The socialisation of the African Union’s norms

Case studies of Gabon, Congo and Cabo Verde

Leslie Sandra Kengne Kamga Mobou

Leslie Sandra Kengne Kamga Mobou S1149350

[email protected] MA Thesis Political Science: International Organisation Leiden University Supervisor: Dr. Kleistra Second reader: Dr. Blarel Word count: 10867 January 12, 2017 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

The socialisation of the African Union’s norms Case studies of Gabon, Congo and Cabo Verde

Introduction Post-election violence is a prominent problem in African states. Although most African states claim to be democracies, election procedures are commonly undemocratic and held in circumstances of severe tension between the incumbent government and the opposition. The contestation of election outcomes, to the point of the use of violence, indicates the degree to which actors in African states do not believe in the democratic nature of their election procedures. Yet, all African states, except for Morocco, are members of the African Union (AU), which aims to promote peace, security and stability in Africa (Moolakkatuu, 2010, p. 151) through democratic governance. One of the primary philosophical reasons for the creation of the AU was to ensure its member states observe certain fundamental values and standards (Kioko, 2003, p. 807). Article 3g of its Constitutive Act states that democratic principles and institutions must be promoted, article 4m emphasises respect for democratic principles, human rights and the rise of law and good governance, and article 4p condemns and rejects changes of power that are unconstitutional (AU, 2000). Furthermore, in 2007 the AU adopted the African Charter on Democracy, Election and Governance, of which one of the focuses is democratic elections, institutions and culture (AU, 2007). Currently, 28 of the AU’s member states have only signed this Charter, and an additional ten states have signed and ratified it (AU, 2007). Some scholars have noted that the majority of such international agreements on democracy are not obligatory under international law, but are political commitments (Kane, 2008, p. 44). Such commitments are too weak to induce compliant behaviour and are difficult to enforce (Dai & Mattiacci 2011:85; ibid). However, the AU is making efforts to monitor and enforce commitment; for example, it obliges every member state that has signed the Charter to inform it of details regarding upcoming elections, and to invite it to send an observer mission (AU 2007, art. 19). Nonetheless, the AU has built itself a reputation of generally tolerating undemocratic governments and their

2 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

practices, as well as being unable to enforce its principles and values (Omorogbe, 2011, pp. 123-124). Therefore, it is puzzling that the AU has a Constitutive Act and a charter that focus on democratic elections, which some states have signed and ratified but many do not adhere to. Hence, the aim of this study is to answer the following question, “Under what conditions, and using which mechanisms, can the African Union encourage its member states’ elite to socialize the norm of democratic elections?” I argue that the AU can use persuasion and social influence to encourage its member states’ elite to socialize the norm of democratic elections, provided some conditions are met, such as that state elite have few prior contradictory norms and beliefs, and they consider the AU as an authoritative structure within the region. Obtaining an answer to this research question is beneficial for four reasons. First, it will provide the AU with a framework to approach its member states to reduce the occurrences of undemocratic elections. Second, it will be one of the few empirical studies on the socialisation of international norms that considers both why and how international organizations can encourage the socialisation of norms. Third, it will contribute to the sparse literature about the socialisation of norms in international organisations that are mostly comprised by undemocratic countries, where the roles of the more democratic powerful states of the world are negligible. Fourth, it will contribute to the sparse literature about how the AU is involved in the elections of its member states. This study will be organised as follows. First, I will briefly review the literature on the role of international organisations in the socialisation of norms. Second, a theoretical framework, together with corresponding hypotheses, will be discussed. Third, a methodology will be presented. Fourth, three cases will be analysed, and lastly there will be a discussion of the results.

Literature review Many international relations scholars have studied the question of how international organisations can influence the behaviour of sovereign states. The goal of this literature review is to present the mechanisms and conditions that those scholars have proposed to explain how states socialise the norms promoted by international organisations.

3 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

There are currently two cognitive models for understanding how international organisations can socialise their member states to the norms they promote. On the one hand, rationalists claim that international organisations can change state behaviour by having an impact on their interests. These interests are based on material gain. On the other hand, constructivists claim that international organisations can change the interests of states. These interests are not based solely on material gain, but rather on identity (Checkel, 2001, p. 443; Johnston, 2001, p. 487). These models will be further elaborated upon.

Rationalist mechanisms Rationalists believe that states only comply with norms promoted by international organizations if there are material rewards or punishments involved (Johnston, 2001, p. 487). Greenhill (2010, p. 128) claims that the European Union’s (EU) member states’ adherence to its core norms is a prime example of this. The EU can give extreme political and economic sanctions to its member states, as well as prospective member states, which fail to adhere to the rules and norms agreed upon within the organisation. For rational mechanisms to work within an international organisation, its member states have to believe that the benefits of being a member of the organisation are greater than the costs of adhering to the norm. The other members of the international organisation also have to be willing to pay the price of reduced cooperation when economic and political sanctions are imposed (ibid). The latter is one of the reasons why the rationalist mechanism is unlikely to be effective in some international organisations. Many international organisations, such as the Association of Southeast Asian Nations (ASEAN) and the African Union (AU), have stressed the importance of continued cooperation of their member states and have prioritised non-interference in domestic affairs (Acharya, 2014, p. 7; Greenhill, 2010, p. 129). Another reason is that the AU does not have an elaborate sanctioning mechanism (AU, 2000) and is often plagued with insufficient resources (Kasaija, 2013, p. 129); material rewards and punishments are rare. The last reason is that Checkel (1997, p. 475) claims that while rationalist mechanisms can be used to (partly) explain the effect of norms in liberal, corporatist and statist states, it fails to explain the socialisation of state-above-society states. He identifies state-above-society states as those which are (i)

4 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

removed from, and exercise considerable power over, the society, and (ii) where top- down policy-making is usual (1997, p. 480). It is precisely such states that dominate the African continent. Constructivist models have more explanatory power than rationalist models in such states.

Constructivist mechanisms of direct effects: Persuasion and social influence Constructivists have provided various mechanisms through which international organisations can alter the behaviour of their member states, both directly and indirectly. Two of the direct mechanisms international organisations can use are persuasion and social influence (Checkel, 1997, p. 475; Johnston, 2001, p. 487). In state- above-society states, state behaviour changes towards compliance when the state elite have been convinced that it is the right thing to do (Björkdahl, 2002, p. 14). International organisations can be used as platforms and agents to raise moral consciousness and to allow for argumentation, dialogue and persuasion of the state elite (Risse, 1999, p. 530), and they can distribute social rewards and punishments (Johnston, 2001, p. 487). There are specific conditions within which persuasion, and to a lesser extent social influence, is likely to be effective. The conditions which Checkel (2001, p. 580), Johnston (2001, p. 487), and Hooghe (2005, p. 861) agree on are (i) the target is new to the environment and motivated to analyse new information, (ii) the persuader is a highly authoritative member within the in-group, (iii) the target has few prior beliefs and/or attitudes that are inconsistent with message of the persuader, (iv) the issue is technical and apolitical and/or the interaction is not politicized, and (v) the target is repeatedly exposed to the message of the persuader. Furthermore, in order for the persuasion to be successful, the international organisation must frame the norm so that it ‘resonates’ and can compete with other norms within the state (Cortell & Davis, 2000, p. 66; Krook & True, 2010, p. 103; Payne, 2001, p. 37). Although there is some literature on the importance of persuasion and state socialisation, as well as the conditions within which they are effective, few have considered them together and few have tested them empirically.

