 as in the 2010 Constitutional Court decision to curtail the power to reform the Constitu­ tion. This section will introduce the core arguments, including the conceptual and methodological implications, that are developed and elaborated in each chapter, and the contributions to institutional theory those arguments make. Deliberative democratic theory, the foundation of discursive institutionalism, contends that authority emanates from communicatively shared validity claims that arise and are tested in deliberative processes (Habermas 1996; March and Olsen 2008, 15). It is somewhat difficult to align this normative core of communicative action with one of the most important analytical concepts of the historical

16 Introduction institutionalist school, critical junctures. After all, as Sanders laments, historical institutionalism turned against the behavioralist accounts of the 1950s and 1960s that took the institutional structure for granted; they, ironically, also rediscovered “the momentous agency of ‘state managers,’” exactly at a time when historians started rejecting the study of “powerful white men” for the histories of the people (2008, 45). Critical junctures, too, highlight the importance of those actors pre­ sent in a juncture, and tie the longevity of specific institutional developments later on to those primary decision-makers. Nevertheless, the application of the critical juncture framework in this book will actually show collective agency, and how it formed the procedures of constitutional genesis in what amounted to an atypical constituent process—atypical in the context of Colombia’s history. Commu­ nicative action, reflected in the actors involved in the drafting of the Constitution, the selections of topics, and the design of processes to deliberate the new political charter, is what set it apart from earlier experiences of institution making in Colombia. My analysis will utilize the Colliers’ critical juncture approach of the study of institutional genesis (2002) that focuses on “(1) a major episode of institutional innovation, (2) occurring in distinct ways, (3) and generating an enduring legacy” (Collier and Munck 2017, 2). Collier and Collier (2002), as well as Collier and Munck (2017), contend that any social event can be of significance to social scien­ tists, but for a critical juncture “[e]pisodes that leave an enduring legacy” are key. Conventionally, such legacy-producing events are identified in “contrasts among cases in comparative analysis”, but my study engages in “comparing outcomes in a single case with counterfactual alternatives” (Collier and Munck 2017, 2). Embed­ ding the creation of the 1991 Constitution in comparative analysis has yielded important results. Negretto (2013) systematically investigates the mechanisms in constitution-making processes and their institutional outcomes. Combining quanti­ tative and qualitative research, he shows that expectations about future political development affect the selection of institutional setups. This chapter will add to these arguments by showing how constituent processes benefit from, and are affec­ ted by, citizens’ participation—a core tenet of “deliberative” theories of democracy. I will disaggregate the constituent process of 1991 with two other profound insti­ tution making episodes in Colombia’s history—namely, the reforms associated with General Rafael Reyes and the National Front. The method of difference dictated their selection (institutional engineering was precipitated by a political crisis and wide­ spread violence) while excluding others (such as the reforms of Carlos Lleras Restrepo, President from 1966–1970, which fall within a more peaceful period in Colombia’s history). The internal disaggregation has the advantage of providing a thick analysis of the processes that constitute the juncture, and a counterfactual causal narrative that is based in the internal variation of the Colombian case. This will show that the 1991 Constitution, by virtue of the events and processes that triggered the collective movement calling for a constitutional renewal, and the implementation of procedures to negotiate the norms of the political charter, differed from previous instances of institutional engineering. A student movement had spontaneously formed after the traumatic assassinations of promising political leaders. It then

Introduction

17

morphed into a genuine, cross-sectional, and multi-ideological social movement that engaged in communicative action to draft suggestions for a new constitution. The procedures to discuss norms were public and gave citizens the opportunity to submit points that had to become the subject of debates in the constituent assembly. Finally, the composition of the assembly itself included various new and formerly marginalized political actors. It therefore shows two important caveats of deliberative democracy: 1) the appearance of communicative power to 2) generate ideal speech situations. Given the institutionalization of this public constituent process, it is not surprising that the result, the 1991 Constitution, was very different from previous constitutions, and, above all, included an expansive rights catalogue with mechanisms to enforce these rights. In fact, there is a causal connection. Eisen­ stadt et. al have affirmed core principles of “deliberative” theories of democracy that value citizens’ participation at the beginning of constituent processes rather than post-factum ratification through referenda (2015). The next chapter will show that constitution making in 1991 was a neo-constitutionalist moment that valued rights, which in turn are rooted in the authorization of political power through publicly shared validity claims. Historical institutionalists provided comparative politics with one of the most powerful tools to conceptualize institutional development to show that institutions reproduce themselves in a path-dependent logic (Pierson 2000). This logic implies that, once a decision is taken, lock-in, increasing returns, and other equilibriumbased mechanisms ensure that further institutional developments follow the direc­ tion of the early decisions (Mahoney 2000). Judicial institutions pose somewhat of a challenge to path dependence, because those lock-in mechanisms function in the context of distributive dynamics around material interests. Judicial institutions, however, are pervaded with normative connotations of deontological right claims consecrated in written text. The institutional contest is therefore not solely about who gets what, but also about the meaning of constitutional clauses in (high-stake) political competitions. This challenge should not go unanswered, however. Rather, after applying the critical juncture framework to the constituent process of 1991, it is my conviction that testing the trajectories of post-genesis legal institutions through the lens of path dependence opens the opportunity to further our under­ standing of institutional change. Moreover, after the implementation of the new Constitution, Colombian jurisprudence evolved by precedent, resulting in a natural valuation of earlier decisions over later ones. Together with the origin in a moment of contingency that could not be predicted on existing institutional patterns, two central conditions for path-dependent development are fulfilled. Therefore, Chapter 4 will test if and how the path-dependent logic holds for the post-1991 evolution of legal institutions, contrasting between predictions from the discontinuous and incremental change models of institutional change. The path-dependence chapter is rich in detail, complex in the interweaving of concepts and scope conditions, and, as a result of both, extensive in length as well as rewarding in terms of the results produced. It begins with a detailed conceptual discussion that will yield a classical conceptualization of path dependence around the contingency of the initial juncture, the reproduction of initial decisions, and

18 Introduction the linearity of further developments down the path of those original decisions. It is hypothesized that a divergence from those three components must render a correction to path dependence in the context of judicial decisions. The profound overhaul of legal institutions in the 1991 Constitution, reflective of the student movement’s agency in the constituent process to redraft the political charter in the first place, demands that a meaningful post-genesis analysis of insti­ tutional development must pay credit to the centrality of rights in the constitutional framework. The tutela—a writ of protection any Colombian citizen can file against rights abuses by public institutions—is evidently concerned with human rights. It has an additional advantage for studying path-dependent trajectories: it creates jur­ isdictional tensions within the judiciary (amongst high courts), and, as a con­ sequence, provides wedges for other actors (i.e. Members of Congress) to exploit for their own material gains. In other words, the constitutional setup can give rise to an intrinsic web of distributive dynamics, often identified as the source of pathdependent developments, which cut through and across the branches government. The Constitution subjects decisions by all public institutions to rights-based judicial review, but also states that the Supreme Court is the highest instance of the ordinary justice system. This created a clash over jurisdiction: the so-called choque de trenes (train crash) evolved around the question of whether the Con­ stitutional Court can review decisions of other high courts. The potential for dis­ tributive dynamics was amplified by the absence of parliamentary immunity in the new political charter. The 1991 Constitution invested the Supreme Court with the task to investigate and judge criminal allegations against Members of Congress. Members of Congress therefore have a material interest, if subjected to criminal allegations of corruption, to utilize the tutela to claim their rights to freely delib­ erate without the threat of prosecution. Together, this sets up a configuration of actors and rules that is valuable for our understanding of institutions and their trajectory, because: 1) various courts with different jurisdictions and contradictory normative principles resulted in a distributive conflict over the application of those norms; 2) these tensions were amplified by the absence of parliamentary immunity (further adding distributive dynamics to the equation); and 3) the connection to rights and rights defense introduced ideational variables that involve the meaning of foundational constitutional principles. In a longitudinal study of two political scandals involving the criminalization of Congress, and the strategies used by Members of Congress to avoid prosecution, I will show that in the course of the proceso 8000 (Process 8000) only few Members of Congress were prosecuted, while in the parapolítica (parapolitics) scandal 102 were investigated between 2006 and 2010, making Uribe’s second term a much more contentious term in office. Three key observations conclude Chapter 3: the continuous production of legal facts—social facts produced in judicial processes— set parapolítica apart from the proceso 8000. This showed, not least, in the exten­ sive response that Uribe and his administration orchestrated against the Supreme Court and its individual judges. Furthermore, the continued production of legal facts was to a decisive part the result of the Constitutional Court refraining from overturning the substance of Supreme Court decisions involving the privileges and

Introduction

19

duties of legislators—during the proceso 8000, the Constitutional Court shut down criminal investigations in an infamous tutela decision. In the end, the tra­ jectory of jurisprudential development in the question of the tutela and the choque de trenes does not become more intelligible with a classical understanding of path dependence that foregoes notions of institutional learning. The outcome of the choque de trenes in the context of the parapolítica scandal speaks to March and Olsen’s reminder that actors reinterpret institutional ideals in a given context. Consequently, change evolves within institutions from “both intra- and inter-institutional dynamics and sources” (2008, 12). Here it is the interpretation of norms and how they apply to criminal processes that have poli­ tical significance. Mahoney and Thelen, too, call attention to the incremental changes that occur when actors are debating compliance with institutional imperatives (2009, 10–11). In these episodes, it was readily evident that actors constantly brought resources to alter the meaning of specific constitutional and legal norms to affect the outcome in their interest. I conclude that the specific path dependence of judicial institutions belongs in the camp of incremental change rather than discontinuous change. Ultimately, the chapter identifies a form of insti­ tutional learning that appeals to the notion that change evolves from argumentation itself. It was simply not argumentatively feasible to defend parliamentary immunity as a bulwark for free deliberation when Members of Congress utilized the norm for impunity. In the context of the parapolítica scandal, Onur Bakiner made an interesting, and convincing, argument to explain the empowerment of the Supreme Court (2020). He writes that the “parapolitics process itself redefined the justices’ inter­ ests, self-perceptions, and, consequently, limits of jurisdiction” in what he terms the endogenous growth of judicial power (604). The 2009 decision to take on cases of Members of Congress who had resigned their posts was an evident divergence from the Court’s prior jurisprudence, and Bakiner is right to assert that the Court accumulated authority by extending its jurisdiction to these cases. My analysis does not forego the validity of Bakiner’s argument—on the contrary, the theoretical reasoning behind the endogenous growth of judicial power finds application in the chapter focusing the Constitutional Court’s empowerment. Nevertheless, beyond focusing on a different court, I am actually trying to do something else in the chapter on path dependence in judicial institutions that focuses on the parapolítica scandal. Rather than zeroing in on an individual court’s behavior, I am interested in the post-genesis institutional trajectory of the judiciary as a whole, and in relation to the other branches of government. It is my contention, however, that this holistic approach—constructed around the internally disaggregated case study—provides nuances to the institutional nar­ rative that are outside the scope of Bakiner’s conclusion. I show that affidavits from paramilitaries were crucial for the evolution of the parapolítica scandal, as these produced “legal facts” that proved to be more difficult to politicize. These affidavits were not endogenous to the Supreme Court—Justice and Peace Courts were separate entities—but not part of the exogenous shock either: Bakiner defined the Constitutional Court’s decision to conditionally allow the Justice and

20 Introduction Peace law as the original exogenous shock. The inter-court relations also play a much more important role in my argumentation, because my longitudinal analysis of the choque de trenes shows more conciliatory relations between the high courts during parapolítica. After all, the Constitutional Court’s decision to back the Supreme Court’s jurisprudential shifts that increased power under its jurisdiction were crucial for the latter’s decisions to remain valid. There are additional, policy-relevant con­ tributions that stem from the holistic approach to study the parapolítica scandal through path dependence that I chose here. For example, several apex courts make it more difficult for executive encroachment over the judiciary (see Pou Giménez 2019). The methodological setup also highlights the relevance of the transitional justice process itself and how it affects subnational and national institutional relations. As explained, the fact-finding of Justice and Peace Courts was crucial for the pro­ duction of legal facts and the political capital other branches of government could employ to further their legislative agenda. (We will see that the legislature was char­ acterized as debilitated and the executive encroaching on its mandate.) Moreover, the transitional justice process affected the armed group behavior of the paramilitaries. Already the negotiations to the Justice and Peace Process created centrifugal forces within the paramilitary organization, pitting the various factions against each other. The latter point will require more comparative research, given the ongoing peace process with the FARC that is now in its implementation phase (see CONPEACE Programme at University of Oxford). What is clear at this point is the value of ana­ lyzing the Court’s behavior in full view of the institutional context. After inquiring constitutional origin and post-genesis trajectory, Chapter 4 explores the empowerment of the Constitutional Court, manifest in its decision to curtail the viability of constitutional reform through the ordinary, legislative route. That analysis is an answer to Epstein and Knight’s call to look at the argumentation of judicial decisions to understand what drives judges in their reasoning (2013). The separation of powers model of judicial behavior borrows the modulation of agency adopted in rational choice models and presumes a utility-maximizing and forwardlooking judge. The strength of these models is their precise conceptualization of the interaction between behavior and institutions that are amenable to generalization, theory building, and falsification. The weakness lies in the image of human nature that misses important facets of human motivation (Hall and Taylor 1996, 950). Chapter 4 of this book will explicitly deal with judicial empowerment in the course of constitutional reforms in Colombia. It embeds the analysis in the tensions between majoritarian and constitutionalist ideals of democracy, and contrasts the strategic and discursive constitutional judge. The substitution doctrine was the critical doctrine to disallow a third presidential term. It is essentially a dogmatic expression of the proportionality principle in con­ stitutional adjudication (Bernal 2013). It submits constitutional reforms to a test of whether the new norms contradict the original norms to the degree that it substitutes and not merely reforms them. The critical step in this test requires the judge to establish axiomatic principles in the constitution, and subsequently expound the relation of the reform to those principles. Here, the judge exercises a form of public reasoning that weighs between positions.

Introduction

21

The reconstruction of the creation and evolution of the substitution doctrine will yield four important empirical insights. Firstly, the substance of the doctrine itself filled an important gap in constitutional theory of how to legally regulate constitutional replacement (see also Dixon and Landau 2015). Secondly, the need to fill that gap and create a doctrine to understand constitutional replacement in legal terms arose because of Uribe’s will to transform the constitutional framework through populist and plebiscitary appeals. Thirdly, there is little evidence that the Court acted as a prudent strategizer that carefully built its jurisprudence with consideration to other (potentially more) powerful actors. The jurisprudence made the most important steps in cases that were far from inconsequential to the administration’s agenda. It was exactly the power and influence Uribe wielded in Colombian politics that urged judges to act rather than tread carefully and cau­ tiously. Fourthly, at a crucial point during the deliberation of the constitutional control of re-election reform, a majority of judges inside the court insisted on applying the substitution test. It was this moment that resulted in declaring the reform unconstitutional not merely on procedural grounds, but on quasi-substantive grounds, highlighting the necessity to fully appreciate the importance of the implementation of deliberation inside constitutional chambers. For those reasons I conclude that the magistrates on the Constitutional Court in Colombia behaved like discursive rather than strategic actors. In order to classify the importance of that decision for theories of judicial beha­ vior, we must remember that legal rationality is not identical with the certainty of outcomes—an agreement that I share with Hübner Mendes (2013) and Alexy (1989). This is one of the reasons why the conceptualization of the deliberative judge goes beyond the confines of the legalistic school of judicial behavior, which reifies law into something that is already out there needing to be found by the constitutional judge. Rather, the ontology of legal decision-making in general, and constitutional adjudication in particular, evolves around the uncertainty of what the law vis-à-vis a particularly pertinent constitutional issue actually is. The core case, the creation of the substitution doctrine, is a case in point, because the doctrine itself, as well as the question of constitutional reform to extend presidential terms, had never been raised in Colombia. If there is indeed latitude and reasonable controversy in law application, it is because “the concept of law combines a factual and an ideal dimension in realizing ‘optimal coupling of legal certainty and justice’”. This entails that you cannot know “the law just by reading its public and printed black-letters”, and addition­ ally means that legal deliberation remains an activity of using public reasoning to come to politically consequential decisions (Hübner Mendes 2013, 57). Norma­ tively, the indeterminacy of law, judges’ lack of direct democratic legitimation, and the burdening to uphold principles of justice and the rule of law commands par­ ticipants in constitutional decisions to be open “to transform their preferences in the light of well-articulated arguments” (18). The deliberative judge remains open to change her opinion in the face of the better argument—an event that occurred in the 2010 decision that curtailed presidential re-election in Colombia—and thereby relies on the soundness of procedures of deliberation. The insistence of

22 Introduction the majority in 2010 to apply the substitution test was justified with the imperative to properly follow the procedures of constitutional adjudication, and the evidence will show that this serendipitous moment runs counter to the predictions of the strategizing judge, who has fixed preferences. Rather than prudently strategizing (Rodríguez-Raga 2011), and cautiously acting on the preferences of the other branches of government (Helmke 2004), the Constitutional Court acted vigilantly and deliberately in cases of immense political importance. This insistence on cor­ rect procedure also adds to Bakiner’s notion of endogenous growth of judicial power, he identified in the Supreme Court’s decision to re-take cases of Members of Congress in the parapolítica affair. (He identified the sources in justices shifting self-definition, snowballing effects of judicial decisions, and conditional civil society groups’ support that entrenches a court’s self-definition; 2020, 603.) Moreover, while Nunes (2010) is certainly correct to stress the importance of ideas and “idea­ tional carriers” in the design of constitutional guarantees and their application in constitutional adjudication, the 2010 decision speaks to a different quality of dis­ cursive institutional development. First, it is not readily evident what ideal con­ secrated into a constitutional norm in the 1991 Constitution is relevant for the substitution doctrine. After all, it was a doctrine, or principle, not mentioned in the original text. On the contrary, arguably the clause that the Constitutional Court could only review procedural defects of constitutional reforms, and the constituents’ will to have an easily transformable political framework (see Chapter 2), runs counter to the substitution doctrine. Second, the evolution and eventual application of the doctrine in 2010 shows that procedures are key for ideas finding their way into the constitutional jurisprudence of high courts. The contributions of this book are to incorporate notions of deliberative democracy and discursive institutionalism to the (comparative) study of judicial institutions. What is more, I aim to fully extrapolate the importance of procedure for institutional change while discursive institutionalism has conventionally focused on the substance of deliberation, namely ideas and how they travel (Schmidt 2008). Deliberative democracy is associated with critical reason giving amongst equals that occurs prior to actually taking the decision (Warren 2017; Habermas 1996). A key caveat of deliberative democracy as a model is that it is “primarily a theory of communicative responses to disagreement, preference formation, and collective will formation, focused on mediating conflict through the give and take of reasons” (Warren 2017, 40). It is therefore not a theory of power or inequality, but of ethically coming to an understanding. I contend that it is for that reason that it is in fact a good theory to explore the processes of coming to judicial decisions in high courts, as well as those processes that create the constitution itself, and assert meaning of constitutional clauses in a post-genesis context. Generically, my analysis aims to show that institutions must also be understood as discursive structures. Institutions, Douglas North famously noted, are the “humanly constructed constraints that shape human interaction”. They are at the root of historical change by devising the incentive structure in human exchange, be it “political, social or economic” (1990, 3). O’Donnell defines institutions as “regularized patterns of interaction that are known, practiced, and regularly

Introduction

23

accepted (if not necessarily normatively approved) by given social agents” (1994, 60). In the processes assessed in this book we will come across institutional fra­ meworks that do not simply function as the rules of the games that structure the expectations of strategic actors. Neither are they cultural norms that discipline actors into subjection. Rather, we will see that institutions are structures, in which actors explore and negotiate the meaning and significance of norms—rules in other words—with reference to a constitutional text and the intention of the constituents that drafted the charter in the first place. The text as basis for nego­ tiation ensured at critical moments that the negotiating process was not simply an expression of the “survival of the fittest” that benefits the actor with the most profound resources. Texts and intention of constituents entailed that these pro­ cesses involved interpretive dimensions that required actors to make arguments as to how the written text of the Constitution applies in a given context. This introduces a logic of public reasoning that forces even strategic actors with fixed interests (such as defense and prosecution in a court room) to not solely engage in rhetorical action but attempt to persuade an audience with argumentative ration­ ality (Risse 2002; Habermas 1996). When Schmidt summarized the advantages of a discursive institutionalism, she noted four key propositions: 1) regardless of definitional distinctions, it takes ideas and discourse seriously; 2) it sets these ideas and discourses in institutional contexts; 3) ideas carry “meaning” and discourse follows a “logic of communication”; and 4) discursive institutionalism induces a dynamic into our understanding of institutional change that the other three insti­ tutionalisms, “equilibrium-focused” as they are, cannot comprehend, because the locked-in equilibria often pose as insurmountable (2008, 304). This study attests to the logic of communication and the open-endedness of discursively contested institutions. To assert an argumentative rationality on processes of constitution making, and when the meaning of constitutional clauses was at stake, was con­ tingent on discursive structures rooted in a fairness of procedures that facilitate the collective search for the truth.

Case Selection and Data Collection This case study aims to do three things: 1) explore a macro-institutional outcome, namely, the contingency of the 1991 constituent process and resulting rights-based constitution; 2) understand post-genesis institutional trajectories—specifically, the indeterministic development of the choque de trenes between Colombia’s high courts; and 3) offer a novel interpretation of micro-institutional decision-making processes that manifested in the creation of the substitution doctrine by Colombia’s Constitutional Court. It therefore disaggregates the Colombian case into three subcases of institution building to explore the genesis of the 1991 Constitution; it contrasts two political scandals involving the criminalization of Congress to gain analytical leverage for a sophisticated understanding of the post-genesis jur­ isprudential trajectory of the tutela contra sentencias; and it reconstructs the delib­ eration of the substitution doctrine in all relevant decisions between 2002 and 2010. These analyses build on an extensive secondary literature review of both

24 Introduction Colombian and international scholarly work. I held over 40 in-depth, structured, elite interviews with judges of the Supreme and Constitutional Court, clerks on either court, legal scholars, Members of Congress, and journalists. Amongst others, I interviewed the entire Constitutional Court that delivered the 2010 elections, and a large share of the clerks working for those judges. These interviews were tran­ scribed and analytically ordered according to orders of observation. In addition, the research integrated systematic content analyses of Colombia’s most important print publications—Semana, El Tiempo, and El Espectador—together with online pub­ lications such as La Silla Vacía. In addition to the inclusion of these content ana­ lyses in the main body of the book, Appendix A includes a detailed exploration of the entire time Uribe occupied the presidency (2002–2010) based on the quality of discourse in Semana, which is accompanied by comments that explain the events occurring in each year. I surveyed the entire content of Semana’s “Nación” pages for those years, and disaggregated the articles according to the space devoted to critical or positive reporting on Uribe’s policies. Finally, I also gathered quantitative data on the conflict in Colombia, opinion polls, and judicial caseloads. Together, these elements of a qualitative research design provide the evidence for making the causal claims outlined above—and expounded in the following chapters. They also imply methodological caveats that this section of the introduction will address. Methodologically, this study faces particular challenges because of its case study design (highlighting validity issues of causal inferences), and focus on judicial processes (emphasizing data validity issues). Any social science research involves trade-offs regarding its validity and generalizability. Case studies in particular are good in showing some things while weaker in disclosing others. Therefore, we need to be fully aware and self-conscious about what this case of Colombia can in fact tell us about (judicial) institutions without making too sweeping general­ izations. Case studies can be very useful for closely analyzing the applicability of specific theories in special cases and generating new hypotheses. The data validity issues in this inquiry arises for two reasons: 1) the focus of this book is on judicial institutions and reasoning; and 2) dissatisfied with the explanations from the neo­ institutionalist views, this book looks at the importance of discursive practices to develop explanations for the outcome. Judicial decision-making is particularly challenging to study for social scientists, because, as a general rule, it takes place behind closed doors. Colombia’s high courts, and the Constitutional Court in particular, are no exception to that. On the contrary, one of the specific features of the Colombian Constitutional Court that this book will analyze and include in the explanation of the outcome is that the Court went to great length to shield the deliberation from public pressure and influence. Discourse, however, implies that policy-makers communicate between actors, and justify their ideas against a back­ ground of overarching philosophies. How then can a discourse that specifically shields itself from being too public be a form of public reasoning? Here I will try to provide satisfactory answers to these methodological admonitions and thereby highlight the value of this analysis for broader social science research. The value of studying the outcome of the 2010 constitutional decisions derives, first, from the fact that Colombia constitutes a special case within Latin America’s

Introduction

25

political history with little experiences of authoritarian rule but long periods of inter­ nal violence. Second, it had historically exhibited a peculiar form of presidentialism, in which the executive enjoys expansive formal powers, but is constrained through informal clientele networks organizing the distribution of political and material rewards (see above). President Uribe’s administrations were uncharacteristic, because, armed with personal charisma, he was able to amass great popular support, dominate legislative politics, and discipline a factionalized and decentralized Congress. Third, even though judicial independence has had precedents in Colombia’s history, the 2010 Constitutional Court ventured into uncharted legal territory by invalidating a constitutional reform based on the distinction between reforming and replacing a constitution. It was, as explained at the outset, a decision that reinterpreted con­ stitutional norms in a novel way, and it did so in a context inauspicious for the growth of judicial power and independence. All of these features provide credible hints that the Colombian case veils new hypotheses for explaining the growth of judicial power that we need to recover in order to sharpen our understanding of institutional development. The case study is defined as the “intensive study of a single case for the purpose of understanding a larger class of cases (a population)” (Gerring 2009, 95). There­ fore, a properly, and self-consciously, applied case study method is always involved in the broader context of a universe of cases and does not stand on its own analysis. As McKeown put it, social scientists, even those employing “standard quantitative methods, are ‘interactive processors’ [and intuitively] move back and forth between theory and data” (2004, 159). They are folk Bayesians. The case study does exactly that and carefully treads between theory and new data. The first imperative to take away for this inquiry then is to understand the existing literature and theory in order to place my original work in the context of the state of the art. This inquiry there­ fore begins each chapter with a conceptual discussion on the critical juncture (Chapter 2), path dependence (Chapter 3), and the strategic judge (Chapter 4) to properly place the Colombian case within the literature, and eventually explain the contributions of this inquiry. The most important advantage of a case study lies in its capacity to generate new hypotheses that imply causal relationship. A shared observation amongst theorists of social science research is that the “relationships discovered among different elements of a single case have prima facie causal connection” (Gerring 2009, 100; see also McKeown 2004; Ragin 1997). This prima facie causal connection helps to explain the advantages of case studies as well as the context traits more amenable to case study approaches. A case study generates new hypotheses rather than testing them, provides internal validity, discloses causal mechanisms (and not effects), and provides a deep scope to the proposed hypotheses (Gerring 2009, 97). The empirical char­ acteristics of cases mean that case studies are better suited for heterogeneous cases, for cases when strong causal relationships are to be expected (since weak causal relation­ ships are particularly opaque), for cases with rare variation, and limited data availability. The value of the Colombian case then hinges on what makes it special in a universe of cases, and what is the specific goal of this book. Rather than test

26 Introduction hypotheses from comparative politics, this study envisions to understand the con­ stitutional judge in a novel way: as a discursive actor. To my knowledge, there has not been a discursive explanation for why a court’s authority increases. This book looks closely at the interaction in specifically arranged discursive places, which I hypothesize to have had an impact on the outcome. Colombia belongs in the category of a least likely case to experience the growth of judicial independence. Gerring refers to such a case as one that “on all dimensions except the dimension of theoretical interest, is predicted not to achieve a certain outcome, and yet does so, [providing] the strongest sort of evidence possible in a non-experimental, single case setting” (2008, 659). Critical for the effectiveness of this method are the facts of the case and the predictive capacity of the theory at hand. Colombia under Uribe’s charismatic leadership is a highly unlikely case for judicial indepen­ dence not solely due to the background conditions of a long history of violence and clientelism, but also because his governance style created an unusual situation for Colombia itself: he was capable of dominating public policies to the degree that a fractioned Congress united in a strong and disciplined coalition, supporting his leadership. Despite these parameters judicial independence did materialize, and the Constitutional Court’s powers did grow appreciably. The close inspection of the development of the substitution doctrine in Chapter 4 will further show that the outcome was not only surprising in a generic sense, but inexplicable for con­ ventional theory. Strategic accounts of the constitutional judge have the advantage of providing testable, or in the Popperian sense, risky predictions (Gerring 2008, 660). In this chapter we will see that the most common predictions of the strate­ gic account that test the conditions for the creation of judicial independence do not hold in this instance. It is thus a very good case to develop new theories and hypotheses. Chapter 3 begins with the notion that institutions in the historical institution­ alist perspective develop along dependent paths. I hypothesize, however, that this might be different for judicial institutions, since they involve claims to rights and justice, which carry foundational normative innuendos. To test the application, I disaggregate the Colombian case into two sub-cases with differing outcomes. In short, I engage the most similar research design. The most similar research design can do two things: it can be used as an exploratory or confirmatory tool of analysis (Gerring 2008, 668). Here we use it first as a hypothesis testing and then gen­ erating device, moving to thick analysis to propose that the dynamic of judicial institutions must be placed in the discursive context, in which apparently settled questions remain open to contestation, and change can develop from within the institution. Finally, the second, and thereby first chapter after this introduction, investigates the 1991 Constitution in historical perspective. It, too, incorporates the most similar research design and takes three instances of constitutional amendments and replacement that share the basic characteristic in that they fundamentally altered the way political interaction was organized in Colombia. There is of course a legal distinction between constitutional replacement and amendment, in that the former disrupts the validity of the existing charter, while the latter preserves the

Introduction

27

legality of the existing constitution. As Negretto explains, however, “both modify the constitution by introducing textual changes, and these formal alterations are sometimes comparable in scope and importance” (2013, 19). This was indeed the case for the three sub-cases: in each instance, electoral, legislative, executive, and judicial institutions were reformed. In addition, even the amendment processes had plebiscitary components and were submitted to popular confirmation in referenda. Yet, the outcome in the 1991 Constitution differed on most of these parameters. Furthermore, these amendments and replacements warrant selection for closer inspection, because they were the only amendments that fundamentally altered judicial institutions (see next two sections on the specific contributions and set up in each chapter). Thus, the analytical setup of this book provides us with a narrative that moves from historical comparisons through comparisons of the 20-year period following the Constitution’s genesis, to a close inspection of processes within the chambers of the Constitutional Court. As seen, each chapter is internally divvied up to provide analytical leverage and increase the study’s value for comparative politics. The second important methodological caveat, after case selection, is data col­ lection. Data collection for judicial processes and reasoning poses heightened methodological problems, because they are said to be particularly disagreeable to public exposure. The analysis in Chapter 4 will reveal in detail that Colombian constitutional judges deliberately curtailed exposure to the public spotlight. Thus, without active participation in public debates, how can we discern the impact of discourse and ideas on decision-making processes inside the Court? The simple answer to that challenge is to go and ask the participants of these processes to describe them to the researcher. This brings us to the centrality of elite interviews in this book. At least since King et al., elite interviews have somewhat become the bad apple of social scientists’ research. They warn the researcher against letting the interviewee “do our work for us” and establish causal inferences with references to motivation. We ought to take them as statements of facts at best, but certainly not draw causal inferences with only that data (1994, 112n). Of course, King et al. are right to warn against simply believing answers given in interviews—not least by politicians, whose job description entails hiding actual intentions behind a smokescreen of rhetorical devices. However, these warnings gloss over the incredible value that elite interviews can provide when used in conjunction with properly triangulated data. The first contention then is to establish the factual correctness. Most research that engages with elite interviews faces the problem that interviewees might not answer truthfully or in some way manipulate their role in specific decisions or processes. The only advantage researchers who interview politicians have over those who speak with judges is that the former subjects are more likely to be on the public record. There are remedies to procure the accuracy of interview data. The first impor­ tant effort is to secure a safe environment for the subject. Ethics requirements for research on Colombia are exceptionally high due to the violent history of Colombia. I therefore went to great lengths to ensure the privacy and anonymity of interviewees, explaining that the protection of subjects was a contractual

28 Introduction obligation for my research. If you interview subjects who are lawyers by training and are engaged in legal work, presenting contracts that legally secure their anonymity is of course a highly valued assurance. Additionally, being an outsider and researcher rather than a journalist from Colombia provided extra assurance that the interviews were not compromised by short-term interest to create a good story, but were indeed intended to provide long-term insights. Judges in particular felt compelled to help with social scientific research, because, as one interviewee said, this work was important, since “the country had suffered tremendously”.14 The sincerity of sharing the interest was confirmed, not least, by the fact that no interview lasted shorter than an hour and often developed into discussions that lasted over 90 minutes or even 150 minutes. The most important tool for confirming the validity of interview, without a doubt, is the triangulation with external or otherwise independent data. This book builds on two sources of triangulation. The interview data presented in this book has been disaggregated according to first, second, and third order observation. This distinction pays tribute to the fact that subjects face different validity con­ cerns. The first order observers are actors inside the processes under investigation, in this case legal deliberation (i.e. judges), while second order observers are close participants in the process with less personal stakes (i.e. clerks). Thus, the first step of triangulation for me was to compare and contrast first and second order observations in order to establish the validity of reported facts. Third order observations are undertaken by individuals outside of those processes but with detailed knowledge of the subject matter (such as journalists and scholars). Third order observers provide interpretative data rather than factual observations of the processes themselves. The next important, and extremely valuable, source to triangulate the flow of deliberation inside the Court were the Actas compiled by the General Secretary of the Constitutional Court. She is the only other person present during deliberation inside the plenary chamber and reports the minutes of the discussion in a docu­ ment that remains sealed for five years. I could collect the most important Actas for four essential decisions by the Constitutional Court. These include, for exam­ ple, deliberations over why the Court chose to abandon public hearing in favor of being solely provided with written statements by concerned parties. As we will see in Chapter 4, public hearings were cut from the process, since they provided a context that could benefit rhetorical action over actual reasoning. The reasoned decisions themselves, too, are a critical source for triangulating the processes under investigation. The importance of discourse shows, not least, in the ability to change opinions and perform changes in position. These do in fact appear on the public record, even though judges do not engage in public debate. Colombia’s Constitutional Court decisions are published and include a lengthy and detailed discussion on the various arguments put forward by the different actors involved in a decision. For example, the first re-election decision was over 800 pages in length, and the second was over 500 pages in length. Moreover, constitutionality verdicts include dissents by the individual judge on a specific merit of the case (not solely the entire decision). The decision to allow a second

Introduction

29

re-election included a dissent by magistrate Humberto Sierra Porto. He did not disagree on the constitutionality of the law but explained his disagreement with the substitution doctrine that was applied in that decision. (The Court applied the doctrine but found that the reform did not violate inherent principles of the Constitution.) In the choque de trenes, too, Eduardo Cifuentes published an extensive dissent in 1999 that outlined his disagreement with the Court’s tutela decision, explaining that it constituted a decisionistic act. Thus, we see discursive dynamics appear on the public record in the reasoned decisions published. The most important evidence for the discursive nature of decision-making inside the Court is the ability to change opinions when the force of the better argument compels individual judges to do so. As I will show, there were indeed critical instances in which individual judges voted against their previous stances on particular doctrines (we will see this occurring in regard to the substitution doc­ trine) as well as apparent ideological preferences on generic social issues. Magis­ trate Nilson Pinilla voted against the substitution doctrine in the 2010 decision, but two years later penned a decision that argues with the substitution doctrine. Furthermore, he is known as a small-c conservative with stringent social values but affirmed gay rights in constitutionality decisions. All of this, I contend, provides credible evidence to show that reasoning and, above all, the process of reasoning is central for judicial independence and a properly institutionalized separation of powers.

Structure of the Analysis The structure of this inquiry follows a logic that runs from macro to micro processes. The next chapter places the 1991 Constitution in historical context and develops a comparison with two other instances of profound institutional engineering. It commences with a conceptual discussion of Collier and Collier’s framework of the critical juncture, and then plugs in a comparative baseline through a brief discussion on the origin of Colombia’s weak state institutions and inequitable social relations. The chapter then moves to a structured comparison of the Reyes reforms, the National Front agreement, and the 1991 constituent pro­ cess. In the end, it argues that the 1991 Constitution can indeed be understood as a critical juncture that shows characteristics of an instance in Colombia’s history that shuffles actors and institutions fundamentally anew. Chapter 3 inspects the post-genesis institutional development and investigates if it followed a path-dependent logic. The chapter is divided into four parts. The first will outline the conceptual stipulations of path dependence and its application to the Colombian post-1991 context. In addition, it focuses on the normative con­ notations of the tutela. From this follows longitudinal division of sub-cases that explore the jurisprudential evolution of the tutela contra sentencias (legal junctures against legal decisions) in the course of the proceso 8000 and parapolítica scandals. The analysis devotes one part to each episode, and contrasts the political and normative evolutions. The final part concludes the findings and ties the analyses of each political scandal together to develop a conclusion on the suitability of path

30 Introduction dependence to describe the institutional behavior when it involves normative values. This chapter will show that the rather static vocabulary of path dependence has difficulties to account for the openness of legal decisions, which depends on the justification and argumentation as well as the specific legal context. Chapter 4 will explicitly deal with the re-election reforms and embed the analysis in the tensions between majoritarian and constitutionalist ideals of democracy, and contrasting understandings of the constitutional judge as a strategic or discursive actor. It will first outline the contours of each tension, and then move to contrasting both decisions in a reconstruction of the evolution of the substitution doctrine. This chapter will counter the understanding of the constitutional judge as a forwardlooking utility-maximizing actor. It insists on the deliberative capacities of magis­ trates who act in carefully designed discursive spaces that aim to protect deliberative patterns. The final chapter, Chapter 5, will conclude the findings and tie them together to accentuate the contributions for the study of institutionalism(s).

Notes 1 The term itself invokes the heritage of the 1948 German Basic Law and the Spanish Constitution from 1978, which stood as models to the constituent assembly that delib­ erated on the Constitution. In general terms, the social state of law combines the liberal state, and its focus on individualism and diversity as requirements for democracy, with the materialization of human rights as benchmarks of justice (Eslava 2009). Borrowing, and further developing, the concept of a social state that is based on the rule of law, Colom­ bia’s new Constitution embraced five defining traits of Latin America’s turn towards neo­ constitutionalism: 1) a Kelsenian system of constitutional review; 2) an expansive rights catalogue including private, social, and collective rights; 3) introducing international law—specifically human rights treaties—as constitutionally binding; 4) incorporating new and independent ombudsman institutions as well as specific mechanisms (the writ of injunction or tutela) to enforce human rights; and 5) addressing governmental corruption through better functioning judicial institutions (Van Cott 2000). 2 The denotation of Constitutional Court decisions follows a specific pattern: the letter spe­ cifies the type of case (C = accion de constitucionalidad decided by the sala plena; T = tutela by a revision chamber; SU = unification decision of tutelas by the sala plena; A = auto), the middle number its chronological position within the year, and the final number after the dash the year of the decision. M.P., omitted in the intext citation but noted in the biblio­ graphy for a complete reference, stands for magistrado ponente, the judge writing the initial study and decision (see also Cepeda Espinosa and Landau 2017, 9–14). 3 Interview with Senator of Republic of Colombia, 12 March 2013. 4 Congress of Colombia, Diario Oficial, 2004, Acto Legislativo 02 de 2004. 27 December 2004. http://www.alcaldiabogota.gov.co/sisjur/normas/Norma1.jsp?i=15519. 5 The case was the first hard case test for the theory of competence that the Court established in its 2003 decision (C-551/03) on Uribe’s first political reform of the Constitution (see Chapter 4). For the first time, a member of the Court dissented on the parameters of the substitution doctrine, but the doctrine was nevertheless applied and utilized to strike down parts of the law. 6 Semana, “Estas son las demandas”, 14 February 2005. http://www.semana.com/porta da/articulo/estas-demandas/70872-3. 7 Semana, “Declaración de Uribe respecto al fallo de la Corte”, 25 February 2010. http://www.semana.com/politica/multimedia/declaracion-uribe-respecto-fallo-corte/ 143458-3.

Introduction

31

8 Interview with Constitutional Court judge, 20 November 2012. 9 Already the 1853 reforms to the 1851 Constitution introduced universal male suffrage, which was then limited to literate men above 21 years of age in the 1886 Constitution. This war repealed in the 1936 reforms, and women gained suffrage in the National Front reforms of the 1950s (Fergusson and Vargas 2013). 10 These are: 1) “war making” to eliminate foreign rivals; 2) “state making” to eliminate internal rivals; 3) “protection” to eliminate the rivals of their clients; and 4) “extrac­ tion” to acquire the resources to fulfill the previous tasks. All of these are contingent on exercising the monopoly of violence (Tilly 1985). 11 As Appendix C makes readily apparent, the security situation was indeed dire: intentional homicides had almost reached all-time highs again, and the number of kidnappings peaked at unprecedented heights. 12 Kline, in his optimistically titled book Showing Teeth to the Dragons, rejected the hypothesis that Colombia during the Uribe administration showed worrying signs of a delegative democracy. According to him, Uribe only successfully defined the greater good of the nation with his hard line against the guerrilla (otherwise he was constrained by Congress and the judiciary); he fulfilled his campaign promise of ending the insur­ gency; he had the support of a party and had no intention to form his own movement; and even though he considered Congress and the judiciary a nuisance, he did follow the courts’ verdicts (2009, 182). Kline’s verdict might have been a little too optimistic, given that he developed the book two years prior to the end of Uribe’s term. His second term in office was mired in scandal, which together paint the picture of a President who thought himself to be equipped with delegative powers. 13 Uprimny, Rodrigo, “¿Estado de Opinión o de Derecho?” El Tiempo, 17 August 2009. http://www.elespectador.com/columna156600-estado-de-opinion-o-de-derecho. 14 Interview with a Constitutional Court judge, 8 April 2013.

References Ackerman, Bruce. 1991. We the People, Volume 1: Foundations. Cambridge, MA: The Bel­ knap Press of Harvard University Press. Alexy, Robert. 1989. A Theory of Legal Argumentation. The Theory of Rational Discourse as Theory of Legal Justification. Oxford: Oxford University Press. Archer, Ronald P. 1989. “The Transition from Traditional to Broker Clientelism in Colombia: Political Stability and Social Unrest”. Paper Presented at the Annual Meeting of the Latin American Association in Miami, Florida. Archer, Ronald P., and Matthew Soberg Shugart. 1997. “The Unrealized Potential of Pre­ sidential Dominance in Colombia”. In Presidentialism and Democracy in Latin America, edited by Scott Mainwaring and Matthew Soberg Shugart, 110–159. Cambridge: Cam­ bridge University Press. https://doi.org/10.1017/CBO9781139174800.004. Bakiner, Onur. 2020. “Endogenous Sources of Judicial Power: Parapolitics and the Supreme Court of Colombia”. Comparative Politics 52 (4): 603–624. https://doi.org/ 10.5129/001041520X15743890683814. Bejarano, Ana María, and Eduardo Pizarro. 2005. “From ‘Restricted’ to ‘Besieged’: The Changing Nature of the Limits to Democracy in Colombia”. In The Third Wave of Democratization in Latin America, edited by Frances Hagopian and Scott P. Mainwar­ ing, 235–260. Cambridge: Cambridge University Press. https://doi.org/10.1017/ CBO9780511791116.009. Bernal, Carlos. 2013. “Unconstitutional Constitutional Amendments in the Case Study of Colombia: An Analysis of the Justification and Meaning of the Constitutional

32 Introduction Replacement Doctrine”. International Journal of Constitutional Law 11 (2): 339–357. https://doi.org/10.1093/icon/mot007. Botero, Felipe, Gary Hoskin, and Mónica Pachon. 2010. “Flash Report: A Key Decision in Colombia”. Vancouver. Cameron, Maxwell A., Ana-Maria Blanaru, and Leslie Burns. 2006. “The Separation of Powers Reconsidered: Presidentialism and the Rule of Law”. In Annual Meeting of the American Political Science Association, Marriott, Loews Philadelphia, and the Pennsyl­ vania Convention Center, Philadelphia. 31 August 2006. Cepeda Espinosa, Manuel José, and David Landau. 2017. Colombian Constitutional Law: Leading Cases. Oxford: Oxford University Press. Collier, David, and Ruth Collier. 2002. Shaping the Political Arena. Critical Junctures, the Labor Movement, and Regime Dynamics in Latin America. Notre Dame: University of Notre Dame Press. Collier, David, and Gerardo L. Munck. 2017. “Building Blocks and Methodological Challenges: A Framework for Studying Critical Junctures”. Qualitative and MultiMethod Research 15 (1): 2–9. Dix, Robert H. 1967. Colombia: The Political Dimensions of Change. New Haven: Yale University Press. Dix, Robert H. 1980. “Political Oppositions under the National Front”. In Politics of Compromise: Coalition Government in Colombia, edited by Albert Berry, Ronald G. Hellmann, and Mauricio Solaún, 131–180. New Brunswick: Transaction Books. Dixon, Rosalind, and David Landau. 2015. “Transnational Constitutionalism and a Lim­ ited Doctrine of Unconstitutional Constitutional Amendment”. International Journal of Constitutional Law 13 (3): 606–638. https://doi.org/10.1093/icon/mov039. Eaton, Kent. 2006. “The Downside of Decentralization: Armed Clientelism in Colombia”. Security Studies 15 (4): 533–562. https://doi.org/10.1080/09636410601188463. Eisenstadt, Todd A., A. Carl Levan, and Tofigh Maboudi. 2015. “When Talk Trumps Text: The Democratizing Effects of Deliberation during Constitution-Making, 1974–2011”. American Political Science Review 109 (3): 592–612. https://doi.org/10.1017/ S0003055415000222. Epstein, Lee, and Jack Knight. 2013. “Reconsidering Judicial Preferences”. Annual Review of Political Science 16 (1): 11–31. https://doi.org/10.1146/annurev-polisci-032211-214229. Eslava, Luis. 2009. “Constitutionalization of Rights in Colombia: Establishing a Ground for Meaningful Comparisons”. Revista Derecho Del Estado 22: 183–229. Fergusson, Leopoldo, and Juan F. Vargas. 2013. “Don’t Make War, Make Elections— Franchise Extension and Violence in XIXth-Century Colombia”. SSRN Electronic Jour­ nal. https://doi.org/10.2139/ssrn.2232070. Finkel, Jodi. 2008. Judicial Reform as Political Insurance: Argentina, Peru, and Mexico in the 1990s. Notre Dame: University of Notre Dame Press. Francisco, Leal Buitrago. 1984. Estado y Política En Colombia. Bogotá: Siglo XXI. Gerring, John. 2008. “Case Selection for Case-Study Analysis: Qualitative and Quantitative Techniques”. In The Oxford Handbook of Political Methodology, edited by Janet M. BoxSteffensmeier, Henry E.Brady, and David Collier, 645–684. Oxford: Oxford University Press. https://doi.org/10.1093/oxfordhb/9780199286546.003.0028. Gerring, John. 2009. “The Case Study: What It Is and What It Does”. In The Oxford Hand­ book of Comparative Politics, edited by Carles Boix and Susan C. Stokes, 90–122. Oxford: Oxford University Press. https://doi.org/10.1093/oxfordhb/9780199566020.003. 0004.

Introduction

33

González, Lucas. 2014. “Unpacking Delegative Democracy: Digging into the Empirical Content of a Rich Theoretical Concept”. In Reflections on Uneven Democracies. The Legacy of Guillermo O’Donnell, edited by Daniel Brinks, Marcelo Leiras, and Scott L. Mainwaring, 240–269. Baltimore: The Johns Hopkins University Press. Habermas, Jürgen. 1996. Between Facts and Norms: Contributions to a Discourse Theory of Law and Democracy. Cambridge, MA: MIT University Press. Hall, Peter A., and C.R. Taylor. 1996. “Political Science and the Three New Institutionalisms”. Political Studies 44: 936–957. https://doi.org/10.1111%2Fj.1467-9248.1996. tb00343.x. Hartlyn, Jonathan. 1988. The Politics of Coalition Rule in Colombia. Cambridge: Cam­ bridge University Press. Helmke, Gretchen. 2004. Courts under Constraints. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9780511510144. Helmke, Gretchen, and Julio Ríos-Figueroa (eds). 2011. Courts in Latin America. Cam­ bridge: Cambridge University Press. https://doi.org/10.1017/CBO9780511976520. Hübner Mendes, Conrado. 2013. Constitutional Courts and Deliberative Democracy. Oxford: Oxford University Press. King, Gary, Robert O. Keohane, and Sidney Verba. 1994. Designing Social Inquiry: Scien­ tific Inference in Qualitative Research. Princeton: Princeton University Press. https:// doi.org/10.2307/2076556. Kline, Harvey. 2009. Showing Teeth to the Dragons. State-Building by Colombian President Álvaro Uribe Vélez. 2002–2006. Tuscaloosa: University of Alabama Press. Kneip, Sascha. 2011. “Constitutional Courts as Democratic Actors and Promoters of the Rule of Law: Institutional Prerequisites and Normative Foundations”. Zeitschrift Für Vergleichende Politikwissenschaft 5 (1): 131–155. https://doi.org/10.1007/ s12286-011-0096-z. Lijphart, Arend. 1994. “Presidentialism and Majoritarian Democracy. Theoretical Obser­ vations”. In The Failure Presidential Democracy, edited by Juan J. Linz and Arturo Valenzuela, 91–105. Baltimore: The Johns Hopkins University Press. Linz, Juan J., and Arturo Valenzuela. 1994. The Failure of Presidential Democracy. Balti­ more: The Johns Hopkins University Press. Mahoney, James. 2000. “Path Dependence in Historical Sociology”. Theory and Society 29 (4): 507–548. https://doi.org/10.1023/A:1007113830879. Mahoney, James, and Kathleen Thelen (eds). 2009. Explaining Institutional Change: Ambiguity, Agency, and Power. Cambridge: Cambridge University Press. https://doi. org/10.1017/CBO9780511806414. Mainwaring, Scott P. 1991. “Clientelism, Patrimonialism, and Economic Crisis: Brazil since 1979”. In Congress of the Latin American Studies Association. Washington, D.C., 4–7 April 1991. Mainwaring, Scott, and Timothy R. Scully. 1995. Building Democratic Institutions: Party Systems in Latin America. Stanford: Stanford University Press. March, James G., and Johan P. Olsen. 2008. “Elaborating the ‘New Institutionalism’”. In The Oxford Handbook of Political Institutions, edited by R.A.W. Rhodes, Sarah A. Binder, and Bert A. Rockmann. Oxford: Oxford University Press. https://doi.org/10. 1093/oxfordhb/9780199548460.003.0001. Martz, John D. 1997. The Politics of Clientelism. New Brunswick: Transaction Publishers. McKeown, Timothy J. 2004. “Case Studies and the Limits of the Quantitative Worldview”. International Organization 53 (1): 139–169.

34 Introduction McLean, Phillip. 2002. “Colombia: Failed, Failing, or Just Weak?” The Washington Quarterly 25 (3): 123–134. https://doi.org/10.1162/01636600260046280. Negretto, Gabriel L. 2013. Making Constitutions. Making Constitutions: Presidents, Parties, and Institutional Choice in Latin America. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9781139207836. North, Douglass C. 1990. Institutions, Institutional Change and Economic Performance. Cambridge: Cambridge University Press. North, Douglass C., and Barry R. Weingast. 1989. “Constitutions and Commitment: The Evolution of Institutions Governing Public Choice in Seventeenth-Century England”. The Journal of Economic History 49 (4): 803–832. https://doi.org/10.1017/ S0022050700009451. Nunes, Rodrigo. 2010. “Ideal Justice in Latin America: Interests, Ideas, and the Origins of Progressive Judicial Activism in Brazil and Colombia”. University of Texas, Austin. O’Donnell, Guillermo A. 1973. Modernization and Bureaucratic-Authoritarianism. Studies in South American Politics. Berkeley: Institute of International Studies University of California, Berkeley. O’Donnell, Guillermo A. 1994. “Delegative Democracy”. Journal of Democracy 5 (1): 55–69. Page, Scott E. 2006. “Path Dependence”. Quarterly Journal of Political Science 1 (1): 87–115. https://doi.org/10.1561/100.00000006. Pierson, Paul. 2000. “Increasing Returns, Path Dependence, and the Study of Politics”. American Political Science Review 94 (2): 251–267. https://doi.org/10.2307/2586011. Powell, John Duncan. 1970. “Peasant Society and Clientelist Politics”. American Political Science Review 64 (2): 411–425. https://doi.org/10.2307/1953841. Pou Giménez, Francisca. 2019. “Supreme and Constitutional Courts: Directions in Con­ stitutional Justice”. In Routledge Handbook of Law and Society in Latin America, edited by Rachel Sieder, Karina Ansolabehere, and Tatiana Alfonso Sierra, 187–204. New York: Routledge. Ragin, Charles. 1997. “Turning the Tables: How Case-Oriented Research Challenges Variable-Oriented Research”. Comparative Social Research 16: 27–42. Ríos-Figueroa, Julio. 2011. “Institutions for Constitutional Justice in Latin America”. In Courts in Latin America, edited by Gretchen Helmke and Julio Rios-Figueroa, 27–54. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9780511976520.002. Risse, Thomas. 2002. “Law and Politics beyond the Nation State: Areas of Conversation and of Common Ground”. In Conference Law and Politics, Max Planck Project Group ‘Common Goods: Law, Politics, and Economics’, Bonn, May 24–25. Rodríguez-Raga, Juan Carlos. 2011. “Strategic Deference in the Colombian Constitutional Court, 1992–2006”. In Courts in Latin America, edited by Gretchen Helmke and Julio Rios-Figueroa, 81–98. Cambridge: Cambridge University Press. https://doi.org/10. 1017/CBO9780511976520.004. Sanders, Elizabeth. 2008. “Historical Institutionalism”. In The Oxford Handbook of Political Institutions, edited by R.A.W. Rhodes, Sarah A. Binder, and Bert A. Rockmann. Oxford: Oxford University Press. https://doi.org/10.1093/oxfordhb/9780199548460.003.0003. Schmidt, Vivien A. 2008. “Discursive Institutionalism: The Explanatory Power of Ideas and Discourse”. Annual Review of Political Science 11 (1): 303–326. https://doi.org/10. 1146/annurev.polisci.11.060606.135342. Tilly, Charles. 1985. “Bringing the State Back In”. In Bringing the State Back In, edited by Peter B. Evans, Dietrich Rueschemeyer, and Theda Skocpol. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9780511628283.

Introduction

35

Van Cott, Donna Lee. 2000. “A Political Analysis of Legal Pluralism in Bolivia and Colombia”. Journal of Latin American Studies 32 (1): 207–234. https://doi.org/10. 1017/S0022216X99005519. Warren, Mark E. 2017. “A Problem-Based Approach to Democratic Theory”. American Political Science Review 111 (1): 39–53. https://doi.org/10.1017/S0003055416000605.

Court Cases Republic of Colombia. Constitutional Court of Colombia. C-551/03. M.P. Eduardo Montealegre Lynett. Republic of Colombia. Constitutional Court of Colombia. C-1040/05. M.P. Manuel Jose Cepeda Espinosa, Rodrigo Escobar Gil, Marco Gerardo Monroy Cabra, Humberto Antonio Sierra Porto, Álvaro Tafur Galvis, Clara Inés Vargas Hernández. Republic of Colombia. Constitutional Court of Colombia. C-141/10. M.P. Humberto Antonio Sierra Porto.

2

The Novelty of the 1991 Constitution A Critical Juncture in Colombia’s History

Introduction Social scientists in general, and political scientists in particular, have a fascination with change: what causes change, what initiates it, and what arrests change? In fact, it has become the central focus of the institutionalist schools, which view the absence of change as a function of the role institutions play in basic interactions. They structure expectations, determine the rules of the game, and exclude certain options from patterns of behavior. In short, institutions’ defining feature is their persistence over time. In addition, writing and reforming constitutions—arguably one of the most profound modes of changing and rearranging political institu­ tions—has become a particularly pertinent focus of constitutionalist and comparative politics scholarship alike (Beasley-Murray et al. 2009). It appears that 80 years after the debate between Carl Schmitt and Hans Kelsen, the contentious issue between constitutionalism and democracy, constituent and constituted power, has once again become the center of academic disputes and debates, ranging from legal studies to political science. The specific questions guiding this chapter are: what made the 1991 Constitution (and the Constitutional Court created with that Constitution) peculiar in Colombia’s history, and were the processes around the implementation of the 1991 Constitution reflective of a critical juncture? Articulating the question in the critical juncture framework provides clear methodological tools: critical junctures are those instances in political history that place the patterns of political life on a distinctly new and consequential path (Collier and Collier 2002; Mahoney 2000). The Colliers’ articulation of the cri­ tical juncture framework centers on three components: preceding (base-line) con­ ditions resulting in a crisis, the juncture of institutional engineering, and the legacies left by the decisions taken during the juncture to overcome the crisis (2002). Each one of those components involves clear methodological imperatives that I will outline in the first, conceptual, section of this chapter. Usually, studies of critical junctures look at several different cases. I will engage the most similar research design and look at various instances of institutional engineering to place the 1991 Constitution in the context of Colombia’s own political history. I will define the base-line conditions with an exploration of the historical weakness of Colombian state institutions and then contrast the Rafael Reyes reforms (1905– DOI: 10.4324/9781003229285-2

The Novelty of the 1991 Constitution

37

1910) and the National Front reforms (1953–1957) with the genesis of the 1991 Constitution. These processes have in common that they were the political response to profound regime crises that had existentially destabilized the country, plunging Colombia into violent conflict. Despite the similarities in context, these junctures evolved very differently, suggesting different processes at play. I will first detail the foundations of Colombia’s weak institutions in the postindependent period of the 19th century arguing that Colombia’s weak state and the creation of patronage systems were intrinsically linked with each other through the traditional two-party system. Looking at the initial crises and preceding conditions, I will then show that, while the two earlier experiences of institution-building were responses to intra-elite disagreements and the incapacity of Colombia’s party system to incorporate the mobilized working class, the 1990 constituent process has its origin in a discursive cleavage: political reality in the 1980s made evidently clear that the democratic regime could not live up to the normative promises of a liberal democracy. The drug economy had created a human rights crisis of unprecedented proportions, with thousands of deaths. At the same time, political elites, sectors of the army, and landowners morphed into a heterogeneous coalition that was com­ plicit in the systematic annihilation of an entire political party (the Unión Patriotica, UP). This laid open the flawed, restrictive, and often exclusionary, political frame­ work inherited from the (officially terminated) National Front that had made meaningful political participation a virtual impossibility. There was thus a cleavage between the normative claim of a democracy and political reality. The junctures themselves, too, followed very different trajectories, suggesting the presence of differing mechanisms of production and reproduction. The 1905–1910 and National Front reforms in the 1950s were instigated by party elites, facilitated with the help of the armed forces, and negotiated behind closed doors to only be ratified post-factum in popular referenda. The 1991 Constitution has its roots in a genuine, cross-sectional student movement, which had wide public support, and was deliberated in public debates with important citizen input. It also included previously marginalized groups from demobilized guerrilla groups, indigenous groups, and independent civil society groups. Thus, in contrast to 1905–1910 and the National Front in 1953–1957, the 1991 Constitution was created in a communicative process genuinely placing collective human agency at the nexus between word and deed (Arendt 1958). In the end, these junctures also produced different legacies. After 1905–1910 and 1953–1957, violence diminished, economic development ensued, but patronage systems reconstituted themselves within the two-party system. In 1991, violence continued, and in some instances even expanded, creating new actors, while patronage systems reconstituted themselves outside the traditional party system, slowly but surely eroding said party system. Furthermore, in 1991, the constitutional text went the furthest in overhauling the judicial branch, not only stipulating an expansive human rights catalogue, but also incorporating new organizations (the Constitutional Court and the Superior Council of the Judiciary) and mechanisms to enforce human rights claims (the tutela). The choices made by constituents showed in the genesis of a new judicial discourse accompanying the

38 The Novelty of the 1991 Constitution creation and implementation of the new Constitution. Nuevo derecho (New Law/ Right) fully embraced human rights principles and the importance of neo-con­ stitutionalist public reasoning in the exercise of constitutional interpretation. Viejo derecho, Old Law, associated with the old Constitution from 1886, had held the legality principle supreme which stipulated a formalistic interpretation of constitutional norms. In sum, I put forward the argument that the 1991 Constitution, in contrast to the Rafael Reyes reforms (1905–1910) and the National Front reforms (1953–1957), is traceable to a profound normative and discursive cleavage in the preceding condi­ tions, followed a novel path of citizen mobilization that sidelined traditional patron­ age networks during the assembly, and created a rights-based institutional framework. That was not a coincidence: its creation had a distinctly communicative origin and the constraints placed upon constituents also differed from the first two cases. It therefore is perfectly sensible to infer that the 1991 Constitution amounted to a critical juncture.1

Critical Junctures and the Study of Colombia’s History The critical juncture framework harbors the danger of being used as a fancy way of saying “important point in history”. The point here, however, is to identify a juncture that sets itself apart even from other important points in history. The Colliers, in their seminal study on critical junctures in Latin America, are quite poetic in that regard. They begin with an analogy of the bifurcating path in Robert Frost’s “The Road Not Taken” to vie that their endeavor is to detect more than solely important episodes in the region’s history, but ones that have had an identifiable legacy; where taking the one road over the other “made all the differ­ ence” (2002, 27). We must be able to identify these critical instances of institu­ tional transformation. They are “periods of significant change, which typically occur in distinct ways in different countries or in other units of analysis” and result in long(er) lasting, distinctive legacies (Collier and Collier 2002, 29). This defini­ tion also makes readily apparent that critical junctures do not appear out of the blue, nor are their consequences random. On the contrary, critical junctures are observable by three crucial components: their preceding conditions resulting in a moment of crisis, the institutional engineering during the juncture, and the lega­ cies left by the decisions taken during the critical juncture to overcome the crisis. Each of these components has clear methodological and conceptual caveats, and therefore require a brief grounding to set the path for the historical comparison of Colombia’s instances of institutional engineering. One pertinent problem of the critical juncture framework, one that receives repeated criticism from scholars, is its supposed incapability of making intelligible predictions (March and Olsen 2008; Sanders 2008; Hay 2008). There is indeed a problem with the predictability of critical junctures, but that is rooted in human agency itself. In fact, determinism may not even be desirable. The criticism that critical junctures are unintelligible, however, misses the mark. The most important works do indeed propose explanations and conceptualizations for what is behind profound moments of change (Lipset and Rokkan 1967; Collier and Collier

The Novelty of the 1991 Constitution

39

2002). In these and other works analyzing historical processes, moments of pro­ found social change—from revolutionary change through cyclical interactions to political incorporation—were preceded by social and cultural cleavages. Such clea­ vages create tensions and crises that cannot be contained with existing political mechanisms of conflict resolution, and require a reconfiguration of the rules gov­ erning socio-political interaction. Thereby, they also create the slate of options that is available to actors in critical junctures. Collier and Munck attest that scholars view “the uncertainty of outcomes and substantial degrees of freedom in actor choices as a defining feature of critical junctures” (2017, 4). We need to remember that part of the heritage of the critical juncture framework was born from cul-de-sacs in neo­ classical economics, which has had difficulties accounting for drastic turns in devel­ opment that result in inefficient outcomes. As Mahoney eloquently states, “without the assumption of initial contingency, path-dependent processes cannot be linked to ‘unpredictability’ and ‘inefficiency’” (2000, 515). Crucially, the potential ineffi­ ciencies of the outcome do not signal the absence of rationality, but rather signal the presence of a political rationality that is contingent on the interests and resources of the actors involved in the process that we can then identify as a critical juncture (Negretto 2013). The range of options, uncertainty of outcomes, and the agency of actors are not equivalent to randomness. As Mahoney asserts, “to argue that an event is contingent is not the same thing as arguing that the event is truly random and without antecedent causes” (Mahoney 2000, 513). On the contrary, “knowl­ edge of antecedent conditions is essential for explaining the distinct ways the critical juncture occurs across cases” (Collier and Munck 2017, 5). It is for that reason that a central task for a researcher engaging the critical juncture framework is to analyze antecedent conditions. They are “an important source of rival hypotheses for explaining the outcomes attributed to the critical junc­ ture” (Collier and Munck 2017, 5). This is particularly true for a within-case study that looks for internal variation. Only with a clear perspective on pre­ ceding conditions can an intelligible variation of potential outcomes be mapped onto a diverse number of within-cases. Finally, critical junctures depend on identifying “a major episode of institutional innovation [that] generates an enduring legacy. In short: no legacy, no critical junc­ ture” (Collier and Munck 2017, 6). The issue of legacy and enduring features of a critical juncture is a rather contentious debate in the institutionalist literature. Collier and Munch contend that “critical juncture scholars focus on factors that yield a selfreinforcing outcome over a longer time horizon” (6). However, Mahoney and Thelen (2009) have outlined a paradigm of incremental change in the process of institutional implementation that deviates from the “path dependence” and “sensitive dependence on initial conditions” cemented in the legacies of critical junctures. Related, the centrality of an “enduring legacy [as] a defining characteristic of a critical juncture” begs the question how much hindsight is in fact required to ascertain that the legacy has endured (Collier and Munck 2017, 7). The question of how much institutional change occurs in post-genesis contexts is the theme of further explora­ tions in the next chapter. Here, I focus on relatively salient changes in political

40 The Novelty of the 1991 Constitution practices occurring after the relevant junctures to identify a case that diverges from the constant causes identified through the analysis of preceding conditions. Looking at critical junctures and analyzing their appearance is therefore not an exercise akin to identifying important moments post-hoc with little intelligible data to recognize the causes behind the change. Rather, it requires the researcher to identify cleavages that carry transformative potential. These cleavages become intelligible if contrasted with a “base-line” of conditions that form the analytical foundation of antecedent conditions. If additionally compounded by external shock, these cleavages result in a moment of crisis that triggers the critical juncture (Collier and Collier 2002, 30). In all of this, it is indispensable to remember that agency is not lost in the moment of profound decision taking. A juncture, as the crossroad in “The Road not Taken”, provides options, from which the agent, however she will be defined and in whatever way she is constrained, can choose one. The juncture becomes critical, if the eventual choice and its consequence could not be predicted by existing patterns of behavior, but reflects the presence of new actors, interacting under specific con­ straints, giving rise to new institutional frameworks that structure (more predictably) future interactions (Mahoney 2000, 514). Collier and Collier (2002) tie the loose ends together into a concise package of characteristics of a critical juncture. It consists of: 1) antecedent conditions representing a baseline against which the critical juncture and the legacy are assessed; 2) the cleavage or crisis emerging out of the antecedent conditions; and 3) rival explanations involving constant causes (30–31). These are the three guiding analytical elements that provide a plan for the next sections of this chapter: identify a base-line against which to explore cleavages potent enough to result in a transformative crisis, discover the appearance of a critical juncture, and, finally, analyze the resources and behavior of actors involved in the juncture that result in choices structuring future interaction, which then increasingly reproduces itself. The research design of this chapter poses some difficulties for the analysis of critical junctures. As alluded to above, critical junctures are said to appear in more than one unit of analysis, which usually implies that the researcher investigates more than one case. I focus on one case: Colombia. I therefore have to increase the observational data points by turning inward and longitudinally analyze instances in Colombia’s history to help me establish a base-line and identify the variations in outcomes. This is why I put the 1991 Constitution in historical context, which enables us to utilize a most similar research design to discover different causal mechanisms at different per­ iods in time (Skocpol and Somers 1980; Gerring 2008). Moreover, the critical junc­ ture design requires cross-sectional data points, which this study will contribute by observing the effects in different subunits—namely, the legislature, executive and bureaucracy, and judiciary. First, I will have to establish a base-line that explains the fundamental structure of political interaction in Colombia. This trajectory reaches back to the 19th cen­ tury and the post-independence era. Historical monographs of Colombia’s history show that 19th century post-independence evolution was marked by two co-rela­ ted developments: the creation of a weak state infrastructure and the evolution of

The Novelty of the 1991 Constitution

41

patronage systems through the two-party-system. From there I can move to ela­ borating potential junctures with the capacity to switch Colombia’s history onto new paths, resulting in novel structures of political interaction. From my inter­ views, as well as secondary literature, I have identified three cases of institution building in Colombia (1905–1910; 1953–1957; and 1991) that share a basic, ideal-type, trajectory (Figure 2.1).

Political Crisis and Violence

Institutional Engineering Process

New Institutional Framework

Figure 2.1 Ideal-type trajectory of sub-cases of institution building in Colombia (1905– 1910; 1953–1957; and 1991)

According to the critical juncture framework, the question becomes whether these periods of institutional change, which provide identifiable time horizons, were preceded by social, cultural, or other cleavages. If we can recognize cleavages, we must also be able to see an aggravating crisis that transforms the cleavage into an institutional crisis, from where we can move to the most interesting part: the action during the juncture. It is here where decisions are taken that have transformative potential. Consequently, the analysis will look closely at the institutional context in the original political crisis, the actors involved in the institutional engineering process, the constraints they faced during the juncture, as well as the resources at their disposal, the arenas and procedures to negotiate the institutional restructuring, and, finally, present the results of each process in terms of the institutional setup as well as the immediate socio-political development. This comparison will highlight a fundamental difference in outcome: judicial institutions after 1991 were most fundamentally transformed. Without getting too far ahead of myself, this outcome is related to an identifiable cleavage that existed prior to the 1991 constituent process. This type of cleavage, however, was differ­ ent from the ones theorized by Lipset and Rokkan (1967) as well as Collier and Collier (2002). As will be shown in more detail further below, the 1980s was a decade when violence associated with the drug wars made it painfully obvious that Colombia’s political system failed to live up to the promises of a liberal democracy that it claimed to espouse. Death squads repeatedly assassinated several pre­ sidential candidates as well as activists from the political party, the UP. It was a discursive cleavage between the normative claim of a democracy and the political reality of a very violent country. This cleavage left a mark on the process that restructured the political arena in the constituent process, since a heterogeneous student movement pegged their demand for an overhaul of the constitutional regime to Colombia’s abhorrent human rights situation and the inability of the state to provide even miniscule channels of meaningful political participation. In

42 The Novelty of the 1991 Constitution addition, the student movement’s agency was instrumental in the course of the negotiations as well, since students’ framing of the issue centrally affected the outcome, namely the 1991 constituent assembly, as well as important features of the document it produced. It is my contention that the 1991 constituent process “cannot be explained on the basis of prior historical conditions”, but must take into account the occurrences during the juncture itself (Mahoney 2000, 507). In other words, it is my assessment that this does indeed constitute a critical juncture in Colombia’s political history.

Identifying a Base-Line: Violence, Parties, and the Structuration of a Fragile State The critical juncture framework disciplines the researcher to understand the variation between continuity and discontinuity. Collier and Collier (2002) adamantly iterate that the identification of any legacy is essentially contingent on an explicit comparison “with the antecedent system” (34). It is what forms the base-line of their conceptual framework. When we turn to Colombian history, the most apparent constant condition of the Colombian state in the 19th century was its virtual absence. If the state is the human organization that monopolizes the means of violence (Weber 1946), Colombia’s central governments repeatedly and consistently fell short of that benchmark. Antonio Nariño himself wrote: Three years have passed, but, nevertheless, none of the provinces have a treasury, armed forces, canons, powder, schools, streets, nor banks; only a few have a number of considered functionaries that consume what little rents have been left, and they defend with all force the new system that benefits them. (Nariño, cited in Valencia Villa 2012, 79)2 The central governments of Gran Colombia, New Granada, and then the Republic of Colombia, not only failed to fully control the entire territory that was nominally under their control, but also lagged behind other South American territories at the time in terms of their ability to generate taxes and provide public goods. The base­ line explanandum, however, not only consists of the historical weakness of the Colombian state, but also includes relatively democratic institutions (Colombia’s two-party system) combined with a legalistic discourse. I build on classical expla­ nations for historic state weakness that focus on war making capabilities (Tilly 1985), arising from struggles with external competitors (Centeno 2002), and geographical challenges that inhibit the ability to project territorial authority (Herbst 2000). While necessary for understanding the historical weakness of the Colombian state, they are not fully sufficient to explain the aforementioned insti­ tutional caveats. Rather, it is important to note that the genesis of the “two-party state” coincided with the establishment of what Colombian sociologists and his­ torians have termed the period of traditional clientelism. Webs of patronage formed around natural leaders, who parasitically lived off the country’s formal institutions and thereby undercut the central government’s ability to extend its

The Novelty of the 1991 Constitution

43

authority to the entire Colombian territory. This helps to isolate the base-line conditions of the critical juncture framework. As a consequence of the underlying, archaic, and un-democratic social relations, Colombia’s political system suffered from a discrepancy between appearance and being: a fairly progressive and liberal discourse embedded in highly unequal socio-political relations. Two features reflect the clientelistic disposition of the Colombian polity. Firstly, parties were cross-sectional in leadership and followership: socio-economic elites at the top of both parties, and lower classes amongst the rank and file, reflecting the identity of their patron. Secondly, normative expectations, which in the context of political institutions that left very little space for meaningful socio-political change must result in disappointment, were restrained through a specific legal discourse. It became known as viejo derecho and, espousing an organic legalism and formalism in the interpretation of constitutional stipulations, was amenable to the exercise of socio-political power in a context of high social inequality. Actors with access to power legitimized their position with reference to the supremacy of the legal order, discursively shielding themselves from the force of social conflict. At least since Charles Tilly’s famous analogy of state making as organized crime (1985), theories on the origin of state power have focused on war making as a crucial component for the genesis of effective state bureaucracies. More recent iterations of that bellicose school of the state have isolated external warfare as a key ingredient to modernize the bureaucracy and create the emphatic “We” in nation-building processes. Conversely, the preponderance of internal warfare has an inhibiting effect on state building. It stops the evolution of a proper national identity and prevents the provision of public goods (Centeno 2002). Thus, when we ask about the reasons for the historical weakness of the Colombian state, we need to look for territorial divisions and internal strife. Incidentally, the territory that would become the Republic of Colombia underwent numerous name changes, broke off into different nation states, and experienced a number of civil conflicts. As a consequence, it lagged behind regional counterparts in terms of taxation levels and provision of public goods. The development from independence to modern Colombia was far from linear. The commotion of the Napoleonic Wars in Europe and resulting alternation on the Spanish throne opened space for Creoles in Nueva Granada to break with the Spanish state. Proper and complete independence, however, was only achieved in 1822 (López-Alves 2000, 96). During the Patria Boba (“Foolish Fatherland”), from 1810 until 1816, the liberators fought amongst themselves and failed to confront Spain as a united front, which allowed Spain to control the territory again between 1816 and 1819. From 1819 until 1830, Simon Bolívar united the entirety of the former Spanish viceroyalty under the Constitution of Cúcuta as the Republic of Gran Colombia. It lasted until 1830, when the various regions dis­ solved and formed autonomous nation states. One of them was the Republic of New Granada that occupied the territory of modern-day Colombia. In 1858, the nominally centralist republic, under the pressure of political realities, decentralized into the Granadine Confederation, and in 1863 formed the ultra-federalist republic of the United States of Colombia. Its constitution, the Constitution of

44 The Novelty of the 1991 Constitution Rionegro, was one of the most progressive constitutions at the time, protecting freedom of speech, the press, and religion, as well as the right to association, and universal male suffrage (Posada-Carbó 2006). It could not, however, stabilize political relations into a peaceful equilibrium, resulting in three major national conflicts between 1863 and 1886. Conservatives sought to centralize rule and succeeded under the leadership of former Liberal, Rafael Núñez, who proclaimed a program of national regeneration, known as La Regeneración (regeneration), and imposed a new, centralizing, constitution—the 1886 Constitution. Despite cen­ tralizing governing authority, La Regeneración did not lead to more political stabi­ lity, either. Small-scale conflicts continued until the end of the century, culminating in the most dreadful internal bloodshed, the so-called War of a Thousand Days and the secession of Panama. Peace and then prosperity only came with the reforms associated with General Rafael Reyes, discussed in the next section. Changes in the nomenclature of the various political entities in the 19th century were usually accompanied by violence. Deas (1997, 352) counts nine civil wars that “could claim to be national” (1830–1831, 1839–1842, 1851, 1854, 1860–1862, 1876–1877, 1885, 1895, 1899–1902) and another fifty local conflicts spread around the country (391). Bushnell combines civil conflict and coups as examples of acts of violence aimed at changing power at the national level, and comes to a number of 14 of such events for the 19th century (1993).3 Importantly, throughout the time of these altercations, the military of the central government remained very weak. In fact, only in the 1920s was a general draft introduced in Colombia. Given this historical context, Centeno’s argument (and by extension Tilly’s assertion) that state making is intrinsically tied up with war making finds credible evidence in Colombia’s history; even more so if contrasted with other nations in the Southern Cone that have devel­ oped more effective bureaucracies. Centeno makes the observation that for Colombia and Chile the relation between internal and external warfare was exactly reverse: Chile has endured external pressure (and war) but enjoyed domestic tranquility, while Colombia has enjoyed international peace for more than a hundred years, but “endured murderous domestic conflict” (2002, 67). As Tilly (1985) showed, weak military capacity is directly related to the incapa­ city to tax citizens. Colombia was no exception. In 1871 Colombia’s tax revenues remained at only half of what Mexico could raise at the same time. Moreover, post-independence was also marked by fiscal disequilibria between the different regions. Bushnell showed that some regions made use of their autonomy, but most “lacked the resources to do very much for either good or ill”, since the central government kept the customs, the most lucrative resource, to itself (Bushnell 1993, 26). For the most part, there existed numerous and confusing tax codes that resembled the interests of landowners and caudillos (López-Alves 2000, 102); what did not exist was a functioning, centralized, and autonomous authority. Herbst (2000) contends that projecting state authority is a particularly intricate endeavor in challenging geographical conditions with a sparse population density, because it increases the marginal costs of extending authority to frontier territory. Both conditions were evidently met in 19th century Colombia: the broken-up terri­ tory stretching across multiple climate and ecological zones, eclipsing 6,000-meter­

The Novelty of the 1991 Constitution

45

high mountain ranges and tropical rain forests made effective recruitment of armed men, raising taxes, and constructing ways of communication to build effective insti­ tutions intrinsically difficult (Bushnell 1993, 74). What is today the Republic of Colombia also lagged behind other regional hubs in the 19th century in terms of population. In 1870 (60 years after independence), only 2.6 million people popu­ lated the territory that is now Colombia. The major urban centers also accounted for a minuscule proportion of the general population. Bogotá and Medellín made up 2.5% of the population, while the figures in other primary urban centers in the region around the same time were much higher (e.g. Chile had an urbanization of 11% in 1875, and Venezuela 7%; Palacios 2006, 2). In sum, throughout the 19th century, Colombia’s central government could only claim a very deficient monopoly of vio­ lence, faced numerous internal conflicts, and was hindered by almost insurmountable geographical challenges to project authority and provide public goods in a territory that was only sparsely populated. As a consequence, the trajectory of socio-economic development evolved very unequally and always contingent on local politics. Clearly, the aforementioned theories of state building have a high degree of validity for the Colombian historical context. My contention is that there are more details important to establish base-line conditions for a proper critical juncture analysis. The difficult geography not only inhibited state making from the center; it also incentivized local power centers built around clientele relations. Bushnell made the curious observation that Colombia’s topography affected state building not only in that it made transportation and communication difficult, but also gave “each of the main population clusters of the interior a complete range of ecologi­ cal zones” (Bushnell 1993, 76–77). A complete range of ecological zones allowed each population cluster to live close to a complete range of crops and basic food­ stuffs, providing incentives for autarchy and disincentives for trade amongst clus­ ters. This was a crucial asset for the autonomy each of these clusters exercised under the domination of a traditional leader, who became the nucleus of the party system that evolved soon after independence and dominated politics for the next 150 years. The argument is that behind the propensity for internal warfare lie patronage systems that tied together soaring inequalities between the social classes, inequitable access to political power, yet stable and deeply entrenched political parties. The essential addition of my analysis is the institutional component that emphasizes how formal and informal relations conjoined in the party system. In the end, this further disincentivized the monopolization of violence and pro­ vided opportunity structures for the consolidation of archaic social relations and factionalist conflict. The Colombian political system in the 19th century was “dysfunctional in its capacity to contain civil violence [while] supremely functional in shielding the elite from the full political consequences of social conflict” (Bergquist 1992b, 5).4 This arrangement “continually obscured, distorted, and channeled the discontent” of social struggles into fluid political relations between patrons and clients, resulting in a disencounter with liberalism: a very weak notion of citizenship and incomplete differentiation between the private and public utilization of political power (Pala­ cios 2006). Above all, it resulted in a party system, consisting of two traditional

46 The Novelty of the 1991 Constitution parties that were cross-sectional in their support base and mostly oligarchic in their leadership. Both parties engaged highly normative discourses around different constitutional and legal values, but never consistently applied their principles in political practice, thereby protecting and embedding the socio-political status quo. The seeds for the compartmentalization of Colombia and localized clusters of clientelism were planted early, because the trajectory of the wars of independence opened space for private actors to consolidate their coercive capacities. Bolívar’s armed forces liberated what became Colombia first and then moved to other parts of Nueva Granada, essentially demilitarizing the area without an authority backed by coercive forces (Payne 1968, 121; see also Centeno 2002, 232). Mercenary troops began looking for “employment on the payroll of the central government [but were eventually] absorbed by landlord party militias” (López-Alves 2000, 135). Already during the Patria Boba, factionalism split the territory into different Juntas conglomerated in the most important urban hubs—Bogotá, Medellín, Calí, Barranquilla/Cartagena. There was not one central authority that was delegating, let alone dominating, these different centers, leaving each region under the con­ trol of regional militias and their corresponding caudillo (Deas 1997; Palacios 1998 and 2006). Thus, localized power centers became the foci of coercive orga­ nizations and public policy evolved around them, not a national political project. While regional land-cum-warlords built informal power networks, the creation of the party system around the two traditional parties, the Conservative and Liberal Party of Colombia, added the crucial formal institutional framework that cemented the compartmentalization into the political fabric. Lipset and Rokkan showed that European party systems originated in social or cultural cleavages (1967). In Colom­ bia, this did not occur: the apparent schism between the liberators Bolívar and San­ tander did not create long-lasting cleavages, as Bolívar’s followers were essentially muted from public politics by 1830, and both parties essentially became Santander­ ista (Bushnell 1993, 22; Valencia Villa 2012, 112).5 The social stratification of society was evident in either party, as both parties were cross-sectional in their support base and oligarchic in their leadership (Dix 1989); cultural divisions were miniscule, focusing on the role of the Catholic Church in public life, but not reflective of general confessional divisions (Dix 1989; Bushnell 1993). Rather than reflecting social, poli­ tical, or confessional cleavages, parties fulfilled functional imperatives in a territory that lacked a clear and generalized authority. Political parties were essential for dealing with collective action problems arising out of the exercise of political power in the context of numerous coercive actors that had built their power on relations of socio-political dependence in a highly stratified society in the absence of a strong centralizing force. Without established central power, these warlords relied on their own confines to resolve property rights and power issues amongst themselves. As López-Alves explains, “in the context of a weak army and a weak state, small-town and rural elites often sought the support of neighboring caudillos and their militias” (López-Alves 2000, 98). Political parties served as a coordination point between rival factions vying for political control over material resources. Rural caudillos and urban politicians began to understand that if they pooled their capacities together for electoral

The Novelty of the 1991 Constitution

47

purposes and control of the territory in rural regions, they could all benefit by “controlling and later dividing the spoils without really threatening the stability of the system they came to dominate” (López-Alves 2000, 98; see also Hoskin 1998). Scholars have shown that voter turnout fluctuated significantly despite very similar socio-economic indicators and even reached 100% in a highly illiterate society (Payne 1968, 123–129; Bushnell 1971; see also Bushnell 1993, 115–116). All of this indicates that patrons mobilized their clients to go to the poles and cast their vote based on affection for the patron, and not for programmatic considera­ tions of the particular political party. The structuration of political power through an embedded party system that undermined the state’s autonomy from social forces meant that as parties grew in importance, state structures became weaker and less independent from social forces (Leal Buitrago 1984). The result was that in the 19th century, parties were the main forces tying the loose nation together, with the caveat that they lacked the organizational autonomy of a functioning bureaucracy. The fact that the Lib­ eral and Conservative Parties “were partly electoral and partly military organiza­ tions, with proportions fluctuating over time” created a negative feedback circle. Without the constraints imposed by the authority of an autonomous state, parties could always retreat to the trenches to solve political conflicts at the expense of the state’s capacity to effectively coordinate society (Mazzuca and Robinson 2009, 287). The violent conflicts of the 19th century were essentially a sign of these coordination points breaking down. Importantly, Colombian conflicts never pro­ duced a clear victor, were resolved only by exhaustion and compromise, and thereby resulted in frail equilibriums until the next clash (Centeno 2002, 67). Lastly, the legal discourse that evolved alongside the traditional party system was also agreeable with the exercise of political power through informal patronage networks. Despite the relatively progressive and normatively imbued narratives each party professed at times, they never implemented progressive policies. As Palacios argues, parties were essentially “mouthpieces of an order which [accep­ ted] the principles of citizenship but [were] incapable of applying them” (Palacios 1998, 38–40; see also Valencia Villa 2012). Whenever the discourse became too subversive, elites changed their party allegiance and moved to the other side of the aisle without breaking off personal relations with the elites of the other party. Viejo derecho (Old Law/Right), as the classical expression of legal discourse in Colombia is known, functionally served the stabilization of clientele relations rather than pro­ viding normative yard sticks for substantive claims of justice. Fifteen national con­ stitutions were enacted between 1811 and 1886 (Valencia Villa 2012, 30). The doctrines evolving from these “Cartas de Batalla” (battle charters; Valencia Villa 2012) had to fulfill two functions in order to stabilize clientele relations: 1) subdue factionalism below the constitutional level by providing remedies in the face of normative disappointments; and 2) assuage the tension resulting from high inequality paired with a constitutional republic that never could provide a satisfac­ tory level of public goods. Viejo derecho achieved this by espousing very formalistic and legalistic norms. Firstly, it affirmed the supremacy of legal reasoning over poli­ tical reasoning: “it uses the language of law to solve its problems”.6 While certainly

48 The Novelty of the 1991 Constitution in line with classical liberal thought, it is important to note that this was discursive but not necessarily a speech act that entails an obligation to live up to the utterance. Rather, it meant that arguments had to “dress up juridically” and take the form of legal arguments.7 Secondly, and related to the primacy of legal argument over political arguments, was the embrace of a strict legalism. Passed down from both liberators’ affection for American and French republican thought, viejo derecho placed a lot of importance on administrative law in the organic structuration of the constitutional state, subsuming substantive claims involving rights of individuals or groups under questions of procedure and competence. Cynically speaking, in Colombia constitutional change was often an exercise in changing everything so nothing changes. One Constitutional Court judge explained that an overly legalistic culture “has many institutional conjunctures, but not many consequences”.8 This was not a coincidence. With the material context retreating behind the formalities of law, the resultant doctrinal ethos embraced an over­ confidence in the transformative capacity of law alone, which, in fact, decoupled law from addressing foundational socio-political grievances and practices. Together, these legalistic parameters of viejo derecho were not only amenable to the exercise of social and political power, but directly contributed to the stabilization of socio-political inequality. Couching substance in the imperatives of legal discourse and the supre­ macy of formal aspects of legal interpretation served the interests of those in power, shielding them from claims of justice and equality. Indeed, it consecrated a cult of order that Colombians colloquially continue to describe in an idiom to this day: “La ley es para los de la ruana!” (“the law is for those with the poncho”). The ruana is a Colombian poncho worn by the campesinos, meaning that the law is only applied to those of the lower classes and managed by the ones from the upper classes.9 From this brief discussion on the genesis of institutions in 19th century Colombia follow a number of observations that form the base-line against which the subsequent sections will analyze junctures of institution building in the 20th century. As seen, public goods and their generalized distribution—above all security in the form of the monopolized means of violence—remained a rarity throughout the 19th century. Partisan caudillos and their accompanying militias fought each other over property claims and the nomenclature of the constitutional order until an equilibrium amongst the warring factions was found. The result was an ineffective institutionalization of linkages between the center and the peripheral regions in the Andean nation. The base-line conditions are therefore informal insti­ tutions embedded in patronage webs, which extended into formal, and relatively democratic, institutions. The contradictions were assuaged by a formalistic and lega­ listic discourse focused on the organic structuration of the constitutional state.

