LEGITIMATE DEMOCRATISATION: UNCONSTITUTIONAL UPRISINGS VS MASS DEMONSTRATIONS FOR REGIME CHANGES IN AFRICA. Prof Dirk Kotzé

Transcription

1 LEGITIMATE DEMOCRATISATION: UNCONSTITUTIONAL UPRISINGS VS MASS DEMONSTRATIONS FOR REGIME CHANGES IN AFRICA INTRODUCTION Prof Dirk Kotzé University of South Africa During the past decade changes in the governments of many African states have not followed the constitutional route of electoral decisions but have been in the form of popular uprisings (notably in North Africa) or demonstrations supported by the military and external powers. The incidents of overt and unpopular military coups d état (like in Mali, Mauretania, Guinea-Bissau, Togo, Central African Republic, etc) are not included here. Some of these changes have been accepted by the international community and regional organisations and others not. While all of them have been unconstitutional in nature, the question is whether there are norms both in political theory and in practice (diplomacy) that can determine the conditions of legitimacy and acceptability, or illegitimacy? This paper is therefore located in the domains of democracy, democratisation, democratic regression (or Huntington s reverse waves), transitions, elections, constitutionalism and the international legal norms for recognition of governments. These issues are not confined to African states. Twenty years ago the fall of the Berlin Wall and the popular uprisings in several Eastern European countries were internationally accepted as legitimate regime changes. About ten years earlier the same happened in Iran and with the exception of the USA, most other states recognised the new government. In 1997 the Erbakan government was forced in a post-modernist coup by the Turkish military to resign and to call for a new election. Again, the international community accepted it. In February 2012 the same happened with the constitutionally-elected Pres. Mohamed Nasheed in the Maldives also without any international objections (Time, April 16, 2012: 60). The majority of examples mentioned here involved governments that were elected. Does it mean that these legitimate regime changes also challenge us to reconsider the general acceptance of elections as the most authentic element of democracy? This first considers the available theoretical tools to guide us towards possible answers for the issues raised here. It is followed by the normative framework developed over time by the Organisation of African Unity (OAU) and the African Union (AU) regarding unconstitutional changes of governments. Thirdly, we shall look at examples or cases of such unconstitutional changes and why some were unacceptable and others were more 1

2 acceptable. Finally, in a comparison of the two sets of processes we shall attempt to identify factors that can be considered as norms to guide our judgments about unconstitutional changes. i)theoretical apparatuses for unconstitutional changes Unconstitutional changes are arguably premised on the theoretical premise that the means of change will determine the end or outcome; in other words, procedural legitimacy. Huntington & Dominguez (1975: 12) however, refer to theories in which the outcome depends on whether the transition was violent or peaceful, and whether socioeconomic change was rapid or gradual. They also mentioned another set of theories in which the outcome is decisively shaped by the nature of the previously existing traditional society. Centralised bureaucratic empires will be modernised into centralised authoritarian regimes, while feudal systems will be modernised into democratic and pluralistic systems. Yet the phenomena mentioned in the beginning as examples of unconstitutional change do not really fit these theoretical expositions. Later Huntington (1991) concentrated more on democratisation (and the third wave) as a theoretical realm to understand transitions. For analytical purposes he identified three types of democratisation processes, namely transformation (or Juan Linz s reforma), replacement (similar to Linz s ruptura) and transplacement (Huntington 1991: 114). Replacements are the closest to unconstitutional changes discussed here. Huntington identified cases like Argentina (1983), Portugal (1974), Greece, Romania (1989), East Germany (1989) and the Philippines (Marcos) as representative of replacements. In most of these cases the turning point was when the military withdrew their support from the regime. The popular image of these transitions, according to Huntington, was that they were brought down by people power in the form of mass mobilisation. In East Germany it was in the form of massive departures of citizens from their country and thereafter massive street demonstrations in Leipzig and Berlin. The emphasis in the regime change was on a sharp, clean break with the past. The new regime used forward legitimacy to justify their rule, in other words what they intend to do about the future and their lack of involvement in the previous regime (Huntington 1991: , ). Huntington s theoretical discussion is primarily empirical comparisons and not really theoretical generalisations that can assist us in reaching clarity about the normative nature of developments much closer in time to events than his historical analyses several years after the fact. For example, how could a political scientist in 2011 approach the developments in Libya as they were unfolding? These theoretical approaches provide not much assistance. 2