5 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

Constructivist mechanisms of direct effects: Copying and learning Some scholars have minimized the active direct role of the international organisations, and have emphasised that international organisations are sites for copying and learning. These scholars argue that states are more likely to change their behaviour in adherence with the norm when (i) the norm is already internalized by other states that are widely considered as successful (Finnemore & Sikkink, 1998, p. 906), and (ii) and when their own experiences, as well as others’ experiences, provide new evidence to change their beliefs (Dobbin et al, 2007, p. 449; Greenhill, 2010, p. 129). This perspective assumes that there are states to learn from. Yet, this is not often the case with new norms, and certainly not the case in environments where norm- violation is not exceptional. In the case of the AU, who should the state elite learn from when the majority of the member states’ leaders frequently violate the norms promoted by the AU? Therefore, rather than only focusing on the states and its elite, some scholars have opened up the ‘black box’ and have considered the different institutions that states comprise of, and the indirect influence international organisations can have.

Constructivist mechanisms of indirect effects: Networks among domestic and transnational actors Many constructivists have been criticized for focusing on the state as a unitary actor and not considering the different institutions within it, which can also play important roles as intermediaries between the international organisations and the state elite (Checkel 1997, p. 475; Dai & Mattiacci, 2001, p. 97-98; Finnemore & Sikkink, 1998, p. 888; Thomas, 2002, p. 25). In Risse and Sikkink’s (1999, p. 3) ‘five phase spiral model’, Finnemore and Sikkink’s (1998, p. 895) ‘norm cycle’, and Alderson’s (2011, p. 417) ‘three steps to state socialisation’, the evolution of norms and the socialisation of states are highly dependent on the strength of transnational networks and domestic groups. These groups advocate for the socialisation of norms. States comply with norms because of the pressure put on them by these groups, as well as by other states and international organisations. This mechanism is problematic for explaining the socialisation of African state for several reasons. First, within undemocratic governments, opposition is usually

6 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

oppressed or immature. Hence, Checkel (1997, p. 475) claims societal pressure on the elite will not be effective. Second, constructivists’ explanations of the role of these groups have rationalist components. They claim that, initially, state elite do not change their behaviour because their identity and interests have changed, but because of the pressures exerted on them by societal groups, which have negative impacts on their abilities to obtain their interests (ibid.). It is unclear in the models at what point the identities of the state leaders change, and at what point they actually start to believe that complying is the right thing to do. Although it is likely that all the variables and models put forward in the literature thus far partly explain why states socialize to the international norms promoted by the international organisations, they are not all suitable for the context of the AU. The undemocratic and oppressive nature of most of the states, the limited capabilities of institutions within the states, the power centralization around the head of states, and AU’s limited resources suggest that theories which focus on rational choice, learning, and networks amongst domestic and international actors have less explanatory power. The effects of persuasion and social influence seem more likely. Although there is a substantial amount of literature on persuasion and social influence, relatively little research has been done on social influence in the field of international relations. This is surprising considering persuasion on its own does not comprise of a feedback mechanism which informs states how well they are doing in terms of compliance. Moreover, there are few studies that empirically test the possible effects of persuasion and social influence, separately as well as together.

Theoretical framework In this section the assumptions of the constructivist approach will be addressed, a theoretical model will be suggested, and the key concepts and scope conditions will be defined.

Assumptions The key assumption of constructivism is that state interests are defined in the context of internationally held norms, and they must be understood in relation to the identities of

7 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

state actors (Hafner-Burton & Tsutsui, 2005, p. 1381). Constructivists believe that identities define the range of behaviours that are possible and appropriate (Hafner- Burton & Tsutsui, 2005, p. 1381; Wendt, 1992, 399). They also argue that international organisations are environments of socialisation and have the capacity to make states internalise the goals and values that are ingrained in international law (Hafner-Burton & Tsutsui, 2005, p. 1381; Wendt, 1992, p. 396). Based on these assumptions and previous literature, I will argue that international organisations can encourage states to internalize the norms they promote by using persuasion and social learning, under specific conditions.

The model Persuasion and social influence are the mechanisms that international organisations can use to encourage state socialisation of their norms.

Norm of democratic elections Norms are defined in sociology as ”rules and standards that are understood by members of a group, and that guide and/or constrain social behaviour without the force of law” (Cialdini & Trost, 1998, p.152). In political science, international norms are defined as “generalised standards of conduct that delineate the scope of state’s entitlements, the extent of its obligations, and the range of its jurisdiction” (Björkdahl, 2002, p. 13). The power of a norm is dependent on its setting and the importance that it is given within that setting (Björkdahl, 2002, p. 13; Cialdini & Trost, 1998, p. 152). There are two categories of norms, namely ‘descriptive norms’ which describe the behaviour of other actors in similar situations, and ‘injunctive norms’ which are proscriptive and have a dimension of ‘oughtness’ that is reinforced by the possibility of social sanctions (Björkdahl, 2002, p. 13; Cialdini & Trost, 1998, p. 132). Given the research question involves the encouragement of changes of state behaviour, the focus will be on the AU’s injunctive norm of democratic elections. Three aspects of this norm must be considered, namely the (a) technical, (b) political, and (c) moral aspects. The technical aspect refers to the expected characteristics of elections, and whether the necessary resources and facilities are present to ensure that elections are democratic, such as (i) frequent and timely elections, (ii) adequate resources for the

8 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

polling booths, and (iii) autonomous election commissions and courts. The political aspect refers to the political culture within the country, and considers issues such as (i) the relationship between political parties, (ii) the electoral system, and (iii) the comprehensiveness of the suffrage. The moral aspect refers to the beliefs of the state elite with regards to democracy and democratic elections, and whether they believe that democratic elections ‘are the right thing’.

State socialisation The dependent variable, state socialization, is defined as the “the process by which states internalize norms originating elsewhere in the international system” (Alderson 2011:415). This process takes place by means of social interaction, which results in the state exhibiting desired and expected “ways of thinking, feeling, and acting” (Johnston, 2001, p. 487). The internalization of the norm implies that the norm is ‘taken-for- granted’ (ibid.); meaning the moral aspect of the norm has been accepted. Although the definition of norms does not require a force of law, an indicator of state socialization is institutionalization. This can be observed by the incorporation of the norms in the legal system (i.e. judicial codes, laws, and constitutions), and the creation of bureaucratic actors who have the duty of promoting norms and ensuring they are enforced (Alderson, 2011, pp. 419-420; Checkel, 2001, p. 560). Another important indicator of the degree of state socialisation is norm-compliant behaviour.

Persuasion The independent variable persuasion is defined as “the representation of persuasive arguments” (Petty & Cacioppo, 1984, p. 69). During social interaction, international organisations can challenge the truth claims that are ingrained in the identities and perceived interests of the state elite (Checkel, 2001, p. 560; ibid.). Social interaction is defined as any exchange of information, acts and behaviour by two or more actors, intentionally and mutually oriented towards each other (Rummel, 1976). Hence, the target is persuaded after having carefully considered the information that has been presented to it by the persuader (Petty & Cacioppo, 1986, p. 125). Two factors that increase the persuasive impact of a message are the strength of the message and the number of arguments provided. A strong message is defined as one that

9 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

provides persuasive evidence (such as statistics and data), and provides argumentation for why a norm should be upheld. As the persuader increases the number of arguments, the target is provided with more information to think about (Petty & Cacioppo, 1984, p. 69). When persuasion is successful, state interests are redefined (Checkel, 2001, p. 560; Risse, 1999, p. 530) and state socialisation is more probable. Therefore the hypotheses are:

H1: The stronger the message that the persuader provides is, the likelier the target will exhibit norm-compliant behaviour and will incorporate the norm in its legal and bureaucratic structures.

H2: The more arguments the persuader provides in the message, the likelier the target will exhibit norm-compliant behaviour and will incorporate the norm in its legal and bureaucratic structures.