Colombian Consociationalism: Military Tutelage, the Exclusion of the Public, and Elite-Driven Institution Building After having outlined the methodological critical juncture framework, this section turns to (potential) junctures in Colombia’s history. Internal variation in a case study design helps to counteract the “analytical equivalence problem” of comparative­

The Novelty of the 1991 Constitution

49

historical research that often surfaces when comparisons are made between cases that differ to such a degree “as to undermine the idea that it really involves the same cri­ tical juncture” (Collier and Collier 2002, 31; original italics). In addition, contrasting the continuities and discontinuities in specific sub-systems mitigates the lack of external differentials, so that we are indeed observing “different ways in different cases” (Collier and Collier 2002, 31). This section will look at the reforms associated with General Rafael Reyes and the reforms that established the National Front gov­ ernments, analyze the appearance of cleavages prior to institutional engineering, detail the political interaction during the juncture, and contrast the outcomes to evaluate the profundity of change induced during the juncture. At its core, the argument is this: in both cases, reformers responded to crises that essentially began as intra-party disputes, which then morphed into violent inter-party conflicts that eventually escalated to civil war proportions. Violence did not erupt from deep social or cultural cleavages, but broke its path due to party elites’ inability to compromise and strike deals that satisfied the entire range of either party’s mem­ bership and leadership. In both instances, high levels of internal violence signaled to elites that the original political arrangement was insufficient for containing the clashes of interest and identity between them. This resulted, in both cases, in elite-driven and controlled processes of institutional reform, which eventually established new arrangements, giving each party significant say in the path future politics were to take. Eventually, these arrangements, too, were insufficient in containing pressures on the political regime, opening a new round of violence and, eventually, institutional engi­ neering. Most importantly, reforms in 1910 and 1957 conserved forms of clientelism by placing them on a more stable formal institutional setting without fundamen­ tally altering the underlying socio-political interaction. Unpacking each juncture and detailing its trajectory from crisis (violent conflict) through juncture (actors negotiating new institutions) to legacy (institutional framework) sets up a pro­ ductive comparison with the 1991 juncture. It will help to specify differences in each component of the critical juncture framework. The Political Origin of the Thousand Days’ War and La Violencia Identifying the origins of conflicts in Colombia is a difficult task, because socio­ economic grievances constantly interacted with political contentions, blurring the lines between them (Bergquist 1992b). The instances of institution building ana­ lyzed here are no exception to that. In both cases, socio-economic relations were strained by either the sharp decline in coffee prices prior to the Thousand Days’ War (Bergquist 1986) or the disappointment of the mobilized working class before the breakdown of order in 1948 prior to La Violencia (Collier and Collier 2002). At the same time, political interactions between elites in both cases had reached impasses that could not be assuaged by the existing institutional setup and eventually resulted in violent escalation. Given the constancy of social inequality beneath the political arrangements that reaches back to the 19th century, I argue that both junctures were not the result of particularly crisis-prone (socio-eco­ nomic) cleavages, but resultant of failures to be found in intra-elite interactions. I

50 The Novelty of the 1991 Constitution hold that this becomes apparent, once we contrast the junctures explored here to the changes in the 1980s when a very profound (normative) cleavage emerged. Amongst the numerous violent, internal conflicts that have paved Colombia’s history, the Thousand Days’ War (1899–1902) and the period simply known as La Violencia (The Violence; 1948–1953) were some of the bloodiest. In the end, the Thousand Days’ War is estimated to have cost 100,000 lives and La Violencía 170,000 (Hartlyn 1988, 20 and 44; this is a conservative estimate; others go as high as 400,000; see, for example, Sanchéz and Meertens 2001). The Thousand Days’ War had another debacle in store for Colombians with the secession of Panama, when the United States took advantage of the internal turmoil within Colombia and protected a secessionist movement in Panama in order to get the Panama Canal built under favorable conditions. Some historians viewed the origin of this conflict in fluctuations of demand for Colombian products. In the case of the Thousand Days’ War, violence coincided with the fall of coffee prices, which resulted in a deep depression that lasted from 1898 until 1910 (Bergquist 1986 and 1992a). Additionally, fighting always broke out first in areas of the country that were most deeply affected by shifts in international demand for Colombia’s products. Together this appears to suggest that socio-economic strain resulted in violent conflict (Safford and Palacios 2002). Recent research has shown, however, that political fracturing and economic strain reinforced each other, producing impasses that caused the institutional breakdown and subsequent violent altercations. Mazzuca and Robinson (2009) traced the beginning of the violent clashes to a dysfunctional political framework that perma­ nently excluded Liberals from power and representation. It aggravated the economic crisis, because it set incentives for using public funds for the suppression of opposition contestation rather than productive investments. “Economic turmoil was” therefore “an indication of the impact of political conflict on the growing fiscal crisis and the associated monetary expansion of the 1890s” (307). In other words, socio-economic strains were compounded to crisis proportion by the failure of political institutions to coordinate between factions and utilize public funds to counter-cyclically soften the effects of fluctuating world prices in key export commodities. In the 1890s, each party split into one more and one less radical faction. On the conservative side, Nationalists and Historical Conservatives disagreed on the rigidity of centralization and importance of foreign trade. On the other side, Lib­ erals divided into one faction around the old Radical oligarchy (more radical in name only, but in fact more centrist in nature), and a younger and more bellicose faction, called the Nuevos (“the New Ones”). Nationalists and Nuevos vehemently opposed one another, while Historical Conservatives and the Radical oligarchy of the Liberal Party were more peace-oriented due to their shared commitment to foreign trade (Safford and Palacios 2002, 250). The institutional framework strengthened the more radical faction on each side, putting the political system on a road to institutional breakdown. After Con­ servatives had won the civil war in 1885 and imposed the 1886 Constitution, they implemented two institutional mechanisms that essentially sidelined Liberals to political obscurity (Mazzuca and Robinson 2009). Firstly, they undercut Liberal

The Novelty of the 1991 Constitution

51

influence at the regional level by implementing the appointment of Governors as a recentralization mechanism. Secondly, they took advantage of extra-ordinary powers granted in Law 61 of 1888 to “unilaterally ban or repress political activities that the president himself considered ‘offensive of public order’” (293). Increas­ ingly frustrated by being delegated to political irrelevance, Liberals issued appeals for more equality before the law as thinly veiled demands for a better and more equitable access to power. They even succeeded to convince the Historical faction of Conservatives, but to no avail. Nationals dominated the Senate and quashed all hopes for electoral reform that would have allowed Liberal minority representation in Congress (a point that will later be picked up as a mechanism to ensure peace; Mazzuca and Robinson 2009, 295–296). As a consequence, the younger genera­ tion Liberals rebelled in October 1899 setting in motion an escalation pattern. The central government issued civil and military authority with the power to decree loans to governors, who used that power to force affluent Liberals to con­ tribute funds. This pushed the more centrist Radicals to the Nuevos and reinforced party identity, dividing “Colombians along party lines more than along those of socio-economic classes” (Safford and Palacios 2002, 250). For the next three years, Colombians fought each other along those party lines, resulting in the most ensanguined internal conflict Latin America had seen yet. The prelude to La Violencia followed a similar pattern as the runup to the Thousand Days’ War. It was also marked by socio-economic contentions. As elsewhere, the Great Depression had an impact on Colombia, and, together with raised expectations resultant from the mobilization of the working class, the demand for socio-economic reforms became apparent. As in the late stages of the 19th century, however, political institutions were not only inept to incorpo­ rate lower class demands, but contributed to a radicalizing, first intra-party, fracturing that pushed factions towards conflict with one another, and eventually into violence along party lines. In the 1920s, organized labor had begun to mobilize and develop political linkages with the Liberal Party, beginning the process of labor incorporation (Collier and Collier 2002). The speed of incorporation increased when propor­ tional representation was introduced in 1929. Liberals around Senator Luis de Greiff campaigned for its introduction and succeeded in their endeavor by pro­ viding support to a Conservative candidate, Alfredo Vásquez Cobo. As in the late 19th century, the Conservative Party had split into two opposing blocs prior to presidential elections in 1930 (Mazzuca and Robinson 2009, 307–308). The concession by the Vásquez Conservatives to Liberals angered the rest of the Conservative caucus, which decided to run a candidate on their own, splitting the Conservative vote in the 1930 election between Guillermo Valencia and Alfredo Vásquez Cobo. This gave Liberals the opportunity to retake the pre­ sidency with 44.9% of the national vote. (The Conservative candidates received 29.2% and 25.9%, respectively.) The violent crisis that escalated into La Violencia had its origin in the abrogated incorporation of Colombia’s working class into the party system. In contrast to other countries in the region, labor incorporation arrived relatively late in

52 The Novelty of the 1991 Constitution Colombia in 1929. However, it came “early in relation to the development of the Colombian labor movement” itself (Collier and Collier 2002, 271). As a result, workers were not essential for winning the majority of votes (but rather an asset), nor did they exercise crucial electoral control. Additionally, the Liberal Party, in contrast to other incorporating parties the Colliers studied in Shaping the Political Arena, was not specifically founded by workers to bring labor into the formal political system, but was itself an old traditional party with roots in the 19th cen­ tury. These two factors combined to produce the crisis that resulted in the bloodshed of La Violencia by developing a dynamic that split the Liberals into two factions. The core of the incorporation period consisted in the Revolución en Marcha (Revolution on the March); the relatively modest New Deal–type policies of President Alfonso López Pumarejo (Collier and Collier 2002, 305). The radical wing, which would eventually turn to labor lawyer Jorge Eliécer Gaitán as their leader, wanted to continue on that path and push for farther-reaching reforms, while the more moderate wing, backed by the landowning oligarchy in the Liberal Party, already considered the López reforms as too extreme. By the early 1940s, the internal divide inside the Liberal Party escalated and pro­ vided the Conservatives with an opening to regain power, setting in motion the familiar escalation path. While Gaitán was capable of coalescing wide public support behind his leadership, it antagonized the “moderate” wing inside the Liberal Party and pushed them towards the Conservative Party. The Conservative Party had not run candidates since the 1934 elections, but, motivated by the Liberal split, returned to the polls. Moreover, goaded by Gaítan’s populism, former President López moved to support the moderate Conservative candidate Ospina, who was elected to the presidency in 1946. The proclaimed unity government between moderate Con­ servatives and Liberals could not prevent the Gaítan wing from controlling the Lib­ eral caucus in Congress, which announced him as their candidate for the 1950 presidential elections. As the Liberal split deepened, Conservatives began to violently retake control of the state at the regional level, transforming the intra-party fracture into an inter-party conflict. As a consequence, the coalition between moderates on both sides of the aisle collapsed, and after Gaítan’s assassination in April 1948, order broke down entirely: first in Bogotá and then in the entire country. In November 1949 the Conservative Gómez was elected president unopposed and, “with the exclusion of Liberals from all areas of government, both national and regional, the institutional breakdown was complete” (Collier and Collier 2002, 312). As in 1899, elite politics as conversation among gentlemen failed, resulting in a drastic escalation. Together, the preceding conditions of the descent into lawlessness and infight­ ing in Colombia in the 1890s and 1930s/1940s show remarkable similarities. While it is true that socio-economic tensions—as always in Colombia—bubbled beneath the formal institutional level, it was political intransigence that resulted in the crises that could no longer be contained by the political framework. The majoritarian electoral system of the 1886 Constitution propelled Conservatives to power and ensured their dominance in national politics. They utilized their posi­ tion to further marginalize Liberals from the exercise of power—above all at the

The Novelty of the 1991 Constitution

53

regional level—and set the incentives in the political system to confrontation. As Bushnell explains, in a two-party system, in which parties have existed long enough for their followers to develop strong attachments to them, the strength of party loyalties when added to the natural advantages of incumbency makes a government almost impossible to dislodge unless the ruling party itself becomes irrevocably divided. (Bushnell 1993, 18) Prior to the institutional juncture that followed the Thousand Days’ War, the situation escalated not as a consequence of external pressure—even if the economy felt the strain of falling commodity prices—but rather as a consequence of an escalating dynamic that evolved from within. The Colombian political evolution in the 1930s and early 1940s seems to suggest a crisis and critical juncture that is rooted in macro processes of incorporation into the world economy, creation of an urban working class, and mobilization of a labor movement. If placed in the context of its own history, however, the process in the 1930s reveals familiar pat­ terns that have destabilized the political regime beyond the brink of civil war before. Unequal social relations conjoined with formal institutions stood in the way of creating avenues to properly change the political interaction and give voice to lower class demands. This resulted in a dynamic, in which political elites meandered between factions and coalitions, effectively destabilizing the formal political system to the degree that war as other means of politics became the only way out. In neither case did an external shock aggravate a particularly potent cleavage to the violent crisis. Rather, internal, political maneuvering created unintended consequences that paved the way to civil war. Political Junctures: Stabilizing Institutions to Contain Violence Since mechanisms of the political system to facilitate peaceful interaction amongst elites failed, and the bloodshed of the Thousand Days’ War and La Violencia resulted, consociational institution building to augment inter-party cooperation was an apparent panacea. Indeed, I argue that institution building was the response to a political system run aground on the inability to find compromising deals. Amongst other things, we will quickly face the peculiar role the military played at these important junctures in Colombia’s history. The episodes following the Thousand Days’ War and La Violencia were the only instances when military generals were in power in Colombia in the 20th century. Equally quick arises the realization that these periods of military rule were very different from the authoritar­ ian bureaucratic regimes elsewhere. Military rule in Colombia bought time for refor­ mers to alleviate problems with civilian rule that had become evident in prolonged periods of violence. The trajectory in both instances was remarkably similar: new intra-elite pacts, facilitated by military rule, mended fences and stabilized the informal system of clientelistic relations conjoined to Colombia’s formal political institutions.

54 The Novelty of the 1991 Constitution All of these processes reflected an elite-driven logic and the implemented frameworks satisfied elite exigencies, rather than bowing to genuine civic activism and popular demand. In the end, the generals returned power—not entirely voluntarily, but without the use of violence or widespread civic activism—to civilians, and civilian rule continued on its path. These crises were reflective of political impasses rather than profound social or cultural cleavages. The governments of General Rafael Reyes, in power from 1904 until 1909, and General Rojas Pinilla, in power from 1953 until 1957, were the closest cases of dictatorial rule in Colombia. Yet, their ascendance to power as well as their respective demise from rule was anything like the rise and fall of other military dictators of the bureaucratic-authoritarian regimes in the Southern Cone. General Reyes had built up his prestige in the civil wars of the late 19th century and was nominated for the presidency by the Conservative Party. Rojas Pinilla was “essen­ tially thrust into power in the midst of La Violencia by a segment of the Con­ servative Party and elements of the military” (Hartlyn 1988, 48). Importantly, neither General utilized the armed forces to attain the presidency against the will of civilian politicians; forces within the formal political establishment sought a neutral arbitrator to end the violence and “invited” them to take power. Even though both developed a taste for what Latin Americanists term con­ tinuismo (McConnell 2010), trying to overstay their welcome, the same forces of the political elite facilitated their peaceful exits from power. General Rafael Reyes stayed in power until 1909, when he was dislodged by a coalition of partisans from both parties (Solaún 1980, 3; Esguerra Portocarrero 2012, 84). Favoritism and corruption were eroding Rojas’ position in power in May 1957. His last attempts to hold on to power and arrest both parties’ presidential candidate, Valencia, resulted in widespread demonstrations by students, bankers, and industrialists. To be sure, the demonstrators did not come from the popular classes, as for example in 1948. Rather, they were instigated by industrialists and banks, which later assured labor support. In addition, support amongst the military for General Rojas was also waning, in part due to the corruption that spurred fears their own role might be in jeopardy (Hartlyn 1988, 58–59). He then nationalized banks, disallowed interest and convened a constituent assem­ bly to approve his re-election, but all in vain: a five-man military junta was placed in charge and Rojas fled into exile, while representatives of each party negotiated the terms of the National Front.10 The way that institutional reforms were instigated also bore surprising similarities in both instances and reflected elite control of these processes. The bipartisan move­ ment to dispose of Reyes in 1909, the Partido Republicano, never consolidated as a new party. It did, however, institute a constituent assembly to overcome the partisan and factionalist tensions that had fueled internal conflicts and civil wars. Crucially, the republican movement did not root its support in a mass movement, but sought sup­ port from merchants and industrialists and largely “foreshadowed the National Front” (Hartlyn 1988, 27). The disposition of General Rojas in 1957 opened the door for bipartisan negotiations. While the military junta took over the daily affairs in Colombia, representatives from both parties came together and finalized negotiations

The Novelty of the 1991 Constitution

55

that had commenced in 1956 in the city of Benidorm. The final pact was signed in Sitges in July 1957, and it contained the most important parts of the National Front, under which Colombia was (officially) governed until 1974. One minor difference between the constitutional reforms of 1910 and the National Front reforms was that the latter were given to the people for (post-factum) ratification in a plebiscite. Nevertheless, in neither case were popular classes or civil society involved in the negotiations that set up the institutional arrangement. As will be shown below, this is a crucial difference to the origin and negotiations of the constituent assembly that wrote the 1991 Constitution. The institutional changes implemented in each instance of institutional engineering involved the legislature, judiciary, and, in the case of the National Front, also the executive and bureaucracy. As both are examples of consociational agreements, they aimed at securing mutual guarantees and representation of each party in the most important national institutions, with the National Front going further than the 1910 reforms. General Reyes had already embraced a form of minority representation in the armed forces, encouraging Liberal families to send their children to the armed forces (Bushnell 1993, 156). The national assembly in 1909 incorporated the principle of minority representation for the electoral system and replaced the majoritarian vote of the legislature with the incomplete vote, thereby blunting the antagonizing effects of the old electoral regime. The caveat of the incomplete vote was that the first and second position in a given election were still at stake, but the relative institutional power that resulted from those elections was pre-established. It assigned fixed portions to the winner and the runner-up, irrespective of the actual popular vote (Mazzuca and Robinson 2009, 289). This guaranteed the Liberals a greater share of representation in Congress, even though the Conservatives retained control of the executive until 1930. In terms of mutual guarantees, the National Front agreement went further than the bipartisan reforms at the beginning of the century with the institutionalization of parity and party-alternation in Colombia’s bureaucracy and political institutions. Pre­ viously elected offices were depoliticized and alternated between the two parties. Congressional elections were constrained and depoliticized: neither party was allowed to attain a majority in either of the two chambers.11 Parties in Congress then had to agree on a candidate to be elected as President, switching between each party. Finally, equal share of representation was extended to most public offices other than the mili­ tary. In the civil service, restrictions on political activities of government officials were imposed as well as prohibitions against political discriminations in hiring practices. In both instances of institutional engineering, judicial institutions also underwent significant reform. While the Supreme Court was introduced in the 1886 Constitu­ tion, it did not have a monopoly of interpreting the constitutional validity of norms; questions regarding the constitutionality of laws heavily involved both, the executive and legislature, in investigating claims of such nature. The 1910 reforms changed that dramatically and introduced an unprecedented novum. While the US had seen the principle of judicial review enforced in the famed Madison vs. Malbury decision, the Colombian reforms made constitutional review a quintessential citizen right. The aptly called acción popular de inconstitucionalidad (popular action of uncon­ stitutionality) anticipated Hans Kelsen’s introduction of a similar acción in the

56 The Novelty of the 1991 Constitution Austrian Constitution after World War I, which marked the beginning of constitu­ tional control by an independent court in Europe (Esguerra Portocarrero 2012, 107; Cepeda Espinosa 2007, 17; Mendieta 2010, 72). The National Front reformed the selection of Supreme Court magistrates and placed their nomination fully into the hands of the judiciary (at arm’s length of the executive and legislature). The crucial caveat was contained in Article 12. It sti­ pulated that judges of the State Council select the replacements of vacant positions on the Court, fully depoliticizing the selection process. In Colombia, this is refer­ red to as cooptación. While it has been lamented that this has resulted in a form of institutional corporatism undermining meritocracy in the legal career, it is undoubtedly true that it significantly increased the independence of the judges and the judicial body itself.12 In sum, these reforms increased the independence of the judges, while those at the beginning of the century fortified the authority of the Court and eased the access to that institution. It is important to note, however, that establishing independent judicial institutions does not contradict the elitedriven logic of the processes. On the contrary, as Finkel showed, they can serve as a political insurance mechanism in times of uncertainty (2008). This also holds true in this case (Hartlyn 1988). Conspicuously, rights and the protection thereof did not play a more significant role in either instance, but would feature very prominently in the 1990 constituent process, reflecting the popular origin of the movement instigating the reform process. It is my argument that elite-driven constraints in both institutional engineering processes also showed in the results of either process, and therefore indicate a prima facie causal relation. This causal relation is further strengthened when we look at the post-juncture evolution of both processes. The aftermath of both junctures reinforce the notion that political stability came at the cost of democratic quality. It appeased the country, brought economic development, but eventually ran out of steam and thereby contained the seed for further contentions down the road. Consociationalism and its Discontent: Political Stability at the Cost of Democratic Quality It is beyond doubt that both institutional junctures improved the lives of Colom­ bians. Appeasing the country helped to propel economic growth. Eventually, however, in line with the elite logic of each process, development expired, and the problems of lower-class disenfranchisement, political exclusion, and socio-economic immobility resurfaced. The institution building processes dealt with specific, political contentions, and the solutions imposed reflected the actors’ interests present at the negotiating table. As a consequence, the implemented institutional remedies could solve the issue of intra-elite intransigence, but did not provide deeper resolutions to issues pertaining to rights guarantees, democratic participation, and economic opportunities. The most important successes of both reforms were the noticeable decline of violence. After the reforms of 1910, internal violence all but disappeared from Colombia and the reason is causally linked to those reforms. Politics in the

The Novelty of the 1991 Constitution

57

independent republic had favored power monopolization and thus “forced the opposition into revolutionary tactics and the government into violent repression” (Mazzuca and Robinson 2009, 286–289). The electoral system in particular had favored power monopolization. The assurances implicit in the incomplete vote suf­ ficed to switch the incentives from confrontation to accommodation (see also Maya Chaves 2012). The National Front also achieved its primary goal of appeasement. From the average of 25,000 deaths per year during La Violencía, the numbers declined to 3,600, 5,243, 3,243, 2,631, and 2,868 deaths per year for the first period of the National Front (1957–1961; García Villegas 2001, 318). The last vestiges of violence were contained to the countryside, where the first guerrilla movements started to form. Not only did both consociational agreements consolidate a more peaceful national environment; they also provided a context conducive for economic and administrative development and (selective) modernization. The epoch commen­ cing with the Reyes reforms is opportunely termed the Coffee Republic (Safford and Palacios 2002, 266). Colombia’s economy experienced an unprecedented boom through its incorporation into the world coffee economy and challenged Brazil’s uncontested position at the realm of the coffee world trade (Bergquist 1986 and 1992a). The export economy grew between 1910 and 1920 by a rate of more than 10% per year. In 1898, before the disruptions of the Thousand Days’ War, coffee contributed almost half the country’s foreign trade. By 1924, it con­ stituted 80% of exports and, in total numbers, grew six-fold from 1896. It was not only coffee production that grew in Colombia. Other industries sprung up and consolidated growth: the textile industry in Medellín, beer brewing in Bogotá, the petroleum industry in the Magdalena Valley and Barrancabermeja, and fruit plan­ tations in the Santa Marta region of the Atlantic. Infrastructure, too, developed, and, although not matching the Porfiriato in Mexico, state builders expanded the railway and highway system in that period (Bushnell 1993, 170–180). As with the Reyes reforms, the National Front arrangement also provided a context, in which economic development could flourish. Notably, this occurred even though the coffee bonanza had ended, and world prices were in decline (Hartlyn, 1988, 111; Safford and Palacios 2002, 276). Despite relatively inop­ portune external conditions, Colombia experienced an average GDP growth of 5.15% per year from 1957 until 1981. In contrast to most other South American countries, it did so by following “moderate” and “eclectic” economic policies, refraining from “pendular” patterns witnessed in Argentina, Chile, and Peru at the time (Hartlyn 1988, 103). The regime focused on key macro-economic variables and abstained from populist politics at the national level. The state and its bureaucracy also underwent some modernization and capacity increases under the National Front arrangement. By eliminating struggles for control of the state, consociationalism permitted both parties to benefit from the spoils of a more functioning state, thereby allowing its capacity to develop. Cru­ cially, though, this strengthening was selective and contingent on the support of key elites—most notably the President. If he made it his prime focus to develop state capacity, advances could be made in the area of state planning and

58 The Novelty of the 1991 Constitution developmental strategies. The most important political figure associated with such policies was Carlos Lleras Restrepo (1966–1970). From 1960 until 1970 public spending grew from 16% to 31% in relation to GDP (Hartlyn 1988, 127; see also Bejarano and Segura 1996). Despite this progress, the eclectic nature of develop­ ment revealed the constraints imposed by the nature of the regime itself—namely its embeddedness in patronage systems. When we take a more long-term perspective on these consociational deals, one pertinent deficiency, evident in both cases, becomes readily apparent: the new insti­ tutional setups never addressed socio-economic inequities and uneven access to power. On the contrary, both arrangements conserved the socio-economic status quo and blocked important transformations. In comparison to other South American states, which degenerated into bureaucratic-authoritarian regimes, Colombia might have appeared as a viable, and relatively democratic, alternative between radical revolution and reactionary authoritarianism. Yet, its underlying informal institutions were far from supremely democratic. They built on what Acemoglu and Robinson termed “extractive institutions” and as a consequence never created sustainable con­ flict mechanisms that could effectively address collective action problems. Extractive institutions, in contrast to inclusive institutions,13 prevented what they dubbed “creative destruction”: it shielded socio-political elites from change. Consequently, economic development that rested on extractive institutions inevitably had to expire, because it could not sustain innovation (Acemoglu and Robinson 2012, 76 and 377). The trajectories of Colombia’s political regime from the beginning of the 20th cen­ tury until 1948 and then again after the National Front confirm Acemoglu and Robinson’s institutionalist hypothesis. The reforms of 1910 and 1911 aimed at institutional re-accommodation. Conservatives at that point simply traded more political power for political sta­ bility. The system ensured some “patronage and material benefits” for the Lib­ erals, “while the Conservatives retained for themselves the lion’s share of power and privilege” (Martz 1997, 48). Minority representation was not pushed upon elites by the masses, but rather developed as a “means for elites to share power among themselves in a way that would avoid infighting” (Robinson 2013, 44). Above all, the constitutional arrangements were based on, and intended to pro­ long, the “exclusion of the majority from political life”, delegating the masses not only in politics to a subservient role, but also in terms of social, economic, and cultural leadership (Dix 1967, 78). The coffee economy, too, played its part in prolonging elite rule. It had a very “conserving effect on society”, in that rural masses were more willing to accept elite leadership. This included the capitalist order as well as the Liberal-Conservative control of politics (Bushnell 1993, 174). Conversely, sinking prices resulting from the global depression created a regime-antagonizing dynamic. As shown, Colombia’s consociational arrangement showed signs of fracture in the 1920s when “issues of class conflict and the new ideological positions posed by the emergence of an organized working class” surfaced (Collier and Collier 2002, 313). During the coffee bonanza, peasants had taken up loans to invest. As prices decreased during the Great Depression, they were increasingly exposed, forcing them from their

The Novelty of the 1991 Constitution

59

lands. Intense urbanization meant that economic depression would be felt in city centers and not far away in the periphery (Dix 1967, 79–80). Liberals wanted to exploit this and built ties with the urban working class. They gained power, however, not because of greater electoral sway, but because of a political schism within the Conservative Party. It gave the Liberals the executive the first time in over 50 years, inaugurating the incorporation period. Despite rather modest New Deal–type poli­ cies, integration eventually failed due to the cross-sectional nature of the incorporat­ ing party, the Liberals. Its regional bosses had tight control over rural peasants and when policies were becoming increasingly progressive in the 1930s (to satisfy workers in urban areas), it alienated exactly those regional bosses, creating rifts, which the Conservatives benefited from to regain power in 1946. The oligarchic nature of political elites, the anchoring of political parties in the regions, and most importantly, the cross-sectional composition of both parties, meant that neither party could develop functional ties with the working-class movement (particularly unions), with­ out offending an important part of their constituent and party elite (Collier and Collier 2002, 300–313). The consociationalism of the National Front, too, eventually succumbed to crisis, resulting from immobilism, lack of responsiveness to popular demands, and policy incoherence. At the root: patronage systems in the periphery. From its very beginning, the political setup of the National Front had contradictory effects. Political leaders pursued three goals: generate popular support for the agreement, defuse inter-party conflict, and prevent political alternatives outside the initial agreements from taking root (Hartlyn 1988, 148). However, parity and alterna­ tion in power were in fact disincentives for generating high voter turnouts, while the existence of factions within the two parties that were opposed to the agree­ ments as well as alternatives outside of it (for example, Rojas Pinilla’s ANAPO movement) meant that every election was essentially a plebiscite on the legitimacy of the agreement itself. The dreaded immobilism and policy incoherence then surfaced as a consequence of party leaders’ peddling to both sides: they “con­ tinued to collaborate in government and simultaneously attempted to serve as political opposition” (Hartlyn 1988, 148). Caught in this constant limbo, national elites increasingly relied on patron-client and brokerage ties with regional power brokers to consolidate their electoral position. Thus, the National Front could never address the grievances that were at the root of political instability in the first place: social inequality (particularly in rural areas) and inequitable access to poli­ tical power. It was therefore not a coincidence that land reform persistently failed. When Lleras Restrepo attempted only minor land reforms, the party elite from the Caribbean coast protested, because they were in fact landowners themselves (Gutiérrez Sanín et al. 2007, 20). Pacted Democracy in Colombia: Building Institutions without Public Constraints Before moving to the processes around the implementation of the 1991 Constitution, it is prudent to collect some of the insights discussed so far. First,

60 The Novelty of the 1991 Constitution and foundational to the base-line condition of the critical juncture framework in this analysis, we saw that formal political institutions were conjoined with highly unequal socio-economic relations and patronage systems. Importantly, these patronage sys­ tems did not disappear after the conflicts in 1899–1902 and 1948–1953, nor did political reforms after those civil wars modify the institutional setup to a degree that leveled the socio-political playing field; far from it. These informal relations survived institutional engineering, only to be attuned to the changes in the formal setup. We saw that the violence was neither revolutionary nor the prelude to a social revolution. Rather, the crises in each instance evolved from within the political establishment. Consequently, these crises did not fundamentally reshuffle the structuration of elite actors and the constrains placed upon them, but eventually led to situations that allowed elite actors to realign and assuage the conflictual potential of institutions that had polarized the political system in the first place. The engineering processes had the convenient effects that they could largely continue on a path that guar­ anteed traditional elites’ position at the pedestal of socio-economic power. The problem with consociational institution building was that it failed to address central grievances. Implementation of institutional modifications followed an elite dynamic and could relieve the polity from the symptoms of social inequality and poli­ tical clientelism only for some time. In the long run, however, both resulting regimes were insufficient to contain arising challenges, because they did not modernize the rather archaic patronage systems. The reforms at the beginning of the 20th century put the traditional forms of clientelism inherited from the 19th century on formal footings, thereby erasing conflict prone tendencies. Yet, this was not sufficient to integrate the working class into the party system, which maintained its bipartisan and oligarchical characteristic. The National Front reformed patronage systems and induced some modernity by creating brokerage systems around which patron–client relations clus­ tered; hence the term broker clientelism. It appeased conflicts, but the resultant restrictive democracy could not conceal the underlying socio-economic tensions. On the contrary, by outlawing legitimate political contestation, it opened the way for new forms of violence: revolutionary violence on the left and reactionary violence on the right. Finally, both instances of institutional reform share the fact that reforms were not imposed upon elites by an active citizenry but designed amongst elites. In both instances, a truly civic front was neither included in negotiations nor exerted pressure during negotiations. Citizens had very little agency in the proceedings, lacked a genu­ ine right to justification and contestation, and could only confirm the final result once elite consultations had concluded. The next sections will show that the 1991 con­ stituent assembly differed precisely on those parameters of public reason.

The 1991 Constituent Assembly: Students Movements and the Implementation of Public Reason The implementation of the 1991 Constitution fundamentally differed from the junctures at the beginning of the 20th century and the National Front reforms. It differed in terms of all conceptual components of the critical juncture framework: preceding crisis, juncture, and legacy. Not only did the creation of the new

The Novelty of the 1991 Constitution

61

Constitution fulfill the conditions of the critical juncture framework; the phenomena that surfaced during each component of that framework were remarkably discursive. It is for that reason that I argue that the 1991 Constitution has a distinct commu­ nicative origin. The institutional crisis leading up to the juncture signaled a much deeper predicament. Internal institutional shortcomings, above all the immobility incurred from the National Front regime, were aggravated by an external shock, namely the booming international narcotics trade, and resulted in a profound regime crisis. The anchoring of the cocaine trade in Colombia induced a dynamic of criminalization that Colombia’s political regime, despite its democratic surface pos­ ture, was incapable of containing. It created new actors with unprecedented resources at their disposal to challenge the state. The result was a profound cleavage between the normative claim of a democracy and the political reality of a regime incapable of upholding basic human rights and maintaining even minimal channels of political participation. As such, I am inclined to call it a normative cleavage between normative claim of a democracy and the political reality of a country besieged by violence. This cleavage had immediate effects on the resulting crisis. Even during the 1970s and 1980s, reform-oriented political leaders had attempted to renew the political system from within, yet failed either in Congress or in the Sala Constitucional (con­ stitutional plenum) of the Supreme Court. That is why, the path toward rewriting the constitution did not originate in Colombia’s formal institutions. In 1989, the assas­ sination of Luis Carlos Galán prompted 20,000 students to take to the street in a silent march. From this protest, a smaller group of activists formed the nucleus of a movement to call for an all-encompassing institutional reform. Their contention was that the violence and ubiquitous terror were symptoms of a much deeper crisis. Colombia still had not thrown off the shackles of the constraining National Front regime and the violence was a sign of the restrictive nature of that regime. It required a new constitution, they argued. This nucleus of an elite student organization was later joined by a broader student coalition that could legitimately claim to represent the entirety of Colombia’s students. While distinct from the original student organi­ zers, they shared their demands and justification for those demands (Dugas 2001). It was students and their movement that gave the project of reform an unmistakably democratic legitimacy and drove the process until a national plebiscite was held alongside the presidential elections in 1990. Thus, the origin of the constituent assembly was genuinely civic and associational, rooted in a large sector of civil society. The path through Colombia’s formal political institutions, too, was distinct and resulted in the rupture of the Conservative–Liberal hegemony of public politics in Colombia. Traditional politicians had made their own attempts of constitutional reform, always failed, and essentially jumped on the bandwagon, once the stu­ dents’ initiative gained popular traction. After the plebiscite to implement a con­ stituent body, President Elect Gaviria and traditional politicians somewhat dominated the process again. However, the basic prescriptions and demands the students made did not disappear, and crucial links between civic society and poli­ ticians in the constituent assembly were maintained. Finally, the composition of the assembly and its decision-making process—public and mostly by consensus—

62 The Novelty of the 1991 Constitution fundamentally differed from the previous constitutional reforms following regime crisis in the 20th century. Succinctly stated, the subsequent discussion will show that a heterogeneous stu­ dent movement consisting of two different organizations is the nucleus of the push for foundational constitutional reform. They were sidelined as the reform process was making its way through the formal institutions. However, political elites could never fully co-opt the process, nor exclude civic involvement entirely. Finally, the actors that debated and negotiated the wording of the new Constitution in the assembly involved groups that had been historically excluded. These debates differed from other negotiations by their publicness. The result was a very progressive poli­ tical charter, which, amongst other things, had deep effects on the normative canopy of the new Constitution termed nuevo derecho, which contrasted the viejo derecho of the old constitutional framework. Above all, the new Constitution embraced a far-reaching rights framework reflecting the initial grievance that led students to the streets: Colombia’s abhorrent human rights situation. The Aftermath of the National Front: Hyper-Fragmentation of a Regime in Crisis While the debt crisis crippled most of Latin America during the 1980s (Drake and Hershberg 2006), Colombia suffered a different tragedy, which in the words of the late Gabriel García Márquez reached biblical proportions (“un holocausto bíblico”, cited in Valencia Villa 2012, 206). Some of Colombia’s most talented politicians, investigative journalists, lawyers, and civil activists fell victim to the violence inflicted by the drug cartels, consolidating paramilitary groups, and increasingly powerful guerrilla groups. Its institutions, too, suffered tremendously from the brutality inflic­ ted by Escobar et al. The criminal justice system came to an almost complete collapse. The significance of these developments cannot be stressed enough, because the vio­ lence besieging Colombia at this point was a new violence, markedly different from the partisan violence of the previous century and mid-20th century. The actors dif­ fered and so did their potential to exert damage on Colombia’s national institutions. Accordingly, I argue that the crisis crippling Colombia, too, differed from the crises prior to the Thousand Days’ War and La Violencia. Delegitimizing and criminalizing all political contestation outside the two par­ ties in the National Front agreement opened the door for revolutionary violence and reactionary vigilantism to take root in Colombia. While the National Front’s development model brought some economic growth, its political economy opened the windows of opportunity for certain groups only, had little to offer for the lower and lower middle class that had achieved more access to education, and, as a consequence, did not automatically subordinate to elite rule (Duncan 2006, 214). The first revolutionary group formed out of Liberal armed groups fighting in rural areas during La Violencia that continued fighting and turned Castroist after the success of the Cuban Revolution. After an army attack in 1964, they constituted themselves as the Revolutionary Armed Forces of Colombia, known by the Spanish acronym FARC (Fuerzas Armadas Revolucionarias de Colombia).

The Novelty of the 1991 Constitution

63

The ELN (Ejercito de Liberaión Nacional or National Liberation Army) quickly followed the FARC in 1964, and the M-19 formed after the fraudulent elections of 1970. Most of the guerrilla activity was contained to rural areas, where land­ owners reacted to guerrilla incursions onto their territory by establishing their own vigilante forces. In 1968, the state formalized decree 3398 into Law 48, which allowed landowners to mount their own private armies thereby essentially priva­ tizing the provision of security inside its legal territory. From this moment on (until at least the peace processes with AUC paramilitaries and guerrilla), the Colombian conflict was essentially separated into three distinct parties: left-wing guerrilla groups, right-wing “self-defense” groups, and the state’s armed forces. There were intrinsic linkages between vigilante groups and the state—not least due to the vigilantes’ legal entrenchment in the law from 1968 turning them into the state’s “bastard child or parasite” that undermined the monopolization of violence from within (Gutiérrez Sanín and Barón 2006; Acemoglu et. al. 2013). Another institutional caveat that laid the foundation for the cocaine bonanza and its accompanying violence to take root was the increasing internal fragmentation of the party system. As shown, the National Front had increased the political weight of regional party bosses vis-à-vis the center. Their positions were particularly vulnerable to the effects of the drug trade, because “at both levels—smuggling and cultivating— the nexus with well-connected politicians played a key role”. Illegal merchants needed political links to protect their business “when the independence of electoral barons from the center boomed” (Gutiérrez Sanín et al. 2007, 22). It is therefore true that “illegal drug production and trafficking are associated with institutional and behavioral factors of the societies where they flourish” (Thoumi 2005, 10). In Colombia, key factors were the disenfranchisement of the popular classes as a con­ sequence of the National Front, and the resultant empowerment of regional political barons (Gutiérrez Sanín et. al. 2007, 22; Thoumi 2005; see also Duncan 2006, 237). Part of the uniqueness of the crisis preceding the 1991 Constitution is that the immense profits of the illicit economy produced entirely new actors—even a new social class in rural Colombia—that produced new crisis dynamics. Pablo Escobar stood out, because of his flamboyance and incredible wealth, but it was his founding of the MAS in 1981 (Muerto a las Secuestradores, “Death to Kidnappers”) that proved most con­ sequential. For the first time, it brought together legal forces—the rural landowning class, the security forces in rural Colombia, and local politicians—with illegal actors from the cocaine trade (Duncan 2006). At the same time, mid-ranking associates from the cartels moved into the countryside, bought land to launder their profits from the drug trade, and thereby became landowners themselves. Together, this created a het­ erogeneous group with a shared interest, because by 1982 the FARC had made the strategic decision to aggressively expand into key areas of the drug trade in rural Colombia (Gutiérrez Sanín and Barón 2006, 44). Cattle ranchers and landowners were the first to feel the effects of that offensive, but could not raise sufficient funds for their protection, while the narco groups could.14 For these narcos, their newly gained relationship with traditional landowners enabled them to utilize structures that were deeply embedded in the social fabric of rural society and thereby conveyed the aura of legitimacy. It essentially merged both into a new landowning class (Duncan 2006).15

64 The Novelty of the 1991 Constitution

2011

2009

2007

2005

2003

2001

1999

1997

1995

1993

1991

1989

1987

1985

1983

90 80 70 60 50 40 30 20 10 0 1981

Homicides per 100 00 citizens

Inten�onal homicide rate

Year Graph 2.1 Intentional homicide rate, 1981–2011

Statistics tracing intentional homicide rates for Colombia began to rise drama­ tically in the 1980s. From a “low” of just above 30 deaths per 100,000 people in 1984, average murder rates more than doubled to a figure of over 80 per 100,000 in 1991. The evident, and very dramatic, increase in the homicide rates in the second half of the 1980s (Graph 2.1) reflected two developments. Firstly, after Pablo Escobar was exposed as a narco-trafficker by Justice Minister Lara Bonilla and had to resign from legal politics in 1984, he fought the Colombian state in a terror war to evade extradition to the United States. Lara Bonilla was the first of many politicians to be assassinated in 1984. Secondly, the coalition of landowners and drug traffickers increasingly turned against all activists who threatened their social power and aimed to alter the socio-economic status quo, annihilating almost an entire political party, the Unión Patriotica (Patriotic Union; hereafter UP) that had united the political wing of the FARC with the Communist Party of Colom­ bia. At this point, Colombia’s legal system was brought to the brink of collapse. Impunity reigned high and meaningful political participation became impossible, as was evident in the fate of the UP and its presidential candidates (three were murdered in the second half of the 1980s, as were a number of its Members of Congress). It was against this background that the student movement launched its activism to implement a new constitution. The Extra-Institutional Path: The Communicative Action of the Student Movement(s) While aspects of the political framework that provided the context in which the violence spiraled to crisis proportions in the 1980s bore some similarities to pre­ vious crises (intransigence in the formal political system), the actors responsible for the violence did not. In line with the critical juncture framework, the international cocaine boom constituted an external shock that effectively created new actors—

The Novelty of the 1991 Constitution

65

social and political. The path towards the juncture—to utilize Robert Frost’s ana­ logy used by the Colliers—also bifurcated from the conventional route taken in previous instances of institutional engineering. There are two central elements to the process that resulted in the implementation of the new constitution: 1) a het­ erogeneous student movement that formed outside of Colombia’s formal institu­ tion setting the discursive framework of the constituent process; and 2) a constituent process taking its path through Colombia’s formal institutions, but outside the hegemony of the two traditional parties. Both elements, I argue, show that this juncture critically differed from the institutional junctures discussed above. Moreover, the communicative action of the student movement, and evi­ dently deliberative tone of the negotiations, suggests to me that the new con­ stitution is indeed beset with communicative power, and the process a form of discursive institution making. Reflecting the chronological order of events, I first look at the trajectory outside Colombia’s formal institutions that culminated in the consolidation of the student movement, and then turn to the development inside its formal institutions. The student movement itself was not monolithic, but consisted of two student organizations that each had a distinct organizational structure and represented different constituents within the student body. As a consequence of their distinct configuration and composition, each part contributed resources that com­ plemented one another, and played crucial parts in the overall success of the movement. The assassination of Juan Carlos Galán on 18 August 1989 was the decisive catalyst that brought students to the streets: 20,000 students from uni­ versities in the nation’s capital marched silently to the Liberal presidential can­ didate’s grave, and concluded the protest with a declaration, disavowing all violence and demanding respect for human rights. The protest note culminated in a call for a general purification of the armed forces, the police, the govern­ ment, and the political parties by convening the people to redress the institu­ tional inhabitations at the root of the current political crisis. Following the silent protest march, students gathered in workshops at Bogotá’s (elite) universities. In long debates and discussions, which greatly reduced the number of participants to a core group of dedicated students, they earmarked seven institutions and norms in most despairing need of reform to improve Colombia’s democracy: 1) Congress; 2) civil rights and social guarantees; 3) the judicial system; 4) state of siege measures; 5) economic planning mechanisms; 6) administrative decen­ tralization; and 7) mechanisms for future constitutional reform (Dugas 2001, 831). To be sure, this framework essentially picked up ideas that academics had explored before (see Leal Buitrago 1984; Valencia Villa 2012). Nevertheless, the long debates “convinced [students] of its validity” and it became the basic set of demands for the entire student movement that spawned another, more grass roots, national movement (Dugas 2001, 832). They published their first mani­ festo calling for reform in El Tiempo in October of 1989 under the name that lent the movement its name, Todavia Podemos Salvar a Colombia (“We can still save Colombia”).

66 The Novelty of the 1991 Constitution While the initial student movement was fairly elitist, the second was much more diverse: the Movimiento Estudiantil por la Constituyente (Student Movement for a Constituent). Activists in Podemos Salvar movement were mostly middle and middle-upper class, limited to small groups of law students from a few universities in Bogotá. The Movimiento Estudiantil brought students from around the country together. It was also much more representative, because it included students from large public universities as well as small private institutions, not limited to law stu­ dents. Most importantly, they reflected the entire ideological spectrum, ranging from anarchists, the Communist Youth, and militants of the former M-I9 guerrilla movement on the left, to political independents and traditional party members on the right. The Movimiento Estudiantil por la Constituyente could legitimately claim to be truly national in its composition and as a consequence organized meetings that drew thousands of students (Dugas 2001, 832). These contrasting resources helped the students to get their constitutional initia­ tive on the political agenda, gain the backing of key elite allies, and rally broad public support for it. Importantly, in the process of initiating constitutional reform, student resources proved more successful than attempts by traditional politicians to reform the political framework from within Colombia’s formal institutions. President Barco had attempted to initiate bipartisan reform of key political institutions, but failed in Congress and in the courts. He, a Liberal, first tried to sign a pact with Conservative leader Misael Pastrana to call for a referendum to reform the Constitution in 1988, which the State Council declared unconstitutional. Following this defeat, the Minis­ ter of Government and Communications (and later President), César Gaviria, aimed to redress constitutional contentions via a constituent assembly. The legislation for that assembly died in Congress (see also Cajas-Sarria 2008). At the time when Barco’s attempt failed in Congress, students advocated for voters to submit a seventh ballot voicing support for a constituent assembly—the so-called séptima papeleta (seventh ballot) campaign—in the congressional elections scheduled for March 1990. The seventh ballot initiative was backed by more than two million votes and led Barco to support the idea of a constituent process by initiating a plebiscite held in conjunction with the 1990 presidential election. Four candidates for that election also supported a constituent process. In the end, 5,236,863 Colombians voted in favor of the stu­ dents’ propositions, representing 86.6% of the votes cast (Dugas 2001, 810). The plebiscite was the movement’s greatest success and by the time the assembly actually convened for deliberation, the movement had fractioned off. Nevertheless, the point of their activism was not to control the agenda and implement a specific framework, but rather to introduce a new “process that could lead to making this Constitution” (cited in Dugas 2001, 826). Therefore, students proposed “a ‘political pact’ among representatives of all Colombians, rather than a document imposed by a limited set of victors”, as was the case with the historical constitutions of the 19th century (Dugas 2001, 830; Valencia Villa 2012). The logic of their demands was that a truly democratic and inclusive constituent assembly could open avenues of citizen participation and regain the legitimacy lost by the patronage-infested institutions of the old republic. The reformed institutional structure would then provide the func­ tional basis on which to address Colombia’s other problems (Dugas 2001, 830).

The Novelty of the 1991 Constitution

67

In sum, it is fair to say that the student movement was the result of a genuine communicative process, rooted in a heterogeneous base, from which leadership resulted organically. Initial leadership arose not from foregone indicators such as class or race—even if the original nucleus came from upper-class echelons—but from longlasting substantive debates. Hannah Arendt noted in her writings on the student movements in the United States that leadership amongst the students crystallized from communicative processes (1972). In Colombia, too, students came together around ideas and programmatic positions, and then proceeded to take action. Deed followed word. Importantly, after the initiation of a diverse movement, student acti­ vists themselves created political opportunity structures amenable to constitutional change through the seventh ballot initiative and subsequent plebiscite, rather than waiting for an opening in the formal political framework. To be sure, elite allies (such as President Virgilio Barco) helped to push their ideas. Most important, though, was that students broadcasted their own reasoning and argumentation of the framing and the solutions that they suggested. It provided a decisively public forum. In the end, Barco himself accepted the framing of the students and not the other way around. Thus, even though their influence diminished after the plebiscite, their action created the valid framework of reference for the debates in the constituent assembly. The Institutional Path: Breaking the Conservative-Liberal Hegemony By the time the Barco administration in May 1990 took up the path for constitutional reform proposed by the student movements, the formal institutional path paved with (elite) political support was essentially dead (Cajas-Sarria 2008, 86). The constitu­ tional reform process needed the extra-institutional support from a genuine—and popular—grass roots movement. I argue here that the process that followed the “seventh ballot” initiative was vested with authentic constituent power, as the con­ stitutional chamber of the Supreme Court confirmed in two ground-breaking deci­ sions. The selection process of the assembly then broke the Conservative–Liberal hegemony of politics that had dominated the previous institutional engineering pro­ cesses. These are two crucial caveats that again reinforce my contention that this constituent process was indeed a critical juncture. After the successful “seventh ballot” initiative, the constitutional reform process took a turn into the limitless domain of extra-legal constituent power. Barco decided to implement the plebiscite by utilizing decree powers to initiate the constitutionmaking process. Invoking the special powers granted to the executive by the state of siege in Art. 121 of the 1886 Constitution, he issued a decree that a plebiscite would be held along the presidential elections (Decree 927 of 1990). Resistance came from the Attorney General of the Nation, Alfonso Gómez Méndez, who lamented the lack of limitations placed upon a constituent assembly. Gómez Méndez warned that it gave too few people too many powers to decide on fundamental issues concerning the organization of law in Colombia. Even Carlos Lleras Restrepo, who himself had attempted constitutional reform during his presidency and supported the students, warned against the extra-judicial origin of the new constitution. For Humberto de la Calle, it had the taste of a constitutional coup (Ahumada 1995, 3–5). Yet, to most

68 The Novelty of the 1991 Constitution observers’ surprise, the Supreme Court affirmed the decree in a split decision and argued that the re-establishment of public order, which justified the proclamation of the state of internal commotion, i.e. state of emergency, in the Constitution, did not solely rely on repressive measures. If reforms had the explicit and objective goal of assuaging the problems underlying the situation of internal commotion, they were also covered by state of emergency clauses (the vote in the Court was 14–12; see also Cajas-Sarria 2008, 88).16 The Supreme Court further ensured the truly constituent character of the assembly, after the successful plebiscite and the election of César Gaviria as pre­ sident. The latter invited the two bosses of the Conservative Party and leaders from the demobilized guerrilla M-19 group to decide on the continuity of the process. They selected 9 December 1990 as the date of the election for the constituent assembly and agreed upon a number of issues that the assembly should consider for reform (Congress, the judiciary, the attorney general’s office, public administration, human rights, political parties and opposition rights, departmental and municipal regimes, mechanisms of participation, state of siege and economic matters). Pre­ sident Gaviria set the date of the election and the number of elected constituents, together with a thematic program of the constituent process, in Decree 1926 in August 1990. The Supreme Court, though, denied the constitutionality of the clause outlining programmatic limitations upon the constituent assembly itself. It argued that a national constituent assembly could not have restrictions concerning its agenda and regulations of deliberation, since its mandate comes from the sover­ eign people. It thereby transformed the constitution-making body into a true con­ stituent body that is not submitted to the powers of the existing constitutional order (Echeverri Uruburu 2012, 457; Cajas-Sarria 2008, 90).17 The Supreme Court’s decision is important for the argument put forward here, because it undermined elite control and ensured that citizen control of the agenda was not limited to the act of voting for the assembly. To ensure citizen participation, workshops were held in more than 1,500 locations in the entire country prior to the elections to the constituent assembly. These workshops, open to all citizens, produced 350,000 projects of consideration, which the delegates in the constituent assembly were obligated to debate. Aside from citizen participation in the agenda setting phase of the constituent assembly, the elections for the constituent assembly also undermined the traditional Liberal–Conservative hegemony of the political process. It had a very poor turnout: 84% abstained from their right to vote. The nature of the constituent assembly, which decided on constitutional norms and not the distribution of public goods, might have disincentivized traditional politicians to mobilize support for their parties (Echeverri Uruburu 2012, 458). They had done so in the congressional elections held only a couple months before, when over 50% of eligible voters went to the poles and Liberals and Conservatives received close to 94% of the vote share. Now only 45 of the 70 delegates came from the two traditional parties (64%) and, most notably, 19 came from the demobilized M-19 guerrilla group (see Graph 2.2). This was a much more diverse assembly than previous national legislatures.

The Novelty of the 1991 Constitution

69

Graph 2.2 Composition of constituent assembly 1991

In its result, the assembly united traditional politicians, representatives of ex-guer­ rillas, Christians, intellectuals, and indigenous peoples for the first time in Colombia’s history in an institution-making setting. In addition, none of the forces represented had a majority in the assembly, which induced a collegial atmosphere (Negretto 2013). The presidency over the assembly rotated between the Liberal, Conservative, and M-19 faction. The minority representatives received presidencies and vice-pre­ sidencies of the five commissions that studied the topics of debates: 1) fundamental rights; 2) territorial organization; 3) the organization of the legislative, executive branch, as well as international relations; 4) the organization of justice; and 5) socio­ economic issues. Notably, the presidency over the fourth commission was given to Fernando Carrillo Flórez of the student movement and the vice-presidency to Jaime Fajardo Landaeta of the demobilized guerrilla group, EPL. The decisions eventually taken by the assembly also reflected the collegial character of the constituent body: 74% of the norms approved in the Constitution were decided unanimously. Of those decided by majority vote, the alliances shifted between the various factions, meaning that the two traditional parties did not exercise its past hegemony either. Thus, the traditional Liberal–Conservative hegemony was no longer effective (Echeverri Uru­ buru 2012, 461) and the incorporation of new actors—as well as novel (transparent) procedures of debates that incorporated citizens’ opinions—were reflected in the constitutional charter produced in the assembly. Institutional Novelties of the 1991 Constitution Constituent power, or poder constituyente primario (primary constituent power) as it is referred to in Colombia (Valencia Villa 2012, 205), summoned in the

70 The Novelty of the 1991 Constitution constituent assembly and produced an entirely re-written constitution. The new Constitution’s norms created a wholly novel institutional structure embedded in a strong normative framework, which, for the first time in Colombia’s history, included a vast catalogue of rights at the ranks of constitutional norms. Even some critics of historical constitutionalism in Colombia noted that the novel Constitu­ tion was reflective of the democratic moment at its origin, not adhering to an ideologically closed worldview imposed by the victor of a conflict, but rather a document of compromise and reconciliation that embraced liberal, conservative, nationalistic, internationalistic, and, above all, social democratic ideas (Valencia Villa 2012, 205–212). In short, it was markedly different from previous constitu­ tions and institutional frameworks. It was, and still is, a neo-constitutionalist document (Bonilla Maldonado 2013; Hailbronner 2017). The longer-term legacy effects of the 1991 Constitution are the foci of the next chapters. I will, however, address nuevo derecho that was imprinted into the constitutional framework to argue that the legacy of this juncture fundamentally differs from the junctures discussed in the previous sections of this chapter. The three branches of government of the state (executive, legislature, and judiciary) were defined as public powers. In addition, the Constitution acknowl­ edged two public powers independent and autonomous from those three: the procuraduría (Inspector General) overseeing the conduct in the exercise of public offices, and the controlería (Controller General) overseeing the management of public fiscal resources. These two offices were not designed as judicial institutions but administrative powers. The legislature maintained its traditional bicameral organization, divided into the House of Representatives and the Senate, but with different electoral systems for each chamber of Congress: Regional circumscription for the House of Representatives and national circumscription and the D’Hondt method for the Senate. The latter norm was incorporated in order to set incentives for national campaigns rather than regionally concentrated cam­ paigns that had benefited the utilization of local patronage systems. The role of the legislator in the law-making process remained similar. The laws to be passed by Congress were divided into a tripartite taxonomy: ordinary laws for fiscal matters that had to be approved by both chambers and the government (essentially giving the executive a veto power), organic laws concerning pro­ cedural regulations in Congress and territorial competences in the country, which required absolute majorities in both chambers; and statutory laws for legislations involving fundamental rights of persons, the administration of justice, organization of political parties as well as electoral functions, citizens’ participatory mechanisms, and states of exceptions, which also require abso­ lute majorities in Congress and approval by the Constitutional Court in an automatic constitutional review. A novelty of the 1991 Constitution was the disciplinary code imposed on Members of Congress. Since corruption in the legislature was a grave concern for constituents in the assembly, they imposed an extensive code of proscriptions on legislators that could cost them their investiture (if found guilty by the State

The Novelty of the 1991 Constitution

71

Council). In addition, the new Constitution delegated the imperative to handle criminal investigations against Members of Congress to the Supreme Court. Thus, not Congress but the judiciary had the powers to impose disciplinary and criminal sanctions on members of the legislative branch, effectively divesting it of legislative immunity. This limitation on legislative immunity became a major factor in the so-called parapolítica affair during the second Uribe term when over a hundred Members of Congress were investigated for ties with illegal armed groups (Echeverri Uruburu 2012, 478–481). The President remained the head of the executive under the new Constitution. There were never earnest debates of switching to a parliamentary system (possibly due to the role of President Gaviria in the process leading up to the constituent assembly; see Negretto 2013). Executive powers were to some extent limited by new norms, but still vast. It lost the power to nominate governors and control monetary policy in Colombia, which was delegated to an independent body, the Junta del Banco de la Repúplica. In addition, the Constitution curtailed executive prerogatives most substantially on issues concerning extra-ordinary powers of the state. Declarations of states of exception were strictly formalized, could only last 90 days, and, above all, required confirmation by the Senate and the Constitu­ tional Court. During a state of emergency, no permanent legislation could be passed, nor could the exception suspend the law of war (Article 212–215). The new Constitution disallowed re-election of a President once he has taken office, altering the constraints of the previous Constitution, which only disallowed immediate re-election. This had effectively given ex-presidents a lot of power after their term had ended, because they could threaten to run again. Evidently, Uribe’s constitutional reform in 2004 changed this until it was changed back in 2015 (Echeverri Uruburu 2012, 482). In terms of executive-legislative relations, the presidency retained its powers to nominate the government, delegate funds for socio-economic programs, initiate and close each session of Congress, promulgate laws, and, at the beginning of each session, present administrative projects for the entirety of the legislative year (Article 188–199). When students initially took to the streets of Bogotá, they were protesting the abhorrent human rights situation in Colombia. It is therefore not surprising that the judiciary received the most far-reaching overhaul. The 1991 Constitution created new courts (the High Council of the Judiciary, the Constitutional Court; see Table 2.1), a new attorney general’s office inside the judiciary (Fiscalía Gen­ eral), and new mechanisms to enforce rights (tutela). The Constitutional Court became the final instance of abstract judicial review of laws, concrete rights review, and judicial review of constitutional reforms. Judicial review of constitutional reforms was limited to procedural defects and excluded powers to test the material substance of a constitutional reform (Article 241, 1–3). The last norm was the point of contention in the substitution theory that I will discuss in the chapter on the empowerment of the Constitutional Court. The Supreme Court remained the highest court in the ordinary justice system and the State Council the highest court of the contentious administrative jurisdiction (adjudicating complaints against public servants). The High Council of the Judiciary fulfilled administrative

72 The Novelty of the 1991 Constitution Table 2.1 Colombia’s high courts in the new constitutional regime Constitutional Court

Supreme Court

State Council (Consejo del Estado)

Superior Council of the Judiciary (Consejo Superior de la Judicatura)

9 judges

23 judges

31 judges

13 judges

3 chambers: Plenary chamber (Sala Plena) for constitutionality review (all 9 judges); tutela selection chamber (3 judges); tutela review chamber (3 judges)

3 chambers: criminal law (9 judges); civil law matter (7 judges); labor law (7 judges)

2 chambers: contentious administrative chamber (27 judges); consulting cham­ ber (4)

2 chambers: administrative chamber (6 judges); disciplinary cham­ ber (7 judges)

Abstract and concrete constitutional review; constitutional review of con­ stitutional reforms

Highest appel­ late court in criminal, civil, labor law; investigates Members of Congress

Highest court concerning the contentious administrative jur­ isdiction; consulting cham­ ber for the government

Administration of the judiciary and disciplinary actions against functionaries of the judiciary

Source: Compiled by the author

and disciplinary tasks within the judiciary. To provide a quick overview, Table 2.1 lists all high courts of the 1991 Constitution, their respective compositions, func­ tions, and number of chambers. The results of civic activism—above all by the student movement—were not only identifiable in the new formal political structure and overhaul of Colombia’s judicial institutions, but also readily apparent in the strong normative connota­ tions. The 1991 Constitution defined Colombia as a social state of law and stipu­ lated a vast rights catalogue at the level of constitutional norms. The inclusion of rights and the shift to the definition of Colombia as social state of law manifested itself in the shift away from the viejo derecho of the 1886 Constitution to the nuevo derecho of the 1991 Constitution. As such, nuevo derecho is a key component of the legacy of the 1991 Constitution and important for understanding the devel­ opments in the so-called choque de trenes and substitution doctrine analyzed in Chapters 3 and 4, respectively.18 Viejo derecho attempted to emulate Montesquieu’s vision of the judge as the mouthpiece of the law. This had three interdependent consequences: 1) the most important principle by which judges interpreted the constitutionality of laws was the principle of legality; 2) precedents, if any, had only an extremely miniscule role in the jurisprudence of the Supreme Court; and 3) rights were subsumed under the civil code and had no constitutional standing on their own.19 The strict formalism in legal questions evolved around questions of procedure and compe­ tence and not about substantive contentions, favored an adherence to order, the

The Novelty of the 1991 Constitution

73

subjection of rights to legality and, ultimately, were reflective of the generic ambiguity in Colombia’s history: a bipolarity between legality and illegality, democracy and clientelism, peace and violence.20 The focus on legal forms and vocabulary fueled a high confidence in the transformative capacity of law, but at the same time created a veil of ignorance for real socio-political grievances. Tra­ ditional constitutionalism in Colombia, thus, was always an exercise in “changing everything, so nothing changes”. For that reason, students not only lamented the corruption of Congress, but also the immobilism of the Constitution itself. In their analysis, both produced Colombia’s intractable position at the end of that decade, which required change.21 The novelty of the 1991 Constitution becomes readily apparent in the first Article. While the 1886 Constitution proclaimed a unitary character that arose from the sovereignty of the nation (Article 1 and 2), without stipulating what this nation consisted of, the 1991 Constitution principally defined Colombia as a social state of law whose primary task was to protect the human dignity of its pluralistic society, conjoined in solidarity to strive for the general good (Article 1). As we will see in the decision concerning re-election, pluralism, solidarity, and human dignity were not just fashionable phrases, but concepts that the Constitutional Court filled with meaning in its jurisprudence (see Chapter 4). Nuevo derecho contended that the protection of inalienable rights requires direct interpretation to alleviate grievances existing in the lifeworld of rights claimants—another break with the formal positivism of viejo derecho. 22 The new Constitution gave expression to this direct interpretation of rights with an expansive rights catalogue, which included not only private and social rights, but also collective cultural rights, and rights-enforcing mechanisms.23 The inclusion of cultural rights was a particularly groundbreaking result, as it reflected the indigenous and Afro-Colombian heritage, and foreshadowed developments in other South American states by more than a decade. Again, this can be traced to the plurality of the constituent assembly, which gave observer status to an indi­ genous guerrilla group: the Movimiento Armado Quintin Lame. One Constitu­ tional Court judge said that the 1991 Constitution was a peace pact. “While one sector of the guerrilla, the FARC, did not participate. Others did, including indigenous guerrilla groups and other small guerrilla groups.”24 Others confirmed: the new Constitution contributed a great deal towards making those previously marginalized groups visible in the Colombian public, giving them a voice of their own. To practically enforce rights, the new Constitution added the acción popular de tutela (Article 86) to the already existing acción popular de inconstitucionalidad (Article 241). The tutela gave every citizen the right to submit a constitutional complaint against violations of their rights with any judge in the country to be deci­ ded within ten days before being passed on to the Constitutional Court for review. The Constitutional Court received the power to select cases for its docket. For the most part, tutelas have, inter alia, character, but, as they accumulate, gain erga omnes characteristics. The normative grounding, and politics, of the tutela will be examined in more detail in Chapter 3. Here it suffices to point out that they are part of the Court’s jurisprudence to directly interpret rights.

74 The Novelty of the 1991 Constitution The centrality of rights and pluralism in the identity of the new Constitution also affected viejo derecho’s most central principle: the principle of legality, part of the “disposition to have legislated right”.25 The constitutionalism of nuevo derecho has not delegated it to futility. For example, Article 230 of the Constitution states that judges are exclusively accountable to the empire of law, upholding the importance and centrality of the written text of the Constitution. Nevertheless, nuevo derecho added tests for the proportionality, rationality, and efficacy to the constitutional review process.26 The proportionality principle, a staple of modern constitutionalism to weigh competing claims that have arisen with the co-equal proliferation of rights and increasing number of state functions (Kumm 2010), was critical for turning the Constitution from a “written Constitution to a living Constitution”. It incorpo­ rated legal context into constitutional adjudication.27 The valuation of the legal context not only placed a much higher value on precedents than the positivism implied in viejo derecho; it also entails that derecho substantial (substantial law/ right) has precedence over derecho formal (formal law/right). In sum, pluralism, rights, and new principles of interpretation meant that the new Constitution valued rights not as functioning outside of the legal context but placed them in the circumstances of human existence. It deviates decisively from the textu­ alism of the old constitutionalism. One Constitutional Court judge put it nicely: The interpretation of law has to be an interpretation that is teleological and takes into account the value of justice, equality, and the value of due process. In the Constitution, the norm acts in an integral manner. It is much more intended as a way to resolve problems.28 The new Constitution shifted the role of the Constitutional Court judge from a dogmatic interpreter of laws and norms to a pragmatic problem solver, mediating between norms that existed on paper and actors that functioned in real life. This was arguably the farthest-reaching consequence of the 1991 Constitution that broke the ground for a number of new constitutions and courts that, in sub­ sequent years, embraced the rights affirming constitutionalism of the Colombian Constitutional Court (for example, South Africa and India; Bonilla Maldonado 2013; Hailbronner 2017). The 1991 Juncture: The Atomization of the Nation The prior discussions on the legacy trajectories of the National Front and the reforms at the beginning of the 19th century showed that the new institutional frameworks had immediate effects. Most notably, violence subsided almost entirely, but clientelistic relations also stabilized within the formal two-party system. The aftermath of the 1991 Constitution is more complex and displayed contradictory developments resultant from the institutional juncture and the effects of the (still) burgeoning drug economy. Violence did not subside immedi­ ately, but, fueled by the drug economy, continued unabated. It created new vio­ lent actors (AUC) and strengthened existing ones (FARC). Arguably, the state’s

The Novelty of the 1991 Constitution

75

capacities further contracted resulting in a situation in which Colombia’s democ­ racy was more open and inclusive, but also besieged by violent, non-democratic forces in the periphery of the country that extended their reach into the national political institutions (Bejarano and Pizarro 2005). Patronage systems, which prior to the 1991 Constitution functioned from within the traditional party system, now slowly morphed into informal relations that functioned from outside of that system. This created a new, even more vicious and cynical form of clientelism: armed clientelism. It built on the internal fragmentation and eventual collapse of the (unintended) traditional party system. The national circumscription of the Senate vote provided incentives for regional politicians to seek alliances with actors that could generate the votes in exchange for legislative support. Politicians essentially became electoral entrepreneurs selling their services to the highest bidder. Regional warlords from the AUC were quite capable and willing to gen­ erate these votes through “violent campaigning”. As a consequence, large parts of the legislature were co-opted by groups at the margin of legality. These evolutions not only formed the backdrop of the conflicts between President Uribe and the courts that unfolded in the early 2000s, but also showed that decisions taken in the 1991 juncture had longer-lasting, albeit contradictory, legacy effects. The political evolutions following the implementation of the new Constitution must be viewed with consideration to changes in the political economy of the drug trade. The chase of Pablo Escobar propelled different actors from within the drug industry to the summit of the business and made them independent of the big capos at the head of the cartels. The eventual heads of the AUC were enforcers, who had worked for the cartels to make contracts obligatory and eliminate rivals (Gutiérrez Sanín et al. 2007; Duncan 2006). Their expertise in the application of violence became the critical asset once the epoch of the big cartels was coming to an end, and provided them with a competitive advantage once the effects of the new Con­ stitution started applying to the electoral and party system. When the state started to turn on the capos of the eponymous cartels—first Escobar of the Medellin, and then the Rodriguez brothers of the Cali Cartel in the course of the proceso 8000 (see Chapter 3)—the remaining bosses increasingly relied on protection from those midranking enforcers that had moved into the countryside and gained territorial control there (Duncan 2006, 236–237; Gutiérrez Sanín 2007, 497). This commenced a process referred to as traquetización: the utilization of violence became the central competitive advantage in the drug business to not only enforce rules but protect it politically.29 Eventually, when the capos looked to the paramilitary groups for pro­ tection from the state, they had no choice but to relinquish control over the business. Legal exposure turned their source of power, control over the entire production chain of the drug trade and its resources, into a liability, while paramilitaries’ terri­ torial control in rural Colombia mitigated legal exposure. Consequently, these deals of protection trapped the capos in contractual obligations where the other side had no incentives to comply, turning them into dependents of those warlords (Duncan 2006, 234; Ávila Martínez 2010, 109ff). After the implementation of the new Constitution, Colombian governments continued the failed policy of privatizing security to vigilante groups. The Escobar

76 The Novelty of the 1991 Constitution hunt had deepened informal webs of cooperation between sections of the security apparatus with these emerging armed groups on the extreme right.30 The con­ tinued strength of the FARC—particularly in rural Colombia—led Governors of Colombia’s departments to press for a policy to legalize private security groups, so-called CONVIVIRs (Cooperativas de Vigilancia y Seguridad Privada; Vigilante Cooperatives and Private Security). One of the most vocal supporters was the Governor of Antioquia: Álvaro Uribe Vélez (Ávila Martínez 2010, 112). For the paramilitaries, CONVIVIR was the perfect vehicle to expand systematically into Córdoba, Antioquia, and the South of Santander, and expel the FARC from important drug trafficking zones in Colombia. From CONVIVIR groups, paramilitaries developed into a proper political pro­ ject. International pressure over the human rights abuses committed by CON­ VIVIRs31 forced the Samper government to shut down the program. The new government of Andrés Pastrana commenced the ultimately failed peace process with the FARC in La Caguan. As a consequence of this rapprochement between national political elites and the leadership of the FARC, regional political elites feared the legalization of the FARC could endanger their social power that was based on the control of land in their regions. They wanted to continue the coun­ terinsurgency policy and needed a new legitimizing mechanism (Ávila Martínez 2010, 118; Duncan 2006, 261; Romero 2000). Carlos Castaño and other heads of paramilitary groups in the country, such as Salvatore Mancuso and Rodrigo Tovar Pupo, alias “Jorge 40”, commenced the process of uniting the various paramilitary groups under an umbrella organization. This was the beginning of the AUC. The Castaños’ ACCU (Autodefensas Campesinas de Córdoba; Peasant Self-Defense Forces of Córdoba) as the largest group assumed the natural leadership role within the AUC, but every other group within this confederation retained its coercive capacities. There are three important and distinct changes between the AUC and the various predecessor organizations: 1) for the first time the AUC proposed a coordinated expansion plan that aimed at the entire national territory; 2) it intro­ duced internal organizational statutes, including different military fronts; and 3) Carlos Castaño defended their violence as a response to the absence of the state and the presence of the guerrilla. Significantly, in their entire public relations campaign, they positioned themselves on the side of the state and democracy. Constituted as the AUC, the paramilitaries and their political allies in Colombia’s regions wanted to complete the political project, which resulted in the effective parainstitutionalization of Colombia’s political regime: the co-optation of its democratic institutions—above all the national Congress, but not limited to it—by illegal armed groups at the margins of legality. In 2001, they sat down in Santa Fe de Ralito and signed a contract to “refound the nation” (López Hernández 2010).32 This constituted the preliminary conclusion of the paramilitary project, which “united a good sector of Colombian society: ganaderos (agro-industrialists), industrialists, narco-traffickers, militaries, some national politicians, and the entire regions and locales where they had dominated before” (Ávila Martínez 2010, 120). The national circumscription for the election of the Senate threatened elites’ regionally clustered power bases by potentially opening the doors for policy

The Novelty of the 1991 Constitution

77

alternatives in those regions. Paramilitaries, for their part, pursued a strategy of counter-agrarian reform, and systematically displaced large populations from frontier lands. They robbed the peasants’ lands and repopulated them with their own cli­ entele. For them, a central goal of the political paramilitary project was the legali­ zation of that agrarian counter-reform (López Hernández 2010). The solution to each side’s problem was “armed political campaigning”: it provided paramilitaries with key political allies and plausible claims to legitimacy by positioning themselves on the side of the democratic state to protect their riches and create a more legit­ imate aura for their project. Political elites benefited in that they could fend off other political adversaries and protect themselves from the FARC (Ávila Martínez 2010, 117). The politicians–paramilitary coalition was a functional coalition, but one with profound consequences. In essence, it shielded political elites from the electoral changes of the 1991 Constitution (Rodríguez-Raga 2002, 222–225). In sum, there were two coinciding evolutions following the implementation of the 1991 Constitution. It marked the commencement of the third wave of paramiltiarism in Colombia, which brought politicians (of all levels) and groups at the margin of legality closer together.33 Paramilitaries’ political clout grew as they morphed from CONVIVIR groups to the formation of the AUC. The evolution of the third wave of paramilitarism functioned in tandem with the atomization of Colombia’s parties and complete fragmentation of the two-party system into electoral entrepreneurs. These entrepreneurs did not rely on the party label and formed small political movements that garnered regionally concentrated votes. Paramilitaries created alliances with these politicians to facilitate their election in exchange for legislative support as well as reduced criminal exposure. As Gutiérrez Sanín documented, the collapse of the tra­ ditional party system was paramount for the criminalization of Colombian politics. Under the stability of the old, albeit fragmented, two-party system, criminal influence was contained to the regions; now it could extend its tentacles, via Congress, to the national level (Gutiérrez Sanín 2007). The dependencies between both electoral entrepreneurs and illegal armed groups created an equilibrium disadvantageous to the proliferation of public good (in particular security) and the rule of law (Boesten 2014). When Álvaro Uribe was elected to the presidency in 2002, the internal frag­ mentation of the traditional party became openly evident as the congressional elec­ tion propelled 45 different parties into Congress (Pachón and Hoskin 2011).

Conclusion: The 1991 Constitution as a Critical Juncture in Colombia’s History This chapter set out to investigate if the 1991 constitutional moment constitutes one of those instances in political history that places the patterns of political life on a distinctly new and consequent path: a critical juncture. I want to answer this question in the affirmative and add the claim that the 1991 constitutional moment was an instance of discursive institution making, following a crisis that exposed a cleavage between the normative claim of a democracy and the reality of political regime unable to uphold minimal channels of political participation, which ulti­ mately resulted in a neo-constitutionalist political charter.

78 The Novelty of the 1991 Constitution I built on the Colliers’ definition of critical junctures as “periods of significant change, which typically occur in distinct ways in different countries or in other units of analysis” and result in long(er)-lasting, distinct legacies (2002, 29). These junctures are distinguished from other important historical periods by three com­ ponents: their preceding conditions, the moment of crisis, and the legacies left by the decisions taken during the critical juncture to overcome the crisis. Since this inquiry concentrated on one specific case, Colombia, and a constitu­ tional decision that was taken in a specific constitutional and political context, I had to increase the observational data points by turning inward and longitudinally analyze instances of institution building in Colombia’s history to identify variations in the outcomes. These had to share a basic, ideal-type, trajectory (Figure 2.2).

Political Crisis and Violence

Institutional Engineering Process

New Institutional Framework

Figure 2.2 Ideal-type trajectory of sub-cases of institution building in Colombia (1905– 1910; 1953–1957; and 1991)

The Thousand Days’ War (1899–1902) ushered in the Reyes Reforms, La Violencía (1948–1957) the National Front Pacts, and the drug violence of the 1980s the 1991 Constitution. In order to gauge the effects of each juncture on Colombia’s political system, I differentiated between several sub-systems and held them constant in the analysis of each observational point. Figure 2.3 helps us to summarize the analysis. The reforms associated with General Reyes and the National Front agreement essentially evolved along the lower ideal-type trajectory (colored in black arrows), while the 1991 constituent process followed the top ideal-type trajectory (colored in grey arrows). To recap, this chapter first detailed the base-line conditions by exploring the fundamentals of Colombian institutionalism that are rooted in the 19th century postindependence epoch. Historically, the state exercised only a very deficient monopoly of violence. The weak distribution of public goods—above all security—was rooted in the intertwinement of state infrastructure with patronage systems during the genesis of the “two-party-state”. Regional caudillos built local power strongholds by tying a followership to their domus and coordinated with each other through the party labels of Colombia’s traditional parties. Two features reflected the clientelistic disposition of the Colombian polity. Firstly, parties were cross-sectional in leadership and following. Socio-economic elites dominated the leadership in both parties, while lower classes populated the followers, reflecting the membership of their patron. Secondly, the nor­ mative expectations, which in the context of political institutions that left very little space for meaningful socio-political change must result in disappointment, were restrained through a specific legal discourse. It became known as viejo derecho and is most centrally focused on a legalistic formalism in the interpretation of constitutional norms.

The Novelty of the 1991 Constitution Antecedent Conditions

•

Crisis

Contradictions amongst elites compounded by exogenous shock creates normative cleavage

Juncture

Institutional Result

Rights-based constitution

•

79

Legacy

• No Peace • Collapse of two-party state

Public negotiation amongst new actors

• Armed clientelism • Nuevo Derecho

•

•

Clientelism paired with democratic institutions Cross sectional parties

• Peaceful intraelite relations •

• Economic development

Protected “gentlemen” negotiations

Consociational power sharing

• Institutional immobilism from internal contradictions • Viejo Derecho

Figure 2.3 The 1991 constitutional moment as critical juncture

Against these base-line conditions, this chapter revealed that reformers in the first two instances responded to crises that were not rooted in deep social or cultural cleavages, but evolved from within the political elite. In essence, these crises con­ sisted of the inability of elites to compromise on deals that satisfied the entire range of either party’s membership and leadership. In both instances, high levels of inter­ nal violence had signaled to elites that the original political arrangement was insuf­ ficient in containing the clashes of interest and identity between them. This resulted, in both cases, in an elite-driven and controlled process of institutional reform, which eventually established a new arrangement, giving each party significant say in the path future politics were to take. Eventually, these arrangements, too, were insuffi­ cient in containing pressures on the political regime, opening a new round of vio­ lence and, eventually, institutional engineering. Most importantly, reforms in 1910 and 1957 conserved forms of clientelism by placing them on a more stable formal institutional setting without fundamentally altering the underlying socio-political interaction. What was lacking was an institutional reform that opened egalitarian paths for participation, socio-economic and socio-political involvement, which could have resulted in sustainable growth and development. The implemented changes of the 1991 Constitution do not only go the furthest in altering the formal political framework; the crisis preceding the juncture resul­ ted from profound insecurity that laid open what I call a discursive cleavage: the incongruity between the normative claim of a democracy and the political reality of a system besieged by violence and incapable of providing channels for political participation. The crisis resulted in a moment that would transform the founda­ tions of political interaction. Importantly, it could neither be predicted nor

80 The Novelty of the 1991 Constitution explained on the basis of prior conditions but resulted from students sponta­ neously coming together for a silent march to mark the assassination of a political leader. This initiated a process of communicative action, during which students first debated in small settings in Bogotá’s universities, subsequently joined ranks with students from the entire country, and then started to publicly and forcefully organize a push for foundational constitutional reform. Thus, to continue with Robert Frost’s crossroad analogy, the movement to implement the constitutional reform took the extra-institutional path rather than the road through Colombia’s formal institutions. The unprecedented trajectory from cleavage and crisis continued through to the actual juncture: the constituent assembly that convened in 1991. Its composition featured a number of actors that had been historically excluded from politics: the leftist party movement founded amongst survivors of the M-19 guerrilla group and representatives of indigenous groups. In contrast to negotiations during the previous instances of institutional engineering, the Conservative–Liberal hege­ mony of formal public affairs was broken. This much more diverse spectrum of actors involved in the public negotiations had a clear impact on the final product, the text of the 1991 Constitution, which included an expansive rights catalogue and novel mechanisms for their enforcement. Gauging the immediate legacy effects of the constitutional transformations in the charter revealed that the decisions taken in the assembly indeed resulted in very profound transformations. The clientelistic relations engulfing Colombia’s formal institutions slowly moved from the inside to the outside of the traditional two parties, taking the party system as it was known down with it. Politicians became electoral entrepreneurs, who (particularly in rural areas) deliberately aligned with forces at the margin of legality in order to reap electoral triumphs. Those armed groups at the margin of legality benefited by gaining the aura legitimacy required to shield them from legal prosecution. In the middle of these transformations stands the shift towards an electoral system that was intended to boost the national circumscription of the Senate and the D’Hondt system. Con­ trary to expectations, regional politicians succeeded in pooling regional votes and sought support from armed campaigners that organized the sufficient electoral threshold in areas under their control. The most profound changes, without a doubt, concerned the judicial system. Two new formal bodies were introduced (the Constitutional Court and the Superior Council of the Judiciary), three different types of rights were introduced to the level of constitutional norm, and a novel mechanism (tutela) was imple­ mented to enforce rights. All of these novelties reflected students’ demands and grievances arising from the human rights situation in Colombia. From the abundance of extrajudicial killings perpetrated against political actors, human rights activists, and union leaders, it follows naturally to situate rights more centrally in the constitutional framework and implement mechanisms that aim at their protection and affirmation. In short, the implemented changes best reflect the movement’s agency. To appreciate the contribution here, we need to recall that Vivien Schmidt long criticized that neo-institutionalists do not fully

The Novelty of the 1991 Constitution

81

appreciate the centrality of ideas, even though politicians’ main task in democ­ racies is to communicate ideas and thereby legitimate their actions. The argu­ ment put forward in this chapter is congruent with her proposed shift to a deliberation-focused institutionalism that centers on the volatile relationship between word and deed (Schmidt 2008, 304). The next question then is how these formal institutional novelties translated into trajectories after the implementation of the Constitution. As explained in the methodological discussion of this chapter, critical junctures are intrinsically tied up with path-dependent development that, as is theorized by historical institutionalist accounts, follows as post-genesis mechanisms of reproduction. Since rights viola­ tions constituted the most important grievance that legitimized calls for renewal, judicial institutions received the most comprehensive overhaul. As shown in this chapter, we could already identify a shift in interpretive principles for constitu­ tional litigation. Consequently, path-dependent development should be most visible in exactly that area of rights adjudication. Since the tutela was the central mechanism to enforce rights and give them practical significance, jurisprudence of the Constitutional Court on the applicability of the tutela provides a most valuable opportunity to test the path-dependent logic of institutional development. This is the task of the next chapter.

