UNIVERSIDADE ESTADUAL DE CAMPINAS INSTITUTO DE FILOSOFIA E CIÊNCIAS HUMANAS CLARISSA CORREA NETO RIBEIRO OVERLAPPING ORGANIZATIONS, POLITICAL CRISES, AND COEXISTENCE: COMPLEMENTARITY AND FRAGMENTATION IN SOUTH AMERICAN AND AFRICAN REGIONALISMS SOBREPOSIÇÕES ORGANIZACIONAIS, CRISES POLÍTICAS E COEXISTÊNCIA: COMPLEMENTARIDADE E FRAGMENTAÇÃO NOS REGIONALISMOS SUL- AMERICANO E AFRICANO CAMPINAS 2020 CLARISSA CORREA NETO RIBEIRO OVERLAPPING ORGANIZATIONS, POLITICAL CRISES, AND COEXISTENCE: COMPLEMENTARITY AND FRAGMENTATION IN SOUTH AMERICAN AND AFRICAN REGIONALISMS SOBREPOSIÇÕES ORGANIZACIONAIS, CRISES POLÍTICAS E COEXISTÊNCIA: COMPLEMENTARIDADE E FRAGMENTAÇÃO NOS REGIONALISMOS SUL- AMERICANO E AFRICANO Tese apresentada ao Instituto de Filosofia e Ciências Humanas da Universidade Estadual de Campinas como parte dos requisitos exigi- dos para a obtenção do título de Doutora em Relações Internacionais, na Área de Institui- ções, Processos e Atores. Thesis presented to the Institute of Philosophy and Human Sciences of the University of Campinas in partial fulfillment of the require- ments for the degree of Doctor, in the area of International Relations. Supervisor/Orientador: Prof. Dr. Shiguenoli Miyamoto Co-supervisor/Coorientador: Profa. Dra. Karina Lilia Pasquariello Mariano ESTE TRABALHO CORRESPONDE À VERSÃO FINAL DA TESE DEFENDIDA PELA ALUNA CLARISSA CORREA NETO RIBEIRO, E ORIENTADA PELOS PROFS. DRS. SHIGUENOLI MIYAMOTO E KARINA L. P. MARIANO CAMPINAS 2020 Ficha catalográfica Universidade Estadual de Campinas Biblioteca do Instituto de Filosofia e Ciências Humanas Paulo Roberto de Oliveira - CRB 8/6272 Ribeiro, Clarissa Correa Neto, 1992- R354o RibOverlapping organizations, political crises, and coexistence : complementarity and fragmentation in South American and African regionalisms / Clarissa Correa Neto Ribeiro. – Campinas, SP : [s.n.], 2020. RibOrientador: Shiguenoli Miyamoto. RibCoorientador: Karina Lilia Pasquariello Mariano. RibTese (doutorado) – Universidade Estadual de Campinas, Instituto de Filosofia e Ciências Humanas. Rib1. Integração regional. 2. Proteção democrática. 3. América do Sul - Política e Governo. 4. África - Política e Governo. 5. Regionalismo - América Latina. 6. Regionalismo - Africa. I. Miyamoto, Shiguenoli, 1948-. II. Mariano, Karina Lilia Pasquariello. III. Universidade Estadual de Campinas. Instituto de Filosofia e Ciências Humanas. IV. Título. Informações para Biblioteca Digital Título em outro idioma: Sobreposições organizacionais, crises políticas e coexistência: : complementariade e fragmentação nos regionalismos Sul-Americano e Africano Palavras-chave em inglês: Regional integration Democracy protection South America - Politics and Government Africa - Politics and Government Regionalism - Latin America Regionalism - Africa Área de concentração: Instituições, Processos e Atores Titulação: Doutora em Relações Internacionais Banca examinadora: Karina Lilia Pasquariello Mariano [Coorientador] Brigitte Weiffen Detlef Nolte Haroldo Ramanzini Junior Regiane Nitsch Bressan Data de defesa: 21-02-2020 Programa de Pós-Graduação: Relações Internacionais Identificação e informações acadêmicas do(a) aluno(a) - ORCID do autor: https://orcid.org/0000-0002-7653-7635 - Currículo Lattes do autor: http://lattes.cnpq.br/2069725223189178 Powered by TCPDF (www.tcpdf.org) http://www.tcpdf.org UNIVERSIDADE ESTADUAL DE CAMPINAS INSTITUTO DE FILOSOFIA E CIÊNCIAS HUMANAS A Comissão Julgadora dos trabalhos de Defesa de Tese de Doutorado composta pelos Profes- sores Doutores a seguir descritos, em sessão pública realizada em 21 de fevereiro de 2020, considerou a candidata Clarissa Correa Neto Ribeiro aprovada. Profa. Dra. Karina Lilia Pasquariello Mariano Profa. Dra. Brigitte Weiffen Prof. Dr. Detlef Nolte Prof. Dr. Haroldo Ramanzini Junior Profa. Dra. Regiane Nitsch Bressan A Ata de Defesa com as respectivas assinaturas dos membros encontra-se no SIGA/Sistema de Fluxo de Dissertações/Teses e na Secretaria do Programa de Pós-Graduação em Relações Internacionais do Instituto de Filosofia e Ciências Humanas. This thesis is dedicated to Carolina, my little sister and great treasure; to my family; and to my adored husband, Rapha, for their unconditional love and support. AKNOWLEDGMENTS This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior – Brasil (CAPES) – Finance Code 001. I would like to thank my supervisors, prof. Shiguenoli Miyamoto and prof. Karina Mariano for their guidance through this thesis and for being professional inspirations. I thank professor Shiguenoli for accepting to supervise my work and for all the support provided during my Unicamp years. To Karina, I am immensely grateful for the work we did together over the past six years, the opportunities you provided me, and for accompanying me so closely on this path, offering your wisdom and friendship. I have not enough words. I thank the San Tiago Dantas Graduate Program, Unesp’s IPPRI, its faculty, staff, and student body, in the name of profs. Samuel Alves, Suzeley Khalil and Héctor Saint-Pierre, for providing a stimulating environment for the development of knowledge. Special thanks go to Graziela Oliveira, Giovana Vieira, Isabela Silvestre and Camila Magalhães, who do everything to make the student’s bureaucratic life easier with grace and sympathy. I am also very grateful to Unicamp and CAPES for the resources provided for the realization of this thesis. Besides the PhD Scholarship, I had a research stay abroad fully funded by the CAPES PDSE Program, and was also awarded a fieldwork grant to visit the African Union and do in loco investigations, and I am very thankful and conscient of how privileged I am because of it, in such times of economic constraints. This thesis would not have been the same without the opportunity of visiting the German Institute of Global and Area Studies – GIGA Hamburg, a center of excellency for regional research that had an essential role on shaping the thesis aspirations. I would like to greatly thank the Institute, on the figure of the academic director of the GIGA Doctoral Programme, Dr. Maren Wagner, and of Prof. Detlef Nolte, through whom I got to know the GIGA, and who accepted to serve as my supervisor for the research stay. Special thanks also are due to my colleagues at the doctoral program, without whom my German experience would also not have been the same. I feel very grateful for having on my committee such distinguished professors that have participated on my academic development during many of its stages. I thank prof. Britta Weiffen for participating on my qualification exam and for contributing so much for regionalism and international studies in Brazil through her actions at the Cátedra Martius. To prof. Nolte, I am very grateful for the help during my GIGA experience, to your kindness and availability. Britta and Detlef were among the first to conceptualize the phenomenon this thesis aimed to investigate, and I am thankful for the opportunity of discussing the findings and learning with them. To prof. Haroldo Ramanzini Júnior, thank you for the incentives to explore Comparative Regionalism on this research, for supporting my activities, for debating and contributing to my work from a very early stage. To prof. Regiane Bressan, my acknowledgments and thanks have their roots on my master’s degree and grew through my doctoral studies, as she acted as my supervisor for Teaching Assistantships, provided me with opportunities to participate on events and contributed to the research development as a member of this committee. Haroldo and Regiane are also members and coordinators of the research group I am happy to participate, Rede de Pesquisa em Política Externa e Regionalismo (REPRI), and I am extremely grateful for having their presence during my academic development and on this committee. Professors Roberto Goulart, Alexandre Fuccille, and Cairo Junqueira were the alternate members chosen to compose this committee, and also members of the research groups REPRI and Observatório de Regionalismo (ODR), and I am privileged for having their presence during my academic development. At REPRI, special thanks to Bruno Theodoro Luciano who is an honorary member of this committee and helped me so much on this journey. A great perk of the field of International Studies is its contemporaneity and connectivity. I feel very lucky that through this work, I was able to personally meet or exchange emails with many of the researchers that are a fundamental part of my case specific literature and received nothing but kindness and support. The composition of the judging committee is also a proof of it. In this sense, many thanks are due to prof. Ulf Engel, who helped me with the agenda setting for the African Union fieldwork, his advices and openness. I also thank dr. Frank Mattheis for the comments during this process and for the opportunity of participating with prof. Karina with a chapter on his and professor Engel’s book. I thank profs. Philippe de Lombaerde and Michel Levi, for the opportunity of participating at the UNU-CRIS/ UASB 6th Doctoral School on Comparative Regionalism, for the support received on a form of a scholarship for participation on this enriching event and for the thoughtful comments. To Prof. Camilo López, who was my supervisor during a master’s degree research stay at the Universidad de la Republica in 2015, but whose support accompanied my doctoral journey. I am deeply indebted to his sharp comments on my research and for his incentives during doctoral workshops. To all of these professors and researchers, in the name of many more, I am also grateful for the opportunities I was given even at an early stage on my career. I would like to thank the people at the African Union Headquarters in Addis Ababa for the receptivity and willingness to help during my fieldwork. Special thanks go to the officials of the AU Department of Political Affairs, the Peace and Security Council and the archivists who took their time to talk to me and help me gather the empirical data for this research. Being at the AU was one of the most unforgettable experiences of my life and reaffirmed the goals of producing knowledge in order to favor regionalism. On a more personal note, I would like to thank the support provided by the friends I had the pleasure of having by my side through the years. Lis Barreto and Matheus Oliveira once shared a home with me during the two years of the master studies, and we have ever since shared so much more from life and dreams, I could not be more thankful. I deep heartedly thank Sharinee Jagtiani and Matheus Hebling who made my life better by sharing the GIGA experience with me. And I also thank Julia Borba and Barbara Neves on behalf of all of my colleagues at the research group Observatório de Regionalismo, for the companionship in concretizing the project of developing and making regionalism studies available to the public. I definitely would not have made through academia without your support during the many moments I felt like giving up. To my life friends and to my cousins, I am grateful for the laughs, the confidences and the rest you have offered me. I especially thank Amanda Ruas, who besides her constant and supportive presence, provided for a careful English language revision before publication. To my family, I am extremely thankful for all the incentives and support I received from my dear grandparents, aunts and uncles, especially my aunt Flavia, who was present in every step of the way, and to my grandma Edir, who unfortunately did not live to see me complete this journey and is so terribly missed. I am also very grateful to my in-laws for being such a safe haven over the years in São Paulo. All of the abovementioned gratitude is not enough to express my feelings for a very special group of people: my loving husband, Raphael, who has been the greatest partner I could have ever asked for, in every sense of the word; my parents, Claudia and