2. An experiment in regional criminal justice: The emergence of the African Criminal Court

The African Criminal Court is intended to be a special Chamber under the ‘roof’ of the African Court of Justice and Human Rights. This court has been the outcome of a merger of the African Court of Human and People's Rights and the African Court of Justice which materialized by a Protocol on 1 July 2008.Footnote ⁶ However, that protocol never received sufficient ratifications and is now superseded by the Malabo Protocol that has reinstated the old name (African Court of Human and People's Rights) and provides that three types of jurisdiction (human rights, general affairs, and international crimes) will be exercised by its separate chambers.Footnote ⁷

It is generally acknowledged that the initiative to establish a special criminal chamber within the architecture of the African Court was spurred by the discontent at the exercise of universal jurisdiction over African people, including high state officials, by Western states and the selective policy of the ICC, which is often perceived to be exclusively interested in targeting African countries.Footnote ⁸ The concrete event that prompted mounting tensions between the African Union and the ICC was the latter's decision to issue arrest warrants for Sudan's incumbent president Al Bashir. The general opinion was that the initiative was ill-timed and thwarted attempts to achieve peace and reconciliation by political means, thereby displaying callous disregard for African solutions.Footnote ⁹ In a resolution issued in July 2009, the African Union referred to the ‘unfortunate consequences that the indictment has had on the delicate peace processes underway in Sudan and the fact that it continues to undermine the ongoing efforts aimed at facilitating the early resolution of the conflict in Darfur’. The resolution enjoined African Union members not to co-operate with the Court when asked to surrender Al Bashir and requested the Security Council to defer the situation in Darfur in conformity with Article 16 of the Rome Statute, a request that was largely ignored.Footnote ¹⁰

In connection with what was perceived as the excessive exercise of universal jurisdiction against African high officials by European criminal courts, the African Union acknowledged that ‘universal jurisdiction is a principle of International Law whose purpose is to ensure that individuals who commit grave offences such as war crimes and crimes against humanity do not do so without impunity and are brought to justice’, but it pointed to an ‘abuse of the principle of universal jurisdiction by judges from some non-African States against African leaders’.Footnote ¹¹ More specific allegations were levied against European states, revealing old grievances:

The African Union's decision to vest the African Court of Justice and Human Peoples’ Rights with criminal jurisdiction cannot be viewed in isolation from such deep-rooted misgivings.Footnote ¹³ It is plainly an attempt to pre-empt both the universal jurisdiction of European states and the ICC. It is telling in this regard that the Preamble of the Malabo Protocol explicitly refers to the African Union Assembly's Decision on the Abuse of the Principle of Universal Jurisdiction.Footnote ¹⁴ In short, the African Criminal Court is intended to hold the ICC and Western states aloof and to render ‘African justice for the Africans’. Meanwhile, the situation has reached a stalemate. While the Gambia and South Africa have recently recanted their decision to dissociate themselves from the ICC, the African Union has backed the call of some of the African leaders to leave the Court en masse.Footnote ¹⁵ For all that, the African Criminal Court is still not operational. This final fact should, however, not discourage us from giving it due attention.

The subject matter jurisdiction of the African Court is extremely broad, covering both the traditional ‘core crimes’ and a number of so-called ‘transnational crimes’.Footnote ¹⁶ The elements of crimes are meticulously elaborated in subsequent provisions (Articles 28B–28M). The definitions of the core crimes are copied almost verbatim from the Rome Statute, whereas the other crimes are derived from legal and political instruments of the African Union and reflect African realities. According to Article 46H, the African Court is intended to be complementary to domestic jurisdictions. The provision is largely modeled after the parallel Article in the Rome Statute (Article 17) although there is one subtle, but possibly far-reaching, difference: in the Malabo text the qualifier ‘genuinely’ is missing. Moreover, the Malabo Protocol is completely silent about the relationship between the African Court and the ICC.Footnote ¹⁷ Another conspicuous feature of the Malabo Protocol is the introduction of corporate criminal liability (Article 46C). As indicated in the introduction, the proposal to add the crime of ‘unconstitutional change of government’ was contentious and caused a delay of the enactment of the Malabo Protocol. It is to this interesting and controversial crime that I will now turn.

3. The constituent elements of unconstitutional change of government

According to Article 28E of the Malabo Protocol, several different acts can constitute the crime of ‘unconstitutional change of government’. What they all have in common is that they threaten democratically elected governments and procedures. In other words, the entire provision serves to protect institutions and (democratic) processes of political decision-making. In this respect, the offence is not dissimilar to existing international crimes. It is a truism that international criminal justice is primarily concerned with the dismal fate of individual human beings that have suffered – and continue to do so – under flagrant human rights violations amounting to the most heinous international crimes.Footnote ¹⁸ This does not exclude, however, that communities, (political) institutions or even procedures can be the victims of international crimes. The most conspicuous example is obviously the crime of aggression. Article 8bis of the Rome Statute defines the crime of aggression as an ‘act of aggression’ – invasion or attack by armed forces, bombardment, blockade of the ports or coasts etc. – of one state against another state. In other words, the provision aims at the preservation of the integrity of a state.Footnote ¹⁹ In a similar vein, terrorism as an international crime is not only characterized by the special intent to intimidate the population, but can also have the purpose to force a government to do something or abstain from doing something or can be directed at the demise of economic, social or political institutions.Footnote ²⁰