Notes 1 In addition, Colombia’s moment of constitution building that deepened and fostered a new, more participatory kind of democracy, coincided with profound transformations in Eastern Europe and authoritarian South America that Huntington summarized as the third wave of democratization. It was therefore a moment of democratic deepening that occurred in other places as well—albeit in categorically different political contexts (Kline 1996; Huntington 1991). 2 Antonio Nariño was the most progressive of the three founding fathers of independent Colombia (the other two were Simon Bolívar and Francisco de Paula Santander). Having translated, and then reprinted, the Declaration of the Rights of Man from French into Spanish, Nariño is considered the man of rights. His discourse before the Congress of Cucuta outlined a constitutional project, which, ahead of his time, involved not only basic rights, but also a declaration of rights of people, and a special tribune to guard the constitution. It was rejected, but we can identify in the speech the outline of institutional designs that have an affinity with innovations of the 1991 Constitution. 3 Osterling counts 28 wars and 63 mini or regional civil wars. While the numbers may vary, the resulting analyses are in agreement that the abundance of internal conflict is testament to a weak institutionalization of the nation state (Osterling 1989, 46–47). 4 Bergquist is not the only historian who makes this observation. Bushnell, too, notes “the relative immunity of Colombian presidents to violent overthrow, despite a political culture marked by fairly high levels of violence, is often overlooked amid the emphasis placed on the violence itself” (Bushnell 1992, 18). Similarly, Malcom Deas noted that conflicts usually commenced as political conflicts and then “much else entered in: rob­ bery, banditry, land-grabs, acts of private vengeance, even Marxist revolution” (Deas 1997, 366). 5 Amongst the founding fathers of the Conservative Party were leaders such as Márquez and Mariano Ospina Rodríguez, who had supported Santander after his break from Bolívar and even took part in the foiled plot to assassinate Bolívar (Bushnell 1992, 22–24).

82 The Novelty of the 1991 Constitution 6 7 8 9 10

11 12

13

14 15

16 17 18 19 20 21 22 23 24 25 26 27 28 29

Interview with Constitutional Court judge, 8 March 2013.

Interview with Constitutional Court judge, 16 May 2013.

Interview with Constitutional Court judge, 17 May 2013.

Interview with constitutional scholar, 6 May 2013.

Rojas later returned to Colombia, formed the populist ANAPO movement, and ran in the 1970 presidential election. Supporters of Rojas claimed that only electoral fraud ensured the National Front candidate, Misael Pastrana, to win the competition, and dubbed it the “theft of the elections”. The M-19 guerrilla movement traced its origin to this episode. See Article 2 of the Pact of Sitges, cited in Plazas Vega (2012). Notably, it was from this moment on that the Supreme Court, which later incorporated a Sala specifically tasked with review of constitutional reform and the constitutionality of laws (Sala Constitucional), played an increasing role in preventing constitutional reform (for better or ill; see Cajas-Sarria 2008). They argue that inclusive institutions allow and encourage participation by the great masses of people in economic activities in order to make use of their talents for the good of society. To be inclusive, institutions must feature private property, an unbiased system of law, and a provision of public services providing a level playing field. It must also permit the entry of new business. In their account, in particular the latter carries the potential for creative destruction, which in turn poses an inherent risk for elites of the status quo (Acemoglu and Robinson 2012, 75). A demobilized paramilitary reported that “the cattle rancher can not maintain let’s say 500 people, medicines for combatants, uniforms, clothes, he simply does not have the capacity” (cited in Gutiérrez Sanín and Barón, 2006, 13). A case from the Uribe family that is making its way through the courts illustrates the potential connections between land interest and privatized security. Santiago Uribe was arrested in February 2016 on charges that he founded and then commanded the para­ military death squad “Doce Aposteles” (“Twelve Apostles”) from his family finca in Antioquia. A police officer, who arrived in the region in 1994, told investigators that superiors obliged him to work with the Apostles. He also reported uniformed militias with heavy guns entering and leaving the Uribe Finca called “La Carolina”. See The Guardian, “Brother of Colombian ex-president arrested as alleged death squad leader”, 29 February 2016. https://www.theguardian.com/world/2016/feb/29/colombia -president-alvaro-uribe-brother-charged-death-squad. Republic of Colombia, Supreme Court of Colombia, Sentencia No. 59 of 1990. Republic of Colombia, Supreme Court of Colombia, Sentencia No. 138 of 1990. This decision contains some aspects that the Constitutional Court later takes on when it develops the substitution doctrine. See Chapter 4. This section builds entirely on the interviews conducted in Bogotá from September 2012 until May 2013. See Appendix B. Interview with auxiliary judge of the Constitutional Court, 15 May 2013. Interview with constitutional scholar, 7 May 2013. This is also the reason why the new Constitution has relatively flexible reform mechanisms and lacks anything resembling eternity clauses such as the German Basic Law. Interview with Constitutional Court judge, 8 March 2013. Interview with Supreme Court judge, 17 April 2013. Interview with Constitutional Court judge, 3 May 2013. Interview with Constitutional Court judge, 3 May 2013. Interview with Constitutional Court judge, 17 May 2013. Interview with Constitutional Court judge, 8 March 2013. Interview with Constitutional Court judge, 8 March 2013; see also Cifuentes (1995). Colombians gave those associated with the drug trade the name traquetos after the onomatopoetic sound of their guns: “trr trr trr”.

The Novelty of the 1991 Constitution

83

30 “Don Berna” in a recent autobiography confirms the importance of the Escobar chase. In fact, he states it was his brother who killed “El Patron”. Importantly, this is what Carlos Castaño purportedly said upon hearing the news of Escobar’s death: “Now that Pablo is dead, we must begin a new struggle. It will be against the guerrilla. We must build a political-military organization. With the resources, weapons and contacts that remain in the war against Escobar, we can begin. Berna and Seemilla, from this moment, you are part of this new organization as subversive members of the Estado Mayor.” Semana, “Mi hermano mató a Pablo Escobar: ‘Don Berna’”, 27 July 2014. http://www.semana.com/ nacion/articulo/mi-hermano-mato-pablo-escobar-don-berna/397197-3. 31 In 1997, a total of 193 massacres were registered in Colombia. Antioquia was the hardest hit with 75 massacres, followed by Cesar (19), Magdalena (15), and Bolívar (13). Several human rights organizations and the Interamerican Court for Human Rights have con­ demned the close cooperation between armed forces and the CONVIVIR paramilitary groups in those crimes (see Human Rights Watch 1996; Ávila Martínez 2010, 113; InterAmerican Court of Human Rights. Massacres de Ituango vs. Colombia 2006). 32 The original contract can be found here: “Texto original del ‘Acuerdo de Ralito’”. http://www.derechos.org/nizkor/corru/doc/ralito1.html. 33 The legal sanctioning of landowners’ self-defense forces in 1968 constituted the first, and the creation of drug lords’ own private armies in the 1980s the second, wave. The third wave essentially consisted of the grouping of the two under the umbrella of a counter-insurgency war.

References Acemoglu, Daron, and James A. Robinson. 2012. Why Nations Fail. The Origins of Power, Prosperity, and Poverty. New York: Crown Publishers. Acemoglu, Daron, James A. Robinson, and Rafael J. Santos. 2013. “The Monopoly of Violence: Evidence from Colombia”. Journal of the European Economic Association 11 (1): 5–44. https://doi.org/10.1111/j.1542-4774.2012.01099.x. Ahumada, Consuelo. 1995. “The Constituent Assembly and the New Constitution in Colombia: Political Openness or Authoritarian Trend?” Working paper presented at Annual Meeting of the Latin American Studies Association in Washington, 28–30 September. Arendt, Hannah. 1958. The Human Condition. Chicago: University of Chicago Press. Arendt, Hannah. 1972. Crises of the Republic: Lying in Politics. Civil Disobedience. On Violence. Thoughts on Politics and Revolution. New York: Harcourt Brace & Company. Ávila Martínez, Ariel Fernando. 2010. “Injerencia Política de Los Grupos Armados Ilegales”. In Y Refundaron La Patria. De Cómo Mafiosos y Politicos Reconfiguraron El Estado Colombiano, edited by Claudia López Hernández, 79–215. Bogotá: Corporación Nuevo Arco Iris. Bejarano, Ana María, and Eduardo Pizarro. 2005. “From ‘Restricted’ to ‘Besieged’: The Changing Nature of the Limits to Democracy in Colombia”. In The Third Wave of Democratization in Latin America, edited by Frances Hagopian and Scott P. Mainwar­ ing, 235–260. Cambridge: Cambridge University Press. https://doi.org/10.1017/ CBO9780511791116.009. Bejarano, Ana María, and Renata Segura. 1996. “El Fortalecimiento Selectivo Del Estado Durante El Frente Nacional”. Controversia 2 (169): 9–35. Bergquist, Charles. 1986. Coffee and Conflict in Colombia, 1886–1910. Durham: Duke University Press. Bergquist, Charles. 1992a. “The Labor Movement and the Origins of the Violence”. In Violence in Colombia. The Contemporary Crisis in Historical Perspective, edited by

84 The Novelty of the 1991 Constitution Charles Bergquist, Ricardo Peñaranda, and Gonzalo Sánchez, 51–72. Wilmington: Scholarly Resources Inc. Bergquist, Charles. 1992b. “Introduction: Colombian Violence in Historical Perspective”. In Violence in Colombia. The Contemporary Crisis in Historical Perspective, edited by Charles Bergquist, Ricardo Peñaranda, and Gonzalo Sánchez, 1–10. Wilmington: Scholarly Resources Inc. Beasley-Murray, Jon, Maxwell A. Cameron, and Eric Hershberg. 2009. “Latin America’s Left Turns: An Introduction”. Third World Quarterly 30 (2): 319–330. https://doi. org/10.1080/01436590902770322. Boesten, Jan. 2014. “The Generalization of Particularized Trust. Paramilitarism and Struc­ tures of Trust in Colombia”. Colombia Internacional 81 (August): 237–265. https://doi. org/10.7440/colombiaint81.2014.08. Bonilla Maldonado, Daniel (ed.). 2013. Constitutionalism of the Global South. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9781139567114. Bushnell, David. 1971. “Voter Participation in the Colombian Election of 1856”. The Hispanic American Historical Review 51 (2): 237–249. https://doi.org/10.2307/ 2512474. Bushnell, David. 1992. “Politics and Violence in Nineteenth-Century Colombia”. In Vio­ lence in Colombia. The Contemporary Crisis in Historical Perspective, edited by Charles Bergquist, Ricardo Peñaranda, and Gonzalo Sánchez, 11–30. Wilmington: Scholarly Resources Inc. Bushnell, David. 1993. The Making of Modern Colombia. A Nation in Spite of Itself. Ber­ keley: University of California Press. Cajas-Sarria, Mario Alberto. 2008. El Control Judicial a La Reforma Constitucional: Colombia, 1910–2007. Cali: Oficina de Publicaciones. Facultad de Derecho y Ciencias Sociales. Universidad Icesi. Centeno, Miguel Angel. 2002. Blood and Debt War and the Nation-State in Latin Amer­ ica. University Park: Penn State University Press. Cepeda Espinosa, Manuel José. 2007. Polémicas Constitucionales. Bogotá: Legis. Cifuentes, Eduardo. 1995. “El Constitucionalism de La Pobreza”. Dereito 4 (2): 53–77. Collier, David, and Ruth Collier. 2002. Shaping the Political Arena. Critical Junctures, the Labor Movement, and Regime Dynamics in Latin America. Notre Dame: University of Notre Dame Press. Collier, David, and Gerardo L. Munck. 2017. “Building Blocks and Methodological Challenges: A Framework for Studying Critical Junctures”. Qualitative and MultiMethod Research 15 (1): 2–9. Deas, Malcolm. 1997. “Violent Exchanges: Reflections on Political Violence in Colombia”. In The Legitimization of Violence, 350–404. London: Palgrave Macmillan UK. https:// doi.org/10.1007/978-1-349-25258-9_9. Dix, Robert H. 1989. “Cleavage Structures and Party Systems in Latin America”. Com­ parative Politics 22 (1): 23. https://doi.org/10.2307/422320. Dix, Robert H. 1967. Colombia: The Political Dimensions of Change. New Haven: Yale University Press. Drake, Paul W., and Eric Hershberg. 2006. State and Society in Conflict Comparative. Perspectives on the Andean Crises. Pittsburgh: University of Pittsburgh Press. Dugas, John C. 2001. “The Origin, Impact and Demise of the 1989–1990 Colombian Student Movement: Insights from Social Movement Theory”. Journal of Latin American Studies 33 (4): 807–837. https://doi.org/10.1017/S0022216X01006216.

The Novelty of the 1991 Constitution

85

Duncan, Gustavo. 2006. Los Señores de La Guerra. De Paramilitares, Mafiosos y Autodefensas En Colombia. Bogotá: Editorial Planeta Colombiana. Echeverri Uruburu, Álvaro. 2012. “La Constitución de 1991. De Una Democracia Restringida y Bloqueada a Una Democracia Social y Participativa”. In Historia Con­ stitucional de Colombia. Tomo II, edited by Jaime Vidal Perdomo and Augusto Trujillo Muños, 439–489. Bogotá: Ediciones Academica Colombiana de Jurisprudencia Collec­ ción Investigaciones. Esguerra Portocarrero , Juan Carlos. 2012. “La Reforma Constitucional de 1910”. In Historia Constituticional de Colombia. Tomo II, edited by Jaime Vidal Perdomo and Augusto Trujillo Muños, 83–114. Bogotá: Ediciones Academica Colombiana de Jur­ isprudencia Collección Investigaciones. Finkel, Jodi. 2008. Judicial Reform as Political Insurance: Argentina, Peru, and Mexico in the 1990s. Notre Dame: University of Notre Dame Press. García Villegas, Mauricio. 2001. “Constitucionalismo Perverso. Normalidad y Anormalidad Constitucional En Colombia: 1957–1997”. In El Caleidoscopio de Las Justicias En Colombia. Tomo I, edited by Boaventura de Sousa Santos and Mauricio García Villegas, 317–371. Bogotá: Siglo del Hombre Editores y Universidad de los Andes. Gerring, John. 2008. “Case Selection for Case-Study Analysis: Qualitative and Quantitative Techniques”. In The Oxford Handbook of Political Methodology, edited by Janet M. BoxSteffensmeier, Henry E.Brady, and David Collier, 645–684. Oxford: Oxford University Press. https://doi.org/10.1093/oxfordhb/9780199286546.003.0028. Gutiérrez Sanín, Francisco. 2007. Lo Que El Viento Se Llevo: Democracia y Partidos Politicos En Colombia 1958–2006. Bogotá: Grupo Editorial Norma. Gutiérrez Sanín, Francisco, Tatiana Acevedo, and Juan Manuel Viatela. 2007. “Violent Liberalism? State, Conflict and Political Regime in Colombia, 1930–2006. An Analytical Narrative on State-Making”. Working paper published in Crisis States Research Centre London. 19:2. Gutiérrez Sanín, Francisco, and Mauricio Barón. 2006. “Re-Stating the State: Paramilitary Territorial Control and Political Order in Colombia (1978–2004)”. Crisis States Research Centre, DESTIN, LSE. Hailbronner, Michaela. 2017. “Transformative Constitutionalism: Not Only in the Global South”. The American Journal of Comparative Law 65 (3): 527–565. https://doi.org/ 10.1093/ajcl/avx016. Hartlyn, Jonathan. 1988. The Politics of Coalition Rule in Colombia. Cambridge: Cam­ bridge University Press. Hay, Colin. 2008. “Constructivist Institutionalism”. In The Oxford Handbook of Political Institutions, edited by Bert A. Binder, Sarah A. Rhodes, and R.A.W. Rockman, 56–74. Oxford: Oxford University Press. https://doi.org/10.1093/oxfordhb/9780199548460. 003.0004. Herbst, Jeffrey. 2000. States and Power in Africa: Comparative Lessons in Authority and Control. Princeton: Princeton University Press. Hoskin, Gary. 1998. “The State and Political Parties in Colombia”. In Colombia. The Pol­ itics of Reforming the State, edited by Eduardo Posada-Carbó, 45–70. London: Mac­ millan Press Ltd. Human Rights Watch. 1996. Colombia’s Killer Network. The Military-Paramilitary Part­ nership and the United States. New York: Human Rights Watch. Huntington, Samuel P. 1991. The Third Wave: Democratization in the Twentieth Century. Oklahoma: University of Oklahoma Press.

86 The Novelty of the 1991 Constitution Kline, Harvey. 1996. “Colombia: Building Democracy in the Midst of Violence and Drugs”. In Constructing Democratic Governance. South America in the 1990s, edited by Jorge I. Domínguez and Abraham F. Lowenthal, 20–41. Baltimore: The Johns Hopkins University Press. Kumm, Mattias. 2010. “The Idea of Socratic Contestation and the Right to Justification: The Point of Rights-Based Proportionality Review”. Law & Ethics of Human Rights 4 (2): 142–175. https://doi.org/10.2202/1938-2545.1047. Leal Buitrago, Francisco. 1984. Estado y Política En Colombia. Bogotá: Siglo XXI. Lipset, Seymour Martin, and Stein Rokkan. 1967. Party Systems and Voter Alignments. Cross-National Perspectives. New York: Free Press. López-Alves, Fernando. 2000. State Formation and Democracy in Latin America. 1810– 1900. Durham: Duke University Press. López Hernández, Claudia. 2010. Y Refundaron La Patria … De Cómo Mafiosos y Políticos Reconfiguraron El Estado Colombiano. Bogotá: Debate. https://doi.org/10.1016/ 1359-0189(86)90036–90031. Mahoney, James. 2000. “Path Dependence in Historical Sociology”. Theory and Society 29 (4): 507–548. https://doi.org/10.1023/A:1007113830879. Mahoney, James, and Kathleen Thelen. 2009. “Explaining Institutional Change”. In Explaining Institutional Change: Ambiguity, Agency, and Power, edited by James Mahoney and Kathleen Thelen. Cambridge: Cambridge University Press. https://doi. org/10.1017/CBO9780511806414. March, James G., and Johan P. Olsen. 2008. “Elaborating the ‘New Institutionalism’”. In The Oxford Handbook of Political Institutions, edited by R.A.W. Rhodes, Sarah A. Binder, and Bert A. Rockmann. Oxford: Oxford University Press. https://doi.org/10. 1093/oxfordhb/9780199548460.003.0001. Maya Chaves, María José. 2012. “Discordia, Reforma Constitucional y Excepción de Inconstitucionalidad”. Revista de Estudios Sociales 42 (April): 118–128. https://doi. org/10.7440/res42.2012.11. Martz, John D. 1997. The Politics of Clientelism. New Brunswick: Transaction Publishers. Mazzuca, Sebastián, and James A. Robinson. 2009. “Political Conflict and Power Sharing in the Origins of Modern Colombia”. Hispanic American Historical Review 89 (2): 285–321. https://doi.org/10.1215/00182168-2008-085. McConnell, Shelley A. 2010. “The Return of Continuismo?” Current History 109 (724): 74–80. https://doi.org/10.1525/curh.2010.109.724.74. Mendieta, David. 2010. “La Acción Pública de Inconstitucionalidad: A Propósito de Los 100 Años de Su Vigencia En Colombia”. Universitas 120: 61–84. http://www.scielo. org.co/pdf/vniv/n120/n120a03.pdf. Negretto, Gabriel L. 2013. Making Constitutions. Making Constitutions: Presidents, Parties, and Institutional Choice in Latin America. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9781139207836. Osterling, Jorge P. 1989. Democracy in Colombia. Clientelist Politics and Guerrilla War­ fare. New Brunswick: Transaction Publishers. Pachón, Mónica, and Gary Hoskin. 2011. “Colombia 2010: Análisis de Las Elecciones Presidenciales y Legislativas”. Colombia Internacional 74 (July): 9–26. https://doi.org/ 10.7440/colombiaint74.2011.02. Palacios, Marco. 1998. “Colombian Experience with Liberalism: On the Historical Weak­ ness of the State”. In Colombia. The Politics of Reforming the State, edited by Eduardo Posada-Carbo, 21–44. London: Palgrave Macmillan UK. https://doi.org/10.1007/ 978-1-349-26050-8_2.

The Novelty of the 1991 Constitution

87

Palacios, Marco. 2006. Between Legitimacy and Violence. A History of Colombia, 1875–2002. Durham: Duke University Press. Payne, James L. 1968. Patterns of Conflict in Colombia. New Haven: Yale University Press. Plazas Vega, Mauricio A. 2012. “El Frente Nacional”. In Historia Constitucional de Colombia. Tomo II, edited by Jaime Vidal Perdomo and Augusto Trujillo Muñoz, 193–284. Bogotá: Ediciones Academica Colombiana de Jurisprudencia Collección Investigaciones. Posada-Carbó, Eduardo. 2006. La Nación Soñada: Violencia, Liberalismo y Democracia En Colombia. Bogotá: Editorial Norma. Robinson, James A. 2013. “Colombia: Another 100 Years of Solitude?” Current History 112 (751): 43–48. https://doi.org/10.1525/curh.2013.112.751.43. Rodríguez-Raga, Juan Carlos. 2002. “¿Cambiar Todo Para Que Nada Cambia? Repre­ sentación, Sistema Electoral y Sistema Politico Colombiano?” In Degradación o Cambio. Evolución Del Sistema Politico Colombiano, edited by Francisco Gutiérrez Sanín. Bogotá: IEPRI Norma. Romero, Mauricio. 2000. “Changing Identities and Contested Settings: Regional Elites and the Paramilitaries in Colombia”. International Journal of Politics, Culture and Society 14 (1): 51–69. Safford, Frank, and Marco Palacios. 2002. Colombia. Fragmented Land, Divided Society. Oxford: Oxford University Press. Sánchez, Gonzalo, and Donny Meertens. 2001. Bandits, Peasants and Politics. The Case of “La Violencia” in Colombia. Austin: University of Texas Press. Sanders, Elizabeth. 2008. “Historical Institutionalism”. In The Oxford Handbook of Political Institutions, edited by Bert A. Binder, Sarah A. Rhodes, R.A.W. Rockman. Oxford: Oxford University Press. https://doi.org/10.1093/oxfordhb/9780199548460.003.0003. Schmidt, Vivien A. 2008. “Discursive Institutionalism: The Explanatory Power of Ideas and Discourse”. Annual Review of Political Science 11 (1): 303–326. https://doi.org/10. 1146/annurev.polisci.11.060606.135342. Skocpol, Theda, and Margaret Somers. 1980. “The Uses of Comparative History in Mac­ rosocial Inquiry”. Comparative Studies in Society and History 22 (2): 174–197. https:// doi.org/10.1017/S0010417500009282. Solaún, Maurico. 1980. “Colombian Politics: Historical Characteristics and Problems”. In Politics of Compromise: Coalition Government in Colombia, edited by R. Albert Berry, Ronald G. Hellmann, and Mauricio Solaún, 1–58. New Brunswick: Transaction Publishers. Thoumi, Francisco E. 2005. “The Colombian Competitive Advantage in Illegal Drugs: The Role of Policies and Institutional Changes”. Journal of Drug Issues 35 (1): 7–26. https://doi.org/10.1177/002204260503500101. Tilly, Charles. 1985. “Bringing the State Back In”. In Bringing the State Back In, edited by Peter B. Evans, Dietrich Rueschemeyer, and Theda Skocpol. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9780511628283. Valencia Villa, Hernando. 2012. Cartas de Batalla. Una Critica Del Constitucionalismo Colombiano. Bogotá: Panamericana Editorial. Weber, Max. 1946. “Max Weber: Politics as Vocation”. In Max Weber: Essays in Sociology, 77–128. Oxford: Oxford University Press.

3 The Choque de Trenes between Colombia’s High Courts Path Dependence and Legal Argumentation

Introduction If the 1991 Constitution represents a critical juncture in Colombia’s institutional history, as argued in the previous chapter, does its post-genesis evolution follow a path-dependent logic? All three major schools of institutional analysis—histor­ ical, rational choice, and sociological institutionalism—have difficulties account­ ing for change, because they treat institutions as the enduring features that connect the present political and social life to the past (North 1990). In many iterations, only events and processes akin to the 1991 Constitution open space for agency and fundamental change, which subsequently again develops towards an equilibrium (Hall and Taylor 1996, 942; Collier and Collier 2002; Katznel­ son 2003; Capoccia and Kelemen 2007; for a review see Mahoney and Thelen 2009, 3–10). In this discontinuous change model, path dependence implies that initial decisions in one direction induce further movements in that same direc­ tion later on (Pierson 2000, 252). They are relatively inert, Mahoney and Thelen contend (2009, 4). Consequently, they, as well as Streeck and Thelen (2005), challenge the discontinuous change model with the incremental change model, arguing that implementation and enforcement of institutional norms are key political arenas of institutional transformation which, therefore, evolves less deterministically and more fluidly. The 1991 Constitution in Colombia was a moment of contingency and most fundamentally redesigned the judicial branch of Colombia’s separation of powers system: it introduced new institutions (Constitutional Court, Superior Council of the Judiciary), a new rights catalogue, and mechanisms to enforce rights (tutela or writ of protection). As a central novelty of the 1991 Constitution, the tutela unified the ideals and promises of the new Constitution, investing every citizen with human dignity, but at the same time the Constitution created conflicting jurisdictions aris­ ing from the duties of the various judicial bodies. In addition, this occurred in a separation of powers system in the context of a highly criminalized political system. After the implementation of the new Constitution, this resulted in a con­ frontation between the high courts, idiomatically referred to as the choque de trenes (train crash). This collision of jurisdictions, conveniently iterating a metaphor close to that of path dependence, evolved around the question of whether judicial DOI: 10.4324/9781003229285-3

The Choque de Trenes

89

decisions by other high courts—namely the Supreme Court and the State Council, who wield the highest legal authority in the ordinary and administrative field of justice, respectively—are open to rights review by the Constitutional Court. This prosaic question over legal authority and jurisdiction becomes crucial for our understanding of institutional change in separation of powers models, because of the absence of parliamentary immunity in the 1991 Constitution. Colombia’s Constitution tasks the Supreme Court with investigating criminal allegations against legislators, who, in turn, fulfill the same function vis-à-vis the executive. The tutela, in turn, is a mechanism for rights review of all decisions by public organs of the state and therefore also open to lawmakers, who have been accused and convicted for criminal allegations. This chapter tests the discontinuous and incremental change models of institu­ tionalism on the altercation between Colombia’s high courts over the application of the tutela vis-à-vis judicial decisions. It asks whether judicial institutions also evolve path-dependently. There are three elements that make the choque de trenes an important case for the study of institutional evolution: 1) the proliferation of new courts and contradictions between normative principles in the new Constitution result in a contest over jurisdiction between the Constitutional and Supreme Court; 2) the legal functions assigned to each branch of government in Colombia’s checks and balances system turns the fight over jurisdiction between the high courts into a political question that involves majoritarian institutions; and 3) the affixture of rights to the standing of tutela complaints adds a layer to the analysis that implicates the meaning of foundational norms of the new Constitution, and their effects on con­ stitutional adjudication. The precedent setting in the development of jurisprudence implies a natural tendency to value earlier decisions. Jurisdictional struggles raise expectation for distributional effects. Both caveats are central path-dependent conditions of insti­ tutional change that result in the reproduction of initial decisions into a given direction. Despite these conditions, we will see that, in the context of judicial institutions, earlier decisions are not automatically reproduced. Constitutional norms adjudication involves deontological rights claims, the constitutional sig­ nificance actors assign to norms, and how they apply them in factual situations that go to the core of the separation of powers system. This opens up the possi­ bility that certain questions can be re-opened again under the condition that the legal context has changed. In order to test the application of either model of institutional change, this paper contrasts sub-cases in Colombia’s post-1991 evolution that involve the criminalization of its representative institutions: the so-called proceso 8000 affair and the parapolítica scandal. In the 1990s, the Cali Cartel funneled funds into the presidential campaign of Ernesto Samper, who won the electoral contest in 1994. The origin of funds became public after his election and all but condemned Samper to fight off accusations for the entire term. He eventually survived the entire time in office, because legislators cleared him from wrongdoing (despite fairly conclusive evidence to the contrary). Citizens wanted to sue Members of Congress in the Supreme Court for breach of duty, but legislators succeeded in

90 The Choque de Trenes stopping these investigations through tutela complaints they submitted to the Constitutional Court. The Court followed their argumentation and stopped the Supreme Court’s criminal investigations. In the 2000s, during the Uribe years, Members of Congress were accused of having cultivated close relations with paramilitaries associated with the drug trade for electoral purposes. In the course of these investigations, legislators used the same tool to fight off investigations, submitting tutelas against Supreme Court investigations, yet, by the year 2010, 102 either faced investigations or were indicted for conspiring to commit a crime by collaborating with paramilitaries. The intention of this chapter is to explain this outcome and see how the explanation fits into the conceptualization of models of institutional change and path dependence. At the end of the analysis stand three key observations: 1. The continuous production of legal facts set the parapolítica apart from the proceso 8000. In the proceso 8000, legislators could utilize the tutela mechan­ ism to foil criminal investigations, while in the parapolítica, this strategy eventually failed to protect them from criminal liability. As a consequence, courts continued their investigations and provided the supplementary pro­ duction of “legal facts”; facts that are probed in juridical processes and thereby assume a higher degree of validity. This forced President Uribe to invest a lot of political capital in order to delegitimize the court’s claims. These campaigns eventually transgressed into illegality. 2. The continued production of legal facts was, to a decisive part, the result of a normative agreement between the Supreme and Constitutional Court in tutela decisions implicating the role, privileges and duties of legislators. 3. The trajectory of jurisprudential development in the question of the tutela contra sentencias (tutelas against judicial decisions)1 did not reflect the pre­ dictions of the classical path dependence model, but evolved along discursive trajectories that left decisions open to (reasoned) contestations. As a consequence of these findings, I argue that an intelligible reading of path dependence for judicial institutions must be able to incorporate the following points: � �

�

Once a decision is taken, the cost of reversal increases, but not exponentially so. Rather, reversals require good explanations rooted in precedent and change of context. Earlier decisions do matter more than later decisions, since discourse (in par­ ticular legal discourse) builds on logical coherence. However, earlier decisions might also open new questions further down the road that can significantly alter the path. There cannot be a practical point of no return, because all decisions are the­ oretically forever open to contestation.

The notion that legal questions can be re-opened under specific conditions in particular proves difficult for path dependence and gives merit to incremental

The Choque de Trenes

91

change models of institutional development. As will be shown, rather than closing down options permanently, legal questions (and their answers) can be re-opened discursively under certain conditions. Moreover, earlier decisions can help to pro­ vide the argumentative context to revisit these precedents, restructure their nor­ mative meaning, and apply them anew to a new empirical context. Therefore, earlier decisions do not accumulatively reduce options, but discursively open new paths and options as well. We will see this logic play out in the jurisprudential trajectory of the tutela contra sentencias that is at the root of the choque de trenes, and how it affected constitutional adjudication in cases that weighed legislative autonomy against criminal accountability of legislators. From these considerations follows a four-part organization of the chapter. The first section develops an ideal-type rendition of path dependence and explains the specifics of Colombian constitutionalism after the implementation of the 1991 Constitution. It will detail the specific institutional design of the three branches of government with a particular focus on appointment powers and investigative functions, before highlighting the normative connotations of the tutela and the choque de trenes, which evolve around the legal figures of the cosa juzgada (“the judged case”; or res judicada) and the via de hecho (“by way of fact”, which entails that a judge creates her own law that diverges from the Constitution). The second and third parts of this analysis focus on the specific empirical cases, the proceso 8000 and parapolítica, respectively, and contrast their political and normative evolutions. The examination of the political developments will result in a separa­ tion of the trajectory of each scandal into different phases. The discussion of the normative evolution closely inspects key court decisions and how they affected the evolution in each political scandal. The fourth part will conclude the findings and connect the empirical analysis with the theoretical considerations.

Part I: Path Dependence and Colombia’s New Institutions Path dependence, as a concept, is a risky theory (Popper 1968) that “forbids cer­ tain things to happen”, while expecting other things to take place (Popper, cited in Gerring 2008, 659). Risky predictions are those that feature precision and determinacy and exclude factors that are outside of the applied theory’s scope. In this sense, good theory is said to be not only falsifiable, but also include specific causal mechanisms that result in the predicted outcomes (Gerring 2008, 660). Path dependence is such a well-defined concept that has travelled from studies in sociology and economics to political economy, and, finally, political science. First popularized by David’s study of the “QWERTY” keyboard layout, which argues that built-up legacies have consolidated sub-par outcomes persisting over time (1985), it shares a lot of characteristics with how economists conceptualize equi­ librium (Page 2006). The Colliers (2002), whose study of party system transfor­ mations and labor integration in Latin America was another benchmark for the concept’s entry into comparative political science in particular, argue that path dependence arises from the mechanisms of reproduction; a “type of constant cause” that embodies the distinct legacy of the critical juncture (35). Therefore, it

92 The Choque de Trenes conjoined characteristics of institutional resistance identified in earlier studies, such as the rule of oligarchy (Michels 1915), the creation of vested interest (Stinch­ combe 1968), and the role of sunk cost that perpetuate institutional frameworks (Krasner 1984). David (1985) and Page (2006) assert, independently from each other, that path dependence involves either small initial advantages or random shocks along the way that alter the course of history—with the focus being the legacy of those historical changes. As Mahoney insists, “path dependence char­ acterizes those historical sequences in which contingent events set into motion institutional patterns or event chains that have deterministic properties” (Maho­ ney 2000, 507). Thus, path dependence has specific observational patterns that result from initial contingency and subsequent legacy. These patterns—namely, the centrality of initial decisions, the exponential increase of cost of reversal (of those decisions), the mechanical reproduction of those decisions, and the impossibility to change course until the occurrence of another rupture of equili­ bria—must hold in the case of the choque de trenes for path dependence to apply for judicial institutions. Conversely, if the observational patterns do not match the empirical analyses, we must amend the conceptualization of path dependence for judicial institutions. Without getting too far ahead of myself, I will argue that contrary to the mechanical predictions of the path dependence logic, judicial decisions do remain open to course corrections. This then invites a discursive interpretation of path dependence that is informed by Mahoney and Thelen’s model of incremental change that focuses on the effects of actors’ resources on implementation. While predominantly the terrain of historical institutionalists who “accent path dependence, the unfolding of outcomes or events over time in particular contexts, and endogenous funnels of causality” (Lichbach 2009, 47), it is not alien to rational choice theorists, either. Margaret Levi, for example, writes that “[p]ath dependence has to mean […] that once a country or region has started down a track, the costs of reversal are very high”. Even with other options further down the path, “the entrenchments of certain institutional arrangements obstruct an easy reversal of the initial choice” (Levi, cited in Pierson 2000, 252). Further exploring the suitability of path dependence cum rational choice, she maintains that “the starting point of the game affects and often determines the end point, but only once the proper payoffs are incorporated”. In effect, “the existence of multiple equilibria makes path dependence both interesting and important”. A crucial addendum is that self-enforcement mechanisms that create institutional setups include players’ belief systems when alternatives to the eventual outcome become increasingly unattractive. She affirms that “beliefs are certainly affected by historical experience, [but] they also are affected by what actors know of the other players within the current context” (Levi 2002, 117). This concurs with Mahoney’s contention that legitimation also evolves in path-dependent trajec­ tories: an actor’s subjective orientations about what is morally correct are affec­ ted by the positive feedback cycle of self-reinforcement, because initial precedent about what is appropriate “forms a basis for making future decisions about what is appropriate” (523).

The Choque de Trenes

93

In essence, the importance of initial decisions and the closing down of further options are the key definitional traits of path dependence. Path dependence has a methodological caveat. It obliges the researcher to identify an outcome and trace its origin to a particular set of historical events to show the contingency of those events. Mahoney goes on to distinguish between two types of path-dependent developments: self-reinforcing sequences and reactive sequences. Contingent historical events may trigger decisions that corresponds with the initial adoption or a breaking point. Either event result in the stable reproduction of that institution over time (Mahoney 2000, 535). Mahoney then makes three crucial observations on the nature of path dependence that inform the observable implica­ tions of this study. Firstly, the contingency of the initial juncture setting the sub­ sequent development on its path defies deterministic explanation rooted in conditions prior to that juncture. Rather, these are stochastically related to recurrent patterns of development. Secondly, earlier events in the pattern matter more than later ones, even if the latter ones hold the potential to create deeper effects. In other words, path dependence describes a sequence of events, whose order matters, because when things happen also affects how things continue to happen (Mahoney 2000, 510). Thirdly, the patterns of reproduction display a high degree of “inertia” and resistance to abrupt change. In sum, the original moment of the creation of a path-dependent pattern defies theoretical explanation (theory cannot predict this event), but its recreation by reinforcing mechanisms follows standard explanations (see also Pierson 2000, 263). Finally, there are at least four causal mechanisms for path-dependent sequences: increasing returns, self-reinforcement, positive feedbacks, and lock-in. Increasing returns entails equally increasing benefits as one actor makes the same decision. Self-reinforcement means that a choice involves a set of forces or institutions that encourage the same choice in the future. Positive feedbacks involve other actors, in that an action creates positive externalities if other actors make the same deci­ sions. Lock-in entails that one choice becomes better than other choices because a sufficient number of actors have already made that decision (Page 2006, 88). Together, there are key observable implications: � � �

Earlier decisions matter more than latter decisions (Mahoney 2000); Once a decision is taken, the cost of reversal increases exponentially, moving institutional development in a path of (mechanical) reproduction (Mahoney 2000; Pierson 2000; Thelen 1999; Levi 1997); The point of reversal becomes a practical impossibility once costs have incur­ red beyond a certain point (although the theory of path dependence does not predict this point in any theoretical way). Only another punctuated equili­ brium can reverse the initial decisions.

Finally, conceptual considerations also imply questions of ontological suitability (Hall 2003), or if path dependence is actually helpful for understanding the trajectory of judicial institutions? As seen, the 1991 Constitution was a critical juncture in Colombia’s history, and path dependence is said to follow such junctures as they produce ruptures in equilibrium-based evolutions. Consequently, the basic condition

94 The Choque de Trenes that path dependence is hypothesized to follow critical junctures applies in the Colombian case. Moreover, if path dependence has applicability to the study of institutional development, there is no obvious contention that a priori disqualifies judicial institutions. In the study of judicial behavior, political scientists have con­ sistently applied tools from behavioral studies to understand how judges decide and produced important insights—even when these models do not account for the entire range of complexities involved in judicial decision-making (as I argue in Chapter 4). The theorists of path dependence have also said that ideational components are very much part of a path-dependent framework. Therefore, legal institutions should not be automatically disqualified for their rootedness in normative rights claims, either. Rather, the question should precisely be whether path dependence applies to judicial institutions. The potential for conceptual specification of path-dependent institutional developments lies in the proposed case: the jurisprudential evolution of the tutela, and its role in cases concerning the criminal liability of legislators. On the one hand, its interpretation and application can have two potential sources for distributive dynamics: 1) the jurisdiction of courts in criminal law cases; and 2) legislators’ lack of parliamentary immunity. On the other hand, the tutela has evident normative con­ notations that reflect the intentions of the movement that initiated constitutional reform in the first place. Even if, as Thelen already lamented in 1999, path depen­ dence is too contingent, the conceptualization of initial junctures as blank slates is unrealistic, and the mechanical reproduction too perfunctory (1999, 385), path dependence still retains its utility as an epistemological tool. It is, as stated at the outset, a risky theory that implies fairly specific predictions. Moreover, the point of ideal-type formulations for the study of human action is not that we assume interac­ tion to follow these predictions consistently, but that we use them to open up analy­ tical space. The point of good Weberian ideal-type sociology is to understand those human secrets that “baffle probability” by devising epistemological predictions, and compare these with the real-world outcomes. Here, the benefit of developing a rig­ orous reading of post-genesis evolution helps us to identify the period when institu­ tional development moved off the beaten path, and then specify mechanisms that help to explain what was behind that shift. The preceding summary of the core attributes of path dependence provides the key questions for the analysis of institutional trajectory in the tutela contra sentencias: � � �

�

Did the jurisprudence regarding the tutela follow a mechanical and linear line from its inception in the 1991 Constitution? Did earlier decisions matter more than later decisions without the possibility to reverse course in later cases, or do earlier decisions create new possibilities for later decisions on the same merit? If the Constitutional Court changed or modified its jurisprudence, did the modification follow mechanically from earlier decisions, or did the Court base its change in principled explanations engaging a legal argumentation evolving around legal values and shifts in legal context? If there was a clear break, did that break in turn result in a logically following pattern, indicating reactive path dependence?

The Choque de Trenes

95

Engaging the path dependence framework and modifying it for the purpose of analyzing legal argumentation will yield important results: judicial decisions do remain open to course corrections. Such a path dependence of legal argumenta­ tion shares features with the incremental change model, Streeck and Thelen (2005) and Mahoney and Thelen (2009) proposed. The next sections of this chapter will show that compliance with institutions can become a most important, and intricate, part of institutional development. In the 1990s, at the end of the era of the big drug cartels, the Constitutional Court upheld legislators’ autonomy against criminal investigations by the Supreme Court. In the early 2000s, during the Uribe years, the Constitutional Court shifted the weighing in its jurisprudence of parliamentary immunity from legislative autonomy to a stricter application of the Supreme Court’s privilege to hold them accountable in the face of the co­ option of Congress by paramilitary groups. The result was that numerous Mem­ bers of Congress were prosecuted for their illicit relations and received lengthy prison sentences. Given the argumentative procedures within a court, the internal shift was more a response to the disclosures of the Supreme Court and the legal facts it produced, resulting in something akin to institutional learning. In the end, it was simply not argumentatively feasible to defend parliamentary immunity as a bulwark for free deliberation, when Members of Congress utilized the right to freely deliberate and vote in Congress as a smoke screen to cover up their impli­ cation in gross human rights crimes (including mass atrocities such as massacres and extra judicial killings). Courts and the New Constitutional Regime: Selection, Jurisdiction, and the Interdependence of the Branches of Government At least since Hamilton’s Federalist Papers is the importance of the judiciary and independent institutionalization of its judges a well-articulated feature of a separation of powers system (Vanberg 2008, 101). In paper number 78, he famously argued that the general liberty of the people could never be endangered from that quarter as long as it remains “truly distinct from both the legislature and the executive” (1788). In order to fully comprehend the dynamics of the conflict unfolding during parapolítica, it is critical to briefly explain the institutionalization of all four high courts in Colombia’s judicial system, as well as the interdependent relations between the courts and the three branches of government. This includes institutional design, the functions each court exercises, as well as the selection method for judges that differ for each court. In fact, the evolution of the para­ política scandal is testament to the relation between judicial independence and appointment procedure, as will become clear in the prosecution of Members of Congress described below. The 1991 Constitution created a complex web of interdependencies between the branches of government through appointment procedures and investigative functions. Figure 3.1 shows the four high courts, the individual chambers within each court, and how judges were appointed to those chambers in order to high­ light the institutional relations between the branches and their respective

96 The Choque de Trenes Cons�tu�onal Court (9 judges)

Appoint 1 judge

Lists Proposes 3 lists of 3

Proposes 3 lists of 3

Elect from lists

State Council (31 Judges)

Supreme Court (23 Judges)

Elect their replacement

Lists

Lists

President

Inves�gates and judges

House of Rep.

Senate

Inves�gates and judges (Commission of Accusa�on)

Congress Appoint 3 judges

Appoint 2 judges

Elect

Lists Supreme Council of the Judiciary

Admin. Chamber (6 Judges)

Discipl. Chamber (7 Judges)

Figure 3.1 The courts and the branches of government in Colombia’s 1991 constitutional regime

capacities. In 2012—after the period under study here—the decision was made to eliminate the Superior Council of the Judiciary, which was concluded in 2015 with Legislative Act 2. However, until 2022 judicial reforms failed to permanently establish a new disciplinary body in the judiciary. The shift from the “American Model” of constitutional adjudication (with one Supreme Court to review legislative and administrative acts as well as criminal sentences) to the Kelsenian model with a specialized court at the helm of the constitutional jurisdiction (Kelsen 1945), without a doubt, represented one of the most fundamental institutional changes imposed by the 1991 Constitution. The Colombian Constitutional Court exercises abstract, erga omnes, constitutional review through the acción popular de inconstitucionalidad. An important facet of this power in the Colombian context is that the Court can impose conditions for declaring a law or constitutional reform legal as happened, for example, in the first re-election decision. It required a statutory law to even the electoral playing field between the incumbent and competitors in the election before the law to allow one consecutive re-election could pass.2 In its decision on the Justice and Peace Law, the Court imposed conditions upholding victims’ rights to truth (García Villegas et al. 2010, 324–325).3 In addition, the Court has the authority to ulti­ mately decide concrete rights complaints through the acción de tutela (tutelas can be filed with any court in the country, including the disciplinary chambers). While these are nominally inter alia only concerning the parties involved in the dispute, tutela decisions can attain erga omnes effects in so-called unification decisions

The Choque de Trenes

97

(Cepeda Espinosa 2007, 31 and 103; see below). Ordinarily, reviews of tutelas are exercised by a plenum consisting of three randomly selected magistrates, while abstract constitutionality review is exercised by the entire plenum in the Sala Plena (plenary chamber). In the course of the choque de trenes, one important concession made by the Constitutional Court to the other courts was that high court decisions, by principle, should always be reviewed by the entire plenum. The 1991 Constitution implemented the mixed method for selecting constitu­ tional judges, combining the advantages of a political selection process (legitima­ tion) with a technocratic process (institutional independence). In the eyes of the constituents in 1991, the selection of constitutional judges by co-optación (the selfselection of vacant positions on the court by the court itself) proved too opaque.4 Under the new regime, the Senate elects all magistrates on the Constitutional Court, yet, the shortlists are prepared by three different bodies: the Supreme Court, the President, and the State Council. Each body presents a list of three candidates for a total of three positions on the Court. Thus, of the nine magis­ trates sitting on the bench, a maximum of three could have come from lists pre­ sented by the executive. Prior to the re-election reform in 2005, only every second President at the end of his term could have had the chance to nominate shortlists for the bench (assuming that all judges stay for their entire term of eight years). This is referred to as the periodization of offices that aimed at curtailing executive influence of horizontal oversight offices (Pérez-Liñán and Castagnola 2009, 93).5 The new Constitution divided the Supreme Court into three chambers: one responsible for criminal law (nine magistrates), one for civil and agrarian law (seven magistrates), and one for labor law (seven magistrates, for a total of 23 magistrates in the entire Court). As stipulated in Article 234, the Supreme Court remained the highest court of the ordinary justice system. Furthermore, as the highest court of the ordinary justice system, it also received the function to inves­ tigate and sentence members from either chamber of Congress (Article 235, 2). The new Constitution retained the co-optacion method for the selection of Supreme Court magistrates: the full chamber selects replacements from lists pre­ pared by the administrative chamber of the Superior Council of the Judiciary. Additionally, an important administrative task the Constitution bestowed on the Supreme Court was to elect the attorney general from a shortlist provided by the President (Article 189, 19). At the end of Uribe’s second term, the election of the Attorney General became the last confrontation between the Supreme Court and the President, because the Supreme Court repeatedly (and in the end successfully) turned down all candidates the President put forward. The State Council, taken over from the old Constitution, remained the highest court in the administrative jurisdiction with 31 magistrates subdivided into two chambers. They are also elected through the co-optation method: the full cham­ ber elects positions for upcoming vacancies from lists prepared by the adminis­ trative chamber of the Superior Council of the Judiciary (Articles 236–238). Finally, the Superior Council of the Judiciary was probably the most peculiar high court. Divided into two chambers, the administrative and disciplinary section,

98 The Choque de Trenes each fulfills eponymous functions in internal affairs of the judiciary. The adminis­ trative chamber is tasked with compiling lists, from which the State Council and the Supreme Court select the replacements for vacancies on their respective courts. The chamber itself consisted of six magistrates, who were appointed by the Constitutional Court (one judge), Supreme Court (two judges), and the State Council (three judges), thereby ensuring that the co-optation method for select­ ing judges on the Supreme Court and the State Council is at arm’s length from the legislative and executive branches. The election process of the disciplinary chamber, however, was entirely political: Congress elected the seven magistrates on this body from lists sent from the executive (Article 254–257). It was a com­ pletely politicized selection process, which, as will be shown below, had a huge impact on the independence of the judges (or rather their lack thereof). It became, in the words of Constitutional Court magistrate Nilson Pinilla Pinilla, a “decomposed and politicized body” acting in the interest of discredited Members of Congress.6 In 2012, President Santos started a process to reform the Superior Council of the Judiciary, which has not been concluded until 2022. As this brief description makes readily evident, the 1991 Constitution created a complex web of dependencies and safeguards between the various courts and bran­ ches of government to impose accountability and bestow democratic legitimacy at the same time. The selection of the Constitutional Court’s magistrates is particularly informative of these two imperatives: on the one hand, they are elected by one of Colombia’s representative bodies, the Senate; on the other hand, two-thirds of the positions are filled from lists prepared by autonomous legal bodies, who in turn are kept at arm’s length from both the executive and the legislature. It also created contradictions. The lack of parliamentary immunity served to impose accountability, but at the same time colluded with the imperative that legislators ought to freely deliberate in a constitutional regime. As we will see, the tutela can be a wedge driven into the Supreme Court’s accountability function vis-à-vis Congress. Rights claimants, in this case Members of Congress, can duly claim their rights to be vio­ lated. After all, legislators’ free and uncoerced deliberation is a valuable good in the constitutional order. At the same time, legislators’ criminal deeds were exactly those abuses constituents had in mind when they conceived of the tutela mechanism as a way to improve Colombia’s abhorrent human rights situation. Together, this cre­ ated a situation in which competing rights bearers’ claims and constitutional values could not be easily solved with reference to the universality of rights, but required a careful weighing of the validity of the specific claims in the specific contexts. While the re-ordering of the separation of powers system in Colombia in the 1991 Constitution created jurisdictional clashes prone to distributive effects, it is my contention that weighing competing rights claims is also intrinsically bound up with the exercise of public reasoning, adding the crucial normative layer of this analysis. I will show that it requires the judge to make validity claims herself in order to evaluate the precedence of a specific rights claim over another. Not least, such an evaluation involves the significance and meaning actors assign to the constitutional order as a whole, its specific organizations therein, and how parti­ cular norms reflect the meaning of the constitutional order. In order to better

The Choque de Trenes

99

understand this layer, we need to recall the neo-constitutionalist connotations of human rights in the normative grounding of the 1991 Constitution in general, and the tutela in particular. The Choque de Trenes and Nuevo Derecho If constitutionalism is the exercise of political power limited by constitutional rules, neo-constitutionalism involves further limitation placed upon the exercise of power through the invocation of rights and human dignity (Sartori 1962; Kumm 2010). It is not a coincidence that the tutela epitomizes the promises of the 1991 Constitution. The deliberative origin of student movement and constituent pro­ cess ensured that the quintessential demand to address human rights in Colombia found its way into the Constitution through the neo-constitutionalist principles embraced in its norms. The tutela is intrinsically bound up with the Constitution’s promise to protect the human dignity of all Colombian citizens, echoing the general call for the respect of human rights in the prelude of the constituent assembly in 1991.7 It also has evident precedents in the German Verfassungs­ beschwerde and the Spanish recurso de amparo. Students as well as constituents in the assembly took note of the neo-constitutionalist experiences of Germany with its Basic Law and Spain and its Constitution from 1978 after those countries’ encounters with totalitarian and authoritarian dictatorships. They therefore inclu­ ded the tutela to give the rights catalogue some teeth to be effective. In order to comprehend the unfolding and evolving confrontation between the Constitutional Court and the Supreme Court over the jurisdiction of the new mechanism, it is therefore indispensable to understand not only its formal aspects, but also the normative imperative that justified its incorporation into the Constitution. As a Constitutional Court judge explained, the tutela not only strengthened con­ stitutional review but turned “all judges in the country into judges on behalf of human rights”.8 It is the teleological rationalization of the tutela that dis­ cursively anchored the evolution of the proceso 8000 and the parapolítica inves­ tigations in the exercise of public reasoning. It will become clear that opinions and positions in disputes over its applications, regardless of the strategic moti­ vations that may have driven specific actors, always must account for the inclu­ sion of rights and human dignity in the definition of the Colombian subject in the 1991 Constitution. Article 86 of the 1991 Constitution states: Every person has the right to file a writ of protection before a judge, at any time or place […] for the immediate protection of his/her fundamental con­ stitutional rights when that person fears the latter may be violated by the action or omission of any public authority. The protection will consist of an order issued by a judge enjoining others to act or refrain from acting. The order, which must be complied with immediately, may be challenged before a superior court judge, and in any case the latter may send it to the Constitu­ tional Court for possible revision.

100 The Choque de Trenes Of these attributes outlining the norm of the tutela in the 1991 Constitution, the immediacy and ubiquity of its jurisdiction are the most central neo-constitutionalist qualities. It connects the empirical life world of the subject in the Constitution with the legal entitlements outlined under Chapters 1 and 2 of the Constitution (namely, the fundamental rights of every Colombian). Moreover, this protection of rights is purpo­ sely bound up with the promise to uphold human dignity, defined not only in Article 1 of the Constitution, but also in conjunction with the stipulation of socio-economic rights (Article 42, 2; Article 51), and cultural rights (Article 70). Arango and Cepeda Espinosa—the former a most distinguished legal scholar in Colombia and the latter a former magistrate and President of the Constitutional Court—both stress that the tutela offers the possibility, “in objective situations of urgency, to preserve human dignity” (see Cepeda Espinosa 1993, 199; Arango 2003, 147; see also Aponte 1994; Cifuentes 1995, 76–77). The preservation of human dignity—essentially a contested concept9 despite its inclusion in the Universal Declaration of Human Rights already in 1946—is best understood as a procedural and gradual exercise to achieve substantive results. In other words, the tutela embraces a democratic proceduralism to improve human rights grievances that occur in a real-life context. It is not metaphysics. Human dignity and the principle of urgency serve as interlocutors between constitution and human relationships. As Arango explains, the reservation against the realization of material social rights—one of the most important fields of application of the tutela (Rodríguez Garavito and Rodríguez Franco 2010)— usually takes the route of inferring a type of judicial decisionism, which legislates the transfer of funds from the haves to the have-nots. Arango responds that such a voluntarist view of the state and materialist understanding of social relations “is both reductionist and incompatible with the Constitution” of 1991 and con­ stitutionalism in general (2003, 147).10 Rather, the neo-constitutionalist frame­ work of the 1991 Constitution relies on an affirmative view of human dignity and consequentialist theory of rights. Its central validity claim concerning rights is that “a right is a normative position based on valid and sufficient claims, whose unjustified non-recognition imminently harms the right-bearer” (Arango 2003, 144). A critical point to note here is that the non-recognition must be justified. It plants the seed for a discursive understanding of the validity and enforcement of rights, ultimately rooted in public reasoning (i.e. reason giving). Of course, legal and political philosophy has long grappled with the recognition of rights in a post-metaphysical world. For neo-Kantian thinkers à la Habermas, such recognition must be bound up in communicative processes to avoid transcending into metaphysics (and thereby neglect validity in human life worlds), without at the same time becoming a victim of post-modern moral relativism. In his discursive con­ ceptualization on the origin of rights and law, the self-legislating citizen must be able to understand herself as the addressee and author of the law. To that end, “we cannot meet this requirement simply by conceiving the right to equal liberties as a morally grounded right that the political legislator merely has to enact” (Habermas 1996, 120). Rather, everyone is involved in disclosing the meaning of rights in specific social and political contexts. Such a contextualization of constitutional adjudication is also evident in the jurisprudence of Colombia’s Constitutional Court. As we have seen,

The Choque de Trenes

101

the new constitutionalism of the 1991 Constitution invokes not only formal princi­ ples of interpretation, but also deliberative ones such as the principle of proportion­ ality or the principle of efficacy. Both result in a process of legal argumentation that takes statutes and norms as given, but weighs the legal consequences of each party in a constitutional dispute (Cifuentes 1995, 66). Connecting rights with communicative action, which balances infractions against potential benefits from a public perspective, runs the risk of jeopardiz­ ing the very idea of rights as deontological restrictions (even against popular demands). Worse still: it might open the gates of moral relativism, placing the entire system of rights protection in limbo. However, as Kumm and Walen showed, human dignity understood as humans treated as ends rather than means is always at stake—even in such simple contractual obligations as buying an apple, when we use the fruit seller as a means to satisfy a desire for fruit. It is the condition of consent that makes that transaction compatible with human dignity. They showed furthermore that proportionality testing can indeed cap­ ture more important deontological restrictions such as long-term preventive detention, because it requires the decision-maker to refrain from mechanical reproduction of statutes. Rather, it engages the judge in a balancing moral reasoning that “is thoroughly deontological and always grounded in human dignity” (2013, 22).11 In sum, the logic of rights protection in the tutela is not that of an auto­ matic panacea. Rather, it is a mechanism to gradually improve the situation of the rights bearer that avoids seeing natural or social disadvantages as objective limitations to the recognition and enjoyment of rights, opening spaces for dif­ ferences of capacities (Arango 2003, 144). It combines seven traits of rightsbased constitutionalism: 1) it is minimalist, not maximalist; 2) it is constitu­ tional, not legal; 3) it is objective, not ad hoc; 4) it arises from precedents, not from legislation; 5) it is diffused, not centralized (meaning tutelas can be sub­ mitted with any judge in the country and then be selected by the Constitu­ tional Court); 6) it prioritizes equity, not legality; and 7) it is gradual, not absolutist (Arango 2009, 302). It is exactly these traits that make the tutela such a good candidate to test the institutionalist hypothesis. Its normativity as well as its formal application provides credence to the assumption that it procedurally augments and builds a thick jur­ isprudence, which is, over time, less and less susceptible to reversal, and thereby introducing precedent into a civil law system (López Medina 2000; Aponte 1994; Cifuentes 1995). On the other hand, it is fully deliberative. The development of rights constitutionalism is not a foregone process that evolves along deterministic patterns but develops along argumentative constraints. In regard to the institu­ tionalist hypothesis of path dependence, the question then becomes whether pre­ cedent in norm creation builds as linearly in one direction, as occurred, for example, in the incorporation of the labor into party systems (Collier and Collier 2002). The competing hypothesis implies that legal decisions do not set a deter­ ministic path in motion, and are rather open to rights-based contestation and therefore also susceptible to course corrections.

102 The Choque de Trenes Via de Hecho: Legal vs. Constitutional Certainty The constituents of the 1991 Constitution identified the bifurcation of constitu­ tional jurisdiction between the State Council and the Supreme Court under the 1886 Constitution as a barrier to effective constitutional rights protection. The logical remedy was to unify the constitutional jurisdiction under one high court. Therefore, the Kelsenian framework with a sole, specific institution at the judicial helm to ensure observance of the constitutional hierarchy was a natural result of the desire to unify constitutional jurisprudence (Cepeda Espinosa 2007, 95). The choque de trenes had its origin in this desire of the constituents for unification and efficacy of constitutional rights enforcement, which had to clash with the institu­ tional interest of the other high courts to decide without the interference from the Constitutional Court. The basic normative contours of the choque de trenes were laid very early in the history of the new Constitution. It first appeared in the intermission period between the implementation of the new Constitution and the election of the first proper Constitutional Court. In the dissent of the intermediary Court’s decision C-543/92, Eduardo Cifuentes penned the doctrine of the via de hecho (“by way of fact”) that gave rise to the tutela contra sentencias (tutela against legal sentences; Aponte 1994). Explaining this doctrine further corroborates the normative connotation of the choque de trenes as well as the importance of pre­ cedent in Colombia’s new constitutional order that enables us to look at this process through the lens of path dependence. The via de hecho is a legal figure that intends to achieve two things: 1) uphold the Constitutional Court as the highest court in the constitutional order and thereby as the guardian of the integrity of the Constitution; and 2) impose the centrality of precedent for the jurisprudence of the 1991 Constitution. Therefore, it was a quintessential expression of the novel orientation of the Court to widen its interpretative perspective and become the unifier of constitutional jurisprudence (Cepeda Espinosa 2007, 97; López Medina 2000, 116). The figure of the via de hecho originated in the question of what constituted a cosa juzgada (“judged case”) in the new Constitution—a central component of legal certainty and its detestation for re-opening legal questions already settled in a concluded legal process. The rule of law in general, and the principle of habeas corpus in particular, accept only legal processes to establish “legal truths”. From the same premise fol­ lows the principle non bis in idem. Legal security dictates that an individual cannot be prosecuted and judged twice for the same crime—the prohibition of double jeopardy. The majority in C-543/92 argued with this principle against the tutela contra sentencias, stating that the principle non bis in idem was incompatible with tutelas against completed judicial decisions, because it constituted the reopening of a finalized process (C-543/92, 2). The dissent in that decision disputed this interpretation of the cosa juzgada visà-vis the procedure of the tutela. Magistrates Cifuentes, Angarita, and Martínez contended that the imposition of a time limit in judicial processes, implicit in the principle of cosa juzgada, is in no way disputed in the application of the tutela against judicial decisions. Rather, the tutela contra sentencias simply entailed that

The Choque de Trenes

103

in the case of fundamental rights violations the Constitutional Court suspended this time limit until after the conclusion of the tutela review. Since there could not be tutelas against tutelas, it did not constitute a prolongation of the legal process ad infinitum. Crucially, in the case of the tutela, the litigated issue was not the concern of the deliberation in the Court, but rather whether a fundamental right of the plaintiff was violated by a court’s decision. They reasoned that the con­ stitutional judge who revises a judge’s legal decision does not undermine the autonomy of the judge in taking her decision. After all, the condition for a tutela application as the ultima ratio for individual right’s protection still applied. Only if all ways in the ordinary legal system were exploited could a plaintiff submit a tutela against a legal decision. Therefore, the tutela contra sentencias did not create chaos beset with legal uncertainties but upheld the primacy of substance over procedure. The pre-eminence of the Constitution necessitates that its axiomatic principles superseded those principles lower in its hierarchy. The competence of the Con­ stitutional Court in its function to revise the tutela was reflective of its position as the final interpreter of the Constitution. Consequently, the Court had the duty of unifying national jurisprudence when it concerned fundamental rights. Logically, the minority then argued, the Court’s authority extended to all other jurisdictions, if questions of rights were involved (see salvamiento de voto in C-543/92). Critical for the dissent was the contention that the tutela redresses factual situations in contradiction to rights in the Constitution. Situations of facts (de hecho) occurred most prominently in the course of ordinary procedures and forms (Cifuentes 1998, 171). The tutela was perceived precisely to amend such situa­ tions of facts, where the ordinary procedure was not just insufficient to uphold rights, but also complicit in their violation. A simple example: a plaintiff’s pleas of innocence in a criminal case are confirmed by further evidence after the comple­ tion of the entire appellation procedure. A formal interpretation of the cosa juz­ gada excluding the tutela contra sentencias could not rectify the plaintiff’s rights violation – being in prison despite being innocent—because of its insistence on the formal completion of the appeal process as the sine qua non of legal certainty. A tutela, however, could, because, as a mechanism of rights litigation, it can also incorporate shifts in the legal context.12 The election of the first proper Constitutional Court in February 1992 added Carlos Gaviria, Antonio Barrera, and Hernando Herrera—all known nuevo derecho lawyers—to the early dissenters Cifuentes and Martínez (Angarita had left the Court). They favored substance over form when interpreting the constitutionality of tutelas and fully developed the doctrine of via de hecho in T-079/93. Building on the dissent of C-543/92 that outlined the substance of a factual judicial situa­ tion, they argued that an erroneous judicial decision, which diverted in such magnitude from the constitutional order, did not, properly speaking, constitute judicial providence. Rather, by not applying the stipulations of the constitutional order, and violating its most fundamental guarantees, the judge establishes her own judicial fact, a via de hecho (Cifuentes 1998, 171 and 174). Furthermore, they reasoned that a via de hecho could contravene two basic principles of rightsbased constitutionalism: 1) undermine the hierarchical ordering of norms and

104 The Choque de Trenes principles by placing form over substance; and 2) create its own legal order by diverting substantively from the valid constitutional order. These very early cases set the tone for the dispute over the application of the tutela. The public—and sometimes polemical—discussion of whether the novel mechanism applied to judicial decisions was not simply a matter of political authority and power. The debate involved fundamental, ontological, and constitutional con­ cerns that were reflective of the normative shift towards neo-constitutionalist prin­ ciples in the 1991 Constitution. The disparate sides in the debate on the tutela contra sentencias clashed over the pre-eminence of substantive constitutional claims over the formal application of legal rule. In other words: substance or form. Magistrate Alejandro Martínez explained it was reflective of the gap between a lit­ eral interpretation of the words of the Constitution, and the meaning of those words for values of justice (cited in Uprimny 2007, 59).13 Whereas opponents of the tutela contra sentencia argued that it created legal uncertainty by undermining the procedural soundness of constitutional adjudication, proponents of the tutela contra sentencias countered that their denial would result in an uncertainty of the Constitution itself by devaluing its most basic guarantees. The first part of this chapter served to provide a conceptual grounding of the analysis in the following sections. The admittedly more extensive route taken owes to the fact that the question of whether post-genesis institutional development in Colombia followed a path-dependent logic combines focuses on political and normative implications. Beginning with the manifestation that the creation of the 1991 Constitution was a critical juncture in Colombia’s history (as shown in the previous chapter), I explained the basic definitional traits of path dependence that theory suggests follow junctures of contingent change. Central to path depen­ dence is that early institutional decisions reproduce themselves and thereby create institutional stability, or even inertia. My contention of novelty here is to include normative ramifications of institutional change into theories of path dependence. For that end, I explore not only the institutional changes of the judiciary in the 1991 Constitution that are amenable to distributive effects, as (conventional) path dependence suggests, but also normative innovations of nuevo derecho and how these apply to the tutela. The legal figure of the via de hecho is of particular importance because it introduces the centrality of precedent into a civil law system to unify constitutional jurisdiction. Precedent setting has characteristics akin to path dependence, because earlier decisions also have an inflated importance for later ones, which have to address the argumentation set out in the precedent. It is therefore valid to hypothesize that jurisprudence evolves in a path-dependent logic. Early decisions invoking the via de hecho center around the importance of rights for the constitutional order in general and the tutela in particular. As seen, the centrality of rights in the argumentation mirrors the normative claims that underlie the implementation of the tutela in the constitutional order in the first place. It connects the practice of rights-based adjudication in Colombia with constituents’ grievances and intentions for drafting the Constitution. The next parts will analyze empirical cases, and focus on how the application of the tutela featured in cases involving the criminalization of Congress (the proceso

The Choque de Trenes

105

8000 in part II, and parapolítica in part III). I will then relate the normative dis­ pute of the tutela contra sentencias with institutional conflicts, centering on how those contentions within the judiciary affect the other branches of government. These are high-stake cases in a context without parliamentary immunity (see Figure 3.1). Even in Pablo Escobar’s heyday, institutional co-optation was largely contained to specific clusters, and only to an infinitesimal part captured national politics. López Hernández, for example, argues that in the epochs of large cartels the vote accumulated by Escobar amounted to only 0.2% of the national vote. During the proceso 8000, Senators eventually sentenced for accepting drug money had obtained 8% of the national vote. In the parapolítica scandal, Members of Congress indicted and sentenced for their relations with paramilitaries had accu­ mulated around 25% of the vote and 35% of seats in Congress (López Hernández 2010, 31). The proceso 8000 was the first political scandal evident of the increasing criminalization of Colombia’s political institutions. It essentially involved the bankrolling of the presidential campaign of the eventual winner of the 1994 con­ test, Ernesto Samper, with drug money from the Cali Cartel. The parapolítica scandal evolved around mutually benefiting informal relations between Members of Congress and paramilitaries. Politicians indicted in the course of the proceso 8000 directly benefited from drug money, and, while “parapolítica also involved money from narco-trafficking”, Members of Congress also “received collaboration by ‘coaction’; not only finances mattered, the paras coerced people to vote for the candidates, too”.14 Despite the apparent differences, the two cases warrant com­ parison, because the criminalization of Congress continued through the 1990s until the early 2000s, not least because Members of Congress were not held accountable for their transgressions during the proceso 8000. The fact that this did not occur was a consequence of the evolution of the choque de trenes itself. Judicial processes were stopped, because the Constitutional Court upheld legislative autonomy against the Supreme Court’s imperative to investigate legislators’ alleged criminal transgressions. Utilizing the tutela to blunt criminal investigations was a strategy Members of Congress also used during the parapolítica scandal. This time, however, to no avail. The Constitutional Court backed the Supreme Court’s jurisprudence in key tutela decisions. These two differing outcomes will enable us to see if the jurisprudential evolution did indeed follow path-dependent sequencing, or if we need to specify path dependence to properly understand what drove court decisions with strong normative implications that produce profound institutional ramifications.

Part II: The Proceso 8000 and the Inviolability of the Vote in Congress My analysis divides the proceso 8000 scandal, as the parapolítica scandal below, into five phases. For the proceso 8000 inquiry, I build on Hinojosa and Pérez-Liñán (2003) as well as Pérez-Liñán (2007), who traced Samper’s success to hold on to power to building a legislative shield against impeachment. He strategically dis­ tributed earmarked funds to tie opposition Members of Congress to his govern­ ment, essentially pulling the teeth from the impeachment process. The fifth phase,

106 The Choque de Trenes which I have added, is essential for understanding the transformation of the separation of powers in Colombia in this phase. It builds on my own content analysis of journalistic publications in Colombia (El Tiempo, El Espectador, and Semana) as well as my interview data. Together, the analysis of these five phases will disclose a field of strategic interactions that not only involved the executive and legislature, but also regional governors and illicit actors. Ultimately, it reflects how formal and informal institutions in Colombia have historically been inter­ twined. Importantly, it also connects the trajectory in Colombia’s majoritarian institutions with the normative evolution of the via de hecho, the tutela, and the choque de trenes. In this fifth phase, the constitutional jurisprudence in the litiga­ tion over parliamentary inviolability meandered between a limited perception based on the need to balance powers with obligations, and a more expansive conceptualization protected by parliamentarians’ civic right to freely deliberate. Empirically, the uncertainty over the deontological constrictions contributed to the stalling of investigations, and epistemologically, the trajectory contradicts the predictions of a path-dependent logic—evidence that will be further corroborated with the evolution during parapolítica affair when the normative argument shifted back to a more restrictive interpretation (see below). The table below (Table 3.1) provides a chronological overview of the scandal and the analytical periodization that I propose here.

Table 3.1 Phases of the proceso 8000 scandal Phase I

Phase II

Phase III

Phase IV

Phase V

June 1994– April 1995

April 1995– January 1996

January 1996– June 1996

June 1996– July 1998

April 1997–January 1999

- Publication of narcocassettes - External pressure by the US on Colombia - No effect on Samper’s popularity

- Opening of the proceso 8000 by Attorney General - Investigation of Campaign Treasurer - External and internal pressure - Effect on popularity ratings

- Implication of Minister of Defense and former Campaign Manager Botero - Investigation in Congress

- Absolution in Congress - “Fizzling away of Scandal” - Continuation of strained relations with US - Beginning of legal proceedings against Congresistas

- Investigations of Members of Congress for Breach of Duty - Tutela Complaint by indicted Members of Congress - Eventual ter­ mination of investigations with tutela decision

June 1994 (post-election)– April 1995

April 1995– January 1996

January 1996– June 1996

June 1996– July 1998

April 1997–Jan­ uary 1999

The Choque de Trenes

107

Phase I

Phase II

Phase III

Phase IV

Phase V

-

- Opening of investiga­ tion and production of legal facts - Normative debate over illicit enrichment

- Confirma­ tion of illicit enrichment as autono­ mous crime - Confirmation of double function of Members of Congress - Confirma­ tion of legal obligations to prosecute

- Submission of criminal complaints against Members of Congress - Continuous production of legal facts based on duty to prosecute

- Declaration of parliamentary inviolability of deliberations - Termination of production of legal facts

No legal investigations and no pro­ duction of legal facts

Source: Compiled by the author

The proceso 8000 spanned a little over four years, from June 1994 until January 1999, when the Constitutional Court essentially terminated the Supreme Court’s investiga­ tions of Members of Congress. The top row describes the most important political developments, while the bottom row traces the most important normative evolutions. In the next two sections, I will follow this differentiation and analyze the political evo­ lution, which encompassed domestic and international political ramifications. I will then retrace the normative evolution of the proceso 8000, focusing on the jurisprudence of illicit enrichment and the inviolability of the Members’ votes in Congress. I hold that, against the conceptual implications of the path-dependent framework, it is difficult to construe a linear line of argumentation from the earliest decision outlining the via de hecho to the decision clearing Members of Congress of criminal liability. Political Contentions: Samper’s Ability to Remain in Power The five phases of the proceso 8000 scandal in Colombia not only describe President Samper’s ability to protect himself by creating a legislative shield against impeach­ ment. They also indicate the increasing ability of regional power brokers to act autonomously of party hierarchy and pursue their own interests. Both Samper’s selfprotection from impeachment through pork barrel spending and regional power brokers’ empowerment are important caveats of the transformation of Colombia’s majoritarian institutions that were increasingly tied to illicit informal institutions of indirect governance in the country’s periphery (Gutiérrez Sanín 2019). The first phase of the proceso 8000, the “external phase”, lasted from the immediate aftermath of the 1994 presidential election until early 1995, when the political ramifications of the scandal were limited to Colombia’s relations with the US, but did not affect Samper’s domestic standing. The scandal had its origin in the need to have a run-off election in the 1994 presidential election, which required an extra three weeks of campaigning. The Samper campaign was caught off-guard and was short of funds. They turned to illicit money from the Cali Cartel, which had been involved in developing social capital with the political elites of the

108 The Choque de Trenes country to protect their business from criminal prosecution. They aimed to legalize the profits of their business (Hinojosa and Pérez-Liñán 2003, 6). Information of the illicit nature of campaign finances surfaced with the so-called “narco-cassettes” immediately after Samper’s victory in June 1994, in which journalist Alberto Giraldo and members of the Cali Cartel appeared to be discussing the influx of drug money into the Samper campaign. After being informed by police colonel Carlos Barragán, the defeated Pastrana camp took the evidence to outgoing President Gaviria and the US Embassy.15 At first, the public in Colombia viewed these revelations as the attempt to take revenge for a lost election, allowing Samper to distance himself from the accusations (Restrepo 1996, 47). Externally, however, the accusations were taken much more seriously. Throughout 1994 and the early part of 1995, pressure was mounting from the United States, with former DEA director Joseph Toft characterizing Colombia as a Narco-Democracy. The Senate in the US and the Clinton administration began the process of decertification of Colombia as punish­ ment for unsatisfactorily cooperating in the War on Drugs.16 During this first phase of the scandal various actors were unable to put effective pressure on the Samper administration. The revelations did not even significantly damage Samper’s approval ratings in opinion polls. As Pérez-Liñán reports, they remained at 79% until the end of 1994 (2007, 105; Hinojosa and Pérez-Liñán 2003, 68). There are a number of reasons why Samper was not immediately affected. For once, there was no direct relation drawn between the alleged drug money and him personally, and, secondly, he successfully posited himself as an anti-establishment candidate, opposed to the neo-liberal agenda of his predecessor César Gaviria (who came from the same party, but a different faction within that party; Pérez-Liñán 2007, 104). This enabled him, while not riding on a huge wave of public support, to enjoy a “presidential honeymoon” and even pass his “Social Leap” program—a gov­ ernment investment program aimed at increasing health care benefits, housing, and education for Colombia’s poor and improve their economic situation by strategically strengthening small businesses and the export sector (Hinojosa and Pérez-Liñán 2003, 72–74). In addition, Samper utilized two strategies to counteract international pressure by the United States. Firstly, he went on the offensive and initiated the cap­ ture of several members of the Cali Cartel, including its bosses, the Rodríguez Ore­ juelas brothers. Secondly, he protected his own position by calling on nationalistic sentiments against impositions of US imperialism and portrayed himself as a victim thereof (Gutiérrez Sanín 2007, 343–411). The second phase commenced when revelations created intensive internal pres­ sure on Samper. His luck changed when new Attorney General Alfonso Valdivieso opened investigations and remitted the names of Alvaro Benedetti, Jaime Lara, José Guerra de la Espriella, Alberto Santofimio, Armando Holguín Sarria, Ana de Petchal, Rodrigo Garavito, Yolima Espinosa, and María Izquierdo to the Supreme Court (under the number 8000, giving the scandal its name “process 8000”). Liberals suspended their investiture in Congress, but, most importantly, Valdivieso began to incriminate individuals close to Samper. When his campaign treasurer, Santiago Medina, was implicated in preliminary investigations in April 1995, the domestic situation began to unravel for Samper.

The Choque de Trenes

109

In July of that same year, Medina admitted that Samper’s campaign had accepted money from the Cali Cartel, resulting in his indictment and detainment for illicit enrichment as well as false testimony. In prison, Medina accused not only Minister of Defense Fernando Zea Botero (campaign manager during the presidential race), but also the President himself of being aware of the origin of the money in his campaign. Samper reacted by denying the claims and called on the Commis­ sion of Accusations and Investigations in Congress to look into the allegations against him. Medina’s narrations and the official (even only preliminary) investi­ gations increased the political cost for Samper dramatically, because accusations and public rumors now acquired legal meaning. As a consequence, his personal exposure to scandal increased considerably and his approval ratings fell to 45%. Clearly, any type of presidential honeymoon was over (Pérez-Liñán 2007, 105). In August 1995, the attorney’s office detained minister Botero and a month later Samper responded to the accusations in the Commission of Accusations. Representative Heine Mogol­ lón, who himself had admitted to the utilization of conspicuous funds during a campaign for Congress, was the head of that Commission. Not surprisingly, Mogollón suggested closing the investigations against Samper for lack of criminal evidence. However, this was not the end of the story.17 The third, and politically worst, phase for Samper began after Botero announced on television that Samper knew about the presence of drug money in his campaign, which prompted Samper to call on Congress to reopen the investigations. Internal pressure combined with external pressure, as US–Colombia relations reached their historical low point. The US decertified Colombia and revoked Samper’s travel visa. As a consequence of the increasing pressure on Samper coming from inside and out­ side of Colombia, his approval ratings further dropped to 36% (Pérez-Liñán 2007, 105). On 27 February 1996, the Commission of Accusation opened formal pro­ ceedings against Samper and suggested on 23 May 1996 that the accusations against Samper did not merit opening criminal investigations (the result in the Commission was 10:3). The Commission, however, did not have the final say and passed its sug­ gestions on to the full Chamber of Representatives.18 The full chamber convened on 12 June 1996 and voted with an overwhelming majority (111:43) against opening criminal investigation for lack of evidence. The fourth phase of the scandal commenced after these two votes had absolved Samper, which effectively took impeachment off the table and significantly reduced the internal political pressure. Externally, the United States maintained its position vis-à-vis Colombia and Samper for the duration of his time in office, but he did not face more substantive threats to his position as President. Hinojosa and Pérez-Liñán characterize this phase as a fizzling away of the scandal (2003). Samper’s success to appease Congress was contingent on his ability to construct a “legislative shield” by coalescing several factions within the increasingly frag­ mentized two-party system behind his leadership. In Congress, Samper success­ fully exploited the factionalism to convince not only the majority of his own, Liberal, caucus (divided into a Gavirista and Samperista faction), but also a majority from the other caucuses. In the vote in the House of Representatives, which was divided amongst 59% Liberals, 31% Conservatives, and 10% other

110 The Choque de Trenes parties (89% of Liberals, 43.8% of Conservatives, and 63.5% of third parties voted against impeachment). The Comisión Ciudadana that, above all, pork barrel spending from the co-finances fund “were ultimately instrumental in building support among opposition members” (Hinojosa and Pérez-Liñán 2003, 76–77). While the average Member of Congress received 3.9 billion pesos in co-financing funds for the period between 1995 and 1998, the average for opponents of impeachment received 4.1 billion (those in favor only received 3.4 billion on average). Samper’s survival was also helped because Colombian politics was in a process of rebalancing an equilibrium between different poles of socio-political power after the implementation of the new Constitution which had put in place some obsta­ cles to the way politics used to function under the old Constitution. As has repeatedly occurred in Colombia’s history, such rebalancing towards an equili­ brium eclipsed formal and informal institutions (see Chapter 2). There were three interdependent developments. First, the party system was in a process of implo­ sion, bringing the previously latent atomization of Colombian political society to the surface. Now, regional power brokers pursued their own interest independent of the national party. Second, the mid-1990s was a crucial juncture in the evolu­ tion of the drug trade, shifting from being dominated by centralized cartels to more decentralized armed groups exercising territorial control in peripheral regions of the country. Third, political elites in the regions needed paths to cir­ cumvent the national circumscription of the vote for the Senate inscribed in the 1991 Constitution to continue to benefit from their localized pool of votes. As consequence, Samper had no choice but to leave regional power brokers a lot of autonomy to manage the counterinsurgency strategy in Colombia’s periphery. They wanted a free hand to utilize armed campaigning to secure their own majo­ rities in elections. It was therefore not a coincidence that, as Samper’s situation became more contentious in Congress, regional Senators and Governors, among them Álvaro Uribe, pressed for the CONVIVIR program. Essentially, Colombia’s formal institutions linked up with informal institutions, resulting in a hetero­ geneous and functional symbiosis between legal and illegal forces (Acemoglu et al. 2013; Ávila Martínez 2010; Gutiérrez Sanín 2007, 112; Duncan 2006). Hinojosa and Pérez-Liñán (2003) as well as Pérez-Liñán (2007) are right to contend that with the decision to absolve Samper, the scandal essentially fizzled away and that, within the majoritarian institutions, Samper’s exposure to impeachment had been reduced to nil. This worked through the reallocation of funds and political concessions. However, to understand the implication for the jurisprudence of the tutela contra sentencias, we need to add the fifth phase, which involved the high courts and Congress. The dissatisfaction with Congress led citi­ zens to submit criminal complaints against Members of Congress for breach of duty with the Supreme Court that had the potential to produce tangible legal facts—as happened during parapolítica. However, when the Constitutional Court upended citizens’ criminal complaints with a tutela decision that protected con­ gresistas’ right to deliberate against criminal investigations, it also terminated the production of legal facts. As a consequence of the unification decision by the

The Choque de Trenes

111

Constitutional Court (SU-047/99), the Supreme Court investigated only a rela­ tively small number of Members of Congress for their involvement in receiving money from illicit sources and legal processes against Samper himself all but col­ lapsed. Senators and representatives enforced their right by petitioning the Con­ stitutional Court in a tutela against criminal investigations that the Supreme Court had commenced against them for the vote to absolve Samper. It is here where Samper’s resilience in office also relates to the choque de trenes and the tutela contra sentencias. Normative Contentions: Illicit Enrichment and the Inviolability of Parliamentary Votes The tension between the Supreme Court and the Constitutional Court over the application of the tutela crossed path with the proceso 8000 in the normative spe­ cification of the constitutional meaning of illicit enrichment (enricimiento illicito), and the inviolability of Members of Congress’s votes in the legislature (inviolabil­ idad parliamentaria). The publication of the narco-cassettes with journalist Alberto Giraldo resulted in the process to jurisprudentially specify illicit enrich­ ment. The vote in the Commission of Accusation and Investigation absolving Samper of any wrongdoing opened the question regarding the jurisprudential status of parliamentarians’ criminal liability while exercising constitutionally dele­ gated duties and functions. Both were tutela cases reviewing decisions by other judicial institutions. Regarding illicit enrichment, the jurisprudential issue con­ cerned whether illicit enrichment was an autonomous crime. In other words: did it require a legal sentence by a criminal court on the illegality of the Cali Cartel’s activities to prove the illicitness of the enrichment by politicians? The Constitu­ tional Court in its final decision agreed with the Supreme Court and the Attorney General’s office that illicit enrichment could be prosecuted autonomously.19 Per­ taining to the inviolability of Members of Congress, Members of Congress dis­ puted the constitutionality of criminal investigations by the Supreme Court that sought to investigate them for prevarication of their judicial duties when they absolved Samper. They saw the inviolability of their votes, inscribed in Article 185 of the Constitution, compromised. Critically, the jurisprudential evolution and contention turned on the question of when the Court identified a via de hecho or fabricated a cosa juzgada. In that trajectory, the Constitutional Court shifted between more or less extensive interpretations of the right to freely deliberate, nor could it offer a consistent interpretation of the temporality of the via de hecho and cosa juzgada. Ultimately, a split majority on the Constitutional Court protected Members of Congress from prosecution for the votes given in the Chamber and terminated further criminal investigation. The very first revelations of the proceso 8000, the narco-cassettes with a record­ ing of a conversation between journalist Alberto Giraldo López and Cali Cartel capo, Gilberto Rodriguez Orejuela, boasting that they had bought the presidency with 3.7 million US dollars, also raised the issue of illicit enrichment (Serrill 1994). Giraldo turned himself in and admitted that he had accepted money from

112 The Choque de Trenes the Cali Cartel. He explained that his task was to forward the money to both presidential campaigns, essentially making him the middleman between the illicit world of the drug trade and the licit world of Colombia’s political system. The question then became if and how the laws against illicit enrichment in Colombia’s Penal Code (Codigo Penal de Colombia, CPC) applied to his case.20 The CPC contained two important clauses determining the illicit and undue nature of enrichment. It defined the enrichment as an unjustified, direct or indir­ ect equity increase (incremento patrimonial no justificado), and as illicit, if the enrichment derived in one or another form from criminal activities (Art. 148 CPC). The defendants (Medina and Botero) argued that, strictly speaking, there was no equity increase, since the entire funds were spent in the campaign and did not enrich them personally. In addition, the clause “enrichment derived from an illicit activity” in the law requires that the origin of the money had to be, in legal terms, tied to a litigated crime. In other words, “B” and “C”, who have received money from “A” that they spend in a political campaign, could only be guilty of illicit enrichment if A had been sentenced for a crime—even if the origin of the funds is relatively clear. The Fiscalía argued against this interpretation. Electoral triumph, helped by “hot money”, was an augmentation of status and power, and therefore an equity increase. Furthermore, Attorney General Valdivieso posited that illicit enrichment was an autonomous crime not contingent on a previous legal decision having determined the illicit origin of the funds. A reasonable and documented certainty of the illicitness behind the funds sufficed. In other words, “B” and “C” can be prosecuted without “A” having been legally sentenced for the activity (Uprimny 1996, 103). The Supreme Court further legitimized the prosecution’s doctrine when it detained various Members of Congress—among them prominent Liberal politicians Rodrigo Garavito, Alvaro Benedetti, and Alberto Santofimio (Uprimny 1996, 104).21 Congress then attempted to evade prosecution in a sudden flip in their senti­ ment vis-à-vis the Constitutional Court. The Constitutional Court’s activist pos­ ture displayed in its first years had not been well received amongst Colombia’s traditional politicians in the legislature. They detested the Court as arrogant and usurping, contesting the appropriation of their functions as legislators (Uprimny 1996, 108). Nevertheless, in December 1995, first the Senate, and then the House of Representatives, passed legislation that made the Constitutional Court’s decisions obligatory for all other courts in the country; apparently taking the side of the Constitutional Court in the debate over the application of the tutela (Law 190 of 1995).22 This change of sentiment had an evidently strategic motivation. In a previous decision (C-127/93), the Court had upheld that illicit enrichment is a contingent felony. Making the Constitutional Court’s doctrine obligatory for all other courts, Congress essentially attempted to undermine the illicit enrich­ ment clause, barring it from application in the proceso 8000. The press termed Congress’s attempt to off-load the political cost for non-prosecution onto the Constitutional Court narcomico. 23 The Constitutional Court eventually concluded that the legislation was a selfserving attempt to give Members of Congress impunity from prosecution.

