A ruler’s shield? Re-evaluating the norm against unconstitutional change of government in Africa

 Chidi Anselm Odinkalu
 BL (Hons) (Nigerian Law School) LLB (Hons) (Imo State) LLM (Lagos) PhD (London-LSE)
 Professor of Practice of International Human Rights Law, Fletcher School of Law and Diplomacy, Tufts University, Medford MA, United States of America; Chairperson, Governing Council, Chukwuemeka Odumegwu-Ojukwu University, Anambra State, Nigeria
 https://orcid.org/0009-0008-2408-4946
 This email address is being protected from spambots. You need JavaScript enabled to view it.

 The author expresses appreciation to the anonymous reviewers as well as to Chepkorir Sambu of the Faculty of Law, Kabarak University, Nakuru, Kenya, whose kind and thoughtful comments assisted in bringing greater clarity to the article.

 Edition: AHRLJ Volume 25 No 1 2025
 Pages: 1 - 36
 Citation: CA Odinkalu ‘A ruler’s shield? Re-evaluating the norm against unconstitutional change of government in Africa’ (2025) 25 African Human Rights Law Journal (2025) 25 African Human Rights Law Journal 1-36
 http://dx.doi.org/10.17159/1996-2096/2025/v25n1a1
 Download article in PDF

Summary

The broad contours of Africa’s inter-state system were configured in the murderous overthrow of the continent’s governance systems in the aftermath of the Berlin (West) Africa Conference in 1885. A strain of violent authoritarianism that did not care for bases in popular legitimacy established thereafter around the continent was resilient even into the age of post-colony. Following independence, African states retreated behind a carapace of domestic jurisdiction and it was understood that how government acquired its mandate to rule was not the concern of other states. Constitutional instability ensued, mostly characterised by an epidemic of coups and conflict. By the last decade of the twentieth century, this understanding had begun to erode, and at the turn of the millennium, the continent had instituted a new normative order establishing democratic elections as the basis of the mandate to rule, and prohibiting coups and unconstitutional changes in government (UCG). However, the meaning of the UCG was largely left to be determined and, in turn, the content and application of the norm prohibiting it came to be defined by an appearance of inconsistency.

Over time, a norm designed to defeat authoritarianism appeared instead to have become appropriated or instrumentalised for the exact opposite, triggering a self-inflicted crisis. This article examines the scope and meaning of the norm against UCGs in Africa in light of its evolution and history. It shows that far from being a shield for interminable rulers, the norm embodies three propositions in support respectively of democratic elections, against interminable rule, and for the protection of legitimate mandates. It argues that popular uprisings are insurrectionary reactions to the failure of Africa’s continental and regional institutions to respect this package, and calls for African states and institutions to return the norm to its original design.

Key words: coups; norm; unconstitutional change of government; popular uprising; Africa

1 Introduction

The subject matter of unconstitutional change of government (UCG) has risen in international priority and global attention in recent years. At the turn of the millennium, African Heads of State and Government in the Organisation of African Union (OAU) expressed ‘grave concern about the resurgence of coups d’état in Africa’,1 while also recognising that coups ‘are a threat to peace and security of the continent and they constitute a very disturbing trend and serious setback to the on-going process of democratisation in the continent’.2 Nearly one decade later, the Assembly of Heads of State and Government of the African Union (AU) complained of a ‘resurgence’ of UCGs on the continent.3 In January 2022 the Peace and Security Council of the AU once again expressed ‘deep concern over the resurgence of military coups d’état’, concerned that these ‘undermined democracy, peace, security and stability in the continent’.4 Drawing upon the public health metaphor of a contagion in October 2021, the Secretary-General of the United Nations (UN), Antonio Guterres, complained of ‘an epidemic of coup d’états’ [sic], calling on the UN Security Council to take urgent steps to arrest what he identified as a global trend.5 More recently, then judge-president of the Court of Justice of the Economic Community of West African States (ECOWAS), Edward Amoako Asante, has equally described UCGs in the states of the Community as a ‘grave danger to West Africa’s march towards the consolidation of democratic governance in the region’.6

This recent rise in UCGs in Africa appears impervious to more than three decades of intense normative innovation on governance within the institutional mechanisms of regional governance in Africa, including the AU, its predecessor, the OAU, as well as the continent’s regional economic communities, especially ECOWAS.7 The idea of a resurgence in UCGs in the midst of this norm inflation is itself reason for a re-evaluation of both the efficacy of the norm and of the gap between norm inflation and reality.

Powell and Thyne argue that any computation of UCG incidents would be dependent upon how the concept is defined or operationalised.8 West Africa arguably is at the epicentre of what the UN and the AU have respectively called a ‘global epidemic’ or a regional ‘resurgence’ in UCGs.9 The AU ‘has declared a total of fourteen (14) instances of UCG since 9 July 2002 when it was established: three in North Africa, seven in West Africa, one in East Africa, one in Southern Africa and two in Central Africa’.10 In West Africa, this recent raft of supposed UCGs has included three in Mali,11 two in Guinea,12 as well as one popular uprising,13 and two coups in Burkina Faso.14 McGowan, for instance, reports 169 ‘military interventions of some type’ in 16 West African countries (including Mauritania) in the half century from 1 January 1955 to 31 December 2004, with Cape Verde being the only country in the region to not have reported a significant coup incident.15 This list excludes the more problematic category of attempted coups,16 the most recent claims in the sub-region of which were reported in Guinea-Bissau,17 The Gambia18 and in Sierra Leone.19

This study grapples with this problem by exploring the normative scope of the UCG through an examination of the institutional practice of Africa’s regional institutions. As a convenience of usage, it employs the word ‘coup’ as illustrative shorthand for the UCG. In reality, the two are not necessarily coterminous. A coup is a form of a UCG but, as will be evident later, it does not necessarily exhaust the concept.

The persistence of the UCG in Africa, despite the appearance of an established norm against it, calls for a reappraisal of the norm that seeks its prohibition. The AU Peace and Security Council had in fact suggested the need for such a reappraisal in 2021.20 Inspired by that, this study examines the evolution and continuing relevance of the norm against UCGs in Africa. It argues that this norm emerged in response to a particular history of interminable political rulership and denial of effective rights of participation around the continent and that it is a composite norm that embodies three propositions, not one, including norms concerning credible elections and legitimate access to political power; on executive term limits; and against unconstitutional changes to legitimate governments. An investigation of its evolution, it is argued, will show that far from a shield for those in power, which it appears to have become, the norm against UCG evolved from a concern for the protection of the right to participation in government in the African inter-governmental and human rights system.