3 Another set of theoretical perspectives are associated with transitions. Well-known in this regard is the collection edited by Guillermo O Donnell, Philippe C. Schmitter and Laurence Whitehead (1986). These studies are not primarily theoretical in nature but rather comparative and empirical. Whitehead (1986: 4) for example, also noted that most of their cases involved transitions from authoritarianism of the right and in countries with political cultures that included liberalism and constitutionalism. Most of the African examples used in this discussion do not comply with these criteria. In conclusion, the existing theoretical instruments in political science are not particularly appropriate for dealing with unconstitutional government changes. The intention here is not to develop another theoretical framework but to identify factors or considerations which can arguably become part of a comprehensive theoretical approach. For the purpose of this discussion aspects of democratic theories, legitimacy and constitutionalism will be used. The focus now moves to the progress already made by African institutions to develop a normative framework on unconstitutional changes. ii) A normative framework regarding unconstitutional changes developed by African institutions The African continent has been characterised by several forms of regime changes since the 1960s in the sense of anti-colonial and independence wars, military coups d état, demonstrations and uprisings but seldom by means of elections. During the period of third wave democratisation in the 1990s the dilemma arose about how to make a judgement about legitimate versus illegitimate changes of government. The OAU adopted an approach that linked it to constitutionalism. In 1995 the OAU took its first initiative by establishing a committee of ambassadors to investigate ways of dealing with unconstitutional changes in governments. They produced the Report of the OAU Central Organ Sub-Committee on the Preparation of a Blue-print for Dealing with Unconstitutional Changes of Government in Africa (Sub- Cttee/Central Organ/Rpt. (III) 1996). In 1996 the states of the Great Lakes region decided to impose a regional blockade against the Burundi government following the unconstitutional change of government. The following year the OAU condemned the coup in Sierra Leone and called on all African countries and the international community at large to refrain from recognizing the new regime and lending support in any form whatsoever to the perpetrators of the coup d état (CM/Dec. 357 (LXVI), para. b (1997). According to Odinkalu (2008: 3) it was the first time in the OAU s history that it reached a consensus about its refusal to recognise a new regime. 3

4 In 1999 the OAU adopted the Algiers Declaration on Unconstitutional Changes of Government and it was followed a year later by the Lomé Declaration for an OAU Response to Unconstitutional Changes of Government (AHG/Decl. 5 (XXXVI) 2000). For the first time a more precise definition of unconstitutional changes was provided it included: Any putsch or coup d état against a democratically elected government, or Any intervention by mercenaries to replace a democratically elected government, or Any replacement of a democratically elected government by armed dissidents or rebels, or Any refusal by an incumbent government to relinquish power to the winning party or candidate after free, fair and regular elections. The Declaration also provided enforcement measures for such decisions: the OAU could suspend the membership of the affected state and economic sanctions and other forms of isolation could be imposed. The Declaration was embraced by the new AU and also included in its Constitutive Act. In Article 4 it was identified as one of its founding constitutional principles. Article 30 also states: Governments which shall come to power through unconstitutional means shall not be allowed to participate in the activities of the Union. In 2007 the AU adopted the African Charter on Democracy, Elections and Governance (AU 2007). It formalised the AU s stance on a number of matters, including elections and also unconstitutional changes of government. Article 23 adopted the four descriptions of unconstitutional change in the Lomé Declaration but added a fifth one: Any amendment or revision of the constitution or legal instruments, which is an infringement on the principles of democratic change of government (AU 2007: 10). One thinks immediately of attempts by Presidents to extend their terms of office beyond the two terms in many countries: Pres. Nujoma in Namibia succeeded with it; Pres. Tandja in Niger failed in 2009 and was removed by the military; in 2002 Pres. Bakili Muluzi in Malawi failed also in his effort. Article 25 in the Charter set a number of sanctions on perpetrators of unconstitutional change. For example, the Peace and Security Council must suspend the state s participation in the AU. It also determines that the perpetrators shall not be allowed to participate in elections after a transition or hold any public positions. It even allows for sanctions against another state that instigated or supported the unconstitutional change (AU 2007: 10). 4