The Choque de Trenes

113

Following the passage of legislation in Congress, Giraldo submitted a tutela against his incarceration on the grounds that there had not been a conviction against the Rodriguez brothers for drug trafficking. Without this conviction there were no grounds to uphold his incarceration for illicit enrichment, which, he argued, unduly violated his due process rights. The Constitutional Court analyzed the constitutionality of the decrees and laws defining the felony of illicit enrich­ ment. In a split decision, the Court differentiated between “criminal activity” and “criminal record” and held that due process implications of Article 248 applied to the latter. Furthermore, the Court argued that illicit enrichment was autonomous because of its particularly injurious influence on public morale, requiring only one judge to decide on the illicit nature of the enrichment and on its undue purpose of the transaction. In July of 1996, it thereby upheld the Fiscalía’s and Supreme Court’s decisions (C-319/96; see also T-820/99).24 The more consequential normative debate concerned the legal contention over Members of Congress’s immunity from criminal prosecution that followed Con­ gress’s vote of absolution. It is here where the Constitutional Court, contrary to the predictions of a path-dependent logic of institutional development, executed a peculiar divergence from its earlier jurisprudential decisions on legislators’ right and duties. As explained, the pertinacity of this debate lay, in part, in the changes of the new Constitution itself that stripped Congress of parliamentary immunity.25 The specific claims arose because of the double function legislators exercised in Congress as lawmakers and their investigative, that is legal, functions vis-à-vis the executive (Article 174), creating a tension between the rights of a parliamentarian and the duties of public servants in Colombia’s legal system. In the legislature, Members of Congress have the right to freely deliberate, vote, and decide (Article 185). Public servants in Colombia’s judiciary, however, have the duty to duly investigate criminal claims. Colombia’s Criminal Code penalizes breaches as “pre­ varication” of public duty (Article 413, CPC, and Article 414, CPC). The criminal complaints citizens filed with Judge Jorge Anibal Gómez Gallego of the Supreme Court centered on “prevarication” of public duty. Senator Viviane Morales Hoyos, on the other side, argued in her tutela that the Supreme Court had vio­ lated her rights as legislator, because the failure to announce preliminary investi­ gations injured her due process rights and criminal investigations unduly infringed the inviolability of legislators’ votes and opinions. In the end, the Constitutional Court diverged from the argumentation of the Supreme Court and the Attorney General and ordered the termination of investigations against Members of Con­ gress for their votes to absolve Samper;26 controversially, it did so in contrast to its own previous decisions. The foundational precedents of the 1999 tutela were laid in 1994 (C-198/94) and in 1996 (C-037/96). Both decisions essentially confirmed Congress’s double function. In 1994, the Constitutional Court argued (in a unanimous decision) that the 1991 Constitution built on a long tradition of delegating legal powers to Congress for the accusation and indictment of the highest offices in the country. Already prior to the new Constitution, Congress had a role to investigate the Pre­ sident. The only new component of the 1991 Constitution was that Congress now

114 The Choque de Trenes also had the task of investigating magistrates of the Consejo Superior de la Judica­ tura (C-198/94). In the 1996 decision, the Court further stressed the importance of legislative and judicial functions combined in the congressional offices and affirmed a statutory law that established the Commission of Investigation and Accusation in Congress; the commission that probed Samper (C-037/96). C-222/96 and C-245/96, both in 1996, outlined public servants’ duty to investigate, upholding the liability that came with Congress’s legal functions. Magistrate Naranjo Mesa explained that legislators were never entirely free from obligations. On the contrary, the very same Constitution that delegates repre­ sentatives’ autonomy also imposes parliamentary obligations: political responsi­ bility to their constituents that elected them to Congress (Article 133), as well as a regime preventing conflict of interests, the misappropriation of public funds, and the peddling of influences (Article 183). The Court argued that in a system of separation of powers, each branch of government had authorities that are filled with certain powers and privileges. These powers and privileges, however, could not be limitless, because the purpose of a constitutional system is to limit and control the exercise of public power. Power and privilege, by necessity, require balancing with obligations. These obligations differ functionally amongst the dif­ ferent branches because of their role in a system of separation of powers: legisla­ tors are held to faithfully represent the interest of their constituents while prosecutors are held to dutifully investigate. Therefore, a consistent separation of powers model that delegates lawmaking and investigative functions to Members of Congress must also penalize breaches of legislative and prosecutorial duties.27 From this rather clear and unequivocal position, the majority in SU-47/99 in 1999 shifted to a different weighing of privileges and duties. Importantly, the Court did so by constructing a controversial interpretation of the via de hecho, a pro­ spective via de hecho. Already the selection of Viviane Morales’ tutela strained the relations between the Supreme and Constitutional Court. First, disagreement between Constitutional Court magistrates Vladimiro Naranjo and Fabio Morón about the selection of Viviane Morales’ tutela led the entire Sala Plena to move that decision to judges Antonio Barrera and Alfredo Beltrán, who affirmed the selection. Then the Constitutional Court was accused by the Supreme Court of currying favor by selecting the tutela. Viviane Morales was represented by former magistrate and President of the Court, Jorge Arango Mejia (from 1993–1998), as her legal council. Curiously, Arango Mejia sided with the majority and only dissented on the ques­ tions of the secrecy of the vote in the previous decision (C-245/96), but now argued that he had undervalued the double sanctioning of Congress through the disciplinary and penal code. This, he explained, was inconsistent with the inviol­ ability of votes inscribed in Article 185.28 After the selection of the tutela, three magistrates had to recuse themselves, because they had discussed generic legal questions concerning the inviolability of the vote in public: magistrates Jose Gre­ gorio Hernández and Vladimiro Naranjo, who had voted against the selection of Morales’ tutela, declared themselves impeded, and so did Alfredo Beltran.29 The decision itself was split. The majority opinion began its reasoning by addressing the precedent established in C-222/96 and C-245/96, arguing it

The Choque de Trenes

115

required more clarifications. The magistrates maintained that it was permissible to diverge from a previous path, and that the tutelas was, in fact, the most suitable mechanism. After all, they argued, tutelas can have inter partes or erga omnes effects and therefore require specification after having accumulated a variety of meanings. Unification tutelas were designed precisely to serve that purpose, allowing for the possibility to modify previous positions in a decision; be they obiter dicta or ratio. They admitted that the interpretation in SU-047/99 differed from the path taken in C-222/96 and C-245/96, but contended that those cases left essential values of the Constitution in danger; namely parliamentary inviolability, a central pillar of the social state of law and the separation of powers therein that embrace the free and democratic debate. As such, the Supreme Court investigations constituted a vía de hecho prospectiva, a prospective judicial decision that is contrary to the law and Constitution, and the course of action itself would constitute a res judicada. They affirmed Morales’s complaints and ordered the Supreme Court to halt all other investigations into Members of Congress’s votes (SU-047/99). It is my argument that this decision constituted a clear divergence from the direction that the Constitutional Court’s jurisprudence regarding the via de hecho and tutela contra sentencias had previously treaded; particularly if contrasted with the dissent that was penned by President of the Constitutional Court, Eduardo Cifuentes Muñoz. In his 93-page dissent, he argued that the majority in the Constitutional Court had declared a cosa juzgada by shifting away from the jur­ isprudence outlined in C-222/96 and C-245/96 by constructing the “via de hecho prospectiva” (a prospective factual act). As explained above, in Colombia’s constitutional jurisprudence a “via de hecho” was an illegal act already performed by a public administration, and the tutela the mechanism against such illegal acts by public authorities. In addition, the Court’s jurisprudence regarding the tutela, thus far, had implied that the tutela was not against a judicial decision per se, but against the arbitrariness of a legal decision. As such, the legal decision must have incurred factual, organic, or procedural defects of intolerable dimensions to the rule of law. By definition, a “via de hecho prospectiva” could not attain these qua­ lities for its lack of factuality, since it penalizes a potential violation occurring in the future. The Court’s President then warned that such unwarranted changes in the jurisprudence could fundamentally harm the legitimacy of Court, because it allowed Congress to slip behind the veil of inviolability of votes, and enjoy an undue amount of impunity that constituents did not envision.30 The Supreme Court for its part uttered bewilderment that a tutela had pre­ cedence over decisions of constitutionality and, referring to C-222/96 and C­ 245/96, could diverge this fundamentally from earlier interpretations on the same issue. To the President of the criminal chamber of the Supreme Court, Jorge Aníbal Gómez Gallego, it confirmed the legal insecurity arising from the tutela contra sentencias. 31 Gómez’s sentiment reflected a widespread sentiment of anger of the Supreme Court leveled toward the Constitutional Court. A Constitutional Court judge, who earlier had served on the Supreme Court during the proceso 8000, said in an interview with me that in the proceso 8000 “the Constitutional Court interfered through a tutela with a decision that was taken by the Supreme

116 The Choque de Trenes Court”.32 Another Supreme Court judge explained that while the tutela was a good tool to defend rights, “the Constitutional Court intervened in the decisions that the Supreme Court had taken, many times to revoke them or change them so the sentences did not do much”. Pointedly, he concluded that “this was grave in the 1990s in the investigations against congresistas. The Constitutional Court stopped investigations of 100 congresistas, and the process died there”.33 There are three concluding notes to be taken from the analysis of the proceso 8000. The first observation must be that without legal investigations and inde­ pendent prosecutions there would not have been a “proceso 8000, nor a trial against the president, nor a political crisis” (Uprimny 1996, 120). The discussion on the political evolution made readily apparent that benchmarks in the investi­ gations also constituted benchmarks in the political crisis facing Samper. Every time the judicial process moved forward against Samper, or one of his close advi­ sors in the campaign, pressure started to mount domestically. Since he could evade external pressure from the United States with relative ease, it meant that these were the more significant pressures that forced Samper to go on the offensive and spend valuable political capital—publicly and in Congress. Thus, this analysis already foreshadows a conclusion of this chapter; namely the production of legal facts—social facts validated in a judicial process—is important for the trajectory of political scandals. They have a more profound institutional effect than other socially produced facts (as the analysis of the unfolding of the parapolítica scandal during the Uribe years will confirm). In the end, Samper’s ability to shift the investigations into his potentially criminal behavior from judicial to majoritarian institutions—namely Congress—essentially stopped the production of legal facts. The second observation concerns what effectively undercut the production of legal facts. It is normative in nature. The eventual disagreement between the Supreme and Constitutional Court over key aspects of the norms outlining par­ liamentary inviolability and Congress’s duties to investigate criminal allegations terminated inquiries into those lawmakers’ nefarious relations. It is important to recall here that those lawmakers benefited from the evolution of formal and informal institutions in the periphery of the country that brought them into a heterogeneous alliance with non-state armed actors. At the time, when these parapolíticos entered these nefarious relations with paramilitaries, courts and legal institutions could not proceed with their investigations, because the Supreme and Constitutional Court had arrived at conflicting interpretations of the inviolability of the vote of Members of Congress. These Members of Congress could utilize the rights enforcement mechanism, the tutela, to evade criminal investigation. The next sections of the trajectory parapolítica will in fact show that both courts did not have a normative disagreement on the meaning of parliamentary inviolability in the context of those investigations. They continued. Finally, given the weighing of legislators’ powers with their duties in the final decision, and how it contrasted with the two previous constitutionality decisions, it is difficult to construct a line—let alone a straight line—that runs from the creation of the Constitution in 1991 to the 1999 tutela decision to protect par­ liamentarians against criminal investigations. These observations already cast doubt

The Choque de Trenes

117

over an understanding of jurisprudential trajectories as moving toward equili­ brium. As a preliminary conclusion, this episode already suggests that we need to understand institutional development of legal institutions as more open to change than path dependence demands. It is my argument that the parapolítica investi­ gations discussed in the next part will prove this point beyond reasonable doubt, because in that scandal the jurisprudence of the Constitutional Court shifted back to placing more value on the obligations of parliamentarians in the context of increasing (legal) evidence of the criminalization of Congress. Thus, legal facts will play an outsized role in that institutional trajectory as well.

Part III: Parapolítica and the Parainstitutionalization of Congress The parapolítica scandal is almost synonymous with the parainstitutionalization of Colombia’s political institutions: the co-optation of its democratic institutions by illegal armed groups at the margins of legality (López Hernández 2010; Garay Salamanca et al. 2010). AUC paramilitarism represented the conclusion of a pro­ cess that united drug traffickers with vigilante groups under the umbrella of a political project that arose and grew very powerful in the absence of the state (Gutiérrez Sanín 2019). Eventually, they formed functional and voluntarist coali­ tions with politicians in the territories that they dominated, which in the end benefited both sides, and were signed into informal pacts. Parapolítica essentially evolves around these pacts, but projects a meaning that goes beyond the signed deals. It denoted a particular form of domination and exercise of power: armed clientelism. Traditional, regional elites circumvented the formal rules of the democratic game to ensure their position of dominance by utilizing the groups’ coercive capacities (García Villegas and Revelo Rebolledo 2010; Eaton 2006; Dávila and Delgado 2001; Peñate 1999). With that, it constituted not only a peripheral grievance, but was at the root of the crisis of representation in Colom­ bia (Mainwaring et al. 2006; Bejarano and Pizarro 2005). In addition, para­ política has been closely associated with Uribismo, because of the high number of congresistas from the Uribe coalition that ended up being indicted by the Supreme Court (López Hernández 2010; Gutiérrez Sanín 2010). From the genesis of the scandal, its unfolding during Uribe’s second term, and its conclusion with numerous Members of Congress jailed, we can draw a number of inferences and findings. First, the evolution of the scandal is a strong example of how judicial activism can protect the quality of democracy by holding Members of Con­ gress criminally accountable, and redress issues of governance and representation. Second, the jurisprudential path corroborates the evidence that developments are not linear, but responsive to changes in the legal context. Third, the Supreme Court’s investigations produced legal facts about these nefarious relationships that were not only dangerous for Uribe’s political interest, but also helped in the creation of con­ stitutional meaning, when the Constitutional Court deliberated a potential third term. Legal facts are those “facts” that are resultant from a legal process and therefore have the potential to constitutes legal truths. It is for this reason that the production of legal facts is part of a process of constitutional meaning making.

118 The Choque de Trenes To provide an overview, the table below (Table 3.2) shows the five-part peri­ odization of the scandal.34 The table identifies the time bracket of each phase, and a condensed list of the most important events during that defined phase. The first phase lasted for the duration of Uribe’s first term. Negotiations for a demobiliza­ tion process between the government and the AUC commenced in 2002, which, by producing centrifugal forces, altered the internal hierarchy of the AUC, and produced accusations of nefarious relations between politicians and armed actors. These did not produce legal facts but only an official complaint. Official investi­ gations during the second phase produced the first legal facts when investigations were opened, and the first Members of Congress indicted on criminal counts. The third phase was characterized by Uribe’s fierce attacks against the Supreme Court that followed the Court’s opening of investigations of (the president’s cousin) Senator Mario Uribe and Senator Araujo. The fourth phase entailed the period when the Supreme Court stopped investigations of congresistas, who had resigned their legislative privileges. The recommencement of investigations in September

Table 3.2 Phases of the parapolítica scandal Phase I

Phase II

Phase III

Phase IV

Phase V

Before March 2006

March 2006– January 2007

January 2007– September 2008

September 2008–March 2009

March 2009– 2010

- AUC peace process - Infiltration of DAS by AUC - Atypical voting in the regions - Report filed with Supreme Court

- Auxiliary Court investigates parapolítica

- Salvatore Man­ cuso incrimi­ nates Mario Uribe

- Supreme Court stops investiga­ tions of Congresistas who have resigned their investiture

- Semana reveals the DAS wir­ etap scandal

Phase I

Phase II

Phase III

Phase IV

Phase V

Before March 2006

March 2006– January 2007

January 2007– September 2008

September 2008–March 2009

March 2009– 2010

- Constitutionality of Justice and Peace

- First indict­ ments of Members of Congress

- Supreme Court investigates Senators Mario Uribe and Araujo - Uribe begins public attack against Supreme Court - Extradition of paramilitaries

- No produc­ tion of legal facts

- Resump­ tion of Supreme Court investiga­ tions against resigned

Source: Compiled by the author

The Choque de Trenes

119

2009 initiated the final phase, which produced the most detrimental consequences for President Uribe. This part of the chapter proceeds similarly to the discussion on the proceso 8000 by first explaining the political evolutions in each Uribe term. It then moves to explaining the normative evolutions that evolved around the tutela, the cosa juz­ gada, and the via de hecho. I will explain how the Constitutional Court confirmed the Supreme Court’s position in the most important tutelas and investigations against Members of Congress continued as a consequence thereof. Political Contentions: The Politics of Parapolítica and the Production of Legal Facts For Uribe, the two terms in office evolved very differently. The first one appeared as an unmitigated success story, encompassing the (largely successful) roll-out of Uribe’s Democratic Security policy that fueled his popularity and justified con­ stitutional change to allow his second term, which he won in a landslide. The second term was marred in scandals that ranged from embezzlement of funds and other forms of relatively benign corruption to gross human rights violations such as the false positives. These scandals tarnished his government’s reputation, even though he personally still commanded extremely high approval ratings. Uribe’s first term covers the first phase of my periodization, while phases two through five evolved in the second term. The scandal arose complimentarily to the peace pro­ cess with the paramilitaries that commenced once Uribe took office. Incidentally, the peace process with the AUC, which was a public process (with the help of the Courts), destabilized the internal structure of the armed group, inflaming internal (violent) clashes, and ultimately subverting equilibrium relations between paramilitaries and politicians. It did not yet result in the production of legal facts, however. The negotiating phase of the peace process with the AUC in Uribe’s first term concluded with important conditions placed upon the constitutionality of its legal framework. Also, criminal complaints were leveled against Members of Congress for their companionship with paramilitaries. It was the implementation of the transitional justice process, and the criminal investigations stemming from those complaints, that involved judicial processes with deeper political ramifica­ tions reshaping relations between executive, legislature and judiciary in the pro­ cess. While it never threatened Uribe’s personal position in power, it forced him to spend a lot of political capital to fend off accusations, defend his majorities in Congress, and prepare the ground for the second constitutional reform process to prolong his time in office. In addition, this section will show that the fuel for the Supreme Court’s continuing curiosity about para-political relations were paramilitaries’ testimonies in Justice and Peace Courts, the truthfulness of which the Constitutional Court imposed as a condition for the constitutionality of the Justice and Peace Law, as well as anonymous accusations submitted by citizens to the Supreme Court. Equally important, there was a qualitative escalation in Uribe’s confrontation with the Supreme Court following the indictments of Senators Álvaro Araújo Castro and Mario Uribe (the President’s first cousin). Because of

120 The Choque de Trenes these investigations, the President began to overtly attack the Supreme Court and covertly design intimidation campaigns threatening magistrates in their personal lives. Thus, there is an evident correlation between the continued production of legal facts by the Supreme Court and the amount of political capital spent by Uribe to delegitimize the Court. Finally, when the covert campaign against Colombia’s high courts became known, the Supreme Court reversed a critical jurisprudential position, retaking the bulk of parapolítica cases of Members of Congress. Uribe rose to political stardom in the 2002 election on a “tough hand, open hearts” ticket. His aggressive posture against the guerrilla movement partly stem­ med from personal experience (he alleged the FARC had assassinated his father), and partly from structural vestment (his family belongs to the wealthy cattle ranching class). In Antioquia, he had built a reputation for focusing on security and calling for the privatization of security through the CONVIVIR groups. His Democratic Security program had three elementary imperatives that formed the core of the public policy in his presidency: 1) establishing the clear authority of the state by amplifying its monopoly of violence against internal enemies; 2) fostering a direct relation between rulers and ruled to assuage social cleavages and create a strong bond of social cohesion; and 3) buttressing of capitalistic institutions to invite foreign direct investment (Botero Campuzano 2008; Gaviria Vélez 2004). During the campaign, a particularly potent move was to focus on his bellicose position vis-à-vis the FARC. In early 2002, the FARC had let the peace negotia­ tions in Caguán fail, and kidnapped presidential candidates Ingrid Betancourt and Clara Rojas. The notion that “peace is the daughter of authority” seemed to be pristine for the moment (Uribe, in Botero Campuzano 2008, 103), and helped Uribe’s rise in the polls from 9% to 53% within four months.35 Once in office, Uribe followed up on his promise to bring peace to the nation with two contrasting strategies for the state’s internal competitors. While he favored a militaristic strategy vis-à-vis the FARC and other guerrilla groups, he wanted to involve the paramilitaries in a negotiated settlement. Already in 2002, preliminary conversations commenced, and shortly thereafter Luis Carlos Restrepo was appoin­ ted High Commissioner for Peace to facilitate formal negotiations. The Santa Fe de Ralito I agreement from 15 July 2003 marked the beginning of said negotiations, and the Santa Fe de Ralito II agreement from 13 May 2004 marked the first benchmark. It set up a “concentration zone” (zona de ubicación) in Tierralta for the purposes of suspending arrest warrants for AUC members, verifying a ceasefire, and preparing the demobilization and disarmament process under the auspices of the OAS. Finally, the peace process between the Colombian Government and the AUC culminated in the Justice and Peace Law (Law 975 of 2005) passed by Congress 22 July 2005. It provided the legal framework for the demobilization and subsequent transitional justice process (Theidon 2007, 72–73).36 These negotiations did not proceed without quarrels and conflicts. On the con­ trary, indiscretions and public pronouncements with conflict potential surfaced reg­ ularly. Already in 2002, Salvatore Mancuso, one of the most powerful commanders of the AUC, publicly declared, without going into further detail, that around 35% of the recently elected Members of Congress were “friends” of paramilitaries.37 Also

The Choque de Trenes

121

as early as 2002 and 2003, left-wing opposition Senator Gustavo Petro denounced the influence of the paras in Congress and bureaucracy. As these stories were gain­ ing traction in the national and international media, the closing of the Santa Fe de Ralito II agreement was followed by one of the most controversial episodes of the Justice and Peace Process: the paramilitary visit to Congress in Bogotá. In July 2004, Salvatore Mancuso held a speech outlining the AUC’s interests in producing a stable peace. Not only was it very audacious for a certified terror organization, culpable of war crimes, crimes against humanity, and narco-trafficking, to visit the center of Colombia’s democratic institutions, but the “state visit” also disclosed some very obnoxiously intimate relations between commanders of the group and certain politicians. For example, representatives Rocío Arias (from Caucasia) and Eleonora Pineda (from Tierralta) not only awaited Mancuso at El Dorado Airport and organized the transport to the Plaza Bolivar in downtown Bogotá; they also arranged to buy their wardrobe to wear in the legislature.38 The courting not only caused anger amongst victims of paramilitary violence; it also raised suspicions about the exact nature of the relation between paramilitarism and politics. The public began to wonder how Arias and Pineda could become such central political figures. Prior to their election in 2002, they had only managed to win in small municipal elections and effectively increased their vote share by over 1,000% in the 2002 congressional election. In 2005, Claudia López found more such “atypical” election results, where majorities apparently shifted overnight and bordering electoral districts had entirely contradictory results even though the socio­ political composition was virtually identical. Finally, Vicente Castaño, in an interview with Semana, confirmed Salvatore Mancuso’s claim that 35% of Congress were elected with paramilitary help.39 As the indiscretions and peculiar publicity stunts were shoring up public wariness about the synergies between politicians and paramilitaries, the internal structure of the AUC, too, was shaken by the increasing public exposure. As described, the AUC only constituted an umbrella organization of various self-defense groups and warlords. It left the regional command structures under the control of those local warlords, who therefore had every opportunity to pursue their own interests. This fragmented structure became readily apparent in the early 2000s, when various fac­ tions within the AUC were fighting an internal war over the role of narco-profits in the financial structure of their organization. Carlos Castaño and Carlos Mauricio García (alias “Doblecero”) were both founding members of the vigilante project as part of the Pepes. In addition, they were driving forces behind the political evolution of the paramilitaries, with Castaño speaking on national television to legitimate their aspirations. In the early 2000s, as the first negotiations between the government and the AUC had commenced, “Doblecero” publicly decried the influence of narco-traffickers in the paramilitary project, and Carlos Castaño was said to have been negotiating with the government and the DEA to exit from the mountainous jungle to a secure exile. The fellow commanders of the various blocs turned against them in the fear of becoming the price for their exit from the narco-vigilante business. Wars erupted in the remote regions of the country, and members of the organization they helped to create assassinated both of them—Carlos Castaño by the order of

122 The Choque de Trenes his brother, Vicente, no less.40 This clash of interests became explosive exactly at the moment when the AUC was seeking public approval for their organization. Clearly, the internal structure of the AUC—above all its financial structure—was irreconcilable with a convincing public relations campaign.41 As Uribe’s first term concluded and his government’s attention rested on the faith of the Justice and Peace legislation as well as the re-election reform legislation in the courts, a number of events unfolded, which would prove to be very con­ sequential in the second term. Clara López Obregon submitted an official complaint to the Supreme Court to investigate the rumors and establish whether they amounted to indictable offenses in June 2005. This became the catalyst of para­ política (Gómez Quintero 2007, 40). The official complaint led investigators to uncover the infiltration of DAS by elements associated with paramilitary forces. DAS is the internal security agency directly responding to the President’s office. Rafael Garcia, who was the chief IT administrator, had become the focus of investigators, because he was caught erasing criminal records of paramilitaries. In December 2005, he started to talk about the close collaboration between the DAS and the AUC. Even more, he confessed that paramilitaries campaigned with Uribe’s allies in the region.42 This put the spotlight on the agency’s director, Jorge Noguera, who organized Uribe’s election on the Caribbean Coast in 2002, which in turn secured him the directorship of DAS. Noguera became one of the closest political allies of Uribe indicted and sentenced for collaboration with paramilitaries (in 2011, the Supreme Court sentenced him to 25 years in prison). In April 2006, the Constitu­ tional Court declared the Justice and Peace Law constitutional for the most part, but imposed critical conditions, above all victims’ rights to the truth. Furthermore, it excluded, for procedural reasons, the felony of “conspiracy to commit a crime” from the list of crimes falling under the Justice and Peace jurisdiction (C-370/06). The effects of these decisions would only fully surface in the second term, because already in March 2006 congressional elections provided Uribe with an even stron­ ger majority in both chambers of Congress, and two months later, he himself was re-elected with 62% of the vote. The political evolutions in Uribe’s first term coincide with what I understand as the first phase of the parapolítica scandal. The negotiation phase of the Justice and Peace Process with the paramilitaries set in motion centrifugal forces within the paramilitary organization. The decentralized organizational structure of the AUC could not contain the pursuit of self-interest by regional commanders, and inflic­ ted lethal consequences on the leadership that purported to pursue political over criminal goals. These negotiations also put tensions on the relations between politicians and paramilitaries—particularly, when they were becoming public and increasingly accompanied by spectacle. As paramilitaries gave interviews and spoke in Congress, the media started to question unusual election results, and politicians found themselves increasingly pushed against the wall. For example, aforemen­ tioned Arias and Pineda were purged from their caucus lists prior to the 2006 elections, and ostracized by all other lists in Congress for fear of the contagion they would bring with them (see below). It is important to note, however, that public scrutiny certainly put a stress on the equilibrium between paras and

The Choque de Trenes

123

Members of Congress, but it did not fully break these symbiotic relations. It is my argument that the courts’ activism largely played out in Uribe’s second term, which was crucial for turning the interest alignment between AUC leaders and Members of Congress into a misalignment. Critical were Clara López Obregon’s official complaint with the Supreme Court, and the Constitutional Court’s April 2006 decision imposing the conditionality on the Justice and Peace Law (Law 975 from 2005) that victims’ rights to the complete truth were upheld (Ramírez Bas­ tidas and Socha Salamanca 2007, 53–54; García Villegas et al. 2010, 324–325).43 This clause entailed that non-compliance would result in foregoing the possibility of being sentenced under the Justice and Peace jurisdiction, almost certainly leading to long prison sentences given the atrocities committed by paramilitaries. Accordingly, the incentive structure for AUC commanders was tilted towards revealing the nature of the secret arrangements with politicians. The Supreme Court established the auxiliary court specifically tasked with investigating parapoliticos in September 2006 (after the election), as a response to the Constitutional Court’s affirmation of the imperative to know the truth (Gómez Quintero 2007, 41). It justified its proceedings against Members of Congress exactly on the clau­ ses containing victims’ right to truth in C-370/06 and C-454/06 (Ramírez Bas­ tidas and Socha Salamanca 2007), inviting citizens to submit anonymous accusations producing a continuous stream of leads to prosecute.44 Their testi­ monies in Justice and Peace Courts together with anonymous allegations filed by citizens were critical in helping the Supreme Court in its investigations and indictments of Members of Congress.45 These are the legal facts that had impor­ tant institutional ramifications and lead me to understand the next developments as different phases of the parapolítica affair. The affair attained a decisively legal characteristic, resulting in the second phase of the scandal, after the first Members of Congress (Jose Enrique, Eric Norris, and Alvaro Garcia) received subpoenas to appear before the Court. The prospects of official prosecutions and the Court’s activism caused turmoil amongst Members of Congress but did not make the President exceedingly nervous. Aforementioned Representatives Rocio Arias and Eleonora Pineda were purged from all lists in Congress and failed to gain entrance to other lists of the Uribista alliance in the legislature. They tried, to no avail, to enlist their “Let the Monkey Play” move­ ment (its actual name) in Mario Uribe’s Partido Colombia Demócratica. Evidently, their proximity to the leaders of the AUC had become an outsized liability in the face of Supreme Court investigations. In this period emerged a key piece of evidence: a computer belonging to paramilitary leader “Jorge 40”, which the Attorney General’s Office had recovered in a raid in March of 2006. After finally breaking the sophisticated encryption code in September, the hard disk became one of the central pieces of evidence that incriminated the AUC leader in crimes against humanity, detailed his violations of Justice and Peace provisions, and exposed the scope of relations between politicians and illegal armed groups. Most importantly, the files mentioned several strategic meetings between the leadership of the AUC and influential politicians, regional and national, that took place in and around

124 The Choque de Trenes the city of Santa Fe de Ralito.46 These strategic meetings are a critical element of parapolítica, because they reflect the scope of the symbiotic relations between armed groups and politicians, making the parainstitutionalization of Colombia’s democratic institutions readily apparent. The content of negotia­ tions that took place in Cordova were revealed soon after the discovery of the computer. The way it surfaced speaks to the almost comical dimension of Colombian politics at the time. In a meeting with the President, Senators Miguel de la Espriella and Juan Manuel Lopez Cabrales (from Salvatore Man­ cuso’s home province Cordova) accused each other in a slapstick-like manner of being more deeply involved with the paramilitaries. The minutes of this event were leaked to the press, resulting in the Supreme Court asking further questions about their respective relations with the AUC. 47 Being exposed in public, Senator de la Esperiella confessed to the media that the meetings had explicit strategic purposes that served the interests of both sides: politicians wanted to be elected and paramilitaries wanted to build a national strategy to avoid extradition after demobilization. Even more, he implied that Uribe’s first election to the presidency in 2002 resulted because of those relations.48 Finally, in January 2007, Salvatore Mancuso in his testimony before the Attorney General’s office filled in the missing details and incriminated numer­ ous high-ranking politicians in an accord signed between them and the lea­ dership of the AUC. This became known as the Pact of Ralito, wherein the AUC leadership and politicians proclaimed the intention to refound the nation (López Hernández 2010).49 Graph 3.1 below indicates the importance of these events for the Supreme Court investigations. It combines the number of preliminary investigations of the Supreme Court, the calls of Members of Congress to appear before the Court, and the incidents when Members of Congress were sentenced

Graph 3.1 Number of parapolítica investigations per quarter, 2006–2011

4th Quarter 2011

3rd Quarter 2011

1st Quarter 2011

2nd Quarter 2011

4th Quarter 2010

3rd Quarter 2010

1st Quarter 2010

2nd Quarter 2010

4th Quarter 2009

3rd Quarter 2009

1st Quarter 2009

2nd Quarter 2009

4th Quarter 2008

3rd Quarter 2008

1st Quarter 2008

2nd Quarter 2008

4th Quarter 2007

3rd Quarter 2007

1st Quarter 2007

2nd Quarter 2007

4th Quarter 2006

3rd Quarter 2006

1st Quarter 2006

20 18 16 14 12 10 8 6 4 2 0 2nd Quarter 2006

Number of Investigations

Parapolí�ca Inves�ga�ons

The Choque de Trenes

125

(excluded are acquittals). It shows that during the fourth quarter of 2006 the number of investigations rose significantly and, after a slump in the first quarter of 2007, augmented dramatically and persistently in the second quar­ ter of 2007. Graph 3.2 below indicates the reporting on Uribe and his allies in Congress by tracing the domestic news pages of Colombia’s most important weekly, Semana. While negative and positive reports were essentially even in 2006, in 2007 almost twice as many articles in Semana reported negatively than positively about Uribe and his coalition in Congress.

Quality of Discourse Positive/Negative Ratio

2.5 2 1.5 1 0.5

December 2007

October 2007

November 2007

September 2007

July 2007

August 2007

May 2007

June 2007

April 2007

March 2007

February 2007

January 2007

December 2006

October 2006

November 2006

September 2006

July 2006

August 2006

May 2006

June 2006

April 2006

March 2006

January 2006

February 2006

0

Graph 3.2 Quality of discourse vis-à-vis Uribe administration, Semana 2006–2007

Up until this point, Uribe was relatively reserved regarding the ongoing devel­ opments. They did not seem to affect his personal popularity with Colombians— and never did throughout his presidency. As Graph 3.3 below shows, his popu­ larity took a small hit from the implication of numerous Members of Congress that supported his government, yet it remained at a very high level; 80% of Colombians approved in mid-2006 (prior to his election), and approval only dipped to the mid-60s before climbing significantly again in March 2007 (to unprecedented heights in 2008; see Graph 3.3). Regardless of consistently high, and in fact improving, approval ratings in 2007, his government fundamentally shifted its response to the Supreme Court investiga­ tions from relative indifference to a concerted effort to undermine the Court’s efforts and legitimacy. Ostensibly, two investigations initiated by the Supreme Court in this period triggered this change in strategy that eventually crossed the line of

126 The Choque de Trenes

Approval Ra�ngs

Sep-09

Sep-08

Mar-09

Mar-08

Sep-07

Sep-06

Mar-07

Sep-05

Mar-06

Sep-04

Mar-05

Mar-04

Sep-03

Disapproval Ra�ngs Sep-02

90

80

70 60 50 40 30 20 10 0

Mar-03

In percentage

(Dis) approval ra�ngs

Month and Year

Graph 3.3 (Dis)approval ratings of President Álvaro Uribe, 2002–2010

legality. In February 2007, the Supreme Court opened an official inquiry against Alvaro Araújo, brother of the Chancellor in his administration, María Consuelo Araújo. The DAS technician Rafael Garcia incriminated him and laid open his relations to paramilitary leader “Jorge 40”. The second case involved Senator Mario Uribe, head of the Partido Colombia Demócratica in Congress and one of the most powerful leaders of the President’s Uribista bench in Congress. He appeared on investigators’ radar after Salvatore Mancuso’s testi­ mony in the Justice and Peace chambers of the Supreme Court, who denun­ ciated several political allies of paramilitaries. Concretely, he implicated the President’s cousin in two meetings with him prior to the 2002 election that were held to facilitate cooperation before and after the election that gave Uribe the presidency.50 The commencement of these investigations resulted in what I term the third phase of the parapolítica scandal as the government’s approach became noticeably aggressive, involving an evasion and intimidation campaign, consist­ ing of four different strategies to avoid or thwart the ongoing investigations: 1) Uribe repeatedly launched public and personal attacks against the Supreme Court and its members, which Human Rights Watch described as a “concerted campaign to smear and discredit the Court”;51 2) he proposed constitutional reforms with the intention to remove parapolítica investigations from the jur­ isdictions of the Supreme Court (none of these attempts succeeded); 3) he blocked and opposed meaningful efforts to reform Congress in the so-called “silla vacia” affair (empty seat); and 4) he extradited the main AUC leaders. The concerted effort by the executive to delegitimize the Supreme Court is the qualitative difference that characterizes the third phase of the parapolítica affair. Each of these components of the strategy deserves at least a brief explanation:

The Choque de Trenes

127

1) Personal Attacks When the Supreme Court officially opened Mario Uribe’s case in September 2007, the President started to get personally involved and take action.52 Accord­ ing to a judge on the Court, these investigations created an incredibly tense situation that made the declaration of a state of emergency to terminate the Court’s investigation a possibility.53 He first inquired about a paramilitary named “Tasmania”, whose area of action was in the Southeast of Antioquia—the home province of the Uribe family. On 8 October 2007, only a few days after the Senator’s official summon, President Uribe accused the primary investigator of parapolítica, Iván Velásquez, of having paid Tasmania to smear the Senator in an orchestrated campaign that eventually was supposed to hit the President himself. It was a clear attempt to paint the accusations as politically and economically motivated and reliant on obscure sources.54 The attacks against the Supreme Court and its auxiliary unit tasked with the parapolítica investigations followed this pattern. First, the President, often personally, tried to taint the evidence to be the result of bribes and irregularities, which most often involved individuals related to paramilitaries and/or the drug trade. Then the President attempted to portray the Court as acting politically against his administration, being motivated by interests that have their roots in institutional clientele networks aimed at guaran­ teeing jobs for friends within the judiciary. In these cases, the Supreme Court reacted by backing its magistrates and asking the Attorney General to investigate the claims. The Attorney General cleared the magistrates of wrongdoing and the accusers themselves admitted to having orchestrated the accusations. Furthermore, the witnesses also testified that Uribe’s advisor, José Obdulio Gaviria, and Mario Uribe were behind the smear campaigns against the magistrates.55 2) Judicial Reform In July 2008, Minister of Justice and the Interior Fabio Valencia proposed a judicial reform, which amongst other items would have transferred the authority to conduct criminal investigations of legislators to the Attorney General’s office, making the Superior Tribunal Court in Bogotá the trial court and the Supreme Court’s criminal chamber the appellate court. The government justified the initiative with the attempt to depoliticize the courts and address the lack of an appellation instance for Members of Congress. Critically, the reform proposal included a norm imposing that its stipulation would apply retroactively, including to the ongoing parapolítica investigations. It was sugar-coated with proposals to increase magistrates’ payments and pensions. However, the judiciary unequivocally opposed the reform. Observers as well as former politicians also described it as an effort by the government to offer its coalition in Congress impunity (Human Rights Watch 2008). Finally, the remaining Members of Congress in the legis­ lature feared that aligning with the government’s proposal would aggravate the Supreme Court and place them in danger of being investigated. Thus, the pro­ posal eventually died in Congress, because Uribe and his minister had

128 The Choque de Trenes underestimated the strong opposition of the entire judiciary and “cowardice” of Congress (Uribe’s words).56 3) Reform of Congress and the Empty Seat (La Silla Vacia) Reform of Congress acquired the palpable name “La silla vacia” (the empty seat) for the vacuum indicted Members of Congress left in the caucuses of affected parties and movements. Thus, the question became what to do with those empty seats, and it had deep political and normative connotations; normatively, because it touched on fundamental democratic principles, since punishing the parties and Members of Congress for their misgivings and purging their seats also punished the electorate, and inhibited, post-factum, their voting right. Allowing caucuses to fill the seats at gusto, on the other hand, would have set incentives that alliances with criminal forces had no negative implications for the party organization. For the executive, the issue was critical, because Uribe relied on his majorities in Congress for passing legislation (in particular another re-election reform), yet to boldly push through his interest was dangerous considering that he himself was increasingly pressed on his relations to illegal forces in the country.57 His govern­ ment undertook an attempt in 2008 to press reform with some penalties for the parties of indicted Members of Congress, but with stipulations on the empty seat norms that would carry him through that legislative period. The reform died in the Senate, because the self-removal of indicted members had decreased the number of Members in the first committee below the threshold required for leg­ islation to move to the second stage. Thus, the Supreme Court’s investigations, which resulted in Members of Congress resigning their investiture (see below), effectively killed the bill in Congress.58 Eventually, Congress passed reform in 2009, which not only included the empty seat legislation, but also several provi­ sions regulating the appointment practice in public offices. Most notably, the selection of family members was drastically curtailed.59 4) Extradition of Paramilitaries In all of this, it must have dawned upon Uribe and his administration that the paramilitaries were a profound source of uncertainty. Salvatore Mancuso and alias “H.H.” were the most willing AUC members to reveal their knowledge before courts. At the beginning of 2008, “Don Berna” and “Jorge 40” also signaled that they might be willing to talk to investigators about their relations with politicians. Evidently, the President and accused Members of Congress from his bench in Congress had a profound interest in cutting this constant supply line of revela­ tions. Extradition to the US proved to be the solution to their problems and drug trafficking charges provided the suitable pretext and justification. In May 2008, Uribe’s government decided to transfer almost the entire AUC leadership to the United States (Human Rights Watch 2008, 6). This proved to be the most suc­ cessful move to slow down revelations, because, even though some paramilitary leaders provided further details from US prisons, their admissions never acquired

The Choque de Trenes

129

legal meaning since this did not occur in Colombian courts. Mancuso, for exam­ ple, stated that he and his armed men had helped Uribe’s elections on various occasions, but declined to submit his testimony officially in court for fear of retri­ bution against family and lawyers that remained in Colombia. The shift to the fourth phase of the parapolítica was caused by the Supreme Court’s reaction to the strategies employed by Members of Congress to avoid prosecution and resign from their investiture. At first, the Court declined to fur­ ther investigate those Members of Congress, because it assumed it had only immediate standing vis-à-vis active Members of Congress. The fifth phase com­ menced when the Court changed its jurisprudence and retook these cases (see below). The next section will specify the details of the normative contentions; here it suffices to quickly summarize the tactics Members of Congress employed to evade investigations. First, they attempted the strategy utilized in the course of the proceso 8000: submitting tutelas on the grounds that criminal prosecution violated their parliamentary prerogatives (which the Constitutional Court rejected; see below). Then they complained—also via tutelas—that the lack of an appeal instance violated their due process rights. As explained above, the Constitution replaced parliamentary immunity with the special privilege to face accusations in the Supreme Court rather than the lower courts. Finally, when the Constitutional Court dismissed this argumentation as well, the parapoliticos attempted to protect themselves by resigning their investiture to be submitted to the ordinary justice system. This resulted in the fourth phase, between 2008 and 2009, when the Supreme Court decided to refrain from investigating and the numbers of inquiries reduced (see Graph 3.1 above).60 Tellingly, Members of Congress submitted their rights appeals with the disciplinary chamber of the Superior Council of the Judi­ ciary, whose members are elected in a highly politicized method (by representa­ tives from lists put forward by the President; see Figure 3.1). This turned out to be at least temporarily successful (Revelo Rebolledo 2009).61 While the parapolítica investigations were slowing down during this period, the confrontation between President Uribe and the Supreme Court remained acri­ monious. Notably, parapolítica was not the only source of profound hostility between the Supreme Court and the executive. Deeply damaging for his legiti­ macy, and therefore particularly toxic for inter-branch relations, was what would become known as Yidispolítica. It evolved around former representative Yídis Medina’s role in the first re-election legislation. Medina had gained notoriety in her role as a replacement for Representative Díaz Mateus from Santander in 2004. Together with Teodolindo Avendaño, who had replaced Arcila Córdoba, she cast the decisive vote at the congressional committee stage to pass the first re-election reform. The turn of events had already caused suspicions, because only a few days before that vote, they both appeared to be opposed and could only cite frivolous arguments such as divine intervention for their impulsive change of mind.62 In April 2008, however, journalist Daniell Coronell published Medina’s confessions in his column for the weekly Semana. 63 She admitted to having accepted bribes from Uribe and his minister Sabas Pretelt for her vote (López Hernández and Sevillano 2008, 78). As a consequence, the Supreme Court issued an arrest

130 The Choque de Trenes warrant for Medina later that month, who then officially confessed and incrimi­ nated Avendaño in Court as well. In May, Avendaño was arrested and made similar confessions, which resulted in the final arrest of Iván Días Mateus. In June of that year, the Supreme Court ruled that Medina did in fact act illegally by sell­ ing her vote to the government, effectively casting doubt over the entire reform process that gave Uribe his re-election. Uribe responded with according fury, and accused the Court of meddling in political affairs, even suggesting to re-submit the first re-election reform to a referendum by the people—knowing full well that his term would be approved given his extremely high popularity rating at his point, when the military had just liberated Ingrid Betancourt (see Graph 3.3 above). His plans eventually became superfluous, because the Constitutional Court decided in July of 2008 that it could not re-open a case without committing a cosa juszgada. New evidence did not suffice; the context needed to have substantially changed.64 When the Supreme Court altered its jurisprudence in March 2009 (Process No. 27032) and re-assumed the cases of resigned members of Congress, the scandal entered its fifth and final phase. It was a response to the executive’s most aggres­ sive tactics. It kept tabs on several institutions, including the Supreme and Con­ stitutional Court, and harassed judges of the Supreme Court with the help of obscure forces from within the bureaucracy (DAS; see Human Rights Watch 2008, 5). The foundation for that escalation, though, was laid earlier. When the Supreme Court took Senator Araújo into custody, the executive initiated the chuzadas (wiretaps), employing the DAS to install tabs in the rooms where the plenary chamber met to discuss matters concerning the whole court.65 In March 2009, Semana revealed that DAS had kept wiretaps on seven of the nine judges in the criminal chamber of the Supreme Court. In addition, judges reported that their family and friends were followed by obscure persons and approached by individuals with questions about their activities (later revealed as DAS agents). Furthermore, the weekly disclosed that DAS listened to over 1,900 phone calls made by Iván Velásquez, the auxiliary judge responsible for the parapolítica inquiry.66 Over the next months, the DAS wiretapping scandal exposed a network of relations that reached all the way up to the presidency, implicating former DAS director Jorge Noguera, and his successor María del Pilar Hurtado.67 The heads of the security agency are directly subjected to the President’s office and in the trials against former DAS agents, witnesses have repeatedly incriminated Uribe’s former Chief of Staff Bernardo Moreno and José Obdulio Gaviria (the President’s closest political advisor) as the masterminds behind the wiretaps. The apparent goal was to stay informed on actors that inhibited Uribe’s exercise of power. US sources confirmed the plot behind the campaign to intimidate and observe judges.68 After the revelations of the chuzadas, relations between President Uribe and the Supreme Court remained acrimoniously tense until the end of his term in office in 2010. This became readily evident in the nomination of the Attorney General, who in Colombia’s constitutional system is confirmed by the Supreme Court from lists put forward by the President. In his last year in office, Uribe repeatedly put forward names to replace the Attorney General only to see his candidates fall through in the vote in the Supreme Court. This was not unimportant, particularly

The Choque de Trenes

131

for the investigations into the wiretapping by the DAS that were ongoing at the time of the replacement of the Attorney General. As a State Department cable explains, Attorney General Mario Iguaran’s promise to get to the bottom of the revelations could have been fruitless, because “the GOC (Government of Colombia) may use the selection of his replacement—Iguaran departs in July—to limit the investigation’s scope”.69 The obstruction of Uribe’s nominations pre­ vented this and the DAS inquiry continued until after Uribe left office in 2010. In the end, from initially thirty Members of Congress investigated at the beginning of parapolítica, the number grew to 83 in June 2008, and 102 by 2010. In addition, the Justice and Peace Unit of the Attorney General’s office argued that one-third of all mayors, governors, and Members of Congress were promoted to office with the help of “narco-paramilitarism” (López Hernández 2010, 30–33). All of this essentially validated what Mancuso and Vicente Castaño had said at the beginning of the parapolítica saga: around 35% of Members of Congress had been friends of the paramilitaries. Revealingly, eight of ten parapoliticos investigated were part of Uribe’s coalition in Congress, providing evi­ dence that paramilitarism and uribismo were two closely related phenomena and movements (Gutiérrez Sanín 2010).70 To conclude this section on parapolítica and Uribe’s two terms in office—and to confirm my periodization of the scandal itself—the analysis showed that the shifts to phases three, four, and five of the scandal resulted from specific Supreme Court acts to investigate (or refrain from investigations). Combined with the drastic reactions by the executive to specific inquiries, this already suggests the importance of legal facts for institutional trajectories. Furthermore, we can say that the Supreme Court’s investigations did affect Uribe’s ability to govern, even though his own personal popularity was never properly in danger. First, the Supreme Court’s activism was instrumental in controlling the excesses of nepotism in the legislature. Its perseverance in the prosecution of corrupted politicians ser­ iously undermined the promises of success for “armed campaigns”. In addition to holding criminal legislators to account, the Supreme Court also undermined Uribe’s control of the law-making process during the empty seat episode. Sec­ ondly, parapolítica revelations created an incredibly negative public discourse, which discredited almost the entire political system. As shown, once parapolítica investigations acquired legal meaning—i.e. when the Supreme Court officially indicted Members of Congress in September 2006—the discourse turned notice­ ably negative and it never recovered until the end of his presidency (see Graph 3.2 above). It forced Uribe to spend a lot political capital to fight off accusations, which in turn fueled critical reports about his style of governance, accusing him and his government of abusing executive power. The fact that Uribe’s second term became identified with the excesses of presidentialism was also noted amongst judges in the Judicial Palace.71 In the final analysis of the constitutional decision regarding presidential re-election, we will see that the substitution doc­ trine, instrumental for denying a potential third term in office, built on the realworld conditions of excessive presidentialism. It is for this reason that the pro­ duction of legal facts not only affected the public debate about Uribe, but also

132 The Choque de Trenes influenced the constitutional meaning of specific clauses. In the end, the US embassy in Bogotá is correct to conclude that the ongoing persistence by the Supreme Court to investigate Members of Congress close to the government shows “the strength of Colombia’s democratic institutions” and posed serious problems “for a possible third Uribe term”.72 A key ingredient to this institutional strength was how the courts in general, and the Supreme and Constitutional Court in particular, coordinated during their divisions over the tutela contra sen­ tencias during the parapolítica affair. The final sections of this part will be devoted to analyzing the normative development. This will show that the Constitutional Court in its jurisprudential decision concerning criminal investigations did not insist on dogmatically applying nuevo derecho interpretations, but also opted for fairly formalistic decisions. Normative Contentions: Parapolítica and Legislators’ Constitutional Privileges in the Tutela Contra Sentencias A key observation of the analysis of the proceso 8000 was that the discontent between the Supreme and Constitutional Court over the inviolability of parlia­ mentary debate, and the concomitant tutela decision, undercut the production of legal facts. The most important developments and decisions in the parapolítica scandal occurred during Uribe’s second term in office. The first phase in my dif­ ferentiation of the parapolítica scandal spanned the entire first term in office until criminal investigations began in the second term. Uribe’s unprecedented sense of mission, his expansive constitutional reform agenda—normatively rooted in a vision of a communitarian state, and practically concerned with eliminating term limits—left marks on executive-court relations in that period (Valencia Villa 2012, 24–25). In the first two years of his presidency, Uribe explicitly marked the Con­ stitutional Court for reform, while in the second two years of his first term he somewhat retreated from these plans. Incidentally, in the last two years fall important reform projects needful of Constitutional Court approval: the Justice and Peace law and the re-election reform. During this time, relations between the Supreme and Constitutional Court remained hardened, and both courts insisted on their respective position regarding the tutela contra sentencias. Uribe’s re-election and beginning of the second term coincided with the com­ mencement of phase II of the parapolítica scandal. Shortly after his re-election, official investigations started to produce first legal facts (indictments of Members of Congress Muriel Benito Rebollo, Erick Morris, Alvaro Garcia). AUC com­ mander Mancuso’s evidence given in a Justice and Peace Court turned the Supreme Court’s attention to the President’s cousin, Mario Uribe. This resulted in the third phase of the scandal, in which the President openly attacked the Court, alleging the meddling in political affairs. The Court’s decision to not further pro­ secute legislators that had revoked their constitutional privileges led to phase IV of the scandal; the final phase began with the recommencement by the Court to take on such cases. During these events, relations between the Supreme and Constitu­ tional Court became very conciliatory. In the most important decisions, the

The Choque de Trenes

133

Constitutional Court backed the Supreme Court in its interpretation, and the Supreme Court, in turn, built on verdicts the Constitutional Court published— particularly its Justice and Peace Law jurisprudence. Thus, this section will show that a key condition for the continuous production of legal facts during the parapolítica scandal, which in turn were critical for the trajectory of that scandal, was a relatively cohesive relationship between the high courts within the judiciary. This relatively cohesive relationship consisted in a general agreement over norma­ tively appealing outcomes reflective in reasoned decisions backing up the Supreme Court’s actions vis-à-vis Members of Congress, while maintaining normative dis­ agreement over the principles of the tutela contra sentencias. It was here that the Constitutional Court also opted for more legalistic and formalistic interpretations to avoid the tutela contra sentencias becoming a wedge between the high courts. This differed not only from the proceso 8000, but also from the first Uribe term. The symbiotic relations between Members of Congress and paramilitaries exposed during the third to fifth phase of the scandal veiled the disagreement over the principles of the tutela behind the concern over the apparent parainstitutionalization of Congress. The acrimonious approach taken by Uribe and his administration ensured that the constitutional stakes were always part of the equation in the calculations of judges and courts. Relations between the courts were aggravated by the discontent over the Viviane Morales tutela upholding her special privileges as a legislator. Two important decisions marked the choque de trenes during Andres Pastrana’s pre­ sidency (1998–2002). Immediately after the notorious SU-047/99 decision acquitting Members of Congress, the Constitutional Court decided a tutela sub­ mitted by Jairo José Ruiz Medina, complaining that Supreme Court investigations for embezzlement of funds violated his parliamentary privileges. The Constitu­ tional Court rejected the claims, because criminal procedures against embezzle­ ment and corruption do not fall under the terms of the constitutional protection granted, since they are not functionally related to the task of voting and opining in either chamber of Congress (SU-786/99). In 2000, President Pastrana issued Decree 1382, specifying the institutions to review tutela contra sentencias: tutelas against any public corporation were to be remitted to a higher authority within that specific corporation, and tutelas submitted against a national institution were to be reviewed by the administrative tribunal of Cundinamarca.73 Thus, while there was some specification regarding the tutela contra sentencias, the core of the contention between the high courts, namely the applicability of the tutela regarding judicial decisions of the highest courts of the ordinary and administrative system, was not resolved at the beginning of the Uribe presidency. Con­ sequentially, this disagreement carried over into both terms. Uribe began his time in office with a large surplus of political capital: he had won a landslide victory and already had unprecedentedly high popular support. He intended to employ that political capital for a wide-ranging constitutional reform agenda. Amongst other items, the judiciary was very high on the list of that reform agenda—not least due to the continuously surfacing conflict between the high courts over the tutela contra sentencias (the next chapter analyzes Colombian

134 The Choque de Trenes politics of constitutional reform, including Uribe’s reform agenda, its normative grounding in communitarian notions of leadership, and resulting tensions with constitutionalism and the separation of powers). His first appointment to the Ministry of Justice, Fernando Londoño Hoyos, a lawyer steeped in the traditions of viejo derecho, decried legal uncertainty arising from tutela, and vowed to reform the Constitution to mitigate its effects, arguing the country had become ungo­ vernable (Londoño Hoyos 2001). His reform project entailed that the jurisdiction of the tutela would only extend to violations of fundamental rights, and explicitly excluded social, economic and collective rights as well as judicial decisions by other courts.74 Critically, the minister adamantly iterated that the Constitutional Court’s ability to review reform is limited to procedure in a very evident effort to push back the Court’s activism, as it was developing the substitution doctrine (see Chapter 4). The Supreme Court fervently supported the reform project, while the Constitutional Court opposed it vehemently. The President of the Constitutional Court, magistrate Eduardo Montealegre, discredited it as the most open attack against the integrity of the Constitution and the social state of law, akin to Fuji­ mori’s assault on democracy in Peru. He argued that Congress has limits in reforming the Constitution, implicit in the term itself: it can reform, but not replace the Constitution through an ordinary act of legislation.75 Colombians supported the Constitutional Court’s position and took to the streets in defense of the Court’s jurisdiction and the tutela in general. They gathered in demonstra­ tions before the Palacio de Justicía in Bogotá. Two interview subjects, including one Constitutional Court judge, noted that this support was rooted in the Con­ stitutional Court’s social rights jurisprudence: “they knew that the tutela had helped them to get medicine”.76 Eventually, the reform project collapsed, because Londoño Hoyos had to resign in November 2003 for corruption allegations in connection with a public company, Ecopetrol. This would later result in a verdict that barred him from public office for 15 years. The relations between Supreme and Constitutional Court became most toxic from 2004 until 2006, during the executive’s strategic withdrawal from the dis­ pute over the tutela contra sentencias. The new Minister of the Interior and Jus­ tice, Sabas Preteld de la Vega, took on a much more conciliatory position and wanted to bring all actors to the table before initiating judicial reform. At this point, however, judicial reform had diminished in importance, and the peace pro­ cess with the paramilitaries was much higher on the agenda. It entered its most important phase and would eventually require the approval of the Constitutional Court. In addition, a potential reform to extend the term limits to two consecutive terms also appeared on the agenda. Constitutional reforms, too, necessitate the approval from the Constitutional Court, making it a prudent act for the executive to defer to the Constitutional Court’s position. The Supreme Court, on the other hand, openly defied the Constitutional Court, and refused to apply 50 tutelas the Court had issued; neither did the Supreme Court remit them for review, provok­ ing a response from the Constitutional Court. In February 2004, the Constitu­ tional Court retorted by issuing Auto 004/04, which stipulated that tutelas could now be submitted to any judge of any institution, and anywhere within the legal

The Choque de Trenes

135

hierarchy, opening the path for lower courts to potentially review decisions of courts higher in the hierarchy (Oliver Ortiz 2012, 8). The second phase of the parapolítica scandal began when the Supreme Court reacted to citizens’ unanimous complaints about illicit relations of their repre­ sentatives after López Obregon had filed her official complaint. As one judge on the Supreme Court confided to me, “the reports kept coming in, so we continued our investigations until it became clear that they required a special unit for coor­ dinating the investigative efforts”.77 Additionally, the revelations from the “Jorge 40” computer forced the Court to centralize efforts, and submit all investigations against Members of Congress to the Auxiliary Office of magistrate Ivan Velaz­ quez’s office. Once investigations commenced, the Justice and Peace Law context is the other critical component. Not only did it require paramilitaries to tell the entire truth of their past crimes for submission under the transitional justice regime (C-370/06): the evidence from paramilitary testimonies resulted in the indictment of Uribe’s cousin, Senator Mario Uribe, and escalated the third phase. The first normative contention that occurred in this third phase of the parapolítica scandal evolved around the issue of sedition and conspiracy to commit a crime in the Justice and Peace framework. In July 2007, José Roberto Páez Varón had submitted a tutela against a lower court decision that sentenced him to 40 years (later reduced to 25) in prison for aggravated homicide. The tutela argued against either sentence, stating that it violated the stipulation of the Justice and Peace Law, which had imposed reduced sentences for crimes committed in association with the paramilitaries. In its decision, the Supreme Court argued, however, that the Constitu­ tional Court, when it decided the Justice and Peace Law, had excluded the subsection on sedition and conspiracy to commit a crime on procedural grounds. It could therefore not be subsumed under the Justice and Peace administration and serve as a justification for reduced criminal sentences. In addition, the Court held that Article 71 of Law 975 from 2005 (Justice and Peace Law) amounted to a material violation of the Constitution, because it unduly conflated ordinary crimes with political ones, resulting in a via de hecho (Ramírez Bastidas and Socha Salamanca 2007, 48–49).78 Uribe’s reaction to the Supreme Court’s verdict was to attack the Court for threa­ tening the peace process in Colombia and undermining the legal framework of the demobilization of paramilitaries. For denying the political nature of the crime, he accused the Court of political interference, because it applied a double standard to the crimes committed by guerrillas and paramilitaries. They were involved in identical activities and therefore ought to be treated equally in Colombia’s penal courts, he argued. Finally, the President accused the Supreme Court of enacting a via de hecho by diverging from the Constitutional Court’s interpretation.79 The Constitutional Court, though, sided with the Supreme Court’s interpretation on formalistic grounds. In 2008, the Court issued an Auto, which affirmed the Supreme Court’s interpretation that sedition does not fall under Justice and Peace jurisdiction and cannot receive transitional justice benefits, because it had decided on the issue in its decision on the Justice and Peace Law, and could not reopen the case without deciding a cosa juzgada (Auto 089/08). It therefore deferred to the Supreme Court’s position with the result that investigations continued.

136 The Choque de Trenes Aside from invoking the Justice and Peace jurisdiction, Members of Congress attempted to avoid or end prosecutions by copying the strategy employed in the proceso 8000: 1) assert immunity as parliamentarians in the exercise of their votes and opinions in Congress; and 2) claim due process rights violations resultant from the constitutional privilege to be criminally investigated by the highest court without an appeal instance. The Constitutional Court, as seen above, had already refined the clause “free exercise of votes and opinions” as not pertaining to investigations for the embezzlement of public funds, because such investigations do not impinge opinions or votes submitted in Congress (SU-786/99). In 2004, the Constitutional Court argued that relations with paramilitaries could be crim­ inally investigated because such conduct was not protected by the freedoms arising from the exercise of legislative functions; these relations were testament of poten­ tially illicit activities outside of Congress (see also C-1174/04). Then Members of Congress filed tutela complaints that took aim at the lack of an appeal instance as a violation of due process rights. The Supreme Court rejected these claims, and insisted that Members of Congress are investigated in the nation’s capital in Bogotá, either by the attorney general or the special division of the Supreme Court. The President of the Supreme Court, César Julio Valencia, argued that investigations in the home departments of the political representative of that department created opportunities to unduly affect the outcome because of the influence they exercise over public institutions in those regions.80 The Constitu­ tional Court agreed in C-545/08 and imposed the condition that the investigat­ ing judge could not be the same as the deciding judge. In 2009, a unifying tutela (SU-811/09) added that constituents in the assembly in 1991 specifically designed the special regime of criminal accountability for Congress to avoid its criminalization at the hands of the drug business. Judges on both courts valued the constituents’ intention to impose accountability measures on legislators as legitimate given the context of the drug economy that had befallen Colombia. The commencement of investigations of Senators Álvaro Araujo and Mario Uribe induced the shift from phase II to phase III. The change to phase IV involved the same Senators, because they were also the first to resign from their investiture to avoid prosecution from the Supreme Court and have their cases handled by the ordinary justice system. There were three strategic considerations for that plan: 1) the ordinary justice system offered them additional appeal instances; 2) in the ordinary justice system, they could delay the prosecution through tutelas and hope for estoppels to terminate prosecution altogether; and 3) it allowed them to plead guilty in exchange for reduced sentences. In September 2008, the Supreme Court refrained from investigating allegations against Mem­ bers of Congress who had resigned their position in Congress, delegating the investigations with the attorney’s office.81 As shown, this decision had an impact on the progress of parapolítica (see Graph 3.1 above), as the numbers of investi­ gations in the Supreme Court declined. This situation, however, was only temporary and lasted until Semana reported that DAS had infiltrated the judicial palace with wiretaps. This commenced the fifth phase, when the Supreme Court retook the cases of resigned Members of

The Choque de Trenes

137

Congress. In March 2009, the Court argued in its proceedings against Senator Edgar Eulises Torres that the constitutional privilege imposed on Members of Congress requires a functional relation between the criminal deed and the pur­ poses of Members of Congress (see Supreme Court of Colombia Process No. 31.653). The signing of pacts with paramilitaries with the evident intention to be elected to Congress fulfills this functionality. Therefore, parapoliticos, who bene­ fited from the armed campaigns paramilitaries orchestrated on their behalf, could be indicted under the constitutional provisions accorded to legislators. Further­ more, the Supreme Court differentiated between the office in Congress and the privileges associated with those offices. The former can be simply relinquished with resignation, but the latter are more difficult to discard, because they are connected to the constitutional system of the separation of powers. They have a certain stickiness that outlives the resignation from that post. Together, this justi­ fied the Court to follow through with investigations against Members of Congress who had resigned their seats. Again, one of the most prominent was Senator Mario Uribe, whose case was once again taken up, and eventually concluded with a prison sentence of seven years in February 2011. The Constitutional Court confirmed this interpretation (SU-198/13). The yidispolítica affair, while not directly associated with the parapolítica investi­ gation, still touched on jurisprudential issues pertaining to the investigation of Members of Congress. For once, it highlights the strategic importance of the Superior Council of the Judiciary for congresistas’ tactics to reduce exposure to Supreme Court investigations. As a reminder, and as Figure 3.1 above showed, representatives in the lower house of Congress elect magistrates to that chamber from the President’s list. It was an evidently politicized process, and several magis­ trates on that court in fact had “campaigned” in Congress prior to their election with the claim that they would defend the double instance for Members of Con­ gress. As a result, many “parapoliticos” submitted tutelas in the course of legal pro­ ceedings with the judges of this chamber, and the disciplinary chamber subsequently largely followed the argumentation put forward by Members of Congress (Revelo Rebolledo 2009, 270–275).82 It is noteworthy that the Constitutional Court had opened this avenue with its decision in 2004 (Auto 004/04) to enable citizens to submit tutelas with any judge in the country. The jurisprudence in this case is additionally important, because the eventual decision by the Constitutional Court affirmed the Supreme Court’s assertion that resigned Members of Congress still fall under the jurisdiction of the Supreme Court if the alleged criminal act is related to the exercise of that office. Moreover, it had a particular standing in Uribe’s second term, because it directly affected the legality of his second term in power. In the yidispolítica affair, Yidis Medina and Teodolindo Avendaño, as well as Senator Iván Díaz Mateus, whom Medina had replaced when the vote transpired in the committee stages, faced criminal proceedings for their respective roles. Díaz Mateus appealed against the investigations, submitting a tutela in August 2009 with the disciplinary chamber of the Superior Council of the Judiciary. That Council ordered his liberation on the grounds that the Court did not have jur­ isdiction to investigate resigned Members of Congress. The Supreme Court

138 The Choque de Trenes ordered Díaz’s recapture on the grounds that its decisions were not open to tutela revision, reiterating its contentious argumentation at the core of the choque de trenes. 83 The Constitutional Court reversed the disciplinary chamber’s decision, and sided with the Supreme Court even though it did not agree on the subject of the tutela contra sentencias. It stressed that the Supreme Court had competence in criminal investigations of Members of Congress who had resigned their seats, and reiterated that the Supreme Court was the highest court of the ordinary justice system (T-965/09). When it reviewed Congress’s constitutional privilege, it refused to intervene in the criminal court’s jurisprudence. The Constitutional Court denied that the Supreme Court had committed a via de hecho for three reasons: 1) Article 186 manifests that the Supreme Court has the duty to investi­ gate any Member of Congress; 2) the temporal separation from the functions of the Member of Congress by resigning does not undo Member of Congress’s constitutional investiture with the office; and (3) even if the objective accusations had nothing to do with the debate on re-election in Congress’s committee (that was the substantive content of yidispolítica), the accusations still aimed at abuse of the investiture. The case not only confirmed the central normative position that was critical during the fifth phase of the parapolítica scandal—that the Supreme Court could investigate Members of Congress who had resigned from their seats—it additionally resulted in an open attack by the Constitutional Court against the disciplinary chamber of the Superior Council of the Judiciary for identically copying Members of Congress’s legal positions. The President of the Constitutional Court, magistrate Pinilla Pinilla, accused it of mixing “political aspirations with judicial decisions” and as consequence having turned into a “decomposed organ of the judiciary”.84 In sum, there were four strategies that Members of Congress tried to utilize to evade prosecution: 1) invoke the transitional justice jurisdiction of the Justice and Peace legislation; 2) submit tutelas on the grounds of their special rights of par­ liamentarians to freely opine without the threat of coercion; 3) claim due process rights violations resultant from the constitutional “privilege” to be criminally investigated by the highest court without the chance of appeal; and 4) resign from their seat and thereby also renounce the congressional investiture that involves that investigations are directly passed to the Supreme Court (thereby hoping to not only avoid the reach of the Supreme Court, but also be submitted to the jur­ isdiction of regional courts where they could more easily exert influence). Even­ tually, and in contrast to the proceso 8000, the Constitutional Court confirmed the Supreme Court’s interpretation in tutela cases foiling these strategies to evade justice. One outcome of the DAS infiltration of the Judicial Palace in Bogotá and the illegal wiretapping of judges on the Supreme and Constitutional Court is indicative of the different nature of the relations between these two judicial bodies. While during the proceso 8000, the Presidents of each court attacked decisions and actions by the other body, they now joined together to condemn the actions by the President. In a joint comunicado, signed by all high courts and the Fiscalía, the judicial bodies of Colombia implored Uribe to terminate the DAS infiltration, called for respect of the ongoing investigations, most emphatically rejected the

The Choque de Trenes

139

invasion of privacy of members of the Supreme Court, and called on Colombia’s society as well as international human rights bodies to protect the Supreme Court’s integrity. This was a powerful, and unprecedented, act of public solidarity.85 The results were twofold. Firstly, it enabled the Supreme Court to continue the production of legal facts about relations Members of Congress had cultivated with illegal armed groups for electoral benefit. Secondly, as these congresistas were to a large extent supporters of Uribe in Congress or other political allies (and in one case, a member of his family), it forced Uribe to invest a lot of political capital in a campaign to delegitimize the Court and its investigations. As seen, this campaign eventually transgressed into illegality, when the President’s office orchestrated the DAS wiretapping of the Supreme and Constitutional Court. Together, this not only held Members of Congress accountable and therefore improved the rule of law; it also counteracted what scholars identified as a crisis of representation in Colombia (Mainwaring et al. 2006).

Part IV: Path Dependence of Legal Argumentation and the Production of Legal Facts during the Proceso 8000 and Parapolítica Scandals The question this chapter posed followed logically from the conclusion of the previous chapter: if the 1991 Constitution is a critical juncture, as shown in that chapter, does the post-genesis evolution follow a path-dependent logic as hypo­ thesized by historical institutionalist theory? Path dependence is defined as the idea that initial decisions in one direction induce further movements in that same direction later on. The reason for the concept’s predictive capacity lies in the dis­ tributive tendencies of political struggles over resources and jurisdictions that reproduce causal patterns in the direction of the most resourceful actor. In other words, institutional outcomes at a later stage reflect the interests of (powerful) actors that had taken critical decisions at earlier stages of the institutional trajec­ tory. The novelty of my approach is that this analysis examines if the same dis­ tributive dynamic that develops towards equilibrium over time in majoritarian institutions can be identified in the evolution of judicial institutions. There are two reasons to consider path dependence to be applicable to post-1991 development in Colombia’s judicial institutions. Firstly, the inquiry into the creation process of the 1991 Constitution showed that the new charter was the product of a moment of contingency, and most fundamentally redesigned the judicial branch of Colombia’s separation of powers system, introducing new institutions (Constitu­ tional Court, Superior Council of the Judiciary), new rights, and mechanisms to enforce rights (tutela). Secondly, since its creation, the Constitutional Court’s jurisprudence involves precedent setting and therefore implies a natural tendency to value earlier decisions over later ones. Thus, I posit that while it is well under­ stood how majoritarian institutions evolve along distributive dynamics of resource and jurisdiction struggles, the same cannot be said of the trajectory of judicial institutions that also have meanings rooted in rights claims of the Constitution.

140 The Choque de Trenes For such a context I hypothesize that discursive dynamics may also come into play. I further hypothesize that discursive institutionalism is not only involved in meaning making, which would be a preposterously general claim as meaning is constantly being produced, but rather that a key feature of discursive processes is that questions remain open to contestation. The institutional trajectory under discursive dynamics must therefore, in contrast to the predictions of conventional path-dependent developments, remain open. This study into judicial institutional trajectories essentially juxtaposes “closed” and “opened” path dependence as desiderata of the institutionalism at play. Colombia’s post-1991 developments have the benefit of sharing context condi­ tions that make the case particularly valuable. First, the judicial branch was the branch that exited the 1991 constituent process most profoundly refurbished, and, second, includes contradictory norms concerning the application of the tutela to legal decisions of high courts. Consequently, this neo-constitutional writ of pro­ tection, on the one hand unites the ideals and promises of the new Constitution that vests every citizen with human dignity, while on the other hand implies con­ flictual potential that incentivize distributive dynamics. The Constitution submits decisions by all public institutions to rights-based judicial review, but also states that the Supreme Court is the highest instance of the ordinary justice. The potential for distributive dynamics between the high courts becomes amplified through resource contentions in another branch of government, the legislature, by the absence of parliamentary immunity in the new political charter. The 1991 Constitution invested the Supreme Court with the task to investigate and judge criminal allegations against Members of Congress, who in turn exercise the same task vis-à-vis the President. Thus, the so-called choque de trenes between the high courts over the jurisdiction of the tutela also touched on the Supreme Court’s imperative to investigate Members of Congress, which in turn encroaches on the institutional interest of the legislature to deliberate freely. These preliminary considerations leave us with three elements that outline the value of the analysis in this chapter: 1) the proliferation of new courts and con­ tradictions between normative principles in the new Constitution result in a con­ test over jurisdiction between the Constitutional and Supreme Court; 2) the legal functions assigned to each branch of government in Colombia’s checks and bal­ ances system turns the fight over jurisdiction between the high courts into a poli­ tical question that involves majoritarian institutions; and 3) the affixture of rights to the standing of tutela complaints adds a layer to the analysis that implicates the meaning of foundational norms of the new Constitution and their effects on constitutional adjudication. These elements not only indicate the importance of this study, but also create the aforementioned juxtaposition between a path dependence of increasing returns and discursive contentions. While new actors, vying for authority and jurisdiction, implies distributive dynamics, the third aspect involves the significance actors assign to norms, and how they apply them in fac­ tual situations that go to the constitutional core of the separation of powers. This chapter employs a longitudinal research design that focused on two poli­ tical scandals evolving around the criminalization of Colombia’s Congress, and

The Choque de Trenes

141

how the courts interacted as political events played out. In the 1990s, the Cali Cartel funneled funds into the presidential campaign of Ernesto Samper, who won the electoral contest in 1994. The origin of those funds became public after his election, casting a shadow over the entirety of Samper’s term in office. He sur­ vived in office, because legislators cleared him from wrongdoing. Citizens sued legislators for breach of duty on the account of this vote of confidence, who in turn submitted tutela complaints arguing that criminal investigations infringed on their constitutional prerogative to opine and vote free from coercion. During the Uribe years (2002–2010), Members of Congress were accused of having culti­ vated close relations with paramilitaries associated with the drug trade for electoral purposes. In a process that essentially commenced with the peace negotiations between the AUC and the government soon after Uribe’s election, and dramati­ cally escalated in Uribe’s second term, the Supreme Court investigated and con­ cluded numerous leads against Members of Congress. They attempted to evade justice by again filing tutela complaints, but to no avail. Building on the conceptual discussion of path dependence, four questions guided the empirical analysis: � �

�

�

Did the jurisprudence regarding the tutela follow a mechanical and linear line from its inception in the 1991 Constitution? Did earlier decisions matter more than later decisions, and if so, did they create lock-in effects that forego the possibility to decide differently in later questions on the same merit, or did earlier decisions leave open and create new possibilities for contentions on the same merit in later decisions? If the Constitutional Court changed or modified its jurisprudence, did the modification follow mechanically from earlier decisions or did it base the change in principled explanation, engaging a legal argumentation evolving around legal values and shifts in legal context? If there was a clear break, did that break in turn result in a logically following pattern, indicating reactive path dependence?

Answers to these questions provide us with intriguing details on the institutional development in Colombia’s democracy after the new Constitution was imple­ mented. First, it is important to point out that the basic disagreement between the Supreme Court and the Constitutional Court over the application of the tutela to judicial decisions remained constant throughout the proceso 8000 and parapolítica scandal. As seen in the contentious Díaz Mateus decision, the Supreme Court continued to argue that its decisions were not open to rights review. Conversely, the Constitutional Court always reiterated its own jurisprudence vis-à-vis the review of judicial decisions, insisting that any factual decision by any public insti­ tution is subject to rights reviews. Furthermore, the institutional framework did not alter the functions of courts, nor was the tutela itself regulated by legislation. The only decree issued in 2000 minimally specified the order of review. The fact that the broad parameters of the institutional clash over the application of the tutela remained the same for the entire period suggests that we cannot identify

142 The Choque de Trenes fundamental shifts that decisively turned the path in one direction. Rather, the institutional development occurred more subtly and incrementally in this instance. The second major finding of this discussion is that, in both affairs, the production of legal facts was of utmost importance. Legal facts are those “facts” produced in a legal process; they have a deeper institutional impact than other socially produced “facts”. On the proceso 8000, Uprimny remarked that without legal investigations and independent prosecutions there would not have been “a proceso 8000, nor a trial against the president, nor a political crisis” (1996, 120). The increasingly hostile public debate was consistently pushed forward by new revelations that resulted from the legal processes against Samper’s campaign managers, and eventually the processes against Members of Congress. Similarly, the most potent revelations of the para­ política scandal were those revealed or affirmed by judicial investigations. While the initial revelations about potential paramilitary infiltrations of the political class—above all the declarations by the AUC commanders themselves—are significant, the scandal received a much more explosive dynamic once these facts were confirmed and further investigated by judicial bodies. We saw this correlation between the production of legal facts and public fallout in the news coverage of the second Uribe administration that turned negative on aggregate when transitional justice courts of the Justice and Peace Process revealed connections with criminal law implications for Members of Congress (especially of those with close connections to Uribe). Third, even though the macro institutional configurations about the application of the tutela remained the same, the outcome in each affair differed. This became evident in the number of investigations each affair produced, and the political fallout each scandal fabricated. In the proceso 8000 only a few Members of Con­ gress were eventually sentenced for their collusion with drug traffickers, while during the parapolítica 102 Members of Congress were investigated between 2006 and 2010. Consequently, the political fallout of the scandal was also much more significant and severe than in the proceso 8000 saga. The lengths to which the Uribe administration went to admonish the Supreme Court’s authority, in both its capacities to administer legally binding decisions and radiate legitimacy, is well documented. The President invested heavily in political and institutional capital to evade and veil the accusations against his coalition allies in Congress and family. Equally indicative of the (adverse) institutional significance was the collective reaction of Colombia’s high courts to the strategies the executive employed—both publicly (they published a joint communique condemning the attacks against the Supreme Court) and jurisprudentially (see below). It is my contention that there is a narrative involving Colombia’s judicial insti­ tutions, specifically the two high courts, Supreme and Constitutional Court, that explains this diverging outcome. My interviewees repeatedly argued that the 1999 tutela decision in favor of Viviane Morales fundamentally obstructed the Supreme Court’s work to further investigate criminal behavior. A judge from that court implored that “the tutela was a good tool for protecting rights, but through this tool the Constitutional Court intervened in the 1990s in the investigations against congresistas and stopped” them.86 Similarly, the same judge explained that during the parapolítica affair, the Constitutional Court did not interfere investigations

The Choque de Trenes

143

and often backed their interpretation of normative clauses: “The Council [of the Judiciary] has a political origin. They are selected by Congress, so they always go to them with their tutelas. But fortunately, the Constitutional Court threw many of them out. This was important for the investigation into parapolítica.”87 There is also evidence in the behavior of those indicted in the judicial proceed­ ings, namely parapoliticos in Congress, that affirms the outsized role of courts in the parapolítica investigation. I documented in this chapter how Members of Congress repeatedly attempted to avoid investigations and utilized four strategies: 1) invoke the transitional justice jurisdiction of the Justice and Peace legislation; 2) submit tutelas on the grounds of their special rights as parliamentarians to freely opine without the threat of coercion; 3) claim due process rights violations resultant from the constitutional “privilege” to be criminally investigated by the highest court without the chance of appeal; and 4) resign from their seat and thereby also renounce the congressional investiture that involves that investigations are directly passed to the Supreme Court (to evade the reach of the Supreme Court and increase their own influence on judicial proceedings in regional courts). Only the fourth tactic listed here was successful in slowing down the penal process against incriminated legislators, because the Supreme Court itself decided to refrain from investigations. Once it changed its jurisprudence, the Constitutional Court also supported that interpretation.88 A central investigator within the Supreme Court stated it concisely: the Constitutional Court was very protective during parapolítica of the Supreme Court. When many Members of Congress wanted to escape investigations, the Constitutional Court rejected their writ of protection claims. So there was no difference in the investigative process (to the processo 8000 investigations), but there was a difference in judicial backing between the Courts.89 This convergence between the Supreme Court and the Constitutional Court regarding the jurisprudence of parliamentary immunity in tutela decisions is the most important factor that explains the prolongation of the parapolítica affair. Firstly, it enabled the Supreme Court to continue the production of legal facts about relations Members of Congress had cultivated with illegal armed groups for electoral benefits. The deepening and broadening of investigations showed the big picture of the network of relations between Members of Congress and illegal ele­ ments associated with the internal conflict and drug trade. It made the governance issues that scholars have identified as the crisis of representation in Colombia readily apparent. Secondly, as these congresistas were, for a large part, supporters of Uribe in Congress or other political allies (in one case a member of his family), it forced Uribe to invest a lot of political capital in a campaign to delegitimize the Court and its investigations. The strategies employed went as far as to orchestrate illegal and intrusive smearing campaigns against individual magistrates from the Supreme Court, and thereby revealed the poisonous combination of a powerful executive with a co-opted legislature. As seen, this campaign eventually

144 The Choque de Trenes transgressed into illegality, when the President’s office orchestrated the DAS wir­ etapping of the Supreme and Constitutional Court. In the research design guiding this chapter’s analysis, the tutela contra sentencias serves as the institutional test case for path-dependent trajectories of judicial insti­ tutions. The question therefore becomes whether the normative jurisprudence of tutela decisions in each scandal corresponds to a linear trajectory that has its origin in the critical juncture. In the discussion of the most important tutelas during the proceso 8000 we saw that the substance of SU-047/99 can be connected to nor­ mative claims that value parliamentary debate as the essence of participatory and pluralistic democracy. As explained, the new constitutional regime of the 1991 Constitution does indeed embrace such values in the articulation of its core prin­ ciples. Viewed through this lens, criminal investigations against legislators then do appear as an infringement of legislators’ prerogatives to freely deliberate. We have seen, however, that constituents intentionally included a special liability regime for legislators to protect the legislative bodies from the criminal elements in the country. Moreover, we also saw that prior to the SU-047/99 decision, the Con­ stitutional Court confirmed these special liabilities as paramount elements of the institutionalization of the separation of powers. In C-222/96 and C-245/96, the Court argued that each branch of government fulfills specific functions that imply specific powers as well as obligations. According to those decisions, breaches in the fulfillment of duties and constitutional function carried the possibility of criminal liability. Together, these observations already indicated that post-genesis develop­ ment in judicial institutions remains open to change and did not lock in a specific configuration. In the parapolítica affair, there is no recognizable line of argumentation that connects the earlier decisions with the later ones, either. Rather, judges in both courts appeared to incorporate the context into their decision-making when they deliberated on individual cases. In the end, the decisions by either court moved much more in the opposite direction from the weighing in SU-047/99 that interpreted the legislative prerogative so expansively that it curtailed the Supreme Court’s abilities to investigate Members of Congress. In 1999, the Constitutional Court held that embezzlement does not fall under the legislative prerogative to freely opine and vote (SU-786/99). When Members of Congress wanted to be judged under the sedition clause of the Justice and Peace legislation, the Con­ stitutional Court excluded the conspiracy to commit a crime from the “political crime” clause of Justice and Peace. When congresistas argued that their constitu­ tional privilege to be investigated by the Supreme Court violated their due process rights as citizens, the Constitutional Court upheld the Supreme Court’s inter­ pretation that the office and the constitutional privileges associated with that office have varying half-lives. The office can be resigned, but the privileges are connected with the constitutional system of separations of powers and therefore remain intact. Finally, when the Constitutional Court specified the special privilege, as it did in C-545/08 and SU-811/09 (designating that the investigating and deciding judge cannot be the same individual in criminal cases of Members of Congress), it

The Choque de Trenes

145

explicitly argued with constituents’ intentions to hold legislators accountable to a higher standard of criminal liability. The interpretation of the constitutional imperatives that outlined implications for legislators’ criminal liability was much more textual and originalist than in the 1999 decision that expanded the legislative prerogative to freely deliberate. It is prudent to cite a Colombian legal scholar at length, who explained the different connotations of nuevo and viejo derecho explicit in the jurisprudence during either affair: The difference between the proceso 8000 and parapolítica is very institutional. In the proceso 8000, it was the Constitutional Court and the decision it took. It said that the congresistas have no responsibility in the exercise of their vote, when they excused Samper, and took away the right of the Supreme Court to judge them. … The Constitutional Court used a very much nuevo derecho interpretation that involved Viviane Morales’ fundamental rights and placed rights over forms. In contrast to that, five years later during parapolítica, they condemned the con­ gresistas only using the letters of law. Very formalist. It is absolutely legalistic.90 This serves to show that there is not a linear normative line from the constituent moment, which undoubtedly embraced neo-constitutionalist foundations such as the freedom of deliberation, to jurisprudential outcomes that tread on those nor­ mative foundations. Rather, there is an interpretative evolution of norms that countenances the legal context with the normative goals of the Constitution. A Constitutional Court judge, and towering scholar in Colombia’s legal culture, confided to me that the “second Constitutional Court with Sierra Porto, Cepeda, and Montealegre thought it was not wise to revise the tutelas in regard to para­ política, because the Supreme Court was pushing a progressive agenda”.91 In addition to normative considerations, the legal context also weighed in on inner branch relations. An auxiliary judge commented that: to annul a decision by the Supreme Court against a congresista (through a tutela) would have been judicial suicide by the judge. Congress is a fairly discredited institution. There are fairly clear relations of some congresistas with terrible forces. To annul one of these decisions was lethal to the judge’s prestige.92 It is noteworthy that the lethality to a judge’s reputation emanates from the legal context that is evidently enmeshed with illicit relations. What institutional narrative do these outcomes suggest? When we contrast these findings with the definition of path dependence along the lines of the historical institutionalist school that theorizes punctuated equilibria, increasing returns in distributive processes, and mechanisms of reproduction, there are some evident incongruities between the predictions of that model and the discussion of case decisions in this chapter. Rather, accounting for the trajectory of legal argu­ mentation, the first thesis must be:

146 The Choque de Trenes �

Once a decision is taken, the cost of reversal increases, but not exponentially so. Rather, reversals require good explanations rooted in precedent and change of context.