In effect, the norm against UCG evolved as part of an implicit bargain that created guardrails for political contestation on the understanding that those who seek political office would abide by the rules governing access to power and not convert incumbency into interminable rulership. However, the scope of the norm appears to be at odds with the narrative around it. This has fostered perceptions of dissonance between intimations of a rule, on the one hand, and folklore around it, on the other, turning the norm into a tool for the protection of what Omorogbe has called ‘a club of incumbents’.21 In particular, it would appear that those dimensions of the norm concerning both term limits and credible elections have suffered retrenchment in favour of turning the norm against UCGs, arguably into a shield for interminable rule.22 Nantulya observes that one consequence of this tendency has been to turn elections in Africa into ‘perfunctory formalities’ with mostly predetermined outcomes in many places.23 In reaction to this trend, Afrobarometer’s most recent surveys have found a co-incidence of statistically significant fall of confidence in elections, rising levels of openness to military rule (as the only effective tool to remove indispensable rulers) and even higher levels of support for presidential term limits.24

This situation calls for an urgent re-examination of the normative package as well as a strengthening of its regional implementation.25 The norm on UCGs itself does not make sense outside the context that necessitated its evolution. The argument in this study will seek an equal balance between narrative reconstruction of the norm and an exploration of its conceptual scope. Accordingly, it begins with an exploration of the scope and meaning of the UCG and the norm against it. As part of this, it also examines the subject of popular uprisings, illustrating an important conceptual distinction between unconstitutional and extra-constitutional changes in government. Thereafter, it considers the context that necessitated a continental response to UCGs in Africa. Before concluding, it highlights some of the challenges associated with the implementation and application of the regional norm against UCGs in Africa.

This study is informed by a concern that a narrow focus on the supposed norm against UCGs in Africa without equal attention to the norms against executive tenure indeterminacy and on credible elections, creates an increased risk of what has variously been called ‘the totalitarian type of democracy’26 or, at best, of ‘competitive authoritarianism’,27 both of which endanger government founded on popular legitimacy and pose adverse consequences for development in Africa. It is also inspired by a realisation of the normative and practical significance of participation in government as a legal and civic entitlement that anchors the legitimacy of political power, whose frustration, as the AU has acknowledged,28 is conducive to a contagion of constitutional instability which the continent cannot afford.

2 An elusive concept: The unconstitutional change of government

Although supposedly at the heart of a path-breaking norm of continental significance, Africa’s institutions left the meaning of the UCG undefined, ostensibly allowing it to evolve from the practice of African states. The UCG in Africa was for a long time mostly associated with the capture of political power by soldiers outside the bounds of constitutional channels. This kind of event was classically called a coup d’état, and Welch famously said of it that it was a ‘clear event, easy to date and (if successful) possible to document’.29 Powell and Clyne dismiss any such hope of clarity in the definition or understanding of coups and UCGs as ‘too optimistic’.30 This elusiveness of the concept of the UCG may explain the challenge that African institutions have had in operationalising an effective continental norm concerning it.

The African Union High Level Panel on Egypt acknowledged this in 2014 when it complained about ‘difficulties encountered in applying the AU norms on unconstitutional changes of government’.31 A year earlier, on the fiftieth anniversary of the establishment of the OAU in 2013, the AU had adopted a ‘Solemn Declaration’, which reaffirmed its commitment against UCGs, but also ‘recognise[d] the right of our people to peacefully express their will against oppressive systems’.32 This affirmation left it unclear what cognitive or evidentiary thresholds were required to be crossed in order for a system to be recognised as oppressive for this purpose; how such peaceful expression of the will of the people against an oppressive system was to be treated, or what ‘peaceful’ indeed meant in this context. It is, therefore, essential to unbundle the elements of the UCG as a concept.

2.1 Conceptual elements of the unconstitutional change of government

Two dimensions of the norm concerning UCG specifically are worth distinguishing here. One is how to access legitimate power, and the other is how to end an incumbent’s term in office. In chronological terms, the latter precedes the former by creating the vacancy to be filled with access to power. A coup d’état brings down a government with a view to replacing it, but it is not always that the deposition of a government is procured by people who set out with a desire or plan to usurp it. The AU has established the popular will through elections as the preferred basis for access to or retention of power,33 but seems ambivalent on the question of whether elections are the sole means for bringing the legitimate exercise of political power to an end. In this connection, the AU’s own Africa Governance Report 2024 complains about ‘situations of “popular uprisings” that lead to UCGs’, implying that there may be popular uprisings resulting in the overthrow of governments which may not, however, be described as UCGs.34 This leads to some uncertainty as to the scope of the prohibition against UCG in the African inter-state system.

Three constitutive elements of the norm concerning UCGs are discernible. The first dimension concerns the legitimate assumption of power. In the AU, it is now clear that the preferred and primary mechanism of acquiring legitimate power is through credible elections. The Preamble to the African Charter on Democracy, Elections and Governance (African Democracy Charter) declares the explicit goal of entrenching ‘in the continent a political culture of change of power based on the holding of regular, free, fair and transparent elections’.35 The Charter also contains extensive provisions concerning the standards of elections on the continent and extending to the observation of elections by the AU.36 Separately, the Constitutive Act of the African Union prioritises, among other things, ‘respect for democratic principles, human rights, the rule of law and good governance’,37 and condemns and rejects unconstitutional changes of governments.38 In particular, article 30 of the Constitutive Act makes it clear that ‘governments which shall come to power through unconstitutional means shall not be allowed to participate in the activities of the Union’.39 These commitments have been described as both ‘quite far reaching and long overdue’,40 crystallising decades of normative evolution in Africa concerning norms for access to legitimate power.

Second, as will be shown below, there is within the composite of the norm against UCGs a prohibition against executive tenure indeterminacy. Pointing to the intensity and frequency of the removal of term limits around the continent for illustration, Omorogbe acknowledges that ‘one key driver of coups is the difficulty of democratic change in Africa’.41 Siegle and Cook record a correlation between the lifting of term limits and the return of UCGs in Africa, noting that in nearly two decades, from 2002 to 2020, incumbents in 18 African states lifted or altered term limits 24 times.42 In approximately the same period, the continent recorded an escalation in UCGs, with 21 successful coups.43 Underscoring the normative implications of this correlation, the AU has clarified that ‘constitutions shall not be manipulated in order to hold on to power against the will of the people’.44 The AU Peace and Security Council implicitly acknowledged this as a problem when it requested the AU Commission, following a military coup in Guinea in 2020, to investigate the ‘root causes and impact of the unconstitutional change of government arising from non-consensual and/or politically manipulated democratic processes’.45

The idea that a politically manipulated process can nevertheless be democratic as implied in this formulation lends credence to the suggestion that the AU remains uncertain as to the exact scope and meaning of UCG. It is even less clear when what is manipulated is not the constitution as such but institutional processes thereunder, such as the judicial process for the certification of eligibility of an incumbent for an election. In EACSOF v Attorney-General of Burundi the Appellate Chamber of the East African Court of Justice considered the case initiated against the tenure elongation by Burundi’s President, Pierre Nkurunziza. On 5 May 2015 Burundi’s Constitutional Court had dismissed an application by some senators seeking a declaration that a third term by President Nkurunziza was incompatible with both Burundi’s Constitution and the Arusha Peace Accord on which it was based and a violation, therefore, of the rule of law. The First Instance Chamber of the East African Court of Appeal dismissed the case on the jurisdictional ground that the Court did not sit on appeal against decisions of national courts. On appeal, the Appellate Chamber affirmed jurisdiction and remitted the case to the First Instance Chamber for determination on its merits.46 In December 2019 the First Instance Chamber again dismissed the case on the merits, a decision which triggered another appeal. While the case was pending on appeal, President Nkurunziza died in June 2020. In its decision in November 2021, the Appellate Chamber determined that the remedies sought had become moot in light of the death of President Nkurunziza and the succession that occurred shortly thereafter. Crucially, however, it held that the decision by the Constitutional Court offering him a third term in 2015 violated the East African Community Treaty.47 This decision suggests that the question of tenure extension could increasingly become a matter for review by the judicial organs of regional institutions in Africa.