Edgard, and my siblings, Fabricio and Carolina, who were tireless in rooting for me, supporting my decisions and understanding my moments. Even the completion of this PhD thesis writing would not have happened without them. This kind of love has no price, and I really have no words for it, as I cannot thank or express how much I love you enough. Last, but not least, I feel eternally grateful and loved by God, who stood by me through every step of this journey. The real tragedy of our postcolonial world is not that the majority of people had no say in whether or not they wanted this new world; rather, it is that the majority have not been given the tools to negotiate this new world. (ADICHIE, C. N., 2007, p. 129, highlights by the author). ABSTRACT The present research has the occurrence of overlapping regionalisms as its object, with a comparative cut for South America and Africa. The purpose of this thesis is to study types of institutional interaction and to analyze the effects of overlapping in the coexistence between regional processes in order to understand their dynamics in the above-mentioned regions. The observation of organizational behaviors and interaction is made through cases of political crises on the selected regions, as we assume that the interactions between overlapping regional organizations in those situations would tend to be more fragmenting than complementary as they go beyond the traditional membership-mandate relationship. Therefore, we also analyze overlapping performances and seek to understand how regional organizations will behave and interact while dealing with political crises, which constitute critical junctures, in accordance with the theoretical and methodological framework of the thesis. This work aims to produce a broad qualitative analysis of the given regions by considering applicable cases that cover all of the sub-regions and hence to contribute to the field of comparative regionalism by providing generalization and institutional learnings which are not derived by a given model of regionalism, but nonetheless based on a cross regional approach. The data is processed through document analysis and further systemized via the construction of presence and absence exercises, consisting on testing of hypotheses and dyadic relationships. Keywords: regionalism; overlapping organizations; regional integration; South America; Africa; political crises; democracy protection. RESUMO A presente pesquisa tem por objeto a ocorrência de regionalismos sobrepostos, com recorte comparativo para a América do Sul e a África. O propósito desta tese é estudar tipos de interação institucional e analisar os efeitos da sobreposição na coexistência entre processos regionais, de modo a entender sua dinâmica nas regiões supramencionadas. A observação de comportamentos organizacionais e interações é feita através de casos de crises políticas nas regiões selecionadas, uma vez que se assume que ditas interações entre organizações regionais sobrepostas tenderiam a produzir efeitos mais fragmentantes do que complementares para o espaço, uma vez que as mesmas ultrapassam a tradicional relação entre filiação e mandato. Portanto, se analisa também a sobreposição de performances enquanto se procura compreender como as organizações regionais irão se comportar e interagir ao lidar com crises políticas, que constituem conjunturas críticas, de acordo com o referencial teórico- metodológico da pesquisa. A tese pretende produzir uma análise qualitativa de grande dimensão das regiões dadas, considerando casos aplicáveis que abarquem todas as sub-regiões e, assim, contribuir para o campo do regionalismo comparado, fornecendo generalizações e aprendizagens institucionais que não são derivadas de um dado modelo de regionalismo, mas, baseadas em uma abordagem regional cruzada. Os dados são processados através da análise de documentos e posteriormente sistematizados através da construção de exercícios de presença e ausência, consistindo no teste de hipóteses e relações diádicas. Palavras-chave: regionalismo; organizações sobrepostas; integração regional; América do Sul; África; crises políticas; proteção democrática. RESUMEN La presente investigación tiene como objeto la superposición de regionalismos como su objetivo, con un corte comparativo para América del Sur y África. El objetivo de esta tesis es estudiar los tipos de interacción institucional y analizar los efectos de la superposición en la coexistencia entre procesos regionales para comprender su dinámica en las regiones mencionadas. El objetivo de esta tesis es estudiar los tipos de interacción institucional y analizar los efectos de la superposición en la coexistencia entre procesos regionales para comprender su dinámica en las regiones mencionadas. La observación de los comportamientos e interacciones organizacionales se realiza a través de casos de crisis políticas en las regiones seleccionadas, ya que suponemos que las interacciones entre las organizaciones regionales superpuestas en esas situaciones tenderían a ser más fragmentarias que complementarias, ya que van más allá de la relación tradicional de membresía-mandato. Por lo tanto, también analizamos desempeños superpuestos y buscamos comprender cómo se comportarán e interactuarán las organizaciones regionales al enfrentar crisis políticas, que constituyen coyunturas críticas, de acuerdo con el marco teórico y metodológico de la tesis. La tesis tiene como objetivo producir un análisis cualitativo amplio de las regiones dadas al considerar los casos aplicables que cubren todas las subregiones y, por lo tanto, contribuir al campo del regionalismo comparativo al proporcionar generalizaciones y aprendizajes institucionales que no se derivan de un determinado modelo de regionalismo, pero basado en un enfoque regional cruzado. Los datos se procesan a través del análisis de documentos y se sistematizan aún más mediante la construcción de ejercicios de presencia y ausencia, que consisten en la prueba de hipótesis y relaciones diádicas. Palabras clave: regionalismo; organizaciones superpuestas; integración regional; Sudamerica; África; crisis políticas; protección democrática. LIST OF FIGURES Figure 1 – Consequences of institutional overlap.............................................................. 26 Figure 2 – Variants of overlap…………………………………………………………... 38 Figure 3 – Explaining ROs horizontal and vertical overlappings.…………….……….... 39 Figure 4 – Overlapping regional mechanisms in South America...................................... 47 Figure 5 – Scramble for Africa ……………………….………………………………… 50 Figure 6 – Overlapping regional mechanisms in Africa………………………………… 54 Figure 7 – Successful coups d’état in Africa 1963-2018………………………………... 62 Figure 8 – ROs’ reactions to political crises in overlapping scenarios………………….. 151 LIST OF TABLES Table 1 – Types organizational behavior and corresponding regional result………………... 29 Table 2 – Compilation of events treated by the AU and the RECs as UCGs after the estab- lishment of the legal instruments 2004-2017……………………………………… 60 Table 3 – Compilation of events treated by the AU and the RECs as UCGs after the establishment of the legal instruments 2004-2017, with West Africa in highlights. 81 Table 4 – “Democratic Density” on African sub-regions……………………………………. 109 Table 5 – Compilation of the sub-regional normative framework on UCGs……………........ 111 Table 6 – Authoritarian regimes in South America until 1990…………………………….. 114 Table 7 – Democratic breaches in South America with overlapping ROs………………….. 115 Table 8 – Commonalities among provisions on representative democracy at the OAS and the AU……………………………………………………………………………... 145 Table 9 – Updated typology of organizational behaviors……………………………………. 152 Table 10 – Dyadic analysis of overlapping ROs’ behaviors in Africa………………………… 154 Table 11 – Dyadic analysis of overlapping ROs’ behaviors in South America……………….. 154 Table 12 – Regional organizations’ mandates and actions on democratic breaches………...… 158 LIST OF ACRONYMS AEC African Economic Community AFISMA African-led International Support Mission to Mali ALBA Bolivarian Alliance for the Peoples of Our America AMIB African Union Mission in Burundi AMU Arab Maghreb Union APSA African Peace and Security Architecture AQIM al-Qaeda in the Islamic Maghreb AU African Union CAN Andean Community CELAC Community of Latin American and Caribbean States CEMAC Central African Economic and Monetary Community CEN-SAD Community of Sahel–Saharan States CEPGL Economic Community of Great Lakes Countries CNDD-FDD Conseil National pour la Défense de la Démocratie et Forces de Défen- se de la Démocratie CNRDRE Comité National pour le Redressement de la Démocratie et de la Res- tauration de l’État COMESA Common Market for Eastern and Southern Africa COPAX Council for Peace and Security in Central Africa EAC East African Community EASBRICOM East Africa Brigade Coordination Mechanism ECCAS Economic Community of Central African States ECLAC (CEPAL) Economic Commission for Latin America and the Caribbean ECOMIG ECOWAS Military Intervention in The Gambia ECOWAS Economic Community of West African States FOMAC Force Multinationale de l’Afrique Centrale FTAA (ALCA) Free Trade Area of the Americas HAT Hight Authority of the Transition IADC Inter-American Democratic Charter ICG-CAR International Contact Group - Central African Republic ICG-M International Contact Group - Madagascar ICGLR International Conference on the Great Lakes Region IGAD Intergovernmental Authority on Development IOC Indian Ocean Commission JMT-M Joint Mediation Team - Madagascar LAFTA (ALALC) Latin American Free Trade Association LAIA (ALADI) Latin American Integration Association MAPROBU African Prevention and Protection Mission MARAC Central African Early Warning System MCPMR OAU Mechanism for Conflict Prevention, Management and Resolution MDP Mechanism for Democracy Protection MERCOSUR Southern Common Market MINUSCA United Nations Multidimensional Integrated Stabilization Mission in the Central African Republic MISCA African-led International Support Mission for the Central African Re- public MNLA Mouvement National de Libération de l'Azawad MoU Memorandum of Understanding on Cooperation in the area of Peace and Security between the African Union, the Regional Economic Communities and the Coordinating Mechanisms of the Regional Standby Brigades of Eastern Africa and Northern Africa MRU Mano River Union MUJAO Mouvement pour l'Unicité et le Jihad en Afrique de l'Ouest NAFTA North America Free Trade Agreement NARC North Africa Regional Capability OAS (OEA) Organization of American States OAU Organization of African Unity OIF International Organization of Francophonie PSC Peace and Security Council RECs Regional Economic Communities RISDP Regional Indicative Strategic Development Plan RMs Regional Mechanisms RO Regional Organization SACU Southern African Customs Union SADC Southern African Development Community SADCC Southern African Development Coordination Conference SAFTA (ALCSA) South American Free Trade Area SIPO Strategic Indicative Plan for the Organ UCGs Unconstitutional Changes of Government UDEAC Central African Customs and Economic Union UN United Nations UNASUR Union of South American Nations UNECA United Nations Economic Commission for Africa WAEMU West African Economic and Monetary Union SUMMARY 1 INTRODUCTION ......................................................................................... 20 1.1 1.2 1.3 1.4 1.5 2 2.1 2.1.1 2.1.2 2.2 2.3 2.3.1 2.3.2 2.3.3 2.4 2.4.1 2.4.2 2.4.3 2.4.4 2.5 3 3.1 3.1.1 3.1.2 3.2 3.2.1 3.2.1.1 Research question and central argument………………………………….. Theoretical and methodological framework………………………………. Case selection and thesis’ scope…………………………………………….. Research procedures………………………………………………………... Summing up: contributions of the cross regional institutional behavior analysis………………………………………………………………………. REGIONALISM IN THE GLOBAL SOUTH: INSTITUTIONAL MAPPING OF THE SCENARIOS OF OVERLAPPING………………... Overlapping regional organizations and regional interplays ……………. Configurational types of overlap……….