If we try to dissect the several manifestations of the composite crime of unconstitutional change of government, we can point out at least three conspicuous features. First of all, it might be observed that the term ‘unconstitutional change’ is in a sense a misnomer, because it does not only encompass dynamic attempts to topple and replace sitting governments, but also the (illicit) perpetuation of power. After all, Article 28E refers to the ‘refusal by an incumbent government to relinquish power to the winning party or candidate after free, fair and regular elections’. The provision is thus a compound of a dynamic and a static aspect and by no means aims to preserve the political status quo. Both the dynamic and static versions give rise to separate questions. Article 28E suggests that coup d’états can be accomplished by making use of mercenaries or rebellious armed groups. Probably, these eventualities are included as separate offences, in order to prevent that the true political beneficiaries might disguise their involvement and get off scot-free by hiding behind the back of their associates.Footnote ²¹ In respect of the static form of unconstitutional ‘change’ of government, a pertinent question is whether the ‘refusal of incumbent government to relinquish power’ also includes tenure prolongations and third term agendas. Shola Omotola convincingly denounces such techniques as clever tricks to circumvent criminal responsibility.Footnote ²² Secondly, Article 28E of the Malabo Protocol refers to legislative initiatives that would enable the execution of unconstitutional changes of government. Obviously, such measures would serve to render these practices a semblance of legality. And finally, the provision repeatedly emphasizes that the victims or targets of this crime are democratically elected governments and candidates that have surfaced victorious after free, fair, and regular elections. It connotes the idea that these political factions bear no guilt in any attempt to depose them and that the provision precisely serves to protect democratic processes and fair elections. Moreover, it would appear that the provision does not apply in case of rebellion or insurrection of the population against an incumbent tyrant. I will address this issue in more detail below. While the drafters may be quite sincere when they contemplated the provision, it begs the question who will decide on the legitimacy of a popular uprising against an (oppressive) regime.Footnote ²³

4. On the relationship between unconstitutional change of government and core crimes

As mentioned earlier, the decision to include ‘unconstitutional change of government’ as a crime under the jurisdiction of the future African Criminal Court has been inspired by the specific African situation where internecine struggles for political power have preceded and triggered mass human rights violations. As the ICC has almost exclusively focused on African situations, one might expect that the connection between unconstitutional change of government and international crimes would have surfaced in the case law of (Pre-)Trial Chambers. To a certain extent, this has been the case, although indirectly and rather sparingly. Causal connections between unconstitutional change of government and subsequent core crimes have emerged in the context of inquiries by the ICC Pre-Trial Chambers into the admissibility of cases before the ICC. After all, the assessment of admissibility requires an investigation into domestic criminal proceedings – if any – in which sedition, the crushing of insurgency, and international crimes are often conflated. Such inquiries would shed light on possible inextricable connections between these crimes, potentially impeding further prosecution by the ICC. As is well known, the admissibility assessment involves a two-pronged test.Footnote ²⁴ First of all, it must be explored whether the state has developed any activity in the realm of criminal enforcement at all, with the ICC being able to proceed if the outcome is negative. Next, Article 17(1)(a) of the Rome Statute stipulates that a case is inadmissible where it is being investigated or prosecuted by a state which has jurisdiction over it, unless the state is unwilling or unable genuinely to carry out the investigation or prosecution. In the Lubanga case, Pre-Trial Chamber I held that, for the purpose of the assessment of admissibility, the person subject to the domestic proceedings had to be the same person against whom the proceedings before the Court are being conducted and that the domestic proceedings had to cover substantially the same conduct that was under scrutiny before the Court.Footnote ²⁵ This so-called ‘same person, same conduct’ test has become consistent case law of the Court. In investigating whether the national proceedings indeed related to the ‘same (f)acts’, the ICC had – at least theoretically – the opportunity to gain insight in the violence that was applied by accused to seize political power or repel contenders.

In the Situation in Kenya case, the Kenyan government challenged the admissibility of the case against Kenyatta (and others), arguing that it ‘currently investigated crimes arising out of the 2007–2008 Post-Election Violence’.Footnote ²⁶ The case was not conducive to a substantive discussion of domestic proceedings, as the government erroneously opined that it was not necessary to investigate the same persons, but that investigation of ‘persons at the same level in the hierarchy’ would suffice for the purpose of the assessment of (in)admissibility.Footnote ²⁷ In view of the lack of information that pointed to ongoing investigations against the three suspects (Kenyatta cum suis), the Chamber concluded that a situation of inactivity remained which prompted it to determine that the case was admissible.Footnote ²⁸ As the domestic criminal proceedings probably involved both charges of electoral fraud, (illegitimate) claims to political power and international crimes, the Chamber missed the opportunity to address the relationship between these crimes.