As seen, the Constitutional Court addressed the prior decisions (C-222/96 and C-245/96) concerning the legislator’s powers and duties in the infamous SU-047/99, and affirmed legislators’ rights to freely deliberate. Later on, the Constitutional Court responded in its unification decisions to the parameters set in SU-047/99, but curtailed, in an apparent U-turn, congresistas’ privilege, and confirmed the Supreme Court’s jurisdiction in criminal cases against Members of Congress (SU-811/09). The second thesis is: �

Earlier decisions do matter more than later decisions, since discourse, in par­ ticular legal discourse, builds on logical coherence. However, earlier decisions might also open new questions further down the road that can significantly alter the path.

The first decision on the rights and duties of parliamentary immunity interpreted legislators’ prerogatives very extensively, essentially shielding Members of Con­ gress from criminal scrutiny (SU-047/99). At the same time, it postulated new questions that had to be addressed in the future. Above all, the next tutelas and constitutionality decisions had to clarify what the parameters of the constitutional prerogative are, and what counts as the exercise of parliamentary duties (e.g. C­ 545/08; SU-811/09; T-965/09; SU-198/13). The reproductive mechanism is thus not one of increasing returns that narrows the scope of reasoning, but one that generates unforeseen, new, or novel questions that might arise from the conclusions of the previous questions. Finally, from the two preceding points follows that we cannot infer a point of no return. The deliberative nature of judicial decision-making retains the possibi­ lity that all decisions are open to contestation and reversal, if judicially relevant arguments can be put forward against the conventional interpretation. The final element of a path dependence of legal argumentation therefore must be: �

There cannot be a practical point of no return because legal decisions remain open to contestation.

In conclusion, the institutional argument in this chapter is that the openness of jurisprudential outcomes contradicts the predictions of classical path dependence for post-genesis trajectories, and better matches the conceptualization of more recent interpretations of incremental institutional change. Beginning with a cri­ tique of all neo-institutionalist schools, Mahoney and Thelen lament that the focus on inertia and stability has created a blind spot for the gradual shifts in institutional development that is not the result of exogenous, punctuating shocks. Above all, viewing institutions as stabilizing expectations about other actors’ behavior has

The Choque de Trenes

147

endogenized enforcement “in the sense that the expected costs and extent of noncompliance are factored into the strategic behavior of the actors” (Mahoney and Thelen 2009, 10). Mahoney and Thelen hold, however, that written norms can never be attuned to cover the full range of possibilities of real-life situations and always require interpretation. This interpretive process induces “its own dynamic of potential change” into the process of enforcing compliance (10). Institutional ambiguity and objective interest by actors to continue to benefit from specific institutional applications means that actors will continue to bring their resources to the table to resolve ambiguities in their own favor. The proceso 8000 and parapolítica are testament to the fact that the implementation of, and com­ pliance with, relatively stable written norms is indeed a sphere that involves poli­ tical action, as congresistas’ numerous strategies to evade the application of criminal law to punish their deeds makes readily apparent. Inserting compliance as a critical variable for institutional development leads Mahoney and Thelen, as well as Streeck and Thelen (2005), to outline four different paths of incremental, endogenous institutional change: layering, conversion, drift, and replacement. Layering involves the planting of new institutions on top of existing ones, conversion the changed enactment of existing rules due to their strategic deployment, drift the changed impact of institutions due to changes in the environ­ ment, and replacement the removal of existing institutions and reinstatement with new ones (Mahoney and Thelen 2009, 15; Streeck and Thelen 2005). This investigation of the application of the tutela in criminal investigations of Members of Congress suggests a more discursive process of institutional learning. First, there is the Constitutional Court’s jurisprudence regarding parliamentary inviolability: it moved from restrictive to expansive back to restrictive. We need to recall that the learning that induced judges to spell out the more restrictive inter­ pretation of parliamentary deliberation during parapolítica arose from the findings of transitional justice courts, the aforementioned legal facts. They showed how discredited Congress was and provided evidence for the relations numerous Members entertained with (illegal) armed non-state actors for their own benefits. On aggregate, such relations undermine a core promise of a constitutional democracy: namely, free, fair, and popular representation. It was the knowledge of these effects that meant that changing these decisions via the tutela would have meant “judicial suicide” for Constitutional Court judges, killing their prestige in the eyes of the other judges within the judiciary.93 Reputation seeking amongst peers and the indeterminacy of behavior in the choque de trenes and the tutela contra sentencias appeals to the notion that change evolves from argumentation itself. It was simply not argumentatively feasible to defend parliamentary immu­ nity as a bulwark for free deliberation, when Members of Congress utilized that norm for impunity. Given the argumentative procedures within a court and the continuous production of legal facts about nefarious relations congresistas enter­ tained, we can term this process institutional learning, in which norms are applied differently to result in a normatively more appealing outcome. In that context, what constitutes a more appealing outcome must result argumentatively from the decision-making process inside the court.

148 The Choque de Trenes In the context of the parapolítica scandal, Onur Bakiner made an interesting, and convincing, argument to explain the empowerment of the Supreme Court (2020). He writes that the “parapolitics process itself redefined the justices’ interests, selfperceptions, and, consequently, limits of jurisdiction” in what he terms the endo­ genous growth of judicial power (604). The 2009 decision to take on cases of Members of Congress who had resigned their posts was an evident divergence from the Court’s prior jurisprudence, and Bakiner is right to assert that the Court accu­ mulated authority by extending its jurisdiction to these cases. My analysis does not forego the validity of Bakiner’s argument; nevertheless, this chapter has attempted to do something else by focusing on path dependence in judicial institutions. It there­ fore also had different conclusions and contributions. First, the holistic approach to the study of path dependence in judicial institutions analyzed intra- as well as interbranch constellations, and how these affected the institutional outcome. Therefore, this internally disaggregated case study provides nuances to the institutional narrative that are outside the scope of Bakiner’s conclusion. I show that affidavits from paramilitaries were crucial for the evolution of the parapolítica scandal, as these produced “legal facts” that proved to be more difficult to politicize. These affidavits were not endogenous to the Supreme Court—Justice and Peace Court were separate enti­ ties—but not part of the exogenous shock, either: Bakiner defined the Constitutional Court’s decision to conditionally allow the Justice and Peace law as the original exo­ genous shock. Second, the inter-court relations also play a much more important role in my argumentation, because my longitudinal analysis of the choque de trenes shows more conciliatory relations between the high courts during parapolítica. After all, the Constitutional Court’s decision to back the Supreme Court’s jurisprudential shifts that increased power under its jurisdiction were crucial for the latter’s decisions to remain valid. It is for that reason that we can make some policy-relevant prescriptions, because it appears readily evident that several apex courts make it more difficult for executive encroachment of the judiciary (see Pou Giménez 2019). Finally, the methodological setup also highlights the relevance of the transitional justice process itself and how it affects subnational and national institutional relations. As explained, the fact-finding of Justice and Peace Courts was crucial for the production of legal facts and the political capital other branches of government could employ to further their legislative agenda. Moreover, the transitional justice process affected armed group behavior of the paramilitaries. Already the negotiations to the Justice and Peace Process created centrifugal forces within the paramilitary organization, pitting the various factions against each other. In sum, there is very evident value in analyzing a court’s behavior in full view of the institutional context. The next chapter will turn more explicitly to the internal, argumentative procedures within the high courts that are central for contending that institutional learning and argumentative procedures not only reinforce one another, but are mutually con­ tingent. To conclude this chapter of this book, we must first manifest that the con­ tinuous production of legal facts is what set the parapolítica apart from the proceso 8000, which showed, not least, in the extensive response that Uribe and his adminis­ tration orchestrated against the Supreme Court and its individual judges. Furthermore, the continued production of legal facts was to a decisive part the result of a normative

The Choque de Trenes

149

agreement between the Supreme and Constitutional Court in tutela decisions impli­ cating the role, privileges, and duties of legislators. Finally, the trajectory of jur­ isprudential development in the question of the tutela and the tutela contra sentencias does not become more intelligible with a classical understanding of path dependence. The shifts and differing outcomes were the result from endogenous changes that occurred within the institutions. The allusion made to the internal prestige suggests that the decision-making processes inside a court are critical for understanding out­ come of judicial behavior. The next chapter turns to the re-election reforms and the decisions by the Constitutional Court. It will also show the feedback circle effect of criminal investigations of Members of Congress into constitutional law, as judges on the Constitutional Court incorporated those investigations into their weighing of the effects of the legal context of (excessive) presidentialism on the separation of powers. That constitutional imperative, and the negative effects incurred through three consecutive terms in the presidential office on its validity, in turn, were the central contention in the litigation on the constitutionality of re-election reforms.

Notes 1 The official term for the subject of the choque de trenes between the high courts is tutela contra providencias judiciales. The slightly abbreviated form is tutela contra sentencias. I use them synonymously. 2 Interview with auxiliary judge of the Constitutional Court, 13 May 2013. 3 The Constitution also provided means to contextualize decisions. To that end, the Court can, prior to deliberation, call on experts as well as cabinet ministers to address concerns it might have. In addition, it can also summon public hearings—a practice that it terminated, however, because it undermined what one interviewee called the tran­ quility of process by increasing the public exposure. Interview with Constitutional Court secretary, 10 May 2013; see also Chapter 4. 4 Interview with professor of sociology, 13 May 2013. 5 While presidential re-election gave every president, if re-elected, the chance to nominate shortlists for the bench, it did not undermine the periodization of Constitutional Court judges and the President, because any given President could never influence the selec­ tion of more than three judges concurrently serving on the Court. 6 Semana, “Los parias de la justicia”, 2 February 2009. http://www.semana.com/na cion/articulo/los-parias-justicia/100371-3. 7 Interview with professor of philosophy, 7 May 2013. 8 Interview with Constitutional Court judge, 17 May 2013. 9 There is a wide-ranging debate on the exact definition of human dignity in legal philoso­ phy. Of course, Kant’s formula that humans should be treated as ends rather than means is probably the most prominent rendition of definitional traits, but not the sole contribution. This is hardly the space to survey the entire literature on human dignity, let alone further develop it. It is worthwhile noting, however, that Jeremy Waldron situates human dignity in the exercise of democratic politics. He showed that the term dignitas is associated with the social phenomenon of honor. Dignitaries used to be persons of rank, whose special treatment was grounded in a birthright. As a consequence of democratization this birth­ right was extended to everyone. As Waldron writes, “once associated with hierarchical differentiations of rank and status, ‘dignity’ now conveys the idea that all human persons belong to the same [high] rank” (2007, 201). This situatedness in democratic politics is interesting for the purposes of this investigation, because the tutela does not hold the

150 The Choque de Trenes

10

11

12 13 14 15

16 17

18 19 20

21

22 23 24

promise to automatically solve all human rights problems, but to gradually improve the situation so that undignified treatment of humans disappears. It is noteworthy that voluntarist views of human dignity and human rights are most often developed for well-ordered societies. However, even in a well-ordered society, the attribute “well-ordered” implies the inclusion of some form of guarantee of basic social rights, as Rawls (1993) and Habermas (1996) argued about. Jürgen Habermas (1974) and Robert Alexy (1989) both have shown that the ration­ ality rules of discourse and the pragmatics of legal reasoning exclude certain outcomes as impossibilities. For example, it is impossible to sell oneself, even consensually, into slavery. Alexy has shown that legal discourse never fully sheds its dependencies on general practical reasoning (and thereby incorporates its weaknesses), yet the struc­ turation of legal reasoning mitigates some of those weakness (292). Interview with auxiliary judge of the Constitutional Court, 12 April 2013. El Tiempo, “Quieren acabar con la acción de tutela”, 1 October 1992. http://www. eltiempo.com/archivo/documento/MAM-214170. Interview with Constitutional Court judge, 8 April 2013. Semana, “Este es el hombre que entregó los narcocasetes a Pastrana”, 29 November 2013. http://www.semana.com/nacion/articulo/andres-pastrana-recibio-los-narcoca setes-de-manos-de-carlos-barragan/366430-3; see also Hinojosa and Pérez-Liñán (2003, 66); Pérez-Liñán (2007, 22). Semana, “El 8000. Dia a Dia”, 16 January 1996. http://www.semana.com/nacion/a rticulo/el-8000-dia-dia/27509-3. There is later testimony by the son of a Cali Cartel capo, William Rodríguez Abadía, that the cartel bought Samper absolution in the Commission. Mogollón denied these. El Espectador, “Heyne Mogollón dijo que no recibió dinero por absolver a Samper”, 15 August 2013. http://www.elespectador.com/noticias/politica/heyne-mogollon-dijo­ no-recibio-dinero-absolver-samper-articulo-440268. El Tiempo, “Comisión pide no acusar a Samper”, 24 May 1996. http://www.eltiempo. com/archivo/documento/MAM-302134. Uprimny stressed the importance of legal investigations and observed that without an independent prosecution, there would not have been a “proceso 8000, nor a trial against the president, nor a political crisis” (1996, 120). Semana, “El relato de una vida”, 25 September 2005. http://www.semana.com/na cion/articulo/el-relato-vida/75036-3; El Tiempo, “El diario oculto de Alberto Giraldo”, 4 June 1995. http://www.eltiempo.com/archivo/documento/MAM-339247. His testi­ mony actually provides credence to the suspicion that the net between illegal money from the drug trade and Colombia’s political society was much more expansive than reckoned. Giraldo López was a longtime friend of the Rodríguez Orejuela brothers from their time in the early 1980s as militants in the Liberal party and the Nuevo Liberalismo movement. He was also close to various politicians and former presidents from Colombia’s Liberal Party and stressed that Samper was not the outlier case, but rather the norm. For instance, he indicated that the Barco government also received help from the Orejuelas. El Tiempo, “Colombia esta semana”, 2 June 1996. http://www.eltiempo.com/a rchivo/documento/MAM-307493; see also El Tiempo, “Niegan libertad a Rodrigo Garavito”, 11 December 1998. http://www.eltiempo.com/archivo/documento/ MAM-834911; El Tiempo, “A Juicio Rodrigo Garavito”, 13 June 1996. http://www. eltiempo.com/archivo/documento/MAM-314385. Garavito later confessed to having received illicit funds. He was sentenced to 94 months in prison. Congress of Colombia, Diario Oficial, 1995, Ley 190 de 1995, 6 June 1995. https:// www.funcionpublica.gov.co/eva/gestornormativo/norma.php?i=321. El Tiempo, “Cámara enterró el Narcomico”, 15 December 1995. http://www.eltiemp o.com/archivo/documento/MAM-483894. In another contentious issue, the Constitutional Court had to decide on a statutory law that would have terminated Attorney General Valdivieso’s term in April 1996.

The Choque de Trenes

25

26 27 28 29 30 31 32 33 34 35 36 37 38 39

40 41

151

Valdivieso was selected as a replacement for Gustavo de Greiff. De Greiff had alienated the United States and criticized the one-sided approach in the war on drugs that solely consisted in militarily constraining the supply side and neglected the demand structure in the overseas market. The legal question was whether Valdivieso would only termi­ nate de Greiff’s term or serve an entire term of four more years. Some of the con­ gresistas, who voted on the legislation evidently tried to avoid prosecution. The Constitutional Court, however, sided with the Consejo de Estado and annulled the statutory provision limiting the term of the Attorney General. See Uprimny (1996, 108); El Tiempo, “Período del Fiscal es cosa juzgada: Corte”, November 1995. http:// www.eltiempo.com/archivo/documento/MAM-469452. El Tiempo, “Corte Suprema apoya sentencia del Consejo de Estado”, November 1995. http://www.eltiempo. com/archivo/documento/MAM-467606. Under the old Constitution, legislators were inviolable and could not be prosecuted without consent of their chamber. Given the collegiality amongst congresistas in the 1980s and the resulting unwillingness to prosecute, citizens considered the framework a passageway to impunity and requested it to be studied in the constituent assembly in 1991. Constituents followed delegate María Teresa Garcés’ suggestions and replaced congressional immunity from criminal prosecution with the special privilege, which implied that the highest court of the ordinary justice system investigated and judged Members of Congress (Article 234; see also Pinilla Pinilla 2012, 818). The Court unified the jurisprudence regarding the inviolability and criminal liability of Members of Congress in a sentencia de unificación. For a chronology of these events, see SU-47/99. Rights and duties are formulated in Articles 174, 175, 178–3, 178–4, and 199 of the Constitution. For the Constitutional Court’s early interpretation, see C-245/96 and C­ 222/96. El Tiempo, “Arango hace mea culpe ante la Corte”, 21 October 1998. http://www. eltiempo.com/archivo/documento/MAM-830358. Gomez, Luis Carlos, “Se afianza division en Corte Constitucional”, El Tiempo, 16 November 1998. http://www.eltiempo.com/archivo/documento/MAM-821247. Cifuentes Muños, Eduardo, Salvamiento de voto, SU-047/99. El Tiempo, “Es un golpe para la justicia: Corte”, 2 February 1999. http://www.eltiemp o.com/archivo/documento/MAM-876667. Interview with Constitutional Court judge, 8 April 2013. Interview with Supreme Court judge, 18 April 2013. Interview with Supreme Court judge, 18 April 2013; interview with Supreme Court judge, 22 April 2013. Interview with Member of Congress of the Republic of Colombia, 12 March 2013; interview with Member of Congress of the Republic of Colombia, 16 April 2013; interview with Member of Congress of the Republic of Colombia, 13 March 2013. Before becoming law, the Constitutional Court reviewed the legislation and imposed critically important conditions, above all victims’ rights to truth; see below. See López Hernández and Sevillano (2008). Semana, “Votas a punta de fusil”, 19 Feb­ ruary 2002. http://www.semana.com/nacion/articulo/votos-punta-fusil/49357-3. Verdad Abierta, “Parapolítica: Rocío Arias Hoyos”, 20 August 2013. http://www. verdadabierta.com/politica-ilegal/parapoliticos/4771-parapolitica-rocio-arias-hoyos. Semana, “Hablar Vicente Castaño”, 5 June 2005. http://www.semana.com/porta da/articulo/habla-vicente-castano/72964-3; López, Claudia, “Votaciones atípicas en las eleciones de congreso de 2002”, Semana, 11 September 2005. http://www.sema na.com/on-line/articulo/votaciones-atipicas-elecciones-congreso-del-2002/74746-3. Semana, “La cacería de ‘Doblecero’”, 29 September 2003. https://verdadabierta. com/la-caceria/. Already in September 2004, Semana publicized reports that a number of narco-traf­ fickers were at the negotiating table and questioned whether they should receive

152 The Choque de Trenes

42 43 44 45 46

47 48 49

50 51

52 53 54

55 56 57 58 59

clemency for political crimes, when their endeavor was clearly motivated by plain crim­ inal aspirations. Semana, “Revelaciones explosivas”, 27 September 2004. https://www. semana.com/nacion/articulo/revelaciones-explosivas/68391-3/. Semana, “Cómo se hizo fraude”, 6 April 2006. http://www.semana.com/nacion/arti culo/como-hizo-fraude/78258-3. See also C-370/06 and C-454/06. Interview with Supreme Court judge, 22 April 2013; interview with Supreme Court judge, 18 April 2013; interview with Supreme Court judge, 17 April 2013. Interview with Supreme Court judge, 22 April 2013. Semana, “El Computador de ‘Jorge 40’”, 2 September 2006. http://www.semana. com/nacion/articulo/el-computador-de-jorge-40/80765-3. Semana, “El imperio de ‘Jorge 40’”, 30 September 2014. http://www.semana.com/nacion/articulo/el-imp erio-jorge-40/81239-3. Semana, “El computador de ‘Jorge 40’ puede ser el inicio de un nuevo proceso 8.000”, 8 October 2006. http://www.semana.com/on-line/arti culo/el-computador-jorge-40-puede-inicio-nuevo-proceso-8000/81379-3. Interview with Supreme Court judge, 22 April 2013. Semana, “La exitosa estrategia política de Ralito”, 28 November 2006. http://www. semana.com/on-line/articulo/la-exitosa-estrategia-politica-ralito/82332-3. Verdad Abierta, “De la Espriella: ‘Fui puente entre Uribe y Auc para negociación de Ralito’”, 25 September 2012. http://www.verdadabierta.com/component/content/a rticle/-/4232-de-la-espriella-y-pineda-eran-el-enlace-de-los-paras-con-uribe-mancuso. Semana, “Pacto con el diablo”, 20 January 2007. http://www.semana.com/nacion/a rticulo/pacto-diablo/83048-3. Interview with Supreme Court judge, 22 April 2013. Human Rights Watch, “Human Rights Watch Comments to the Office of the US Trade Representative Concerning the US-Colombia Free Trade Agreement”, 15 Sep­ tember 2009. https://www.hrw.org/news/2009/09/15/human-rights-watch-comm ents-office-us-trade-representative-concerning-us-colombia. Semana, “Mancuso vincula mas politicos”, 16 May 2007. http://www.semana.com/ on-line/articulo/salvatore-mancuso-vincula-mas-politicos-autodefensas/85941-3; interview with Supreme Court judge, 22 April 2013. Interview with Supreme Court judge, 22 April 2013. Semana, “Uribe contra el mundo”, 12 October 2007. https://www.semana.com/uri be-contra-mundo/88767-3/; Semana, “Asedio a la Corte Suprema”, 28 February 2009. http://www.semana.com/nacion/articulo/asedio-corte-suprema/100578-3; interview with Supreme Court judges, 22 April 2013 and 17 April 2013; Human Rights Watch 2008, 15–16; See Coronell, Daniell, “Los secretos de Tasmania”, Semana, 15 January 2011. http://www.semana.com/opinion/articulo/los-secretos-ta smania/234098-3: “Tasmania” himself revealed that he received money from the Pre­ sident’s advisors to plant these accusations. Embassy Bogotá, “No end in sight to Uribe-Supreme Court power struggle”, Wiki­ leakscable: 08BOGOTA3007_a, dated 15 August 2008. https://www.wikileaks.org/p lusd/cables/08BOGOTA3007_a.html. Embassy Bogotá, “High courts oppose GOC judicial reform proposal affecting parapolitician investigations”, Wikileakscable: 08BOGOTA2891_a., dated 5 August 2008. http://www.wikileaks.org/plusd/cables/08BOGOTA2891_a.html. Semana, “Para-política y reelección”, 19 April 2008. http://www.semana.com/na cion/articulo/para-politica-reeleccion/92233-3. Semana, “Se murió la reforma política”, 10 June 2008. http://www.semana.com/ on-line/articulo/se-murio-reforma-politica/93233-3. Radio Caracol, “Aprobada la Reforma Política con la ‘silla vacía’”, 17 June 2009. http://www.caracol.com.co/noticias/actualidad/aprobada-la-reforma-politica-con-la -silla-vacia/20090617/nota/830745.aspx.

The Choque de Trenes

153

60 Semana, “La Corte no juzgará los casos de congresistas que renunciaron al fuero”, 17 September 2008. http://www.semana.com/on-line/articulo/la-corte-no-juzgara-ca sos-congresistas-renunciaron-fuero/95316-3. 61 Importantly, once the Supreme Court changed its jurisprudence and took on these cases, the tutela complaints against these, too, were unsuccessful as the Constitutional Court turned down arguments by the disciplinary chamber of the Superior Council of the Judiciary and reinstated Supreme Court sentences (see below). 62 Interview with Constitutional Court judge, 3 May 2014. 63 Coronell, Daniel, “La historia no contada”, Semana, 5 April 2008. http://www.sema na.com/opinion/articulo/la-historia-no-contada/91968-3. 64 Interview with Constitutional Court judge, 8 March 2013; interview with auxiliary judge of the Constitutional Court, 15 March 2014. Nevertheless, this affair made the importance of legal facts not only for public policy readily apparent but also for insti­ tutional developments, since Uribe had to spend a lot political capital against the deci­ sion by the Supreme Court. My interviewees confirmed that had the Supreme Court produced this decision prior to the constitutionality decision that enabled the first re­ election, the Court would have decided differently. As an interviewee explained, had Medina’s misgivings been legally sanctioned prior to the Constitutional Court’s deci­ sion in 2005, it would have merited a procedural violation of the reform process. The same judge said, “what mattered is if they were sanctioned or not”; interview with Constitutional Court judge, 8 March 2013. 65 Interview with Supreme Court judge, 22 April 2013. The evidence comes from Alba Luz Flórez, who was the waitress in the Court passing on information to DAS and installing the bugs to listen into conversations. See El Espectador, “‘Mata Hari’ pide a María del Pilar Hurtado contar la verdad de las chuzadas”, 7 July 2014. http://www.elespectador.com/ noticias/judicial/mata-hari-pide-maria-del-pilar-hurtado-contar-verdad-de-articulo-502799. 66 Semana, “Asedio a la Corte”, 2 March 2009. http://www.semana.com/nacion/arti culo/asedio-corte-suprema/100578-3. 67 The DAS installed wiretaps on the political opposition (Cesar Gaviria, Gustavo Petro), members of the government coalition itself (Francisco Santos), several Supreme Court judges (Alfredo Gómez Quitnero, Sigrifredo de Jesús Espinosa, César Julio Valencia, Mauro Solarte, Camilo Humberto Tarquino), as well as Con­ stitutional Court judges (Clara Inés Vargas, Rodrigo Escobar). See Semana, “El espionaje era peor”, 25 April 2009. http://www.semana.com/nacion/articulo/ el-espionaje-peor/102429-3. 68 Colombiareports, “Reports on illegal wiretaps were meant to inform Uribe: Witness”, 11 June 2013. http://colombiareports.co/reports-on-illegal-wiretaps-were-meant-to-in form-uribe-witness/. Colombiareports, “Former intel official insists Uribe’s former chief of staff ordered wiretaps”, 1 March 2012. http://colombiareports.co/colombias-form er-chief-of-staff-accused-of-spying. 69 Embassy Bogotá, “DAS domestic spying scandal deepens”, Wikileaks Cable: 09BOGO­ TA1412_a. dated 4 September 2009. http://www.wikileaks.org/plusd/cables/ 09BOGOTA1412_a.html. 70 Gutiérrez Sanín explains that there was a clear relationship between Uribismo and paramilitarism. He argues that of the entire cohort of Senators (100) elected to the upper chamber in 2006, 42 ended up being investigated for paramilitary, FARC, or ELN relations. Of those 42, 35 were part of the Uribista coalition in Congress (all investigated for relations with paramilitaries). The entire cohort belonging to Uribe’s coalition consisted of 70 Senators. Thus, the likelihood of an Uribista Senator to have relations with paramilitaries was at 50% percent, whereas amongst non-Uribista Senators that correlation was only ~23% (12; see also López Hernán­ dez 2010). 71 Interview with Constitutional Court judge, 26 November 2012.

154 The Choque de Trenes 72 Embassy Bogotá, “Arrest of U Party leader president for paramilitary ties shows con­ tinued Supreme Court commitment to parapolitical investigations”, Wikileaks Cable; 08BOGOTA2702_a. dated 28 July 2008. http://www.wikileaks.org/plusd/cables/ 08BOGOTA2702_a.html. 73 El Tiempo, “Reglamentan el reparto de las tutelas”, 14 July 2000. http://www.eltiemp o.com/archivo/documento/MAM-1304429. 74 El Tiempo, “No a tutelas contra sentencias”, 25 September 2002. http://www.eltiemp o.com/archivo/documento/MAM-1331741. 75 El Tiempo, “Londoño quiere un estado monárquico y sin controles”, 3 August 2003. http://www.eltiempo.com/archivo/documento/MAM-1038668. See also El Tiempo, “Reform agita congreso”, 1 August 2003. http://www.eltiempo.com/archivo/docum ento/MAM-1034549. 76 Interview with auxiliary judge of the Constitutional Court, 24 April 2013; interview with Constitutional Court judge, 8 April 2013. See also Rodríguez Garavito and Rodríguez Franco (2010). 77 Interview with Supreme Court judge, 18 April 2013; interview with Supreme Court judge, 22 April 2013. 78 See Process No. 26945 from 11 July 2007 (against Orlando César Caballero Mon­ talvo), and also the decision by the tutela chamber of the Sala Casación Penal of the Supreme Court of Colombia: “Decisión: Resolver la acción de tutela promovida por José Roberto Páez Varón”, 26 July 2007. 79 El Tiempo, “Se calienta pulso por sedición”, 28 July 2007. http://www.eltiempo. com/archivo/documento/MAM-2591024. 80 El Espectador, “Sería un pésimo ejemplo que primara la impunidad”, 28 July 2007. 81 Semana, “La Corte no juzgará los casos de congresistas que renunciaron al fuero”, 17 September 2008. http://www.semana.com/on-line/articulo/la-corte-no-juzgara-ca sos-congresistas-renunciaron-fuero/95316-3. 82 Other cases, which the disciplinary chamber decided on evidently political grounds, were those of Senator Miguel de la Espriella and Minister Diego Palacio—the former in connection to parapolítica and the latter in relation to yidispolítica. While campaigning for their seats in Congress, the chamber’s judges supported creating a double instance for indicted Members of Congress (Revelo Rebolledo 2009, 264). 83 El Espectador, “Corte Suprema ratifica orden de captura a Iván Díaz Mateus”, 1 September 2009. http://www.elespectador.com/noticias/judicial/articulo159092-corte-suprema-ra tifica-orden-de-captura-ivan-diaz-mateus. 84 El Espectador, “Sala Disciplinaria de la Judicatura es un órgano descompuesto: Nilson Pinilla”, 12 January 2010. http://www.elespectador.com/noticias/judicial/articu lo181553-sala-disciplinaria-de-judicatura-un-organo-descompuesto-nilson-pini. 85 El Tiempo, “Un pronunciamiento directo al presidente Uribe por el caso de las chuza­ das, pidieron las Cortes”, 5 May 2009. http://www.eltiempo.com/archivo/docum ento/CMS-5141428. 86 Interview with Supreme Court judge, 18 April 2013. 87 Interview with Supreme Court judge, 18 April 2013. 88 Interview with Supreme Court judge, 18 April 2013; interview with Supreme Court judge, 22 April 2013; interview Constitutional Court judge, 8 April 2013. 89 Interview with auxiliary judge of the Supreme Court, 22 April 2013. 90 Interview with legal scholar, 6 May 2013. 91 Interview with Constitutional Court judge, 17 May 2013. 92 Interview with auxiliary judge of the Constitutional Court, 15 April 2013. 93 Interview with auxiliary judge of the Constitutional Court, 15 April 2013.

The Choque de Trenes

155

References Acemoglu, Daron, James A. Robinson, and Rafael J. Santos. 2013. “The Monopoly of Violence: Evidence from Colombia”. Journal of the European Economic Association 11 (1): 5–44. https://doi.org/10.1111/j.1542-4774.2012.01099.x. Alexy, Robert. 1989. A Theory of Legal Argumentation. The Theory of Rational Discourse as Theory of Legal Justification. Oxford: Oxford University Press. Aponte, Alejandro. 1994. “La Tutela Contra Sentencias: Del Juez Como Garante de Los Derechos Fundamentales”. Revista de Derecho Público de La Universidad de Los Andes 3: 97–112. Arango, Rodolfo. 2003. “Basic Social Rights, Constitutional Justice, and Democracy”. Ratio Juris 16 (2): 141–154. https://doi.org/10.1111/1467-9337.00229. Ávila Martínez, Ariel Fernando. 2010. “Injerencia Política de Los Grupos Armados Ile­ gales”. In Y Refundaron La Patria. De Cómo Mafiosos y Politicos Reconfiguraron El Estado Colombiano, edited by Claudia López Hernández, 79–215. Bogotá: Corporación Nuevo Arco Iris. Bakiner, Onur. 2020. “Endogenous Sources of Judicial Power: Parapolitics and the Supreme Court of Colombia”. Comparative Politics 52 (4): 603–624. https://doi.org/ 10.5129/001041520X15743890683814. Bejarano, Ana María, and Eduardo Pizarro. 2005. “From ‘Restricted’ to ‘Besieged’: The Changing Nature of the Limits to Democracy in Colombia”. In The Third Wave of Democratization in Latin America, edited by Frances Hagopian and Scott P. Mainwar­ ing, 235–260. Cambridge: Cambridge University Press. https://doi.org/10.1017/ CBO9780511791116.009. Botero Campuzano, Libardo. 2008. Los Potros de Bàrbaros Atilas. La Razón de Los Ina­ movibles Del Presidente Uribe En El Debate Sobre El Llamado “Acuerdo Humanitario”. Bogotá: Fundación Centro de Pensamiento Primero Colombia. Capoccia, Giovanni, and R. Daniel Kelemen. 2007. “The Study of Critical Junctures: Theory, Narrative, and Counterfactuals in Historical Institutionalism”. World Politics 59 (3): 341–369. https://doi.org/10.1017/S0043887100020852. Cepeda Espinosa, Manuel José. 1993. “Participatory Democracy and Respect for Human Rights. Colombia’s New Constitution”. Bogotá: Presidency of the Republic of Colombia: Office of the Advisor for the Development of the Constitution. Cepeda Espinosa, Manuel José. 2007. Polémicas Constitucionales. Bogotá: Legis. Cifuentes, Eduardo. 1995. “El Constitucionalism de La Pobreza”. Dereito 4 (2): 53–77. Cifuentes, Eduardo. 1998. “Tutela Contra Sentencias (El Caso Colombiano)”. Ius et Praxis 4 (1): 147–187. Collier, David, and Ruth Collier. 2002. Shaping the Political Arena. Critical Junctures, the Labor Movement, and Regime Dynamics in Latin America. Notre Dame: University of Notre Dame Press. David, Paul A. 1985. “Clio and the Economics of QWERTY”. American Economic Review 75 (2): 332–337. https://doi.org/Article. Dávila, Andres, and Natalia Delgado. 2001. “La Metarmorfosis Del Sistema Politico Colombiano: ¿clientelismo de Mercado o Nuevo Forma de Intermediación?” In Degra­ dación o Cambio? Evolución Del Sistema Politico Colombiano, edited by Francisco Gutiérrez Sanín. Bogotá: Norma – IEPRI. Duncan, Gustavo. 2006. Los Señores de La Guerra. De Paramilitares, Mafiosos y Auto­ defensas En Colombia. Bogotá: Editorial Planeta Colombiana.

156 The Choque de Trenes Eaton, Kent. 2006. “The Downside of Decentralization: Armed Clientelism in Colombia”. Security Studies 15 (4): 533–562. https://doi.org/10.1080/09636410601188463. Garay Salamanca, Luis Jorge, Eduardo Salcedo Albarán, and Isaac de León Beltrán. 2010. “Garay Salamanca, Luis Jorge. Eduardo Salcedo Albarán. Isaac de León Beltrán”. In Y Refundaron La Patria. De Cómo Mafiosos y Politicos Reconfiguraron El Estado Colombiano, edited by Claudia López Hernández, 215–267. Bogotá: Corporación Nuevo Arco Iris. García Villegas, Mauricio, and Javier Eduardo Revelo Rebolledo. 2010. Estado Alterado, Cli­ entelismo, Mafias y Debilidad Institucional En Colombia. Dejusticia. Bogotá: Dejusticia. García Villegas, Mauricio, Javier Eduardo Revelo Rebolledo, and Rodrigo Uprimny Yepes. 2010. “Impacto Constitucional e Institucional de La Ley de Justicia y Paz”. In Y Refun­ daron La Patria. De Cómo Mafiosos y Politicos Reconfiguraron El Estado Colombiano, edited by Claudia López Hernández, 317–333. Bogotá: Corporación Nuevo Arco Iris. Gaviria Vélez, José Obdulio. 2004. Reelección. Que El Pueblo Decida. Bogotá: Editorial Planeta Colombiana. Gerring, John. 2008. “Case Selection for Case-Study Analysis: Qualitative and Quantitative Techniques”. In The Oxford Handbook of Political Methodology, edited by Janet M. BoxSteffensmeier, Henry E.Brady, and David Collier, 645–684. Oxford: Oxford University Press. https://doi.org/10.1093/oxfordhb/9780199286546.003.0028. Gómez Quintero, Alfredo. 2007. “La Justicia Penal y La Parapolítica”. Corte Suprema de Justicia. Revista, 23: 38–43. Gutiérrez Sanín, Francisco. 2007. Lo Que El Viento Se Llevo: Democracia y Partidos Politicos En Colombia 1958–2006. Bogotá: Grupo Editorial Norma. Gutiérrez Sanín, Francisco. 2010. “Reflexiones a Propósito Del Libro ‘Y Refundaron La Patria’”. In Y Refundaron La Patria. De Cómo Mafiosos y Politicos Reconfiguraron El Estado Colombiano., edited by Claudia López Hernández, 9–28. Bogotá: Corporación Nuevo Arco Iris. Gutiérrez Sanín, Francisco. 2019. Clientelistic Warfare. Paramilitaries and the State in Colombia (1982–2007). London: Peter Lang UK. Habermas, Jürgen. 1974. Theory and Practice. Boston: Beacon Press. Habermas, Jürgen. 1996. Between Facts and Norms: Contributions to a Discourse Theory of Law and Democracy. Cambridge, MA: MIT University Press. Hall, Peter A. 2003. “Aligning Ontology and Methodology in Comparative Politics”. In Comparative Historical Analysis in the Social Sciences, edited by James Mahoney and Dietrich Rueschemeyer, 373–407. Cambridge: Cambridge University Press. Hall, Peter A., and C.R. Taylor. 1996. “Political Science and the Three New Institution­ alisms”. Political Studies 44: 936–957. https://doi.org/Article. Hamilton, Alexander. 1788. “The Federalist Papers. The Judiciary Department”. http:// thomas.loc.gov/home/histdox/fed_78.html. Hinojosa, Víctor J., and Aníbal Pérez-Liñán. 2003. “Presidential Impeachment and the Politics of Survival: The Case of Colombia”. In Checking Executive Power: Presidential Impeachment in Comparative Perspective, edited by Jody C. Baumgartner and Naoko Kada, 65–79. Westport: Praeger Publishers. Human Rights Watch. 2008. “Breaking the Grip? Obstacles to Justice for Paramilitary Mafias in Colombia”. New York. Katznelson, Ira. 2003. “Periodization and Preferences: Reflections on Purposive Action in Comparative Historical Social Science”. In Comparative Historical Analysis in the Social Sciences, edited by James Mahoney and Dietrich Rueschemeyer, 270–301. Cambridge: Cambridge University Press.

The Choque de Trenes

157

Kelsen, Hans. 1945. General Theory of Law and State. Cambridge, MA: Harvard University Press. Krasner, Stephen D. 1984. “Review: Approaches to the State: Alternative Conceptions and Historical Dynamics”. Comparitive Politics 16 (2): 223–246. Kumm, Mattias. 2010. “The Idea of Socratic Contestation and the Right to Justification: The Point of Rights-Based Proportionality Review”. Law & Ethics of Human Rights 4 (2): 142–175. https://doi.org/10.2202/1938-2545.1047. Kumm, Mattias, and Alec D. Walen. 2013. “Human Dignity and Proportionality: Deontic Pluralism in Balancing”. 13–03. Proportionality and the Rule of Law. New York: Public Law & Legal Theory Research Paper Series. https://ssrn.com/abstract=2195663. Levi, Margaret. 1997. “A Model, a Method, and a Map: Rational Choice in Comparative and Historical Analysis”. In Comparative Politics: Rationality, Culture, and Structure, edited by Mark I. Lichbach and Alan S. Zuckermann, 19–41. Cambridge: Cambridge University Press. Levi, Margaret. 2002. “Modeling Complex Historical Processes with Analytic Narratives”. In Akteure—Mechanismen—Modelle. Zur Theoriefaähigkeit Makro-Sozialer Analysen, edited by Renate Mayntz, 108–128. Frankfurt: Campus Verlag. Lichbach, Mark Irving. 2009. “Thinking and Working in the Midst of Things: Discovery, Explanation, and Evidence in Comparative Politics”. In Comparative Politics. Ration­ ality, Culture, and Structure, edited by Mark Irving Lichbach and Alan S. Zuckerman, 18–71. Cambridge: Cambridge University Press. Londoño Hoyos, Fernando. 2001. “La Economía En La Constitución Del 91”. Revista Javeriana 137 (678): 639–648. López Hernández, Claudia. 2010. Y Refundaron La Patria … De Cómo Mafiosos y Políticos Reconfiguraron El Estado Colombiano. Bogotá: Debate. https://doi.org/10.1016/ 1359-0189(86)90036–90031. López Hernández, Claudia, and Oscar Sevillano. 2008. “Balance Político de a Parapolítica”. Bogotá. López Medina, Diego Eduardo. 2000. El Derecho de Los Jueces. Bogotá: Legis. Mahoney, James. 2000. “Path Dependence in Historical Sociology”. Theory and Society 29 (4): 507–548. https://doi.org/10.1023/A:1007113830879. Mahoney, James, and Kathleen Thelen. 2009. “Explaining Institutional Change”. In Explaining Institutional Change: Ambiguity, Agency, and Power, edited by James Mahoney and Kathleen Thelen. Cambridge: Cambridge University Press. https://doi. org/10.1017/CBO9780511806414. Mainwaring, Scott, Ana Maria Bejerano, and Eduardo Pizarro Leongómez. 2006. The Crisis of Democratic Representation in the Andes. Stanford: Stanford University Press. Michels, Robert. 1915. Political Parties: A Sociological Study of the Oligarchical Tendencies of Modern Democracy. New York: Heart’s International Library Company. North, Douglas C. 1990. Institutions, Institutional Change and Economic Performance. Cambridge: Cambridge University Press. Oliver Ortiz, Juan Esteban. 2012. “Tutela Contra Sentencias, Un Cambio En El Paradigma Jurídico Colombiano”. Diálogos de Derecho y Política 4 (11): 1–19. Page, Scott E. 2006. “Path Dependence”. Quarterly Journal of Political Science 1 (1): 87–115. https://doi.org/10.1561/100.00000006. Peñate, Andres. 1999. “El Sendero Estratégico Del ELN: Del Idealismo Guevarista Al Clientelismo Armado”. In Reconocer La Guerra Para Construir La Paz, edited by María Victoria Llorente and Malcolm D. Deas, 53–98. Bogotá: Grupo Editorial Norma.

158 The Choque de Trenes Pérez-Liñán, Aníbal. 2007. Presidential Impeachment and the New Political Instability in Latin America. Cambridge: Cambridge University Press. https://doi.org/10.1017/ CBO9780511510335. Pérez-Liñán, Aníbal, and Andrea Castagnola. 2009. “Presidential Control of High Courts in Latin America: A Long-Term View (1904–2006)”. Journal of Politics in Latin America 1 (2): 87–114. Pierson, Paul. 2000. “Increasing Returns, Path Dependence, and the Study of Politics”. American Political Science Review 94 (2): 251–267. https://doi.org/10.2307/2586011. Pinilla Pinilla, Nilson. 2012. “Balance Del Cambio Constitucional En Cuanto a La Inves­ tigación y El Juzgamiento de Los Altos Servidores Del Estado. Funciones Judiciales Del Congreso de Colombia. Investigacion y Juzgamiento a Congresistas”. In La Constitu­ ción 20 Años Despues. Bogotá: Universidad Javeriana, Fundación Konrad Adenauer. Popper, Karl R. 1968. The Logic of Scientific Discovery. New York: Harper & Row. Pou Giménez, Francisca. 2019. “Supreme and Constitutional Courts: Directions in Constitu­ tional Justice”. In Routledge Handbook of Law and Society in Latin America, edited by Rachel Sieder, Karina Ansolabehere, and Tatiana Alfonso Sierra, 187–204. New York: Routledge. Ramírez Bastidas, Yesid, and Julio Enrique Socha Salamana. 2007. “Sedición y Concierto Para Delinquir”. Corte Suprema de Justicia. Revista 23: 44–63. Rawls, John. 1993. Political Liberalism. New York: Columbia University Press. https:// doi.org/10.1177/0090591790018003001. Restrepo, Luis Alberto. 1996. “El Ejecutivo En La Crisis: Dimensiones, Antecedents y Perspectivas”. In Tras Las Huellas de La Crisis Política, edited by Francisco Leal Bui­ trago, 47–75. Bogotá: Fundación Friedrich Ebert de Colombia. Revelo Rebolledo, Javier Eduardo. 2009. “El Consejo Superior de La Judicatura: Entre La Eliminicaión y La Cooptación”. In Mayorías Sin Democracia. Desequilibrio de Poderes y Estado de Derecho En Colombia, 2002–2009, edited by Mauricio García Villegas and Javier Eduardo Revelo Rebolledo, 248–282. Bogotá: Dejusticia. Rodríguez Garavito, César, and Diana Rodríguez Franco. 2010. Cortes y Cambio Social: Cómo La Corte Constitucional Transformó El Desplazamiento Forzado En Colombia. Bogotá: Dejusticia. Sartori, Giovanni. 1962. “Constitutionalism: A Preliminary Discussion”. The American Political Science Review 56 (4): 853–864. Serrill, Michael S. 1994. “The Narco-Candidate?” Time Magazine. Stinchcombe, Arthur L. 1968. Constructing Social Theories. Chicago: University of Chicago Press. Streeck, Wolfgang, and Kathleen Thelen. 2005. Beyond Continuity: Institutional Change in Advanced Political Economies. Oxford: Oxford University Press. Theidon, K. 2007. “Transitional Subjects: The Disarmament, Demobilization and Reinte­ gration of Former Combatants in Colombia”. International Journal of Transitional Justice 1 (1): 66–90. https://doi.org/10.1093/ijtj/ijm011. Thelen, Kathleen. 1999. “Historical Institutionalism in Comparative Politics”. Annual Review of Political Science 2 (1): 369–404. https://doi.org/10.1146/annurev.polisci.2.1.369. Uprimny, Rodrigo. 1996. “Jueces, Narcos y Politicos: La Judicialización de La Crisis Polí­ tica”. In Tras Las Huellas de La Crisis Política, edited by Francisco Buitrago. Bogotá: Fundación Friedrich Ebert de Colombia. Uprimny, Rodrigo. 2007. “¿En Qué Esta Debate Colombiano de La Tutela Contra Sen­ tencias?” In Estado de Derecho y Sentencias Judiciales. Seguridad Juridica y Garantismo, edited by Rodrigo Uprimny, Juan Jaramillo, Catalina Botero, and Libardo Herreño, 57– 76. Bogotá: Instituto Latinoamericano de Servicios Legales Alternativos.

The Choque de Trenes

159

Valencia Villa, Hernando. 2012. Cartas de Batalla. Una Critica Del Constitucionalismo Colombiano. Bogotá: Panamericana Editorial. Vanberg, Georg. 2008. “Establishing and Maintaining Judicial Independence”. In The Oxford Handbook of Law and Politics, edited by Keith E. Whittington, R. Daniel Kele­ men, and Gregory A. Caldeira. Oxford: Oxford University Press. https://doi.org/10. 1093/oxfordhb/9780199208425.003.0007. Waldron, Jeremy. 2007. “Dignity and Rank”. European Journal of Sociology 48 (2): 201– 237. https://doi.org/10.1017/S0003975607000343.

Court Cases Constitutional Court of Colombia. Auto 004/04. M.P. Clara Inés Vargas Hernández.

Constitutional Court of Colombia. Auto 089/08. M.P. Humberto Antonio Sierra Porto.

Constitutional Court of Colombia. C-543/92. M.P. José Gregorio Hernández.

Constitutional Court of Colombia. C-127/93. M.P. Alejandro Martinez Caballero.

Constitutional Court of Colombia. C-198/94. M.P. Vladimiro Naranjo Mesa.

Constitutional Court of Colombia. C-037/96. M.P. Vladimiro Naranjo Mesa.

Constitutional Court of Colombia. C-222/96. M.P. Fabio Moron Diaz.

Constitutional Court of Colombia. C-245/96. M.P. Vladimiro Naranjo Mesa.

Constitutional Court of Colombia. C-319/96. M.P. Vladimiro Naranjo Mesa.

Constitutional Court of Colombia. C-370/06. M.P. Manuel Jose Cepeda Espinosa, Jaime

Córdoba Triviño, Rodrigo Escobar Gil, Marco Gerardo Monroy Cabra, Álvaro Tafur Galvis, Clara Inés Vargas Hernández. Constitutional Court of Colombia. C-1174/04. M.P. Alvaro Tafur Galvis. Constitutional Court of Colombia. C-454/06. M.P. Jaime Córdoba Triviño. Constitutional Court of Colombia. C-545/08, M.P. Nilson Pinilla Pinilla. Constitutional Court of Colombia. SU-47/99. M.P. Carlos Gaviria Díaz, Alejandro Mar­ tinez Caballero. Constitutional Court of Colombia. SU-786/99. M.P. Hernández Galindo. Constitutional Court of Colombia. SU-811/09. M.P. Nilson Pinilla Pinilla. Constitutional Court of Colombia. SU-198/13. M.P. Luís Ernesto Vargas Silva. Constitutional Court of Colombia. T-079/93. M.P. Eduardo Cifuentes Muñoz. Constitutional Court of Colombia. T-820/99. M.P. Alfredo Beltrán Sierra. Constitutional Court of Colombia. T-965/09. M.P. María Victoria Calle Correa. Supreme Court of Colombia. Process No. 26945 of 2007 (against Orlando César Caballero Montalvo), 11 July 2007. Supreme Court of Colombia. Sala Casación Penal. “Decisión: Resolver la acción de tutela promovida por José Roberto Páez Varón”, 26 July 2007. Supreme Court of Colombia. Process No 31.653 of 2009 (against Edgar Eulisess Torres), 1 September 2009.

4 Discursive Institutionalism in Colombia’s Constitutional Court The Deliberative Judge and Norm Creation in a Politicized Context

Introduction This book set out to explain why the Constitutional Court decided differently on the merits of constitutional reforms in 2005 and 2010. These two decisions are in fact embedded in a process of norm creation and understanding the changes elucidates how the Court’s authority tangibly increased in that period. This chapter will explore behavioral aspects of judges’ decision-making processes to highlight discursive struc­ tures during the deliberation of consequential constitutional litigation. The most recent developments in the research on judicial behavior have shown that the strategic judge can advance judicial independence and authority in specific situations and under certain conditions. The question for this chapter is whether the deliberative judge is also capable of buttressing judicial authority. In order to make the case, this chapter has to show two things: 1) the decision to withhold a potential third term does indeed represent an augmentation of judicial power; and 2) the processes leading up to the decision itself cannot be explained with reference to a strategizing, forward-looking actor, but requires an understanding of a judge that discursively engages in some form of communicative reasoning and deliberation. The deliberative judge comes to conclusions not because of strategic considerations, but because of convictions of the correctness of her interpretation. Moreover, these convictions themselves are not preordained and given, but crafted in the process of deliberation itself. The argument is as follows: the decision to bar Uribe from another potential term in office constituted an objective increase in judicial power, and restrained an otherwise powerful, popular, and often abusive executive. We defined judicial power as the degree of judges’ institutional independence, and their authority in the exercise of judicial review. The outcome in this case was the result of a process of norm creation—the process by which a doctrine attains legal validity in a given jurisdiction. In its reasoning, the Constitutional Court asserted its authority to review (and eventually reject) constitutional reforms by drawing on the “substitu­ tion doctrine”. This doctrine increased the Court’s powers to review constitutional reforms by means of a reinterpretation of norms enshrined in the 1991 Constitu­ tion. In the 1991 Constitution, the norms limited the Constitutional Court’s ability to review constitutional reforms to “errors of procedure” only (Article 241, 1–3). The substitution test involves determining axiomatic principles (which are DOI: 10.4324/9781003229285-4

The Deliberative Judge

161

not explicitly stated in the Constitution) and comparing these with the parameters of the constitutional reform. Critics lament that this represents a form of sub­ stantive review that goes beyond written text of the Constitution. Regardless of the jurisprudential connotations, the 2010 decision most certainly constitutes an increase in judicial power, as it expanded the authority the Court asserts in its judicial review capacity, and unequivocally applied a doctrine to legally validate constitutional reform that had evolved over the span of Uribe’s two terms. This analysis will show in detail that the importance of this decision not only lies in the fact that it constituted an objective increase of power, but also in the context of very evidently majoritarian tendencies in the Uribe administrations. Uribe’s unique position in Colombian history shows in his capability to amass popular and institu­ tional power. Conventionally, Presidents in Colombia were capable power brokers, skilled in forming coalitions between elites through backroom deals, and repaying electoral favors with earmarked funds from the national coffers. The self-styled Anti­ oquian rancher-turned-politician not only commanded the game of government as conversation amongst gentlemen, but enjoyed broad popular support for his Demo­ cratic Security policy. He put his popularity to work by combining it with commu­ nitarian notions of citizenship, and embarked on a very broad reformist agenda that aimed at increasing his own power and curtailing the powers of horizontal control institutions. Therefore, the increase in judicial power, and the Court’s confirmation of its independence, must also be viewed in the context that juxtaposed majoritarian and constitutionalist claims of citizenship and legitimacy. Principled legal reasoning and the institutionalization of deliberative structures within the court, as supposed to utilitarian strategizing, better explains this augmen­ tation of authority. This chapter will trace the evolution of the substitution doctrine’s jurisprudence together with data on Uribe’s popularity, his majorities in Congress, and the specific votes for the legislative projects in both chambers of Congress. This will show that the Constitutional Court deliberately, and deliberatively, planted, devel­ oped, and eventually affirmed the substitution doctrine against the excessive constitu­ tional reformism of the Uribe administration. Importantly, at critical moments—not least in February 2010—its application relied on clear deliberative procedures inside the Court, and was less the result of a court that has struck strategic accords with other actors. More than potential retaliation, or the lack thereof, the cohesiveness of the discourse and coherence of argumentation affected opinions amongst judges. Conjoined with an analysis of the deliberation inside the Court, this chapter will then show that the constitutional magistrate must be understood as a deliberative judge, who must justify her opinions to colleagues that only accept judicial reasoning. The implementation of this deliberation plays a crucial role in the exercise of independent and autonomous decision-making inside Colombia’s high court. I will proceed by first explaining the theoretical and conceptual connotations of the strategic approach to the study of judicial behavior of courts, and then move to the specific majoritarian/constitutionalist divide that surfaced in the re-election debate (and eventual decisions). Against Uribe’s majoritarian sense of mission, born out of the theorization of a communitarian state with corresponding claims of citizenship that postulates an estado de opinión (state of opinion) as superior to the

162 The Deliberative Judge estado de derecho (state of law or constitutional state), this chapter will reconstruct the jurisprudence of the substitution doctrine. This doctrine centers on the question of how far Colombia’s majoritarian institutions can go in reforming constitutional norms without changing the central characteristics of the Constitution. We will see that the strategic paradigm cannot properly account for its creation and evolution, and that any intelligible explanation requires a deliberative understanding of the judge, who incorporates legal context, and evaluates decisions according to their discursive coherence. The analysis will close by carefully examining the deliberative process in that decision. It will show that the decision reflected the careful imple­ mentation of deliberation, which aims at guaranteeing the tranquility of the process so that in the final conclusion it is the law that matters (la ley cuenta), and not personal preferences nor strategic considerations.

Conceptual Implications of the Empowerment of Colombia’s Constitutional Court: Rational Choice versus Deliberative Institutionalism Colombia under Uribe—his majoritarian convictions turned into public policy and constitutional reform—exemplifies the tension inherent to positive law, pitting a powerful president’s claim of speaking for a unified political will against a court positioning itself as the protector of the integrity of the Constitution. Incidentally, this tension corresponds with the boundary problem between law and politics, which just as naturally brings us to the study of courts and their behavior. Judicial behavior follows in the footsteps of Max Weber’s sociological perspective of law and contends that if judges are inserted into a political game, they become political actors vying for influence in the distribution of public goods. Moreover, they are not just any political actors, but ones that interact at the nexus between law and politics, and, above all, define where that boundary specifically runs. Therefore, any investigation into judicial behavior must contend with the basic premise that “judi­ cial politics can be law or politics, but frequently is both” (Segal 2008, 20). When scholars in the positivist school of judicial behavior ask what “judges do and why they do it” (Segal 2008, 20), they conventionally focus on: 1) ideological pre­ ferences that inform judges’ attitudes (Segal and Spaeth 2002); 2) incentives or disincentives that constrain judges’ strategies (Epstein and Knight 1998); and 3) legal texts and precedents that structure judges’ legal context (Whittington 2000; Gillman 2001 and 2002; Friedman 2006). James Gibson is credited with stating it most poignantly (Ríos-Figueroa 2016), arguing that judges’ actions can best be explained as results of processes that combine “what they prefer to do, tempered by what they think they ought to do, [and] constrained by what they perceive is fea­ sible to do” (2008, 516). The purpose of this section is threefold. Firstly, it shows that the legalistic and attitudinalist schools are not helpful for explaining the out­ come in the Colombian case. Secondly, it serves to outline the specification of the hypothesis of the Colombian Constitutional Court judge as a strategic actor. Thirdly, it provides the specification of the constitutional judge as a deliberative actor, outlining discursive constraints and the corresponding hypothesis.

The Deliberative Judge

163

From Political Misalignments to Prudent Deference: The Implications of the Strategic Judge The three dominant schools of judicial behavior evolve around three analytical focal points: judges’ ideological preferences are the focus of the attitudinalist school, institutional constraints and the feasibility of their intended decisions the explanandum of the strategic school, and the strictly legal context consisting of constitutional/legal texts and precedents the focus of the legalistic school. I argue here that the legalistic school holds on to an overly metaphysical conception of law claiming judges are there to “find the law”. It also has difficulties explaining judicial outcomes that are not precipitated by change in the Constitution. The attitudinal model under-appreciates the distinction between political and legal reasoning, and simplifies political attitudes into a two-poled spectrum between conservative and liberal. It also has difficulties traveling outside the US context. Against these con­ tentions, the advantage of the conceptualization of the strategic judge lies in the parsimony of its predictions, and the ease with which the conceptual scope condi­ tions travel across contexts—amongst other places to the Colombian Constitutional Court itself. Therefore, I will weigh the strategic model’s hypothesis, and its obser­ vable implications, against the explanation of the Colombian Constitutional Court judge as a deliberative judge. The deliberative judge does not equate with the judge of the legalistic school. Even if the focus of deliberation in courts is legal questions, and my interviewees stressed the profundity of legal issues and the application of the law, judges remain political actors—but in the Aristotelian sense of the Zoon Politikon rather than the partisan political animal. They cannot evade political socialization that results in some understanding of what constitutes the good life. Such political socialization entails, above all, that judges, too, hold political opinions as citizens of what is right and wrong. The understanding of judges as mere mouthpieces of the law engaged in finding the law is unrealistic, turning subjects into automatons of the constitutional order. Crucially, too, is that processes of legal decision-making are most often open-ended and display an internal relation between political and legal opinion. Ríos-Figueroa poignantly observes, however, that the legalistic judge conceives of constitutional changes as “the primary and sole causes of interpretive changes of that constitution” (2016, 37), yet constitutional interpretation often develops without textual changes to the Constitution. The substitution doctrine is a case in point as it was a principle not explicitly mentioned in the Constitution, and arguably contradicts a textualist interpretation of the Constitutional Court’s judicial review clause that limits judicial review of constitutional reforms to “pro­ cedural defects only”. The legalistic model cannot properly account for such out­ comes, nor can it incorporate a sophisticated conception of the contradicting contents of constitutions themselves. After all, more often than not, these docu­ ments reflect diverse constitutive moments that brought together heterogeneous groups of actors with a variety of interests (Negretto 2013). The proliferation of rights means that the articulation of different rights in the same document is not always compatible, or sometimes even discordant. Coming to legal decisions

164 The Deliberative Judge therefore entails that judges must balance or weigh constitutional norms against one another with the help of different interpretive principles, such as proportion­ ality or efficacy (Kumm 2012). Therefore, theories of judicial behavior must incorporate the complexity of the legal decision-making process itself, and that means appreciating the importance of deliberation and argumentation. While the legalistic model under-appreciates the role of politics in judges as political beings, the attitudinalist model over-appreciates it. This is partly due to the origin of this model: scholars studying judicial behavior with a focus on judges’ political attitudes are mostly confined to the studies of the US judiciary in general, and the US Supreme Court in particular. The model essentially argues that policy preferences along a liberal–conservative continuum not only correlate with the outcome—judges’ decisions—but in fact determine where their votes fall on specific issues. This model is strongly based on the American legal and political system, and therefore faces external validity issues. Its core argument is that in the context of institutionalized appointment and removal mechanisms, judges’ beha­ vior can best be predicted by measuring their political orientation. The problem is that such a one-dimensional measurement of political ideology is even more unrealistic outside of the United States. More importantly, as Ríos-Figueroa rightly points out, it relies on political characteristics found almost exclusively in the United States: life tenure, no chance of removal, and no higher professional achievement for lawyers than reaching Supreme Court (2016, 37). All of this means that outside the US, the ideological composition of the court is only a very weak indicator for the outcome of legal decisions, because the requirements to intelligibly and reliably measure ideology, without institutional context variables interfering, are not given. In Colombia, one judge on the Constitutional Court during the Uribe administration was a known social conservative, who held posi­ tions that homosexuality constitutes something unnatural, and even voiced those positions publicly. In legal decisions concerning gay rights, however, the same judge decided in favor of expanding rights for homosexual couples. The attitudinal school cannot explain such apparent contradictions or rapid shifts in opinion. Rather it suggests that legal opinionizing differs from political reasoning, and deliberation plays an important role for explaining why judges hold different views in their role as guardians of a Constitution than as citizens. The deliberative process inside the Court imposes different constraints and persuades judges to come to different conclusions on the same or similar questions. This leaves the strategic model for our analysis. Its strength lies in the parsimony and clarity of its conceptual features, which, among other things, enabled its appli­ cation in contexts beyond the confines of US institutions. It is, like most models, based in rational choice theory, a risky modulation of human behavior that makes intelligible and falsifiable predictions (Popper 1968). Latin America specifically has provided ample cases and developments, which, viewed through the strategic prism, have boosted our knowledge on when, and under what conditions, judges act the way they do. The strategic account of the constitutional judge refers to the part in Gibson’s equation of judicial behavior that explores judges’ constraints and their perception of what is feasible to accomplish. These constraints arise from the

The Deliberative Judge

165

institutional context, the other branches of government in a separation of powers system, offering a synonym for strategic model. The separation of powers model evolves around the assumption that judges, too, act strategically when they take decisions. It presumes an actor with a sophisticated knowledge of the institutional context and legislative process. It then inserts the judge in a game with other forward-looking actors, who behave pur­ posively on the merits of well-defined preferences (Spiller and Gely 2008, 35). In its empirical application, the model predicts that a court maneuvers between the positions of the legislative and executive branch that fall under the “contract curve” between their policy preferences. As an actor that reacts to incentives and constraints, the median judge then has a strong inducement to decide in a manner that is unlikely to be overturned by either the legislature or executive. Epstein and Knight conveniently tie the assumptions of the separation of powers model into three core suppositions: � � �

“Under strategic accounts, it is up to the researcher to specify a priori the actors’ goals.” “Strategic accounts assume that when goal-oriented justices make their deci­ sions, they take into account the preferences and likely actions of other rele­ vant actors, including their colleagues, elected officials, and the public.” “Strategic accounts assume that judging takes place within a complex institu­ tional framework—that sets of rules structure the judges’ interactions among themselves or with other relevant actors” (2013, 12).

These pillars of the separation of powers model evidence its apparent strength: the ability to make parsimonious and falsifiable predictions, based on intelligible conceptualizations. One of the most important contributions Latin Americanists employing the strategic model have made is this: judicial independence often arises in politically uncertain or politically unstable contexts. Research into institutional trajectories in authoritarian, one-party, hegemonic, and democratic regimes with competi­ tive elections has shown that divisions between factions within (even autocratic) governments, as well as between branches of government, creates political uncertainty. That uncertainty, also evident in the context of imminent regime change in authoritarian states, opens a political vacuum that independent high courts are willing to fill—or are even sometimes asked to fill by those political actors entrapped in an uncertain political game. This is the foundation of the political uncertainty argument of judicial independence. The plethora of autocratic dictatorships in the Southern Cone has given rise to questions: why have courts not done more to prevent, or at least investigate, the abundant human rights violations? Have courts in Argentina, Chile, and Brazil stood aside because of institutional incapacity or indisposition? The analyses of the institutional structures of authoritarian regimes, and judicial culture amongst judges, in places such as Chile and Argentina during different periods of their respective political histories have provided important insights into the trajectories of judicial

166 The Deliberative Judge bodies under authoritarian regimes, including what constraints and incentives for defection they may face. Hilbink has argued that the Chilean judiciary’s traditionally conservative legal culture has turned them into apolitical actors that let Pinochet’s abuses go unchecked (2007), while Couso (2003) interprets the Chilean judiciary’s deference to General Pinochet as a consequence of strategic decisions to avoid con­ troversial cases and maintain a minimum of institutional independence. Yet, just like competitive regimes, authoritarian regimes also provide openings for a judiciary to exercise some form of judicial independence. Barros showed that the existence of different factions within military regimes increased the need for coordination between groups. Principle–agent dilemmas together with the need to coordinate and bridge internal division open spaces, or niches rather, for courts to exercise authority (Barros 2002). In other words, these cleavages, even inside autocratic regimes, can be a source of political uncertainty, requiring bargaining mechanisms, and courts can fulfil that function to coordinate between different factions. Coordination, however, is contingent on a miniscule degree of institutional independence, without which the ability to harmonize and delegate between factions was not credible. Semi-competitive regimes in processes of transition also generate an uncertain context that creates conditions favorable for judicial independence. Building on Ramseyer’s findings (1994) that courts in Japan tended to subordinate to the executive in the course of long terms of one-party dominance, Domingo found that PRI hegemony in Mexico had the same effect on courts (2000). When oneparty rule started to crumble, however, and PRI dominance showed signs of open decay, ruling parties were willing to negotiate legal guarantees. This is what hap­ pened under President Ernesto Zedillo in Mexico in 1994 (Finkel 2005 and 2008; Ríos-Figueroa 2007). In 2000, Vicente Fox became the first opposition party president in Mexico since the Revolution. The logic of the uncertainty argument also holds for non-autocratic regimes. In competitive democracies, space for judicial dissent opens when political interest is not aligned. Helmke found such behavior in her seminal study of Argentinean courts. Robert Dahl had argued that justices appointed by a previous administra­ tion acted in accordance with the legal political cycle, meaning they defected from the incoming administration at the beginning of the term and deferred to their position at the end of the term. Helmke, on the other hand, argues that towards the end of a term, judges defer to the position of the incoming administration to avoid being purged after the change of office holder. For Argentina, the uncer­ tainty argument holds for judges active during the military regime, as well as under post-transition democratic regimes (Helmke 2004). Moreover, Chávez, Ferejohn, and Weingast showed that Argentina and the United States are fairly similar in that regard: courts are more likely to challenge the executive without fear of political retribution if disagreement exists between elected officials (2011). Finally, the strategic school has also helped to elucidate the behavior of Colombia’s Constitutional Court with the analogy of the prudent strategizer. Differentiating between the horizontal (interbranch arbitrator) and vertical control (rights adjudicator) functions of constitutional courts,1Rodríguez-Raga (2011) contends that while the Colombian Constitutional Court has acquired the reputation

The Deliberative Judge

167

of a rights-enforcing institution (see also Bonilla Maldonado 2013; Hailbronner 2017), it has been relatively weak when it comes to adjudicating constitutional dis­ putes between the various branches of government (horizontal control). Analyzing data from the Colombian Constitutional Court and the acción popular the incon­ stitucionalidad (1992–2005, thus prior to the 2010 decision!), he found evidence that pointed toward a relation between strategic deference and the “costs for the players associated with both giving up policy and clashing with each other” (95). Succinctly stated, he argues, that the Court most often acts like a prudent strategizer, laying the jurisprudential groundwork for expanding its jurisdiction by upholding the principle in prominent cases, but deferring to the executive on the substantive merits of those cases. In less prominent cases in its docket, the Court then moves to enforce these principles against the executive (2011).2 To sum up this short review of the classical schools of judicial behavior, in par­ ticular its strategic/rational choice variant, it is sensible to list the most important findings. This is already a collection of findings that applies to this study of the application of the substitution doctrine in Colombia. At the end of this section, following the discussion of the deliberative judge, these will be articulated as concise hypotheses to be tested against the predictions of the deliberative judge. Without a doubt, the strategic school showed the importance of uncertainty for facilitating a higher degree of judicial independence and authority. Specifically, inflated ambiguities about future political constellations opened political space for courts to occupy. Intriguingly, we saw that the strategic school found evidence of the beneficial effects of political opacity for judicial independence across different type of regimes—ranging from authoritarian through semi-competitive to com­ petitive. Given these findings, we need to explore if we can identify a political opening for the court to exploit. Was there indeed a misalignment between executive and legislature that helped the court to act with defiance, or even a major regime transformation? The other key finding from studies that evaluate the costs and constraints of other players in zero-sum games was Rodríguez-Raga’s position that the Colom­ bian Constitutional Court up until 2005 had acted akin to a prudent strategizer. In effect emulating Marshall’s strategic action in Marbury v. Madison for estab­ lishing the principle of judicial review in the US Constitution, the Colombian Constitutional Court has deferred to the executive or legislature on the substance in a litigation, while asserting the validity of constraining principles. It would then assert these principles later and on the substantive merits of those cases. Therefore, we need to explore if this strategy of prudent strategizing also applies to the sub­ stitution doctrine—a process that falls outside the time frame Rodríguez-Raga examined in his study. Finally, there is an ontological caveat to rational choice paradigms that perceive policy and institutional outcomes as the result of games played between strategic actors: the rules of the games, as well as preconceived preferences of actors, remain constant throughout the game. The stability of preferences will be a crucial para­ meter in the examination of deliberation during the re-election decisions. Without getting too far ahead of myself, I want to posit here that this notion is particularly

168 The Deliberative Judge problematic for deliberative conceptualizations of the constitutional judge, because, as I will explain below, deliberation is ontologically tied up with the ability to change one’s mind in the face of the better argument. A Competing Explanation: Deliberative Constraints and the Ability to Change Your Opinion Against the hypothesis of the constitutional judge as a strategic actor, I contend that the micro foundation of practicing law, deliberation, has been conspicuously missing (this counts for all the models explained above). Yet, when asked, judges stress that collective contemplation and argumentation form the core of their interaction inside courts. In fact, the imperative of coming to well-reasoned con­ stitutional decisions offsets a (political legitimacy) dilemma: the Constitutional Court must decide politically consequential decisions, but its composition is not legitimized through electoral processes. Therefore, the task of this section is to develop an alternative model of judicial decision-making processes—a model that focuses on the process of legal argumentation. Such a model would not only help us to explain what the deliberative judge looks like, but also let us differentiate between discursive and strategic incentives. This is a very timely endeavor. Epstein and Knight have raised concerns that the strategic paradigm’s insistence that policy preferences drive judges’ behavior is lopsided. They argue that legal scholars, who contend that law does indeed matter, should be taken seriously and (legal) argu­ mentation should become a central variable in social scientific inquiry of judicial behavior (2013). Yet, while studies on the essence of legal argumentation occupy a central place in normative legal theory, conceptualizations for comparative studies are not as abundant. I argue here, borrowing from normative legal theory, that such a conceptual operationalization of the deliberative judge must pay credit to the notion that the constitutional judge engages in the practice of law and is employed in legal argumentation. With that the deliberative judge stands between the liberal model of a quasi-prodigious arbitrator (Rawls and Dworkin), and the textualist interpreter of constitutions that assuages the counter-majoritarian difficulty promi­ nent in conservative models. Translated into the key observable implication that sets it apart from other models is the expression of discursive agency in the deliberative judge: the ability to change her mind when faced with better arguments. Montesquieu’s ideal of the judge as the mouthpiece of the law fell on fertile ground in the US context, where originalism or textualism is held in high regard amongst conservative lawyers. It presupposes the normative contention “that there can be deliberation in lawmaking, but not in law application”, as it would unduly politicize adjudication and, thereby, law itself (Hübner Mendes 2013, 54). The judge ought to be an apolitical arbiter. In this book, I have already utilized the incremental change model that suggests that the implementation of rules provides space for actors to influence the outcome, because abstract rules can never cover the entire range of real-life possibilities, making interpretation of rules an essential process of implementation. Hübner Mendes’ rebuttal to the imperative of the constitutional judge as neutral arbiter voices a similar sentiment. He argues that

The Deliberative Judge

169

the imperative to solely apply the law is misleading: “[I]ts gross mistake is not exactly the very distinction between lawmaking and law application, but rather the notion that, in the latter case, there is no room for debate, no latitude or reason­ able controversy” (56). After all, constitutional norms are particularly abstract to uphold law’s neutrality. Borrowing from Alexy, Hübner Mendes contends that “the concept of law … combines a factual and an ideal dimension in realizing ‘optimal coupling of legal certainty and justice’”, meaning that you cannot know “the law just by reading its public and printed black-letters” (2013, 57). When placing constitutional courts in the context of modern democracy, in which actors engage deliberatively with each other, it becomes additionally diffi­ cult to assume that the court as an institution is void of linkages with the greater public, and the judge an automaton or phonograph of written law. Incidentally, Latin America has experienced the reverse of the politicization of law through the judicialization of politics (Sieder et al. 2005). The reasons for that trajectory are telling as they highlight the correcting function of judicial institutions to political pathologies of young democracies. First, the increasing importance of courts in Latin America (Sieder et al. 2005) is closely related to their functions in separation of powers models, where, ideally, they exercise the monopoly over the interpreta­ tion of laws and their application to concrete cases (Cameron 2013, 42). This function to translate text into language and deeds also involves a representative function, but, in contrast to political institutions engaged in the process of law­ making, it is representative of arguments and not votes or interests (Alexy 1989, 16). Kumm, for example, argues “it is not clear what the issue of deep principle could be, that would condemn judicial review, but not electoral representation” because “there are no plausible reasons to identify ‘the people’ with the voice of one institution, even when that institution is a parliament”, and not with another (Kumm 2007, 26). After all, “in the real world of modern territorial democracy, the right to persuade a court to veto a policy is at least as empowering as the right to change policy” (Kumm 2007, 27). Legal language put into writing in the form of legislation requires specification (Alexy 1989). Courts fulfill that function, and participation in argumentative processes before courts is therefore empowering. Further rebuttals to the counter-majoritarian difficulties can be found in liberal models of the adjudicating court. The problem for this analysis is that such nor­ mative theories load the institution to constitutionally review legislation with expansive cognitive capacities. Both Rawls and Dworking place constitutional courts accordingly central in their respective system to coordinate between power and rights—Rawls as a public reasoner (Rawls 1993), Dworkin as the Herculean arbitrator that coordinates from an Archimedean point (Dworkin 1986). Both accounts address the demand for legitimacy of positive law that demands obedi­ ence from free citizens. Rawls’ contention that the Supreme Court—both accounts are largely descriptive of the US Supreme Court, but prescriptive of judicial review at large—faces larger constraints as an exemplar of public reason burdens the judge to “refuse to engage in a comprehensive doctrine of the good, and remain within the bounds of a strictly political conception of justice” (Hübner Mendes 2013, 14). Dworkin designates the court that exercises judicial review as

170 The Deliberative Judge the “forum of principle” whose pillars reflect the ideals of the community as a whole. Judges’ constituents are therefore the entirety of the political community, because “an elected branch cannot be sufficiently trusted as the ‘forum of princi­ ple’ [due to] the counter incentives it faces” (Hübner Mendes 2013, 15). Brazilian constitutionalist Conrado Hübner Mendes provides the pivotal cri­ tique of the liberal conceptualization of the constitutional judge. He argues that Rawls bequeaths the judge with overly expansive cognitive capacities that places her outside of social and political forces, while Dworkin assumes a Herculean arbitrator that coordinates from an Archimedean point (Dworkin, 1986). As Michelman argued before him, “Dworkin’s mythic judge is a loner [and] much too heroic. His narrative constructions are monologies” (Michelman, cited in Hübner Mendes 2012, 18). As Hübner Mendes exalts, and I confirm in the ana­ lysis of decision-making within the Colombian Constitutional Court, law is delib­ erative in the inter-institutional sense, as well as in the intra-institutional sense: “judges deliberate with each other within collegiate adjudication” that “requires participants to display the reasons why they support a particular stance. … Silence is acceptable neither during nor after the process” (2013, 53 and 15). Our con­ ceptualization should therefore pay attention to how the constitutional judge interacts with other branches of government as well as with her peers on the court. The Socratic model of the constitutional judge is well placed to crystallize the key components of the deliberative judge as a model for judicial behavior. It incorporates notions of deliberative democracy, and connects them with judicial institutions that actively, but not paternalistically, engage in the political process. As Kumm explains, Socratic contestation is a “practice that gives institutional expression to the idea that all legitimate authority depends on being grounded in public reason, that is, justifi­ able to others on grounds they might reasonably accept” (2007, 4). Courts ascertain whether the settlement of the disagreement between the public authorities and the rights claimant is in fact reasonable, but they are not, as Kumm stresses, in the busi­ ness of settling reasonable disagreements. The advantage is that, as a normative model of what judges do, it does not claim to pursue higher, metaphysical truths, but nevertheless excludes certain arguments from the path towards attaining legal truths: such as claims rooted in tradition, ideological claims trespassing the limits of public reason, and the utilization of coercive power. Crucial is that the rights bearer has access to contestation and justification, and that the court evaluates the proportion­ ality of the claims in a reasonable manner (Kumm 2010). When all relevant social actors can bring their arguments to the table and present their defense, an ideal-type setting ought to force utility-maximizing actors to persuade others with reasons and arguments, diminishing the importance of resources at their disposal. The normative ideal of Socratic contestation might appear elusive, because it functions not from a cosmologically fixed moral universe or normative principles. Yet, this elusiveness is precisely its advantage in a post-modern world lacking a unified ideological canopy. After all, Robert Alexy reminds us that “anyone who equates rationality with certainty must renounce the idea of a theory of legal argumentation” (1989, 293). Socratic contestation, on the other hand, forces us to focus on the validity of the structures, in which different moral universes interact, and

The Deliberative Judge

171

defends the sanctity of deliberative procedure itself. At its core, it has three normative imperatives that emulate Robert Alexy’s rules of reason: (1) prevent a rationally unmotivated termination of argumentation; (2) secure both freedom in the choice of topics and inclusion of the best information and reasons through universal and equal access to, as well as equal and symmetrical participation in, argumentation; and (3) exclude every kind of coercion—whether originating outside the process of reaching understanding or within it—other than that of the better argument, so that all motives except that of the cooperative search for truth are neutralized (Habermas 1996, 230; Alexy 1989, 136–137). This then accepts that judges may have a back­ ground formed by specific political and ideological biases. It can even accept that the ultimate decision overlaps with these biases. It cannot accept, however, that the rea­ soning resulting in these decisions is rooted in these biases and defies the better legal argument that would lead into a different direction. In other words, legal deliberation is not autonomous from politics, but neither deterministically entrenched in judges’ political socialization. It is open-ended. We can again turn to Brazilian constitutionalist Conrado Hübner Mendes, who has made an enormous contribution to outline the specific contours of a deliberative judge (2013). He, too, accepts the “co-legislative” nature of constitutional scrutiny, but stresses that it is not identical to parliamentary debate. Rather, proper legal rea­ soning is situated between the poles of lawmaking and law application, and therefore always involves deliberation (2013, 221). Hübner Mendes first defines legal delib­ eration as a “variant of practical reasoning applied to collective decision-making pro­ cesses” that is contingent on participants being open “to transform their preferences in the light of well-articulated arguments” (18). This definition contains in it seven caveats crucial for our investigation here: 1) it recognizes the need to come to a col­ lective decision, since we are concerned with a political exercise that involves ques­ tions of authority; 2) decision-making does not end with the decision, but only temporarily arrests developments, because continuity, for deliberative theorists, is an integral part of legitimate politics; 3) deliberation combines reason giving as the grounds for a decision; 4) reasons can have many different origins, but share that they are (plausibly) directed at the common good, attempting to conform to some form of impartiality; 5) the inter-subjective engagement to persuade each other requires three conditions of its own: the willingness by participants to review their position, the presence of an ethics of conversation, and the absence of coercion; 6) deliberation assumes an ethics of respect that is based on the mutual recognition of equals; and 7) deliberators’ attitudes reflect a morality of inclusiveness, empathy, and responsiveness to demonstrate that they are fellow members of a political community (14–16). The centrality of the ability to change preferences points to the distinctive impor­ tance of procedures while coming to collectively binding decisions that can result in discursive transformations of institutions. This is not a coincidence: Habermas’s dis­ course ethics shifts the normative nucleus of political legitimation from metaphysics to rational argumentation (1996). The exclusion of every kind of coercion other than the force of the better argument to operationalize the cooperative search for truth (Alexy 1989) carries the normative foundation of deliberation. Incidentally, the dis­ covery of procedure as a key component of deliberative institutional change in the

172 The Deliberative Judge empowerment of courts is a key finding of this book. It is an addition to more recent, discursively institutionalist, contributions that have explored endogenous sources (Bakiner 2020) and the role of “ideational carriers” (Nunes 2010) for the growth of judicial power—in Colombian cases, no less. In sum, the discussion of (normative) theories of the deliberating judge coalesces into one key observable implication for the purpose of this study: the deliberative judge remains open to change her opinion in the face of the better argument. This remains true in the context of political socialization and changing legal contexts. In fact, this brief review of normative models of the adjudicating judge showed that it is entirely implausible to conceive of an apolitical judge. This in turn makes delib­ eration and the force of the better argument indispensable for a legitimate judicial review body. We have seen (Chapter 3) that norms and rules are abstract and require interpretation for the purposes of implementation. The constitutional judge must therefore engage with arguments that arise in the legal context, be they poli­ tical, social, or cultural. The judge as the automaton of the law is simply incon­ ceivable. It is the imperative to justify positions to peers that makes legal positions tenable for a democracy. It is also this process of justification that curtails the pool of available arguments, which in turn facilitates the discursive legitimacy of judicial bodies. A judge must argue with legal principles, not political convenience, and remain open to other arguments. The corollary for this analysis then becomes: is there evidence of consequential, mid-process, opinion changes during deliberation? Hypotheses: Political Uncertainty and Prudent Strategizing vs. Principled Deliberation The importance of the 2010 decision disallowing constitutional reform lies in the fact that it made readily apparent that the Court’s approach has shifted. If in 2005 it still acted with care, by 2010 it had become an assertive adjudicator. It enforced its authority by deciding against a powerful and popular President. Post-2010 develop­ ment affirms this more powerful position of courts within the institutional framework as a consequence of that decision, since the executive now makes deliberate efforts to consult with the high courts prior to commencing with a reform project.3 The evi­ dent task is how to explain this increase of judicial authority, and from the discussion above follows that the central competing hypotheses that this book will return to is whether Colombia’s Constitutional Court’s empowerment is better explained with reference to strategic or deliberative action. The table below (Table 4.1) lists the key questions that this analysis of the creation of the substitution doctrine will explore. As becomes readily apparent, they are based on the review of the key models and theories of the adjudicating judge. In order to get a satisfying answer to whether the evident increase in judi­ cial power is resultant from strategic or deliberative action, we need to examine the possibility of a political opening (i.e. a non-alignment between executive and legislature), explore evidence of prudent strategizing (i.e. laying jurisprudential groundwork in earlier decisions that is exploited in subsequent ones), and ulti­ mately identify if preferences remained constant (i.e. the stability opinions during

The Deliberative Judge

173

Table 4.1 Strategic vs. deliberative action

Was there a tangible political opening for the Constitutional Court to strategically defect?