When the mandate to rule has been acquired legitimately through elections and the person holding it substantially respects applicable guardrails against tenure indeterminacy, normative and logical coherence dictates that the third element of the norm against UCG in Africa would prohibit the unconstitutional removal of that person during the pendency of their tenure. This third element builds on the first two elements and is contingent on both. The practice of the AU and African states, however, has mostly valorised this third element but without acknowledging that it is contingent on the first two. In other words, the AU and African states and institutions have used the norm against UCGs in Africa to protect incumbency with little care for how power is retained. Yet, it is the case that a norm against UCGs would be unnecessary and self-defeating if it existed to protect power that itself was procured by force of arms or in favour of elections that are perennially manipulated to keep a despot in power. Such arrangements would fall into the class of what the Peace and Security Council of the AU has described as ‘non-consensual and/or politically manipulated democratic processes’.48 Where there is a failure to firmly police the first two elements of this norm, therefore, the third element breaks down. This also implies that power originally legitimately acquired could itself subsequently cease to be so exercised if the foundation in the contingent legitimacy of elections is broken.49

2.2 Non-interference and the beginnings of a norm against unconstitutional change of government in Africa

The emergence of the UCG as the dominant method for changing or instituting government in the immediate aftermath of decolonisation in Africa took place in the shadows of a doctrinal disposition in regionalism in Africa that valorised self-regarding sovereignty over democratic pluralism. This history deserves attention. What is now called UCG has manifested itself in Africa mostly in the form of the coup d’état and the continent has experienced a lot of it. At the beginning of 2022, it was said that ‘of 486 attempted or successful coups carried out around the world since 1950, Africa has seen 214, the most of any region, with 106 of them successful’.50 This means that at least 108 were unsuccessful. The continent accounted for 44,03 per cent of the global epidemiology of coups, including 43,8 per cent of the successful ones and 44,26 per cent of the unsuccessful ones. While Sudan accounts for the highest number of attempts (both successful and unsuccessful), with 18, Burkina Faso, with eight successful coups out of nine attempts, accounts for the highest success rate.51 Of the top ten countries in Africa in the league table of coups, only three – Comoros with nine (four successful); Burundi with 11 (five successful); and Sudan with 17 (six successful) – are not in West Africa. Of the remainder, Ghana and Sierra Leone have had ten coup attempts each, resulting in five changes of government in each; Benin, Guinea-Bissau, Mali and Nigeria have each had eight coup attempts with success rates of six, four, five and six respectively. Togo has reported three successful coups from seven attempts.52

The above trend is evidence for the proposition that in the immediate aftermath of the age of independence in Africa in the decade of the 1960s, a contagion of constitutional instability appears to have beset the continent with both domestic and regional dimensions.53 The result was that the egalitarian promise of independence and liberation around the continent was quickly followed by what has been described as a ‘proliferation of authoritarian regimes in newly independent African countries’,54 which institutionalised in many countries the phenomenon of interminable or life presidencies.55 Regionally, the countries of post-colonial Africa quickly retreated behind a curtain of domestic jurisdiction, which was crystallised in the prohibition in the Charter of the Organisation of African Unity against ‘interference in the internal affairs of states’.56 For much of the first three decades of post-colonial Africa, two complementary principles defined the continent’s attitude to coups. First, the question of how a government came to power was regarded as essentially within the domestic jurisdiction of African states and not a matter for external or continental oversight.57 Second, coups or UCGs were said to be ‘the most frequently attempted method of changing government’ and, for the most part, the most successful.58

When 15 member states of the OAU created ECOWAS in 1975, they effectively set up a pioneering African model of regional integration as an exception to the rule in the OAU Charter on sovereignty and non-interference.59 The only question became how deep or extensive the scope of this exception would be to the habits of somewhat anarchical sovereignty then existing within the OAU. It is ironic that the founding treaty of ECOWAS was the first breach in the armour of the OAU’s absolutism on domestic jurisdiction, ultimately inspiring the developments that led to the emergence of the prohibition against UCG in Africa, because seven of the 15 original signatories to the treaty were military rulers.60 The second chink in this armour was the adoption in 1981 of the African Charter on Human and Peoples’ Rights (African Charter) finalised in Banjul, The Gambia.61

The onset of the Liberian civil war, almost contemporaneously with the collapse of the Berlin Wall in 1989, eventually triggered an overdue re-evaluation of the doctrinal foundations of Africa’s inter-state system. At its summit in Addis-Ababa, Ethiopia, in July 1990, the Assembly of Heads of State and Government of the OAU adopted the Declaration of the Assembly of Heads of State and Government of the Organisation of African Unity on the Political and Socio-Economic Situation in Africa and the Fundamental Changes Taking Place in the World, in which they declared themselves to be ‘fully aware that in order to facilitate this process of socio-economic transformation and integration, it is necessary to promote popular participation of our peoples in the processes of government and development’,62 committing themselves to both democratisation and ‘to the consolidation of democratic institutions in our countries’.63

Very importantly, the heads also affirmed that ‘democracy and development should go together and should be mutually reinforcing’.64 This represented a major shift in the doctrines of the OAU and its member states. In the same year, ECOWAS established a plurinational Ceasefire Monitoring Group (ECOMOG) to intervene in Liberia, marking the first time a regional institution in Africa would undertake intervention in this way.65 One year later, the Assembly of Heads of State and Government of ECOWAS followed suite with the adoption of the Declaration of Political Principles, in which they undertook, among other things, to ‘respect and promote human rights and fundamental freedoms in all their plentitude including in particular freedom of thought, conscience, association, religion or belief for all our peoples without distinction as to race, sex, language or creed’,66 and to ‘strive to encourage and promote in each our countries, political pluralism and those representative institutions and guarantees for personal safety and freedom under the law that are our common heritage’.67

Contemporaneously, in 1991, Uganda’s President Yoweri Museveni, who was then the Chairperson of the Assembly of Heads of State and Government of the OAU, together with Nigeria’s then former military head of state, Olusegun Obasanjo, co-convened a high-level Conference on Security, Stability, Development and Co-Operation in Africa (CSSDCA), resulting in the Kampala Document, which declared that ‘[t]here should be periodic renewal of the mandate of political leaders. At the same time, the tenure of elected leaders in various branches of government should be constitutionally limited to a given number of years.’68

A mere four years following the Kampala Document of the CSSDCA, up to ‘37 African states had revised their constitutions and all but four of them inserted presidential term limits’.69 This narrative reinforces the conclusion that the norm against UCG in Africa evolved as part of a composite continental deal to address instability traceable to authoritarian rule and its manifestation in the phenomenon of the interminable or life presidency. Any fruitful examination of pathologies of governance and development in Africa must take account of the two dimensions to the issue: UCGs, on the one hand, and the phenomenon of the interminable presidency, on the other.