Social influence Considering the social and interactive nature of state socialisation, international organisations must provide feedback on whether the state’s behaviour is considered as norm-consistent (Johnston, 2001, p. 487). This can be done by the independent variable social influence, which constitutes of micro-processes where norm-compliant behaviour is encouraged through the distribution of social rewards, and social punishments (ibid). Social rewards include, but are not limited to, back-patting, status maximization and social liking, while social punishments include, but are not limited to, shaming, exclusion, and sanctions (ibid). The process of social influence can have both positive and negative effects on the socialisation of norms, depending on whether it is handled adequately or not (Cialdini & Trost, 1998, p. 151). If handled correctly, it can be used to encourage targets to move towards injunctive norms. However, if handled poorly, for example by giving social rewards when norm-deviant behaviour is exhibited, it can reduce the likelihood of such changes and may even have adverse effects (ibid.). Therefore, the hypotheses are:

H3: An increase in the use of social rewards to praise norm-compliant behaviour will increase the likelihood of the target exhibiting norm-compliant behaviour and incorporating the norm in their legal and bureaucratic structures.

10 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

H4: An increase in the use of social punishments to condemn norm-deviant behaviour will increase the likelihood of the target exhibiting norm-compliant behaviour and incorporating the norm in their legal and bureaucratic structures.

H5: The inappropriate use of the social rewards and social punishment will reduce the likelihood of the target exhibiting norm-compliant behaviour and incorporating the norm in their legal and bureaucratic structures.

Scope conditions Persuasion, and to a lesser extent social influence, is more likely to be effective when done in the following conditions; (i) the elite are new to the environment and motivated to analyse new information, (ii) the elite have few prior beliefs and/or have attitudes that are inconsistent with the international organisation’s norm, (iii) the international organisation is seen as a respectable, credible or powerful organisation by its member states with regards to the norm of interest, (iv) the norm is technical and apolitical and/or the interaction between the international organisation and the elite is not politicized, and (v) the elite are repeatedly exposed to the international organisation’s norm (Checkel, 2001, p. 580; Hooghe, 2005, p. 861; Johnston, 2001, p. 487). Furthermore, it is important that the norm is broad enough to allow for historical and context-specific framing by the international organisation, so it ‘resonates’ (i.e. becomes salient) within the state (Payne, 2001, p. 37). A norm is not fixed; it can be shaped and interpreted (Krook & True, 2010, p. 105). Norms that are framed to align norms already held by state elite are more likely to become salient (Payne, 2001, p. 37). Whether or not a norm ‘resonates’ in a state can be observed by the frequent use of the norm to justify its actions. However, an actor’s frequent use of the norm to justify its actions does not always mean its behaviour is norm-compliant (Cortell & Davis, 2000, p. 66). The model and the scope conditions will be tested empirically by doing case studies of three elections within three states, namely Gabon, Congo and Cape Verde.

11 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

Methodology In order to answer the research question, “Under what conditions, and using which mechanisms, can the African Union encourage its member states’ elite to socialize the norm of democratic elections?”, a qualitative multiple case study was performed. This method has two advantages to other qualitative methods. First, it allows for an explanatory study of the complex causal relationships being explored, while considering the contextual condition (Baxter & Jack, 2008, p. 554). Second, following the constructivist assumption that reality is not fixed but constructed, the use of multiple data sources to analyse the cases allows for a more holistic understanding of the mechanisms of interest and will increase the construct validity (ibid.).

Cases In order to select the cases, a list of the five most recent African Union Election Observation Missions (AUEOM) was made on November 28, 2016. This choice ensured that the cases analysed were cases where the AU had recently been involved in the election procedure. The resulting cases included the missions in the Republic of Congo, Zambia, Gabon, Seychelles and Cabo Verde. Considering time constraints, three of those five cases where chosen for analyses, namely Gabon, the Republic of Congo and Cabo Verde. Gabon’s election was chosen because it has a history of undemocratic rule and post-election violence, and the opposing contestant is a former chair of the AU. The Republic of Congo was chosen because President Sassou Nguesso’s successful push for constitutional changes, which allowed him to run for presidency for longer, most likely resulted in that the 2016 elections were watched closely by other African leaders who are attempting to do the same. Lastly, Cabo Verde was chosen as a contrast to the other two countries, as it is hailed a model for African democracy. It has a tradition of peaceful transition of power in its forty years of independence.

Data source In this study I looked at the interaction between the AU and elite of the above- mentioned countries, in the period leading up to the elections as well as just after the

12 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

election. In order to investigate this, both primary and secondary data were used. First, all documented information in English or French on the interactions between the AU and the state elite were analysed. This included official reports, press releases and newspaper articles. Second, documented information on the comments by the AU officials related to the relevant presidential elections were analysed. These included the previously mentioned sources, as well as key AU documents such as Decisions and Declarations of the Assembly and documents from the AU’s relevant committees.

Control variables In order to be able to claim that the increase in norm-compliant behaviour is due to socialisation, one must control for self-selection, selective recruitment, and utility maximization (Hooghe, 2005, p. 861). Self-selection refers to the possibility that state elite who already share the norm of the international organisation can choose to join the international organisation (ibid). This is not the case within the African Union, as apart from South Sudan, all of the AU’s member states have been members since its creation. Selective recruitment refers to the possibility that the international organisation recruits state elite after they have screened their views (ibid). This again is not the case for the African Union for the same reason as above. Lastly, utility maximization refers to the fact that it is self-interest, and not change in interest, that encourages the state elite to internalize the organisation’s norm. Based on constructivist assumptions and Checkel’s (1997, p. 475) distinction of when rationalist mechanisms are more appropriate, this is unlikely. Yet, it cannot be excluded and must therefore also be considered in the analysis.

Results Norm of democratic elections Although, in essence, a norm is difficult to define and to describe, one can assume that the norm of democratic election exists within the AU, as stated in its official documents. The African Charter on Democracy, Election and Democracy states what is expected of the member states in terms of technical, political and moral aspects of the norm.

13 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

Technically, the Charter states that the member states must hold “regular, transparent, free and fair elections” (AU 2007, art. 3.4), and must “establish and strengthen independent and impartial national electoral bodies responsible for the management of elections” (AU 2007, art.17.1). Furthermore, every member state must inform the Commission of the African Union of upcoming elections and “invite it to send an electoral mission” (AU 2007, art.19.1). The Charter also states what is expected politically. It states that the member states must “recognize popular participation through universal suffrage ” (AU 2007, art. 4.2), should “take measures to ensure and maintain political and social dialogue” (AU 2007, art. 13), and “ensure equitable access by contested parties and candidates to state controlled media during elections” (AU 2007 art. 17.3). It also clearly states that the member states should “promote political will as a necessary condition for the attainment of the goals set forth” in the Charter (AU 2007, art. 44.1b). Finally, the Charter attempts to establish the moral aspects of the norm. It states that every member states should promote adherence “to the universal values and principles of democracy” (AU 2007, art. 2.1), they should “re-affirm their commitment to regularly holding transparent, free and fair elections” (AU 2007, art. 17), they should “take all appropriate measures to ensure constitutional rule, particularly transfer of power” (AU 2007, art. 5), and the binding code of conduct within a member state “shall include a commitment by political stakeholders to accept the results of the election or challenge them in through exclusively legal channel” (AU 2007, art. 17.4).