Strategic Action

Deliberative Action

�

Misalignment between parties in executive and legislature Factionalism within governing coalition Tangible decrease in public support for executive and/or governing coalition in legislature Imminent regime change

�

Laying jurisprudential groundwork in less consequential cases Deference to executive in consequential cases

�

� �

� Was there prudent strategizing by the Constitutional Court?

� �

Did preferences remain constant throughout decisionmaking processes or was there a change of opinion induced by the force of the better argument?

�

No change of opinion during deliberation

� �

�

No political misalignment inside government nor across executive and legislature Constant public support No imminent regime change

Jurisprudential develop­ ment in most consequential cases Defiance towards execu­ tive in consequential cases

� Change of opinion during deliberation

deliberation of judicial decisions). Table 4.1 lists these and connects them with observable implications. The subsequent analysis will ask the questions listed in Table 4.1 to ascertain if the Constitutional Court’s behavior in the creation of the substitution doctrine, and its decision to disallow a third term in the presidential office, is better explained with the strategic or deliberative judge. The argument I will build, namely that the deliberative judge is the better model to explain the outcome in this case, rests on the importance of the tensions in any constitutional order between popular power and constitutional control, and the fact that the political leadership of President Álvaro Uribe heightened this tension with his sense of mission and claim to popular sovereignty.

Constitutional Reform, Competing Claims of Citizenship, and Ethical Discourse in the Process of Norm Creation in Colombia’s Constitutional Court The value of this analysis for our understanding of institutional processes arises from its normative connotations that are wedged between communitarian and constitutionalist

174 The Deliberative Judge principles of citizenship. Its comparativist significance results from Álvaro Uribe’s unique political leadership that delegated political authority at the expense of hor­ izontal control. The epistemological contributions lie in the fact the traits of Uribe’s political leadership run counter to the predictions of the strategic account of judicial behavior that focus on conceivable political openings for judges’ autonomy, while the 2010 decision makes a deliberative account of the adjudicating judge more plausible. It appreciates the importance of procedures of deliberation for the autonomy of prin­ cipled legal reasoning inside collective decision-making bodies such as high courts. Normatively, questions of constitutional reform involve the justification of political power as such, and what constitutes a legitimate political will that provides the nor­ mative scaffolding for legal democratic decisions, including decisions on the con­ stitutional order itself. The majoritarian lineage embraces Rousseau’s General Will as the only source of legitimate power, and views law as the public and solemn declara­ tion of the General Will on an object of common interest (2002, 132). In the legit­ imation of constitutional reform, the General Will reappears in the estado de opinión. For constitutionalists, the General Will remains a performative act without relation to reason if it is not constrained by procedure. In the reality of modern democracy, constitutionalists argue, no political body or entity can be intelligibly connected with one will (Kumm 2007). Rather, any polity consists of a plurality of viewpoints, identities, and, above all, individuals, that must recognize each other as equals. This recognition of equals that constitute a plurality is the foundational claim that justifies and controls democratic decisions—an argument that reappears in the Constitutional Court’s decision to limit presidential re-elections (see, for example, C-141/10; 99). These aspects of the constitutionalist/majoritarian divide, their ontological claims on the nature of the democratic will, and their respective con­ tentions regarding the institution of judicial review, are manifest in the question of constitutional reform. When we ask how far a political reform, initiated by popu­ larly elected bodies, can in fact go, we also pose foundational questions of whether the exercise of political power is legally sanctioned or popularly mandated. In other words: is the power to reform the constitution itself constrained by con­ stitutional principles that must withstand legislative authority, or does legislative authority, coupled with popular sovereignty, supersede constitutional principles? Álvaro Uribe was not like any other President in Colombia’s history. By virtue of his charismatic appeal, and successes of his security policies, he coalesced a substantial amount of popular support behind him, and by virtue of his sense of mission put significant stress on the horizontal control function of Colombia’s constitutional system. He exemplified traits of delegative leaders that utilize extremely high approval ratings to push aside horizontal controls, and he asserted a sense of mission to reform the Colombian polity along lines that, more often than not, served his personal poli­ tical needs. The resulting clash with those institutions that are authorized to uphold the rule of law and integrity of the Constitution was therefore not coincidental. In the normative framework of his Democratic Security policy, Uribe conjoined commu­ nitarian notions of citizenship with paternalistic claims of political leadership that ultimately culminate in the estado de opinión (state of opinion). This discussion will show that Uribe’s rendition of the estado de opinión remained relatively ambiguous,

The Deliberative Judge

175

even contrarian, towards the estado de derecho (state of law; rule of law). It provided the justification of the delegative trends of Colombia under Uribe, not least his ambitious reform agenda. The delegative turn in Colombia’s democracy was not solely theoretical in nature but had practical ramifications for the exercise of politics. A corollary of delegative leaders is that they usurp, or encroach on, constraining institutions. Uribe attempted and succeeded in bringing the majority of Congress behind his political agenda, including his appeals to change the constitution. Throughout his time in office, he could count on healthy majorities in both chambers of Congress—majorities that easily passed constitutional reform to change presidential terms limits. Strategic accounts, however, predict that judicial independence and empowerment are usually precipitated by periods of political uncertainty in the form of misalignment between executive and legislature. No such political uncertainty materialized while Uribe was in power. Moreover, this delegative trend and Uribe’s sense of mission bodes badly for the prudent strategizer argument—that a high court lays out principles while deferring to the executive on the merits in consequential cases for those principles to be used in later cases. The argument I present here shows that the substitution doc­ trine was indeed aimed at Uribe and the most important elements of the theory advanced in highly consequential cases. Thus, it remains an empirical anomaly that the Constitutional Court decided against re-election reform: there was not a political opening, a political space, that the court could exploit. The reconstruction of the jurisprudential development of the substitution doc­ trine will reveal one particularly serendipitous moment within the Constitutional Court: the majority’s insistence to fully apply the substitution test to explore potential competence defects, after the ponente declined to apply the theory of competence. It had such outsized significance for the final outcome, because it was a high-stakes moment that tied several aspects of the deliberative judge together: the strategic opportunity to reject the constitutional reform on mere procedural grounds (which the Court eventually forfeited); the importance of comprehensive procedures as a fortification against any kind of coercion during deliberation (that the Court upheld); and, above all, the ability to change one’s mind in the face of the better argument that occurred during the deliberation in 2010. It is this moment that requires explanation. Ultimately, the explanation is that legal delib­ eration is a “variant of practical reasoning applied to collective decision-making processes” that is contingent on participants being open “to transform their pre­ ferences in the light of well-articulated arguments” (Hübner Mendes 2013, 221). Such deliberativeness is upheld with an intertwined codex of formal and informal rules to warrant that legal reasoning takes precedence over political arguments. I begin by laying out the political foundations of Uribismo, and the appeal to communitarian notions of citizenship that served to justify constitutional reform. From there, I will move to the analysis of legislative politics, specifically the tra­ jectory of majorities in Congress. This sets the stage for conducting an in-depth analysis of the creation of the substitution doctrine in the Constitutional Court’s jurisprudence as well as the process of deliberation during the key decision to reject constitutional reform in 2010.

176 The Deliberative Judge Uribe and the Delegative Turn in Colombian Politics Democratic Security, Uribe’s public policy program, had three focal points: buttres­ sing internal security, inviting foreign direct investment, and creating a tight social fabric through trusting relations between rulers and ruled. The focal shift in this chapter to majoritarian and constitutional tensions in the legitimation of constitu­ tional amendments places the attention on the third aspect of his policy approach: the imperative to create a tightly knit social cohesion between rulers and ruled. Given the foundation in trusting relations between rulers and ruled, it is not surprising that the practical implications of Uribe’s normative positions incentivize delegative political developments. To place trust in rulers not only runs counter to liberal constitutional constraints on popular power that conventionally implore citizens not to trust those in power; it is also a convenient way to delegitimize horizontal control institutions as outside the democratic bond between rulers and ruled. Of course, the nefarious temptation to utilize an ideology that values paternalistic leadership over constitu­ tional constraints becomes immediate in a context where that leader does indeed enjoy extremely high popularity ratings (see Graph 4.1 below).

90 80 70 60 50 40 30 20 10 0

Approval Ra�ngs

Aug-05

Feb-05

Feb-04

Aug-04

Feb-03

Aug-03

Aug-02

Feb-02

Feb-01

Aug-01

Feb-00

Aug-00

Feb-99

Aug-99

Disapproval Ra�ngs Aug-98

In percentage

(Dis) approval ra�ngs

Month and Year

Graph 4.1 (Dis)Approval Ratings of Álvaro Uribe, 2002–2010

With his charismatic appeal and extra-ordinary popularity, it is not far-fetched to place Álvaro Uribe into the context of populist politics. However, being raised in a traditional rancher (ganadero) family, he was neither an outsider to the upper socio-economic stratum, nor unfamiliar with the political machinery in his home department, as conventional models of populism in Latin America suggest (Levitsky and Loxton 2013). On the contrary, as an activist of New Liberalism in Antioquia in his younger years, his political network quickly paid off, opening doors to leadership positions in the bureaucracy and political offices in Antioquia and Bogotá.4 While Uribe was certainly not an outsider to the upper echelons of power in Colombia, he made use of populist strategies to attain the presidency, and utilized plebiscitarian claims to leadership to cement his position in power. In

The Deliberative Judge

177

the end, Uribe treated populism as a potent strategy to attain and maintain power (Weyland 2013). As the late Ana María Bejarano argued: his was an instrumental populism in the service of a reactionary political project (2013). The cultivation of trustful relations between rulers and ruled forms the normative core of his self-understanding in the political arena. Uribismo aspires to transcend the social and political cleavages within a society by affirming the General Will. The source of the cleavages within society is not class divisions and conflict, but dialec­ tical dissociation between groups. Uribe’s most important political advisor, Jose Obdulio Gaviria Vélez, asserted that the “General Will is good and to understand it does not require war, but dialectic” (Gaviria Vélez 2004, 34). Gaviria’s articulation of Uribismo’s normative foundation converges on two specific characteristics of the General Will: 1) the General Will of a nation is above the particular wills that con­ stitute it; and 2) finding the General Will requires special relations between rulers and ruled that ought not to degenerate into conflict, but should be cherished through trustful relations. He arrives there via an elaboration on the distinct origins of the republics in North and South America. While in the North, ideological clea­ vages were assuaged through dialogue and a dialectical search for the General Will, the Southern Cone plunged into warfare and factional hatred between caudillos (28– 33). A Senator from Uribe’s coalition in Congress said: “everything can be nego­ tiated and brought to a conclusion with a majority without the use of violence […] through a permanent dialogue of government with society”.5 By situating the dialo­ gue between rulers and ruled as a way to assuage social cleavages, Uribistas aimed to contrast their normative core to the ontological core of social life that leftist guerrillas in Colombia and Chavistas in neighboring Venezuela in the pursuit of the Bolivarian revolution espouse: the division of historically constructed social classes. Crucially, if facilitating trust between rulers and ruled is imperative to avoid violent conflicts, creating compromises to soften contentious class divisions becomes unnecessary. The foundational role of trust in the normative framework of Uribismo does not only carry ramification for social and cultural life, but is consequential for political society as well. Essentially, situating social cleavages in the context of the dialec­ tical affirmation of the General Will gives rise to a paternalistic sense of leadership. Connecting Burke and ancient Greek philosopher Anaxagoras with the originalism of the General Will, Gaviria argues that true political leadership rests on a coherent vision of life with a corresponding moral universe that transcends particular inter­ ests (2004, 100). In the specific case of Uribismo, said moral universe is a mix of regional peasant folklorism invoking the natural and paternalistic leadership of the hacienda owner and justifies an extremely conservative political project (Bejarano 2013, 336). The result is a veneration of political leadership that does not place Uribe on the side of the people against elites—as Latin American populist Cau­ dillos a la Peron or Chavez proclaimed in the justification of their leadership—but “placed traditional elites back on their pedestal” (Bejarano 2013, 337). The argument is that rulers are to be trusted, since their excellence in political leader­ ship propelled them to the pedestal in the first place. Affirmingly, a Constitutional Court judge explained Uribe’s appeal in a society that was shifting to the right.” He had a very Bonapartian posture (postura bonapartista) with symbols that were

178 The Deliberative Judge very close to the pueblo; of the campesinos (riding horses in the country-side), with the appearance of a traditional landlord in Antioquia.”6 Given these ideological contentions that embrace trust bestowed upon leaders to operate without deontological constraints, it is not surprising that President Uribe inflamed the inherent tension between majoritarianism and con­ stitutionalism with his personalistic traits of leadership, and the resulting sense of mission to reform the political charter. Applied to the political framework, the Uribista rendition of the General Will results in a normative majoritarianism that situates the highest authority in the nation with the political decision of the people in competitive elections and rejects constraints on the people’s free expression other than formal and procedural ones. I argued that constitutional reform always treads on the contentious line between the legitimacy of constraints on popular power and the legality of the reform itself. This tension becomes nowhere as evi­ dent as in the so-called estado de opinión (state of opinion) Uribe used to justify his re-election reform.7 The semantic relation with the estado de derecho (state of law or rule of law) is deliberate, as is the imprecision of whether it is superior or equal to it. For Gaviria, it is the element of democratic governance that relates the mandate holders of political power to the electorate. Similar to the General Will, the estado de opinión precedes constituted power and therefore has qualities asso­ ciated with constituent power and plebiscitary democratic governance. From this original moment of political genesis, the estado de opinión radiates significance that informs practical politics as well as institutional design. In their argumentation on democratic virtues, Uribistas contend that the various organs of the state are sub­ jected to the supervision by public opinion only. The estado de opinión situates the highest authority in the nation with the poli­ tical decision of the people. This decision expresses itself directly through compe­ titive elections or other plebiscitary institutions, and therefore ought not to be filtered through constitutional principles and constraints. In fact, Gaviria held that the imposition of constitutional constraints can be detrimental for good governance—especially when they impede the governance of a good political leader. Referencing Alfonso López Pumarejo and his New Deal–type policy program in the 1930s, Revolución en Marcha, he argues that constitutional constraints cut López Pumarejo’s time in office short, and thereby undermined the success of his policies. The resulting polarization and political crisis that culminated in La Vio­ lencía, he argues, had its roots in the brittleness of constitutional constraints regarding re-election (2004, 103–107). By embracing public opinion as the only source of democratic legitimacy, Uribismo opens a pathway towards constitutional reform that is subject to procedural and formal controls only, but foregoes any substantial controls inherent in (neo)-constitutionalist paradigms of democratic legitimacy (2004, 35). It is the free people, he insists, who will develop mechanisms to “dictate laws, deliver justice, and direct the executive” to practically implement the General Will of the nation (2004, 33). For critics, equating the highest authority with the political decision of the people can have two ramifications: one banal and another consequential and dangerous. Uprimny argues that the estado de opinión, at best, confers that public opinion is foundationally important for democratic

The Deliberative Judge

179

governance, or, at worst, emulates O’Donnell’s characterization of a delegative democracy: “The maximalist idea [of the estado de opinión] is risky, because it means that the truly important control in a democracy is the vertical one, which can then be used to justify a breach of the horizontal controls”.8 As shown, key normative elements of Uribe’s Democratic Security program focus on the core contention implicit in the question of constitutional reform: what is legitimate political power rooted in popular sovereignty, and what legit­ imate constraints does it entail? Evidently, this normative framework provides strong arguments in favor of re-election. Equally evident is that it espouses pro­ cedural constraints only and denigrates substantial constitutional limits on popular sovereignty—be they implicit or explicit. In contrast to a liberal constitutional framework that functions on the basis of distrust toward organized political power (Warren 1999), Uribe and his allies advocated for trust in political leaders. The significance of the political theory behind Uribismo is connected to Uribe’s standing as a political leader in Colombia. As explained in the second chapter, Colombia’s polity has historically been marred by factionalism and regionalism. In such a context, successful political leadership entailed skillful brokerage between those factions: a task entirely devoid of charismatic appeal. Uribe was distinct, because he truly commanded popular support, and his sway over Congress was equally distinct from those successful brokers that had previously occupied the Presidential Palace in Bogotá. The President managed to control and dominate Congress with his popularity, and thereby submit various factions into submission to his leadership. His position above and outside of traditional politics not only provided himself with high popularity ratings, but also made wide sections of Congress dependent on him personally. Thus, there was a clear delegative chal­ lenge to democracy in Colombia under Uribe. This challenge was not merely normative, but, helped through the fragmentation encompassed the dominance over Congress. The Fragmentation and Co-optation of Congress Uribe’s political philosophy coupled with his extremely high and persistent perso­ nal popularity were foundational of delegative trends in Colombia’s democracy. Developments in the legislature reinforce this view: Uribe managed not only to dominate public opinion, but his popularity extended to a control over substantive majorities in Congress. It helped his rise to power that the traditionally stable twoparty system (Mainwaring and Scully 1995) was in the process of imploding (Gutiérrez Sanín 2007), creating its own crisis of representation (Mainwaring et. al. 2006) through an extreme inflation of the numbers of parties represented in Congress. This provided the opportunity to successfully paint himself as an out­ sider to traditional politics, and reap the rewards of a general distrust towards traditional politics in Colombia to control public politics in Colombia in the first place. Once in power, and throughout both terms, he maintained a stable majority in Congress. Thus, the evidence suggests that there was not a sufficient political opening—a substantive decline of the president’s popularity or a misalignment

180 The Deliberative Judge between the President and his coalition in Congress—that explains the Constitu­ tional Court’s decision and assertion of authority. Foundational for Uribe’s high popularity was his success in stabilizing the security situation in Colombia. When he took office in 2002, almost 15% of the country had no police presence (158 of 1,098 municipalities). This number was reduced to zero by 2004.9 The intentional homicide rate curved downwards from 70 to 48 per 100 000 in the same period, and the numbers of kidnappings decreased almost by the same rate that Colombia had witnessed them rise between 1996 and 2000 (Appendix C, Graph C.2). The number of terrorist attacks and massacres also declined significantly. In short, Colombia moved from the brink of becoming a failed state to a state with significant but not insurmountable security issues (see Appendix C). This resulted in a general sense that the situation was indeed improving. A taxi driver told me a common sentiment, namely that people in Bogotá (and other large urban hubs) were enthralled by Uribe’s Democratic Security, because they could securely travel to their fincas and holiday homes on the coast. A Constitutional Court judge explained the connection of the Pre­ sident’s popularity to national security with the “myopia of the guerrilla, the FARC. … Every time, they employed terror tactics against the civilian population, the indicators of the president rose. The guerrilla had no popular support, they were the President’s campaign managers, metaphorically speaking.”10 Irrespective of the alleged upper-class vanity, the clear security improvement did produce consistently high popularity ratings that oscillated between the high 60s and mid­ 80s percentile range (Graph 4.1). Approval ratings reached their highest point in the second term after the liberation of Ingrid Betancourt (85%). The fact that Uribe’s popularity never took a substantial dent from the numer­ ous scandals that befell his government during the second term reached unprece­ dented heights in that term, and again rose substantially in the last months of his presidency—precisely when the Constitutional Court was deliberating constitu­ tional reform—makes readily evident that there was no lack of personal popularity that a strategic court could exploit to assert its authority. Such a political opening simply did not materialize. In addition, Constitutional Court judges were fully aware of the popularity of the President when they deliberated constitutional reform. A Constitutional Court judge said that “the President had enormous popularity ratings. In the second instance [of constitutional re-election reform], he was still very popular, more than 80 percent. That is unusual, because usually the high popularity wears off over time.”11 Another Constitutional Court judge said that “in all of Latin America, the executive is powerful, but in the case of Uribe, this is accentuated by popular power”. The consequence was that “he absorbed or annulled all internal political struggle”, creating a “presidential system, dominated by the President, without opposition in the Congress of the Republic”.12 There is a reason for the suffocat­ ing clout Uribe had over Congress during his time in office. His ascendance to power coincided with the open implosion of the traditional Colombian party system, from one of the most stably institutionalized two-party systems into one of the most fragmented party systems in the Americas, consisting of a high number

The Deliberative Judge

181

of electoral entrepreneurs that organized their electoral victories outside of the traditional party structures. With the help of alliances with regional powerbrokers, he could develop a network that propelled him from fame in the region of Anti­ oquia (but relative national obscurity) to national power in the nation’s capital, Bogotá. It enabled Uribe to run as an independent candidate, despite having his political home with the Liberal Party, whose selection of Horacio Serpa for the candidacy in 2002 caused the schism with Uribe in the first place (Bejarano 2013). It was thus largely circumstantial coincidence that he could portray himself as an outsider to traditional politics. The reconfiguration of legislative politics during Uribe’s time in office was a consequence of the unintended effects of electoral rule changes in the new Con­ stitution. One the one hand, new rules made Congress much more diverse, but, on the other hand, also more fragmented, and, as a consequence of that, also more easily co-opted by illegal forces (Gutiérrez Sanín 2007). As we saw in Chapter 2, before 1991, Colombia exhibited a long history of bipartisan con­ socionalism in Colombia. Yet, in 2002, the National Electoral Council (Consejo Nacional Electoral, CNE) officially recognized 72 parties, of which 45 had repre­ sentation in Congress (Pachón and Hoskin 2011; Pizarro Leongómez 2002). It had become one of the most fragmented party systems in the Americas. Con­ stituents in 1991 hoped to generate more political participation by lowering requirements, but due to its denationalization produced a party system that con­ sisted of many little micro electoral entrepreneurs with few coherent and identifi­ able national policy plans for voters (Pizarro Leongómez 2002; see also Crisp and Ingall 2002). These electoral entrepreneurs were an easily co-opted entry point for illegal forces to influence national politics (Gutiérrez Sanín 2019).13 The advantage for Uribe was that, once in office, he could use his popularity and the prerogatives of the presidential office to incentivize (with pork barrel) and discipline a steadfast coalition out of the many different and particularized political movements represented in Congress. One Member of Congress told me that “he was the general of the entire country, not just this circle of power”.14 Interest­ ingly, the same Senator said of the subsequent Santos government that “the gov­ ernment has the capacity to buy parliamentarians, because our legislators are dependent on caudillo votes” by political brokers in the regions that are con­ ventionally elected as mayors or governors. “The government, too, needs them, because they organize the votes”. In addition, they have the capacity to put public “money in projects there to keep this machinery going. So this means that this parliamentarian [in the coalition of President Santos] has to accept all that comes from the President.”15 There is no plausible reason to believe that the founda­ tional political exchange between national and regional powerbrokers was funda­ mentally different during Uribe’s time in office. In fact, Mejía Guinand et al. (2008) show that infrastructure projects in Uribe’s first term in office were “paved with votes” and that Uribe made liberal use of “spending on road infrastructure as a populist measure” to ascertain support from regional power-brokers (15). Beyond utilizing the pork barrel carrot, Uribe was also capable of wielding the stick to enforce compliance from Congress. After being elected in 2002, Uribe

182 The Deliberative Judge planned to accumulate more power in the presidential office through a refer­ endum, effectively threatening to curtail legislative powers. The Constitutional Court foiled the referendum (see below), but the threat achieved its apparent goal of ensuring a deferential Congress. He could handedly coalesce a broad majority behind him that “approved most of his policy initiatives” (Pachón and Hoskin 2011) and Congress, in both instances of constitutional reform of presidential term limits, pre-emptively initiated the reform for Uribe to modestly accept the demand by the people to extend his time in power. When we look at the specifics of the Uribe coalitions in Congress, we see that the President was capable of assembling coalitions with extensive majorities (the % column in Table 4.2 shows the percentage of each chamber in Congress belong­ ing to the Uribe coalition). As Table 4.2 shows, his governing majority even increased in 2006—the political reform to re-centralize the national vote had no effect on Uribe’s ability to unite a large sector of Congress behind him. Only towards the end of his second administration did the governing coalition lose a partner in Vargas Lleras’ Cambio Radical (Radical Change). Carroll and Pachón relate the loss of Cambio Radical to the efforts to generate another re-election reform during the second term. Another explanation for Vargas Lleras’ break with Uribe could also have been connected with parapolítica. Vargas Lleras survived a bomb attack against him that he attributed to obscure forces with relations to Members of Congress, while Uribe blamed the FARC for the attack. The mas­ termind was never caught, but it is suspicious that relations between the two soured soon after that incident.

Table 4.2 Uribe’s majorities in Congress, 2002–2010 Year

Uribe I

Uribe II

House

Senate

Parties in the Coalition

Seats

%

Seats

%

2002–2005

108

65.10

60

2005–2006

100

62.61

69

58.80 Conservative Party, Cambio 66.97 Radical, Liberal faction, Alas Equipo Colombia, Convergencia Ciudadana, Colombia Democrática, Colombia Viva.

2006–2007

103

62.43

68

2007–2008

118

63.43

83

2008–2009

92

51.96

70

2009–2010

91

52.63

59

Source: Carroll and Pachón (2016)

68.80 Conservative Party, Cambio 70.93 Radical, U Party, Alas Equipo Colombia, Convergencia Ciudadana, Colombia Democrática, Colombia Viva. 57.38 Conservative Party, U Party, 55.14 Alas Equipo Colombia, Convergencia Ciudadana, Colombia Democrática, Colombia Viva.

The Deliberative Judge

183

The estrangement between Vargas Lleras and Uribe is the only potential evidence that could support the hypothesis that a decreased alignment between executive and legislature provided a strategic opening for the Constitutional Court to exploit and prohibit another term in office in 2010. However, the effect of this fracture was not significant enough to substantively alter the support re-election received in the vote calls on the respective laws to initiate constitutional reform—Law 02 of 2004 and Law 1354 of 2009 (Table 4.3). In fact, because the opposition on the left abstained from voting in 2009, Uribe’s coalition consolidated a higher percentage in 2009 (63 of 95 votes, or 66%) in the vote in the House of Representatives than in 2004 (95 of 166 votes, or 57%), and only a slightly less overwhelming majority in the Senate. Table 4.3 Votes in Congress for Legislative Act 02 of 2004 and Law 1354 of 2009

Law 02 of 2004

Chamber of Representatives

Senate

Legislative Session

Legislative Session

Affirmative Votes

Primary Commission (19)

14

Primary Commis­ sion (35)

Law 1354 of 2009

Affirmative Votes 25

Plenum (166)

95

Plenum (102)

68

2nd Commission (35)

29

2nd Commission (19)

13

2nd Plenum (102)

67

Primary Commission (19)

12

2nd Plenum (166)

115

Primary Commis­ sion (35)

28

Plenum (95)

63

Plenum (102)

59

Conciliation

84

Conciliation

53

Source: Constitutional Court of Colombia; CongresoVisible.org

The evidence presented here aims to show two things: 1) Uribe’s personal popularity and institutional support in Congress remained very high throughout his two terms in office; and 2) the President managed to utilize his high popularity ratings to dominate Congress in a similar way to how he dominated public opi­ nion. Akin to the natural leader in his Democratic Security framework, he demanded (and largely received) trust and obedience from his coalition in Congress. Therefore, if under the ideal democratic scenario, the legislature exercises effective (horizontal) control of the executive, Uribe’s hegemony largely obfuscated this function. In light of this evidence, it is not plausible to argue that the Court faced a strategic opening when it decided against re-election in 2010. Rather, legislative politics in the Uribe years reinforce the notion that there was a visible delegative trend in Colombia’s democracy that pushed aside horizontal institutions and pro­ pelled the executive’s position in the institutional matrix. Uribe’s propensity to dominate legislative politics in Colombia did not really recede in the second term. His popularity rose sharply (from already high grounds) in the last three months of his presidency as the reform campaign gained momentum (see Graph 4.1 above).

184 The Deliberative Judge This surge in the polls was properly put to use to maneuver the project through the institutions and place potential objectors under pressure. In effect, this was also the view amongst judges on the Constitutional Court who said that magistrates are not “little kids” and know that constitutional arbitration always involves high stakes: “2005 was no different from 2009”.16 In addition, given that during the para­ política scandal judges were physically threatened, rooms in the judicial palace wir­ etapped, and judges of the high courts subjected to public smear campaigns, the sense of insecurity probably was much more direct than calculations showing a potential political opening could outweigh. Thus, we are left with one crucial hypothesis that identifies the Constitutional Court as a strategic actor. As Rodríguez-Raga impressively shows the Court had built its jurisprudence carefully by restraining itself in prominent cases, but upholding the principle, and then applying the principle in less prominent cases. According to his analysis, it is this strategic behavior that made the Colombian Constitutional Court a prudent strategizer. As said, this analysis stands on data that went until 2005, and therefore predates key decisions of the substitution doctrine’s jurisprudence. To test this hypothesis, we need to investigate whether the substitution doctrine, central to rejecting Uribe’s final constitutional reform, grew strategically or not. The next sec­ tion will address this caveat and show that the origin of the substitution doctrine goes back to the very beginning of Uribe’s time in office and was a direct response to his sense of mission. We will see that the thickening of that jurisprudence did not follow strategic action, but imperatives based in the ethics of deliberation. The Making of the Substitution Doctrine: Neo-Constitutionalism and the Implementation of Ethical Reasoning in the Constitutional Court The reconstruction of the creation and evolution of the substitution doctrine has four important insights in store. First is the substance of the doctrine itself, which filled an important gap in constitutional theory of how to legally regulate constitutional repla­ cement (see also Dixon and Landau 2015; Bernal 2013; Roznai 2013). Second is the fact that the need to fill that gap and create a doctrine to understand constitutional replacement in legal terms arose because of Uribe’s will to transform the constitutional framework through populist and plebiscitary appeals. Third, there is little evidence that the Court acted as a prudent strategizer that carefully builds its jurisprudence with consideration to other institutions and powerful actors. The jurisprudence made the most important steps in cases that were far from inconsequential to the administration’s agenda. It was exactly the power and influence Uribe wielded in Colombian politics that urged judges to act rather than carefully tread a line and remain cautious. Fourth, at a crucial point during the deliberation of the re-election reform, a majority of judges inside the court insisted on applying the substitution test. It was this moment that resulted in declaring the reform unconstitutional not merely on procedural grounds, but on quasi-substantive grounds, highlighting the necessity to fully appreciate the importance of the implementation of deliberation inside constitutional chambers. These four points together lead me to understand the constitutional judge in Colombia as a discursive actor rather than a strategic one.

The Deliberative Judge

185

Given the communitarian invocation of Uribe’s political philosophy, constitutional reform to suit his interest as the natural leader of the Colombian nation is not a sur­ prising outcome. In fact, from the very beginning of his first term, constitutional reform was placed high on the agenda. Not least, since his first minister of the com­ bined Justice and Interior Ministry, Fernando Londoño Hoyos, laid out a clear con­ stitutional reform agenda, arguing the country had become ungovernable under the 1991 Constitution (2001). Uribe followed through and passed several amendments to the Constitution. These ranged from more technical alterations such as the elec­ toral circumscription of the Chamber of Representatives (Legislative Acts 2 and 3 of 2005), the administration of special municipal districts (Legislative Acts 2 of 2007), the District Council of Bogotá (Legislative Act 3 of 2007), to others that directly modified norms concerning issues of fundamental rights, strategic public services, and the incorporation of international norms into national law. Examples are the intro­ duction of the accusatory system in criminal proceedings (Legislative Act 3 of 2002), the alteration of the political regime of parties and elections (Legislative Act 1 2003, Legislative Act 1 of 2004, Legislative Act 1 of 2007, and Legislative Act 1 of 2009), the anti-terrorist statutes (Legislative Act 2 of 2003), and the regulatory system of pensions (Legislative Act of 2005). The table below (Table 4.4) lists all cases before the Constitutional Court until 2010 that invoked the substitution doctrine. They fall under the Court’s jur­ isprudence regarding the norms of “Title XIII” concerning “Constitutional Amendments” (Article 374–379). The first column lists the decision, the second the theoretical focus, the third explains the application of the doctrine, and the last column whether the Court detected formal or substantive defects. A first observation that runs counter to the prudent strategizer argument is that the emergence of the substitution doctrine in the Court’s jurisprudence solely coincided with Uribe’s time in office, even though his predecessor also reformed the Constitution. While the first two Presidents under the new Constitution, Gaviria and Samper, did not introduce any constitutional amendments, Pastrana passed seven reforms. Uribe submitted a total of 15 pieces of legislation that aimed to alter the Constitution (Valencia Villa 2012, 15–18). The subject of constitutional reform therefore had already been a practical and not merely theo­ retical issue. A Constitutional Court judge posited that the “constitution needed to be protected against a very big threat, the government of President Uribe, that wanted to impose 17 reforms in one strike without a debate”.17 Prior to Uribe, as an auxiliary judge of the Constitutional Court confirmed, “they only looked at procedure”, but beginning with his first constitutional reform in 2003, the court “began to address the question of the substitution of the constitution”.18 It was not lost on the Constitutional Court, either, that the appointment of Fernando Londoño as the Minister of Justice and Interior in 2002, put someone “very close to the Supreme Court and an enemy of the Constitutional Court” in charge of the judiciary.19 Londoño had written an essay before becoming minister (2001) arguing that the country was ungovernable with this Constitution and in need of reform. It is therefore fairly reasonable to assume that the theory of competence was indeed a direct response to Uribe’s election, his high popularity, control of the

186 The Deliberative Judge Table 4.4 Constitutional Court decisions between 2003 and 2010 and the substitution doctrine Decision

Theoretical Focus

Application

Constitutionality

C-551/03 (“Political Reform”)

Develops theory of competence

Development of theory, but no application to merits of the case

No defects

C-1200/03

Citing doctrine

No defects

C-816/04 (AntiTerrorism Statute)

Citing and application of doctrine

Unconstitutional due to formal defects

C-970/04 (Changes to Code of Penal Procedure)

Citing and application of doctrine

No competence defects for implementing accusatory system

C-971/04

Citing and application of doctrine

No competence defects

Citing and applica­ tion of doctrine

Re-election is granted, but delegation of legislative functions to state council is unconstitutional on com­ petence grounds

Citing doctrine

No defects

C-1040/05 (First Re-election Reform)

Develops substitution test

C-153/07 (Pension Reforms) C-180/07

Citing doctrine

No defects

C-588/09 (Administrative Career)

Citing and application of doctrine

Partial declaration of unconstitutionality on competence grounds

Citing and application of doctrine

Unconstitutionality on competence grounds

C-141/10 (Second Re-election Reform)

Separation of powers is an axiomatic principle

Source: Compiled by the author from Constitutional Court of Colombia

agenda in Congress, and willingness to transform constitutional foundations.20 Or, as a Constitutional Court judge put it, “the control of competence has to be viewed in the context of excessive presidentialism, quasi caudillismo, in the person of Álvaro Uribe”.21 The jurisprudential reaction of the Constitutional Court was distinct, because the political leadership of Álvaro Uribe was also distinct in the Colombian context: “there was no president until President Uribe that wanted to utilize his popularity to change the constitution so radically as he tried 2003”.22 In short, and as an auxiliary judge to the Constitutional Court stated bluntly, “it was clear that the [substitution doctrine] was a counter Uribe doctrine. Uribe had an ample majority in Congress and was very popular. The argument was that large majorities are dangerous for democracy and the Constitutional Court had a counter-majoritarian role.”23

The Deliberative Judge

187

The core contention against the prudent strategizer argument arises from the fact that the substitution doctrine’s most important jurisprudential developments were made in salient cases to the executive: C-551/03, when the contention of compe­ tence was first articulated in the review of President Uribe’s first major reform project passed under the eponymous title “reforma política” (Political Reform); C-1040/05, when the Court first struck down a portion of a law based on competence grounds in Uribe’s first re-election reform; and C-141/10, when the Court struck down a law in its entirety based on competence grounds in Uribe’s second re-election reform. The reconstruction of the argumentation in each case further shows the deliberative approach of the court, and how it weighed constituents’ intent with legal context at specific moments. All of that suggests that judicial activism was not driven by strategic prudence, but by deliberative intent. Uribe’s “Political Reform” involved three core projects that addressed electoral rules and the creation of prohibitions to battle corruption, make budgetary debates more public, and overrule several Constitutional Court decisions including the legalization of the consumption of narcotics for private use (Cepeda Espinosa 2012, 70). The law intended to place the reform before the people in a refer­ endum. As any constitutional reform submitted to a plebiscitary vote, this reform project had to undergo an all-encompassing constitutionality review by the Con­ stitutional Court. The Court affirmed the conditional constitutionality in a 6:2 decision, but the specifics of the case compelled the Constitutional Court to explore the question of the constitutionality of constitutional amendments them­ selves. It is in the details of this reasoned decision that the Court introduces crucial parameters for expanding the review power of the Constitutional Court (Cepeda Espinosa 2012). The driving force behind the first articulation of the doctrine was magistrate Eduardo Montealegre.24 He had already publicly had asserted that reforming the Constitution is not the same as replacing it in the debate on the tutela contra sentencias in the early 2000s. The substitution doctrine hinges on this distinction between reforming and replacing the Constitution, situated at the nexus between material and procedural review of constitutional amendments. The motivation arose because Uribe’s “proposed [political reform] involved the structure” of the constitutional order. The content of a constitutional reform was not subject to review, because, as Montealegre explained, any reform naturally alters the sub­ stantive meaning of the established political charter. While the Constitution lim­ ited the Court to procedural review in such cases, it was also clear that the three mechanisms to reform the Constitution (legislative act, referendum, or constituent assembly) “are meant to imply limits so that these big treaties cannot be changed in a moment of Odyssean madness”.25 Another auxiliary judge explained that such limits became imperative to thwart the government’s ability to invoke a “theory of the allusion of the constitution to simply change the constitutional reason with the help of Congress, when the Constitutional Court had declared laws unconstitu­ tional for that constitutional reason”.26 The inventive turn in the argumentation of the Court comes when it argued that substance and procedure overlap in the competence of the actor attempting to

188 The Deliberative Judge create a legal consequence. To safeguard against the sirens’ song of habitual con­ stitutional reform, Montealegre and his team of auxiliary judges, “proposed that the power to reform has limits, not only of procedure, but also of competence: com­ petence is a presupposition of procedure”.27 As Colon-Rios explains, competence, or the “legal power of producing a determinate legal consequence”, is a central component of procedure in constitutional reform; “it would be legally meaningless to follow each of the requirements of a determinate procedure if one lacks the competence to produce the desired outcome” (Colón-Ríos 2011, 374; see also Colón-Ríos 2012). Therefore, the Court had to investigate whether the actor implementing the constitutional reform, namely the legislature, was acting ultra vires or within its constitutionally bestowed powers to reform the Constitution. The Court’s deconstruction of procedure involved a second step. To be applicable and have practical relevance, the theory, in addition to bestowing competence defects of constitutional reforms under the Court’s mandate to iden­ tify “procedural defects”, also has to substantiate that reforms themselves face limitations. Again, without any limits contingent on the type of changes a reform is allowed to produce, the power to review the competence of the actors initiating that reform becomes meaningless. It is here where the relation between procedural review of competence becomes most disputatious, because some argued that “it involved substantial limits of the power to reform, not [prescribed] in the original constitu­ tion”.28 Advocates in the Court for limits on effortless constitutional reform29 of foundational principles contended that the “dichotomy between procedure and sub­ stance is a formal dichotomy”. The new Constitution, however, gave “right (derecho) more space” precisely to overcome the “formalism, which conceived norms as autonomous of society.” As such, exploring implicit limits to specific paths of con­ stitutional reform is an example of “nuevo derecho [that] is neither procedure nor substance, because it is not formalistic”.30 Indeed, we will see that the dissolution of this dichotomy between procedure and substance through competence also serves to incorporate legal context into the evaluation of the actor’s competence itself.31 Controversially, the substitution doctrine “certainly involved stronger limits” that arise from the distinction between constituent (poder constituyente) and con­ stituted power (constituente segundario). After all, it essentially contended that “competence cannot be absolute, some basic pillars can only be modified through the constituent and not the legislator”.32 A Constitutional Court judge—even a critic of the substitution doctrine—explained that the basic paradox is that in constitutional reforms “the people can pronounce itself and expedite norms of constitutional rank, [but this] people is not the same as the sovereign people; it is not the constituent”.33 Perhaps somewhat counterintuitively, the Court here agreed with Carl Schmitt’s Verfassungslehre that a constituted body never attains the constituent power of the sovereign people, but always remains inside the sphere of regulated power (Colón-Ríos 2011, 375–376). The innovative twist, and clear departure from Schmitt, is that, as much as this entails the absolute freedom of the sovereign people, it also contains the limitations placed upon amendment procedures that aim at transforming the valid Constitution:

The Deliberative Judge

189

In the development of democratic principles and of popular sovereignty, the constituent power lies with the people, who have and preserve the power to give themselves a constitution. The original constituent power, then, is not subjected to legal limits and implies, above all, the complete exercise of the political power by the relevant individuals. […] On the other hand, the power of reform, or derivative constituent power, refers to the capacity certain organs of the state have, on some occasions by consulting the citizens, to modify one existing constitution, but within the paths determined by the [current] con­ stitution itself. This implies that it is a power established by the constitution, and that is exercised under the conditions set by the same constitution. Such conditions comprise matters of competence, procedure, etc. It deals, therefore, with the power of a reform of the constitution itself and, in that sense, it is of constituent nature, but it is instituted by the existing constitution and it is therefore, derivative and limited. (C-551/03, cited in Cepeda Espinosa 2012, 71) The Court argued that this distinction between constituent and constituted power is indispensable for the application of Article 374 of the Constitution (the Con­ stitution may be “amended by Congress, constituent assembly, or by the people through a referendum”), because “a power of reform without limitations of com­ petence also eliminates the basic distinction between the original constituent power and the derived constituent power” (C-551/03, cited in Cepeda Espinosa 2012, 72). Between 2003 and 2005 the Court solely reiterated the parameters of the doctrine as part of the Court’s jurisprudence. Au contraire to the prudent strate­ gizer prediction, the first time to apply its parameters occurred in the review of the constitutional reform to extend presidential term limits in 2005—by far the most consequential reform for Uribe’s personal and the executive’s institutional power. The reform contained four different articles: 1) legislating the provisions for the incumbent to level the playing field; 2) altering Article 197 of the Constitution to the wording that any individual cannot be president for more than two consecutive terms; 3) specifying the role of the vice-president; and 4) guaranteeing the rights of the opposition in a statutory law. Article 4 included a transitory clause, that stated: [i]f Congress were not able to issue the statutory law within the deadline established, or it were declared void by the Constitutional Court, the Council of State will provisionally issue regulations on the matter during a two-month period. (cited in C-1040/05)34 The Court neither considered alleged procedural defects sufficient to declare the law unconstitutional, nor did a second consecutive term in the presidential office in itself amount to a substitution of the Constitution. However, it did find that the statutory clauses included in Article 4 of the legislative act delegated legislative powers to a

190 The Deliberative Judge judicial body, which substituted the principle of a separation of powers into execu­ tive, legislative, and judicial branch of government. The jurisprudential advancement in this conclusion consisted in the specification of a “substitution test”. Previously, the Court had simply manifested that the Court must analyze the principles and values of the Constitution. In 2005, it defined a substitution of the Constitution as replacing “one of its defining elements by an opposing or wholly different one. Substituting implies that the resulting text contradicts the core of the 1991 Con­ stitution, which therefore is no longer recognizable” (C-1040/05). Defining ele­ ments of the Constitution are those elements that pass a three-tiered test: 1) identifying the core element; 2) referencing multiple constitutional provisions to define its specificity within the 1991 Constitution; and 3) showing its importance in the comprehensive constitutional context. From here the judge can then move on to comparing the original element with the new element to estimate whether the new element is different to the degree of incompatibility with the original document (Cepeda Espinosa 2012, 76). The provision to delegate legislative powers to the administrative court fulfilled the parameters of that test and was therefore declared unconstitutional on competence grounds. Between 2005 and 2010, the Court invoked the doctrine three more times. In C-588/09, the Court went the furthest by, similarly to 2005, declaring part of a law unconstitutional. The next crucial jurisprudential step was the 2010 re-election decision; evidently the most consequential reform legislation for Uribe’s second administration. In 2010, the Court investigated five procedural complaints coupled with competence defects. The latter, as we will see below, was not the result of specific citizens’ complaints, but arose from deliberation inside the court itself: “it was interesting in the second re-election, because it was a close vote (5:4) and the ponente dissented from that opinion. So it was difficult, but these five accepted the precedent and built more arguments on top of that”.35 The implications for my argument are two-fold: 1) the importance of that decision signals the inapplic­ ability of the prudent strategizer argument for the substitution doctrine; and 2) the specifics of its application in 2010 evidences the deliberative nature of that legal decision-making process. The Court first had to answer questions regarding its own competence for reform projects that originate outside the legislature and the executive. The Court affirmed its competence, arguing that its jurisdictional reach covered all procedures leading to a constitutional reform that pass through Congress, even when they originate outside the formal institutions and are plebiscitary in nature. After all, Congress still had to validate referenda as well. It then moved to address the five complaints litigators had submitted to the Court in the litigation period. When a referendum for constitutional reform is initiated, its committee of promoters and supporters must register with the National Register of the Civil State (Regis­ traduría Nacional del Estado Civil) and receive the support of 5% of eligible voters (C-141/10; 11). The promoters raised the funds necessary, but did so with little respect for rules governing such campaigns. As Botero et al. write, “the referendum records [were] murky and plagued with irregularities, such as selfloans between organizers that deliberately attempted to obscure the way in which

The Deliberative Judge

191

the referendum was financed. Promoters spent six times more than the spending cap permitted”, with contributions by individual donors sometimes thirty times as high as allowed (Botero et al. 2010). In addition, Congress violated the require­ ment to have the votes of support confirmed by the National Registrar prior to vote on the bill and went ahead without approval. The next violation concerned the wording of the referendum. The text placed before the electorate actually read that Uribe would run again in 2014. Promoters of the law assuaged this when the bill was already in Congress. The Court argued that Congress extra-limited itself by changing the wording late in the process (Botero et al. 2010). In addition to material violations of the procedures governing amendment processes of the Colombian Constitutions, the Court also found formal defects of procedure. The majority backing reform in Congress had five congresistas amongst their ranks, who had joined Uribe’s party from another party (Cambio Radical). However, the laws against transfuguismo disallowed such maneuvering between party caucuses and Cambio Radical had sanctioned them, essentially invalidating their votes. Without their votes, the reform coalition did not have the required majority. Finally, the bill was not passed within the time limit of the ordinary session that expired 16 December 2008 at 24:00. At that time, Uribe’s coalition had called in an extra-ordinary session that began at 00:05 of 17 December. However, extra­ ordinary sessions of Congress must be published in the Diario Oficial a day in advance, meaning that the extra-ordinary session was invalid (Botero et al. 2010). The ponente of that decision, Humberto Sierra Porto, was a critic of the sub­ stitution doctrine, contending that the substitution test evades any objective para­ meters to “identify the essential elements of the 1991 Constitution” (C-1040/05; 771). His opinion therefore did not apply the substitution doctrine. In the end, seven of nine judges (the two defecting ones were listed by Uribe himself) agreed to declare the law for a referendum unconstitutional for procedural defects. Nevertheless, five judges insisted on the application of the competence test. After further deliberation the majority concluded that the separation of powers is an axiomatic principle of the social state of law. It is inscribed in the first Article of the Constitution as well as in the democratic principle of the Colombian Con­ stitution (Article 137). In addition, it specified that the separation of powers in the 1991 Constitution was institutionalized by what the Court called the “period­ ization of offices”. This entailed that the terms of one office must not align with the periods of offices with horizontal control functions. For example, the Federal Bank was supposed to be independent from political influence. Yet, one re-elec­ tion already enabled presidents to pack the Board of Executives of the Bank; another third presidential term, the majority argued, would make the envisioned independence a travesty. Therefore, those five magistrates, giving credit to the conditions of presidentialism in Latin America, argued that another term diluted these checks and balances (pesos y contrapesos) to the degree that they were no longer recognizable, thereby not only modifying an axiomatic principle of the Constitution, but substituting it for another. For that, Congress and President lacked the competence, because only a constituent assembly wields original con­ stituent power legitimating it to draft reforms of such magnitude.

192 The Deliberative Judge So far, the reconstruction of the jurisprudential genesis and solidifying of the sub­ stitution doctrine showed that the normative trajectory of the theory didn’t merely coincide with Uribe’s time in office; it was a direct response to his sense of mission. The lockstep advancements of key elements of the theory in constitutionality deci­ sions of great importance to the executive, and Uribe himself, corroborates this impression. The final contention that shows the inadequacy of the strategic model to explain this particular outcome in the Colombian case comes courtesy of political events during Uribe’s second term and how they formed the legal context. As docu­ mented, judges insisted the Constitution was “a living document”36 and “the magistrate a mediator between the reality and rights [enshrined in the constitution] so that they apply these principles”.37 Therefore, it is not surprising that nefarious political practices in Congress and the treatment of the high courts crucially informed the normative weighing during deliberation of whether a third consecutive term replaced the separation of powers principle in the Constitution. My interviews with judges and auxiliary judges on the Constitutional Court affirmed three sets of arguments that are rooted in the legal context: 1) confirming the link between Uribe and the advent of the substitution doctrine, they diag­ nosed expansive de-facto presidential powers arising from Uribe’s willingness to utilize his popularity; 2) worrying about the Congress’s docility in face of Uribe’s popular power, they detected an ineptitude to function as a check (contrapeso) on executive power; and 3) differentiating between the quantity and quality of poli­ tical scandals in Uribe’s second term, they noted signs of institutional decay as well as proceedings that constituted gross institutional violations of the Constitution in their own right (e.g. bugging the halls of the Supreme Court equaled a violation of the judiciary’s independence). For the five-count majority in 2010, these argu­ ments tipped the scale in favor of the precedent that limited presidential terms to one re-election over the claim that the popular will ought to choose the number of presidential terms without constitutional/legal constraints. Even though my interviews were structured relatively openly, the question to explain the tension implicit in constitutional democracy (between democracy and constitutionalism) led five magistrates (all from the Constitutional Court) to posit that Uribe had accumulated vast de-facto power in addition to the broad constitu­ tional powers attributed legally to the executive. As already cited, some judges from the Constitutional Court alluded to images of a caudillo. In addition, seven judges noted that Congress appeared to be entirely ineffective in applying its control func­ tion. The reflection on the effects of a quasi-caudillo on the constitutional order deserve to be quoted in full, as it best illuminates the normative connotations explored in the legal decision-making process of the Court: So the Court, faced with the invasion of a caudillo, and the form of this cau­ dillo increased his power, as a constituted power, because this power was inoculated with popular power. In all of Latin America, the executive is pow­ erful, but when Uribe first entered power, he absorbed or annulled all internal political struggle: As a president and the dominus of the Political (“de lo poli­ tico”), he has direct power. So, the presidential system in such conditions,

The Deliberative Judge

193

dominated by the president without opposition in the Congress of the Republic, with the president as a media person that is fully engaged in the communicative universe, he does not face a distinct power other than the Constitutional Court. In Consequence, the Constitutional Court saw in these conditions that the possibility of a second re-election would effectively end the system of the separation of powers and compliment the presidential system in Colombia into something more monarchical.38 Another judge of the Constitutional Court, even a critic of the substitution doc­ trine, confirmed that the Colombian system “is designed to strengthen the Pre­ sident”, and added that “in our system, presidentialism does not only mean that the president has more competence than is formally written; there is a sociological or cultural element through which the President accumulates more powers to delegitimize the other powers”. He concluded that Colombia had reached an “extreme in this presidentialism and the principal victim of presidentialism in Colombia has been Congress”.39 Thus, the empirical conditions of presidentialism in Colombia countenanced the Court’s ambitious intention to control constitutional reform, even for positivists in the mold of viejo derecho in Colombia. The criticism of Congress’s role in Colombia’s constitutional order at the time of the second re-election decision reveals the inventive normative turn the jur­ isprudence took in the specification of the substitution doctrine. As the aforemen­ tioned critic of material review contended, the substitution doctrine is studded with characteristic elements of Latin American constitutionalism: “The written text in the constitutionalism of Colombia can be identical to that of Germany. However, because the social reality is very different, the same written texts have very different [constitutional] meanings”. This differing meaning arises from the context, which in Colombia and Latin America is decisively marked by two things: 1) profound social inequality; and 2) the presidentialism in the region. The same judge illustrated the implication of the later for constitutional review: “If I believe that Congress is cen­ tral for a democratic system and the expression of a democratic will, the only limits that it can have are those formal limits, but if I think if it is not legitimate, I evi­ dently want to put limits to what it can perform”. Damaging to the fortunes of the re-election reform was the fact that those institutions, namely “Congress and Municipal Councils, where society was represented, [were] seriously debilitated”. So there was a “perception of low trust in the capability of the Congress”. To be sure, one reason for this debilitation was the particular Colombian presidentialism under Uribe; “but it [was] not the only reason. The other is the electoral system. Parti­ cularly, the deficiencies that result in a distortion of the representative institutions. What causes these distortions: essentially clientelism and corruption.”40 Affirmingly, another judge stated “[the re-election] decision was a very political decision by the Court. It is not derived from the precedents, not from the text of the constitution, it is simply a form to preserve the Constitution as the norm of norms. As the insti­ tutional architecture.”41 In sum, the inclusion of the legal context as a legitimate argumentative point of constitutional review made the application of the substitu­ tion doctrine under the real-life conditions of Uribe’s presidentialism plausible for its

194 The Deliberative Judge critics—and imperative for those that identified the defects in the legal context as foundational and systematic. It was the nature of the scandals befalling Colombia’s political system, and revealed in Uribe’s second term, that convinced those judges voting with the majority to dis­ allow a second re-election of the institutional damage of Uribe’s presidentialism. Corruption and political scandals are evidently not a novel phenomenon in Colom­ bian politics. Yet, the abundance of scandals was not only noted by many judges (as well as their auxiliary personnel); six high court judges noted a qualitative difference in the type of scandals (and five of their auxiliary judges confirmed this). Interviewees noted some general corruption scandals, involving nepotism and kickbacks (inter­ views were conducted before the Odebrecht scandal). They were mindful of gross human rights scandals such as the false positives but were fairly unequivocal in that the most damaging was the wiretapping scandals orchestrated by the DAS: “I think the infiltration of DAS by paramilitarism, an act of co-optation, that is the gravest”.42 Another judge confirmed, “the wiretapping was the gravest infringement on the institutional fabric of the Court”.43 The chuzadas created a “high level of distrust, angst, and not merely pressure of deciding high-stakes cases, [because] it was not clear if they followed the dialogues of the magistrates of the high courts”. It is pre­ cisely these dialogues, deliberation in other words, that ensure that “the pro­ cess to take a decision is marked by autonomy”.44 In the end, and foreshadowing my argument that deliberative action better accounts for this outcome, judges and their support staff involved in deliberations (auxiliary judges), concluded that the wiretapping of the judicial palace, with the help of the state’s security apparatus (the DAS) as well as nefarious individuals from paramilitary groups, constituted a systemic violation of the democratic order, because it harmfully affected the core of judicial independence: uncoerced deliberation. The 2010 decision to reject another term in office was sui generis and constitutes a new case that cannot be explained with existing theories of judicial behavior. It is my argument that the strategic judge, specifically the notion that courts act like prudent strategizers when introducing consequential new doctrines, cannot explain the creation and development of the substitution doctrine and its application in 2010. There is simply little evidence in the data that suggests strategic interaction: I showed that there was no misalignment between the goals of the executive and governing coalition in Congress regarding the question of constitutional reform. The factionalism resulting from the Cambio Radical’s departure from Uribe’s coalition was meek at best, and not consequential for the legislative path of the constitutional reform in Congress. The raw numbers of the congressional coalitions during either Uribe term, as well as the legislative support for the legislations in the roll calls in Congress, confirm this impression. What is more, the evidence suggests that judges felt compelled to act precisely because of a usurping presidency. In line with that observation, it did not go unnoticed that the President managed to utilize his high popularity ratings to dominate Congress akin to how he dominated public opinion; nor did the institutional weakness of Congress remain unobserved. Neither observation bodes well with the predictions of the strategic model, as neither indicates a political opening: “The Court saw what was happening in the political sphere. Without a

The Deliberative Judge

195

contrasting power, this would result in a pathological growth” of executive power.45 The prudent strategizer argument describes a court that oscillates between defer­ ential action in high-stakes cases and principle-asserting decisions in less important cases. My analysis did not find evidence of such behavior in the creation of the sub­ stitution doctrine jurisprudence, either. First, its ex nihilio creation deterministically overlapped with Uribe’s ascendance to power: it was a direct and deliberate response to his sense of mission. This is not surprising given that he was perceived as a Bona­ partian Caudillo, rich in symbolism of Antioquian rancheros and traditionalist cam­ pesino culture. His embrace of an estado de opinión with qualities of constituent power and plebiscitary democratic governance was also perceived as a systemic, and novel, threat to the constitutional order. The consequence was not to tread carefully but act vigilantly: the most critical steps in creating the substitution doctrine (the review of the 2003 political reform), specifying its meaning (the 2004 presidential re­ election reform), and expanding its application (the second 2009 presidential re­ election reform), did not occur in unimportant cases, but, in fact, cases of high priority to the executive. It is by virtue of this trajectory that the jurisprudential development of the substitution doctrine contradicts the predictions of the prudent strategizer model. A Constitutional Court judge, the decisive one that swung the 2010 decision towards rejecting the reform in its entirety for competence defects, explained that the judges were well aware “that a decision against re-election could result in an institutional crisis, because this president had a lot of acceptance amongst the people. The people felt that he understood their needs.” At the same time, “our country has a high degree of institutionality, and you respect the institutions above everything”. Crucially: “We were mandated in protecting the Constitution. So we had to decline the reform, because it was substituting a central part of the constitu­ tion. Only the primary constituent can do that, which is the people.”46 In short, the application of the substitution doctrine was not a convenience or just legitimate, but in fact a necessity. La Ley Cuenta: The Institutionalization of Legal Reasoning in Colombia’s Constitutional Court and the Rejection of a Third Consecutive Presidential Term The thesis that the 2010 decision to reject constitutional reform for compe­ tence defects is reflective of deliberative rather than strategic action arises from that serendipitous moment during deliberations. First, (relieved) reactions by others on the Court to the (randomized) selection of experienced magistrate Humberto Sierra Porto to be the ponente presenting the initial study for deliberation in the plenary chamber indicate the awareness of the political stakes of the case—a sentiment that evidence can trace to the very end of the case when its decision was announced to the public. Second, the selection of Sierra Porto, a known legal positivist and therefore critic of the substitution doctrine, created an imminent choice for the deliberation to take: magistrates had the options to either solely explore formal defects in the implementation

196 The Deliberative Judge of reform, or competence defects of the actors applying the reform—avoid the Gretchen question between form and substance of constitutional reforms they could not. Third, the eventual choice for venturing into an analysis of compe­ tence defects was precipitated by a genuine change of opinion, motivated by the force of the argument. What is more, behind the insistence to extend deliberations unfolds a deeper structure of formal and informal rules, central to legal decision-making in the Constitutional Court. These observations com­ bined not only fulfill Hübner Mendes’ ontological foundation of deliberation— a collective decision-making process that is contingent on participants being open “to transform their preferences in the light of well-articulated arguments” (2013, p. 18)—the formal and informal rules of deliberation also aim to pre­ serve an uncoercive space of free deliberation. I argue that these elements, and not strategic prudence, were causal in the 2010 decision to apply the compe­ tence test of the substitution doctrine and reject the constitutional reform on the grounds that it substituted the constitution without the competence to do so, for vicios de competencia. To convey the argument that the deliberative rather than the strategic judge is essential for understanding the 2010 decision, it is useful to present the list of formal and informal rules governing procedures of deliberation and the interaction between magistrates. Table 4.5 below is the result of interviews held with magis­ trates and auxiliary magistrates of the Constitutional Court. 47 Table 4.5 Formal and informal rules of deliberation inside the Constitutional Court Formal Rules

Informal Rules

Annual election of President of the Court by magistrates

Termination of public hearings

Presidential right to force vote

Requirement of consensus to terminate deliberation

Selection of ponente by lot

Reduction of public interviews

Exclusion of third parties from deliberations except for Secretary General of the Constitutional Court

No discussion outside of Sala Plena

Decision by absolute majority

No quid pro quo negotiations

Source: Compiled by the author

Recall how Hübner Mendes’ imperatives of a deliberative constitutional judge move the normative substance to the procedure of deliberation rather than the attainment of a metaphysical notion of the good. It also becomes readily evident that deliberation in plenary chambers of constitutional courts relies on formal and informal rules in order pass the test of deliberative democracy. From the founda­ tion that deliberation implies practical reasoning in collective decision-making, combined with the contingency that participants change their preferences in light of better arguments, Hübner Mendes lists seven traits that I summarize in the table below (Table 4.6).

The Deliberative Judge

197

Table 4.6 Hübner Mendes’ theory of deliberative decision-making in constitutional courts Elements of Deliberative Decision-making in Constitutional Courts

1 2 3 4 5

Recognition for collective decisions on questions of authority. Temporality of decisions, because continuity is vital for legitimate politics. Deliberation combines reason giving as the grounds for a decision. Reasons are directed at the common good, attempting to conform to impartiality. Inter-subjective engagement of persuasion requires three conditions of its own: 1 2 3

6 7

the willingness to review their position; the presence of an ethics of conversation; and the absence of coercion.

Deliberation assumes the mutual recognition of equals. Deliberation reflects a morality of inclusiveness, empathy, and responsiveness to demonstrate fellowship in one political community.

Source: Hübner Mendes (2013)

The reconstruction of deliberation for that consequential decision in 2010 not only illustrates how those above-listed formal and informal rules of deliberation play out in a specific case. It also shows that deliberation inside Colombia’s Con­ stitutional Court emulated the normative elements of a deliberative democratic ethos in legal reasoning. The basic implementation of debate inside the Court follows a routine similar to a university seminar. First, a random selection from a pool of lottery balls decides the initial ponente (rapporteur) from the nine magistrates that occupy the Court. The elected ponente, together with a team of auxiliary judges, then prepares the initial, but not necessarily ultimate (see below), study (ponencia) into the juridical questions posed by a particular constitutional issue. It presents his/her take on the legal issues and is distributed to the other magistrates and their clerks, who then have time (usually around ten days) to prepare their own interpretation of the legal issue based on the arguments presented in the first ponencia. Everyone then meets well prepared for deliberation in the Sala Plena, where only the magistrates and the general secretary are present. Deliberations end when the President of the Court puts the ponencia to a vote in the Sala. The President has the right to conclude discussions, but interviewees explained that debates are never arbitrarily terminated, but only if the discussion has come to a natural ending. If the ponen­ cia does not reach the five-vote majority threshold, it is given to another magis­ trate, who introduces his/her project. This can be repeated until a majority has been found. The President of the Court then presents the verdict to the public. What may not become directly evident is that there are three dimensions impor­ tant for our understanding of the way judges deliberate subsumed under the micro-institutional implementation of deliberation: 1) interpersonal interaction between the individual magistrates; 2) public interaction between the magistrates and the broader public; and 3) institutional interaction with the other branches of governments. Here, I will focus on the actions of judges towards each other and

198 The Deliberative Judge the public (as strategic interaction with other branches of government was dis­ cussed above). From the beginning of the process of deliberation on the re-election reform, all indications suggested a certain anxiety amongst judges that was undoubtedly triggered by the political importance of the case itself. The selection of Humberto Sierra Porto was met with relief by those who were not chosen in the lottery lot. A Constitutional Court judge explained that “the distribution of cases is always by chance. When we organized the question of re-election, we all knew that this was one of the most important questions to arrive at the court. And no one wanted their number to be taken”.48 Clearly, and in affirmation of my argument, the Court understood the political importance of the constitutional reform project for the executive, which had the support of a powerful coalition in Congress. A Constitutional Court judge asserted “the Court is used to the notion that it takes decisions that have a big impact on the social situation in Colombia [and] can contribute to an institutional crisis”.49 The personal and institutional stakes were still high, and in contrast to predictions of the strategic paradigm, not favorable for perceptions of political maneuvering space. An auxiliary judge said, “I think there were thoughts that we are going to create institutional instability, because it was a president with a lot of popular respect”.50 In addition, at the time of the delib­ eration in the Constitutional Court in 2009, the wiretapping of the Supreme Court was one of the highest-ranking topics in the media, vividly displaying that undue pressure on high court judges was part of the political game in Colombia under Uribe. The selection of the ponente had an effect on the way discussions evolved in the 2010 decision. The draw placed the task to write the ponencia, the initial study, with Humberto Sierra Porto, a constitutional lawyer with a distinctly positivist orientation to interpret the Constitution. Sierra Porto had dissented in 2005 against the substitution doctrine, arguing that the extrapolation of axiomatic principles in the substitution test was an example of judicial decisionism: “The construction of a ‘major premise’ in the substitution judgement is in short a con­ struction of material parameter to judge the scope of the reform, [because in the end] it is then up to the constitutional judge to identify the essential elements of the 1991 Constitution in order to make the substitution test” (C-1040/05; 771). It was then hardly surprising for other judges on the court that the first study had a decisively positivist orientation. One Constitutional Court judge explained: “he was not in favor of the theory of the substitution of the constitution, [and] argued for an unconstitutionality for formal reasons”.51 As a consequence of Sierra Porto’s selection as ponente, his initial analysis of the legal question in 2009 did not include an inquiry into whether the proposed reform replaced central tenants of the Constitution. The study of the jur­ isprudential questions (and their answers, namely the identification of formal defects) did manage to receive an unambiguous majority (7:2). First, the focus was whether the Court had competence to review the reform. An auxiliary judge explained that “in 2010 the initiative for constitutional change was different, it was a citizen initiative for a referendum. It [was] a new mechanism.” The key

The Deliberative Judge

199

modification was “that it was not Congress that requested this change, but this citizen initiative”.52 Another source confirmed that “the first re-election was dif­ ferent, because it was done through Congress. So, we are not talking about a popular project. In the second time it was going be voted on by the people. This is subtle, but it is a difference.”53 This process comes with a caution, though, as it may deceive to be of constituent origin: “It is very easy to manipulate, it is in reality not the people (el pueblo) that is talking through this initiative, it is not the original constituent, who is talking, but a group of citizens that promote this, with a lot of economic power”.54 The Court explained in this regard that even while a referendum may attain some plebiscitary characteristics, the law to call for a refer­ endum has to pass Congress. It was therefore not a far stretch to argue that the Court did indeed have competence that lie within the procedural parameters of constitutionality review, as Sierra Porto did in his study. Only after affirming its own competence did the deliberation move to the procedure of the amendment process itself. In the end, the procedural flaws were of two kinds: material viola­ tions and formal violations of the constitutional reform procedures. “Cooking the books” of the referendum initiative, changing the wording of the legislation because it contained a typo, and failing to receive approval from the National Registrar prior to the vote in Congress belonged in the former category, while the transfuguismo violations (votes from sanctioned congresistas that were therefore ineligible) as well as the belated calling of an extra-ordinary session of Congress belonged in the second category. As an auxiliary judge explained, even within the category of procedural defects, “some were more formalist than others”.55 The Court then decided that the procedural flaws were of such substantial nature that the reform had to be declared unconstitutional, resulting in a reasonably unified decision of 7:2. For my argument, the critical moment came with this relatively unequivocal decision, because the Court, if it were a strategic actor, could have declared the law unconstitutional on procedural grounds with the backing of a robust majority. Yet, it did not, and the reason is rooted in deliberation itself. As a source explained, while the decision “regarding procedural flaws was very clear”, a min­ ority of four magistrates “wanted to analyze substitution as well. And they did. Amongst all of them. All nine. Why? Because the discussion does not necessarily end with one point [and] the ponente does not control the debate.”56 An aux­ iliary judge explained that the initial decision “left the door open for another referendum”. Therefore, “a group of magistrates, including my boss, was con­ cerned with the question of substitution”. This group of judges, relatively new to the Court, initiated a collective learning process, gathered and asked their teams of auxiliary judges whether “the substitution theory does apply in this case”.57 It was this collective reasoning process that changed the attitude of at least one judge. As the previous source told me, the number of those who favored the substitution doctrine in this case (and more generally) was “four at first and then it grew to five”.58 Confirming this turn of events, another auxiliary judge said that “this court insisted that, if there is another request for constitutional reform, with good procedures, can they do that?” And there were five magistrates: “Dr. Palacio, Dr.

200 The Deliberative Judge Vargas, Dra Calle, Dr. Mendoza, and Dr. Carlos Henao”, who said no.59 What is more, this final vote of 5:4 against the constitutional reform, based on competence defects, relied on a judge that was nominated by Uribe himself: “Uribe nominated three. Maria Victoria [Calle], Mauricio Gonzalez, and Pretelt. I think Gonzalez and Pretelt are clearly Uribista. Maria Victoria is pretty liberal, but when she came here—no one really knew her—but everyone thought she would be Uribista and vote with Gonzalez and Pretelt. But that did not happen.”60 She voted against re­ election. Another auxiliary judge said “the strange case is Maria Victoria [Calle]. She is in the group of five. So it is not only that her vote is part of the majority, but it is also decisive in the political sense.” Only to add, “I think this cost her some friends”.61 The initial fretfulness over the importance of the case absorbed the Court all the way through to the public declaration of the decision in February 2010. In an unusual display of institutional weight, the entirety of the Court joined the pro­ nouncement through the President of the Court at the time (Mauricio Gonzalez): “Normally, the Court never presents the decision together. It is just the President of the Court who speaks.” The circumstances, however, demanded a symbolic gesture of unity, the judge argued: “This was a thing we discussed, since this time the press, national and international, would be there, so we would also all go. So that we show that the entire court is behind the decision even if only the President of the Court speaks.” Intriguingly, the magistrate continued: “this was symbolic for the cameras of television see the entire court; that there is a solid institutionalization”.62 The analysis of the decision in the second re-election case showed two things: the case itself was perceived as an extremely high-stake decision, and the ultimate conclusion of the Court that more than two consecutive presidential terms was precipitated by a genuine change of opinion and an extensive deliberation process. There is more, though: Between the selection of the ponente and the announcing of the decision in the lobby of the Judicial Palace in Bogotá transpired a delib­ erative process, whose core normative element was the exclusion coercion. “The deliberation depended on the formal and informal rules”63 that aimed to exclude rhetorical effects and undue public pressure to facilitate an ethical reasoning; a reasoning that only wades into legal thinking shuns political arguments (or argu­ ments of convenience), eschews rhetorical effects, and facilitates the ability to change one’s opinion by the force of the better (legal) argument. In that regard, it is important to bear in mind that the task of the constitutional judge is to engage in practical reasoning. This has two critical presuppositions: 1) much alike anyone else, judges have experienced a particular political socialization with accompanying biases; and 2) the judge in general, and the constitutional judge in particular, face the specific conundrum of having to exercise great authority with very little democratic legitimization. A Constitutional Court judge explained that delibera­ tion is very interesting because “many different magistrates come with different backgrounds; some more academic, others from the judiciary”.64 Another judge said that “the constitutional magistrate cannot entirely shut out the outside world. He lives in a social environment.”65 Still another magistrate said that the

The Deliberative Judge

201

Constitutional Court “consists of nine judges with very different backgrounds; in terms of thought and of political convictions. Some are more conservative, some more progressive.”66 Another said, “it is a farce to think that all judges arrive with a blank slate to the court, and to deliberation.”67 Recall also that in liberal sys­ tems, the constitutional judge appears to exist outside of those social biases to properly exercise her authority to review democratically legitimized laws. The notion that the constitutional judge engages in practical reason forfeits the possi­ bility to circumvent the legitimacy dilemma of the constitutional judge by asso­ ciating her with neutral political socialization. In contrast to Dworkian or Rawlsian conceptualization, constitutional judges do not have superhuman capacities to reason unmitigated from outside influences. Rather, excluding undue influences on deliberation, reasoning, and arguments must be achieved via the procedure of deliberation itself. Formal and informal rules that pertain to deliberation inside the Constitutional Court encompass public and private interaction; that is the interaction between court and public, and between individual magistrates. Public interaction is evidently crucial in democracies that assert to subject public institutions to popular control. Given the exposed position and peculiar conundrum a court faces arising from the lack of direct democratic legitimacy, we might expect the Court to cultivate a par­ ticularly open “public persona”. For magistrates, however, it was not so much the public interaction that creates accountability and transparency, but the cohesion in the argumentation. The informal rules of engagement followed the imperative to assert the autonomy and independence as a collective body and the dominance of legal reasoning over political reasoning. Therefore, rather than increasing public exposure, the Court aims to reduce public interaction in the attempt to guarantee the tranquility of the process in the decision-making procedure. In the time between when a demand for constitutional review is submitted (or automatically triggered) and the actual decision taken by the Court, there are two ways of interaction between judges and the public at large: public consultations (written and oral), as well as interviews with the media. When the litigation for a case begins, the Court invites various parties from Colombian civil society to draft their opinion regarding the case. This involves civil society groups, religious representa­ tives, and political groups. These groups opine in writing to the Court and the documents produced are the most important input the Court receives from Colom­ bian society. The reasoned decisions of the Court reiterate and summarize the opi­ nions in favor and against a particular law or amendment, before proceeding to the Court’s own interpretation of the legal questions and facts in a given case. Justified with the imperative to consult coherent arguments rather than rheto­ rical effects for legal reasoning, in-person public consultations between the Court and civil society at large have become very under-utilized, even shunned. After the Political Reform in 2003, the Court had made such negative experiences that they terminated public consultations altogether. Curiously, the Court debated the issue of public consultations again during deliberations in the re-election case, including their democratic merits as well as potential pitfalls arising from public exposure. In 2009, non-governmental organizations, the Alianza Democrática and Dejusticia,

202 The Deliberative Judge submitted demands that the Court would implement a public hearing on the case. The majority decided that a public hearing was unnecessary, because a multitude of actors had already submitted positions in writing, and these served the demo­ cratic imperative that citizens are involved in issues of public importance. The Court preferred public consultations per scripturam rather than conducting oral hearings, as the latter could invite rhetorical effects, while the former incentivizes a more complete reasoning and argumentation.68 A source in the Court said that public hearings had “created a bad environment for the court [of] attempted influences. Of course, in the end it did not change anything, [but] it takes the tranquility of the process away.” Therefore, the termination of public hearings was in fact an improvement buttressing “internal guarantees of autonomy”.69 Essen­ tially, the view was that public consultations cut down on the substance of an argument, while written texts discipline authors to think through the argumenta­ tion and present the entirety of its merits. Thus, they better serve the imperative to let legal reasoning prevail over political reasoning. The second aspect of participation by the Court in the public sphere through media interviews is an even finer line to maneuver. As a public insti­ tution that makes crucial decisions of high political importance, the Court cannot withdraw entirely from the debate—in particular, since it faces peculiar legitimation issues. Nevertheless, the posture by the Court in such engage­ ments with the public is best described as extremely cautious. There are inter­ views with the printed press.70 Usually, these interviews are with the President of the Court to give the Court a unified institutional voice. The informational degree of such interviews is fairly limited, because their nature is essentially pedagogical. For example, the President of the Court may outline the legal questions, and differentiate those from political questions that dominate the public debate. Crucially, positions within the Court are kept secret until the final decision is revealed in writing, which provides space for dissenting opi­ nions. This helps to protect the autonomy of deliberation, because it is easier to change an opinion during the process and keep an open mind if no indivi­ dual judge feels the obligation to justify perspectives that are not the final ones. The President of the Court safeguards this autonomy by only speaking in the third person during interviews (“the Court will inquire”), and referencing the Sala Plena when asked specific juridical questions. A judge explained the Court has “a president [who] is the voice of the Court to explain the con­ stitutional questions, which are more complex still. So it is important that the voice of the body does her job of explaining cases, but not to discuss them. To educate!”71 Another judge argued that generally the interaction “with the media is zero [and] we do not do any declarations about what happens with the case”. That counted especially for “this [re-election] decision that was so important”.72 Thus, while the media is evidently critical in a functioning democracy, that does not necessarily mean that as a constitutional judge you need to directly address it: “what I say about the public discourse is that you consult it”. In fact, “all of that is merely part of the process of coming to a decision, in line with incorporating the political history”. In the end, though,

The Deliberative Judge

203

they followed the imperative that “the decisions are there, and do not require more controversy or discussion”.73 Another magistrate explained, “the judge talks to the citizens through her sentences, and here is said what the structure of her thought is. Her constitutional thought. And I don’t think you need to opinionate after this. Nothing!” It is the decision of the Court that commu­ nicates with the citizenry, and if you are, as a judge, “not part of the majority you can dissent and even reflect this to the people (pueblo). And I think this is important.”74 While this may appear as a common and even convenient strat­ egy, letting the decisions speak for themselves, it is very different from the United States Supreme Court, where individual magistrates engage in a sub­ stantial amount of “external deliberation”, while apparently engaging in little internal deliberation. From the interaction between court and public, we move to the inter­ personal relations between individual magistrates and the implementation of deliberation in the Court itself. It evolves around the most sacred institutional space of the Constitutional Court, the Sala Plena (Plenary Chamber): the “coliseum and arena”.75 It was a particularly valued space for high-stake cases such as the re-election reform: “In this concrete case, since it was such a delicate question, no one wanted to talk with colleagues outside the Sala Plena”.76 This sen­ timent that deliberation exclusively takes place inside that plenary chamber was reit­ erated by seven judges I interviewed: all stated that “the most important setting is of course the Sala Plena. Here, deliberations take place.”77 These same judges also stated that deliberation is the sine qua non of institutional autonomy and indepen­ dence. It is therefore only consistent that the magistrates of the Constitutional Court have paid abundant attention to adhering to informal rules that protect that institu­ tional independence. The imperative to let the force of the better argument prevail precludes specific conducts between magistrates. These are not written down in a regulatory statute but are the informal rules of how magistrates govern their interaction.78 These rules have to cope with the reality that judges are social beings. While it is fairly easy to reduce interactions with other branches of government or the media, it is impossible to exclude social interactions with your professional peers before a decision or a deliberation process. The key is to limit case-specific consultations with other magistrates to the plenary chamber. A judge said, “I have a rule: do not discuss the project outside of the Sala Plena” because “the most important place of the Con­ stitutional Court is the Sala Plena. In the end, all that you do is to bring [your project] to the Sala Plena.”79 The corollary of refraining from talking with your peers prior to deliberation inside the Sala Plena is that magistrates discuss specific legal questions only with their dispatch of auxiliary judges. These teams of auxiliary judges that work in the Constitutional Court have often been working in the Court for a longer time and often are distinguished professors of law themselves. An aux­ iliary judge herself stated that “the magistrates depend a lot on the auxiliary judges. In 2010, they were all new judges except for three, but the auxiliary magistrates were always there.”80Auxiliary judges do sometimes interact with one another and potentially speak on legal questions to exchange know-how.81

204 The Deliberative Judge My interviews also disclosed an informal rule of conduct that reaches into deliberation itself. While the working relationships between judges is often colle­ gial and friendly, the professionalism of the task stipulates a very specific collegi­ ality that makes quid pro quo agreements between judges unacceptable. The magistrate explained it best himself for why they do not exchange votes, neither for ideological nor political purposes: [The atmosphere in the Sala Plena] is collegial, but not in the sense that if a col­ league has a project and you vote for it, you can come back the next week and claim the vote from that colleague for a project of your own. If it were like this, that would end the deliberation, this would terminate the Court. The two dan­ gers for a collegial organization are the compromises of collegiality and antag­ onistic band alignments. The first occurs when there are prearranged majorities prior to the case arriving in the Court. The second is the opposite and occurs when the animosity is so stark that they form fixed blocks and that you only vote with yours and never with the other. Fortunately, in this Court there are collea­ gues that work a lot on this, that it does not degenerate into a situation like this.82 It is noteworthy that a scandal befalling the Constitutional Court in 2015 largely confirmed the inadmissibility of such conduct. Admittedly, the nefariousness of the alleged conduct was more severe. Magistrate Mauricio Gonzalez accused then President of the Court, Jorge Pretelt, of soliciting bribes from Fidupetrol to issue a tutela that would annul a sentence that had imposed a fine of $9 million. As described in the epilogue, ultimately the former President of the Constitutional Court was dismissed from the Court, charged with (criminal) corruption charges, and sentenced to six years in prison. In sum, the implementation of deliberation has as its premise that rationality does not equate to (Alexy 1989, 293). Legal reasoning can never eliminate the uncertainty implicit in general practical discourse, because it is itself a form of practical discourse. It only aims, under the constraints placed on judicial and legal reasoning, to “limit the range of the discursively possible in as rational a way as one can” (288). Equating legal rationality with certainty is further compromised by the fact that judges as social actors have individual preferences rooted in parti­ cular political socializations. Legal discourse only claims for itself to establish ways to exclude coercion so as to better engage in a cooperative search for truth because “decisions have many hands in the process, so that after two or three weeks of deliberation, the project [may be] a different one”.83 In that specific endeavor, the key task for the deliberative judge is to “translate a political question into a juridical question, so that the language of convenience is not part of it”.84 It is for that reason that preferences retreat behind the exercise of public reason: reasoning and deliberation is properly implemented through “a process of valida­ tion [of arguments]. A process that engages the law and constitution and is not political.”85 Most importantly, as the re-election case shows: “opinions change!”86

The Deliberative Judge

205

Conclusion: The Deliberative Judge and the Empowerment of Colombia’s Constitutional Court The point of departure for this chapter (as for the book) was that when the Constitutional Court declared a potential third term in the presidency uncon­ stitutional in 2010, it manifestly displayed its own institutional independence against a charismatic President, who commanded expansive de jure and de facto power. Prior to that decision, Colombia’s Constitutional Court has shown a much more careful attitude in inter-branch constitutional disputes. This fun­ damentally changed when the Court for the first time declared a constitutional reform unconstitutional in its entirety on competence grounds. The competence claim, the normative core of the jurisprudence controlling constitutional reforms, was a novel and far-reaching reinterpretation of norms that outline the Court’s review powers. It enabled the Court to limit the ability of Colombia’s majoritarian institutions to alter the political charter ad infinitum. The magnitude of the deci­ sion lies in no small part in the fact that President Álvaro Uribe was not like any other President in Colombia’s history. His Democratic Security policy not only succeeded militarily against the FARC guerrilla, but it also conjoined notions of natural leadership, conservative values, and communitarian social bonds between rulers and ruled to produce expansive transformative capacities. In short, he had translated his power and popularity in a healthy sense of mission that aimed at fundamentally altering the political landscape in Colombia. It is for that reason that the decision to withhold another term also crystallized a generic tension of liberal democracies that vacillate between majoritarian claims of democratic legiti­ macy and constitutionalist/legal constraints placed upon the exercise of political power: The question is how far does the power of a judge reach into society? How far does a judge’s control go vis-à-vis the other political organs of a state? Basically, I want to say: the entire framework of Nuevo Con­ stitucionalismo builds on the idea that courts exercise democratic contra­ majoritarian control. Democracy is not simply a quantitative problem, not only of votes, but also a control of what those votes do. Hitler was elec­ ted. This created a substantial change in a Nuevo Constitucionalismo, that democracy is not only a formal system, but also contains axiomatic princi­ ples that are inscribed into its premise. And the function of the judge is exactly the control of these axiomatic principles. To prevent that democ­ racy ends in a game of majorities.87 In the end, Democratic Legality prevailed over Democratic Security: the exercise of political power must be legally sanctioned and faces limitations arising from the basic fact that a society is pluralistic and cannot be intelligible understood as being represented by one General, even democratic, Will. Exploring what was behind this tangible increase in judicial power juxta­ posed strategic action with deliberative action as plausible explanans. As seen,

206 The Deliberative Judge the context conditions of the case provided compelling rationales to engage the rational choice institutionalism. Following Epstein and Knight (2013), there are three core elements of the strategic model of judicial behavior: 1) the a priori specification of goals; 2) the inclusion of constraining factors such as the preferences and likely actions of other institutions, colleagues, and public; and 3) the institutional framework that structures the interaction amongst themselves as well as with other actors (12). Put to the test in varying con­ texts, not least in Latin American cases of judicial empowerment, the strategic model has given rise to three applicable hypotheses and implications: 1) the political/electoral uncertainty argument: courts tend to act more independently when imminent regime change or executive-legislature misalignment veil the certainty of actors in the other branches of government; 2) the prudent stra­ tegizer argument: courts act akin to a prudent strategizer who upholds princi­ ples but defects on the merits in prominent cases, and then applies the principles in less prominent cases; and 3) the constancy of preferences: pre­ ferences are informed by political socialization and remain constant during deliberations. Following Hübner Mendes (2013), and a host of legal theorists from Alexy (1989) to Kumm (2010), the understanding of the deliberative constitutional court builds on the contention that courts engage in collective decision-making processes on questions of public authority, in which partici­ pants remain open “to transform their preferences in the light of well-articu­ lated arguments” (Hübner Mendes 2013, 18). At its core, deliberation builds on reason giving as the grounds for a decision, which in turn requires a will­ ingness to review positions, establish an ethics of conversation, and ensure the absence of coercion. From these theorizations followed: can strategic behavior best explain the Constitutional Court’s increase of authority, or is a discursive interpretation of judicial behavior better suited? To test these hypotheses, the Colombian case provided an extra-ordinary well-suited context, because the Court answered constitutional questions regarding reform of presidential terms in office twice. Moreover, there were ten constitutionality decisions (including the two re-election cases) between 2002 and 2010 that engaged the substitu­ tion doctrine. Therefore, we could conveniently observe whether the Court’s authority increase followed a strategic or deliberative logic. Table 4.7 below lists the observable implications of either explanans: Table 4.7 Observable implications of strategic and deliberative action Strategic Action

Deliberative Action

� Misalignment between parties in execu­ � No tive and legislature � Factionalism within governing coalition � Tangible decrease in public support for executive and/or governing coalition in legislature � Imminent regime change

political misalignment inside government nor across executive and legislature � Stable public support � No imminent regime change

The Deliberative Judge Strategic Action

Deliberative Action

� Laying jurisprudential groundwork in � less consequential cases � Deference to executive in consequential cases � No change of opinion during deliberation

207

� �

Jurisprudential development in most consequential cases Defiance towards executive in consequential cases Change of opinion during deliberation

Source: Compiled by the author

The closing exclamation of the core part of this chapter—“opinions change!”— amalgamates my argument: the Court did not act strategically but followed dis­ cursive imperatives. It is difficult to explain such opinion changes inside the Con­ stitutional Court with reference to strategic action—and the supplementary evidence corroborates the conclusion that the Court acted like a deliberative court. First, Uribe’s appeal eclipsed Congress, so that he not only dominated public opinion but also the legislature (akin to O’Donnell’s delegative president; 1994). Importantly, there was no evident withering of his hegemony over Colombia’s political arena between 2002–2010. His personal popularity remained unabashedly high throughout both terms. In fact, his second term produced higher survey results than the first one, and the majorities for constitutional reform were clearer the second time that presidential term limits were considered. Democratic Security espoused a natural leader whose command is trusted; Con­ gress showed this trust toward Uribe and followed compliantly. Thus, there was no perceivable political opening or weakness in Uribe’s position. Moreover, the magistrates on the Court knew of the President’s popularity and the resulting potential to create an institutional crisis by upsetting his support base. Nevertheless, they did object to Uribe’s reform project. The argument here is that the Court extended its control not despite Uribe’s power and popularity but because of his “Bonapartian” posture. The lockstep advancements of key elements of the theory in constitutionality decisions of great importance to the executive, and Uribe himself, evidence that the substitution doctrine was a direct response. As shown, it was indeed created ex-nihilio when he entered office and began implementing constitutional reforms—even though his predecessor had already initiated some reform projects—and the most critical steps in creating the sub­ stitution doctrine (the review of the 2003 political reform), specifying its meaning (the 2004 presidential re-election reform), and expanding its application (the second presidential re-election reform in 2009) correlate with highly consequential cases. It is my argument that these observations contradict the predictions of the prudent strategizer argument, making it non-applicable to the creation of the substitution doctrine: the Court acted urgently, not prudently! These observations served to argue why the prudent strategizer argument does not apply to this case. Following that assertion, I outlined the reasons in favor of the deliberative judge as the explanans for that outcome. The reconstruction of the 2010 decision identified a serendipitous moment that made facing the Gretchen

208 The Deliberative Judge question between form and substance of constitutional reform unavoidable. This moment attained its importance for the eventual outcome because of the impor­ tance of procedures of deliberation. In Colombia, “a country full of lawyers”, judges often say that “la ley cuenta” (the law matters). This differs from the banal, and misleading, assertion that judges solely apply or merely find the law. Rather, entrenched within that incomplete conjunction—the law matters for what?—is a claim that connects valid legal reasoning, and thereby decisions, with procedure. Constitutional review, by definition, deals with ambiguity in what the law actually is. It may involve a constitutional question that has never been raised before or a novel situation that has not been addressed through a particular jurisprudential doctrine. Generally, applying abstract norms, consecrated in text, to concrete situations in the production of legal decisions requires at least three steps of interpretation: what are the concrete facts and their legal significance? What is the constitutional norm and the Court’s jurisprudence on these norms? And how do these relate to one another? The specification of the relation between concrete facts with legal significance and constitutional norm is the essence of the legal decision. Ultimately, the application of the substitution doctrine in 2010 was contingent on the change of opinion during deliberation; the change of mind of the fifth magistrate to explore the substitution doctrine. Further, this change of mind depended on a set of formal and informal rules of deliberation inside the Con­ stitutional Court. Had it not been for those internal regulations of deliberation, the Gretchen question would not have arrived in the process of deliberation. Informal rules between magistrates serve the imperative to protect the integrity of the Sala Plena and affect how magistrates interact with the wider public and with each other. Public interaction between magistrates and civil society—be it through interviews or consultations prior to decisions—is limited. Various parties can submit opinions in writing and only the President of the Court submits to inter­ views with the press. The most important informal rule governing the inter­ personal interaction between judges is that social gatherings exclude politicking. Political matters, and much less specific cases, are not to be discussed outside of the Sala Plena. Hereby, the Court attempts to exclude coercion emanating from within the institution itself. This also counts for the stipulation that quid pro quod agreements between judges for votes are strictly prohibited. Votes are not to be sold but argued for! Epstein and Knight have called on rational choice scholars of judicial behavior to incorporate the importance of law for the self-maximizing agent, even though “many political science studies over the past 60 years have sought to refute” the centrality of law (2013, 25). This appears to be a belated acknowledgment of the centrality of the practice of legal reasoning, because self-maximization in legal decision-making processes is only sensible if they are understood as collective efforts around the exercise of public reasoning. Only then can the joy of con­ structing a good legal argument in a debate amongst judges be utility-maximizing. Sound procedure in legal reasoning also mitigates a constitutional court’s democratic dilemma. The other branches of government in a liberal democracy are functionally specified, and democratically elected, to represent society’s various

The Deliberative Judge

209

interests in the law-making process. The constitutional court’s specification is loaded with the more “mysterious goal of upholding the rule of law” while, as an unelected body, lacking direct democratic legitimation (Hübner Mendes 2012, 1). Amplifying the dilemma: magistrates are never fully detached from socio-political biases and do in fact exercise a form of public reasoning (Alexy 1989). The implementation of specific procedures to come to legal decisions is key to bridge the gap between lack of democratic legitimation and the functional role of the constitutional judge to come to consequential decisions regarding the exercise of public authority. The point of legal decision-making in constitutional bodies is to establish truths of what the law of the land is, but the rules governing the process accept at the same time that individuals with biases have to make value judgments. Personal values that individual judges may hold are effectively constrained by the role and function of courts that cognitively establish boundaries of legitimate reason giving. These boundaries, in turn, are upheld in carefully protected dis­ cursive spaces, in which individual magistrates are held accountable by their peers to not tread beyond the boundaries of legitimation. After all, as a Constitutional Court judge emphasized, a magistrate’s most “potent weapon is the coherence of her discourse”.88 It is for these reasons that the Colombian constitutional judge is better understood as a discursive rather than a strategic actor. Deliberation entails that preferences are not static: opinions change. Moreover, for opinions to change procedures must remain constant and fair. This core contention of deliberative democratic theory, the burdening of procedure with normativity, was evident in the 2010 decision. Empirically showing the relevance of the modus operandi of deliberation for judicial empowerment adds to the canon of judicial behavior, going beyond explanations that are even more discursively inclined, stressing the importance of ideas and their carriers (Nunes 2010) or the sources of endogenous growth (Bakiner 2020).