……………………………………. Potential outcomes’ scenarios: why overlappings matter……..……...……… Drivers of regionalism..................................................................................... Latin America: South American studies……................................................ Closed Regionalism........................................................................................... Open Regionalism............................................................................................. Post-hegemonic Regionalism………………………………………………… Africa................................................................................................................ Colonial Arrangements...................................................................................... Pan-Africanism versus post-colonial sub-regionalism……………………….. The not-so-new “New Regionalism”………………………………………… African Union Era……………………………………………………………. Analysis and mapping of organizations: partial considerations…………. OVERLAPPING ORGANIZATIONS AND POLITICAL CRISES IN AFRICA……………………………………………………………………... The African Union and Unconstitutional Changes of Government……... Normative Provisions on UCGs........................................................................ Provisions on the relationship between the AU and sub-regional groupings… Regional and sub-regional responses to UCGs……………………………. East Africa........................................................................................................ The Burundian crisis, 2015: UCG or not UCG?............................................... 21 27 30 33 34 35 36 38 39 41 43 44 45 45 48 48 51 52 53 55 58 61 63 68 73 73 75 3.2.2 3.2.2.1 3.2.2.2 3.2.3 3.2.3.1 3.2.4 3.2.4.1 3.2.5 3.3 4 West Africa…………………………………………………………………… Mali, 2012: ECOWAS, CEN-SAD, AU……………………………………… The Gambia, 2017: ECOWAS, CEN-SAD, AU……………………………... Southern Africa………………………………………………………………. Madagascar, 2009: SADC, COMESA, AU…………………….……………. Central Africa.................................................................................................... Central Africa Republic, 2013: ECCAS, CEN-SAD, AU................................. Northern Africa................................................................................................. Partial Analysis................................................................................................ OVERLAPPING ORGANIZATIONS AND POLITICAL CRISES IN SOUTH AMERICA………………………………………………………… 79 81 86 88 91 97 100 105 107 113 4.1 4.2 4.2.1 4.2.1.1 4.2.2 4.2.2.1 4.2.2.2 4.3 5 5.1 5.2 5.2.1 5.3 6 6.1 6.2 Normative provisions on democracy in the American continent.………... Sub-regional responses.................................................................................... The Andean sub-region…………………………………………………..…... Ecuador 2010: ALBA, CAN, UNASUR, OAS………..……………………... The Southern Cone............................................................................................ Paraguay 2012: MERCOSUR, UNASUR, OAS............................................... Venezuela 2013-onwards: ALBA, MERCOSUR, UNASUR, OAS................. Partial Analysis................................................................................................ COMPARATIVE REGIONALISM, POLITICAL CRISES, DE FACTO OVERLAPPINGS AND REGIONAL COHESION…...….......................... Regionalism and political crises: challenges to shared democracy protection…………………………………………………………………….. Regional behaviors on overlappings situations……………………………. Analysis of dyadic relationships……………………………………………… De facto overlappings: Regional Organizations in action………………… FINAL CONSIDERATIONS……………………………………………….. Potential causalities of overlapping ROs’ behavior choice on African and South American political crises………………………………………... Summing up the observed effects of overlapping actions on political cri- ses for regional cohesion: complementarity or fragmentation?.................. 117 123 124 126 128 130 133 138 142 145 149 153 157 160 161 164 REFERENCES................................................................................................ 167 APPENDIX A – REGIONAL ORGANIZATION BUILDING OVER TIME………………………………………………………………………… APPENDIX B – SUCCESSFUL COUPS D’ÉTAT IN AFRICA 1963- 2018………………………………………………………………………….. APPENDIX C – OTHER CASES………………………………………….. APPENDIX D – POLITICAL CRISIS IN SOUTH AMERICA………….. APPENDIX E – COMPILATION OF MAIN ROs’ BEHAVIORS THROUGH CASE STUDIES………………………………………………. 181 182 183 190 191 20 1 INTRODUCTION “The world of regions is also a world of regionalisms” (BACH, 2016). By adapting Katzenstein’s (2005) comprehension of contemporary world dynamics, Bach affirms the essentiality of understanding the particularities of regional logics in order to navigate the international system. Even though the concept of region is susceptible to interpretation and can change over time, what is interesting to this research is the formal relationship between the stablished regional organizations within those given spaces, as we understand regions as political constructions (POWERS; GOERTZ, 2011). Bach’s words are particularly relevant when we consider that in the last few decades, different kinds of regional projects have been developed in all different regions of the world, creating multilayered interactions within projects, and an “alphabet soup” of acronyms. Therefore, the understanding of the multiplicity, allied to the simultaneity in the coexistence of regional institutions, which many times are overlapping, generates questions about the effects of this coexistence for the functioning of the regions, as well as regarding their cohesion or fragmentation. The comprehension of a segmented proliferation of institutions in South America and its different orientations and purposes guided the author’s master's studies, in which we institutionally analyzed the fragmentation of integration at the regional organizations that Brazil participated, namely the Southern Common Market (MERCOSUR), the Union of South American Nations (UNASUR) and the Community of Latin American and Caribbean States (CELAC), departing from their policy agendas. The studies carried out demonstrated the existence not only of an institutional overlap between different blocs, that is, of multiple regional initiatives, but also that the different moments of regionalism, also known as waves, had produced an intra-bloc sectorial segmentation derived from the institutional broadening of existing initiatives and the expansion of mandates (RIBEIRO, 2016) 1. Consequently, the understanding of simultaneity in regional overlaps in other areas of the world raised questions about how organizations behave and interplay in a complex regional scenario, and their role in generating complementarity or fragmentation. The present study proposes, therefore, to discuss how regional organizations will act and interact, 1 In some cases, discussed in the master's dissertation, it was even perceived that the intra-bloc proliferation of institutions causes effects such as the multiplication of instances with the same sectoral thematic, which generates a greater internal need for coordination efforts (RIBEIRO, 2016). 21 considering the effects of the multiplicity and overlap of organizational initiatives in the selected regions of the world. For this doctoral thesis, the choice of cases to be analyzed was given in a way to cover two different regions of the world, Africa and South America, in order to enable generalizations and comparisons. The choice was made to avoid the dilemma of treating a single experience as a reference (n = 1) (DE LOMBAERDE et al., 2010), valuing the individual experiences of each region, following the methodology of comparative regionalism, as we will further discuss. Accordingly, we proposed to continue the studies initiated in the master's degree regarding South America, and to add another region in the Global South to the analysis as it also avoids an asymmetric comparison with the European experience, by representing the developing world as a fruitful area for the development of regional blocs. Therefore, considering the expansion of the object of analysis, with the addition of another region and the previous inexperience with different conjunctures, we departed from an inductive research to understand regional scenarios and to choose the scope of analysis. 1.1 Research question and central argument The studied phenomena of overlapping regionalisms result from the fact that multiple regional organizations (ROs) coexist in the same space and it occurs when states are affiliated with several of these initiatives (overlap of memberships), which in turn have similar objectives in their policies (overlapping mandates). While in the African case there is a shortage of literature on the term, with more attention given to regime complexity (BROSIG, 2011), in Latin America’s case, overlapping regionalism has been widely discussed by experts in recent years (RIGGIROZZI; TUSSIE, 2012; MALAMUD; GARDINI, 2012; MALAMUD, 2013; WEIFFEN; WEHNER; NOLTE, 2013; GÓMEZ-MERA, 2015; WEIFFEN, 2017; NOLTE, 2014, 2016, 2018). In general, the impressions tend to imply that the existing organizational overlap would be less negative than expected, or more of a “theoretical problem than a practical challenge” (NOLTE, 2018, p.129). More recently, however, the literature has also admitted the possibility of overlapping in action, or the ROs’ actual performance, which constantly engage in policies which are not 22 explicit to their obvious mandate2 (NOLTE; WEIFFEN 2018). This was also one of the conclusions obtained during the master's studies concerning the de facto overlapping, not just de jure. Even though results also showed initially greater opportunities of cooperation than of competition between ROs in the cases analyzed, they have also raised concerns about their segmentation and friction. On this point, the literature has been quite fruitful in recent years in providing possible effects caused by overlapping organizations, but there are still scarce empirical analyses of its actual effects (PANKE, STAPEL, 2016), mainly if we consider that throughout their existence, these organizations usually go through a process of expanding and refining its action in order to achieve the goals of its member states and remain significant for the region they belong. The effects produced by institutional enlargements on the broadening of the scope of action of an organization results then on its original mandate not always comprising the totality of its sectoral performance. Thus, the idea that the plurality of blocs with overlapping mandates and members would already have potential for fragmentation is then reinforced by the realization that, through overlapping in action, regional policy coherence could be further reduced. It should be noted, nonetheless, that when selecting and analyzing the cases, the coexistence of multiple overlapping projects per se is not considered as negative but as an element for the discussion the effects caused by the interplay between these institutions for the coexistence and the construction of regional cohesion. In fact, according to Nolte (2014; 2016; 2018), a scenario of overlap could potentially generate effects for regional governance that would range from indifference and non-interaction to increased friction, competition and consequent regional fragmentation, to potential strengthening and cooperation between projects. These could also be comprehended into four main possibilities: synergy, cooperation, conflict and segmentation, which provide a framework to further elaborate on expected patterns of interaction between ROs. More interestingly, by borrowing from the experience of the European integration project, the author exemplifies the positive potential of institutional plurality, referring to studies by Hoffmann and Mérand (2012) who argue that regional negotiations are less likely to fail when states have the possibility to opt-out of policies that do not correspond with their interests; thus establishing the concepts of regional integration at different velocities and agreements with “varied geometries” (HOFFMANN, MÉRAND, 2 A good example of this statement would be the case of Mercosur. Even though the original or explic- it mandate of the bloc was centered on trade, it has since evolved to include sectoral themes such as health or education. 