Cases The people of Gabon have had three presidents since their independence from France in 1960. Both of the presidential changes occurred following the death of the previous president. The current president Ali Bongo Ondimba came to power in 2009 following the death of his father, Omar Bongo, who ruled the country for 42 years (France24, August 2016). The 2009 election, which Ali Bongo Ondimba won, was widely contested and was followed by violence (The Guardian, September 2016). The most recent election that took place on August 27, 2016 was between president Bongo Ondimba and Jean Ping. Despite complaints by the opposition and imbalances in the media coverage

14 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

of campaigns of the candidates, the voting and counting procedures were peaceful (AUEOM & ECCAS, August 2016). However, the calm atmosphere rapidly changed on August 31, 2016 when it was announced that the incumbent president had won the elections with just 6000 more votes (Köpp, September 2016). Jean Ping immediately contested the results, pointing to the statistically improbable results in Bongo Ondimba’s home province, Haut-Ogooue, where the voter turnout was 99% with 95% of the votes being in favour of Bongo Ondimba. Gabon’s National Autonomous and Permanent Election’s commission did not question this outcome, but Jean Ping’s supporters’ anger at the re-election of Bongo Ondimba escalated to riots in Libreville and other towns. There were street clashes between rioters and police, the country’s national assembly was set on fire, political parties’ buildings were attacked, and over one thousand people were arrested. Furthermore, the government claims that six people have been killed whereas the opposition claims the true figure lies between fifty and one hundred people (Dahir, September 106; Köpp, September 2016; SABC, September 2016). Ping called for a massive rejection of the election and for a re-count of the votes (ibid). Bongo Ondimba responded by ignoring those calls, calling Ping a fraud, and deploying security forces in large numbers. He also implemented Internet curfews, shut down media and text message services, and placed Ping under house arrest (BBC, September 2016; The Guardian, September 2016; Latif Dahir, September 2016). Ping stayed defiant and filed a legal challenge to Gabon’s constitutional court. The court counted the very ballots that the National Autonomous and Permanent Elections Commissions had earlier claimed they had burnt, and validated the original outcome of the election. This outcome was not surprising for Ping, and the opposition in general, who had already questioned the court’s independence; the court is headed by Marie- Madeleine Mborantsuo, who has two children with former president Omar Bongo (Jobe, September 2016; Latif Dahir, September 2016). Even after this validation, Jean Ping maintained that he was the winner of the election (Aljazeera, September 2016) Congo, Gabon’s neighbouring country, held its most recent presidential election on March 20 2016, less than six months after a highly controversial referendum. The referendum was initiated and supported by President Denis Sassous Nguesso, and would allow for constitutional changes, including the increase of the maximum of

15 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

presidential mandates from two terms to three terms and the removal of the maximum age of seventy years to be eligible to run for presidency (Winsor, September 2016). This would allow for 72-year old Sassou Nguesso to run for a third mandate. A large proportion of the opposition boycotted the referendum and 18 protestors were killed. Yet, President Sassou Nguesso won the referendum with 94.3% of the votes in his favour (ibid). Hence, the prospects of the 2016 presidential election were not positive, especially considering the 2002 and 2009 elections, both won by Sassou Nguesso, were boycotted and contested by opposition parties (France24, March 2016a). Nonetheless, following the 2015 referendum, steps were made towards incorporating the norm of democratic elections in the national bureaucratic structure. The Commission Nationale Electorale Independente (CNEI) was established. Yet, the opposition has explicitly expressed its concerns about the credibility and autonomy of institutions involved in the electoral process (EurActive, March 2016b). In early 2016, the opposition had vocalised their doubts about the fairness of the upcoming elections, pointing out documented preparations by the government for widespread fraud (Reuters, March 2016). Voters and journalists reported abnormalities in voter lists, and opposition groups claimed they were denied access to the counting of the polls (ibid). Yet, despite these accusations and the detainment of Sassou Nguesso’s main opponent, Jean-Marie Michel Mokoko, the presidential campaign was largely peaceful (EurActive, March 2016b). Sassou Nguesso eventually won the elections with sixty per cent of the votes. The opposition contested the results; Mokoko called his supporters to peaceful protests (New York Times, March 2016), but strikes and gun battles broke out in Brazzaville (Africa Confidential, April 2016). The government responded by using tear- gas against protestors, arresting opposition leaders, young activists and foreign journalists, as well as implementing a media black-out (ibid.). In contrast to Gabon and Congo, Cabo Verde is praised by many for being a model for African democracy. It is an archipelago with 314,000 residents that gained its independence from Portugal in 1975, following an 11-year liberation war (SNA, October 2016). The multi-party system was adopted in 1990 and it has since experienced two peaceful changes of power between government and opposition (Beck, March 2016). In 2011, Inoncencio Sousa Manuel conceded defeat to incumbent president Jorges Carlos Fonseca. President Fonseca sought re-election, and on October 10, 2016 won a landslide

16 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

victory with 74% of votes (ENCA, October 2016). The opponents Joaquim Monteiro and Albertino Graca both accepted the results. The elections followed parliamentary elections earlier the same year, where the opponent came into power for the first time in fifteen years (ENCA, September 2016). The state has four autonomous bodies that are responsible for the management of elections, namely the (i) National Election Commission (CNE), (ii) General Directorate of Support to Electoral Process (DGAPE), (iii) Census Commission, and (iv) Nucleo Operacional of Sociedad de Informação (AUEOM, October 2016). In general, the election procedure was relatively uneventful; there are no reports of violence, neither of the fear of violence, before, during, and after the election. The international community applauded Cabo Verdean politicians and institutions for their free and fair elections (SNA, October 2016).

State socialisation The Gabonese elite have not any aspect socialised the norm of democratic elections. Although Gabon has held regular elections, the highly improbable outcomes in Haut- Ogooue and the non-transparent practices of the National Autonomous and Permanent Election’s Commission, shows that the technical aspects of the norm have not been met. Furthermore, the imbalance in media coverage of the candidates and the seemingly inability of the candidates to collaborate, as they use violence and security forces rather than dialogue to solve the issue of contestation of the election, shows the lack of political will to solve the problem and lack of political commitment to the norm. Lastly, the moral aspect of the norm is not adhered to either, as both politicians do not seem to be fully commitment to constitutional rule, or to take all necessary measures to ensure this rule is maintained. The actions of the elite do not suggest they truly believe in the norm. The Congolese elite have not socialised any aspect of the norm of democratic elections either. First, although the elections were timely, they were not clear and transparent. Changes to the referendum allowed for the establishment of an electoral commission, but the autonomy of this commission was doubtful. Furthermore, there was no mechanism to deal with the contestation of the votes, suggesting that the technical aspect of the norm was not respected. Second, the political aspect was not respected as the opposition groups had limited access to the polls, and rather than

17 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

dialogue, hostile methods such as detainment were used to deal with problems. Last, the change in constitution and the use of violence, the detainment of individuals and the blacking out of the media shows that the elite were not committed to promoting the values of democratic elections and to deal with contestations using legal methods. The situation is different in Cape Verde, where the norm seems to have been socialised. First, the democratic election was a succession of previous elections, and was applauded for being fair and transparent. The state also has four autonomous bodies responsible for the management of elections, which were not reported to be problematic. This suggests that the technical aspect of the norm was met. Second, the political aspect of the norm was met as there were no reports of conflict between the parties, and the candidates who lost easily conceded their defeats. There were no reports of violence following the elections. Last, there were no reports that suggest the moral aspect of the norm was not met. The contestants did not engage in behaviour that suggested that they were not committed to holding regular elections and to concede elections when they lost.