Siegle and Cook point out that a ‘lack of effective term limits has resulted in Africa having 10 leaders who have ruled for over 20 years and two family dynasties that have been in power for more than 50 years’.70 The sense of correlation between authoritarian or interminable rule, on the one hand, and the phenomenon of the UCG, on the other, appears inescapable. The implication of this is that a focus on UCGs may be fruitless without an effort to effectively address interminable rule. They are interdependent, and it has in fact been argued that the manipulation of national constitutions to extend presidential tenure is itself a UCG.71 Against this background, Manirakiza distinguishes between military and ‘constitutional’ coups, with the latter referring to the manipulation of constitutions for the purpose of extending presidential tenure.72

2.3 Emergence of a prohibition on unconstitutional change of government as continental norm

At the beginning of the 1990s, therefore, it appeared that a broad continental consensus was emerging with three elements. The first was the recognition of an organic nexus between democracy and development. The second was a recognition that government had to be founded on popular will grounded broadly in the effective exercise of the right to democratic participation. The third element of this consensus was an affirmation of a commitment to presidential term limits. If they held firm, these elements together could disincentivise the tendency towards coups around the continent and potentially outlaw them. However, there was still some way to go before this outcome could be realised or guaranteed.

By 1995, the OAU had evolved the outlines of a norm against UCG but failed to assign any meaning thereto. These norms were first deployed by the OAU in response to the 1997 coup d’état in the West African state of Sierra Leone.73 In its Declaration on the Framework for an OAU Response to Unconstitutional Changes of Government, the member states of the OAU

agreed on the following definition of situations that could be considered as situations of unconstitutional change of government:

1. military coup d’état against a democratically elected government;
2. intervention by mercenaries to replace a democratically elected government;
3. replacement of democratically elected governments by armed dissident groups and rebel movements;
4. the refusal by an incumbent government to relinquish power to the winning party after free, fair and regular elections.74

This definition was a partial adoption of a much wider-ranging recommendation by a Sub-Committee of the Central Organ (of the OAU Conflict Resolution Mechanism), which was constituted in 1995 and reported in 2000. The typologies of unconstitutional change of government recommended by the Sub-Committee, included, in addition to the four adopted eventually by the Assembly:75

1. the refusal by a government to call for general elections at the end of its term of office;
2. any manipulation of the Constitution aimed at preventing a democratic change of government;
3. any form of election rigging and electoral malpractice, duly established by the OAU or ascertained by an independent and credible body established for that purpose;
4. systematic and persistent violation of the common values and principles of democratic governance referred to above; and
5. any other form of unconstitutional change as may be defined by the OAU policy.

The omission of these from the initial definition by the OAU implied, first, that the concept of the UCG was contested and, second, that the organisation perhaps agreed to the lowest common denominator among its members. On the margins of the Lomé Summit of the OAU at the turn of the millennium, a consensus position was issued in the form of the Lomé Solemn Declaration on the CSSDCA, based on the principles in the Kampala Document from nearly one decade earlier in 1991, wherein the African Heads of State merely undertook to ‘adhere to the provisions of the law and other legislative enactment promulgated by National Assemblies’ and to ‘respect their national constitutions and adhere to the provisions of the law and other legislative enactment promulgated by National Assemblies’.76 This was an implicit climb-down from the high ambitions of the Kampala Document and a clear claw-back on the elaborations thereafter.

2.4 The African Commission on Human and Peoples’ Rights and the evolution of the norm against unconstitutional changes in government in Africa

The African Commission on Human and Peoples’ Rights (African Commission) played a defining role in underscoring the urgency of a norm against UCGs in Africa, clarifying essential aspects of its scope and calling attention to its relationship to the right to participation in government in article 13 of the African Charter. In particular, the African Commission established that the UCG had ceased to be a matter essentially within the domestic jurisdiction of affected states and had become a matter of regional and continental salience. Three cases decided by the Commission deserve attention in this respect. The assumption of jurisdiction over these cases by the Commission was an implicit indication that the age of tolerance of coups by continental institutions was at an end.

After one decade of military rule, the soldiers in Nigeria on
12 June 1993 organised a presidential election to return the country to civil rule. Ten days after the vote, with the results about to be declared, the military regime casually issued an announcement nullifying the ballot. They also precluded any legal challenge to that decision domestically. A group of Nigerian advocacy organisations approached the African Commission, claiming that the nullification of the election violated the African Charter. The African Commission upheld their claims and held that the nullification by Nigeria’s military violated the rights to participation and self-determination respectively in articles 13 and 20(1) of the African Charter.77 This established an important principle about upholding the sanctity of elections.

Four years after the developments in Nigeria, the military in 1997 overthrew the elected civilian government of Burundi. To bring pressure on the usurper regime, the neighbouring states of East and Central Africa decided to impose a regional blockade on Burundi, a land-locked country. The Association pour la Sauvegarde de la Paix au Burundi, a non-governmental organisation, initiated proceedings ultimately to the African Commission against the blockading neighbours, claiming various violations of the African Charter. If they succeeded, it would have reinforced the previous doctrine that regarded coups as matters essentially within the domestic jurisdiction of affected countries. Departing from this tendency, however, the African Commission affirmed the need for regional action against UCGs in Africa and upheld the regional blockade implemented by Burundi’s neighbours against the usurper regime as ‘legitimate interventions in international law’.78 The Commission explained that ‘the military coup which deposed the democratically elected government constituted a threat to, indeed a breach of the peace in Burundi and the region’.79 In effect, this decision explicitly confirmed that the UCG was no longer a matter essentially within the domestic jurisdiction of the affected state and had become a legitimate matter of regional attention.