23 2012). However, the type of institutions may constitute an important factor when analyzing the potential of the institutional plurality as the theoretical discussion does not dedicate special attention to the fact that the organizational logic of a region may be composed of informal or limited institutionalized systems with scarce capabilities, which could hamper the positive effects of the multeity of ROs. Therefore, some authors diverge from these perspectives, as we have previously discussed3. Malamud and Giardini (2012) argue that overlapping doesn't necessarily produce a competitive regionalism, but rather segments it, facilitating decentralization at the sub-regional level. Another author who has reflected on the effects caused by the complexity of the international regime for regional governance is Laura Gómez-Mera (2015). In her analysis, Gómez-Mera does not disregard the possibilities for synergy and cooperation between organizations. However, she identifies three main mechanisms that sabotage the proliferation of international agreements and the effectivity of cooperation: legal fragmentation; the use of institutional strategies; and constant competition that causes the regional unity to unravel. These challenges proposed by Gómez-Mera also add interesting features to the perception of what regional behaviors could look like, considering the tactics behind the use of such mechanisms. This could contribute to states losing confidence in, or even abandoning regional integration altogether in favor of uni-or multilateral initiatives. The absence of direct conflict between institutions, which might be considered an opportunity for synergy and cooperation, should not be promptly interpreted as a positive effect; often, regional blocs experience serious difficulties in meeting their objectives, while states often opt to negotiate directly rather than in the regional context. In addition, we consider that institutional overlapping can produce outcomes that might not be regarded monolithically. For instance, the same action can be positive for crises’ resolution and at the same time generate friction within actors, such as when two ROs have the same position, thus generating positive reinforcement, but simultaneously, they compete for the leadership on the conflict resolution. This friction generates ambiguity, hijacks institutional legitimacy and therefore can delay or bypass the results. In this way, given the questions and arguments raised, and, as previously discussed, we broadened the research regions to evaluate the interplay of overlapping institutions around the 3 Cf. RIBEIRO; MARIANO, 2017. 24 world, and to propose generalizations about the effects of the phenomenon. As Nolte (2014, p.9) recalls the statements by Brosig (2010, p.37), “competition between organizations is most likely if there is significant overlap in core competence fields or if one core area of one organization threatens to dominate a peripheral policy area of another organization”, whereas overlapping in policy areas of secondary importance has a greater chance of generating cooperation, we have chosen to analyze a particularly sensitive issue to ROs, that would therefore demand more institutional action and provide more data to the research. In fact, we continued the policy agenda analysis used in the previous research stage (RIBEIRO, 2016)4, in order to advance in the studies beyond the member/mandate relationship, but we singled out a specific subject to be studied cross-regionally, one of the most sensitive issues to the blocs as they would be related to their cores of power: political crises. By political crises, we intend to bring together the different terms used in the two regions to refer to the incidents such as coups d’état; attempted coups; undemocratic seizing or maintenance of power; and violations of the constitutional order; which are known as democratic ruptures or attempted ruptures, and invoke the use of mechanisms for democracy protection (MDPs) in South American Regionalism; and as Unconstitutional Changes of Government, by African regulations. This definition is derived from the normative context of democratic clauses, democracy protection and good governance, as specifically described by each ROs’ legal instruments. Those terms will be further addressed as we analyze the normative provisions in both regions throughout the case studies. The choice for political crises is also in accordance with the theoretical framework of the thesis, which departs from the institutionalist perspective, particularly its historical strand (KEOHANE, 1989; PIERSON, 2004). Accordingly, as we study regional organizations, their contextual background, design and moment in time will have key roles in providing explanations on the effects of overlappings to ROs. The theory further stresses that there are moments which are propitious to conduct these analyses, namely critical junctures, or “a relatively short period in time during which an event or a set of events occurs that has a large and enduring subsequent impact” (MAHONEY; MOHAMEDALI; NGUYEN, 2016). Ikenberry (1994) explicitly cites political crises as critical junctures, therefore allowing their choice as object of analysis in an institutionalist framework, as they provide change to the scenarios. 4 The author’s master dissertation was entitled “Overlapping regionalism and institutional prolifera- tion in Latin America: complementarity and fragmentation in regional agendas”. It is available, in Portuguese, on: . 25 Moments of political crises are also specially interesting in the light of current world events, which are giving rise to nationalist claims in opposition to communitarian approaches, as shown by the recent episode of the United Kingdom demand to leave the European Union, which became known as Brexit. As globalization advanced and has transformed international relations into a more porous set of interactions, regions still have central roles on conflict resolution and governance, as crises have also evolved to be regional in their nature, once they hardly stay confined to a country’s national border. Another good example to this last statement would be how the economic and political crises in Venezuela have led to a refugee crisis on neighboring countries. Overlapping regionalisms do not depend on political crises but will have effects more prominently seen on those moments, serving to the testing of our hypothesis on the potentialities of overlappings to generate conflict. The relevance of critical moments is also found on the literature on regime complexity, which has also served to base early approaches on overlapping regionalism and has been specifically used to understand African institutions. The core concept here is related to institutional interplay as stated by Brosig (2011; 2013): interaction between institutions would be triggered by two reasons: diverging norms; and the obligations derived from the institutional competence. Hence, both cases confirm the relevance of moments of crisis as relevant to measure since they foment the need of interaction, as they generate pression on normative fields for the interplay. On its turn, the interplay itself can generate two effects: empowering and integrative actions, or disruptive/conflictive action. The literature has since stated that in matters of political and security cooperation, which include therefore our understanding of political crises, a mixed picture tends to emerge when looking at interactions in instances where several ROs have become involved in managing crises or conflicts considering that cross institutional action has either led to advancing cooperation or to blockades (WEIFFEN 2017; NOLTE; WEIFFEN 2018). Thus, considering all the above-mentioned arguments, some questions could be proposed. According to the institutionalist theory (KEOHANE, 1989), institutions are instruments that would facilitate the trust of neighbors through the sharing of information and increase predictability. However, considering the coexistence of several institutions, how do we predict behavior or consolidate practices? What are the effects of segmented regionalism? Those questions serve as auxiliary as they have helped shaping what would become the central problem of this doctoral research, which is consisted by a two-folded question: 26 • How do overlapping ROs behave when they have shared competence over an issue and how can they influence regional cohesion or fragmentation? The underlying motivation behind the questioning would then be the concern that institutional overlap would be a generator of regional fragmentation through the friction in the coexistence of different organizations. The aim is to conduct not an outcome-driven research, but a process-driven one, as we are not specifically looking at conflict resolution is reached but rather at the strategies adopted by ROs throughout critical events in order to interact in a complex system. This goal can be further illustrated by Figure 1 below on the consequences of institutional overlap. As the redundancy of ROs will ultimately lead to policy outcomes, our focus rests on the two previous stages, namely the actor strategies and inter-organizational dynamics, to which we seek to perceive the presence or absence of interaction patterns in overlapping situation. Ultimately, when we question the effects to regional governance, what we mean to analyze are the outcomes of these organizational behaviors and interactions to regional cohesion, not necessarily in terms of policy outcomes, but in terms of coexistence. Under what conditions interactions generate complementarity between regional blocs or provide friction and fragmentation? Figure 1- Consequences of institutional overlap Source: NOLTE; WEIFFEN, 2018; with our highlight (circle). 