5 Implementation of this framework is determined by many considerations. The controversy about the Libyan situation in 2011 and the absence of a resolution of the situation in Madagascar since 2009 are merely two examples of the complexity. The framework involves a number of aspects that deserve attention. The first is that a coup d état is relatively easy to identify when it involves military forces but when their involvement is more covert in nature, it is less obvious. Francis Ikome (2007) even refers to good coups and bad coups. The examples of Turkey (February 1997) and the Maldives (2012) already mentioned and even Portugal (1974) also complicate objective definitions of a coup. Ikome (2007: 7-8) refers to Edward Luttwak s definition of a coup consists of the infiltration of a small but critical segment of the state apparatus, which is then used to displace the government from its control of the remainder. Ikome is critical of this definition, because the military or any other organised force is not necessarily the defining feature of a coup it should be rather any seizure of state apparatuses by extra-legal tactics (Ikome 2007: 8). The problem with these definitions is that they also include legitimate uprisings like those in North Africa or Eastern Europe, revolutions or legitimate civil conflicts. A narrow interpretation would mean that only elections (or parliamentary and judicial actions) would be legal tactics. Anything else would be a coup. The third possibility of unconstitutional changes, i.e. replacement of democratically elected government by armed dissident groups and rebel movements, has a number of implications. The first is whether it makes the governments of Laurent-Desiree Kabila (DRC), Paul Kagame (Rwanda) or Yoweni Museveni (Uganda) illegitimate or unconstitutional? All of them waged civil wars against their own regimes, claimed victory and assumed government power without any electoral processes. They were accepted as representatives in the UN, OAU and many other international bodies. The second implication is: does it also disqualify an armed uprising or rebellion that did not end in a victory or regime change but that produced a negotiated transition or regime change in which the rebels are one of the participants. One thinks about the negotiations in the DRC during the Joseph Kabila era, the Burundi peace process or the one in Côte d Ivoire before the latest election, as well as the Comprehensive Peace Accord in the Sudan (2005). The third implication is: what about unarmed dissidents? The changes in Europe (especially in Eastern Europe) or in Iran (1979) and more recently in Tunisia and Egypt were not condemned as internationally unacceptable. The situations in Libya and Syria i.e. armed dissidents are more complex but also not condemned as illegitimate. The complicating factor was NATO s involvement to support the armed rebels in Libya. 5

6 However the Transitional National Council was accepted by the AU once it indicated its willingness to become a national unity government. Their representative status was more of an impediment than the tactics they used to come to power. Most African governments consist of a directly-elected executive President together with a Prime Minister or Deputy President and council of Ministers. The AU s normative framework uses the phrase democratically elected government but the question is: who is the government? If the President is replaced by the Deputy President (as in the Maldives) or by the Speaker (as in Guinea-Bissau, 2009) after pressure by the military (or any other state institution) to resign, is it unconstitutional in other words, is it a change in government? When Marc Ravalomanana was forced to resign in March 2009 and to hand over power to a military council which in turn appointed a former civilian Mayor of Antananarivo, Andry Rajoelina, as President, was it a coup and unconstitutional? One of the most complicating implications of the AU normative framework is its use of the phrases democratically elected governments and free and fair elections. Most democratic theorists are adamant that a democracy depends on free and fair elections. Adam Przeworski et al. (2000: 23-28) for example, use Botswana as an illustration to conclude that democracy depends on alteration of governments, in other words, that incumbent governments must have transferred their power to a victorious opposition party before it can be a democracy. However, elections are not necessary synonymous with democracy and can be also undemocratic or can have undemocratic consequences. The requirement that elections must be free and fair is also not particularly helpful, because most observer governments are not generally willing to conclude that an election was unfair. Observer NGOs are indeed more willing to reach such a conclusion although such conclusions do not have any diplomatic implications. Where problems were experienced, elections are often declared as substantially free and fair. The absence of a fair election would create a diplomatic conundrum, because it means that the new government cannot be recognised and therefore no diplomatic relations will be possible. For the sake of diplomatic pragmatism even undemocratic elections are accepted take for example the fact that Pres. Mubarak s National Democratic Party won the parliamentary elections in December 2010 and no government refused to accept the outcome, while about three months later the Egyptian uprising and Mubarak s demise were internationally legitimised. The phrase democratically elected has therefore very little practical normative value in determining the legitimacy of a government. Another implication of the phrase free and fair elections and its impact on the fourth possibility of unconstitutional changes (i.e refusal by an incumbent government to relinquish power to the winning party or candidate after free, fair and regular elections) 6