Notes 1 Helmke’s and Ríos-Figueroa’s edited volume on courts in Latin America (2011) builds on Montesquieu, Kelsen, and Schmitt’s theorizations of judicial review in modern constitutional states. Montesquieu identified in his Spirit of Laws liberties relative to the constitution that result from the separation of powers and the political liberties of the subject from the state. Kelsen found that the functional separation of powers that reg­ ulates the creation of laws is in itself subject to control through a catalogue of rights that delineate the material content of laws. Finally, Schmitt held that the organizational principle facilitated the implementation of the distributional principle and, conjoined, form the Rechststaat component of the modern constitution (Montesquieu 1949; Kelsen 1945; Schmitt 2008; see Helmke and Ríos-Figueroa 2011, 8). 2 Such strategizing is not necessarily novel for newly established courts. In fact, Marbury v. Madison constitutes the paradigmatic case for such (judicial) behavior, because Jus­ tice Marshall refrained from imposing obligations, but established the principle of judi­ cial review, which other courts at later points in time used to strike down legislation (Vanberg 2008; see Ginsburg 2003). 3 Revista Semana, “Vuelve y juega la reforma a la Justicia”, 13 September 2014. http:// www.semana.com/nacion/articulo/vuelve-juega-la-reforma-la-justicia/402640-3. Pre­ sident Santos stated that his first act in the office was to visit the courts and debate a

210 The Deliberative Judge

4

5 6 7

8 9

10 11 12 13

mending of fences between them and the executives. Prior to redrafting another attempt at judicial reform, the President visited the Constitutional Court as well as the Supreme Court, because, as Semana writes, Santos knows that the main reform “he has planned for his second term may not be approved without them”. Born in the midst of La Violencia in 1952, and coming from a wealthy ranchero family in Antioquia, Uribe’s talent already shone in school and university: a prodigy student with an outstanding work ethic, he graduated with exceptionally good grades from school and earned a law degree from the Universidad de Antioquia in Medellin in 1977. He was no outsider to politics either. His parents opened doors in the Liberal Party in Antioquia. Early on, he became the protégé of the regional Liberal Party boss, Bernardo Guerra Serna, a connection that quickly paid off. Already in 1976, he received his first political position in 1976, as Chief of Assets for Public Enterprises in Medellin. Under Liberal President Alfonso López Michelsen, he served as the Secretary General of the Ministry of Labor from 1977 until 1978. The next President Julio César Turbay, also a Liberal, made him director of the national civil aviation department in 1980. All of this took place before Uribe turned 30. When he did turn 30, Turbay’s successor appointed him mayor of Medellin—a post he held for only five months. Uribe then served two terms as Senator for Antioquia in Colombia’s Congress between 1986 and 1994, and was elected Governor of Antioquia in 1995. As Governor of his home department, Uribe became one of the loudest voices in favor of the CONVIVIR policy, building a reputation as tough and uncompromising in security policy: a political narrative fed by the early death of his father at the hands of a FARC commando intent on kidnapping Alberto Uribe. Clearly, Uribe could maneuver his way through the political landscape in his home province, and remained loyal to the Liberal Party of Colombia until the presidential election in 2002—only shifting between factions within the party (Dugas 2003, 1123). Interview with Member of Congress of the Republic of Colombia, 13 March 2013.

Interview with Constitutional Court judge, 3 May 2013.

Semana showed that Uribe used references to the estado de opinión more than a hun­ dred times during the ten months before the decision in 2009. In that year, he opened the legislative session in Congress with a speech defending the estado de opinión as superior to the estado de derecho. See Semana, “La estrategia del estado de opinion”, 14 August 2009. http://www.semana.com/nacion/articulo/la-estrategia-del-estado-op inion/106304-3. El Tiempo, “Qué es el Estado de opinión del que habló Álvaro Uribe en su discurso ante el Congreso?” 27 July 2009. http://www.eltiempo.com/archivo/ documento/CMS-5712361. Uprimny, Rodrigo, “¿Estado de Opinión o de Derecho?” El Tiempo, 17 August 2009. http://www.elespectador.com/columna156600-estado-de-opinion-o-de-derecho. US Department of State. Memorandum of Justification Concerning Human Rights Conditions with Respect to Assistance for Armed Forces. (Washington, DC: US Depart­ ment of State, 2006). http://www2.gwu.edu/~nsarchiv/NSAEBB/NSAEBB327/ doc13_20050000.pdf. Interview with Constitutional Court judge, 3 May 2013. Interview with Constitutional Court judge, 26 November 2012. Interview with Constitutional Court judge, 17 May 2013. The process of vote personalization was somewhat corrected as a consequence in the political reform in 2003. The reform changed the rules that favored excessive frag­ mentation (Legislative Act 01 of 2003; Pachón and Shugart 2010). It introduced a minimum threshold of 2% of the votes (umbral) at the national level, restricted parties to present a single open or closed list, and generated incentives to form electoral coali­ tions through the introduction of the D’Hondt system. Pachón and Shugart showed that as the size of the district in the 2002 election increased, the effective number of parties (ENP) followed suit. In 2006 and 2010, the threshold had the effect that most parties introduced lists in all districts resulting in a higher degree of nationalization, irrespective of the district’s size. The evidence showed, however, that the Chamber of

The Deliberative Judge

14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29

30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47

211

Representatives still suffers from a very high degree of fragmentation (Pachón and Shugart 2010; see also Batlle and Puyana 2011; Carroll and Pachón 2016). Interview with Senator of the Republic of Colombia, 12 March 2013. Interview with Senator of the Republic of Colombia, 12 March 2013. Interview with Constitutional Court judge, 8 March 2013. Interview with Constitutional Court judge, 20 May 2013. Interview with auxiliary judge of the Constitutional Court, 10 April 2013. Interview with Constitutional Court judge, 20 November 2012. Interview with auxiliary judge of the Constitutional Court, 10 April 2013; interview with Constitutional Court judge, 17 May 2013; interview with Constitutional Court judge, 3 May 2013. Interview with Constitutional Court judge, 17 May 2013. Interview with Constitutional Court judge, 20 May 2013. Interview with auxiliary judge of the Constitutional Court, 15 April 2013. Interview with auxiliary judge of the Constitutional Court, 24 April 2013. Interview with auxiliary judge of the Constitutional Court, 24 April 2013. Interview with auxiliary judge of the Constitutional Court, 15 May 2013. Interview with auxiliary judge of the Constitutional Court, 24 April 2013. Interview with auxiliary judge of the Constitutional Court, 24 April 2013. It needs to be said that those insisting on the validity of the substitution doctrine expli­ citly stated that the jurisprudence as it stood did not forego substantial constitutional reform through constituent assembly, even if it went as far as replacing the republican with a monarchical structure. As it were, the limits for such a substitution lied in the procedure of a constituent assembly, for which, one can speculate, the 1991 process would serve as precedent. Interview with Constitutional Court judge, 16 May 2013. Interview with Constitutional Court judge, 16 May 2013. Interview with Constitutional Court judge, 17 May 2013. Interview with auxiliary judge of the Constitutional Court, 15 May 2013. Interview with Constitutional Court judge, 20 November 2012. Congress of Colombia, Acto Legislativo 02 de 2004 (Bogotá: Alcadia Bogotá, 2004). http://www.alcaldiabogota.gov.co/sisjur/normas/Norma1.jsp?i=15519. Interview with Constitutional Court judge, 16 May 2013. Interview with Constitutional Court judge, 16 May 2013. Interview with Constitutional Court judge, 17 May 2013. Interview with Constitutional Court judge, 17 May 2013. Interview with Constitutional Court judge, 8 March 2013. Interview with Constitutional Court judge, 8 March 2013. Interview with Constitutional Court judge, 17 May 2013. Interview with Constitutional Court judge, 17 May 2013. Interview with Supreme Court judge, 13 April 2013. Interview with Supreme Court judge, 26 November 2012. Interview with Constitutional Court judge, 17 May 2013. Interview with Constitutional Court judge, 19 November 2012. I conducted over 40 in-depth elite interviews with persons directly or indirectly involved with the Constitutional Court of Colombia. The interviewees consisted of all magistrates of Colombia’s Constitutional Court, who were on the Court in 2010 when the decision regarding a law to call for referendum to allow a second re-election was decided, save for one, whose auxiliary judge I interviewed. I also interviewed four of the nine members of the Constitutional Court who made the decision regarding the law to permit constitutional change for a first re-election of the President of the Republic. Furthermore, the interviews consisted of auxiliary judges of eight different magistrates of the Court from different periods. The findings on the procedures of deliberation, as well as the evidence suggesting that a genuine change of opinion occurred during

212 The Deliberative Judge

48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88

deliberations, build on these interviews, disaggregated by order of observation, and corroborated by the actas and the reasoned decision itself. Interview with Constitutional Court judge, 30 March 2013. Interview with Constitutional Court judge, 8 March 2013. Interview with auxiliary judge of the Constitutional Court, 15 March 2013. Interview with Constitutional Court judge, 26 November 2012. Interview with auxiliary judge of the Constitutional Court, 10 April 2013. Interview with secretary of Constitutional Court, 10 May 2013. Interview with auxiliary judge of the Constitutional Court, 10 April 2013. Interview with auxiliary judge of the Constitutional Court, 10 April 2013. Interview with auxiliary judge of the Constitutional Court, 24 April 2013. Interview with auxiliary judge of the Constitutional Court, 10 April 2013. Interview with secretary of Constitutional Court, 10 May 2013. Interview with auxiliary judge of the Constitutional Court, 10 April 2013. Interview with auxiliary judge of the Constitutional Court, 10 April 2013. Interview with auxiliary judge of the Constitutional Court, 12 April 2013. Interview with Constitutional Court judge, 30 March 2013. Interview with auxiliary judge of the Constitutional Court, 12 April 2013. Interview with Constitutional Court judge, 8 March 2013. Interview with Constitutional Court judge, 3 May 2013. Interview with Constitutional Court judge, 26 November 2012. Interview with Constitutional Court judge, 20 November 2012. Constitutional Court of Colombia, Sala Plena, “Acta No. 3, 27 January 2010”. Interview with secretary of the Constitutional Court, 10 May 2013. I have not come across interviews on television or radio. Interview with Constitutional Court judge, 26 November 2012. Interview with Constitutional Court judge, 30 March 2013. Interview with Constitutional Court judge, 26 November 2012. Interview with Constitutional Court judge, 26 November 2012. Interview with Constitutional Court judge, 20 November 2012. Interview with Constitutional Court judge, 30 March 2013. Interview with Constitutional Court judge, 8 April 2013. Constitutional Court of Colombia, Reglamento Interno (Bogotá: Constitutional Court of Colombia, 2008). http://www.corteconstitucional.gov.co/lacorte/reglamento.php. Interview with Constitutional Court judge, 20 November 2012. Interview with auxiliary judge of the Constitutional Court, 10 April 2013. Interview with auxiliary judge of Constitutional Court, 15 March 2013. Interview with Constitutional Court judge, 20 November 2012. Interview with Constitutional Court judge, 20 November 2012. Interview with Constitutional Court judge, 26 November 2012. Interview with Constitutional Court judge, 8 March 2013. Interview with Constitutional Court judge, 26 November 2012. Interview with Constitutional Court judge, 30 March 2013. Interview with Constitutional Court judge, 30 March 2013.

References Alexy, Robert. 1989. A Theory of Legal Argumentation. The Theory of Rational Discourse as Theory of Legal Justification. Oxford: Oxford University Press. Bakiner, Onur. 2020. “Endogenous Sources of Judicial Power: Parapolitics and the Supreme Court of Colombia”. Comparative Politics 52 (4): 603–624. https://doi.org/ 10.5129/001041520X15743890683814.

The Deliberative Judge

213

Barros, Robert. 2002. Constitutionalism and Dictatorship: Pinochet, the Junta, and the 1980 Constitution. Cambridge: Cambridge University Press. https://doi.org/10.1017/ CBO9780511606298. Batlle, Margarita, and José Ricardo V. Puyana. 2011. “El Nivel de Nacionalización Del Sistema de Partidos Colombiano: Una Mirada a Partir de Las Elecciones Legislativas de 2010”. Colombia Internacional 74 (July): 27–57. https://doi.org/10.7440/colombia int74.2011.03. Bejarano, Ana María. 2013. “Politicizing Insecurity: Uribe’s Instrumental Populism”. In Latin American Populism in the Twentieth-First Century, edited by Carlos de la Torre and Cynthia J. Arnson, 323–351. Baltimore: The Johns Hopkins University Press. Bernal, Carlos. 2013. “Unconstitutional Constitutional Amendments in the Case Study of Colombia: An Analysis of the Justification and Meaning of the Constitutional Replace­ ment Doctrine”. International Journal of Constitutional Law 11 (2): 339–357. https:// doi.org/10.1093/icon/mot007. Bonilla Maldonado, Daniel (eds). 2013. Constitutionalism of the Global South. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9781139567114. Botero, Felipe, GaryHoskin, and MónicaPachón. 2010. “Flash Report: A Key Decision in Colombia”. Vancouver:Andean Democracy Research Network, Centre for the Study of Democratic Institutions. Cameron, Maxwell. 2013. Strong Constitutions: Social-Cognitive Origins of the Separation of Powers. Oxford: Oxford University Press. https://doi.org/10.1093/acprof:oso/ 9780199987443.001.0001. Carroll, Royce, and Mónica Pachón. 2016. “The Unrealized Potential of Presidential Coalitions in Colombia”. In Legislative Institutions and Lawmaking in Latin America, edited by Eduardo Alemán and George Tsebelis, 122–147. Oxford University Press. https://doi.org/10.1093/acprof:oso/9780198777861.003.0005. Cepeda Espinosa, Manuel José. 2012. “Readings on the Colombian Constitutional Court”. Colloquium Law School at the University of Texas, Austin. Chávez, Rebecca Bill, John A. Ferejohn, and Barry R. Weingast. 2011. “A Theory of the Politically Independent Judiciary. A Comparative Study of the United States and Argentina”. In Courts in Latin America, edited by Gretchen Helmke and Julio RíosFigueroa, 219–247. Cambridge: Cambridge University Press. https://doi.org/10. 1017/CBO9780511976520.009. Colón-Ríos, Joel. 2011. “Carl Schmitt and Constituent Power in Latin American Courts: The Cases of Venezuela and Colombia”. Constellations 18 (3): 365–388. https://doi. org/10.1111/j.1467-8675.2011.00643.x. Colón-Ríos, Joel. 2012. Weak Constitutionalism: Democratic Legitimacy and the Question of Constituent Power. London: Routledge. https://doi.org/10.4324/9780203120132. Couso, Javier. 2003. “The Politics of Judicial Review in Chile in the Era of Democratic Transition, 1990–2002”. Democratization 10 (4): 70–91. https://doi.org/10.1080/ 13510340312331294037. Crisp, Brian, and Rachael E. Ingall. 2002. “Institutional Engineering and the Nature of Representation: Mapping the Effects of Electoral Reform in Colombia”. American Journal of Political Science 46 (4): 733–748. https://doi.org/10.2307/3088430. Dixon, Rosalind, and David Landau. 2015. “Transnational Constitutionalism and a Lim­ ited Doctrine of Unconstitutional Constitutional Amendment”. International Journal of Constitutional Law 13 (3): 606–638. https://doi.org/10.1093/icon/mov039.

214 The Deliberative Judge Domingo, Pilar. 2000. “Judicial Independence: The Politics of the Supreme Court in Mexico”. Journal of Latin American Studies 32 (3): 705–735. https://doi.org/10. 1017/S0022216X00005885. Dugas, John C. 2003. “The Emergence of Neopopulism in Colombia? The Case of Álvaro Uribe”. Third World Quarterly 24 (6): 1117–1136. https://doi.org/10.1080/ 01436590310001630099. Dworkin, Ronald. 1986. Law’s Empire. Cambridge, MA: Belknap Press. Epstein, Lee, and Jack Knight. 1998. The Choices Justices Make. Washington, DC: CQ Press. Epstein, Lee, and Jack Knight. 2013. “Reconsidering Judicial Preferences”. Annual Review of Political Science 16 (1): 11–31. https://doi.org/10.1146/annurev-polisci-032211-214229. Finkel, Jodi. 2005. “Judicial Reform as Insurance Policy: Mexico in the 1990s”. Latin American Politics & Society 47 (1): 87–113. https://doi.org/10.1353/lap.2005.0006. Finkel, Jodi. 2008. Judicial Reform as Political Insurance: Argentina, Peru, and Mexico in the 1990s. Notre Dame: University of Notre Dame Press. Friedman, Barry. 2006. “Taking Law Seriously”. Perspectives on Politics 4 (2). https://doi. org/10.1017/S1537592706060178. Gaviria Vélez, José Obdulio. 2004. Reelección. Que El Pueblo Decida. Bogotá: Editorial Planeta Colombiana. Gibson, James L. 2008. “Judicial Institutions”. In Oxford Handbook of Political Institutions, edited by R.A.W. Rhodes, Sarah A. Binder, and Bert A. Rockman, 514–535. Oxford: Oxford University Press. https://doi.org/10.1093/oxfordhb/9780199548460.003.0026. Gillman, Howard. 2001. “What’s Law Got to Do with It? Judicial Behavioralists Test the ‘Legal Model’ of Judicial Decision Making”. Law & Social Inquiry 26 (2): 465–504. https://doi.org/10.1111/j.1747-4469.2001.tb00185.x. Gillman, Howard. 2002. “How Political Parties Can Use the Courts to Advance Their Agendas: Federal Courts in the United States, 1875–1891”. American Political Science Review 96 (3): 511–524. https://doi.org/10.1017/S0003055402000291. Ginsburg, Tom. 2003. Judicial Review in New Democracies: Constitutional Courts in Asian Cases. Cambridge: Cambridge University Press. https://doi.org/10.1017/ CBO9780511511189. Gutiérrez Sanín, Francisco. 2007. Lo Que El Viento Se Llevo: Democracia y Partidos Politicos En Colombia 1958–2006. Bogotá: Grupo Editorial Norma. Gutiérrez Sanín, Francisco. 2019. Clientelistic Warfare. Paramilitaries and the State in Colombia (1982–2007). London: Peter Lang UK. Habermas, Jürgen. 1996. Between Facts and Norms: Contributions to a Discourse Theory of Law and Democracy. Cambridge, MA: MIT University Press. Hailbronner, Michaela. 2017. “Transformative Constitutionalism: Not Only in the Global South”. The American Journal of Comparative Law 65 (3): 527–565. https://doi.org/ 10.1093/ajcl/avx016. Helmke, Gretchen. 2004. Courts under Constraints. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9780511510144. Helmke, Gretchen, and Julio Ríos-Figueroa (eds). 2011. Courts in Latin America. Cam­ bridge: Cambridge University Press. https://doi.org/10.1017/CBO9780511976520. Hilbink, Lisa. 2007. Judges Beyond Politics in Democracy and Dictatorship. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9780511511509. Hübner Mendes, Conrado. 2012. “Deliberative Performance of Constitutional Courts”. Rule of Law Colloquium at the WZB, Berlin. Hübner Mendes, Conrado. 2013. Constitutional Courts and Deliberative Democracy. Oxford: Oxford University Press.

The Deliberative Judge

215

Kelsen, Hans. 1945. General Theory of Law and State. Cambridge, MA: Harvard University Press. Kumm, Mattias. 2007. “Institutionalizing Socratic Contestation: The Rationalist Human Rights Paradigm, Legitimate Authority and the Point of Judicial Review”. European Journal of Legal Studies 1 (2): 1–32. Kumm, Mattias. 2010. “The Idea of Socratic Contestation and the Right to Justification: The Point of Rights-Based Proportionality Review”. Law & Ethics of Human Rights 4 (2): 142–175. https://doi.org/10.2202/1938-2545.1047. Kumm, Mattias. 2012. “Alexy’s Theory of Constitutional Rights and the Problem of Judicial Review”. In Institutionalized Reason: The Jurisprudence of Robert Alexy, 201–217. Oxford University Press. https://doi.org/10.1093/acprof:oso/9780199582068.003.0009. Levitsky, Steven, and James Loxton. 2013. “Populism and CCompetitive Authoritarianism in the Andes”. Democratization 20 (1): 107–136. https://doi.org/10.1080/ 13510347.2013.738864. Londoño Hoyos, Fernando. 2001. “La Economía En La Constitución Del 91”. Revista Javeriana 137 (678): 639–648. Mainwaring, Scott, Ana María Bejarano, and Eduardo Pizarro Leongómez. 2006. The Crisis of Democratic Representation in the Andes. Stanford: Stanford University Press. Mainwaring, Scott, and Timothy R. Scully. 1995. Building Democratic Institutions: Party Systems in Latin America. Stanford: Stanford University Press. Mejía Guinand, Luis, Felipe Botero, and Juan Rodríguez-Raga. 2008. “Paving Roads with Votes? Budgetary Allocations for Road Infrastructure Projects in Colombia, 2002– 2006”. Colombia Internacional 68: 14–42. Montesquieu, Baron de. 1949. The Spirit of Laws. New York: Hafner Press. https://doi. org/10.2307/2593726. Negretto, Gabriel L. 2013. Making Constitutions. Making Constitutions: Presidents, Parties, and Institutional Choice in Latin America. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9781139207836. Nunes, Rodrigo. 2010. “Ideal Justice in Latin America: Interests, Ideas, and the Origins of Progressive Judicial Activism in Brazil and Colombia”. University of Texas, Austin. O’Donnell, Guillermo A. 1994. “Delegative Democracy”. Journal of Democracy 5 (1): 55–69. Pachón, Mónica, and Gary Hoskin. 2011. “Colombia 2010: Análisis de Las Elecciones Presidenciales y Legislativas”. Colombia Internacional 74 (July): 9–26. https://doi.org/ 10.7440/colombiaint74.2011.02. Pachón, Mónica, and Matthew S. Shugart. 2010. “Electoral Reform and the Mirror Image of Inter-Party and Intra-Party Competition: The Adoption of Party Lists in Colombia”. Electoral Studies 29 (4): 648–660. https://doi.org/10.1016/j.electstud.2010.06.005. Pizarro Leongómez, Eduardo. 2002. “La Atomización Partidista En Colombia: El Fenó­ meno de Las Micro-Empresas Electorales”. Working Paper of the Helen Kellogg Institute for International Studies. Notre Dame. Popper, Karl R. 1968. The Logic of Scientific Discovery. New York: Harper & Row. Ramseyer, J. Mark. 1994. “The Puzzling (In)Dependence of Courts: A Comparative Approach”. The Journal of Legal Studies 23 (2): 721–747. https://doi.org/10.1086/467943. Rawls, John. 1993. Political Liberalism. New York: Columbia University Press. https:// doi.org/10.1177/0090591790018003001. Ríos-Figueroa, Julio. 2007. “Fragmentation of Power and the Emergence of an Effective Judi­ ciary in Mexico, 1994–2002”. Latin American Politics & Society 49 (1): 31–57. https://doi. org/10.1353/lap.2007.0011.

216 The Deliberative Judge Ríos-Figueroa, Julio. 2016. Constitutional Courts as Mediators. Armed Conflict, CivilMilitary Relations, and the Rule of Law in Latin America. Cambridge: Cambridge University Press. Rodríguez-Raga, Juan Carlos. 2011. “Strategic Deference in the Colombian Constitutional Court, 1992–2006”. In Courts in Latin America, edited by Gretchen Helmke and Julio Ríos-Figueroa, 81–98. Cambridge: Cambridge University Press. https://doi.org/10. 1017/CBO9780511976520.004. Rousseau, Jean-Jaques. 2002. The Social Contract and Discourses. London: Everyman. Roznai, Yaniv. 2013. “Unconstitutional Constitutional Amendments—the Migration and Success of a Constitutional Idea”. American Journal of Comparative Law 61 (3): 657– 719. https://doi.org/10.5131/AJCL.2012.0027. Schmitt, Carl. 2008. Constitutional Theory. Durham: Duke University Press. Segal, Jeffrey A. 2008. “Judicial Behavior”. In The Oxford Handbook of Law and Politics, edited by Keith E. Whittington, R. Daniel Kelemen, and Gregory A. Caldeira, 19–34. Oxford: Oxford University Press. https://doi.org/10.1093/oxfordhb/9780199208425.003.0002. Segal, Jeffrey A., and Harold J. Spaeth. 2002. The Supreme Court and the Attitudinal Model Revisited. Cambridge: Cambridge University Press. https://doi.org/10.1017/ CBO9780511615696. Sieder, Rachel, Line Schjolden, and Alan Angell (eds). 2005. The Judicialization of Politics in Latin America. New York: Palgrave Macmillan US. https://doi.org/10.1007/ 978-1-137-10887-6. Spiller, Pablo T., and Rafael Gely. 2008. “Strategic Judicial Decision-Making”. In The Oxford Handbook of Law and Politics, edited by Keith E. Whittington, R. Daniel Kele­ man, and Gregory A. Caldeira, 34–45. Oxford: Oxford University Press. https://doi. org/10.1093/oxfordhb/9780199208425.003.0003. Valencia Villa, Hernando. 2012. Cartas de Batalla. Una Critica Del Constitucionalismo Colombiano. Bogotá: Panamericana Editorial. Vanberg, Georg. 2008. “Establishing and Maintaining Judicial Independence”. In The Oxford Handbook of Law and Politics, edited by Keith E. Whittington, R. Daniel Kele­ men, and Gregory A. Caldeira. Oxford: Oxford University Press. https://doi.org/10. 1093/oxfordhb/9780199208425.003.0007. Warren, Mark E. 1999. Democracy and Trust. Cambridge: Cambridge University Press. Weyland, Kurt. 2013. “Latin America’s Authoritarian Drift: The Threat from the Populist Left”. Journal of Democracy 24 (3): 18–32. Whittington, Keith E. 2000. “Dworkin’s ‘Originalism’: The Role of Intentions in Con­ stitutional Interpretation”. The Review of Politics 62 (2): 197–229. https://doi.org/10. 1017/S0034670500029442.

Court Cases Constitutional Court of Colombia. C-551/03. M.P. Eduardo Montealegre Lynett. Constitutional Court of Colombia. C-1200/03. M.P. Manuel Jose Cepeda Espinosa, Rodrigo Escobar Gil. Constitutional Court of Colombia. C-816/04. M.P. Jaime Córdoba Triviño, Rodrigo Uprimny Yepes. Constitutional Court of Colombia. C-970/04. M.P. Rodrigo Escobar Gil. Constitutional Court of Colombia. C-971/04. M.P. Manuel Jose Cepeda Espinosa.

The Deliberative Judge

217

Constitutional Court of Colombia. C-1040/05. M.P. Manuel Jose Cepeda Espinosa, Rodrigo Escobar Gil, Marco Gerardo Monroy Cabra, Humberto Antonio Sierra Porto, Álvaro Tafur Galvis, Clara Inés Vargas Hernández. Constitutional Court of Colombia. C-153/07. M.P. Jaime Córdoba Triviño. Constitutional Court of Colombia. C-180/07. M.P. Rodrigo Escobar Gil. Constitutional Court of Colombia. C-588/09. M.P. Gabriel Eduardo Mendoza Martel. Constitutional Court of Colombia. C-141/10. M.P. Humberto Antonio Sierra Porto.

5

Conclusion The Pre-eminence of Democratic Legality over Democratic Security and Discursive Institutionalism

The Question The ancient Roman God of transitions and new beginnings, Janus, most often depicted with two faces pointing in opposite directions (Graves 1975, 130 and 137), is quite possibly one of the most overused analogies in the social sciences, employed liberally to describe contradictions and inconsistencies. However, in the Colombian case it is just as likely one of the most fitting ones. Colombia con­ stantly traversed legality and illegality, stability and instability. It is therefore nei­ ther a coincidence nor a misgiving to metaphorize the specifically Colombian puzzle of relatively democratic institutions with high levels of (organized) violence with Janus’ face. After all, it is for these reasons that Colombia has for most of the 20th century been the example of a democracy with an adjective (Collier and Levitsky 1997), ranging from “oligarchic”, “restricted”, “controlled”, to “besieged” and “threatened” (Hartlyn 1988; Kline 1995; Archer 1995; Archer and Shugart 1997; Bejarano and Pizarro 2005; Bejarano et al. 2010). In 2010, the country faced a new and, in the context of its own history, unprece­ dented challenge. The question of constitutional reform to allow presidential re­ election, benefiting an extremely popular and powerful President, Álvaro Uribe, put Colombia’s separation of powers model to the test. Colombia’s Constitutional Court had to decide how far the legislature can go in amending the political charter before its basic principles are unrecognizably altered. The Constitutional Court turned down the possibility on the basis of two legal components of constitutional reasoning: formal/procedural defects and competence defects in the creation of the reform law. Convenient for analytical purposes, this was not the first time the Court had decided on the issue of presidential re-election: in 2005 a (re-election) reform project bene­ fiting the same President landed before the Court. On that occasion, the Court deferred to the executive’s position on most counts. The difference between the two outcomes set out the basic problem that this book sought to explain. Following conventional definitions of judicial power—judges’ institutional independence as well as their vertical and horizontal control authority (Kneip 2011)—three key observations formed the foundation of this analysis: 1) President Álvaro Uribe was an extremely popular and powerful President, who had few scruples to utilize his transformative capacities to initiate a far-reaching reform DOI: 10.4324/9781003229285-5

Conclusion

219

project that served his own political agenda; 2) against this evident sense of mis­ sion of a charismatic leader, the Court’s authority appreciably increased between 2005 and 2010; and 3) the jurisprudence of the Court involved a doctrine that is not explicitly mentioned in the Constitution, but is a reinterpretation of the norms outlining judicial review of constitutional reforms. Comparative politics has tools for investigating and better understanding the historical uniqueness of a process, the post-genesis evolution of institutional structures, and the micro processes within an institutional framework. Employing such conceptual tools converted the aforementioned observations into three spe­ cific research questions that guided each chapter: � � �

Does the 1991 Constitution amount to a critical juncture in Colombia’s political history? What did the post-genesis evolution of Colombia’s constitutional jurisprudence look like? What motivated judges to develop and apply the substitution doctrine, which struck down Uribe’s reform to extend the number of terms of the presidential office?

The investigation into the substance of Colombia’s institutional history provided a holistic picture of Uribe’s peculiar position in Colombia’s history and, above all, the novelty of the 1991 Constitution that established the Constitutional Court. Dissecting the post-genesis institutional evolution of legal institutions expounded that incremental change comes via institutional learning. Finally, this book con­ trasted deliberative and strategic action in the norm-creation process of Colom­ bia’s Constitutional Court. It found that the thickening and, ultimately, the application of the substitution doctrine to outlaw a constitutional reform on competence grounds is difficult to grasp with the rationalizing judge. Rather, the specific form of contestation in judicial bodies follows a deliberative logic that institutionalizes the ability to change one’s mind by the force of the better argu­ ment. This conclusion brings these most important findings together and connects them with the contribution to our understanding of the new institutionalism. The centrality of discursive factors in the creation, development, and application of institutions and norms, identified in each chapter, places this analysis in the field of discursive institutionalism.

The 1991 Colombian Constitution: Critical Junctures Revisited Asking if an event, such as the creation of the 1991 Constitution, amounts to a critical juncture inevitably leads to the Colliers’ work on foundational transforma­ tions in Latin American political arenas with the incorporation of the working class into the respective party systems. Ruth and David Collier understand critical junctures as “periods of significant change, which typically occur in distinct ways in different countries or in other units of analysis” and result in long(er)-lasting, dis­ tinct legacies (2002, 29). These junctures are distinguished from other important

220 Conclusion historical periods by three components: the preceding conditions and their pro­ pensity to generate moments of crisis, the actual juncture of institutional engi­ neering, and the legacies left by the decisions taken during the critical juncture to overcome the crisis. Variation in the juncture itself comes via the actors present in decision-making processes, the nature and transparency of the decision-making process, and the resulting (formal) norms. To investigate if the 1991 constitutional moment constitutes one of those instances in political history that places the patterns of political life on a distinctly new path, I had to increase the observational data points by turning inward, and longitudinally analyze instances of institution building in Colombia’s history to identify variations in the outcomes. After all, this inquiry concentrated on one specific case, Colombia, and a constitutional decision that was taken in a specific constitutional and political context. Therefore, the methodological framework dictated to look for other instances of institutional engineering processes in Colombia’s history that followed a similar pattern. This ideal-type trajectory could be described as follows (Figure 5.1):

Political Crisis and Violence

Institutional Engineering Process

New Institutional Framework

Figure 5.1 Ideal-type trajectory of sub-cases of institution building in Colombia (1905– 1910; 1953–1957; and 1991)

The Thousand Days’ War (1899–1902) ushered in the Reyes Reforms, La Violencía (1948–1957) in the National Front Pacts, and the drug violence of the 1980s in the constituent process that resulted in the 1991 Constitution. These events conveniently provided the basic instances of a most similar research design, and to validly measure the effects of each juncture on Colombia’s political system I differentiated between several sub-systems and held them constant in the analysis of each observational point. The result was that the 1991 Constitution does indeed constitute a critical juncture in Colombia’s history. It was the response to an unprecedented crisis, involved different (historically marginalized) actors, con­ veyed in a novel forum with participatory and public procedures. In the end, the 1991 Constituent process created a more profound institutional transformation than the other instances of institutional engineering, and has produced a unique and identifiable legacy that contrasts with those processes. Following the Colliers’ methodological imperatives, Chapter 2 first detailed the base-line conditions by exploring the fundamentals of Colombian institutionalism rooted in the 19th century post-independence epoch. Historically, the state

Conclusion

221

exercised only a very deficient monopoly of violence. The weak distribution of public goods—above all generalized security—resulted from the concoction between state infrastructure and patronage systems during the genesis of the “two-party state”. Regional caudillos built local power strongholds by tying a fol­ lowing to their domus and coordinated with each other through their respective party affiliation, Liberal and Conservative Party. Two features reflect the cliente­ listic disposition of the Colombian polity. Firstly, parties were cross-sectional in leadership and following. Socio-economic elites dominated the leadership in both parties, while lower classes populated the followership, reflecting the party mem­ bership of their patron. Secondly, the normative expectations, which in the con­ text of political institutions that left very little space for meaningful socio-political change must result in disappointment, were restrained through a specific legal discourse in the interpretation of constitutional stipulations. It became known as Viejo Derecho, validating a legalistic formalism as supposed to rights protection. These base-line conditions served as comparative parameters and helped to identify the similarities around the two most violent and destructive instances of internal conflict in Colombia’s 20th century history: the Thousand Days’ War and La Violencia. In both instances, the original crises that led to violent bloodshed essentially consisted of the inability of elites to compromise and strike deals that satisfied the entire range of party membership and leadership. The fighting even­ tually exhausted without producing a clear victor. The destruction signaled to elites that the original political arrangement was insufficient in containing clashes of interest and identity between them. In both cases, an elite-driven, and con­ trolled, process of institutional reform eventually established a new arrangement that guaranteed political participation for each party. Above all, reforms in 1910 and 1957 conserved forms of clientelism by placing them on a more stable formal institutional setting without fundamentally upsetting the underlying socio-political interactions. What was lacking was an institutional reform that opened egalitarian paths for political participation and a more equitable distribution of socio-eco­ nomic growth. The 1991 Constitution fundamentally differed from these previous instances of institutional engineering. First, the original crisis preceding the juncture resulted from profound insecurity that laid open what I call a discursive cleavage between the normative claim of a democracy (to uphold channels of political participation), and the political reality of a system besieged by violence and therefore incapable of providing those channels of political participation. Importantly, neither theory nor prior conditions can convincingly explain why students decided to act at that par­ ticular moment. It was the collective grief felt after the assassination of a talented political leader that led to spontaneous collective action. The initial proclamation at the silent march initiated a process of communicative action, during which stu­ dents first debated in small settings in Bogotá’s universities, then joined ranks with students from the entire country, and ultimately started to publicly organize a push for foundational constitutional reform. Thus, echoing Robert Frost’s cross­ road analogy, the movement to implement the constitutional reform took the extra-institutional path rather than the road through Colombia’s formal

222 Conclusion institutions, and it was therefore not a coincidence that it went the furthest in altering the formal political framework. The 1991 constituent process was not only unique because it evolved from a profound normative cleavage and unprecedented crisis moment that defied pre­ dictable patterns, but also because the actual juncture—the constituent assembly in 1991—followed patterns without precedent in Colombia’s history. Its compo­ sition featured a number of actors that had been historically excluded from poli­ tics: the leftist party movement founded from the M-19 guerrillas and representatives of indigenous groups. In contrast to negotiations during the previous instances of institutional engineering, the Conservative–Liberal hegemony of formal public affairs was broken, and institutional engineering attained an unabashedly public virtue as citizens could get involved at various stages of the process. The product was astounding. The assembly created a very progressive political charter that impressed with a number of constitutional novelties. It is without doubt that the immediate legacy of the 1991 constitutional trans­ formations was profound, but also ambiguous. The clientelistic relations engulfing Colombia’s formal institutions slowly moved from the inside to the outside of the traditional two parties, taking the party system as it was known down with it (Gutiérrez Sanín 2007). Politicians became electoral entrepreneurs, who (particu­ larly in rural areas) deliberately aligned with forces at the margin of legality to reap electoral triumphs. Those armed groups at the margin of legality benefited by gaining the aura of legitimacy required to shield them from legal prosecution. In the middle of these transformations stands the shift towards an electoral system that was intended to boost the national circumscription of the Senate. Contrary to expectations, regional politicians succeeded in pooling regional votes and sought support from armed campaigners that organized the sufficient electoral threshold in areas under their control. The most profound changes concerned the judicial system. Two new formal bodies were introduced (the Constitutional Court and the Superior Council of the Judiciary), three different types of rights introduced to the level of constitutional norm, and a novel mechanism implemented to enforce rights. These novelties reflected students’ original demands and grievances incited by the human rights situation in Colombia. From the abundance of extrajudicial killings perpetrated against political actors, human rights activists, and union leaders, it follows natu­ rally to situate rights more centrally in the constitutional framework and imple­ ment mechanisms that aim at their protection. I argued that the implemented changes best reflect the movement’s agency. In the end, it was readily apparent that the 1991 Constitution—its origin, creation, and implementation—was unprecedented in Colombia’s history and it is therefore fair to conclude that it amounts to a critical juncture in Colombia’s history. Most of all, it is key to observe that a genuinely communicative process produced a very ambitious Constitution that consecrated far-reaching rights guarantees. It is for these reasons that the 1991 constitutional moment was an instance of discursive institution-making. The process had qualities that espoused transparency in its creation and inclusive participation in the actors that

Conclusion

223

ultimately decided on the norms. It was a public process. The ultimately pro­ duced neo-constitutionalist political charter was therefore not a coincidence but a reflection of its discursive origin. As a Constitutional Court judge said: “the philosophical origin lies in the discourse that accompanied the promulgation of the Constitution”.1 Another magistrate from the 2010 Court made the causal relation between process and constitution explicit, and it is therefore worthwhile to cite in detail: The 1991 Constitution created a radical change in the juridical culture of the country, [because] the process in 1991 was much more inclusive, participant, and less presidential. The 1991 constitution changed, democratically, the constitution from 1886. It was a democratic pass to consolidate the demo­ cratic regime. It is important that this constitution was a peace pact, that it was pluralist, that included guerrilla groups, indigenous groups, and AfroColombian groups. Before, the constitution was the constitution of the winners.2 The profundity of changes in the moment of constitutional genesis naturally leads to exploring post-genesis trajectories. After all, the new Constitution not only included far-reaching rights promises; its new actor, the Constitutional Court, showed a high propensity to enforce rights in the following years and embraced a jurisprudence that diverged from the formalism associated with the old Constitu­ tion. Historical institutionalists have theorized, and shown, that path dependency follows critical junctures. The rights enforcing mechanism of the new Constitu­ tion, the tutela, provided a well-suited testing ground.

Institutional Learning in Judicial Institutions The third chapter defined path dependence as the concept that initial decisions in one direction induce further movements in that same direction later on in the institutional trajectory. To get to the ground of post-genesis evolution of legal institutions in Colombia, the analysis turned to the question of whether pathdependent logic can be utilized to portray developments that have strong norma­ tive connotations. The judicial branch was the branch that left the 1991 con­ stituent process most profoundly refurbished, and the tutela was, without a doubt, one of its most important institutional novelties. A priori considerations do not exclude judicial institutions from path-dependent analysis. After all, contingency in the moment of creation as well as the importance of precedents in legal decisions invoke characteristics of path dependency. Nevertheless, the novel part here was that focusing on judicial institutions that are tasked, among other things, with protecting rights and human dignity provides a level of analysis that must take argumentation about constitutional values into consideration. The conflict at the center of this chapter was indeed a complex affair, because it involved several levels of analysis. The so-called choque de trenes, or train crash in Spanish, evolved around the question whether decisions by high courts are open

224 Conclusion to rights review by the Constitutional Court (their jurisdictions are crashing). Addi­ tionally, since the new Constitution stipulates that Members of Congress are investi­ gated by the Supreme Court for criminal allegations and do not enjoy parliamentary immunity, the distributive dynamic between the courts can be amplified by the interest of political actors in the legislature and executive. All of this is embedded in a highly normative and value-laden discourse that invokes the foundational meaning of the Constitution, since the tutela is a central mechanism in the 1991 charter to pro­ tect human rights and dignity. From that followed three critical elements that make this case valuable for institutional theory: 1) the proliferation of new courts and con­ tradictions between normative principles in the new Constitution result in a contest over jurisdiction between the Constitutional and Supreme Court; 2) the legal func­ tions assigned to each branch of government in Colombia’s checks and balances system turns the fight over jurisdiction between the high courts into a political ques­ tion that involves majoritarian institutions; and 3) the affixture of rights to the standing of tutela complaints implicates the meaning of foundational norms of the new Constitution and their effects on constitutional adjudication. Engaging the historical institutionalist literature on path dependence and ela­ borating on key points of an ideal-type definition of the concept, I developed four guiding questions for the investigation: � �

�

�

Did the jurisprudence regarding the tutela follow a mechanical and linear line from its inception in the 1991 Constitution? Did earlier decisions matter more than later decisions, and if so, did they forego the possibility to decide differently in later questions on the same merit, or do earlier decisions open new possibilities for later decisions on the same merit? If the Constitutional Court changed or modified its jurisprudence, did the modification follow linearly from earlier decisions, or did it base its change in principled explanation engaging a legal argumentation evolving around legal values and shifts in legal context? If there was a clear break, did that break in turn result in a logically following pattern, indicating reactive path dependence?

These questions led to designing the inquiry as a longitudinal study, focusing on two political scandals evolving around the criminalization of Colombia’s Congress and the strategies employed by Members of Congress to avoid prosecution (pro­ ceso 8000 and parapolítica). The analysis focused on how the courts interacted as the political drama played out. In the course of the proceso 8000, Members of Congress were sued for their vote of absolution in the investigations against Pre­ sident Ernesto Samper. During the Uribe years in the 2000s, Members of Con­ gress were accused of having cultivated close relations with paramilitaries associated with the drug trade for electoral purposes. In both cases, the investiga­ tions by the Supreme Court became the subject of tutela complaints submitted for revision with the Constitutional Court, thereby placing the conflicts in the realm of rights guarantees.

Conclusion

225

In the final outcome, the investigation showed that neither the institutional design nor normative positions of the most important actors involved shifted. The courts held on to their jurisprudential positions in the choque de trenes, maintain­ ing that their decisions are not open to rights review (Supreme Court), or that any factual decision by any public institution is subject to rights reviews (Constitu­ tional Court). The functions of the courts remained the same, and the tutela itself was not regulated by legislation. The only decree issued in 2000 minimally speci­ fied the order of review. Thus, the basic disagreement at the root of the so-called choque de trenes remained until the end of the Uribe administration in 2010. Even though the generic disagreement over the precise application of the tutela did not subside, the political scandals developed very differently—politically and juridically. In the proceso 8000 only a few Members of Congress were eventually sentenced for their collusion with drug traffickers, while during the parapolítica 102 Members of Congress were investigated between 2006 and 2010. By the same token, the political consequences also varied. In the proceso 8000, political pressure and exposure to the scandal subsided with the Constitutional Court’s decision to uphold the inviolability of legislators. That decision effectively termi­ nated criminal investigations and neither President nor Congress were held accountable for their trespassing. Parapolítica investigations, on the other hand, were not halted by an unfavorable Constitutional Court decision. Inquiries con­ tinued and, in their course, overcame several strategic obstructions legislators uti­ lized in the attempt to prolong the process and evade prosecutions. In the end, thanks to the activism of the Supreme Court, the numbers of investigations and verdicts greatly outnumbered the cases concluded in the proceso 8000, and, toge­ ther, the courts exposed fundamental issues of democratic representation and governance by disclosing the symbiotic relations between Congresistas and armed groups at the margin of legality. President Uribe’s ability to maneuver in the political arena, too, was affected by the courts’ activism. He was forced to spend a lot of political capital to maintain his ability to dominate the political scene and fight off the negative public discourse. The diverging trajectory between the two scandals highlights the importance of legal facts on the development of political scandals and consequently on institu­ tional development. Legal facts are those “facts” produced in a legal process. They have deeper institutional impact than other socially produced “facts”. We saw that even in the proceso 8000, without legal investigations and independent prosecu­ tions there would not have been a “proceso 8000, nor a trial against the president, nor a political crisis” (Uprimny 1996, 120). The increasingly hostile public debate evolved in lock-step with new revelations that resulted from the legal processes against Samper’s campaign managers and Members of Congress. Similarly, the most potent revelations of the parapolítica scandal were those revealed or con­ firmed by judicial investigations. While the initial revelations about potential paramilitary infiltrations of the political class—above all the declarations by the AUC commanders themselves—are significant, the scandal received a much more explosive dynamic, once these facts were confirmed and further investigated by judicial bodies. Thus, legal facts are crucial for constitutional meaning-making.

226 Conclusion Finally, the centrality of legal facts for the continuation of developments required us to specify what caused the investigations to continue in one case (parapolítica), and be abruptly discontinued in the first case (proceso 8000). To find satisfactory answers to these questions, I turned to the normative develop­ ment in each political affair. Here we could identify a “united front” between the Supreme Court and the Constitutional Court during the parapolítica affair as the key difference to the proceso 8000. My interviewees repeatedly argued that the 1999 tutela decision (SU-047/99) in favor of Viviane Morales fundamentally obstructed the Supreme Court’s work to further investigate criminal behavior. Similarly, the same subjects held that during the parapolítica affair, the Constitu­ tional Court did not interfere with the Supreme Court’s investigation and often backed their interpretation of normative clauses. During the parapolítica scandal, Members of Congress repeatedly attempted to avoid investigations and utilized four strategies: 1) invoke the transitional justice jurisdiction of the Justice and Peace legislation; 2) submit tutelas on the grounds of their special rights as par­ liamentarians to freely opine without the threat of coercion; 3) claim due process rights violations resultant from the constitutional “privilege” to be criminally investigated by the highest court without the chance of appeal; and 4) resign from their seat and thereby also renounce the congressional investiture that involves that investigations are directly passed to the Supreme Court (hereby hoping to not only avoid the reach of the Supreme Court, but also be submitted to the jur­ isdiction of regional courts where they could more easily exert influence). Resigning from the congressional investiture proved to work only as long as the Supreme Court itself decided to refrain from investigations. Once it changed its jurisprudence, and the Constitutional Court followed suit in validating the other apex court’s jurisprudence, investigations continued. The final part of Chapter 3 coalesced the findings, in conjunction with the conceptual vocabulary of institutional (path-dependent) change, into three theses of path dependence and legal argumentation: � �

�

Once a decision is taken, the cost of reversal increases, but not exponentially so. Rather, reversals require good explanations rooted in precedent and change of context. Earlier decisions do matter more than later decisions since discourse (in par­ ticular legal discourse) builds on logical coherence. However, earlier decisions might also open new questions further down the road that can significantly alter the path. There cannot be a practical point of no return because all decisions are theo­ retically forever open to contestation.

Historical institutionalists have made advances towards including compliance as a variable for institutional change to provide a holistic picture that can also incor­ porate small and incremental changes. As Mahoney and Thelen maintain, written norms can never be attuned to cover the full range of possibilities of real-life situations. They always require interpretation. Thus, “the need to enforce

Conclusion

227

institutions carries its own dynamic of potential change”, because “of the inter­ pretation and implementation of these rules” (2009, 10). In my analysis of the choque de trenes, I ultimately argue that the outcome in the parapolítica affair was a result of institutional learning that works in tandem with reputation-seeking actors, who yearn for the recognition of their peers. It was simply not argu­ mentatively feasible to defend parliamentary immunity as a bulwark for free deliberation, when Members of Congress utilized the norm for impunity. I inter­ preted the difference in outcome between the proceso 8000 and parapolítica scan­ dal as evidence of a process institutional learning in the courts, in which norms are applied to result in a normatively more appealing outcome. Evidently, this result begs the question of what determines a normatively appealing outcome. For that end, Chapter 4 explored the internal working of the Constitutional Court and how they establish what becomes a norm.

The Substitution Doctrine and the Deliberative Judge The preceding chapters placed the 2005 and 2010 Constitutional Court decisions in the historical and institutional context of Colombia’s turbulent history, which was a long, but nevertheless necessary, route to take. Chapter 4 turned to the contention over Uribe’s re-election bids. These decisions conspicuously bring out the inherent tensions in democratic governance between majoritarian claims to democratic legitimacy and the constitutionalist constraints placed upon the exer­ cise of power. The principal task was to explain the apparent increase in judicial authority and autonomy. This not only required specifying the observable impli­ cations arising from the strategic and discursive model of the constitutional judge, but also made the extrapolation of the constitutional/majoritarian divide indis­ pensable. Only together can we fully appreciate the importance of the imple­ mentation of legal reasoning. This chapter took the separation of powers model and the strategic judge as its starting point, because it can (and has been) applied to contexts outside the United States and generated the most recent contributions to the study of judicial behavior. The state of the literature provides three central hypotheses and impli­ cations applicable to this case: 1) the political/electoral uncertainty argument: courts tend to act more independently when imminent regime change or execu­ tive-legislature misalignment veil the certainty of actors in the other branches of government; 2) the prudent strategizer argument: courts act akin to a prudent strategizer who upholds principles but defects on the merits in prominent cases, and then applies the principles in less prominent cases; and 3) the constancy of preferences: preferences are informed by political socialization and remain constant during deliberations. I juxtaposed the strategic with the deliberative judge, who engages in collective decision-making processes on questions of public authority, in which participants remain open “to transform their preferences in the light of well-articulated arguments” (Hübner Mendes 2013, 18). To test these hypotheses, Chapter 4 elaborated the peculiar and unique role Álvaro Uribe played in Colombian politics, and how the cases evolved around

228 Conclusion democracy’s inherent tension between majoritarian claims of legitimacy and the constitutionalist constraints placed upon the exercise of popular power. Colombia under Uribe—particularly in the second term—displayed trends that O’Donnell first identified as part of a new animal in comparative politics: delegative democ­ racy. Uribe’s popular appeal made him an atypical political leader in Colombia’s history, commanding unmatched de jure and de facto power. He displayed a char­ isma that united a large part of the population behind him and with his popularity disciplined political rivals into obedience. Importantly for the purposes of this study, he had translated his power and status into a healthy sense of mission, aiming at fundamentally transforming and reforming the political landscape in Colombia. Uribistas legitimized the reform program by appealing to notions of natural leadership, conservative values, and communitarian social bonds between rulers and ruled. As a delegative President, he viewed and presented himself as the poli­ tical expression of the entire Colombian nation. His political program, Democratic Security, imagined a polity of tightly knit relations between rulers and ruled that, backed by trust in their capacity to rule, ultimately place leaders back on their pedestal. From such communitarian notions of political legitimacy, he and his associates developed a legitimation to reform the Colombian state rooted in the so-called estado de opinión (state of opinion). With this doctrine, Uribe claimed that only popular elections and opinion are a legitimate control of political power and not horizontal controls that lack such grounding in popular legitimacy. The Uribe years not only stand out for the personal charisma the President himself displayed, but also for the transformations of the party system, which first enabled his rise to power, helped him to cement his grip on Colombia’s institu­ tions, and then displayed a criminalization that made the co-optation of Colom­ bia’s Congress readily visible. Uribe’s election to the presidency was only possible because he ran as a renegade candidate, listed himself as an independent, and presented his policies as distinct from the traditional two-party political game that had dominated Colombia for the previous 150 years. As shown, the 2002 electoral cycle produced a partial implosion of the party system that rendered it unrecog­ nizable due to denationalization into many little micro electoral entrepreneurs that left voters with few coherent and identifiable national policy plans (Pizarro Leon­ gómez 2002). Effectively, this created a very diverse, but also easily co-opted, Congress that responded to the special interest of illegal forces in the country (López Hernández 2010). In other words, the legislature as the most central horizontal control institution of liberal democracy was effectively debilitated. Together, the ineffectiveness of Congress and Uribe’s propensity to rule as he saw fit created a tense situation that placed the democratic order under strain. (Neo)-constitutionalism does not per se reject the notion that the only legit­ imate political authority emanates from the people, but generically stipulates that power is limited by written or unwritten constitutional norms. Fundamentally, while the majoritarian model of democracy holds that “power is at the disposition of a political will as a means for achieving collective goals, regardless of the nor­ mative constraints that authorize it”, constitutionalism reintroduces the

Conclusion

229

importance of the normative content of law to legitimately speak of the legal authorization of power (Habermas 1995, 777; Rawls 1993). Firstly, this implies an adherence to formal rules and procedures in the formulation of positive rules. Secondly, neo-constitutionalism goes beyond the mere legalism of majoritarian rule, and further implies the inclusion of rights—not least, minority rights—to guarantee that the majority’s will is not compromised by social and political structures of power. The normative basis for the constitutionalist paradigm is not the General Will, or other collective entities such as nation and race, but a plurality of individuals that recognize each other as equals. The Constitutional Court itself expressed this sentiment quite poetically in its decision to withhold a third pre­ sidential term: The notion of people that accompanies the conception of constitutional lib­ eral democracy cannot ignore the notion of pluralism, which involves the coexistence of different ideas, races, genders, backgrounds, religions, institu­ tions or social groups. Such a heterogeneous people accepts that all power must have limits and therefore agrees, as a sovereign people, and in accor­ dance with the democratic model, to show self-restraint and establish channels through which to express all its diversity. Therefore, in contemporary states the voice of the people cannot be appropriated by one group of citizens, even if it is a majority, but it arises from the procedures that guarantee a manifes­ tation of this plurality. (C-141/10; translation by the author) Uribe’s motto to create social cohesion through trustful relations between rulers and ruled is the majoritarian caveat of Democratic Security, and the Constitutional Court’s assertion that society consists of a plurality of wills is the counter-major­ itarian caveat implicit in Democratic Legality. In the final analysis, we could not refer to conventional explanations from the field of judicial behavior to clarify why Democratic Legality prevailed over Democratic Security. Even though Uribe’s coalition lost a critical partner in Congress with Cambio Radical, the perception in the Court was not that this made the decision any lighter, nor did magistrates consider the political weight behind Uribe to be much less significant than in 2005. After all, his popularity ratings were as high as in the first term and even peaked in 2008. It was readily apparent that the President was still hugely popular with Colombians and a large majority of them favored another term in office. Ultimately, the Court’s decision is difficult to bring in line with the narrative of a prudent strategizer. Contrary to its predictions, every important development of the substitution doctrine coincided with an extremely important case for the executive; not least the two re-election decisions were essential cases in the trajectory of the theory, which in 2010 ended all Uribista hopes of another term. I argue that the institutionalization of discourse did indeed have an important effect on the outcome in 2010. After the Court had come to the conclusion that procedural norms were violated in the formation of the law that called for a

230 Conclusion referendum, there was the apparent suggestion that the Sala Plena should refrain from a competence test in favor of a more united decision. However, the eventual majority insisted on the complete jurisprudential tests applied to the reform. This then resulted in the verdict that the reform not only incurred formal defects in its creation, but also competence defects: Congress was acting ultra vires. The insistence on the completeness of procedure speaks to the importance of the institutionalization of deliberation and the way magistrates identify with dis­ cursive practices. Legal theory has long embraced that positive law always requires interpretation, and judges always have some inherent biases rooted in their socia­ lization. To counter the alleged bias that follows from the inherent lack of neu­ trality, the procedures of implementing legal reasoning become especially important. It has three imperatives: 1) prevent a rationally unmotivated termina­ tion of argumentation; 2) secure both freedom in the choice of topics and inclu­ sion of the best information and reasons through universal and equal access to, as well as equal and symmetrical participation in, argumentation; and (3) exclude every kind of coercion—whether originating outside the process of reaching understanding or within it—other than that of the better argument, so that all motives except that of the cooperative search for truth are neutralized (Habermas 1996, 230; Alexy 1989, 136–137). In order for law to matter (la ley cuenta), the Court insists on the tranquility of a structured deliberation process. Aware of magistrates’ specific role to uphold the rule of law without direct democratic legitimacy, while socialization makes political neutrality devoid of convictions implausible, the Constitutional Court functioned with a number of formal and informal rules to protect legal reasoning from the encroachment of political rea­ soning. Rules aim to isolate the Sala Plena from the rhetorical effects of public spectacles and guard argumentation from functional coalitions between judges. Individual positions were never revealed to the public prior to the conclusion of deliberation and judges did not engage in quid pro quo trades for votes in the chamber. Subjects viewed the Sala Plena as the coliseum of legal arbitration and its integrity was identical to the integrity of the institution as such. Guarding the deliberation from undue influences had the important effect that judges can and do change their opinion. They have voted against their stated ideological pre­ ferences if the force of the better argument required them to. It is this force of the better argument that mattered in the 2010 decision. The unfolding deliberation process showed that Congress had acted outside of its competence when it legis­ lated the constitutional reform referendum.

The Deliberative Judge, Democratic Legality, and Discursive Institutionalism These conclusions have ramifications for institutional theory. They call for the incor­ poration of discursive variables to paint a more holistic picture of institutions and their transformations over time—especially for the study of legal institutions. Chapter 2 highlighted the centrality of communicative action for the creation of a movement that eventually succeeded in the implementation of a constituent assembly. Chapter 3

Conclusion

231

unearthed the legal fact production of courts in post-genesis developments that decisively factored into the process of defining institutional boundaries and com­ pliance with institutional stipulations. The analysis identified this as a process of institutional learning. Chapter 4 problematized decision-making procedures inside courts by taking the notion seriously that judges are practicing law. The important factor here is that practicing law is tightly connected with deliberative practices of truth finding and constitutional meaning-making. The chapters have iterated that institutionalism has had difficulties accounting for change, unless it comes in an abrupt and impulsive manner that dislodges stable relations. As Mahoney and Thelen lament, “despite many other differences, nearly all definitions of institutions treat them as relatively enduring features of political and social life (rules, norms, procedures) that structure behavior and that cannot be changed easily or instantaneously” (2009, 4; italics in the original). As they lay out, this bias towards continuity at the expense of smaller shifts is present in all three of the classical expressions of neo-institutionalism: historical, socio­ logical, and rational choice institutionalism (6). The contributions of those strands of political analysis are not in doubt, yet the literature has started to focus on the more gradual forms of change, and has begun to build a canon that introduces smaller shifts in rule and policy frameworks. At the most general level, this book is a contribution to this growing body of literature that addresses subtle change. Discursive institutionalism makes the bold claim to better understand agency and gradual change by focusing on the versatility of discourse and how ideas are transported. It theorizes how human agents—rather than actors—generate, inter­ pret, deliberate, apply ideas, and how those processes generate consequential meanings. Vivien Schmidt criticized that neo-institutionalists do not fully appreci­ ate the centrality of ideas, even though politicians’ main task in democracies is to communicate ideas and thereby legitimate their actions. She explains that the tra­ ditional institutionalist schools still have very little to say about how “ideas go from thought to word, to deed, that is, how ideas are conveyed, adopted, and adapted” (2008, 309). She proposes a shift to a deliberation-focused institution­ alism that centers on the volatile relationship between word and deed that has puzzled philosophers for centuries (304). The foundational condition for discursively engaging actors is the presence of some form of public. Actors engage in discursive practice, if there is an audience that listens and expects justification for action. Schmidt is correct to insinuate that this is the case for public policies in all democracies (305). Importantly, this is also true for strategic actors in public spaces, because even strategic actors with fixed preferences who engage in rhetorical action must convince and persuade an audi­ ence with argumentative rationality (Risse 2002; Habermas 1996). As a con­ sequence, this argumentative logic of appropriateness also has validity for the discursive interaction in political settings. The applicability of discourse for the analysis of human interaction falls apart, however, when discourse and publicness is existentially threatened or non-existent. As we saw in the history of paramilitarism in Colombia, until the demobilization process with the AUC forced various actors to take public stances, relations between Members of Congress and paramilitaries

232 Conclusion were very stable because they largely existed in the shadows of the public. At the same time, they also stifled any form of locally engaged, and autonomously exer­ cised, civic action—equilibrated relations of locally privatized violence asphyxiated political interaction (Boesten 2014). Schmidt concisely summarizes four points that set discursive institutionalists apart from their neo-classical siblings: 1) regardless of definitional distinctions, they take ideas and discourse seriously; 2) they set these ideas and discourses in institutional contexts; and 3) ideas carry “meaning” and discourse follows a “logic of commu­ nication”; and 4) discursive institutionalism induces a dynamic into our under­ standing of institutional change that the other three institutionalisms, “equilibrium­ focused” as they are, cannot comprehend, because the locked-in equilibria often pose as insurmountable (2008, 304). Consequently, she argues, the increased versatility of discourse as a concept helps to simultaneously “indicate the ideas represented in the discourse and the interactive processes by which ideas are conveyed” (309). When we look back at the chapters in this book and the results they produced, we saw that meaning-making repeatedly appears in the (re)-structuration of insti­ tutions: first at the broadest level when ideas were turned into text in the con­ stituent process, and then when text is translated into action. In the second chapter, the analysis of the processes leading up to the constituent assembly in 1991 and the debates in the halls of the Luis Angel Arango Library in Bogotá disclosed the centrality of agency in collective action. Even under extremely diffi­ cult and dangerous conditions of a worsening civil war, the people managed to come together for free and open debates, installing a process that produced a formidably progressive document. The student movement owed most of its appearance to the spontaneity of human action that arises when a collective comes together for a common cause (in this case, it was the collective grief over the assassination of a promising political leader). Moreover, throughout the debate, ethical and collective reasoning for the search of a common institutional framework—and not aligned interest or structure—tied the assembly of actors together. The question of constitution-making has become a very prominent topic of com­ parative politics—particularly in the Latin American context. Ginsburg writes that “Latin America is something of a constitutional graveyard, in which formal texts have been replaced frequently over the past two centuries” (2014). Consequently, an increasing number of scholars pay attention to various facets around the issues of constitution-making and the longevity of constitutions. Negretto produced an out­ standing contribution to the study of those constitutional moments that structure politics anew (2013). He understands the volatility and frequency of constitutional change in Latin America as a chance to properly understand the politics of writing a charter and the forces at play at these junctures that until now remained behind the veil of contingency. In his carefully designed study, he shows that similar factors such as actors’ resources and expectations about future political development affect the selection of institutional setups in such areas as the electoral system and presidential powers. Intriguingly, what topics are selected for debates in constituent processes, and how judicial institutions are constructed, were outside the scope of Negretto’s study.

Conclusion

233

This book made a first effort at getting at these questions by positing that the initial crisis provided a discursive context that placed the human rights situation in Colombia in the 1980s at the center of public debate. It suggests that it was not a coincidence that rights and judicial institutions became a central focus of the constituent process, given the normative crisis that preceded the critical juncture. Yet, Negretto is correct to call for more research and a broader scope to study topic selection in constituent bodies. The task is then to increase causal leverage by constructing comparative studies across cases to understand under what con­ ditions rights become the central grievance of constitution-making. A positivistic study could take the result of implementing a Kelsenian high court model as sup­ posed to the American model as an indication that the unification of rights under one jurisdiction was a primary concern of constituents, and then work back to identify conditions at the beginning of such processes. Such a categorization can be quickly filled with data and provide a large set of observations that is amenable to broader investigations. Moreover, to identify actors and their arguments in a constituent process, rights can be disaggregated by their type. As seen the 1991 Constitution not only included private and civic rights, but also social and collec­ tive (cultural) rights. This, too, is amenable to cross-case quantification. Above all, however, this book found that the discursive context and the activism of agents must be closely inspected and incorporated into the analysis. Therefore, future research must make an effort to combine classical approaches (be they quantitative or qualitative) that identify actors, their resources, and the implementation of constitutional negotiations with discursive analyses, because constitutions reflect a societal consensus over the commonly shared values and identities that give future political interaction meaning and significance. These values and identities are most often consecrated in the expression of rights. In the chapter analyzing the jurisprudence of central features of the new Con­ stitution’s rights framework, path dependence proved too static to understand the rootedness of legal argumentation in practical discourse. Thelen already lamented in 1999 that path dependence is too contingent. The image of initial junctures that produce path dependence is unrealistic, because they appear to resemble a blank slate that is rarely present in political interaction. In addition, the mechan­ isms of reproduction become too mechanical (1999, 385). Indeed, we found that the mechanical reproduction of path dependence becomes problematic as a descriptor of legal institutions that emulate a rationality that cannot be equated with “certainty”. Such a reification of discourse “must renounce the idea of a theory of legal argumentation” (Alexy 1989, 293). Classical expressions of path dependence have placed too much emphasis on stasis, pertaining from lock-in trajectories and equilibria-inducing mechanisms. Legal argumentation, on the other side, reopens previous decisions and tests if former arguments still have validity under current context conditions; or, conversely, if the parameters of an earlier decision need to be adjusted. There is an evident discursive nature in these procedures that enables judges to retain their agency and reopen previously closed processes.

234 Conclusion The incremental change model emphatically includes compliance as a crucial variable in explanations of institutional development. The tutela contra sentencias, the choque de trenes between Colombia’s high court, and criminal investigations of Members of Congress—probed in depth in Chapter 3—also politicized com­ pliance. In agreement with the arguments offered by Mahoney and Thelen, actors brought resources to bear when the question of compliance with the Supreme Court’s decisions arose during the proceso 8000 as well as the parapolítica affair. The ability to reopen past decisions, which I identified in the path dependence of judicial institutions, therefore, essentially evolves around this caveat of formal institutions. My contention here is that the incremental change model in fact converges with some stipulations from discursive institutionalism. As Mahoney and Thelen themselves explain, compliance as a variable is crucially important for explaining stability and change, because of the ambiguous origin of institutions and the resultant “degree of openness in the implementation of these rules” (2009, 10). Implementing and enforcing institutions always involves “interpreta­ tion, debate, and contestation” (2009; italics added). In other words, the political action often takes place in a discursive sphere, in which actors engage over the specific meanings of clauses and norms, while general and normatively appealing principles remain relatively static (broad concepts like justice, democracy, the separation of powers, or effective representation in the legislature). An intriguing result of the analysis of choque de trenes is that courts face dis­ tributive dilemmas—having to defend their own jurisdictions—but at the same time have to justify their positions and actions consistently, because they operate without direct democratic legitimation. Moreover, the judiciary and courts are institutions that create new institutions and institutional boundaries. In a separa­ tion of powers model, they have the monopoly over interpreting texts and trans­ lating the written word into speech and action (Cameron 2013). In other words, their task always requires something akin to being aware of Kant’s categorical imperative since their actions always turn into broader laws. Together, this sets incentives to aim for a consistency in argumentation and carefully consider the consequences of one’s decisions in the broader constitutional context. That Members of Congress should not associate with criminal elements and, to the best of their capacities, represent the interest of their constituents fairly is a relatively safe assumption about the role of legislators in constitutional democracies. How­ ever, from the discussion of the proceso 8000 and the parapolítica affairs, we saw that judges became aware that interpreting parliamentary immunity expansively would end in impunity and prestige loss in the eyes of their peers. What I term institutional learning then fits well into the categorization of incremental change, since it evolves from within and does not require abrupt and wholesome change. There were more facets of the processes outlined in the chapter on post-genesis evolution of institutions that had unequivocally discursive caveats. As seen, the capacity to re-evaluate norms in a new context was instrumental for the produc­ tion of legal facts that developed a dynamic in the public discourse impossible for Uribe to contain—despite his broad powers and extreme popularity. When enga­ ging with critics from NGOs—be they domestic or international—Uribe painted

Conclusion

235

those that railed against his policies as friends of the guerrilla and enemies of the Colombian people. Yet, the courts could not as easily be placed on the other side of a friend/foe distinction. After all, Colombians pride themselves as one of the oldest democracies of the continent, and the courts in general embody that tradi­ tion of a relatively stable and democratic country. Uribe’s public defamation strategy used against the Supreme Court continuously insinuated that the evidence was contaminated by the paramilitaries themselves, and that judges were acting out of personal or political spite against the President. Noticeably, he did not suggest the Supreme Court to be a FARC ally. In the end, neither strategy suc­ ceeded in discrediting the Supreme Court and its investigations, forcing the Pre­ sident to employ illegal tactics to silence the magistrates. Evidently, the continued production of legal facts posed a real danger to the President’s power. The continued production of legal facts also played an important role in the processes of giving meaning to constitutional clauses. The absence of timely legal facts about the vote in the committee stage of the first re-election reform that involved Yidis Medina and Teodolindo Avendaño prevented the Constitutional Court from disallowing the first re-election reform on procedural grounds. On the other hand, when legally affirmed knowledge about relations between paramilitaries and Members of Congress was widely available, and cover-up campaigns could be related to the presidential office (as with the wiretapping by DAS of the Supreme Court’s chambers), the production of legal facts factored into the crea­ tion of constitutional meaning. After all, the Constitutional Court’s decision cited the effects of presidentialism on the separation of powers. It is for these reasons that the importance of legal facts for political strategies can be a valuable tool for future, quantitative analyses of communication, because it suggests a category that sets itself apart from other discursively transported facts in social settings. The recognition of the importance of legal facts in this affair also contributes to Baki­ ner’s notion of endogenous growth of judicial power in the Supreme Court. The sources of that growth can well be described as the result of shifts in self-definition (including conditional civil society support that entrenches a court’s self-defini­ tion), and the snowballing effects of judicial decisions (2020, 603). What this book’s analysis shows—not least through the construction of a within case study— is that the process by which socially relevant facts are produced has an impact on judges’ self-definition. In other words, the formalization of allegations in the parapolítica process played a key part in the snowballing effects of judicial deci­ sions—including its ramifications on inter-branch relations. The practice of meaning-making became most evident in Chapter 4, which reconstructed decision-making inside the Sala Plena of the Constitutional Court. It is by now a highly acclaimed accomplishment of the Constitutional Court to have developed a novel approach to a question that constitutional theory in gen­ eral has had difficulties addressing: legally sanctioning constitutional reform. Going back to the dispute between Carl Schmitt (2008) and Hans Kelsen (1945), legal theorists have grappled with the distinction between constituent and constituted power, and what it means for the practice of politics. The Constitutional Court, with its jurisprudence delineating the difference between reforming and

236 Conclusion substituting the Constitution, broke into new territory that even legal theorists themselves have only begun to understand (Cajas-Sarria 2017; Roznai 2016; Dixon and Landau 2015; Bernal 2013). This fact by itself makes the Court a pioneer in giving meaning to complex constitutional issues that have relevance beyond the Colombian case. Still, the jurisprudential evolution of the substitution doctrine told us more about judicial behavior. The substitution test inquires if a reform touches on basic axiomatic principles of the Constitution, and if the reform unduly changes them beyond recognition. In other words, it is a dogmatic expression of the pro­ portionality principle (González-Bertomeu 2019; Bernal 2013). The principle of proportionality analyzes the legitimacy, suitability, necessity, and reasonability of a legal measure when it is engaged in judicial review of laws or constitutional reforms (Kumm 2012). The substitution doctrine enabled judges to properly incorporate the legal context and, for example, engage with the outcomes of the clash between Uribe and the Supreme Court. In that way, the application of the substitution doctrine effectively resulted in specifying the meaning of constitu­ tional clauses—namely the separation of powers. The end result not only validated the clause as a fundamental and axiomatic principle of the Constitution; it also had evident practical ramifications by affirming that it cannot be altered by simple legislation or referendum. Implementing three or more consecutive presidential terms requires a new constituent assembly. Moreover, the production of facts in the context of the parapolítica affair at least indirectly affected the creation of constitutional meaning in this final decision to block the referendum law, because it showed the effects of excessive presidentialism on the constitutional order. Ultimately, even ardent critics of the substitution doctrine accepted its validity in the context of such presidentialism.3 The strategic paradigm of the constitutional judge, whose decisions are rooted in policy preferences, was not sufficient to illuminate the final outcome. It could be interesting, however, to include reputation as a factor in utility maximization equations (Epstein and Knight 2013), because it tacitly incorporates an intersubjective agent. It is conceivable that judges act a certain way because they expect recognition by others for upholding a specific argumentation and reasoning. Arguably, this validates the practice of legal reasoning. Nevertheless, it leaves out important caveats. For once, seeing the answer to judges’ behavior in reputation seeking requires a specification of what actually would increase reputation, or tar­ nish it. In the context of legal decisions, specifying the “right” course of action involves solving complex constitutional questions that, more often than not, defy simple and foregone conclusions. It is precisely the strength of the conceptualiza­ tion of the deliberative judge that it is contingent on a discursive truth-finding process to come to a conclusion of what the right answer may be. The findings in my analysis suggest that practicing of law requires a collec­ tive effort and it is the debate itself amongst judges that motivates them to take specific decisions. Not only does the process itself constitute the sine qua non of coherent and democratic legal decisions, magistrates in other contexts, too, have stressed that they do not have a concise will or even conclusion of

Conclusion

237

the best way to resolve a constitutional question prior to deliberation, nor what can increase one’s reputation in such coliseums of debate (Dieter Grimm, cited in Hübner Mendes, 2013). Deliberation, not least its institutionalization into discursive structures biased toward the collaborative search for legal truths, defines the content of reputation in a collegial space. It also creates the volatility of preferences during legal decision-making processes. Therefore, I hold that legal argumentation itself is ontological of judicial behavior. The notion that procedures of argumentation are constituent of autonomous judicial decision-making adds a caveat for understanding how constitutional ideas find their way into adjudication and application. Ideational carriers (Nunes 2010) do play an important role, but we saw that procedures enable the creation of sui generis jurisprudential ideas, and ultimately facilitated their appli­ cation in a politically consequential decision. It is noteworthy that Epstein and Knight call for a renewal of the discipline of the study of judicial behavior by properly inserting law into the equation. Such an approach was in fact part of the methodology utilized in this investigation: closely inspecting the substantive content of legal decisions. They argue that scholarship should try to provide wholesome and generalizable answers to specify under what conditions what types of argumentation dominate in particular legal decisions. I concur that this would be an interesting strategy, because underlying the call for inspecting the reasoning of court decisions is the assumption that legal argu­ mentation itself is ontological of judicial behavior. Executing such a strategy in future research will undoubtedly be tedious, because the categorization of argu­ ments will require an intimate knowledge of the specific case. For example, ori­ ginalism, one of the categories suggested by Epstein and Knight, in the Colombian case can refer to a traditional understanding of basic legal categories that predate the current Constitution, or imply the novel categories inscribed in the 1991 Constitution that directly contradict the traditional categories of Colombian legalistic thought. It is therefore not straightforward to exactly define what originalism means in a specific case. Nevertheless, social science can make a real contribution by employing these strategies, because it gives credit to what legal scholars and many of my subjects have repeatedly affirmed: la ley cuenta! These results lead me to suggest a definition of institutions that is not meant to replace other definitions but highlight further dimensions of political inter­ action. Douglas North is right to hold that institutions are the “humanly constructed constraints that shape human interaction” (1990, 3), and so is Guillermo O’Donnell, who states that they are “regularized patterns of inter­ action that are known, practiced, and regularly accepted (if not necessarily normatively approved) by given social agents” (1994, 5). I want to add that, if we take the notions that they are humanly created and known by agents, we also need to consider that institutions are discursive structures. This study has shown that actors negotiate the meaning and significance of norms—rules in other words—with reference to a constitutional text and the intention of the constituents that drafted the Constitution in the first place. We saw that the text as basis for negotiation ensured at critical moments that the negotiating process is not simply an expression of the “survival of the fittest” that benefits

238 Conclusion the actor with the most profound resources. Texts and intention of con­ stituents entail that this process often is an interpretive process that requires actors to make arguments as to how text and intention apply in the given context. The discursive judge then must be a judge that values the soundness of argumentation over utility maximization considerations.

Notes 1 Interview with Constitutional Court judge, 3 May 2013. 2 Interview with Constitutional Court judge, 30 March 2013. 3 Interview with Constitutional Court judge, 8 March 2013.

References Alexy, Robert. 1989. A Theory of Legal Argumentation. The Theory of Rational Discourse as Theory of Legal Justification. Oxford: Oxford University Press. Archer, Ronald P. 1995. “Party Strength and Weakness in Colombia’s Besieged Democ­ racy”. In Building Democratic Institutions: Party Systems in Latin America, edited by Timothy Mainwaring, Scott Scully, 164–199. Stanford: Stanford University Press. Archer, Ronald P., and Matthew Soberg Shugart. 1997. “The Unrealized Potential of Pre­ sidential Dominance in Colombia”. In Presidentialism and Democracy in Latin America, edited by Scott Mainwaring and Matthew Soberg Shugart, 110–159. Cambridge: Cam­ bridge University Press. https://doi.org/10.1017/CBO9781139174800.004. Bakiner, Onur. 2020. “Endogenous Sources of Judicial Power: Parapolitics and the Supreme Court of Colombia”. Comparative Politics 52 (4): 603–624. https://doi.org/ 10.5129/001041520X15743890683814. Bejarano, Ana María, Helena Alviar, Felipe Botero, Gary Hoskin, and Mónica Pachón. 2010. “Colombia: ‘Democracia Amenazada’”. In Democracia En La Región Andina: Diversidad y Desafíos, edited by Maxwell A. Cameron and Juan Pablo Luna, 101–164. Lima: Instituto de Estudios Peruanos. https://doi.org/10.7440/2012.43. Bejarano, Ana María, and Eduardo Pizarro. 2005. “From ‘Restricted’ to ‘Besieged’: The Changing Nature of the Limits to Democracy in Colombia”. In The Third Wave of Democratization in Latin America, edited by Frances Hagopian and Scott P. Mainwar­ ing, 235–260. Cambridge: Cambridge University Press. https://doi.org/10.1017/ CBO9780511791116.009. Bernal, Carlos. 2013. “Unconstitutional Constitutional Amendments in the Case Study of Colombia: An Analysis of the Justification and Meaning of the Constitutional Replace­ ment Doctrine”. International Journal of Constitutional Law 11 (2): 339–357. https:// doi.org/10.1093/icon/mot007. Boesten, Jan. 2014. “The Generalization of Particularized Trust. Paramilitarism and Structures of Trust in Colombia”. Colombia Internacional 81 (August): 237–265. https://doi.org/ 10.7440/colombiaint81.2014.08. Cajas-Sarria, Mario Alberto. 2017. “Judicial Review of Constitutional Amendments in Colombia: A Political and Historical Perspective, 1955–2016”. The Theory and Practice of Legislation 5 (3): 245–275. https://doi.org/10.1080/20508840.2017.1407397. Cameron, Maxwell. 2013. Strong Constitutions: Social-Cognitive Origins of the Separation of Powers. Oxford: Oxford University Press. https://doi.org/10.1093/acprof:oso/ 9780199987443.001.0001.