Persuasion Despite obvious differences in the electoral history of Gabon, Congo and Cabo Verde, the AU acted nearly identically in the three cases. Regardless of clear reasons for concern about the 2016 elections of both Gabon and Congo, the AU did not mention those elections in its Decisions and Declarations of Assembly from January 2012 to July 2016. Yet, other elections were mentioned. The AU was not preoccupied with any of the three elections; there are no reports of social interaction between the AU and the political elite, at least not in the form of dialogues between the AU and the political elite of the three countries, or joint drafting of decisions or mechanisms to encourage and ensure that the elections were held democratically. This is remarkable as the AU Commission does give the possibility of sending “special advisory missions to provide assistance to the State Party for strengthening its electoral institutions and process” (AU 2007, art. 18.2). Nonetheless, the AU did deploy observation missions to all three countries. This is in line with the article 19 of the Charter, which obligates member states to invite the Commission to send an electoral observer mission. The aims of these missions are to (a)

18 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

determine if the “necessary conditions have been established and if the environment is conducive to the holding of transparent, free and fair elections in conformity with the principles of the Union governing democratic elections” prior to the elections (AU 2007, art. 20), and (b) observe the missions in a objective, impartial and transparent manner” in order to make sure the elections take place in accordance with the African Union’s instruments (AU 2007, art. 21.3; AUEOM & ECCAS, August 2016; AUEOM, March 2016; AUEOM, October 2016). It is noteworthy that Gabon and Congo have only signed the Charter but have not ratified it, and Cabo Verde has not signed it at all. The observation mission in Gabon was a joint effort by the AU and the Economic Community of Central African States (ECCAS). The observation mission was deployed in two phases. The Long Term Election Experts were deployed from August 7 until September 11, 2016. The Short Term Observers (STO), led by Cassam Uteem, the former president of the Republic of Mauritius and Abou Moussa, a former Special Representative of the UN Secretary General for Africa, were deployed from August 20 until September 2, 2016 to monitor the election on August 27, 2016 (AU, August 2016). The observation mission in Congo, led by Dileita Mohamed Dileita, a former Prime Minister of the Republic of Djibouti, was present from March 15 to March 24, 2016 to monitor the election on March 20, 2016 (AUEOM & ECCAS, August 2016). It is interesting to note that, although Dileita claimed that the elections were a noteworthy political advancement for the country, the EU had declined an invitation to monitor the elections because it believed that the organization of the election was flawed to the point of not being conducive for fair elections (EurActive, March 2016a; France24, March 2016b). This is relevant because the EU had engaged in dialogue with the Congolese elite for several months. The observation mission in Cabo Verde was deployed from September 24 until October 6, 2016 to monitor the election on October 2, 2016 (AUEOM, October 2016). Similarly to Congo, the AU only sent STO, and Manuel Serifo Nhamado, who served as an interim President of the Republic of Guinea Bissau, led this team. It is important to note that many of the observations made by the missions were written in reports, which were not directed to the states of interest but to the Commission, and some were sent after the elections. This suggests that there was in fact limited social interaction between the observer missions and the state elite, and the

19 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

message of the missions often came after the fact. Furthermore, the focus on the missions seems to be on the technical and political aspects of the norm, and not on the moral aspect of the norm. Yet, it was also within the mandates of the missions to meet and consult with key stakeholders involved in the electoral process, namely elite such as government officials, candidates and political parties (AUEOM & ECCAS, August 2016; AUEOM, March 2016; AUEOM, October 2016). However, using the available data sources it is not clear what was discussed. Also, the short time frame between the deployment of the missions and the election dates suggests that the AU did not have sufficient time to present strong messages with numerous arguments, and the state elite probably did not have sufficient time to consider and evaluate them critically.

Social influence Although the AU’s efforts of persuasion were minimal, there were some instances of social influence. More specifically, there were only instances of social reward, regardless of whether the state exhibited norm-compliant or norm-deviant behaviour. Although the Commission is expected to “evaluate the compliance of State Parties (AU 2007, Art. 44.2Aa), the Charter only specifies sanctions in case of unconstitutional change of power (AU 2007, art. 23). The sanctions that the AU considers range from denial of transport and communication links with other member states to the denial of participation in the activities of the Union (AU 2007, Art. 30), but these were not implemented in any of these cases. Despite obvious problems during the election in Gabon, the joint AU-ECCAS mission and the Chairwoman of the AU Commission, H.E. Dlamini-Zuma, did a fair bit of back patting. The observation mission congratulated the Gabonese people for their peaceful voting, “noted with satisfaction” that the open election was peaceful, and stated that the election was “a significant political breakthrough in the Gabonese Republic” (AUEOM & ECCAS, August 2016.). Yet, the AU failed to state whether they believed the elections were free and fair, and whether a re-count was necessary, in contrast to other international organisations such as the United Nations (UN) and the EU who called for a re-count (Köpp, September, 2016). Furthermore, H.E. Dlamini-Zuma expressed

20 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

criticisms about the escalation of violence in Gabon, but the criticism was not directed to specific actors (AU, March 2016.). Similarly, in Congo no social punishment was given despite abnormalities in the election procedure. On March 22, 2016 H.E. Dlamini-Zuma “urged” the government to unblock telecommunications (AU, March 2016), however this language seems mild considering the important role telecommunications play in society. Moreover, again, her comments were not directed at a specific actor and she failed to pinpoint the person or group who was accountable for the situation. President Sassou Nguesso has not only eluded criticism from the AU, he has served as a mediator in regional crises, even after his 2016 re-election (Muisyo, June 2016). Cabo Verde also received numerous social rewards from the AU. Before the election the AUEOM’s report “commended” the candidates’ respect for fairness and pluralism in media coverage of the election (AUEOM, October 2016) and Nhamadjo told the media that Cabo Verde was “an example as regards transparent elections” (Seychelles News Agency, October 2016).). In addition, Khabele Matlosa, the director of Political Affairs at the AU stated that Cabo Verde was an exception on a continent where a considerable amount of African leaders like to stay in power for too long” (ibid.). Interestedly, this could be seen as an indirect criticism to other African leaders, however as it is not directed to them, its impact is likely to be negligible. Despite the numerous praises, it is unlikely that those comments had significant impacts on the beliefs and behaviour of the candidates, as they were made so close to the election date. These data suggests that the social rewards given in the case of Gabon, Congo and Cabo Verde focused predominantly on the technical aspects and ignored the political and moral issues. Statements were made on the transparency and timeliness of the elections, and not so much on the political culture within the states or the commitment of the elite to the norms of democratic elections. The exception seems to be the statements directed toward Cabo Verde.

Scope conditions Even if the AU had attempted more persuasion methods in Gabon or Congo, or if it had been necessary in Cabo Verde, it is unlikely that it would have had large effects on the beliefs and behaviour of the contenders of the elections. This is due to several reasons.

21 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

First, the contenders of the elections in all the three cases were not new to the political environment and already had engrained beliefs. In Gabon, there is President Bongo Ondimba who is the son of former president Omar Bongo, who ruled the country for 42 years. Prior to the death of his father, he had already held positions in the government (BBC, September 2016a). His opponent, Jean Ping, is a former president of the UN’s General Assembly, a former AU Commission Chairman, and an ally-turned-rival of the Bongos (Reuters, September 2016). As the AU Commission Chairman he oversaw and contributed to the legitimatisation of numerous obscure elections. Moreover, prior to holding that position he was always close to the Gabonese ruling party and served as Omar Bongo’s cabinet director (Haque, October 2016). He was also married to, and has children with, Omar Bongo’s daughter (Reuters, September 2016). In Congo, there is president Sassou Nguesso who has ruled for 32 of the last 37 years. He ruled from 1979 until 1992, when he lost re-elections after the introduction of multiparty politics. He regained power in 1997 when his rebel forces ousted President Pascal Lissouba (Winsor, March 2016). His opponent, Mokoko, is his long-time ally. He is a retired general who served as the president’s peace and security advisor (EurActive, March 2016b). In Cabo Verde, president Fonseca is not only the incumbent president, he is also a former minister of foreign affairs (Seychelles News Agency, June 2014). One of his opponents, Monteiro, is a veteran politician who was an influential figure during the fight for independence (ibid). Second, the focus of the norm by the AU commission in the period of the elections was too political. Although the greatest focus was on the technical aspect of the norm, there was more focus on the political than the moral aspect of the norm. Especially in the undemocratic environments, such as that of Gabon and Congo, the political careers of many politicians depend on the outcomes of elections. Hence, state elite in power may attempt to show they abide to the norm in the technical term to relieve pressure on themselves, but they are prepared to violate the political aspects of the norm in order to stay in power. This scope condition is unlikely to be a problem in Cabo Verde where it seems the moral aspect has already been accepted, making the political aspect less sensitive. Third, it is unlikely that the contenders of the elections in the three cases considered the AU as a respectable, credible or powerful organisation in terms of