7 is that disputes about election results have to be considered. In most instances the election management bodies are not independent enough and there are not legitimate dispute resolution mechanisms. Examples of such instances involved Pres. Ratsiraka in Madagascar (2001), Pres. Gbagbo in Côte d Ivoire ( ) and the elections in Kenya and Zimbabwe (2008). Reliance on a narrow interpretation of procedurally free and fair elections might not be sufficient and therefore insistence on legitimate governments might solve some of the normative problems. It will avoid the dilemma that the governments in Egypt and Tunisia were elected and internationally accepted as governments but when they were unconstitutionally overthrown the international community also accepted it. Only if it can be argued that the unconstitutional change was motivated by the illegitimate nature of the government, as judged by the population themselves, can its consequences be legitimised. At the same time it should be possible to argue that if Ravalomanana continued to enjoy significant popular support after his expulsion and that his government before March 2009 was not considered as illegitimate, then the Rajoelina regime cannot be legitimised. Legitimacy depends on value judgments and not merely procedural compliance (for example, to be free and fair). It will also allow for unconstitutional actions such as just wars (Michael Walzer, 1977) in all its different manifestations. Value judgements are very difficult but were used in the cases of the Arab Spring and Eastern Europe. A typical problem was posed by the Libyan situation: on the one hand the Gaddafi regime was not a democratically elected government but on the other hand it was removed by armed rebels with only regional support, and by external intervention, which compromised its legitimacy. Whether the uprising enjoyed sufficient consensus and support was therefore in question. Arguably a regime is a legitimate target if it is isolated from the population and does not have significant popular support beyond the security forces. Legitimacy as a norm will always be in competition with Realist considerations. In North Africa it was less problematic in the cases of Tunisia and Egypt while Libya separated the AU from Europe and the USA. The same applies to Madagascar, especially between the francophone world and the rest of Africa. We shall now proceed to discuss the manner in which African institutions dealt with instances of unconstitutional changes on the continent. The situation in North Africa will be discussed separately. iii) African approach to instances of unconstitutional change 7

8 Since adoption of the Lomé Declaration the OAU and AU had to respond to changes in government in ten states and they had to determine whether they were unconstitutional. The ten states were Madagascar (2001/2 and 2009), Côte d Ivoire (2002 and 2010/11), Central African Republic (2003), São Tome and Principé (2003), Guinea-Bissau (2003 and 2009), Togo (2005), Mauritania (2005 and 2008), Niger (2009/10) and Mali (2012). Förander (2010: 47) also added the Comoros islands as a category 5 unconstitutional change. Francis Ikome reached the conclusion that the AU has not been consistent in its response and their ad hoc approach he called a policy of exceptionalism. Therefore he distinguished between good coups and bad coups. Ikome identified at least four variables that influenced the AU s responses. The first was the constellation of forces at a given time. It therefore depended on which states had an interest in the country in question as well as the power coalition patterns in the AU and in Africa in general. A second variable is the AU leadership at a given time and whether the leadership is able to unite the AU members regarding a particular situation. (It also depends on the leadership qualities and diplomatic skills of the AU president and the stance of the AU Commission chairperson in a particular matter). A third variable is the domestic reaction to a coup or unconstitutional change and finally, it is also influenced by international interests in the country at stake (Ikome 2007: 34-35). The continental response to each of the instances mentioned so far has been the focus of substantial research and publications. The most comprehensive studies are those conducted by Nkosi (2010), Engel (2010), Förander (2010) and Ikome (2007). The North African cases are treated in a different category. Before addressing the individual cases, a few general observations should be made. The first is that unconstitutional changes can repeat themselves within the same state. Madagascar, Côte d Ivoire, Guinea-Bissau and Mauritania are examples where it happened more than once in the past decade. A convincing theoretical explanation for their recurrence has to be developed. The majority of unconstitutional changes occurred in West Africa except for those in the CAR, Madagascar and the election-related conflicts in Kenya and Zimbabwe. A third observation is that a significant number of unconstitutional changes were not in the form of conventional military coups in the sense that the replacement ruler was not a career militarist: see for example in Madagascar (2009), Côte d Ivoire, Togo and Guinea-Bissau (2009). It does not mean that the military were not instrumental in the government change or refusal to accept an elected transfer of power. A fourth observation is that very often a transitional (often unity) government is established soon after the government change. Finally, in most of the presidential elections after a transition the coup leader won the election. It created the contradictory situation for the AU that it had to condemn the unconstitutional change 8