Conclusion

239

Collier, David, and Ruth Collier. 2002. Shaping the Political Arena. Critical Junctures, the Labor Movement, and Regime Dynamics in Latin America. Notre Dame: University of Notre Dame Press. Collier, David, and Steven Levitsky. 1997. “Democracy with Adjectives: Conceptual Innovation in Comparative Research”. World Politics 49 (3): 430–451. Dixon, Rosalind, and David Landau. 2015. “Transnational Constitutionalism and a Lim­ ited Doctrine of Unconstitutional Constitutional Amendment”. International Journal of Constitutional Law 13 (3): 606–638. https://doi.org/10.1093/icon/mov039. Epstein, Lee, and Jack Knight. 2013. “Reconsidering Judicial Preferences”. Annual Review of Political Science 16 (1): 11–31. https://doi.org/10.1146/annurev-p olisci-032211-214229. Ginsburg, Tom. 2014. “Making Constitutions: Presidents, Parties and Institutional Choice in Latin America by Gabriel L. Negretto New York, Cambridge University Press, 2013. 296 pp. $95.00”. Political Science Quarterly 129 (4): 749–751. https://doi.org/10. 1002/polq.12266. González-Bertomeu, Juan F. 2019. “The Colombian Constitutional Court’s Doctrine on the Substitution of the Constitution”. In Constitutional Change and Transformation in Latin America, edited by Richard Albert, Carlos Bernal, and Juliano Zaiden Benvindo, 119–142. Oxford: Hart Publishing. https://doi.org/10.5040/9781509923533.ch-005. Graves, Robert. 1975. The Greek Myths:1. Harmondsworth: Penguin Books. Gutiérrez Sanín, Francisco. 2007. Lo Que El Viento Se Llevo: Democracia y Partidos Politicos En Colombia 1958–2006. Bogotá: Grupo Editorial Norma. Habermas, Jürgen. 1995. “Paradigms of Law”. Cardoso Law Review 17 (4–5): 771–784. Habermas, Jürgen. 1996. Between Facts and Norms: Contributions to a Discourse Theory of Law and Democracy. Cambridge, MA: MIT University Press. Hartlyn, Jonathan. 1988. The Politics of Coalition Rule in Colombia. Cambridge: Cambridge University Press. Hübner Mendes, Conrado. 2013. Constitutional Courts and Deliberative Democracy. Oxford: Oxford University Press. Kelsen, Hans. 1945. General Theory of Law and State. Cambridge, MA: Harvard University Press. Kline, Harvey. 1995. Colombia: Democracy under Assault. Boulder: Westview Press Inc. Kneip, Sascha. 2011. “Constitutional Courts as Democratic Actors and Promoters of the Rule of Law: Institutional Prerequisites and Normative Foundations”. Zeitschrift Für Vergleichende Politikwissenschaft 5 (1): 131–155. https://doi.org/10.1007/ s12286-011-0096-z. Kumm, Mattias. 2012. “Alexy’s Theory of Constitutional Rights and the Problem of Judicial Review”. In Institutionalized Reason: The Jurisprudence of Robert Alexy, 201–217. Oxford: Oxford University Press. https://doi.org/10.1093/acprof:oso/9780199582068.003.0009. López Hernández, Claudia. 2010. Y Refundaron La Patria … De Cómo Mafiosos y Políticos Reconfiguraron El Estado Colombiano. Bogotá: Debate. Mahoney, James, and Kathleen Thelen. 2009. Explaining Institutional Change: Ambiguity, Agency, and Power. Cambridge: Cambridge University Press. https://doi.org/10. 1017/CBO9780511806414. Negretto, Gabriel L. 2013. Making Constitutions. Making Constitutions: Presidents, Parties, and Institutional Choice in Latin America. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9781139207836. North, Douglas C. 1990. Institutions, Institutional Change and Economic Performance. Cambridge: Cambridge University Press.

240 Conclusion Nunes, Rodrigo. 2010. “Ideal Justice in Latin America: Interests, Ideas, and the Origins of Progressive Judicial Activism in Brazil and Colombia”. Unpublished Dissertation. Department of Government. University of Texas, Austin. O’Donnell, Guillermo A. 1994. “Delegative Democracy”. Journal of Democracy 5 (1): 55–69. Pizarro Leongómez, Eduardo. 2002. “La Atomización Partidista En Colombia: El Fenó­ meno de Las Micro-Empresas Electorales”. Working Paper of the Helen Kellogg Institute for International Studies. Notre Dame. Rawls, John. 1993. Political Liberalism. New York: Columbia University Press. https:// doi.org/10.1177/0090591790018003001. Risse, Thomas. 2002. “Law and Politics beyond the Nation State: Areas of Conversation and of Common Ground”. In Conference Law and Politics, Max Planck Project Group “Common Goods: Law, Politics, and Economics”, Bonn, 24–25 May. Roznai, Yaniv. 2016. Unconstitutional Constitutional Amendments: The Limits of Amend­ ment Powers. Oxford: Oxford University Press. Schmidt, Vivien A. 2008. “Discursive Institutionalism: The Explanatory Power of Ideas and Discourse”. Annual Review of Political Science 11 (1): 303–326. https://doi.org/10. 1146/annurev.polisci.11.060606.135342. Schmitt, Carl. 2008. Constitutional Theory. Durham: Duke University Press. Thelen, Kathleen. 1999. “Historical Institutionalism in Comparative Politics”. Annual Review of Political Science 2 (1): 369–404. https://doi.org/10.1146/annurev.polisci.2.1.369. Uprimny, Rodrigo. 1996. “Jueces, Narcos y Politicos: La Judicialización de La Crisis Polí­ tica”. In Tras Las Huellas de La Crisis Política, edited by Francisco Buitrago. Bogotá: Fundación Friedrich Ebert de Colombia.

6

Epilogue

The end of a book contains with almost prophesizing certainty a new question; after all, the scientific method, by definition, consists of doubting all apparently settled questions. This book cannot be an exception to that rule—it too aspires to deepen our scientific knowledge and must deal with the fact that its findings can never attain the attribute of being final, but, as with questions in judicial inquiries, remain open to contestation. By now, the reader will know that Colombia’s past is tumultuous, producing contradictions and puzzling outcomes in factory-akin consistency. It is therefore not surprising that the aftermath of the 2010 decision to reject constitutional reform in Colombia also brought new aspects and devel­ opments to light that deserve at least some brief discussion. To be sure, none contradict my thesis that the constitutional judge is a deliberative judge—on the contrary, there is mounting evidence that supports that conclusion, albeit in para­ doxical shape. For example, there are two scandals, post-2010, that have befallen the high courts in Colombia—both the Supreme and Constitutional Court—and both require explanation. In addition, the Constitutional Court issued further cases that cite the substitution doctrine, which has in fact become fully entrenched in the Court’s jurisprudence. In the same vein, it is testament of the judicial authority that Congress decided to revoke immediate presidential re-election in 2015, justifying the change of mind with the deficiency of the 2004 re-election reform law to uphold the equilibrium between the branches of government, or, in other words, with the importance of the separation of powers for constitutional democracies. Finally, it would be disingenuous not to mention the most important development in Colombia since 2010: the peace process with the FARC to end the longest-lasting insurgency in the Western hemisphere. Incidentally, Uribe ultimately attempted to derail the legal framework of the demobilization process before the Constitutional Court, arguing that it substituted the 1991 Constitu­ tion. Needless to say, Colombia’s courts have played, and will continue to play, an important role in that process as well. However, despite the finalization of the peace treaty against mounting opposition from Colombia’s right, the real test only begins now. Will Colombia fare better in current conditions and incorporate an actor that its armed forces have fought for for the better part of the last century, or will the country experience a resurgence of violence and the rise of new violent actors? This is only to be seen in the next years, and if the past is an indicator, the DOI: 10.4324/9781003229285-6

242 Epilogue most exigent hurdles of transitional justice processes in general—and Colombian peace processes in particular—will have to be overcome now. The integration of former armed actors while at the same time continuing to confront the interna­ tional narcotics trade have in the past proven to be herculean tasks.

The Substitution Doctrine after 2010 The idea of unconstitutional amendments to a constitution is not new; arguably, 1789 is the modern watershed moment that invented the idea of constituted and constituent power and thereby also the differentiation between legal and illegal constitutional amendments (Roznai 2013, 661). Nevertheless, in recent years, events and developments—not least in Latin America—have rekindled the con­ tentious issue amongst constitutional scholars. In fact, the Colombian Constitu­ tional Court’s decisions traced in this book have played no small part in the renaissance of the idea that constitutions contain implicit substantive limits that curtail the amendment power of constituted branches of government (see Gon­ zález-Bertomeu 2019; Dimaté and Vernaza 2017; Roznai 2016; Bernal 2013; Colón-Ríos 2012). More generally, the Latin American context provides suitable ground for comparisons between different experiences with the invocation of constituent power: some that lead down a path towards democratic renewal, which is capable of constraining majoritarian tendencies later on (arguably Colombia is a textbook example; Eisenstadt et al. 2015), and other experiences, where constituent power is instrumentalized for majoritarian and authoritarian ends (Venezuela appears to be the textbook example of this trajectory; Weyland 2013.). In all of this, it has become readily evident that the substitution doctrine does indeed represent an influential and important doctrinal articulation of lim­ itations to constitutional reform that has caught the attention and imagination of theorists, commentators, and practitioners alike. Given the importance of the doctrine for legal theory and discourse—not only in Colombia—it would be dis­ ingenuous to fail to amend the discussion of the substitution concluded in Chap­ ter 4 with a brief discussion of the most important decisions that cited the substitution doctrine since 2010. If the split decision in 2010 (5:4) might have suggested that the doctrine is still a fragile edifice in the jurisprudence of the Court, it has now become fully entrenched (a 2016 decision on judicial reform was abrogated with a vote of 8:1; see below).1 After the famous 2010 decision, the Court cited the doctrine again in 2012. C­ 1056/12 is important for the purposes of my argument not only because it fur­ ther entrenched the doctrine in the Court’s jurisprudence, but also because it “flipped” a judge who had opposed the doctrine in the 2010 decision. My argu­ ment of constitutional judges as deliberative judges relies on the ability of indivi­ dual judges to alter their opinion in the face of stronger arguments. The ponente in 2012 was Nilson Pinilla Pinilla, who had voted against the application of the sub­ stitution doctrine in the re-election decision. In this case, however, new facts have appeared to change his mind regarding the applicability of the substitution doc­ trine. In its decision, the Court found Congress acting ultra vires when it

Epilogue

243

introduced a constitutional amendment that exempted debates on constitutional amendments themselves from the ordinary regulations governing conflict of interests. In the words of the Court, excluding conflict of interest rules from the sanctions against behavior contrary to constitutional principles distorts a critical aspect of the institutionalization of the legislative branch. These functional aspects are not trivial components, but part of the axiomatic, separation of powers, fra­ mework of the 1991 Constitution (C-1056/12). In the end, a slim majority (5:4) found that basic elements of the constitution were replaced, because it infringed on the respect for public morality in a democracy (Cajas-Sarria 2017; Bernal 2013). Commenting on Nilson Pinilla Pinilla, an interviewee noted that as Pre­ sident of the Supreme Court, Pinilla Pinilla had been one of the loudest critics of the Constitutional Court for its expansive rights review. Now, “he changed dras­ tically. Nothing minimalist anymore. He has changed into a defender of rights, for victims, for health. He is a person with a high sensitivity for the law.”2 The next substitution doctrine cases were more controversial and contentious. They did not only further build the jurisprudence, but also re-politicize the issue of constitutional replacement to some extent. In 2016, the substitution doctrine was broadly accepted within Colombia. Yet, as Mario Cajas-Sarria writes, “the two decisions have reminded Colombians of the political nature of judicial review and constitutional justice, [because] they have been widely perceived as an obstacle to the improvement of judicial administration”.3 Given that the cases involved the reform of the judiciary, the image emerged that the Court ruthlessly used its dis­ cretionary power of constitutional review when it concerned its own institutional interest. In addition, these cases took place in the context of scandals inside the Constitutional Court (see below). The first case from June of 2016 (C-285/16) struck down laws that established new bodies for the governance and administra­ tion of the judiciary, replacing the administrative chamber of the Superior Council of the Judiciary. The point of contention for the Court was that the newly estab­ lished Judicial Governance Council (JGC) would replace a structure, which ensured the principle of self-government of the judiciary, which in turn is intrin­ sically linked up with judicial independence and the separation of powers. Repla­ cing these principles, the Court argued, substituted parts of the Constitution. The second decision, from July of 2016, declared the creation of the “Commission of aforados” a replacement of the Constitution. The commission was designed to replace the rather ineffective Commission of Prosecutions and Accusations of the House of Representatives to oversee criminal and dis­ ciplinary prosecutions of high court judges (see Chapter 3). Similar to the aforementioned decision, the Court argued that the creation of the Commis­ sion “replaced the defining axes of the separation of powers, autonomy, and judicial independence” found in the 1991 Constitution by touching on the system of investigations within the judiciary itself. According to the Court, this undermined judicial self-governance.4 While both decisions appeared as selfserving to observers outside the Court, they also signaled the definite entrenchment of the substitution jurisprudence: both were decided with a comfortable margin of 8:1 (C-373/16).

244 Epilogue After Uribe’s time in office, the peace process with the FARC–EP became the most important policy in Colombia—for good reason, given the longevity of the conflict. Álvaro Uribe became the most outspoken critic of the process with the FARC, which culminated in successfully leading the campaign against the Havana Peace Accord in the plebiscite in October 2016. An additional tool Uribe utilized to stop the process was to submit a claim of unconstitutionality before the Con­ stitutional Court that argued that the peace treaty replaced the Constitution and was therefore unconstitutional. In other words, Uribe, at that point a Senator in Congress, argued against the peace process framework with the substitution doc­ trine. Even more ironic, Uribe submitted his complaint with a list of signatures from more than a million citizens, who opposed the peace process. Uribe pressed the Court to strike down a law with a counter-majoritarian tool, the substitution doctrine, with the argument that the majority of Colombians opposed the peace process.5 In a first decision, the Court struck down parts of the peace legislations known as “fast-track”. The fast-track was part of the legislative package within the peace deal to ensure that Congress could quickly introduce and implement guar­ antees such as the amnesty bill. In a second decision, the Court upheld other parts of the peace deal with the FARC–EP. In C-630/17, it allowed the content of a transitional article because it corresponded with norms in international law. A key part of the Constitutional Court case law jurisprudence was the introduction of the Constitutional Block (Bloque Constitucional), which entails that international human rights treaties and international humanitarian law attain constitutional rank in the Colombian Constitution (Cepeda Espinosa and Landau 2017). Ultimately, the Court held that “state institutions have the obligation to comply in good faith with what is established in the agreement and that their interpretations and actions ‘must be consistent’ with the agreement”.6 Moreover, the provisions are manda­ tory for another three presidential periods, substantively anchoring the peace treaty in constitutional law. Thus, it is proven beyond reasonable doubt that the substitution doctrine is now an entrenched doctrine of the Court’s jurisprudence and has, quite rightly, attained critical acclaim amongst legal scholars. It has withstood negative backlash, and even actors who had been on the receiving end of competence decisions— such as Álvaro Uribe—have now accepted its validity. It is further confirmation of the argumentative authority that the Court developed in C-141/10, the second re-election decision, that Congress during Santos’ second term attempted to turn back the clock, and revoke immediate presidential re-elections. The legislation to rescind re-election, conveniently introduced one day after Santos had won his re­ election, submitted that “the decision of Congress, in 2004, to allow the pre­ sidential reelection, disequilibrated the system of democratic checks and bal­ ances”.7 The Centro Democratico, Uribe’s party in Congress, lamenting that the law was directed against Uribe himself, disputed that re-election unhinged the separation of powers. Rather, it was the abuse of the law of guarantees included in the original re-election reform—presumably by the Santos government—that put the different branches of government off balance. In the end, an ample majority in Congress affirmed the switch back to one-term presidencies.

Epilogue

245

Scandals in the Judiciary It might be part of the Colombian paradox (recalling Janus!) that after years of astounding successes and the display of best practices, Colombia’s high courts face the greatest challenge to their reputation since the inception of the new Con­ stitution in 1991. Since 2015, both high courts, the Constitutional and Supreme Court of the country, are engulfed in piteous corruption scandals. Evidently, this is particularly dreadful, given the accomplishments both courts have achieved in the thwarting of nefarious relations between Members of Congress and non-state armed groups, and certainly undercuts the authority each court had commanded because of their displayed autonomy. In their essence, both scandals were similar. In each case, judges on the court solicited bribes to affect the outcome in salient court cases: in the case of the Constitutional Court, a judge offered to decide a tutela in the interest of an oil company, and in the case of the Supreme Court, judges offered to decide cases in the interest of defendants implicated for para­ military relations. A noteworthy difference between the scandals is the source of the accusation that brought the incidences to public light. In the case of the Constitutional Court, it was one of the magistrates himself, who reported the attempt, while in the case of the Supreme Court, the source was an extradition request from the United States for the prosecutor of the anti-corruption unit on corruption charges. In other words, in the former case, the source was internal to the institution, and in the latter case, it was external (even foreign). The scandal at the Constitutional Court broke when magistrate Mauricio Gonzalez accused then president of the Court, Jorge Pretelt, of soliciting bribes from Fidupetrol for $200,000 to issue a tutela that would annul a sentence that imposed a fine of $9 million. Gonzalez was the presiding magistrate in the Octo­ ber 2014 case.8 Fellow magistrate Luis Ernesto Vargas had informed him that Pretelt had approached the Fidupetrol attorney, Victor Pacheco, with the offer to convince Gonzalez to overturn the ruling and fine. Gonzalez, however, rather than awaiting the offer, filed a complaint against Pretelt, which became public in February 2015, when his report was submitted to the Commission of Accusation.9 As the accusations were debated in the Commission of Accusation, the rest of the Court’s magistrates increased the pressure on Pretelt to resign from the pre­ sidency and withdraw from the Court. Pretelt offered to step aside temporarily, but sternly refused to resign from the Court. Maria Victoria Calle, who assumed the presidency in his place, reaffirmed that judges, too, are presumed innocent, but nevertheless insisted that for the good of the institution, Pretelt should step down—a sentiment shared with great vehemence by the rest of the judges that were evidently worried about the reputation of the Court. Regardless of the mounting pressure from within and outside the Court Pretelt stayed put, requiring impeachment procedures.10 In November of the same year, the case against Pre­ telt moved from the Commission of Accusation to a vote in the Chamber of Representatives, which, affirming the judge was guilty of extortion, submitted the investigation to the Senate, which confirmed the conclusion with a 55:5 vote. (Uribe’s caucus in Senate left the plenary in protest.) With the Senate’s

246 Epilogue affirmation, Pretelt not only lost his seat on the Court; he also was stripped of his immunity, opening the path to criminal investigations.11 In October 2017, the hearings before the Supreme Court began,12 and in December 2019, he was sen­ tenced to six and a half years in prison.13 The scandal in the Supreme Court and Prosecutor General’s Office entailed solicitations of bribes in cases that go to the core of Colombia’s governance issues: money in exchange for deciding parapolítica cases in favor of the accused. Thus, it was not solely the personal enrichment that raised the specter of indignity, but the fact that these cases touch on gross human rights violations committed by paramilitaries or other non-state armed actors. Beyond the indignity of the case, I suggest that it is significant that the Constitutional Court displayed a pattern of buoyancy against apparent nefarious influences on decision-making processes inside the court, while in the case of corruption allegations against judges of the Supreme Court this has not been apparent. The initial investigation came from outside the institution—even outside the country—suggesting that the internal institutions to enforce ethical reasoning were not decisive for bringing the case to the forefront of public attention. The scandal of the Supreme Court began with an extradition request from the United States against Luis Gustavo Moreno, the prosecutor responsible for the anti-corruption unit—on corruption charges, no less. Colombian officials from the attorney general’s technical investigation team (Cuerpo Tecnico de Investigacion, hereafter CTI) together with agents from the US Drug Enforcement Administra­ tion (DEA) arrested Moreno and his lawyer Leonardo Luis Pinilla in June 2017 in Bogotá to face charges of “conspiracy to launder money with the intent to pro­ mote bribery and corruption in Colombia”.14 US officials had caught both on tape in a sting operation, in which they demanded more money from former Córdoba governor Alejandro Llyons Muskus. In exchange, they offered to affect the outcome of a corruption case involving paramilitary ties. The accusations and arrest of Luis Gustavo Moreno was only the beginning. The next explosive bombshell was the accusation against former President of the Supreme Court Francisco Ricaurte, who was arrested in September of 2017, and Supreme Court judge Gustavo Malo. Both were accused by Senator Musa Besaile of extorting $700,000 from him. In all of these allegations the pattern is similar: judges or prosecutors solicited money from accused politicians in exchange for favorable decisions.15 Ricaurte was sentenced to 19 years in prison for his role in what has become known as the “Toga Cartel” in Colombia.16 While both affairs undoubtedly constitute grievous events, the fact remains that in the case of the Constitutional Court, judges themselves brought forward the allegations against magistrate Pretelt, while in the case of the Supreme Court initial evidence was presented by external agencies. This does not only shine a negative light on the Supreme Court, whose investigations into parapolítica rela­ tions between 2006–2010 fully disclosed the magnitude of such relations in the first place. It also affirms that internal norms to protect the integrity of the Court (and its deliberation process) are stronger within the halls of the Constitutional Court than within the Supreme Court in Colombia.

Epilogue

247

From Actors of War to Architects of Peace? It was not predictable and far from a foregone conclusion that Uribe’s successor in the presidential palace would opt for a more dovish approach to dealing with the FARC. After all, Juan Manuel Santos had been Minister of Defense in Uribe’s administration and not only sided but implemented Uribe’s tough-hand approach. The notorious false positive scandal rocking the armed forces in 2009 falls in the time of his tenure at the head of the Defense Ministry. Nevertheless, in 2011 he accepted the FARC–EP’s overtures for rapprochement and clandestinely held exploratory meetings in Havana, Cuba. At the same time as these negotiations were taking shape outside the public view, Santos initiated the land restitution law that aimed at returning stolen land to displaced people. Both initiatives infuriated his former boss Álvaro Uribe, and created the schism between Santos and Uribe that has played such a dominant role in Colombian politics since 2012. Between 2011 and the passage of the peace treaty in Congress in December 2016, the negotiations hit several obstacles on the path to conclusion; the most serious one with the failure of the plebiscite to ratify the treaty in October 2016. Despite the ultimate success of the process, Colombia’s security situation remains complex: the second largest guerrilla group, the ELN, and various spin-off groups of former paramilitary groups, as well as different drug trafficking organizations, remain active. Given the multitude of actors involved in the highly profitable drug trade, a full pacification of Colombia is unlikely. Yet, what remains beyond doubt is that signing a peace treaty with the FARC–EP after six decades of internal war is an opportunity of historic proportions to turn the country into a more peaceful place. Equally beyond doubt is that the most taxing part of that process is only to come now, and we will only be able to pass judgment in the coming years. Beyond hope for the future, the commencing integration of former FARC fighters offers the opportunity to study transition processes and highlight, among other things, the role of courts. The peace negotiations officially began in October 2012; first in Oslo and then moved to Havana. In 2013, Uribe started to publicly oppose the peace process under the “No” campaign, which would eventually succeed in mobilizing suffi­ cient numbers of Colombians against the peace treaty in the referendum in 2016. After the failed referendum, Santos introduced a slightly altered accord to Con­ gress, which passed in December of that year. In June 2017, the FARC–EP offi­ cially declared to have handed 100% of their weapons and thereby ceased to exist as an armed group. In September of 2017, the group reconstituted itself as the Fuerza Alternativa Revolucionaria del Común (The Common Alternative Revo­ lutionary Force), retaining its (in)famous acronym FARC (but without the suffix “–EP” for people’s army), and entering legal political life. Of course, observers know that the founding of a political party need not mean the end of violence— the fate of the Unión Patriotica is the tragic testament of the difficulty that lie ahead. As detailed in this book (see Chapter 2), the assassinations of party activists were a critical factor of the political crisis in the 1980s that eventually resulted in the movement to alter the Constitution and the conveying of the constituent

248 Epilogue assembly in 1991. What is most worrying about these prospects is that we can already observe all too familiar patterns re-emerging. Already during the negotia­ tions, threats against social leaders and assassinations went up. According to the Defensor del Pueblo, over 71 peasant leaders, who had returned to their stolen land, were assassinated between 2006 and 2011—either by paramilitaries or their follow-up organizations.17 The number of assassinations has gone up. The postaccord period in particular has been tremendously lethal for social leaders and defenders of human rights. INDEPAZ has traced 566 assassinations of social lea­ ders since the signing of the Havana Peace Treaty until January 2019 (2019). The Fundación Ideas para la Paz (FIP) has traced an increase in assassinations of social leaders between 2010 and 2018 from 38 to over 90 per annum.18Prem et al. (2018) have shown that the assassinations of social leaders has been a direct, albeit unintended, consequence of peace with the FARC–EP, arguing that “the killing of social leaders is driven by the selective targeting of leaders so as to thwart collec­ tive action at the local level, and not by indiscriminate municipal violence” (22). It appears that, despite the demobilization of the FARC–EP, it remains true what Bejarano and Pizarro wrote more than fifteen years ago: in Colombia the problem is not that elections are not held or disrespected, but that activists, journalists, and social leaders are repeatedly killed for simply doing their work. Colombia remains a besieged democracy (2005).

Notes 1 Cajas-Sarria, Mario, “The Unconstitutional Constitutional Amendment Doctrine and the Reform of the Judiciary in Colombia”, Int’l J. Const. L. Blog, 1 September 2016. http://www.iconnectblog.com/2016/09/the-unconstitutional-constitutional-am endment-doctrine-and-the-reform-of-the-judiciary-in-colombia/. 2 Interview with secretary of Constitutional Court, 10 May 2013. 3 Cajas-Sarria, Mario, “The Unconstitutional Constitutional Amendment Doctrine and the Reform of the Judiciary in Colombia”, Int’l J. Const. L. Blog, 1 September 2016. http:// www.iconnectblog.com/2016/09/the-unconstitutional-constitutional-amendment-doc trine-and-the-reform-of-the-judiciary-in-colombia/. 4 Cajas-Sarria, Mario, “The Unconstitutional Constitutional Amendment Doctrine and the Reform of the Judiciary in Colombia”, Int’l J. Const. L. Blog, 1 September 2016. http://www.iconnectblog.com/2016/09/the-unconstitutional-constitutional-am endment-doctrine-and-the-reform-of-the-judiciary-in-colombia/ 5 González-Jácome, Jorge, “Courts and Public Opinion: The Colombian Peace Process and the Substitution of the Constitution Doctrine”, Int’l J. Const. L. Blog, 28 Septem­ ber 2016. http://www.iconnectblog.com/2016/09/courts-and-public-opinion-the­ colombian-peace-process-and-the-substitution-of-the-constitution-doctrine/. 6 Ramírez-Cleves, Gonzalo, “The Colombian Constitutional Court Rules that the Peace Agreement is Mandatory for Three Presidential Terms”, Int’l J. Const. L. Blog, 26 October 2017. http://www.iconnectblog.com/2017/10/the-colombian-constitutional-court-rules -that-the-peace-agreement-is-mandatory-for-three-presidential-terms/. 7 Semana, “El fin de la reelección presidencial”, 3 June 2015. https://www.semana. com/nacion/articulo/congreso-elimina-definitivamente-reeleccion-de-la-constitucion/ 430119-3. 8 El Tiempo, “Magistrado Mauricio González amplió su denuncia contra Jorge Pretelt”, 4 March 2015. http://www.eltiempo.com/archivo/documento/CMS-15340418.

Epilogue

249

9 Semana, “Guerra en la justiciar”, 28 February 2015. http://www.semana.com/na cion/articulo/lo-que-hay-detras-de-la-pelea-entre-los-magistrados-gonzalez-pretelt/ 419364-3. 10 El Universal, “Cronología del proceso que se le adelantó al magistrado Pretelt”, 24 August 2016. http://www.eluniversal.com.co/colombia/cronologia-del-proceso-que­ se-le-adelanto-al-magistrado-pretelt-233989. 11 El Espectador, “¿Adiós a la investidura del magistrado Jorge Pretelt?” 18 August 2017. https://www.elespectador.com/noticias/politica/adios-investidura-del-magistrado-jor ge-pretelt-articulo-649627. 12 El Tiempo, “Jorge Pretelt negó haber exigido dinero en el caso de Fidupetrol”, 19 October 2017. http://www.eltiempo.com/justicia/investigacion/se-inicia-juicio-con tra-el-exmagistrado-jorge-pretelt-por-caso-fidupetrol-142630. 13 Colombiareports.com, “Colombia’s former Constitutional Court president sentenced to prison over corruption”, 19 December 2019. https://colombiareports.com/colombia s-former-constitutional-court-president-sentenced-to-prison-over-corruption. 14 Insightcrime, “Colombia Attorney General’s Office Hit by Corruption Scandals”, 4 July 2017. http://www.insightcrime.org/news-briefs/colombia-attorney-general-s-office­ hit-by-corruption-scandals. 15 Insightcrime, “Colombia Judicial Corruption Scandal Expands With Fmr Top Judge’s Arrest”, 21 September 2017. http://www.insightcrime.org/news-briefs/colombia-judi cial-corruption-scandal-expands-fmr-top-judge-arrest. 16 Colombiareports.com, “Colombia’s former Supreme Court president sentenced to 19 years”, 26 March 2021. https://colombiareports.com/colombias-former-suprem e-court-president-sentenced-to-19-years. 17 La Semana, “Reina impunidad en asesinatos de líderes de tierras”, 4 September 2012. http://www.semana.com/nacion/articulo/reina-impunidad-asesinatos-lideres-tierras/ 256131-3. 18 Fundación Ideas para la Paz (FIP), “Agresiones y Homicidios de Líderes Sociales”, August 2018. http://ideaspaz.org/media/website/infografia-lideres.pdf.

References Bejarano, Ana María, and Eduardo Pizarro. 2005. “From ‘Restricted’ to ‘Besieged’: The Changing Nature of the Limits to Democracy in Colombia”. In The Third Wave of Democratization in Latin America, edited by Frances Hagopian and Scott P. Mainwar­ ing, 235–260. Cambridge: Cambridge University Press. https://doi.org/10.1017/ CBO9780511791116.009. Bernal, Carlos. 2013. “Unconstitutional Constitutional Amendments in the Case Study of Colombia: An Analysis of the Justification and Meaning of the Constitutional Replace­ ment Doctrine”. International Journal of Constitutional Law 11 (2): 339–357. https:// doi.org/10.1093/icon/mot007. Cajas-Sarria, Mario Alberto. 2017. “Judicial Review of Constitutional Amendments in Colombia: A Political and Historical Perspective, 1955–2016”. The Theory and Practice of Legislation 5 (3): 245–275. https://doi.org/10.1080/20508840.2017.1407397. Cepeda Espinosa, Manuel José, and David Landau. 2017. Colombian Constitutional Law: Leading Cases. Oxford: Oxford University Press. Colón-Ríos, Joel. 2012. Weak Constitutionalism: Democratic Legitimacy and the Question of Constituent Power. London: Routledge. https://doi.org/10.4324/9780203120132. Dimaté, Juan Diego, and Santiago Vernaza. 2017. “Control de Constitucionalidad Al Fast Track: ¿Debilitando El Poder Constituyente?” UNA Revista de Derecho 2.

250 Epilogue Eisenstadt, Todd A., A. Carl Levan, and Tofigh Maboudi. 2015. “When Talk Trumps Text: The Democratizing Effects of Deliberation during Constitution-Making, 1974– 2011”. American Political Science Review 109 (3): 592–612. https://doi.org/10. 1017/S0003055415000222. González-Bertomeu, Juan F. 2019. “The Colombian Constitutional Court’s Doctrine on the Substitution of the Constitution”. In Constitutional Change and Transformation in Latin America, edited by Richard Albert, Carlos Bernal, and Juliano Zaiden Benvindo, 119–142. Oxford: Hart Publishing. https://doi.org/10.5040/9781509923533. ch-005. INDEPAZ. 2019. “Informe Especial: Violaciones a Derechos Humanos En Tiempos de Paz”. http://www.indepaz.org.co/wp-content/uploads/2019/09/Informe-Violacio nes-a-los-Derechos-Humanos-en-tiempos-de-Paz.-Septiembre-de-2019-18-09-19. Prem, Mounu, Andrrs Rivera, Dario Romero, and Juan F. Vargas. 2018. “Killing Social Leaders for Territorial Control: The Unintended Consequences of Peace”. SSRN Elec­ tronic Journal. https://doi.org/10.2139/ssrn.3203065. Roznai, Yaniv. 2013. “Unconstitutional Constitutional Amendments—the Migration and Success of a Constitutional Idea”. American Journal of Comparative Law 61 (3): 657– 719. https://doi.org/10.5131/AJCL.2012.0027. Roznai, Yaniv. 2016. Unconstitutional Constitutional Amendments: The Limits of Amend­ ment Powers. Oxford: Oxford University Press. Weyland, Kurt. 2013. “Latin America’s Authoritarian Drift: The Threat from the Populist Left”. Journal of Democracy 24 (3): 18–32.

Court Cases Constitutional Court of Colombia. C-141/10. M.P. Humberto Antonio Sierra Porto. Constitutional Court of Colombia. C-1056/12, M.P. Nilson Pinilla Pinilla. Constitutional Court of Colombia. C-285/16. M.P. Luis Guillermo Guerrero Pérez. Constitutional Court of Colombia. C-373/16. M.P. Alejandro Linares Cantillo, Gabriel Eduardo Mendoza Martelo. Constitutional Court of Colombia. C-630/17. M.P. Luis Guillermo Guerrero Pérez, Antonio José Lizarazo Ocampo.

Appendices

Appendix A: Public Opinion and Public Discourse

Part of the process tracing engaged in this book involved understanding the discourse during Uribe’s years in government. I therefore analyzed the reporting in Colombia’s biggest and most important weekly political magazine—Semana. The purpose was to follow the dominant public debates and discussions and triangulate these with the data obtained from elite interviews. Furthermore, I also endeavored to see if the public debate was as decisively pro-Uribe as his popularity figures suggest (see Graph A.1).

Approval Ra�ngs

Sep-09

Sep-08

Mar-09

Mar-08

Sep-07

Sep-06

Mar-07

Sep-05

Mar-06

Sep-04

Mar-05

Mar-04

Sep-03

Disapproval Ra�ngs Sep-02

90 80 70 60 50 40 30 20 10 0

Mar-03

In percentage

(Dis) approval ra�ngs

Month and Year

Graph A.1 (Dis)approval ratings of President Álvaro Uribe, 2002–2010

Identifying whether a public debate is indeed leaning in favor or against a par­ ticular political leader is an elusive activity, because debates themselves are difficult to categorize. Already Pollock and Adorno in their famous group experiment study found that survey data to trace public opinion is ontologically deficient, because it does not fully reflect the back and forth that is constitutive of debates and discussions (Pollock and Adorno 2011). Moreover, an interviewee reported to me that Uribe best utilized the radio for his political messaging. The time con­ strains and the lack of availability of radio transcripts simply made it impossible to incorporate all different types of media into this analysis. Thus, this study does not

Appendix A: Public Opinion and Public Discourse

253

claim to incorporate all facets of public discourse surrounding Uribe in his pre­ sidency. However, another subject also said that the Constitutional Court took notice that the print media at the time of the second re-election reform was not enticed with the prospects of further constitutional amendments to prolong Uribe’s time in office. In my judgment, it was therefore necessary to at least see if there was some evidence to suggest that the print media did indeed view Uribe in a more skeptical light by the end of the second term. Furthermore, the data pro­ vides the reader with a practical overview of the most important events during Uribe’s tenure in the presidential office and serves the purpose of a timeline. I therefore accompanied each graph with an annotation of the most important events in that specific year. I surveyed the magazine’s “Nación” pages for the entire time of the Uribe presidency from his inauguration in August 2002 until March 2010, when the Constitutional Court had declared the law 1354 of 2009 unconstitutional. I analyzed the articles and disaggregated them according to the space devoted to critical or positive reporting on Uribe’s policies. I omitted pages that had a neutral outlook from the tally. A four-page article that was critical of Uribe’s strategy to secure his first re-election but spent half of the four pages devoted to general rules and tactics without specifically addressing Uribe would be counted as two pages in the critical column, with the other two pages omitted. If, however, there was a four-page article that described the paramilitary rela­ tions of Mario Uribe, the President’s cousin and important political ally in Congress, these would be counted as four negative pages. The analysis of the content of the Revista Semana required a very close reading and understanding of the arguments and facts reported in the articles, because it aimed at interpretively understanding the creation of meaning. My elite interviews suggested that print media reports—Semana’s in particular— provided a more critical context for debates around Uribe and his allies in Congress during the second term. The content analysis therefore required a deep understanding of the context in order to provide any valuable and quan­ tifiable data to trace at least a generic development of debates. The most important caveat in the tracing of debate was whether there were explicit connections drawn between Uribe, his coalition in Congress, and evidently negative developments. The evolution of paramilitiarism during Uribe’s term is telling of the challenges implicit in such an approach to tracing discourse. Prior to the revelations of the parapolítica scandal, paramilitaries were certainly not viewed as positive for the development in Colombia, but only very rarely con­ nected to specific politicians. After the first revelations of close relations between paramilitaries and officials, this fundamentally changed. The reports about Jorge Noguera’s ties with paramilitaries, who was the director of DAS and a close supporter of Uribe during his first election, are essential for this, since after these revelations paramilitarism was not explained as a phenomenon without political linkages. Rather, there were explicit arguments and studies made that tied Uribe’s political support to the influence of paramilitaries. Moreover, after the incrimination of Mario Uribe, these linkages were explicitly

254 Appendix A: Public Opinion and Public Discourse along family lines. As a consequence, prior to parapolítica, reports that focused on paramilitaries and their dealings with politicians (most often as a con­ sequence of the commencing peace process) were neither tallied as negative or positive, but rather as neutral. After the Noguera revelations and explicit con­ nections drawn between parapoliticos and Uribe’s coalition in Congress, reports were much more explicitly negative and tallied as such. Another challenge was to incorporate reports of the FARC. Uribe made tack­ ling the FARC militarily a prime component of his Democratic Security approach. He legitimized this stance with the cruelty of the FARC, the damage it has done to the Colombian state (and democracy), and its undemocratic nature. Particularly in the first two years of his presidency, the FARC attacks that resulted in military and civilian casualties were not that uncommon. Reports about these incidences could be counted as “positive”, “negative”, or neutral. My strategy was to incor­ porate these reports into “positive” for Uribe, because they aligned with the argumentation that Uribistas put forward in support of their tough-hand approach. Only if the military strength by the FARC supported arguments (in editorial pages, for example) for a different approach to the internal conflict (through negotiations or a humanitarian exchange with the FARC) did these reports count as explicitly negative. I was consistent with this weighing throughout. In the end, I had a quan­ titative weighing of Semana reporting on Uribe and his government divided into positive and negative pages reported, from which I could derive a co­ efficient for each month to show if reports were generally more negative or positive. A co-efficient of 1 represents an equal amount of positive and nega­ tive pages. A coefficient over 1 represents a more positive public discourse as it was reflected in the discussions in Semana. We see a shift occurring in 2006, when Uribe was re-elected and the parapolítica revelations were only begin­ ning to come to light. The second term was entirely below 1, with the exception of 2008 when the tensions with Chavez’s Venezuela and Correa’s Ecuador, attacks against the FARC, and the liberation of Ingrid Betancourt countered the generally much more negative reporting on parapolítica, the falsos positivos, and other corruption scandals implicating Uribe’s sons (Zona Franca). It also showed that 2007 and 2009 were the most negative for Uribe in terms of public debates. These were the years when the Supreme Court most forcefully investigated the parapolítica scandals and implicated numerous Members of Congress. In 2007, it opened investigations against Mario Uribe and in 2009 it changed its jurisprudence and recontinued investigations against Members of Congress that have resigned from their seats and thereby declined their parliamentary privileges (see Chapter 4). The comment column denotes the most important events in the year and highlights specific incidences that have had a noticeable effect on the course of the reporting in Semana.

Appendix A: Public Opinion and Public Discourse

255

Table A.1 Quality of discourse in Semana during Uribe presidency, 2002–2010 Quality of Discourse Coefficient per Month

Comments

� The FARC attacks the celebrations around the pre­ sidential inauguration

� President Uribe levels the war tax on the richest Colombians to finance the security war effort

� Salvatore Mancuso mentions that 35% of Congress are paramilitaries’ “amigos”

� Average: 2.81

� Security is notably improving. Roads can be traveled as the FARC continues bomb attacks

� Constitutional Court first mentions the substitution doctrine in its review of Uribe’s Political Reform

� Political referendum fails to achieve the necessary quorum

� Beginning of formal negotiations with the AUC paramilitaries

� First Justice Minister—Londoño—leaves office in November

� Average: 2,79 � Demobilization of paramilitaries � Every community with police presence � Debate over re-election begins at the beginning of the year

� The Constitutional Court is turning down antiterror legislation

� Paramilitaries visit Congress in July � Average: 1.83 � Senator Vargas Lleras survives assassination attempt � Claudia López Hernández publishes article about atypical voting results

� Lopez Obgregon requests official inquiry by the Supreme Court into relations between paramilitaries and Members of Congress � Constitutional Court reviews and passes first re­ election reform � Average: 2.57

� Encilce de Jesus López Romero, “La Gata”, is one

of the first businesswomen indicted for paramilitary relations. She helped finance Uribe on the Car­ ibbean coast � Congress is elected in March, increasing Uribe’s coalition � The DAS IT Technician, Rafael García, becomes the witness for the prosecution in April and incriminates DAS Director (appointed by Uribe for campaign work in the Caribbean) Jorge Noguera

256 Appendix A: Public Opinion and Public Discourse Quality of Discourse Coefficient per Month

Comments

� Constitutional Court reviews Justice and Peace Law and passes it conditionally (victims’ rights to truth)

� Uribe is re-elected in landslide victory (62%) � In September the laptop belonging to “Jorge 40” is decrypted

� First public mention of strategic meetings between politicians and the AUC in Santa Fe de Ralito

� First subpoenas issued to Members of Congress � Average: 1.05 � In January, Salvatore Mancuso gives testimony to � � � �

Justice and Peace Court indicting Uribe’s cousin, Mario Uribe Mancuso provides details of meetings in Santa Fe de Ralito and the Pact of Ralito Álvaro Araújo, the brother of the Chancellor, is indicted. Indictment of Araújo and M. Uribe escalates rela­ tions between President and Supreme Court. He publicly attacks the Court Average: 0.52

� January until March is dominated by tensions with

� � � � �

Venezuela and Chavez. These escalate with the Colombian Army’s attack on a FARC camp in Ecua­ dor that kills Raul Reyes in March of 2008 In April Yedis Medina admits to bribery in relation to the first re-election. It starts Yidispolítica and eventually incriminates her in the Supreme Court Ingrid Betancourt is liberated in September, Opera­ ción Jaque in July In September, the first falsos positivos become public In December, Uribe’s sons are implicated in the socalled Zona Franca scandal Average: 1.5

� In March, Semana reveals that DAS kept tabs on Supreme Court judges. The wiretapping scandal reaches all the way to the President’s office � Supreme Court changes its jurisprudence, and recontinues investigations into Members of Congress who had resigned their seats to avoid prosecution � In December, second re-election law (to call a referendum) passes Congress � Average: 0.38

� Re-election fails. Powerful politicians, such as Vargas Lleras, take a public stand against re-election.

� Average: 0.46

Source: Compiled by the author

Appendix B: Interview Data

In total, I conducted over 40 in-depth elite interviews with persons directly or indirectly involved with the Constitutional Court of Colombia, as well as indivi­ duals who occupy prominent positions in the analysis of the actions by the Con­ stitutional Court, such as professors of law of the most important law schools in the country or persons engaged with NGOs that are involved in actively litigating Constitutional Court cases. The interviewees consisted of all magistrates of Colombia’s Constitutional Court who were on the Court in 2010 when the decision regarding the law to call for referendum to allow a second re-election was decided, save for one, whose auxiliary judge I interviewed. I also interviewed four of the nine members of the Constitutional Court that decided the decision regarding the law to permit constitutional change for a first re-election of the President of the Republic. Furthermore, the interviews consisted of auxiliary judges of eight different magistrates of the Court from different periods and three magistrates from the Supreme Court. Finally, I interviewed three jour­ nalists, four Senators and Representatives from the Chamber of Representa­ tives, ten professors of law and political science with specialization in legal politics and judicial/political culture, and five activists from NGOs litigating Constitutional Court cases. The selection resulted from an analysis of the sen­ tences by the Constitutional Court, a survey of the literature on judicial poli­ tics in Colombia, as well as an archival research of the most important journalistic publications in Colombia (La Revista Semana, El Tiempo, El Espectador, and La Revista Cambio). The interviews were in-depth and focused on the issues resulting from the content analysis of publications. They lasted from one hour to two and a half hours. The questions were structured according to these four topics: 1

Democracy and constitutionality: �

What does the contentious connection between sovereignty of the people and constitutional reform consist of? � What is the role of the judiciary in the Colombian Constitution? � What is the legal culture (Cultura Legalista) in Colombia and how has it changed since the new Constitution in 1991?

258 Appendix B: Interview Data 2

Judiciary and the media: � � �

3

The press and publicity as democratic phenomena. The impact of media discourse on decisions by the Court. How does the Court communicate with the media? Is it legitimate?

Deliberation and institutionalism: �

The role of deliberation amongst magistrates before and during the pro­ cess of reaching a decision. � Can you describe the deliberations regarding re-election?

4

Uribe, democratic security and re-election: � � � � � �

What was the importance of Democratic Security and did it justify re­ election? What were the important changes between the first and second decision regarding constitutional reform and re-election? How did the political scandals affect the decision? What importance did the revelations regarding the so-called Yidispolítica scandal have? How important were the official investigations by the Supreme Court in the parapolitica scandal? Did the image of an institutional crisis resulting from a negative decision regarding re-election matter in deliberations?

As is evident in the outline above, the questions were semi-structured, and there­ fore left the interviewee sufficient leeway in answering the question openly, and also involved follow-up questions for clarification. Following the field research, the interviews were transcribed and analytically structured according to the topics given in the questions. Most often this corre­ sponded to the chronology of the interview, but due to the open-ended nature of the questions, in some cases the interviewee made statements that could also be sub­ sumed under another topic. In the analysis, the structuring of the interview contents according to topics was complemented by an ordering of first-, second-, and thirdorder observations of the position of the interviewees. Thus, magistrates belong in the first category, auxiliary judges in the second, and observer and interpreters of jur­ isprudence of the Court in the third category. The epistemological question remains whether this also implies a hierarchical ordering of the content or whether these are simply descriptive analytical categories. While it is true that the magistrates are the closest to the decisions, since they are actively involved, they are also more likely to produce pre-structured answers to questions regarding the role of the judge in the constitutional politics of Colombia—after all, their imperative is to make judgments based on established constitutional norms and legal categories. Conversely, auxiliary judges can be seen in a position to answer more freely, since they are not directly involved nor is their role defined by norms of independence and autonomy that require aforementioned pre-structured reasoning.

Appendix C: State and State Weakness

Inten�onal homicide rate

Year Graph C.1 Homicide rate per 100,000 Citizens, 1981–2011

2011

2009

2007

2005

2003

2001

1999

1997

1995

1993

1991

1989

1987

1985

1983

90 80 70 60 50 40 30 20 10 0 1981

Homicides per 10000 ci�zens

Colombia’s state weakness has its origins in the 19th century. Chapter 2 makes the argument that it is related to the genesis of the two-party system, predating the establishment of the nation state. Chapter 2 then further explains that the drug trade introduced a new quality in terms of state weakness that undermined the few advances made during the National Front period (Bejarano and Segura 1996). Since the Bringing the State Back In project commenced in the 1980s (Evans et al. 1985), the state’s characteristics have been captured by numerous definitions and measures, but no argument has convincingly diverged in a sub­ stantive way that the state at the most basic level presumes the monopoly of vio­ lence as its defining trait (Weber 1946). The measures below consist of indicators that show deficiencies of the monopoly of violence and efforts by the state to combat this shortcoming. Already the crudest measurement makes the difficulties Colombia faced in the last three decades readily apparent. The observations are fairly evident: violence erupts at exactly the moment Pablo Escobar’s attempt to enter legal politics was foiled and he initiated the terror war against Colombia’s institutions in 1983. Graph C.1 also captures the wave of paramilitary expansion that commenced in 1993 until 2002. As explained in the same chapter, this is the

260 Appendix C: State and State Weakness period of paramilitary expansion and conflict between paramilitary groups and the guerrilla (particularly the FARC). Finally, the graph also shows the “Uribe effect” of his Democratic Security policy. After his election to the presidency in 2002, intentional homicide rates began to steadily fall. In 2010, they stood at the same level as at the beginning of the 1980s. In the second part of the 1990s, kidnappings as a war strategy symbolized the generic security crisis Colombians faced. Conversely, Uribe’s successes in reducing the numbers of kidnappings epitomized the success of his Democratic Security in general. Graph C.2 traces the evolution of kidnapping events in Colombia. As the graph suggests, the security situation effectively deteriorated after 1996 until 2002, confirming what interviewees indicated was a general sense of insecurity covering the entire nation rooted in the very real threat of being kidnapped. Guerrilla organizations (FARC, ELN) were behind a vast number of the recorded kidnappings, having made kidnapping for ransom a central part of their strategy to generate funds for the armed struggle.

2010

2008

2006

2004

2002

2000

1998

1996

1994

1992

1990

1988

1986

1984

1982

1980

1978

1976

1974

1972

4000 3500 3000 2500 2000 1500 1000 500 0 1970

Number of kidnappings

Total kidnappings per year

Year

Graph C.2 Total kidnappings per year, 1970–2010

While kidnapping was key to guerrillas’ war strategy, a central part of the para­ military strategy was to bring territory under their control with the use of coercion and terror. A common feature in that strategy was the use of massacres. Graph C.3 shows the escalation of violence in the late 1990s orchestrated as a response to the Pastrana government’s rapprochement with the FARC in the Caguán negotiations (López Hernández 2010). The graph shows that massacres increased drastically in 1999 and climaxed in 2000, then slowly ebbed after 2002. Chapter 3 also sheds some light on the demobilization process with the paramilitaries. The reduction of numbers after AUC heads and Uribe administration officials began to negotiate in Santa Fe de Ralito in 2002 resembles the events detailed in that chapter: paramilitaries could not and did not utilize coercion as freely as before, and as a con­ sequence massacres were no longer recorded in the high numbers of the late 1990s.

Appendix C: State and State Weakness

261

Massacres per year

250

Number of events

200

150

100

50

1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018

0

Year

Graph C.3 Massacres committed by year, 1988–2018

References Bejarano, Ana María, and Renata Segura. 1996. “El Fortalecimiento Selectivo Del Estado Durante El Frente Nacional”. Controversia 2 (169): 9–35. Comisión Interamericana de Derechos Humanos. 1994. “Informe Comisión Inter­ americana de Derechos Humanos 1993”. https://www.cidh.oas.org/annualrep/93span/ indice.htm. Evans, Peter B., Dietrich Rueschemeyer, and Theda Skocpol (eds). 1985. Bringing the State Back In. Cambridge: Cambridge University Press. https://doi.org/10.1017/ CBO9780511628283. Grupo de Memoria Histórica. 2013. “¡Basta Ya! Colombia: Memorias de Guerra y Dignidad”. https://www.centrodememoriahistorica.gov.co/micrositios/informeGeneral/descargas. html. López Hernández, Claudia. 2010. Y Refundaron La Patria … De Cómo Mafiosos y Políticos Reconfiguraron El Estado Colombiano. Bogotá: Debate. https://doi.org/10.1016/ 1359-0189(86)90036–90031. Pachón, Mónica, and Gary Hoskin. 2011. “Colombia 2010: Análisis de Las Elecciones Presidenciales y Legislativas”. Colombia Internacional 74 (July): 9–26. https://doi.org/ 10.7440/colombiaint74.2011.02. Pollock, Friedrich, and Theodor W. Adorno. 2011. Group Experiment and Other Writings: The Frankfurt School on Public Opinion in Postwar Germany. Cambridge, MA: Harvard University Press. Weber, Max. 1946. “Max Weber: Politics as Vocation”. In Max Weber: Essays in Sociology, 77–128. Oxford: Oxford University Press.

Index

Page numbers in bold refer to figures, page numbers in italic refer to tables acción de tutela 96–97 acción popular de inconstitucionalidad (1992–2005) 55–56, 73, 167 acción popular de tutela 73 accountability, horizontal 2, 11–12 ACCU 76 Acemoglu, Daron 58 administrative law 48 aims 23 Alexy, Robert 170–171, 204, 206 Alianza Democrática 201–202 analysis structure 29–30 ANAPO 59 Arango Mejia, Jorge 114 Araújo, Alvaro 126 Arendt, Hannah 67 Argentina 165–166 argumentative rationality 231 Arias, Rocío 121, 122 armed clientelism 10–11 assassinations 64, 65, 247, 248 AUC 11, 75, 76, 77, 117, 119, 120–121, 121–122, 122, 123–124, 128–129, 142, 225, 261 Austria 56 authority 15–16, 26, 227 Auto 004/04 134–135 autonomy 227 Avendaño, Teodolindo 129–130, 137–138, 235 Ávila Martínez, Ariel Fernando 76 Bakiner, O. 19–20, 22, 148 Barco, Virgilio 66, 67 Barros, Robert 166 base-line conditions 36–37, 42–48, 220–221

Bejarano, Ana María 10, 177 Betancourt, Ingrid 5 Bolívar, Simon 43, 46 Bonilla, Lara 64 Botero, Felipe 190–191 Botero, Fernando Zea 109, 112 branches of government, interdependencies 95–96, 96 Brazil, Hilbink 166 Bushnell, David 44, 45, 53 C-285/16 243 C-373/16 243 C-588/09 190 C-630/17 244 C-1040/05 190 C-1056/12 242–243 Cajas-Sarria, Mario 243 Cali Cartel 107–108, 108, 109, 111–112, 141 Cambio Radical 6, 229 Cameron, Maxwell A. 12 Carrillo Flórez, Fernando 69 case selection 23–27 case study, definition 25 Castaño, Carlos 76, 83n30, 121–122 Castaño, Vicente 121, 131 caudillo 192–193, 221 Centeno, Miguel Angel 44 centralization 50 Chavez, Hugo 177 checks and balances 6–7, 140, 191 Chile 165–166 choque de trenes (train crash) 18, 19, 20, 72, 99–101, 133, 223–227, 234–238; background 88–99; collision of jurisdic­ tions 88–89; judicial decisions 89; legal

Index 263 vs. constitutional certainty 102–105; origins 102; parapolítica scandal 89–90, 117–139; and path dependence 88–89, 90–91, 91–105, 96, 139–149; proceso 8000 (Process 8000) 90, 105–117, 132; production of legal facts 90, 133, 139, 234–235; rights protection 99–101 Cifuentes Muñoz, Eduardo 29, 115 citizen participation 17 citizenship 173–174 clientele relations 9, 10, 47 clientelism 42, 46, 49, 60, 74, 75, 79, 221, 222 coffee 57, 58–59 collective agency 16 Collier, David 16, 36, 38, 40, 42, 49, 52, 78, 91, 219 Collier, Ruth 16, 36, 38, 40, 42, 49, 52, 78, 91, 219 Colón-Ríos, Joel 187–188 Comisión Ciudadana 110 Commission of Accusation and Investiga­ tion 109, 111, 114, 245–246 communicative action 15–16, 64–67, 80, 101 communicative power 17, 65 Communist Party of Colombia 64 comparative politics 8 compartmentalization 46 competence, theory of 30n5 conceptual tools 219 Congress: constitutional privilege 138; criminalization 105, 117, 136–137, 224–225, 228; disciplinary code 70–71; docility 192; domiciliary right 6; double function 113–114; fragmentation and co-optation of 179–184, 182, 183; insti­ tutional weakness 194–195; La silla vacia 128; Law 1354 of 2009 6; Legis­ lative Act 02 of 2004 2–5; members criminal liability 106, 111–117, 145, 226, 227; privileges and duties 114; and proceso 8000 (Process 8000) 109–111, 224–225, 234; role 193–194; Uribe’s majority 179–180, 182, 182, 186 congressional elections, 2006 122 Consejo del Estado (State Council) 4 Conservative–Liberal hegemony 61–62, 67–69, 80, 222 Conservative Party 9, 52, 58–59 consociationalism 56–59 Constituent Assembly, 1991 60–77, 222, 232; and aftermath of National Front 62–64; composition 61–62, 67, 68–69,

68, 80; constituent character 68; critical juncture 65; debates 62; elections 68; extra-institutional path 64–67, 80; insti­ tutional crisis leading up to 61; institu­ tional path 67–69; legacy effects 80–81; minority representatives 69; origins 61–62; presidency 69; publicness 62; student movement and 61, 64–67, 80 constituent power 69–70, 188–189, 242 constituted power 188–189 Constitution, 1851 31n9 Constitution, 1886 38, 44, 50, 52–53, 67, 72–73 Constitution, 1991 14, 30n1; aftermath 74–77; Article 1 Article 4 189–190; 73; Article 86 73, 99–100; Article 230 74; Article 241 3, 73; Article 374 189; and choque de trenes 99–101; communicative power 65; comparative analysis 16; con­ ceptual components 60; congressional disciplinary code 70–71; constituent power 69–70; constituent process 16–17; crisis preceding 61, 62–64; as a critical juncture 77–81, 79, 88, 93–94, 219–221; and decline in violence 74–75; drafting 8, 222; executive-legislative relations 71; high courts 71–72, 72; historical context 36–81; historical per­ spective 26–27; implementation 38, 60–61; institutional novelties 69–74, 72; institutional structure 69–74, 72; inter­ dependencies 95–99, 96; legacy effects 70, 80–81; legal vs. constitutional cer­ tainty 102–105; origins 61; overhaul of legal institutions 18; promise to protect human dignity 99–101; public powers 70; rights catalogue 73–74; rights fra­ mework 233; rights protection 99–101; roots 37; and separation of powers 6–7; text 37 constitutional adjudication 96 constitutional amendment 26–27 Constitutional Block 244 constitutional certainty, and legal certainty 102–105 constitutional change, base-line conditions 48 constitutional constraints, imposition of 178 Constitutional Court 30n2, 37, 47, 50–51, 52–53, 72, 229–230; 1999 tutela deci­ sion 113–117; authority 1, 8; auxiliary judges 203; case distribution 198; case selection power 73; caudillo 192–193;

264 Index choque de trenes 18, 19, 20, 89, 99–101, 102–105, 105–117, 132–139, 223–227; competence 190–191; composition 6–7; creation 71, 80; criminal accountability 136; deliberation flow 28; deliberation rules 196, 201; deliberative constraints 168–172; discursive institutionalism 160–209; empowerment 9, 19, 20, 162–173, 172–173, 205–209; ethical discourse 173–176; illicit enrichment decision 111–113; institutionalization of deliberation 8–9; interpersonal relations between magistrates 203–204; judge selection 97; judicial power 25; judicial review capacity 161; jurisdictional reach 5, 190–191; jurisprudence 21, 100–101, 147, 235–236; Law 1354 of 2009 5–7; legal reasoning 195–204, 196, 197; legal vs. constitutional certainty 102–105; making of substitution doctrine 184– 195, 186; parainstitutionalization 117– 139; parapolítica scandal 19–20, 89–90, 119–120, 122, 123; parapolítica scandal 132–139, 142–143, 144–146, 224–226; path dependence 95, 139, 139–149; power 96–97; powers 26; precedent set­ ting 139;; presidential re-election reform decisions 1–2, 4, 21–22, 195–204, 196, 197205–209, 218–219, 229, 235–236; proceso 8000 (Process 8000) 18–19, 105–117, 132, 144, 224–226; as pru­ dent strategizer 167, 187, 189, 195; public interactions 201–203; reviews 3; rights-enforcing reputation 166–167; rights protection 99–101; rights review 89; role 74; scandals 204, 241, 243, 245–246; social rights jurisprudence 134; strategic vs. deliberative action 172–173, 173; substitution doctrine use after 2010 242–244; tutela contra sentencias position 132–139; and Uribe’s popularity ratings 180 constitutional democracy, tension in 192 constitutional interpretation 38 constitutionalism 99, 101, 228–229 constitutionalist/majoritarian divide 173–174 constitutional judges 201, 236; exposure to the public spotlight 27; ideological preferences 29; selection 97 constitutional reform 1, 12–13, 71, 185, 218–219; approval 134; Constituent Assembly, 1991 80; extra-institutional path 64–67; goals 68; institutional path

67–69; Law 1354 of 2009 5–7; limits to 187–189; ordinary route 3; procedures 3–4; referendum route 5–7; review 2; student movement and 62; substitution doctrine 2; viability 20 constitutional renewal movement 16–17 constitutional replacement 21, 26–27, 184, 243–244 constitutional review 208 constitutional scrutiny, co-legislative nature 171 Constitution of Cúcuta 43 Constitution of Rionegro 43–44 continuismo 54 CONVIVIRs 76, 77, 120 Coronell, Daniell 129–130 cosa juzgada 102–105 Council of Hemispheric Affairs 11 counter-agrarian reform 77 court authority 26 Couso, Javier 166 creative destruction 58 criminal justice system, collapse 62 critical junctures 16, 17, 37, 38–42, 41, 60–61; base-line conditions 42; char­ acteristics 40; components 36; compo­ nents: 220; Constituent Assembly, 1991 65; Constitution, 1991 as 77–81, 79, 88, 93–94, 219–221; data points 40; definition 38, 78; options 40; and path-dependent development 81; predictability problem 38–39 Cuban Revolution 62 cultural rights 73 Dahl, Robert 166 data collection 27–29 data validity 24 David, Paul A. 91 debt crisis, 1980s 62 decision-making 8, 15, 21, 29, 149, 164, 206, 209, 237; deliberative 197; deliberative constraints 168–172 Decree 927 of 1990 67–68 Decree 1382 133 Decree 3398 63 de Greiff, Luis 51 Dejusticia 201–202 de la Calle, Humberto 67 delegative democracy 11, 31n12, 179; measurements 12–14, 13; plebiscitary features 11–12 delegativeness 12 delegative rule 9

Index 265 delegative turn 174–175, 176–179 deliberation 8–9, 206, 230, 237; definition

15, 196; legal 21; presidential re-election

reform decision, 2010 195–204, 196,

197; rules of 196, 201

deliberative action 206–207, 207, 219

deliberative constraints 168–172 deliberative democracy 22

deliberative democratic theory 15–16 deliberative institutionalism, and rational choice 162–173 deliberative judges 160, 163, 227–230; delib­ erative constraints 168–172; key task 204;

opinion changes 171–172; presidential

re-election reform decision 196–204,

205–209; Socratic model 170–171

deliberative structures, institutionalization

of 161

del Pilar Hurtado, María 130

democracy measurements 12–14, 13

Democratic Legality 14, 205, 229

democratic rule, history of 9

Democratic Security policy 2, 14, 119,

120, 174–175, 176, 179, 180, 205,

228, 254, 259, 260–261

Díaz Mateus, Iván 137–138, 141

discontinuous change model 88

discourse 24; importance of 28–29; public

252–254, 252, 255, 256; rules of

150n11

discourse ethics 15

discursive cleavage 78

discursive institutional development 22

discursive institutionalism 14–23, 140,

160–209, 229–230, 231–232; actor

preferences 167–168; Constitutional

Court empowerment 162–173; delib­ erative constraints 168–172; deliberative

judges 160, 163; strategic judges 160,

163–168; strategic vs. deliberative action

172–173, 173

discursive judge, the 238

distributive dynamics 18

doctrinal ethos 48

domiciliary right 6

Domingo, Pilar 166

drugs see narcotics trade Dworkin, Ronald 169–170 Eastern Europe 81n1

economic development 37, 56, 57, 62

Eisenstadt, Todd A. 17

electoral system 55, 70, 193

El Tiempo 65

Epstein, Lee 20

Epstein,Lee 165, 168, 206, 208, 237

equality 3

Escobar, Pablo 63, 64, 75, 75–76, 83n30,

105, 259

estado de derecho (state of law; rule of law)

162, 175

estado de opinión (state of opinion) 7,

161–162, 174–175, 178, 178–179, 195,

210n7, 228

ethical discourse 173–176 executive 71

executive-legislative relations 71

executive powers 71

exports 57

extractive institutions 58

extra-institutional path, Constituent

Assembly, 1991 64–67, 80

Fajardo Landaeta, Jaime 69

FARC 5, 10–11, 20, 62–63, 64, 205,

254, 261; peace process 76, 241, 244,

247–248

Federal Bank 191

Finkel, Jodi 56

foreign trade 50

Freedom House 12

Frost, Robert 38, 65

Gaitán, Jorge Eliécer 52

Galán, Luis Carlos 61, 65

Garcia, Rafael 126

Gaviria, César 61, 66, 108, 185

Gaviria Vélez, Jose Obdulio 177, 178

GDP 57

General Will 14, 174, 177, 178, 229

Germany 99

Gerring, John 25, 26

Ginsburg, Tom 232

Giraldo López, Alberto 107–108,

111–112, 113

Gómez Gallego, Jorge Aníbal 115–116 Gómez Méndez, Alfonso 67

González, Lucas 12

Gonzalez, Mauricio 204, 245

Granadine Confederation 43

Gran Colombia, Republic of 43

Great Depression, the 58–59 Gretchen question, the 207–208 guaranteerist portfolio 7

Gutiérrez Sanín, Francisco 77

habeas corpus 102–105

Habermas, Jürgen 15, 100, 150n11, 171

266 Index Hartlyn, Jonathan 54, 59 Helmke, Gretchen 166, 209n1 Herbst, Jeffrey 44–45 High Council of the Judiciary 71, 71–72, 72 high courts: Constitution, 1991 71–72, 72; institutionalization 95–99, 96; joint comunicado 138–139; jurisdictional clashes 98 Hinojosa, Víctor J. 105, 110 historical institutionalism 15–16, 17, 26 historical uniqueness 219 homicide rates 64, 64, 180, 222, 259–260, 259 horizontal accountability 2, 11–12 House of Representatives 70–71; and proceso 8000 (Process 8000) 109–110 Hübner Mendes, Conrado 168–170, 171, 175, 196, 197, 206, 209 human dignity 223; definition 149–150n9; protections 99–101; voluntarist views of 150n10 human nature 20 human rights 30n1, 37, 61, 62, 71, 76, 119, 150n10, 194, 233 hyper-fragmentation 62–64 ideational carriers 22

ideological preferences, judges 29, 162,

163, 171 Iguaran, Mario 131 illicit enrichment 111–113 inalienable rights 73 inclusive institutions 58, 82n13 inclusiveness 14 incomplete vote, the 55, 57 incremental change model 95, 168–169, 234 infrastructure development 57 institutional ambiguity 147 institutional change 39; paths of 147 institutional consolidation 2 institutional decay 192 institutional development 8, 81, 141, 147 institutional engineering 8, 36, 38, 41, 60, 80, 220–221; for violence containment 53–56 institutional frameworks 22–23 institutional genesis 16 institutionalism 231–232 institutional learning 147, 223–227, 231 institutional path, Constituent Assembly, 1991 67–69

institutional stability 5 institutional structure, Constitution, 1991 69–74, 72 institutional theory 230–238 institutional weakness 37 institution building 8, 26, 37, 41, 77, 220, 220; ideal-type trajectory 78–79, 78; problem with 60; for violence contain­ ment 53–56 institutions: definition 237–238; as discursive structures 22–23; violence containment 53–56 inter-court relations 20 interview data 257–258 interviews 27–28 Japan 166 judges: ideological preferences 162, 163, 171; institutional context 165; legal context 162, 163; strategy constraints 162, 163 judicial behavior 160–209, 236–237; atti­ tudinalist model 164; constraining fac­ tors 206; convictions 160; deliberative constraints 168–172; deliberative judges 160, 163; legalistic model 163–164; opinion changes 171–172; presidential re-election reform decision 196–204, 205–209; rational choice versus delib­ erative institutionalism 162–173; schools 163; strategic judges 160, 163–168; strategic model 164–168; strategic vs. deliberative action 172–173, 173; theories of 21 judicial decision-making 24; deliberative nature of 146 Judicial Governance Council 243 judicial independence 15, 26, 56, 95, 165–166 judicial institutions 17, 26, 80–81 judicial power 7–9, 15, 71; components 7; definition 160, 218–219; disciplinary body 96; growth 22, 25, 172, 235; increase in 160–161, 205–209; Socratic model 170–171 judicial review 71, 140, 161, 169–170, 174, 209n1, 224 judiciary 71–73, 72, 88, 222; correcting function 169; executive encroachment 20; institutional learning 223–227; path dependence 90–91, 91–105, 96; scan­ dals 241, 243, 245–246 jurisprudence 21, 100–101; evolution 17; precedent setting 89

Index 267 Justice and Peace Courts 119–120, 120, 148 Justice and Peace Law 122, 133, 135–136, 148 Justice and Peace Process 20, 121–122, 122, 142, 148 Kelsen, Hans 36, 55–56, 96, 209n1, 235–236 Kelsenian model 96, 233 kidnappings 260 King, Gary 27 Kline, Harvey 31n12 Knight, Jack 20, 165, 168, 206, 208, 237 Kumm, Mattias 101, 169, 206 labor incorporation 51 labor movement 53 La Ley Cuenta 195–204, 196, 197, 208, 230 La Regeneración 44 La Violencia 49–50, 51–53, 53, 54, 62, 78, 220 Law 02 of 2004 2–5, 183, 183 law 48 63 Law 61 of 1888 51 Law 1354 of 2009 5–7, 183, 183 law and law making 70, 74 legal argumentation 168–172; path depen­ dence 139–149, 226–227 legal argument, primacy of 48 legal certainty, and constitutional certainty 102–105 legal decision-making 164, 209 legal deliberation 21, 171 legal discourse, base-line conditions 47–48 legal facts 116, 117, 131–132, 147, 153n64, 236; path dependence 139–149; production of 90, 133, 139, 148–149, 225–226, 234–235 legality 74 legal reasoning 208–209, 227, 236; insti­ tutionalization 195–204, 196, 197 legal system, collapse 64 Legislative Act 2, 2015 96 legislative elections 4–5 legislature 70 legitimation 92 Levi, Margaret 92 Liberal Party 9, 47, 50–51, 52, 52–53, 58–59 Lichbach, Mark Irving 92 Linz, Juan J. 11 Lipset, Seymour Martin 41, 46 Lleras Restrepo. Carlos 67

Londoño Hoyos, Fernando 134, 185 López-Alves, Fernando 46–47 López, Claudia 121 López Obregon, Clara 122, 123, 135 López Pumarejo, Alfonso 52, 178 M-19 group 68 McKeown, Timothy J. 25 Mahoney, James 19, 39, 88, 92–93, 95, 226–227, 231, 234; Mahoney and 146–147 Mancuso, Salvatore 76, 120–121, 124, 126, 128, 131 Marbury v. Madison 167, 209n2 Márquez, Gabriel García 62 Martínez, Alejandro 104 Mauricio García, Carlos 121–122 Mazzuca, Sebastián 50, 57 meaning making 140, 232, 235–236 media interviews 202 Medina, Santiago 108–109, 112 Medina, Yídis 3, 129–130, 137–138, 235 Mendes, H. 15 Mesa, Naranjo 114 methodology 23–29 Mexico 57 military regimes 9–10 military rule 53 minority representation 55–56 modernization 57–58 Montealegre, Eduardo 134, 187–188 Montesquieu, Baron de 168, 209n1 Morales Hoyos, Viviane 113, 114–115, 133, 142, 226 Moreno, Luis Gustavo, extradition request for 246 Morón, Fabio 114 Movimiento Armado Quintin Lame 73 Movimiento Estudiantil por la Constituyente 66 Naranjo, Vladimiro 114 narco-cassettes 108, 111–112 Narco-Democracy 108 narcomico 112 narcotics trade 37, 41, 61, 63–64, 64–65, 78, 83n30, 110 Nariño, Antonio 42, 81n2 National Electoral Council 181 National Front 16–17, 37, 49, 55, 57, 59, 60, 62–64, 78–79 National Register of the Civil State (Registraduría Nacional del Estado Civil) 6, 190–191

268 Index Negretto, Gabriel L. 16, 27, 232–233 neo-constitutionalism 99, 100, 104, 228–229 neo-institutionalism 80–81, 231 Noguera, Jorge 130, 253 norms, significance 140 North, D. 22, 238 nuevo derecho 62, 72, 73–74, 145; and choque de trenes 99–101 Nunes, Rodrigo 22 Núñez, Rafael 44 Obdulio Gaviria, José 127 Odebrecht scandal 194 O’Donnell, Guillermo A. 9, 11–12, 22–23, 179, 228, 238 offices, periodization of 6–7, 97, 191 organized labor 51 pacted democracy 59–60 Pact of Ralito 124 Páez Varón, José Roberto 135 Palacios, Marco 51 Panama 44 parainstitutionalization 5, 117–139 paramilitarism, and politics 121 paramilitary groups 10–11, 20, 62, 248, 253–254, 259–261; influence 121; peace negotiations 120–121; political project 75–77; relationship with Uribismo 153–154n70 parapolítica scandal 5, 12, 18–20, 22, 71, 89–90, 95, 105, 117–139, 184, 224–226, 227, 234, 235, 246, 254; choque de trenes 132–139; Congressional reform 128; extradition of paramilitaries 128–129; fifth phase 129–131, `136–137; fourth phase 129, 136; impact 131–132; investigation 123–126, 125, 131, 136– 137, 142–144; judicial reform 127–128; legal facts 117, 133, 236; negative public discourse 131; negotiation phase 119–123; network of relations 130; offi­ cial complaint 122, 123; path dependence 139–149; personal attacks 127; phases 118–119, 118, 131; politics of 119–132; production of legal facts 148–149, 225–226; reporting on Uribe 124, 125; second phase 132–139, 135; third phase 135–137; third phase strategies 126–129; and tutela contra sentencias 132–139; tutelas 129–131; Yidispolítica. 129–130, 137–138, 235

parliamentary immunity 105, 136, 143, 147, 226, 227, 234 Partido Republicano 54–55 party system 9–10, 51–52, 117, 151n25, 228; base-line conditions 45–46, 47; collapse 75, 110, 179; Pastrana, Andres 133 Pastrana, Misael 66 path dependence 8, 88, 101, 223–227, 233–234; causal mechanisms 93; cen­ trality of initial decisions 92–93; choque de trenes 88–89; closed 140; definition 139; ideational components 94; judiciary 90–91, 91–105, 96; legal argumentation 139–149, 226–227; legal facts 139–149; logic 17–18; methodological caveat 93; ontological suitability 93–94; open 140; parapolítica scandal 139–149; patterns 92; proceso 8000 (Process 8000) 139–149; risk 91; sources 91–92; summary 91–95 path-dependent development 81 patronage systems 37, 41, 42–43, 59, 60, 75 Penal Code (Codigo Penal de Colombia, CPC) 111–113, 113 Pérez-Liñán, Aníbal 105, 108, 110 periodization of offices 6–7, 97 periodization, of offices 97 Petro, Gustavo 121 Pineda, Eleonora 121, 122 Pinilla Pinilla, Nilson 98, 138, 242–243 Pizarro, Eduardo 10 pluralism 74 political contestation, delegitimizing 62 political economy 62 political institutions 43; context 9 political legitimation 15 political participation 61, 221 political power, legal authorization 14 political rationality 39 political will, legitimate 174 Pollock, Friedrich 252 popular sovereignty 179 population 45 populism 176–177 power relations 11 power, uneven access to 58 precedent setting 139 presidential election, 1994 105, 141; Samper campaign 107–108; see also proceso 8000 (Process 8000) presidential election, 2002 11, 77, 120 presidential election, 2010 24

Index 269 presidentialism 5, 10, 12, 193–194, 235 presidential power 6–7, 10, 71, 192, 205 presidential re-election reform 2–5, 218–219, 227–230; aftermath 241–248; competence defects 196; Constitutional Court decision 1–2, 4, 21–22, 229, 235–236; critical moment, 2010 199– 200; delegative turn 174–175, 176–179; deliberative constraints 168–172; discursive institutionalism 160–209; fragmentation and co-optation of Con­ gress 179–184, 182, 183; importance of decision 161; making of substitution doctrine 184–195, 186; momentum 183–184; procedural flaws 199; rational choice versus deliberative institutional­ ism 162–173; rejection, 2010 195–204, 196, 197, 205–209; strategic vs. delib­ erative action 172–173, 173 Preteld de la Vega, Sabas 134 Pretelt, Jorge 204, 245–246 proceduralism 15 proceso 8000 (Process 8000) 18–19, 75, 89, 90, 104–105, 105–106, 105–117, 224–226, 227, 234; 1999 tutela deci­ sion 113–117; allegations against Samper 109; decision to absolve Samper 109–111, 111; duration 106; illicit enrichment charges 111–113; origins 107–108; and parliamentarians criminal liability 111–117; path dependence 139– 149; phases 105–106, 106–111, 107; preliminary investigations 108–109; Samper’s ability to remain in power 106–111; and tutela contra sentencias 111–117, 132, 144 process rights 136 proportionality principle 74 proportionality testing 101, 236 prudent strategizer, the 166–167, 172–173, 173, 175, 184, 187, 189, 195, 207–208, 227 public consultations 201–202 public opinion and discourse 252–254, 252, 255, 256 public powers 70 public reasoning 23 public spending 58 Ramseyer, J.Mark 166 rational choice, and deliberative institu­ tionalism 162–173 Rawls, John 169, 170 referenda 17, 66

reforma política (Political Reform); C-1040/05 187–189 regional power brokers, autonomy 110 Restrepo, Carlos Lleras 16–17, 58, 59 Restrepo, Luis Carlos 120 Reyes, General Rafael 9–10, 16–17, 38, 44, 54–55 Reyes reforms (1905–1910) 37, 38, 44, 49, 55, 56–57, 60, 78–79, 220 rights 70, 72, 73–74, 80, 81, 101, 163–164, 223, 224, 233; protection 99–101, 103; review 89 rights constitutionalism 101 Ríos-Figueroa, Julio 163, 209n1 Robinson, James A. 50, 57, 58 Rodriguez Orejuela, Gilberto 111–112 Rodríguez-Raga, Juan Carlos 166–167, 167, 184 Rojas Pinilla, General 9–10, 54, 59, 82n10 Rokkan, Stein 41, 46 Ruiz Medina, Jairo José 133 rule of law 1, 2, 7 Safford, Frank 51 Samper, Ernesto 89–90, 105, 185, 224; 1994 presidential election campaign 107–108, 141; ability to remain in power 106–111; allegations against 109; approval ratings 108; decision to absolve 109–111, 111; pork barrel spending 106, 110; see also proceso 8000 (Process 8000) Sanders, Elizabeth 16 Santa Fe de Ralito II agreement 121 Santos, Juan Manuel 247 Schmidt, Vivien A. 23, 80–81, 231–232 Schmitt, Carl 36, 188, 209n1, 235–236 security improvement 180 security privatization 75–76, 83n33 Semana 24 Senate, the 70; circumscription of 75 separation of powers 1, 2, 6–7, 11–12, 20, 88, 89, 98, 106, 114, 139, 165, 169, 209n1, 227, 235, 236, 241, 243 séptima papeleta (seventh ballot) campaign 66, 67 Serpa, Horacio 181 Sierra Porto, Humberto 191, 195, 198–199 social inequality 59, 60 social state of law 1 socio-economic development 45 socio-economic indicators 47 socio-economic tensions 52–53, 58

270 Index Spain 99 stability 12 state authority 44–45 state building 45 State Council 56, 97; choque de trenes 89 state power 43 states of exception 71 state weakness 42–43, 259–261, 259, 260 strategic action 206–207, 206–207, 219 strategic judges 160, 163–168, 168–172, 194, 227 Streeck, Wolfgang 88, 95, 147 student movement 16–17, 18, 37, 61, 71, 232; agency 42; communicative action 64–67, 80; composition 66; constitu­ tional reform and 62; formation 65; lea­ dership 67; manifesto 65; organizations 65–66 SU-047/99 decision 133 substitution doctrine 1, 2, 20–21, 23–24, 26, 30n2, 72, 131, 160–161, 162, 163, 172–173, 175, 184–195, 186, 206, 208, 211n29, 227–230, 236, 241, 242–244 Superior Council of the Judiciary 37, 80, 97–98; elimination 96; magistrate appointment 98 Supreme Court 18, 20, 71, 72, 80, 90; activism 131; chambers 97; choque de trenes 89, 99–101, 105–117, 132–139, 224, 226; congressional disciplinary power 71; and Constituent Assembly, 1991 67–68; defamation strategy 235; empowerment 148; function 89; judicial power 15, 235; magistrate selection 56, 97; parapolítica scandal 22, 117, 120, 122, 123, 123–132, 125, 143, 234, 246; path dependence 139–149; power to investigate Members of Congress 140; proceso 8000 (Process 8000) 105– 117, 110–111, 112, 113, 115–116, 234; relations with Uribe 129–131, 135; Sala Constitucional 61; scandals 241, 245, 246; tutela contra sentencias position 111, 132–139; via de hecho 138 Talero, Germán Navas 3 tax revenues 44 Thelen, Kathleen 19, 39, 88, 94, 95, 146–147, 226–227, 231, 234 Thoumi, Francisco E. 63 Thousand Days’ War 44, 49–51, 53, 78, 220 Tilly, Charles 43, 44 Title XIII 185

Todavia Podemos Salvar a Colombia 65, 66 Tovar Pupo, Rodrigo 76 transfuguismo 6, 191 transitional justice courts 147 transparency 14 traquetización 75 triangulation 28 tutela 18, 1, 973, 81, 88, 89, 89–90, 94, 99, 129–131, 141, 149, 223, 224; rights protection 99–101 tutela contra sentencias 90–91, 102–105, 144, 149; choque de trenes 132–139; proceso 8000 (Process 8000) 111–117 Unión Patriotica 64 United States of America 67, 164, 168, 246; Supreme Court 164, 169 United States of Colombia 43–44, 50, 108, 109 Uprimny, Rodrigo 116, 142, 178–179 urbanization 59 Uribe, Álvaro 1, 9–14, 120, 132–139, 227–228, 241; background 10–11, 210n4; Bonapartian posture 177–178, 207; charismatic leadership 26, 228; constitutional reform, 2004 71; defama­ tion strategy 235; delegative democracy 31n12; delegative leadership 174–175, 176–179; democracy measurements 12–14, 13; Democratic Security policy 2, 14, 119, 120, 174–175, 176, 179, 180, 205, 228, 254, 259, 260; dominance 207; election, 2002 11, 77, 120; exces­ sive presidentialism 5; family background 82n15; and FARC peace process 244, 247–248; first term 119–123; fragmen­ tation and co-optation of Congress 179–184, 182, 183; historical position 219; institutional decay under 192; majoritarian convictions 162; majority in Congress 179–180, 182, 182, 186; and making of substitution doctrine 184– 195; notions of natural leadership 14; parapolítica scandal 117–139, 254; parapolítica scandal reporting on 125; parapolítica scandal strategies 126–129; political capital 133–134; popularity 2, 7–8, 9, 25, 124–125, 126, 130, 161, 174, 176, 176, 179–180, 181, 186, 192, 194, 205, 207, 218–219, 229, 252; populism 176–177; position 8; power and influence 9, 21, 205; pre­ sidency 24; public discourse 252–254, 252, 255, 256; re-election 4–5, 132;

Index 271 reform agenda 8, 134; relations with Supreme Court 129–131, 135; scandals 194; second term 5–7, 207; security approach 11; sense of mission 175, 192; style of government 9 Uribe, Mario 127, 132, 135, 254 Uribismo 175, 176–179; relationship with paramilitary groups 153–154n70 Uribista coalition 12, 177, 178, 228, 254

Viejo derecho (Old Law) 38, 43, 47–48, 72–73, 74, 78, 111, 145, 221 violence 9, 25, 44, 49, 60, 79, 259–261, 259; containment 53–56; decline in 56–57, 74–75; kidnappings 260; mas­ sacres 83n31, 260; state monopolization of 63, 221 vote personalization 210–211n13 voter turnout 47

Valdivieso, Alfonso 108, 112, 151n24 Valencia, Fabio 127–128 Valenzuela, Arturo 11 Vargas, Luis Ernesto 245 Velásquez, Iván 127 Venezuela 177 vertical control 166–167 via de hecho 102–105, 115, 135, 138

war making capabilities 42, 43, 44 War on Drugs 108 Warren, Mark E. 22 Weber, Max 162 working class 37, 49; incorporation into the party system 51–52 Yidispolítica. 129–130, 137–138, 235