22 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

democratic elections. Although the Charter states that the “electoral observer mission shall be conducted by appropriate and competent experts in the area of election monitoring”(AU 2007, art. 21.2), the AU’s credibility is undermined because (i) the Chair of the AU, Idriss Deby, has ruled his country for 26 years and recently won a fifth term, while opposition accused him of fraud and the disappearance of soldiers who voted against him (BBC, April 2016), and whose regime has been described as an unchanging electoral authoritarian regime (Lindberg 2006:140), and (ii) the observation missions can be led by politicians such as Nhamadjo, who served as interim President of the Republic of Guinea Bissau following a coup d’état (ASemana, May 2012). Fourth, in the three cases, the AU did not deliberately expose the majority of the contenders to the norm of democratic elections frequently, especially not within their own contexts. It is so that the presidents of the states are also representatives of their states within the AU and that some contenders have represented the AU, so one would expect that they have often been exposed to the general norm of democratic elections. However, their behaviour as well as the choices made by the AU suggests it is likely that the moral aspect of the norm is not “taken-for-granted” within the AU itself. Currently, the same people who need to be persuaded are expected to persuade other people. Fifth, it is interesting to note that although the norm of democratic elections is fairly broad, and at times even vague, the AU puts little effort to frame it to fit the African context. The norm is generally presented within the peace and security framework, but it is not linked to distinguishing African norms. It seems like the AU has taken the Western understanding of democracy, promoted within the international community, and has made little attempts to shape and alter it to make it ‘resonate’ within the African context.

Discussion The aim of this study was to answer the following research question, “Under what conditions, and using which mechanism, can the African Union (AU) encourage its member states’ elite to socialise the norm of democratic elections?” The suggested model indicates that the AU can use persuasion and social influence to encourage its

23 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

member states to socialise the norm of democratic elections, provided certain conditions have been met. The results of the case analyses of the 2016 elections in Gabon, Congo, and Cabo Verde provide some support for this model. The results show that the AU hardly used persuasion in the period leading up to the elections and often used social rewards inappropriately. Although the African Charter on Elections, Democracy and Governance puts forward three aspects of the norm of democratic elections, the AU mainly focused on the technical and political aspect, and not on the moral aspect. This suggests the African Union has not put enough effort into changing the way of thinking and feeling of the states. Furthermore, the deployed observation missions had the clear aim of observing and assessing the elections, and did not have sufficient time to have meaningful social interaction with state elite. Hence, the first two hypotheses, which stated that an increase in the strength of the message and the number of arguments that the African Union provided would result in the acceleration of the process of state socialisation, can be tentatively confirmed. In both Gabon and Congo, the elections were followed by violence, suggesting the AU could have done more to minimize the likeliness of those outcomes. It is noteworthy, however, that the EU’s dialogues with Congo prior to the elections did not yield positive results either. Further analysis of the nature of the dialogue and the framing of the norm is necessary to draw stronger conclusions. It is also noteworthy that Cabo Verde exhibited norm-compliant behaviour despite the absence of persuasion. It is probable that Cabo Verde’s success can be attributed to factors the AU cannot influence, such as its geographic and population size, and its relatively recent independence. The third hypothesis, which states that an increase in the use of social reward to praise norm-compliant behaviour will result in the acceleration of the process of state socialisation can be confirmed in the case of Cabo Verde, keeping in mind the little time between the social rewards and the elections. The fourth hypothesis, which focused on social punishments, could not be tested, as the AU did not give any social punishments. This is understandable in a context where those who have the power of giving social punishments have sometimes come to power undemocratically or lead undemocratic governments themselves. The fifth hypothesis, which states that the inappropriate use of social rewards and social punishments will decelerate the process of state

24 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

socialisation, was confirmed in the cases of Gabon and Congo. The AU’s indiscriminate use of social rewards undermines the potential positive effects social influence. Moreover, most of the scope conditions were not met in the three cases and the norm was not framed to suit the African context. Therefore, it is unlikely that any attempts of persuasion and social influence by the AU would be effective. These results are in line with previous constructivist literature. First, there is no evidence that suggests that rationalist mechanisms play a more important role than constructivist mechanisms. The results suggest that the AU has to put more effort into changing the interests of all parties involved. Second, the results also confirm that the AU member states are state-above-society states, where the civil societies are weak and there are no clear indications of the (strong) presence of networks among domestic and transnational actors. Third, the fact that those sent to monitor the countries are from countries with undemocratic elections shows that either the number of AU member states with democratic elections is limited, or that the most powerful and influential states do not exhibit norm-compliant behaviour. Hence, the mechanisms of learning and copying are unlikely to be effective.

Implications This research is one of the few empirical studies on the socialisation of both why and how international organisations can encourage the socialisation of their norms, especially in international organisations that are largely comprised of undemocratic countries. It has also provided the AU with a framework to approach its member states to maximize norm-compliant behaviour, while focusing on the different aspects an international norm. If the AU wants its member to have democratic elections and wants to minimize the occurrence of post-election violence, it needs to revise its policies and its focus with regards to the state socialisation of its norms, and especially its norm of democratic elections. It must be aware of the potential effects of persuasion and socialisation, and must ensure the technical, political and moral aspects of the norm are addressed during its interaction with member states. It must also consider whether its policies are feasible within its context.

25 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

This can be generalised to similar international organisations, which can use a similar theoretical framework when making their policies regarding the socialisation of their norms. It is important that such organisations’ theoretical and practical considerations on how to induce changes in state behaviour are context-specific.

Limitations Although this study has contributed to the literature on AU and the socialisation of norms, there are some limitations. First, in order to better investigate the role the AU can play in socialising a state, a case study of a state that made the change from non- democratic to democratic elections would have been most appropriate. However, such a state was not an option resulting from the method of case-selection used in this study. Second, although AU’s documents were used, there was a heavy reliance on secondary sources, especially news articles. Further analyses of the AU archives, and minutes of meetings held within the AU and with states elite, would have allowed for a more thorough understanding of the AU’s efforts. Also, the inclusion of analyses of the legal and bureaucratic structures of the states could allow for a more comprehensive understanding of state socialisation. Third, Western theories have been used to analyse a norm within the African context. There are currently no theories that have been derived inductively from the African context, although the African political context and history is significantly different from the Western context, and any other context for that matter. In summary, there is still uncertainty about the answer of the research question because of the limited access to information and theories regarding the topic.

Recommendations for future research In order to have a more certain answer to the research question, additional research should be done with access to more information from the AU as well as its member states. Furthermore, more research should be done on the evolvement of the norm of democratic elections within the AU itself. It must be investigated whether the norm is truly an African norm and whether it is ‘taken-for-granted’ within the organisation. Moreover, a process-tracing analysis of Cabo Verde, or any other African country that has socialised the norm of democratic elections, could allow for a better understanding

26 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

of the process of state socialisation and the role the AU can play. Lastly, an inductive approach can be used to explain the behaviour of state elite with regards to African norms, in order to induce theories that are more applicable in the African context.