9 as a normative judgement but afterwards it had to accept the election outcome also as a normative imperative, though in practice the outcomes of the two events were not dramatically different. In the following section a brief summary is made of some of the individual cases of unconstitutional change. Madagascar (2001): Presidential elections were held on 16 December The incumbent President Didier Ratsiraka received per cent of the votes and his challenger Marc Ravalomanana 46.2 per cent. However, the latter claimed to have received 52 per cent support and therefore that he was elected as President. The AU responded with emergency meetings and diplomacy and did not recognise Ravalomanana as properly elected. His government was recognised only after legislative elections in December The AU also sent a high-level fact-finding mission to Madagascar in January 2003 and its positive report convinced the AU to recognise the government in February The AU s approach was therefore regarded as relatively successful (Ikome 2007: 35). Côte d Ivoire (2002, 2011): In September 2002 the Ivorian military made an unsuccessful attempt to oust Pres. Laurent Gbagbo. Though it did not result in an unconstitutional change it introduced another period of instability for almost ten years. It divided the country effectively in two in which the north was controlled by the rebel movement Forces Nouvelles (Ikome 2007: 36). Several peace initiatives were launched and it resulted in a unity government between Pres. Gbagbo and FN leader Guillaume Soro in April The transition plan had to culminate in a presidential election. In the election Gbagbo was opposed by his former Prime Minister Alassane Ouattara in November Ouattara won the election according to the national electoral commission but the Constitutional Council declard Gbagbo as the winner. The AU s stance was inconclusive (while tacitly favouring Gbagbo) while the regional organisation ECOWAS supported Ouattara. After intensive diplomacy the UN, AU, EU and France considered Ouattara as the elected President but Gbagbo refused to hand over the presidential powers to him therefore a category 4 unconstitutional change (or lack of change). ECOWAS and the AU suspended the country from all their activities. Military intervention by the UN and France forced Gbagbo to surrender his power to Ouattara (UNOCI, Ero 2011, ICG 2009). The AU s response in this instance was ambiguous and contradictory, was not coordinated with ECOWAS s and could not enforce its position. The UN/French intervention coincided 9

10 with the NATO action in Libya and therefore drew much of the attention to the French interventions in Africa. Central African Republic (2003): In March 2003 a popular army general removed the President in a coup. The public was largely supportive of the coup and the military capitalised on it to organise presidential and legislative elections two years later. The coup leaders won the elections and were accepted by the AU (Ikome 2007: 37; Nkosi 2010: 61). According to Ikome (2007: 38) the best explanation for the AU s relatively soft stance on the CAR was because of the public mood in its favour. It was also a relief for the people who challenged the regime for years and therefore it was a good coup. Mauritania (2005, 2008): In 2005 the armed and security forces united under the Military Committee for Justice and Democracy (CMJD) and seized power from the President. The AU Peace and Security Council condemned the coup and suspended the state from all AU activities. The public opinion was, however, in favour of the coup. The AU sent a delegation to Mauritius to negotiate the re-establishment of constitutional order. In response the CMJD presented the AU with a justification of their action and outlined a timetable for the return to constitutional rule. As a result the AU limited its criticism of the coup (Nkosi 2010: 64-66). Factors that also influence the outcome were that the USA and Western powers supported the coup leader, partly because of the recent discovery of oil in the country. Mauritania was also split in terms of loyalty to the AU/ECOWAS and the Arab League. The AU realised that a too harsh condemnation could cause the country s withdrawal from the continental body and closer association with the Arab League (Ikome 2007: 41-42). A second coup was staged in August The President was arrested by the military after he had dismissed some generals whom he suspected of being behind parliamentary opposition against him. The military established the High Council of State and the AU responded as usual to the coup. In June 2009 AU mediators facilitated the Framework Agreement for a transition. A month later presidential elections were held, while the AU lifted the suspension and sanctions already a month before the elections (Engel 2010: 5, 11; Förander 2010: 44-46; Nkosi 2010: 60-67). Niger (2009/10): The unconstitutional change in Niger consisted of two elements. In May 2009 Pres. Tandja dissolved the national Parliament, because it refused to amend the Constitution 10