27 1.2 Theoretical and methodological framework Considered the proposed questions and object, the methodological background adopted relies on Comparative Regionalism in order to understand the common conjuncture presented in the analyzed regions. This field has been subject to dubious assessments in the past, as European integration was put into perspective as a model experience and there was no consensual clarity in the definition of its elementary concepts. Thus, area specialists have also strongly criticized the cultural relativism employed in these studies and stressed the need for a deep multidisciplinary knowledge of complex contexts and peoples (SÖDERBAUM, 2008). However, as the studies developed, a consensus in the field of study concerning the practice of regionalism was reached and as Acharya (2012) points out, regionalism is a socially constructed political movement whose form depends directly on the domestic and transnational interests involved in the process, allowing therefore the comprehension that it encompasses not only state actors but also non-governmental actors within a given region. Considering the social aspect of the construction of regional initiatives, the need for field studies to be conducted from non-European perspectives in order to avoid Euro- Standardization of regional processes conducted in different contexts, is derived as Sbragia (2008), Söderbaum (2008), De Lombaerde et al (2010) and Acharya (2012) further point out. The idea of comparison helps in the prevention of ethnocentric and cultural interpretations, but in order to make it feasible, it is necessary to create new generalizations or invalidate previously established generalizations (AXLINE, 1994). This is why we have chosen to work with two cases, two regions that offer the perspective of the global South to provide a solid basis for analysis. The comparison in studies of comparative regionalism offers the opportunity of avoiding the narrowness of a single experience, or parochialism, while testing the validity of assumptions externally. It allows reflections on the meanings of concepts such as “regional integration” and “regionalism” from non-Eurocentric perspectives and enables the accounting of common challenges and perspectives. Therefore, comparison on the field does not mean the valuative action of determining which region has “the best” approach to regionalism, but it relies on the possibility of enhancing the understanding of regional dynamics on the basis of commonalities among them. Thus, given the conjunctural similarity of the Global South on the studied regions, the analysis of the institutional framework’s evolution could enable the debate on resources to promote development in the existing blocs. In addition, the study of regions that could 28 arguably be considered as global peripheries provides for the analysis of the relationship between colonialism, integration and the development of autonomy in those regions. In order to provide the study of patterns of organizational behavior or response to be compared, in addition to the political and historical contexts, the analysis should focus on the formal institutional structures that make up the regional processes to perceive their dynamics and scope of action, without disregarding the regional organization’s performances that are developed ad hoc or informally. According to Acharya (2012), in this sense, we once again highlight the institutionalist theory of international relations, which starts to question not the validity of regional institutions, but the how they matter. Therefore, as we have previously stated, we intend to analyze the documents and actions from an institutionalist point of view (KEOHANE, 1989). As regarding the how they matter propositions, through the use of the two-level game dynamics (PUTNAM, 1988) also comes a response to the comprehension of the organizational behaviors, as it considers the spheres in which institutional actions are inserted. In this sense, we adapt the author’s propositions, which have been previously used on Foreign Policy analysis5, to reflect on ROs’ behavior as a result of a binomial game. Putnam’s contribution, which states that the position of a State at the international level would be the result of the intersection of the interests of governmental and non-governmental actors is therefore adapted to understand the position of a regional organization. Thus, regional interplay would be the result of the game between the sub-regional and the regional, or the international level, whose common positions and conflicts would lead to the formation of a given region’s interests and would thus shape these actors’ behaviors. These types of interaction would fit then into a model to study strategic choices on given situations of imperfect competition, specifically contributing to the understanding of how actions can affect different actors, making the outcomes of a situation the results of the actors’ performances choices, as we preliminarily draw on table 1 (p.29), and further develop on chapter 5, after the empirical analysis. The types of behavior on the table are scaled from the most cooperative or synergic, which produces as a response a scenario of regional complementarity, through the most conflictive or clashing, that leads to a result of a fragmented regional cohesion. Some indicators of behavioral type that could be fitted into the table involve: the agreement between two ROs and the positive reinforcement of an action, which generates cooperation or a 5 Cf. DA CONCEIÇÃO-HELDT, E.; MELLO, P. Two-Level Games in Foreign Policy Analysis, Ox- ford Research Encyclopedia of Politics, New York & Oxford: Oxford University Press, 2017. 29 synergic regional governance; the agreement between two ROs without clear role definition on action generating competition for regional leadership (even if it results in positive reinforcement), which creates a cooperative but competitive scenario; the decision or incapacity by one of the involved ROs not to act, either in a situation of agreement or disagreement, which generates a regional silence or indifference that neither advances nor hampers regional actions; the disagreement of a RO with a given measure and its decision not to act or to blockade the action, which generates a situation of institutional deadlock, and the veto behavior among players; and lastly, the most conflicting scenario which occurs when two ROs have clashing interpretations on how to act on a given situation, and aim to delegitimize other actor’s performances, generating fragmentation of regional cohesion. Table 1- Types organizational behavior and corresponding regional results Type of organizational behavior Behavior outcome Cooperative-synergic Coordination or complementarity Cooperative-competitive Agreement with competition for leadership Conflictive-passive or cooperative-passive Silence or indifference Conflictive-blockade Deadlock/veto Conflictive-clashing Fragmentation Source: author’s elaboration It is imperative to stress that case studies and institutional responses are not presented monolithically, as ROs can have different positionings over time. The indicators and the actors’ performances choices described above and fitted into table 1 should be further developed in the case analysis’ chapter. On another note, even though the proposition of a behavioral typology exerts the actor’s underlying rational choice, which could contradict the affirmations of the need of context from historical institutionalism, the combination between both strands of thought is made possible by the comprehension of bounded rationality (JONES, 1999). As statements from historical institutionalism and its perception of path dependency could not directly be framed as rational, this does not mean that organizational politics are irrational. As Jones (1999) affirms, bounded rationality asserts that decision makers are intendedly rational, or goal-oriented and adaptive, but their preexisting underlying features, such as their human cognitive architecture can partly compromise its rationality, creating a deviation. Therefore, by bringing together bounded rationality and institutionalism we admit the influence exerted 30 by historical features of ROs, allowing the understanding of path dependence and particular features of institutional trajectory, such as critical junctures as a rationality that results from processing behavioral limits. Lastly, even though we recognize the challenges and the untraditional approach, we rely on a broader perspective comprehension of multiple cases through qualitative testing to address the inquiries of this thesis, since it can help tackle the “how general?” question about institutional interplay, analogously to Gary Goertz’ (2017) proposition of a large-N qualitative analysis. Seeking to produce generalizations and generalizability on the study of ROs patterns of action in overlapping scenarios, we meet Goertz’ requirements by examining a selection of all of the relevant sub-regions for concrete, empirical evidence, as we discuss in the following section. 1.3 Case selection and thesis’ scope Considering the preeminence of multilayered regional relationships in the different regions of the world and the need to develop empirical studies on overlappings in order to consider its effects, the present thesis also aims to contribute to comparative regionalism studies by considering each region individually and valuing the Global South as a producer of regional dynamics. The initial idea was to cover cases on three regions, avoiding Euro-standardizations by considering besides south America and Africa, the Asian continent, to be studied from the perspective of Southeast Asian overlapping organizations. However, as the inductive research was developed on political crises, the central focus of case analysis, Asia, in its particularity, presented a situation of indifference and non-action within ROs, which can therefore be perceived as a non-negative but also as a non-existent effect. As we center the attention of this research on institutional interactions and the form they can present themselves, in the absence of actions and, consequently, interactions, we decided to focus the research on the regions that provide richer material for the analysis. Thus, we proposed to resume and deepen studies on the effects of overlapping organizations in South America, with the intention of providing elements that could contribute to the development of a positive regional dynamic. The choice for South America instead of Latin America derives from the perspective that the disaggregation of the American continent into sub-regions has given birth to such different regional logics within the Caribbean, North and Central American contexts, that even the overlapping dynamics are disaggregated, 31 presenting themselves more attractively and intensely within the sub-regions. Hence, we continue the analysis of the experiences of MERCOSUR, UNASUR and CELAC started at the master’s studies adding regards to the other regional projects with competences on the sub-region, such as the Andean Community (CAN) and the Organization of American States (OAS), in the sectoral cases where they are pertinent. In the African case, however, the preeminence of the continental logic proposed by the African Union (AU) provides an interesting opportunity to analyze the relationship between the pan-African organization, as a continental organization with some of the sub-regional experiences, the so-called Regional Economic Communities (RECs). This study also can contribute to the empirical testing of propositions on overlapping regionalism. For instance, according to Nolte (2014), the presence of more comprehensive central ROs that articulate the multiple sub-regional platforms and build bridges between the different mechanisms while using institutional overlapping to their advantage, could represent a positive aspect for regional governance and cohesion. Therefore, when we look at specific sectoral cases, it will be interesting to see how the interplay between the overlapping organizations work in a situation where a continental mechanism that intends to organize the relationship with sub-regional blocs exists. In this case, following the African Union’s interpretation, eight of the existing sub-regional organizations are categorized as RECs and understood as building blocs, or pillars for the development of continental regionalism. They are the Arab Maghreb Union (AMU); the Common Market for Eastern and Southern Africa (COMESA); the Community of Sahel– Saharan States (CEN–SAD); the East African Community (EAC); the Economic Community of Central African States (ECCAS); the Economic Community of West African States (ECOWAS); the Intergovernmental Authority on Development (IGAD); and the Southern African Development Community (SADC). The plurality of African institutions also functions as way of granting the study with a large enough universe of cases, as we can further develop analysis on dyadic relationships between the AU and each one of the particular RECs, to perceive their behavior and interaction. As we have already justified the choice of political crises as the sectoral cut and scope condition, we move through the geographical selection, and regard the temporal framework. The timeline of analysis was established from normative milestones in the different regions concerning democratic violations and unconstitutional changes of government. In the African