27 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

Bibliography Acharya, A. (2014). Constructing a security community in Southeast Asia: ASEAN and the problem of regional order. New York: Routledge. Africa Confidential. (2016, April 1) Sassou makes it modest. Africa Confidential. Retrieved November 29, 2016 from http://www.africa-confidential.com/article- preview/id/11608/Sassou_makes_it_modesthttp://www.reuters.com/article/us -congo-election-idUSKCN0WM0B2. Africa News. (2016, February 14). Congo's ex-security chief confirms candidacy in March 20 polls. Africa News. Retrieved November 29, 2016 from http://www.africanews.com/2016/02/14/congo-s-ex-security-chief-to-stand- for-march-polls/. Alderson, K. (2001). Making Sense of State Socialization. Review of International Studies, 27(3), 415-433. doi: https://doi.org/10.1017/S0260210501004156. Aljazeera. (2016, September 24). Gabon: Ping rejects court ruling on Bongo election win. Aljazeera. Retrieved January 10, 2017 from http://www.aljazeera.com/news/2016/09/gabon-ping-rejects-court-ruling- bongo-election-win-160923162554678.html. Asemana. (2012, May 17). Cape Verde opposed to selection of Serifo Nhamadjo as Guinea Bissau’s interim president. Asemana. Retrieved December 10, 2016 from http://www.asemana.publ.cv/spip.php?article76301. AU. (2000). Constitutive Act of the African Union. Retrieved on December 12, 2016 from https://www.au.int/en/about/constitutive_act. AU. (2007). African Charter on Democracy Elections and Governance. Retrieved on December 12, 2016 from http://www.achpr.org/instruments/charter-democracy/. AU. (2016, March 22). AU Commission Chairperson calls for calm and restraint in Congo Republic as election results are awaited. AU. Retrieved November 29, 2016 from http://www.peaceau.org/uploads/auc.congo.elections.23.03.16.pdf. AU. (2016a, August 27). Republic of Gabon's 27 August 2016 Presidential Election. AU. Retrieved November 29, 2016 from

28 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

http://www.au.int/en/newsevents/31232/republic-gabons-27-august-2016- presidential-election. AU. (2016b, August 29). AU congratulates Gabon for peaceful polls, calls for respect of results. AU. Retrieved November 29, 2016 from http://www.au.int/en/pressreleases/31316/au-congratulates-gabon-peaceful- polls-calls-respect-results. AU. (2016, September 1). AU condemns the outbreak of violence and urges for peaceful resolution of election-related disputes in Gabon. AU. Retrieved November 29, 2016 from http://www.au.int/en/pressreleases/31324/au-condemns-outbreak-violence- and-urges-peaceful-resolution-election-related. AUEOM. (2016, March 16). Arrival Statement - AU Election Observation Mission for the First Round Presidential Election in the Republic of Congo - 20 March 2016. Retrieved November 29, 2016 from http://www.au.int/en/pressreleases/21376/arrival-statement-au-election- observation-mission-first-round-presidential. AUEOM. (2016, October 4). African Union Observer Mission to the Presidential Election of 2ndOctober 2016 in the Republic of Cabo Verde: Preliminary Statement. Retrieved November 29, 2016 from http://www.au.int/en/sites/default/files/pressreleases/31474-pr- aueom_cv_preliminary_statement_eng_october_2016_.pdf. AUEOM & ECCAS. (2016, August 30). Preliminary Statement: Joint AU - ECCAS Observation Mission for the 27 August 2016 Presidential Election in the Republic of Gabon. Retrieved November 29, 2016 from http://www.au.int/en/pressreleases/31318/preliminary-statement-joint-au- eccas-observation-mission-27-august-2016. Baxter, P., & Jack, S. (2008). Qualitative case study methodology: Study design and implementation for novice researchers. The Qualitative Report, 13(4), 544-559. Retrieved from http://nsuworks.nova.edu/tqr/vol13/iss4/2/. BBC. (2016, April 22). Chad's President Idriss Deby wins fifth term. BBC. Retrieved November 10, 2016 from

29 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

http://www.bbc.com/news/world-africa-36110158. BBC. (2016a, September 4). Ali Bongo: Like father, like son? BBC. Retrieved December 11, 2016 from http://news.bbc.co.uk/2/hi/africa/8238860.stm. BBC. (2016b, September 7). Gabon election: Ali Bongo criticises over vote ‘anomalies’. BBC. Retrieved November 29, 2016 from http://www.bbc.com/news/world-africa-37295175. Beck, J. (2016, March 18). Cape Verde: peaceful political change. Deutsche Welle. Retrieved November 29, 2016 from http://www.dw.com/en/cape-verde-peaceful-political-change/a-19127634. Björkdahl, A. (2002). Norms in International Relations: Some Conceptual and Methodological Reflections. Cambridge Review of International Affairs, 15(1), 9- 23. doi: 10.1080/09557570220126216. Buchanan, E. (2016, November 8). Gabon opposition leader Jean Ping says France should arrest President Ali Bongo. IB Times. Retrieved November 29, 2016 from http://www.ibtimes.co.uk/gabon-opposition-leader-jean-ping-says-france- should-arrest-president-ali-bongo-1590468. Cialdini, R.B., & Trost, M.R. (1998). Social influence: Social norms, conformity and compliance. In Lindzey, G., Gilbert, D.T., & Fiske, S.T. (Ed), The Handbook of Social Psychology, 1. Oxford: Oxford University Press. Checkel, J.T. (1997). International Norms and Domestic Politics: Bridging the Rationalist-Constructivist Divide. European Journal of International Relations, 3(4), 473-95. Retrieved from http://journals.sagepub.com/doi/pdf/10.1177/1354066197003004003. Checkel, J.T. (2001). Why Comply? Social Learning and European Identity Change. International Organization, 55(3), 553-588. doi: 10.1162/00208180152507551 Cortell, A.P., & Davis, J.W.Jr. (2000). Understanding the Domestic Impact of International Norms: A Research Agenda. International Studies Review, 2(1), 65-87. doi: 10.1162/00208180152507551. Dai, X., & Mattiacci, E. (2011). The Compliance Gap and the Efficacy of International Human Rights Institutions. Retrieved from https://kb.osu.edu/dspace/handle/1811/49696.

30 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

Dobbin, F., Simmons, B., & Garrett, G. (2007). The Global Diffusion of Public Policies: Social Construction, Coercion, Competition, or Learning?. Annual Review of Sociology, 33, 449-472. doi: 10.1146/annurev.soc.33.090106.142507. ENCA. (2016, September 6). Cape Verde opposition resigns after poll defeat. ENCA. Retrieved November 29, 2016 from http://www.enca.com/africa/cape-verde-opposition-resigns-after-poll-defeat. ENCA. (2016, October 3) President Fonseca wins second term in Cape Verde vote. ENCA. Retrieved November 29, 2016 from http://www.enca.com/africa/open-the-document-in-a-new-window-x- president-fonseca-wins-second-term-in-cape-verde-vote. Euractive. (2016a, March 21). Internet blackout in Congo as government fears post- election unrest. Euractive. Retrieved November 29, 2016 from https://www.euractiv.com/section/global-europe/news/internet-blackout-in- congo-as-government-fears-post-election-unrest/. Euractive. (2016b, March 18-25). Elections in the Republic of Congo. Euractive. Retrieved November 29, 2016 from http://en.euractiv.eu/wp-content/uploads/special-report/EurActiv-Special- Report-Elections-in-the-Republic-of-Congo-English-1.pdf. Finnemore, M., & Sikkink, K. (1998). International Norm Dynamics and Political Change. International Organization, 52(4), 887-917. Retrieved from https://www.jstor.org/stable/2601361?seq=1#page_scan_tab_contents France24. (2016a, March 24). Republic of Congo’s President Denis Sassou Nguesso, who was reelected for a third term on Thursday, began his political career as a Marxist-Leninist and has become a wealthy strongman determined to continue his 32 years in power. France24. Retrieved November 29, 2016 from http://www.france24.com/en/20160324-congo-president-sassou-nguesso- reelected-first-round. France24. (2016, August 16). Two opposition heavyweights in Gabon are set to back former African Union chief Jean Ping in his bid to oust President Ali Bongo at the ballot box this month. France24. Retrieved November 29, 2016 from http://www.france24.com/en/20160816-gabon-opposition-leaders-back-ex-au- chief-ping-president.