11 so that he could serve a third presidential term. That was an example of category 5 unconstitutional changes that appeared in the African Charter but not in the Lomé Declaration. The second element was when in February 2010 the military detained Pres. Tandja in response to his action against the parliament and they called for the restoration of constitutional order. Niger was immediately suspended from the AU. A transitional government was established and the military announced a return to constitutional rule by February According to Engel (2010: 12) the junta showed the intention to democratise the country and therefore the AU did not condemn it as harshly as other coups. Mali (2012): The events in Mali can be linked to the consequences of the Libyan conflict in 2011 in the sense that armed Tuareg rebels returned from Libya and demanded selfdetermination in the north of Mali. Pres. Touré and his government did not respond to rebellion to the satisfaction of the military and therefore dissatisfied soldiers staged a coup against them in March The AU responded by suspending Mali as a member until effective restoration of constitutional order is achieved without delay. A joint AU- ECOWAS delegation started immediately with negotiations to restore constitutional rule but the military were supported by huge public demonstrations. In April 2012 ECOWAS and the military reached agreement on a transition: a transitional government will be established and the parliamentary Speaker will be the interim President and will oversee elections. The junta also agreed to step down in return for the lifting of sanctions (). Madagascar (2009): In March 2009 Pres. Marc Ravalomanana was forced by military officers to resign after marches were organised him by Andry Rajoelina, the former Mayor of Antananarivo. Ravalomanana transferred his authority to a military council who in turn transferred it to Rajoelina. Ravalomanana won the presidential election in 2007 with a clear majority and his party TIM won the parliamentary election also with a landslide majority. Later it became clear that external factors such as France and Pakistani businesspeople in Madagascar supported the coup. Ravalomanana had to leave the country and went to South Africa. Ravalomanana described the events as follows (verbatim interview, 15 May 2012): All the international condemned the coup. It was clear because the coup was well-prepared in three months. And I remember exactly well the situation at the time, because they called me and said: You have to move out of the office and then the military at the night of 16 th March [2009] ended my Presidency. I can tell 11

12 you it was a coup. Even Pres. Sarkozy said at the time that the coup d état in Madagascar must be condemned. SADC immediately convened an extraordinary summit in Ezulwini, Swaziland, and decided to suspend Madagascar s participation and called on Rajoelina to vacate the Presidency as a matter of urgency. The AU suspended Madagascar a few earlier, because it considered the events as constituting a coup (South Africa.info, 31 March 2009). The SADC Organ on Politics, Defence and Security Cooperation took responsibility for the Malagasy process and former Mozambican President Chissano was appointed as the mediator. Several meetings in Maputo and Addis Abeba produced the three Maputo Accords (9 August 2009), the Charte de la Transition, the Accord Politique de Maputo, the Charte des valeurs and the Acte Additionel d Addis Abeba à la charte de la Transition malgache (6 November 2009). They provided for power-sharing and a unity government but also accepted Rajoelina as the President of the Transition, which was already a deviation from the Ezulwini resolution. Back in Madagascar Rajoelina refused to implement any of the agreements. In response to the emerging stalemate a joint facilitation by Pres. Chissano and Zuma in Pretoria in April 2010 did not make any breakthrough. At the same time an alternative roadmap drafted by France was already in the pipeline which determined that Ravalomanana could not return to Madagascar until the political and security situation was conducive for it therefore its indefinite postponement. The Maputo process came to a halt and the roadmap phase emerged. The original Roadmap was signed by Rajoelina s TGV party and a number of smaller parties, all supportive of him and it became known as the Rajoelina Platform. The mouvances of the three former Presidents (Zafy, Ratsiraka and Ravalomanana) refused to support it. A SADC extraordinary summit in Sandton in June 2011 tried to break the deadlock and amended the Roadmap to allow for Ravalomanana to return unconditionally. As a consequence, the mouvances of Zafy and Ravalomanana signed the amended document in September Ratsiraka returned from France to Madagascar soon thereafter but Ravalomanana failed twice to do it. In 2012 the Mouvance Ravalomanana decided to suspend its participation in all the transitional institutions. In response SADC met in Luanda and urged Rajoelina and Ravalomanana to meet urgently to resolve the outstanding matters, while Ravalomanana s right to return unconditionally was reaffirmed (information based on numerous meetings with Ravalomanana since 2010, official documents and correspondence). 12