Following his ouster from power by soldiers in 1994, The Gambia’s former President, Dauda Jawara, initiated proceedings before the African Commission questioning the powers of the military government to take certain decisions including the exclusion of members of his government from rights of participation in government. In deciding on this complaint, the Commission for the first time addressed the question of the compatibility of UCGs with the African Charter, declaring that the military coup in The Gambia in 1994 was ‘a grave violation of the right of Gambian people to freely choose their government as entrenched in article 20(1) of the Charter’.80 In principle, the overthrow of Dauda Jawara by the military was a coup but it was difficult to see how the people of The Gambia could lawfully bring an end to his tenure. Dauda Jawara’s presidency began in April 1970. At the time of his ouster in July 1994, he had been in power for over two decades. In determining this communication in the way it did, the African Commission focused its condemnation against military rule as a violation of the right to participation in government, leaving aside questions as to how power was retained or for how long.

2.5 A millennial norm

By the turn of the millennium, therefore, a combination of diplomatic and jurisprudential developments had crystallised the need for a norm against UCG in the OAU as well as rationales for it, but the full scope of this norm was far from entirely clear. Over time, different elements of this norm were discernible addressing three related dimensions, namely, whether the question of the mandate to govern essentially is within the domestic jurisdiction of the state affected; how to establish legitimate government; and how to preclude government from overstaying its legitimacy.

With reference to the first, the African Commission established in Association pour la Sauvegarde de la Paix au Burundi that the UCG was a fit subject of regional oversight and action. Concerning the second, the African Commission was able to establish in Jawara that the continent would no longer tolerate the phenomenon of arbitrary termination and replacement of legitimate government by soldiers, mercenaries or armed dissidents. The Nigerian cases laid the foundations for addressing the sanctity of elections and the obligation of incumbents to respect the will of the people expressed through them. Concerning the third, however, there appears to be a failure of clear direction from the continental institutions. These institutions subsequently established that a government could no longer be allowed to stay in power after losing elections.81 The African Democracy Charter promulgates these into a definition of UCGs, together with a prohibition against ‘any amendment or revision of the constitution or legal instruments, which is an infringement on the principles of democratic change of government’.82 The same Charter also seeks to prohibit the perpetrators of UCG from participating in elections for the restoration of democratic form of government and criminalises the act of perpetrating UCG.83

In 2014 the AU adopted the Protocol on Amendments to the Protocol on the Statute of the African Court of Justice and Human Rights, which formally creates the crime of UCG in Africa to mean ‘committing or ordering to be committed’ acts with the aim of illegally accessing or maintaining power through:84

1. a putsch or coup d’état against a democratically elected government;
2. an intervention by mercenaries to replace a democratically elected government;
3. any replacement of a democratically elected government by the use of armed dissidents or rebels through political assassination;
4. any refusal by an incumbent government to relinquish power to the winning party or a candidate after free, fair and regular elections;
5. any amendment or revision of the constitution or legal instruments, which is an infringement of the principles of democratic change of government or is inconsistent with the constitution;
6. any substantial modification of the electoral laws in the last six (6) months before the elections without the consent of the majority of the political actors.85

There are legitimate questions as to whether these provisions provide sufficient clarity to define the elements of a crime, but an answer to those questions must await a time when the court has the opportunity to determine those issues. For now, however, three points may be made. First, it seems quite clear that an institutional consensus was crystallised around the need for and existence of a norm prohibiting UCGs in Africa. Second, former judge president of the ECOWAS Court of Justice, Edward Amoako, calls attention to the multiple dimensions of the norm concerning UCGs in Africa ‘such as zero tolerance for unconstitutional change of government, abolition of tenure elongation beyond the two terms of office, the conduct of free, fair and credible elections, as well as the scrupulous adherence to the separation of powers’.86 Third, that said, the scope of this norm remains far from fully evolved. This last point is demonstrated by the uncertainties as to whether the concept of ‘popular uprising’ is or is not part of the norm against UCGs.

3 ‘Popular uprising’: An extra-constitutional phenomenon

The uprisings in the countries of North Africa, now widely described as the Arab Spring, presented the AU with an opportunity to address an issue that was not before foreseen in the annals of the organisation.87 Since 2011, the continent has witnessed at least seven situations where widespread protests eventually led to the ouster of a government.88 At the AU Summit in July 2011, then Chairperson of the African Union Commission, Jean Ping, characterised these events as ‘popular uprisings’,89 although some others persisted in describing the events in Tunisia leading to the sack of the regime of President Zine Abdine Ben Ali in 2011 as a coup d’état.90 The Africa Governance Report 2024 notes ‘a trend towards popular uprisings in Africa’,91 suggesting that this implies ‘profound state or nation-building challenges in the countries concerned’.92

The concept of popular uprising has never quite been coherently defined or operationalised in the practice of the AU or any of the continent’s regional economic communities (RECs). In August 2019, the Peace and Security Council of the AU cautioned that ‘the concept is not provided [for] in any of the existing AU normative frameworks, in this regard, it is an invention that needs to be interrogated and reflected upon before it is embraced by the Union’,93 describing it as ‘complex, contested and controversial’.94 At least two issues arise with reference to the subject of popular uprisings. One is the substantive scope or elements of the concept. The second is whether this concept is itself embodied within the meaning of UCG or whether it is beyond the scope of the UCG.

With reference to the former, the High-Level Panel on Egypt in 2014 proposed some elements as guidelines in framing the concept of popular uprising. These include:95

1. the descent of the government into total authoritarianism to the point of forfeiting its legitimacy;
2. the absence or total ineffectiveness of constitutional processes for effecting change of government;
3. popularity of the uprisings in the sense of attracting significant portion of the population and involving people from all walks of life and ideological persuasions;
4. the absence of involvement of the military in removing the government; and
5. peacefulness of the popular protests.

These criteria do not lend themselves to easy application or evaluation. For example, the distinction implied in this between total and partial authoritarianism is at best unsafe and could be dangerous to the extent that it could suggest comfort with undefined forms of authoritarianism. There may also be questions as to who assesses or determines the ‘absence or total ineffectiveness’ of lawful mechanisms for change of government or acquisition or renewal of the mandate to rule. The descent of a member state into that kind of situation arguably is impossible to envisage without an implicit admission of gross failure or complicity of the AU’s own mechanisms for oversight of elections under the African Democracy Charter. It may even be impossible to determine what constitutes military involvement in an uprising. Would forbearance of the military from taking action against widespread disorder against an elected government constitute involvement? Additionally, some of the requirements in the above criteria are somewhat fanciful. It is difficult, for instance, to see how a protest that is ‘peaceful’ can force out a government described as ‘total authoritarianism’. The Africa Governance Report 2024 appears to engage with the subject in this spirit, proposing four different scenarios:96

1. a popular civilian uprising resulting in the resignation of an incumbent government;
2. the military stepping into a power vacuum created by the resignation of an incumbent government;
3. the military takes over a civilian mass movement’s demand for resignation of an incumbent government, leading to the resignation of an elected leader;
4. a popular uprising leads to an armed conflict or a civil war.