case, for instance, we depart from the institutionalization of the Peace and Security Council of the African Union (2004), as a rule landmark for the enforcement of its normative framework 32 on unconstitutional changes of government (UCGs). Considering the high number of cases through the years, this selection also makes the research more feasible. However, in the South American case, the same institutional design cannot be perceived, as democratic provisions and interventions are based on charters and normative acts known as democratic clauses or mechanisms for democratic protection. Therefore, the point of departure is the moment when the overlapping competences regarding democratic abuses began in the sub-continent in 1996, as Mercosur signed its first commitment on the subject, hence sharing the competence on the subject with the OAS, which had previously signed a similar resolution in 1991, and a protocol in 1992. Over the following years, other ROs in the continent have come to sign provisions on the subject, thus becoming co- responsible for the democratic stability on the region. Departing from an earlier period is also a research choice based on the literature, as in the case of South America, the recent secondary data has already attempted to systematize the cases of political crises that occurred in recent years, through two main works by Weiffen (2017) and Nolte (2018). The authors also affirmed (2018) that there is a different dynamic on overlapping actions overtime in the region: while in the 1990s the coordinative logic would be preeminent, from the years 2000, a competitive logic has been installed. This perception could therefore provide interesting data for the analysis, allowing the comparison of patterns of interaction over time. Among the two regional cases, we have been able to map multiple within cases that meet the scope conditions of organizational overlap interaction in a situation of political crises in the relevant years. From the mapping we established the following criteria for the case selection and analysis: first all of the continents’ sub-regions organizations should be represented. Therefore, the choice for case-countries considers that all ROs in the region are entitled to debate a scenario of political crisis at least once, even if they choose not to act. The second relevant criterium was the availability of data, which also considers the impact of the crisis on the region and on the literature production. A complete list of the compiled cases can be seen in each regional chapter (chapters 3 and 4). Lastly, due to the advances of academic production on the topic, we were able to work with all of the mapped ruptures to the democratic order in South America that fit the scope conditions and were negotiated with overlapping mandates: 2010 Ecuador; 2012 Paraguay; and 2013-onwards Venezuela. As for Africa, since the number of cases is very high, more selection criteria had to be employed. They are further described on chapter 3, but, for symmetry reasons and according to the methodological approach that favors the possibility of 33 providing generalizations, we analyze within cases in all five African sub-regions and its relationship with the AU. They are: 2009 Madagascar representing Southern Africa; 2012 Mali and 2017 Gambia, representing West Africa; 2013 Central African Republic for Central Africa; 2013 Egypt for Northern Africa; and 2015 Burundi for Eastern Africa. 1.4 Research procedures The research has three main stages of analysis: mapping the regional organizations that compose scenario of overlapping in Africa and South America for political and historical contextualization which will be further used to understand organizational behavior; Case study scrutiny comprising the selection of cases that meet the research conditions in order to capture patterns of organizational behavior; Comparative systemization of convergences and divergences between regional processes to signify ROs behavior accordingly to the process of regional overlap. Thus, in the first stage, we seek to present the general political and historical panorama of regionalism in the two regions and the prominent institutional overlaps, in order to make known the overlapping scenario to be analyzed and allow discussion of the drivers of regionalism, which led to this conformation, allowing its comparison. Next, as we depart from historical institutionalism, we analyze ROs and their action when confronted with critical junctures (political crises), in order to provide evidence about the effects of overlapping on the interaction between ROs and its consequences for regional governance. To do so, we use primary and secondary data to trace institutional patterns of action, throughout the selected cases that meet preliminarily stablished conditions in both continents. The data is then processed through presence and absence to reflect on influent regional conditions and outcomes. Data systematization will also serve to highlight the specificities of each region or process and to enable the listing of variables that might influence the results of both the organizational behaviors and interplays, and their consequences to the concretization of the possible scenarios of regional fragmentation or complementarity, as indicated in the research question. Thus, the study of the processes is done individually, but simultaneously with both regions through the production of correlations between the presence and absence of patterns of organizational behavior, which are studied both in a broad regional perspective, and in a sub-regional dyadic one. The final effort is to develop the main argument of the thesis on institutional 34 interactions in overlapping regionalisms, analyzing the implications of segmented institutional proliferation for the development of regionalism in the global South, considering the different contributions of the regions in which they are inserted. 1.5 Summing up: contributions of the cross regional organizational behavior analysis In conclusion, considering the different theoretical, conceptual and methodological perspectives combined in this chapter, together with the thesis overview provided by the research procedures, we could single out contributions to be achieved by the present doctoral thesis. Firstly, overlapping regionalism is still a relatively new field, whose studies have been developing in the last decade, but still demand empirical exploration. This affirmation is also supported by the fact that most of the works of the literature are focused on a single region. Thus, the research here presented also offers a contribution as it values the cross-regional approach of comparative regionalism, not only by not studying a mainstream region, but by combining the analysis of two of them. Secondly, in this research we are focusing on the process of interactions generated by overlapping competences, not on policy outcomes, as does most of the existing literature. This differentiation allows observation on organizational behavior and institutional learning, which by its turn can provide reflections on how the plurality of ROs affects regional power dynamics from the dyadic interplays between the sub-regional and regional levels. Moreover, when we consider the African continent, the innovation relies primarily on associating studies on regionalism to overlappings as such conceptual problems, since the literature on regime complexity is more preeminent. Finally, regarding the unusual methodological choice of a broad spectrum, the horizontal selection of multiple cases to cover different sub-regions can also provide further developments to the field, as we test approaches to political crises by applying qualitative analysis to institutional interplays at the regional level, enabling not only methodological testing, but also the advance of theory building. 35 2 REGIONALISM IN THE GLOBAL SOUTH: INSTITUTIONAL MAPPING OF THE SCENARIOS OF OVERLAPPING In order to advance questioning the phenomena of overlapping ROs and to provide a contextual framework to the research, a set of preliminary inquiries must be addressed, such as: how are regional scenarios composed in Africa and Latin America? What types of institutions are present, what is their context of creation? What are the factors that lead states to associate around ROs? Does the presence of said factors in the beginning of the processes translate into influence in the results of the interactions of the coexistence of blocs in a complex regional scenario? Are there characteristics common to these regions? The justification for proposing such contextual questions lies in the theoretical framework that underlies the thesis, institutionalism (KEOHANE, 1989), and which, therefore, is centered on the formal and institutional structures that make up the regional processes to perceive their dynamics and scope of action. As we have stated before, according to Acharya (2012), in this sense, we highlight that this framework starts to question not the validity of regional institutions, but how they matter. By associating the motivation behind the conformation of scenarios with the literature of drivers of regionalism, we aim to provide not causal but analytical elements that allow future reflection in the thesis about the participation of the elements that make up the history of regional processes in their later functioning and behavior, in cases of overlapping institutions. In this sense, therefore, we deal with the concept of path dependence, prevenient from historical institutionalism, which perceives the sequences in which events unfold as elementary to understand later consequences (PIERSON, 2004). Thus, the so-called drivers are the factors that motivate and initiate the development of regional processes and they can assume the most varied characteristics, from the interests of national elites and their ability to press governments for integration, to the presence of threats that can be better faced together. The present work does not intend to simplify said drivers, because it is understood that these should not be analyzed in a monolithic way, but as a complex set that influences the formation of preferences for the institutional establishment. Notwithstanding, the intention that underlies the study of drivers is, in accordance with the aforementioned path dependence perspective, to be able to reflect on how its influences might be determining on the shaping of regional relationships between overlapping organizations. Therefore, in this first moment we present the problem through a panorama of the regional organizations in two regions of the world, with the objective of illustrating the 36 display of organizational settings in Africa and South America before proceeding to the case- specific analysis on the next chapters of the doctoral thesis. Consequently, the aim of this chapter is to reflect on possible commonalities between the regions, and to provide future material for analyzing the organizational behavior, through the knowledge and problematization of the scenarios of overlapping regionalisms, their conformation and the drivers that drove the establishment of their institutions. Yet, before moving on to the individual study of each region, we address on the next two sections two fundamental topics for this research. We further develop the theoretical background on overlapping regionalisms and why it is so relevant for this study; and then we enter the discussion on drivers of regionalism, which encourages the discussion of development and trajectory in the regional processes. 