13 The situation in Madagascar has escalated to a level where neither the AU (or PSC) nor SADC can enforce their normative frameworks anymore. France has developed a veto right in the situation. The EU follows the leadership of South Africa in this situation, while the AU (and especially Jean Ping s francophone association) is in competition with SADC. The Mozambican triumvirate (Chissano, his deputy Simão and the SADC Executive Secretary Salamão) are also in competition with the South Africans in SADC, because since August 2011 they have sidelined Chissano as the mediator and assumed the mediation role themselves as the SADC Organ troika. The Mozambicans have been closer to the Ping position. The transitional process has reached the point where the normative framework has lost most of its impact, because no reference to the coup and its leaders are anymore made and Ravalomanana considers it as a process that rewarded the coup leaders and legitimised their regime. A normative framework or standard do not depend solely consist of explicitly formulated charters and declarations but also of practical precedents. The precedents established by the AU and regional organisations vary in nature, as the cases above indicate. All coups are unconstitutional but not all of them are illegitimate. Even more problematic are government changes that are not coups but popular uprisings such as those in North Africa but still unconstitutional. The examples of Tunisia, Egypt and Libya are not discussed here, because they are well-known to most of us. In the next section a comparison between them and the other African cases forms the basis for some conclusions. It is informed by the fact that all the cases were essentially unconstitutional changes but Tunisia and Egypt were much less problematic for the AU and the international community than the others why? iv) Normative considerations for legitimate government change The Chairperson of the AU Commission, Jean Ping, conceded the fact that the revolts in North Africa put the AU normative framework under pressure in He made the following observation (Ping 2011: 1): The popular uprisings that occurred in Tunisia and in Egypt posed serious doctrinal problems because they do not correspond to any of the cases envisaged by the 2000 Lomé Declaration on Unconstitutional Change of Government. While the AU, like other international players, did not anticipate these developments, it nonetheless reacted creatively. Indeed, the AU exhibited the necessary flexibility, basing its response not n a dogmatic interpretation of the existing texts, but rather on the need to contribute to the attainment of the overall AU objective of consolidating democracy in the continent. Notably, the African leaders welcomed the developments in Tunisia and Egypt, stressing that 13

14 they provided an opportunity for Member States to renew their commitment to the AU agenda for democracy and governance, to inject additional momentum to efforts being exerted in this regard and to implement socio-economic reforms adapted to each national situation. For a number of reasons, the democratic revolution in Libya followed a different path from those of Tunisia and Egypt. Ping implied a distinction between constitutionalism and democracy and that the AU made compromises on unconstitutional changes in order to promote democracy in general. We have seen that in the cases of both Tunisia and Egypt elections took centre-stage in the transition and it arguably also reflects the AU sentiment. The problematic implications of elections have been discussed earlier and therefore our preference for a broader perspective in terms of regime legitimacy is suggested. How can be measure or determine the legitimacy of unconstitutional changes? Three propositions are suggested here. The first is that the change should introduce a genuine transition and not merely an interim government that will prepare for new elections. Transitions involve institutional changes, political democratisation, socio-economic improvements and sometimes also constitutional changes. At least these should be stated objectives of a transition the results can only be judged much later. The Egyptian uprising was judged by the AU in these terms. The Peace and Security Council stated on 16 February 2011 its solidarity with the Egyptian people whose desire for democracy is consistent with the relevant instruments of the AU and the continent s commitment to promote democratization, good governance and respect for human rights (quoted by Sturman 2011: 3). The second proposition (or norm) is that changes in which the military is prominent or which is staged by them will undermine their legitimacy. Legitimacy will be enhanced if the military remains neutral during an uprising or if they shift their loyalty to the protestors. An alternative scenario is when the military is associated with the illegitimate regime and defeated by a rebel movement. The latter scenario introduces the controversy of Libya. The Libyan military, supportive of Gaddafi, was engaged in a conflict with the armed rebels of the National Transitional Council. The military were strengthened by mercenaries from Chad, Niger, Mali and other neighbouring countries while the rebels were supported by NATO. At the early stages of the uprising before these military dynamics emerged, the AU had a much clearer normative assessment of the situation. On 23 February 2011 the Peace and Security Council strongly condemned the indiscriminate and excessive use of force and lethal weapons against peaceful protestors, in violation of human rights and International Humanitarian Law [and] underscores that the aspirations of the people of Libya for democracy, political 14

15 reform, justice and socio-economic development are legitimate (quoted by Sturman 2011: 3-4). The third proposition (or norm) is that the uprising must be popular in nature. How many citizens participate in the revolt; how spontaneous are they? Do they serve only sectoral political interests of a particular political group? Is it only confined to a particular region within the state? In general, national consensus (excluding the security forces) about the need for such an uprising is required to regard it is legitimate. It also excludes external intervention in manipulating the domestic politics of a state. If the uprising is meant to promote or protect the interests of an external power, its legitimacy should be questioned. This proposition is illustrated by two divergent cases. When former Pres. Nasheed of the Maldives was asked whether his expulsion was in the form of a popular uprising, he responded (Time, April 16, 2012: 60): Where are the flowers? Where are the people behind them? Let me see one person coming out on the street and being happy about it. And look at the amount of people coming out in our favour, in our support. There s no popularity in the coup. Ravalomanana (interview, 15 May 2012) drew also a comparison by saying: The difference between the Arab Spring (like Libya, Tunisia) they did it on their own initiative with the uprising of the people. But in Madagascar it was so small group of the senior military 150 maximum with a small group of TGV that is the difference. In similar vein, Dersso (2011: 39) identified six principle criteria for legitimate popular revolts. Some of them overlap with these identified above. The first criterion is that a socio-political dispensation must exist that warrants change: like the absence of the rule of law, social equity and free electoral processes. The second criterion is that the legitimacy will depend on the organisation and nature of mobilisation for a revolt. It should be spontaneous and not only for specific political groups. Thirdly, it is determined by the popularity of the revolt, transcending ideological, religious, ethnic or cultural cleavages. Fourthly, the revolt should be peaceful. Fifthly, it will be determined by the degree of the military s involvement in the revolt. Lastly, the role of external actors in the revolt is also relevant. The right to self-determination must be respected and no foreign manipulation should be involved. In conclusion, in a simplified form, an illegitimate unconstitutional change of government will create a non-democratic regime, which would be the source of instability and socioeconomic degradation, and which would require third-party intervention to convince them that a transition and democratisation are necessary. A legitimate revolt, on the 15