In terms of the report, the first scenario will not be treated as a UCG. The second scenario would be addressed by issuing the military with a deadline to return to constitutional normalcy. The third would be a UCG, while the fourth would be treated as an armed conflict.97 According to the report, therefore, a popular uprising could be characterised as

the collective actions of a variety of non-collective actors from a broad section of society, which embody shared concerns of large numbers of ordinary people, whose fragmented but concerted activities trigger socio-political change, irrespective of whether the concerns are guided by an ideology or recognisable leadership. The definition will also include ‘popular military coup’ which refers to a situation in which large numbers of ordinary people publicly support a UCG undertaken by the military.98

This definition does not necessarily bring much clarity to the concept besides making it clear that a ‘popular uprising’ is the result of an irretrievable breakdown in the legitimacy and authority of government. Rather, this definition confuses the concept. The essential proposal in the report is that a popular uprising may or may not be classed as a UCG depending on the facts of each case. This suggestion of a fact-specific determination, which appears to be the favoured practice of regional institutions at the moment, achieves an outcome that appears on its face appealing but ends up in fact conflating regime termination with regime replacement. Even more importantly, this also avoids a fundamental conceptual distinction between ‘unconstitutional’ change in government and ‘extra-constitutional’ end to government. The former relates to a means of taking power while the latter addresses the means of losing it. Chronologically, these are two separate occurrences and the termination of a regime can be envisaged without being accompanied by a plan or even capacity to replace it.

A coup d’état engineers the termination of power in order to usurp those who exercise it, but there is no reason why both need to be engineered by the same people or to occur contemporaneously. They are thus severable. The entire point of a popular uprising in fact is to terminate power, but not necessarily for those who bring this about to themselves capture it or replace those whom they remove from office. The AU definition of the popular uprising concedes this point through its recognition of the possibility that the actors in a popular uprising may be ‘fragmented’ and may, in fact, lack a shared ideology or coherent or recognisable leadership. Such a collective may be able to bring down a government but is unlikely to be configured to replace it.

A popular uprising does not necessarily establish a government; it only terminates one. If and when that happens, the likelihood of a power vacuum is high, as is the possibility of an outbreak of a contest among diverse actors to replace the ousted regime. By contrast, a coup d’état does not just stop at terminating a government; it engineers regime change in order to deliberately install another. A popular uprising that is faithful to the characteristics outlined in the AU’s attempted definition would not be a UCG because it is complete upon the downfall of the regime against which it takes place. The character of any government that results thereafter and the manner of diplomatic or other response that should follow in its wake is liable to be calibrated to the facts and evidence available.

The attempt at a definition by the AU at the moment appears, therefore, to conflate a popular uprising with the resulting regime. While convenient, this is flawed. As a strict matter of chronology and definition, a popular uprising cannot on its own be a UCG because the replacement of a government in the aftermath of a popular uprising could be a necessity of circumstance but not a design of an uprising. An event or undertaking that sets out to replace a legitimately elected government before the expiration of its lawful tenure would not qualify as a popular uprising. Whether what follows in the wake of an uprising can justifiably be called a UCG may be open to debate. In the absence of a constitutive intent to remove the preceding regime, it is difficult to see how a successor in such circumstances can with justification be called a UCG. The one thing that may be said with certainty, however, is that a popular uprising is itself evidence of the failure of both national governments and regional institutions in Africa to hold states accountable to the standards of good government set up in regional treaties and other relevant laws.

4 Governance, participation and development: Three dimensions of the unconstitutional change of government

The subject matter of UCGs in Africa has a rich and complex history connected with the crisis of state legitimacy on the continent, which predates decolonisation. It is also a matter of high salience for the continent’s development and, not unsurprisingly, a priority for Africa’s inter-state system and institutions. The history of the African continent in the aftermath of decolonisation was dominated by UCGs, sometimes accompanied by political assassinations.99 Mazrui explains that African states at independence confronted the two major challenges of civic integration and achieving political legitimacy. Integration, on the one hand, was a problem of hewing together diverse peoples into mutual coexistence and shared recognition of common citizenship. The challenge of legitimacy, on the other hand, concerned the acceptance by the peoples and citizens of the new countries of the authority to rule of the post-colonial political leadership.100 Neither integration nor political legitimacy could be taken for granted at the formal end of colonialism and both challenges were conducive to a rise in what has been called ‘post-formation instability’.101

This context of post-colonial instability had its origins in an underlying crisis of illegitimacy that afflicted the post-colonial African state as well as the regimes established thereby.102 UCGs in Africa do not merely pre-date decolonisation, they created the original sin of illegitimacies at the root of Africa’s crisis of statehood. The invasion of Ethiopia and the overthrow of the government of Emperor Haile Selassie by Benito Mussolini in 1935 was an early example of a UCG on the continent.103 Going further back in time, the same could be said of the Berlin Africa Conference whose adoption of the General Act of Berlin in January 1885 destroyed the pre-colonial governance systems in Africa, sundering the continent into territories under European occupation.104 It thus is arguable that to the extent that its political geography is ultimately traceable to the foundations laid in the General Act of the Berlin Conference, the African inter-state system was the product of a UCG with its origins in Europe.105

To appreciate the need for the norm against UCGs in Africa, it is essential to highlight some beacons in the continent’s history of the phenomenon. On 23 July 1952, army officers known as the Free Officers Movement, led by Muhammad Naguib, an army general, and Gamal Abdel Nasser, a colonel, overthrew the monarchy of King Farouk I, in what would become known as the Egyptian Revolution.106 Six years later, on 17 November 1958, the government in neighbouring Sudan collapsed, a mere 34 months after independence, handing the reins of power to Ibrahim Abboud, a general who at the time was the army chief of staff.107 These two early examples presaged the emergence of a pattern that would largely define access to power in Africa over the next four decades. These developments were to trigger a cascade with far-reaching normative, functional and instrumental consequences for the continent.

4.1 Unconstitutional change of government and the right to participation in government

It is clear from the account in the earlier parts herein that there has been a normative shift in the acceptability of coups d’état or UCGs, globally, and in Africa, particularly. In their Declaration on the Framework for an OAU Response to Unconstitutional Changes in Government adopted at the Lomé Summit in 2000, the Heads of State and Government of the OAU claimed that ‘the phenomenon of coup d’état has resulted in flagrant violations of the basic principles of our continental Organisation and of the United Nations’,108 and that ‘coups are sad and unacceptable developments in our continent, coming at a time when our people have committed themselves to respect of the rule of law based on peoples’ will expressed through the ballot and not the bullet’.109 This was a significant normative evolution from the orthodoxy of the OAU member states. Indeed, for much of its life, the position of the OAU was the exact opposite. West Africa exemplified this trend. Of the 15 Heads of State and Government present at the adoption of the ECOWAS Treaty in May 1975, seven were military rulers and another six were succeeded by soldiers.110