2.1 Overlapping regional organizations and regional interplays As we have introduced on the previous chapter, overlapping regionalism has been con- ceived as instances in which states are affiliated with several regional organizations (ROs), which are autonomous and independent of each other, creating an overlap of memberships; and share similar objectives in their scope and policies, thus producing overlapping mandates. The term has been consolidated by the literature (MALAMUD, 2013; WEIFFEN; WEHNER; NOLTE, 2013; WEIFFEN, 2017; NOLTE, 2014, 2016, 2018; PANKE STAPEL, 2016) mostly on studies of the profusion and coexistence of regional processes in Latin America6. We underscore, however, that for this study, we also consider the overlapping actions propelled by the ROs on a complex scenario, as previous researches (RIBEIRO, 2016) have demonstrated that with institutional enlargements, the mandates not always cover for the entire scope of action of a regional bloc. We further develop the conceptualization and characteris- tics of overlapping in action on chapter 5, after the enlistment of the contributions of the em- pirical analysis, but the prescriptions for the formulation of the dyadic overlapping in action relationship are similar to those concerning overlapping mandates. In short, a situation with two or more ROs that share member states and policy areas constitute a scenario of overlap- 6 A few studies have addressed overlapping regionalisms in Europe and Asia. For Europe, cf. PANKE, D., & STAPEL, S. Overlapping regionalism in Europe: Patterns and effects. The British Journal of Politics and International Relations, 20(1), 239–258, 2018; and on Asia, see YEO, A. Overlapping regionalism in East Asia: determinants and potential effects, International Relations of the Asia- Pacific, Volume 18, Issue 2, Pages 161–191, 2018. 37 ping organizations. If the shared policy areas are ruled by a formal mandate, there is a de jure overlapping, while a de facto overlap is formed through the actions of the involved ROs. The research and conceptualization on overlapping regionalisms are derived from the regime complexity literature (BROSIG, 2011; WEIFFEN, WEHNER; NOLTE, 2013), which produced important insights on how the multiplicity of institutions affects the way they will interact at the international level (BROSIG, 2011; 2013). Orsini, Morin and Young (2013, p.29) define regime complexes “as a network of three or more international regimes that relate to a common subject matter; exhibit overlapping membership; and generate substantive, normative, or operative interactions”. The authors rec- ognize that those interactions are perceived as potentially problematic to the environment they are inserted to independently of if they are managed effectively, or not. Weiffen (2017) clarifies that even if regime complexity studies are focused on interna- tional treaties and conventions, rather than on formal organizations, there are some similari- ties among the two fields of institutional interplay, as they share “the theoretical specification of causal mechanisms underlying the consequences of overlap”. Brosig’s work (2011, p. 150) agrees on the commonalities and states that “the most significant difference between treaty regimes and international organizations is the difference in the degree of agency”. Thus, the author affirms that this divergency between the two regimes would be the organizations’ abil- ity to manage their interactions on the complex scenario, by formulating their external rela- tions pro-actively (BROSIG, 2011, p.150). Brosig’s contributions also provide a theoretical basis for our proposal of studying how ROs behave on overlapping settings, as the author further enunciates that overlaps will gener- ate a normative pressure for the organizations to react to common issues through the actors’ behavior, and thus trigger interaction among them. Those forces for interplay may lead to empowering as well as disruptive effects and the author (BROSIG, 2011, p.152) has catego- rized them under two main factors: the existence of diverging norms and behavior; and the impact of the obligations that an international organization imposes on itself and its members. Consequently, overlaps change the strategies open to national and international actors to achieve their preferences through regionalism (WEIFFEN, 2017), and understanding these settings and organizational behavior thus becomes relevant, considering the outcomes they can produce on the regions they are located at. 38 2.1.1 Configuration types of overlap Weiffen, Wehner and Nolte (2013) have presented one of the first conceptualizations of overlapping organizations, and the possible configuration the phenomenon might assume. Considering the relationship between mandates and memberships, the authors present two variants of overlap, which can be seen on Figure 2 below. In the first constellation, (a) Inter- section, the two circles represent two regional organizations with common mandates and an intersection of states to both organizations. The second diagram, (b) Subset, presents “an over- lapping mandate with nested membership, where all members of a smaller organization are part of a larger organization, but both organizations are autonomous and independent of each other” (WEIFFEN; WHENER; NOLTE, 2013, p.374). Figure 2- Variants of overlap Source: WEIFFEN; WHENER; NOLTE, 2013, p.375 In order to provide further specialization to categories of overlap, we draw on Young’s (2002, apud BROSIG, 2011) formulations about organizational interplay and clarify that we understand as horizontal overlappings the sharing of mandates/actions or memberships between two ROs at the same level (regional or sub-regional), while we refer to vertical overlap when considering the relationship between an organization at the sub-regional and the other at the regional - or continental level as seen in figure 3, on the next page. The sets covered by the concept of vertical overlappings therefore address the levels of interaction between sub-regional and regional; sub-regional and continental; regional and continental organizations; while horizontal overlappings cover organizations at the same level: sub- regional - sub-regional, or regional - regional, and so forth. The notion of vertical overlappings is partly compatible with the idea of nested 39 international organizations as pictured by the literature on regime complexity (WEIFFEN; WEHNER; NOLTE, 2013). However, we choose to present it differently as we also consider the possibility of finding ROs in a vertical overlapping situation where they are not fully nested due to not sharing the totality of member states, only coexisting at different levels. A good example would be the situation between the African Union and the Arab Maghreb Union (AMU) with Morocco until 2017: the country was a member of the sub-regional organization, but not of the continental one, thus making the nesting between organizations imperfect. Also, another possibility would be the existence of nested settings at the same level, meaning, two sub-regional nested organizations that do not represent a vertical overlap, as considered by our definition, as for example ECOWAS and WAEMU. Figure 3 – Explaining ROs horizontal and vertical overlappings Source: author’s elaboration 2.1.2 Potential outcomes’ scenarios: why overlappings matter The existence of overlapping regional organizations is not intrinsically good or bad for regional development, as both outcomes can prevail from the interaction between regional organizations. The specific literature on the subject has presented divided impressions of what the real consequences of the complexity of regional scenarios would be, as we discuss below. Among the positive potential of institutional plurality’s arguments, Hoffmann and Mérand (2012) argue, that regional negotiations are less likely to fail when states have the possibility to opt-out of arrangements that do not meet their interests. Based on the European experience, the authors have proposed the concepts of regional integration at different veloci- ties and agreements with varied geometries (HOFFMANN, MÉRAND, 2012). Nolte (2014) also contributes to this view by stating that combining the flexibility with which states can choose the processes in line with their interests while keeping their loyalty to other regional partnerships could contribute to the articulation of regional governance while fomenting cross-sectoral associations focusing on different themes in multiple regional pro- Vertical overlap Horizontal overlap 40 jects. According to the author, one factor that could help the articulation of the multiple sub- regional platforms and build bridges between the different mechanisms would the presence of comprehensive central institutions that could use institutional overlapping to their advantage and foster regional governance. Other positive views on overlapping scenarios would be that a diverse scenario and its competitive pressures could stimulate development and innovation, cause the reinforcement of regional positions, and foster the division of labor and specialization among ROs. Nonetheless, some points of view diverge from the perspectives outlined before and enlist possible consequences of the organizational plurality. Malamud and Giardini (2012) affirm that overlappings could lead to competitive regionalism, to the appearance or deepen- ing of the divisions between countries, to increased tension and disagreements, and eventually to a hemispheric disintegration based on emerging sub-regionalism. Laura Gómez-Mera (2015) does not disregard the possibilities for synergy and coop- eration between institutions but identifies three mechanisms that sabotage the effectivity of cooperation on overlapping situations: legal fragmentation; the use of inter-institutional strat- egies; and constant competition that causes the regional unity to be segmented. The first mechanism refers to the normative ambiguity generated by institutional over- lapping, which increases the difficulties in complying with regional initiatives, is costly for both the public and private sectors, and demands a considerable legal and technical know- how. The second mechanism is the formulation of inter-institutional political strategies through which negotiations can be shifted from one bloc to the other with the objective to benefit by influencing the nature of their regimes. And the third mechanism is the erosion of regional cohesion and unity through constant competition (GÓMEZ-MERA, 2015). The author’s proposed mechanisms would generate conflict and segmentation as they enable actors to switch between different legal frameworks, effectively forum 'shopping' for most attractive or favorable set of rules to deal with a particular situation; allow actors to de- velop a strategic inconsistency in dealing with regional blocs with the purpose of undermining the legitimacy of one organization; and create inefficiencies and even undermine the effectiv- ity of common efforts in the resolution of international conflicts (GÓMEZ-MERA, 2015, p. 23). Gómez-Mera further stresses that the constant competition among organizations could contribute to states losing confidence in, or even abandoning regional integration altogether in favor of other initiatives. Considered the characteristics and consequences of overlapping regional organizations, we move on to the next section in order to develop the argument about drivers 41 and their influence on the processes of regionalism, and consequently on the formation of the complex regional scenarios. We depart from the theoretical framework proposed by Louise Fawcett and Philomena Murray (2015), whose constructivist perspective, combined with the institutionalist logic, encourages discussion of development and trajectory in the regional processes. In addition, we considered specific factors to each region in order to value diversity and comply with the methodological framework of Comparative Regionalism, as we mentioned on the previous chapter. Thus, some additional questions that could be raised after scrutinizing the concept of overlapping regionalisms would be: can drivers be understood as important factors not only for understanding the composition of regional organizations but also to comprehend overlapping scenarios? Is there a driver that tends to produce more overlaps? Beyond that: are the history and origins of regional processes decisive in shaping their behavior? 