16 other hand, is normally motivated by social and political change and therefore a transition is normally a natural consequence of the revolt. Hence, it does not necessarily require third-party facilitation, such as the transitions in Tunisia and Egypt can confirm. A normative framework with legitimacy as its point of departure, and not constitutionalism, appears to be more useful in addressing the contradictions that confronted the AU in The problem of legitimacy however, is that it is more difficult to define it in procedural or legal terms. BIBLIOGRAPHY Aljazeera 22 March 2012 (Mali mutiny topples President Toure ( AU African Charter on Democracy, Elections and Governance, 30 January 2007 ( pdf) AU-PSC Communique of the 294 th meeting of the Peace and Security Council, New York (USA), 21 September 2011 (PSC/PR/COMM (CCXCIV) Engel, U Unconstitutional changes of government New AU policies in defence of democracy. Working Paper Series no. 9, Graduate Centre Humanities and Social Sciences of the Research Academy, University of Leipzig. Dersso, SA Réflexions sur la competences et les capacities de l AAPS à faire face aux révoltes populaires, en Souarés, IK et Mesfin, B (dir). Regard critique sur les revolutions de 2011 en Afrique du Nord et leurs implications. l Institut d Etudes de Sécurité, Pretoria Ero, C Looming civil war in Ivory Coast. International Crisis Group, 10 March 2011 ( Förander, K Dealing with unconstitutional changes of government The African Union way. Master thesis, Faculty of Law, University of Lund. Huntington, SP The third wave: democratization in the late twentieth century. Norman & London: University of Oklahoma Press 16

17 Huntington, SP & Dominguez, JI Political development, in Greenstein, FI & Polsby, NW (eds). Macropolitical theory. Handbook of Political Science, vol. 3. Reading, Mass: Addison-Wesley Publishing Co. ICG What s needed to end the crisis. Africa Briefing no. 62, 2 July 2009, International Crisis Group ( Ikome, FN Good coups and bad coups: The limits of the African Union s injunction on unconstitutional changes of power in Africa. Occasional paper no. 55. Johannesburg: Institute for Global Dialogue. Maru, MT On unconstitutional changes of government: The case of the National Transitional Council of Libya. African Security Review. 21:1, Nkosi, MS Analysis of OAU/AU responses to unconstitutional changes of government in Africa. Master of Diplomatic Studies, University of Pretoria. Odinkalu, CA Concerning Kenya: the current AU position on unconstitutional changes in government. African Governance Monitoring & Advocacy Project, Open Society Institute O Donnell, G, Schmitter, PC & Whitehead, L (eds) Transitions from authoritarian rule. Baltimore & London: The Johns Hopkins University Press Ping, Jean The African Union and the Libyan crisis: Putting the records straight. Letter from the Chairperson. Issue 1, November 2011 ( %20English%20_2_1.pdf) Przeworski, A et al. (eds) Democracy and development: Political institutions and well-being in the world, Cambridge: Cambridge University Press Sturman, K Unconstitutional changes of government: The democrat s dilemma in Africa. Policy briefing 30. Governance of Africa s Resources Programme, South African Institute of International Affairs, Johannesburg. Time, April 16, 2012 ( 10 Questions ) UNOCI. United Nations Operation in Côte d Ivoire ( 17

18 Whitehead, L International aspects of democratization, in O Donnell, G, Schmitter, PC & Whitehead, L (eds) Transitions from authoritarian rule: Comparative perspectives. Baltimore & London: The Johns Hopkins University Press Personal interview: Marc Ravalomanana (President of Madagascar, ), Pretoria, 5 May