In his foreword to the 1978 revision of Edward Luttwak’s widely acclaimed Coup d’état: A practical handbook, Walter Laqueur declared as a matter of fact that ‘the coup d’état is now the normal mode of political change in most member states of the United Nations’.111 For long, the regime change imperative inherent in the coup d’état was considered by some countries an instrument of foreign policy.112 In the immediate aftermath of decolonisation following the end of World War II, some imperial powers switched from direct colonialism to the UCG as means of retaining power and influence in foreign affairs, especially affecting their former colonies or spheres of influence.113

Following decolonisation around Africa, military law and rule emerged as an accepted norm, legitimised by jurisprudence and legal doctrine across the continent, and taken for granted as an essential rampart of public and constitutional law.114 Outstanding judicial and legal careers were even built on it.115 However, by the turn of the millennium in Africa, Roth felt confident to assert the emergence of a norm to the effect that ‘governments violate international law where they do not predicate their rule on popular will’.116 Contemporaneously in Africa, Cheeseman felt able to affirm a similar normative shift that military rule by men (they were all men) in fatigues had become by and large discredited, such that ‘in the 2000s, elections and term limits replaced death and coup d’état as the most common ways in which African presidents and prime ministers left office’.117 The implication of this was that the manner by which government acquired its mandate to rule could no longer be regarded as exclusively within the domestic jurisdiction of African states.118 It also meant that the right to participation had become established as the foundational anchor for the mandate of government to rule in the African system. Put differently, the legitimacy of the claim of government to political power in Africa had become both a human rights issue and a subject for continental or regional oversight.

4.2 Unconstitutional change of government, despotism and instability

Functionally, the UCG is anchored on a philosophy that splinters political power, on the one hand, from ownership by or legitimacy in the will of the people, on the other. In doing so, it licenses despotism by definition. It was perhaps the pioneering anarchist thinker, William Godwin, who put the functional consequences of constitutional instability into perspective, even if inadvertently, when he juxtaposed despotism with anarchy and came away choosing the latter over the former. He reasoned that ‘[a]narchy is a horrible calamity, but it is no less horrible than despotism. Where an anarchy has slain its hundreds, despotism has sacrificed millions upon millions, with this only effect: to perpetuate the ignorance, the vices, and the misery of mankind. Anarchy is a short-lived mischief, while despotism is all but immortal.’119

The logic of regional or continental action against the UCG, therefore, is not difficult to see or justify, and the African Commission did in fact provide the legal justification for that in Association pour la Sauvegarde de la Paix au Burundi.120 It is not illogical that the instability inherent in UCGs should be a major determinant of development and its outcomes. Acemoglu and Robinson illustrate this point with the example of Mexico:121

Between 1824 and 1867, there were fifty-two presidents in Mexico, few of whom assumed power according to any constitutionally sanctioned procedure. The consequence of this unprecedented political instability for economic institutions and incentives should be obvious. Such instability led to highly insecure property rights. It also led to a severe weakening of the Mexican state, which now had little authority and little ability to raise taxes or provide public services.

4.3 Unconstitutional change of government and development

The instrumental effect of the constitutional instability of the UCG in (West) Africa on the continent’s development has been no different. The concern with UCGs in the region, therefore, acknowledges the ‘blight that political instability has visited on the continent generally and on its individual citizens and inhabitants in particular’.122 Hopkins similarly underscores the relationship between constitutional and governance instability, impoverishment and misery around Africa, describing it as ‘a moral reproach on a scale even greater than that represented by the slave trade in the 19th century’.123 He summarised these consequences in graphic numbers on a continental scale:124

Much of Africa has failed to achieve any growth in per capita incomes since 1960; some countries have seen incomes decline. At the close of the twentieth century, the average life expectancy of a child born in sub-Saharan Africa in 1980 was only 48 years; a typical African mother had only a 30 per cent chance of seeing all her children survive to the age of five; daily calorie intake was only 70 per cent of that of Latin America and East Asia.

This underscores the organic relationship between UCG and development in Africa. Development remains one of the major preoccupations of the international settlement that evolved at the end of World War II, and decolonisation in Africa was a central reason for this.125 Conceptually, however, it is also both complex and slippery. Opinions diverge as to what its most pressing priorities should be from a selection that includes a focus on advancing economic progress, enhancing the frontiers of human agency, growing the institutional capacities of the state, the quality of leadership, or a synthesis of all these factors.126 One basic agreement in this debate is around the centrality of the state in the project of development. This makes state building essential in development.127

The idea of sustainable development is reasonably well established in international law.128 It was first fully articulated in the report of the World Commission on Environment and Development, better known as the Brundtland Commission Report.129 Inter-generational equity and effective participation of citizens are both inherent in this conceptualisation of sustainable development. These, in turn, imply a government founded on popular legitimacy, which alone is the kind of government liable to be held to account by the people for in the event of failure to address or meet the needs of the people. This cannot be the case where government derives its power from anything other than the will of the people democratically expressed.

It is no accident, therefore, that the UN’s 2030 Agenda for
Sustainable Development proclaims that ‘good governance and the rule of law as well as an enabling environment at national and international levels, are essential for sustainable development, including sustained and inclusive economic growth, social development, environmental protection and the eradication of poverty and hunger’.130 At the national level, this requires a government that is founded on and able to guarantee ‘just and inclusive societies that provide equal access to justice and that are based on respect for human rights (including the right to development), on effective rule of law and good governance at all levels and on transparent, effective and accountable institutions’.131 As the African Commission in Jawara v The Gambia, these are not guarantees that come naturally from a government which comes to power by way of an UCG.132

Mirroring these issues, the African Union’s Africa Governance Report 2024 identifies five factors that predispose countries to UCGs. These include integrity of elections; diversity management and human rights; constitutional order and state legitimacy; economic governance; public sector accountability; popular uprising; militarization; and terrorism.133 It concludes that when elections are not considered credible or if the will of the people is subverted by the incumbent administration, instability results.134 A government founded on UCG denies the rule of law, popular legitimacy and participation, and in this way precludes a state that is stable and development that is sustainable. Participation is itself a human right guaranteed by the African Charter,135 which is violated when a government is established by UCG.136

5 Conclusion

The emergence of a norm against UCGs in Africa is evidence of the exponential change that has occurred in the landscape of regional norms in Africa since the independence of African states in the 1960s. From behind the veil of the doctrine of sovereignty and non-interference, this norm emerged to outlaw the UCG as a basis for access to political power, reaching into the very core of domestic jurisdiction. It has an affirmative element in the proposition that elections are the normative ideal for accession to power. It also embodies a prohibition against the overthrow of an elected government. In cases where elections prove to be chronically ineffective or manipulated, popular uprisings can terminate governments that have spent their legitimacy and authority. When that happens, how that government is replaced can also fall within the purview of the norm.