2.2 Drivers Before addressing the theoretical predictions for regional drivers, a conceptual elucidation must be made. “Regional integration” and “regionalism” are two expressions that have been used indistinctively by authors and politicians over the years to address the political movements that brought states together. However, even though all “regional integration” processes are also part of a “regionalism” process, the reciprocal is not true, as currently some consensus has been achieved among specialists (BÖRZEL; RISSE, 2016) that “integration” presupposes autonomy delegation in favor of creating supranational instances. On its turn “regionalism” also comprehends state-led cooperation projects, with intergovernmental characteristics that are created among countries of a common regional space. This clarification was needed in order to avoid conceptual ambiguity, since parts of the guiding literature on regional drivers’ development have concentrated the conceptualization either on “regional integration” or “regionalism”, and yet research has demonstrated that the factors that influence the approximation of regional interests are common to both processes, and thus both contributions are combined on the debate. Louise Fawcett (2015) defines the drivers as “that collective of factors that, historically and at present, may contribute to the process of integration” (FAWCETT, 2015, p. 35, emphasis added). Therefore, the drivers do not only explain the beginning of the processes, but also their changes, and three would be the main elements that serve for the historical explanation and provide bases for the comparison in the regional construction: core 42 states; ideas about the region; and institutional environment. According to the author, the first factor would be linked to the exercise of leadership by key countries in the promotion and development of ROs. As an example, one could cite Brazil's role in shaping the Union of South American Nations (UNASUR), or the leading role that rising countries like Nigeria, South Africa and India assume in their own sub-regions. On the other hand, the factor related to ideas about the region addresses how States and the individuals and groups that promote them act in order to establish parameters for what is acceptable and possible. Finally, the factor that considers institutional environments relates the presence of existing international institutions that act to shape regionalisms. As an example, there is the UN and the proposal of conformation of sub-regional organizations in the sphere of security in various parts of the world (FAWCETT, 2015, p.50). Philomena Murray (2015), in turn, considers that the factors that lead to regionalism are directly linked to two elements: trust and reconciliation between countries. Therefore, region building depends on community building, which can only be done through these two main factors. In turn, community building generates region building and therefore configures the regional scenarios as known. The author lists endogenous and exogenous factors that can be considered as drivers influencing regional integration. In the list of endogenous, elements are described such as how perceptions about sovereignty may function as drivers or inhibitors; commitment and state leadership / capacity; economic interests; business interests; personalist leaderships. The exogenous ones, for example, would be the presence of an external hegemon - notably the US - or the need to compete or cooperate with a regional power – which is the case of China in Asia-; perceptions of threats or conflicts; diffusion of norms; or a regional bloc acting as a driver. Finally, Tanja Börzel summarizes by stating: “The main drivers for regionalism are the (material) gains states expect to reap from trading with each other” (BÖRZEL, 2016). In this way, by delimiting possible factors for the promotion of ROs, it is possible to proceed with the questioning about the relationship between the elements that influence the conformation of the institutions and the way in which they will relate to each other when the regional situations cause their interaction. Moreover, although drivers of regionalism are not always the drivers of overlapping, one can discuss their role in shaping complex regional scenarios. Finally, one can still question the existence and influence of path dependence from drivers when analyzing how regional organizations behave. 43 2.3 Latin America: Scenarios of overlapping, South American Studies In Latin America, the roots of the influences of regional integration processes date back to the independence period of the former, especially Iberian, colonies. Since then, according to Bianculli (2016), two main trends have guided the proposal of competing projects in the region. The first one is based on the ideals of Simon Bolívar, which, from the end of the nineteenth century, proposed the making of political articulations for the union of the neighboring nations, seeking to affirm the non-alignment of the new states not only to the colonial metropolises through their own concertation, but also to distance themselves from the expansion of the United States. Although unsuccessful in institution building in this period, this trend leaves marks for the continuity of processes in subsequent periods. In its turn, the second trend would be aligned with the ideas proposed by the United States itself for the region and goes beyond the Latin understanding of the region to comprise the neighbors of the North. Between the years of 1889 and 1930, the Pan American Conferences, organized by the United States, had the objective of promoting a hemispheric regionalism, with roots in the Monroe Doctrine and the ideal of “America for the Americans”. Pan-Americanism inaugurates the construction of institutions in the region, but its importance is also due to its consequent contestation, as we will discuss later. The two trends for regional approximation have been described by the literature as waves of regionalism, an illustrative expression to address the moments in history where conditions have made the development of regional processes more propitious. The initial classification of the waves’ periods was much based on the European experience, but these moments have also contributed to the understanding of the historical conformation of the Latin American blocs, since the waves presented themselves to the countries as means in the constant search for overcoming the condition of underdevelopment, and even an exclusive wave of regionalism is created in Latin American in the 2000’s. Therefore, it is clarified for the purposes of comparison, that in the Latin American case, where the waves and the phases of creation of ROs are coherent, they will be used for periodization and the perception of the drivers. In the case of Africa, periodization follows its own criteria, as we argue on section 2.4. Something to emphasize, however, is that the understanding of Latin America as a whole, as an area of integration, goes through adaptations. Although the regional blocs proliferated at such moments, initiatives that comprised the entire subcontinent could not achieve integrationist successes or had limitations. In fact, the “disintegration” of Latin 44 America in interconnected sub-regions (Andean America, the Caribbean, Central America, Southern Cone, Northern America) was a solution for the creation of processes in the face of the waves of regionalism and its understanding is essential for understanding the pursuit of integrated development in the region (PHILLIPS, 2005). Thus, in addition to comprehensive projects such as the Latin American Free Trade Association – LAFTA or in its Portuguese acronym ALALC - (1960), which later gave way to the Latin American Integration Association – LAIA or ALADI - (1980) and the Community of Latin American and Caribbean States - CELAC - (2011), there are initiatives in the region that propose different ways of thinking about relationships with neighbors, whether through a redetermination of space or through sub-regions, as will be discussed below. 2.3.1 Closed Regionalism The first cycle of regionalism emerged after World War II, strongly inspired by the successes of the integration process developed on the European continent and was termed closed regionalism (SARAIVA, 1999; HERZ, HOFFMAN, 2004). It was characterized by a concern centered on trade logic and economic protectionism, with a strong institutional framework - including a high degree of autonomy vis-à-vis national governments - and the pursuit of development through the stimulation of greater economic interdependence through trade. It was at this moment that the first regional initiatives of Latin America appeared. While the Pan-American ideal embodied the formation of the Organization of American States (OAS or OEA) in 1948, motivated by the post-conflict world and regional security, other sub-regional initiatives were developed. Reflecting on drivers, it is important to highlight the role played by the Economic Commission for Latin America and the Caribbean (ECLAC or CEPAL) in promoting ideas about how development would be achieved by the countries of the region and fostering regional integration as solution. It was through ideas like the structuralism of Raúl Prebisch in the years 1959-60 that the first regional projects as known today took shape. In reference to Prebisch (1959), Söderbaum (2016) explains that the ideal presented was that “liberalized intra-regional trade in combination with regional protectionism seemed to offer large economies of scale and wider markets, which could serve as a stimulus to industrialization, economic growth, and investment”. Thus, the ideal of achieving development led to the creation of institutions that sought 45 to foster productive integration and imports substitution, such as LAFTA in 1960 and, sub- regionally, the Andean Pact in 1969, which developed an institutional framework inspired on the European model. 2.3.2 Open Regionalism Closed regionalism presented limitations and was replaced by a new regionalist impulse in the 1980s and 1990s, in the context of the intensification of globalization and the “new world order”, which mainly met neoliberal norms - such as free trade and opening markets. Hence, this new moment was called open regionalism and its main provisions suggested the interdependence between neighboring countries should be promoted by the liberalization of economies, in order to strengthen the region's competitiveness and the interaction between nations. One particularity of this moment is the intergovernmental character of the institutionalization adopted, in contrast to the aspirations for supranationalism from the previous period. In the context of open regionalism, one can also see the influence of ECLAC's proposals on institutional formation7, since countries saw in regional integration the opportunity to overcome the peripheral situation through an international competitive insertion. While the opening of economies was considered strategic, the state was no longer the only actor in the process, giving the private sector the opportunity to replace it in the promotion of regional integration. During this period many of the regional agreements still in force proliferate, as well as the re-launching of some existing mechanisms. This is how the failed ALALC or LAFTA is relaunched in 1980 as ALADI (or LAIA) and the Andean Pact becomes the Andean Community (CAN) in 1996. As for the new sub- regional initiatives, proposals such as the Southern Common Market (MERCOSUR) in 1991, and the North America Free Trade Agreement (NAFTA) in 1994. With regard to the United States' proposals for the region, in 1994 a project fo