31 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

Greenhill, B. (2010). The Company You Keep: International Socialization and the Diffusion of Human Rights Norms. International Studies Quarterly, 54(1), 127- 145. doi: 10.1111/j.1468-2478.2009.00580.x. The Guardian. (2016, September 7). The Guardian view on Gabon: recount the votes. The Guardian. Retrieved November 29, 2016 from https://www.theguardian.com/commentisfree/2016/sep/07/the-guardian- view-on-gabon-recount-the-votes. Haque, N. (2016, October 1). A week in post-election Gabon. Aljazeera. Retrieved November 29, 2016 from http://www.aljazeera.com/blogs/africa/2016/10/week-post-election-gabon- 161001064825797.html. Hafner-Burton, E. M., & Tsutsui, K. (2005). Human rights in a globalizing world: The paradox of empty promises. American journal of sociology, 110(5), 1373-1411. Retrieved from http://pages.ucsd.edu/~ehafner/pdfs/paradox_promises.pdf. Hooghe, L. (2005). Several Roads Lead to International Norms, but Few via International Socialization: A Case Study of the European Commission. International Organization, 59(4), 861-98. Retrieved from http://pages.ucsd.edu/~ehafner/pdfs/paradox_promises.pdf. Jobe, A. (2016, September 14). AU to send observers to help Constitutional Court in opposition election appeal case. SMBCNews. Retrieved November 29, 2016 from https://gambia.smbcgo.com/2016/09/14/au-send-observers-help-gabon- constitutional-court-opposition-election-appeal-case/. Johnston, A.I. (2001). Treating International Institutions as Social Environments. International Studies Quarterly, 45(4), 487-515. Retrieved from https://www.jstor.org/stable/3096058?seq=1#page_scan_tab_contents. Kane, I. (2008). The implementation of the African Charter on Democracy, Elections and Governance. African Security Studies, 17(4), 43-63. doi:10.1080/10246029.2008.9627496. Kasaija, P.A., (2013). The African Union (AU), the Libya Crisis and the notion of ‘African solutions to African problems’. Journal of Contemporary African Studies, 31(1), 117-138. doi: 10.1080/02589001.2012.761463.

32 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

Kioko, B. (2003). The right of intervention under the African Union's Constitutive Act: From non-interference to non-intervention. Revue Internationale de la Croix- Rouge/International Review of the Red Cross, 85(852), 807-826. Retrieved from https://www.icrc.org/eng/assets/files/other/irrc_852_kioko.pdf. Köpp, D. (2016, September 2). Opinion: Gabon and the inglorious African institutions. Deutsche Welle. Retrieved November 29, 2016 from http://www.dw.com/en/opinion-gabon-and-the-inglorious-silence-of-african- institutions/a-19523708. Krook, M.L., & True, J. (2012). Rethinking the Life Cycles of International Norms: The United Nations and the Global Promotion of Gender Equality. European Journal of International Relations, 18(1), 103-127. doi: 10.1080/02589001.2012.761463. Latif Dahir, A. (2016, September 23). Gabon’s highest court says Ali Bongo won its controversial presidential elections. Quartz. Retrieved November 29, 2016 from http://qz.com/790385/ali-bongo-has-won-re-election-as-president-of-gabon- extending-his-familys-50-year-rule-2/. Moolakkattu, J.S. (2010). The Role of the African Union in Continental Peace and Security Governance. India Quarterly: A Journal of International Affairs, 66(2), 151-165. doi: 10.1177/097492841006600202. Muisyo, V. (2016, June 28). Congo: Nguesso hosts Kabila for talks on DRC political situation. Africanews. Retrieved on January 10, 2016 from http://www.africanews.com/2016/06/28/congo-nguesso-hosts-kabila-for- talks-on-drc-political-situation/. The New York Times. (2016, March 30). A Marred Election in the Republic of Congo. The New York Times. Retrieved November 29, 2016 from http://www.nytimes.com/2016/03/31/opinion/a-marred-election-in-the- republic-of-congo.html?_r=1. Omorogbe, E.Y. (2012). The African Union, responsibility to protect and the Libyan crisis. Netherlands International Law Review, 59(2), 141-163. doi: 10.1017/S0165070X12000150. Payne, R.A. (2001). Persuasion, Frames and Norm Construction. European Journal of International Relations, 7(1), 37-61. Retrieved from http://journals.sagepub.com/doi/pdf/10.1177/1354066101007001002.

33 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

Petty, R. E., & Cacioppo. J. T. (1984). The effects of involvement on responses to argument quantity and quality: Central and peripheral routes to persuasion. Journal of Personality and Social Psychology, 46(1), 69-81. Petty, R. E., & Cacioppo, J. T. (1986). The elaboration likelihood model of persuasion. Communication and persuasion, 1-24. Retrieved from http://www.psy.ohio- state.edu/petty/documents/1986ADVANCESsPettyCacioppo.pdf. Risse, T., & Sikkink, K. (1999). The socialization of international human rights norms into domestic practices: Introduction. Cambridge Studies in International Relations, 66, 1-38. Retrieved from http://assets.cambridge.org/97805216/50939/sample/9780521650939wsc00. pdf. Risse, T. (1999). International Norms and Domestic Change: Arguing and Communicative Behavior in the Human Rights Area. Politics & Society, 27(4), 529-559. Retrieved from http://journals.sagepub.com/doi/pdf/10.1177/0032329299027004004. Reuters. (2016, March 20). Protest, tear gas in Congo as Sassou Nguesso seeks to extend rule. Reuters. Retrieved November 29, 2016 from http://www.reuters.com/article/us-congo-election-idUSKCN0WM0B2. Reuters. (2016a, September 8). African Union's Gabon crisis mediators delay visit as mission leader 'tired': Gabon. Reuters. Retrieved November 29, 2016 from http://www.reuters.com/article/us-gabon-election-mediation-idUSKCN11E2AR. Reuters. (2016b, September 9). Gabon says African Union’s mediation mission is delayed. Indian Express. Retrieved November 29, 2016 from http://indianexpress.com/article/world/world-news/gabon-says-african- union-mediation-mission-is-delayed-disputed-election-results-recount- 3021348/. Rummel, R.J. (1976). The Conflict Helix. News Brunswick: Sage Publications. SABC. (2016, September 7). US welcomes AU engagement in Gabon. SABC. Retrieved November 29, 2016 from http://www.sabc.co.za/news/a/d54f87804e26fb109e53bf4bf75cdd95/US- welcomes-AU-engagement-in-Gabon-20160709.

34 Sandra Kamga Dr. Yvonne Kleistra S1149350 Thesis Political Science: International Organisation Word count: 10876 Deadline: 12th of January 2017

SNA. (2014, June 16). The Cabo Verde miracle - An African island state with a winning record. SNA. Retrieved November 29, 2016 from http://www.seychellesnewsagency.com/articles/742/The+Cabo+Verde+miracle +-+An+African+island+state+with+a+winning+record. SNA. (2016, October 2). President Fonseca eyes second term in Cape Verde vote. SNA. Retrieved November 2016 from http://www.seychellesnewsagency.com/articles/6012/President+Fonseca+eyes +second+term+in+Cape+Verde+vote. Shaban, A.R.A. (2016, September 20). Gabon's constitutional court 'throws out' AU observer mission. Africa News. Retrieved November 29, 2016 from http://www.africanews.com/2016/09/20/gabon-s-constitutional-court-throws- out-au-observer-mission/. Thomas, D.C. (2002). Boomerangs and Superpowers: International Norms, Transnational Networks and US Foreign Policy. Cambridge Review of International Affairs, 15(1), 25-44. doi: 10.1080/09557570220126225. Wendt, Alexander. (1992). Anarchy is What States Make of It. The Social Construction of Power Politics. International Organization, 46(2), 391-425. Retrieved from https://www.jstor.org/stable/2706858. Winsor, M. (2016, September 19). Republic of Congo Presidential Election 2016: Guide to Candidates, Issues, Rules and Results. IB Times. Retrieved November 29, 2016 from http://www.ibtimes.com/republic-congo-presidential-election-2016-guide- candidates-issues-rules-results-2339135.

35