In practice, the implementation of the norm has not been without challenges. First, coordination between regional institutions has sometimes proved problematic. For example, the AU complains about ‘a strong suggestion of an overlap of mandates between the AU and ECOWAS in responding to military coups’,137 expressing the hope that ‘the two institutions must thus coordinate their efforts’.138 There are a similar overlaps and contradictions between regional institutions in West Africa. For instance, in connection with the measures imposed following successive coups in Mali in 2020 and 2021, the Court of Justice of the Economic and Monetary Union of West Africa (UEMOA) lifted the sanctions imposed by the ECOWAS.139 With eight of the 15 member states of ECOWAS also party to and bound by the orders of the UEMOA Court of Justice, the regional sanctions regime against Mali effectively collapsed.

A second issue is the effect of the sanctions levied by regional institutions on states that experience UCGs. Again, with reference to Burkina Faso, Mali and Guinea, the ECOWAS Court of Justice on 30 September 2021, acting under Rule 78 of its Rules of Procedure, suspended cases from countries under sanctions from the Community as a consequence of respective cases of UCG. Some groups criticised this at the time as an act of ‘obstruction of justice for victims of human rights violations’.140 This suspension lasted for nearly 14 months until 22 November 2022, when the Court lifted it.141 This was odd at best because the ECOWAS Supplementary Protocol on Good Governance makes it clear, among other things, that while states are under sanctions for UCG, ‘ECOWAS shall continue to monitor, encourage and support the efforts being made by the suspended member state to return to normalcy and constitutional order’.142 The effect must be that the ECOWAS Court should retain jurisdiction over the country for the purpose of supervising return to Community law and standards, including standards of compliance with human rights, in particular, during the period of the suspension. The Court cannot afford to yield up jurisdiction over violations alleged to have taken place prior to the suspension, including, in fact, over the legality of the UCG.

Third, an appearance of inconsistency in the implementation of the norm against UCG around the continent could also feed regional contagion. The AU and ECOWAS, for instance, sanctioned as UCGs military takeovers in Burkina Faso, Guinea and Mali, but failed to do so in Chad where the military clearly bypassed constitutional mechanisms after the killing of the incumbent President in April 2021, claiming that Chad was responding to a security imperative imposed by attack from foreign mercenaries.143 This exemption was not deemed available to Burkina Faso which the AU suspended in January 2022 after a military coup,144 despite the fact that the country had been under a prolonged assault from foreign Jihadists.145

Fourth, there are considerable uncertainties as to the meaning of the UCG itself arising from or connected with the idea of popular uprising. The popular uprising in many cases is an extraordinary last resort that ensues when all lawful options for removing an ineffective, unpopular or illegitimate government have failed. This calls attention to the need for a prophylactic approach to the phenomenon of UCGs in the form of consistent application and implementation of regional norms governing elections and good governance. It is often because the implementation of these norms is uneven and patchy that UCGs and popular uprisings ensue. This subject may be suitable for the exercise of the advisory jurisdiction of regional courts and tribunals in Africa, including the African Court on Human and Peoples’ Rights.

These inconsistencies have both substantive and spatial dimensions. As to the latter, the overlaps between the spatial scope of regional institutions complicates any effort to determine the precise geographical reach of the applicable norms. The AU, ECOWAS and the UEMOA are separate institutions with distinct geographical reach but all member states of the UEMOA are in ECOWAS; and all member states of ECOWAS are in the AU. This spatial incoherence, in turn, has substantive implications, suggesting that the norm against UCG in Africa is either still inchoate or in the process of fuller crystallisation.

Fifth, in the face of this appearance of inconsistent application of the norm, it is no accident that there are complaints that the efficacy of the prohibition against UCGs in Africa ‘seems to be waning’.146 In the nature of the original formulation and evolution of the norm, the prohibition against the UCG as a coup against an incumbent was contingent on both legitimate elections and the absence of tenure indeterminacy. In its practice, the AU casualised the latter two and only emphasised the protection of tenure with no acknowledgment of its contingent character. This fission of the norm was both convenient and unprincipled and had the effect of placing it at the feet of incumbents to be instrumentalised. By turning the norm into a shield for the protection of the ‘club of incumbents’ in the AU rather than an affirmation of the foundation of the mandate to govern in popular legitimacy and the right to participation, the regional institutions created a crisis of sustainability for the norm. In effect, the current crisis afflicting the norm against UGCs in Africa, therefore, is entirely self-inflicted at the instance of the self-same regional institutions that instituted it. One piece of good news inherent in this is that they also have the capacity to bring the crisis to an end.

The concept of popular uprising itself proves the limits of the norm on UCGs as well as the flaws in the implementation of regional norms on good governance. If these norms were applied or implemented consistently and effectively, countries would be able to lawfully change governments that have spent their contingent legitimacy or which prove to no longer be fit for purpose. However, when governments can perform badly, afflict their people with manifestly malign outcomes, or manipulate elections chronically in order to retain power without consequences, they make popular uprisings inevitable. Collier points out, additionally, that this compromises development because ‘crooks will replace the honest as candidates’.147

This argues for clarity in the norm against UCGs and consistency in its implementation. Neither of these can presently be said about the norm. When regional institutions consistently certify crooked elections or ballots that are manifestly lacking in credibility, they lose standing as credible arbiters on questions of UCGs. This calls for greater clarity concerning the applicable standards to govern the credibility of elections. The African Court on Human and Peoples’ Rights offered an example of what is possible in this direction with its 2021 advisory opinion on elections and COVID-19.148 Regional courts of justice can complement this role.149

UCGs in many ways are both cause and consequence of the unravelling of governance and development on the continent. It is clear that the UCG undoubtedly causes a setback to the pursuit of development. The UCG could also be a consequence or evidence of the failure of regional compliance with standards of good governance. Returning to the language of public health with which this study began in the description of UCGs as a regional ‘epidemic’, it seems clear that an effective approach to addressing this problem requires both prophylactic and therapeutic dimensions. By way of prophylaxes, regional supervision of elections must be returned to a place of credibility, which is not the case presently. That would make it possible to assert and implement the sanctions against UCGs much more firmly. Absent a credible prophylactic programme, it will be difficult for the region to fully preclude the opportunism of UCGs or the desperation of popular uprisings. In acknowledgment of this fact, the Peace and Security Council of the AU has called for an investigation into the resilience of the pathology of UCGs in Africa. This study has sought to offer a diagnosis to aid this process.

Just as important as prevention, if not more important, is the need for effective accountability as a deterrent to would-be UCGs. It is surprising that nearly one decade after its adoption, no African state has ratified the Malabo Protocol, which makes the UCG an international crime in Africa. This fact calls into question the commitment of African states to rooting out the phenomenon of UCGs on the continent. This sense of equivocation may itself be evidence of a lack of commitment to the kinds of governance habits that the norm against UCGs requires. The entry into force of the Malabo Protocol will ultimately provide the best metric of the preparedness of the continent to match its declamations with requisite tools of accountability.