The Situation in Libya case, encompassing the cases against Saif Gaddafi and Al-Senussi, offered more interesting material for exploring the connection between political violence and core crimes. The Pre-Trial Chamber summarized the case before the Court as concerning:

After having compared the case before the Court with the case subject to domestic proceedings, the Chamber was satisfied that ‘the same conduct alleged against Mr Al-Senussi in the proceedings before the Court is subject to Libya's domestic proceedings’.Footnote ³⁰ The Pre-Trial Chamber observed that, while the discriminatory intent to target victims on political grounds – which is required for the crime of ‘persecution’ – was not a factual aspect of the domestic proceedings, it nevertheless served as an ‘aggravating factor which is taken into account in sentencing under articles 27 and 28 of the Libyan Criminal Code’.Footnote ³¹ The Chamber concluded that this demonstrated that Al Senussi was prosecuted in Libya for the same facts that sustained his indictment before the Court. The Chamber acknowledged thus that in both the international and domestic proceedings the core crimes had been inspired by the political motive to repel popular insurrections in the quest to remain in power. The Pre-Trial Chamber did not pronounce any verdict (not even implicitly) on the question whether the rebels had a right to rise against the oppressive Gaddafi regime, nor – for that matter – whether Gaddafi was authorized (or not) to crush the insurgency, for the simple reason that those issues are beyond the jurisdictional competence of the ICC. But the recognition that political persecution emerged from the incumbent regime's attempt to suppress an insurrection is strong evidence of the Court's understanding of a close connection between political turmoil and international crimes.

The most conspicuous example of domestic criminal proceedings in respect of political crimes that did not coincide with core crimes (at least in which this connection was not demonstrated) – and therefore did not impede the admissibility of a case – was provided in Situation in the Republic of Côte d'Ivoire, to wit, the admissibility challenge relating to the case against Simone Gbagbo.Footnote ³² This case concerns:

The government of Côte d'Ivoire challenged the admissibility of the case, arguing that some economic crimes and crimes against the state that were prosecuted domestically were preparatory acts to the commission of the crimes with which Mrs. Gbagbo had been charged by the ICC Prosecutor.Footnote ³⁴ Unfortunately, the criminal acts have been redacted in the ICC Pre-Trial Chamber's decision and the Appeals Chamber's judgment, so we are precluded from determining their relationship with the core crimes. In any event, the Appeals Chamber held that:

The wording of the judgment is quite interesting. The Appeals Chamber does not deny that a close connection might have existed between the crimes against the state and international crimes, but only contends that the state had failed to prove that such a relationship would actually conflate both crimes into one ‘case’ impeding the ICC from further investigation and prosecution. As in the case against Gaddafi and Al Senussi, the ICC would not be authorized to adjudicate any state crimes allegedly committed by Mrs. Gbagbo. Apart from the legal impediments, there may be political considerations why the ICC is rather reluctant to scrutinize the political context of core crimes and to that purpose engage in the assessment of (for example) unconstitutional changes of governments. After all, the ICC will eschew becoming embroiled in politics. In the context of the case against Laurent Gbagbo, the ICC Prosecutor, Ms Fatou Bensouda, stated that:

While these efforts to keep aloof from politics are commendable, the concern expressed by Ms Bensouda is perhaps a bit exaggerated. As indicated above, the jurisdictional limitations impede the ICC from judging crimes against the constitutional order. Its assessment of the political background of some core crimes is more subtle and indirect. It can either take any political events – like a rebellion or the crushing of an insurgency – into consideration, when the very nature of some core crimes (like persecution on political grounds) makes such assessment inevitable (as in the Al Senussi and Gaddafi cases). Or the ICC can, in assessing the admissibility of a case, decide that charges on purely political crimes and international crimes in domestic proceedings are so inextricably linked that they impede the ICC from pursuing a case (a decision the ICC has not yet taken). In both cases, the ICC confirms the causal relationship between the political crimes (including possibly unconstitutional change of government) and core crimes, without expressing any judgement on the former.

5. The right to rebel

There is a potential tension between criminalization of assaults on incumbent governments and the right of rebellion. In the opening sentences of his searching article, Tony Honoré captures succinctly the predicament:

For the purpose of this article, it is important to figure out whether the introduction of the crime of ‘unconstitutional change of government’ can be reconciled with the right to rebel. Such an inquiry into the normative compatibility of these notions requires a further investigation of the content of, and limitations on, the right to rebel as well as by whom it is recognized. I am, therefore, less interested in the question of whether there exists a remedy to enforce the right to rebel, because the focus is on the crime of ‘unconstitutional change of government’ and whether the introduction of that crime can be trumped by the right to rebel, as a normative counterweight.Footnote ³⁸ Whether the right to rebel can actually be enforced is less important.

Some light is shed on both the essence of the right to rebel and its limitations by a famous sentence in the Preamble of the Universal Declaration of Human Rights: ‘. . . whereas it is essential, if man is not to be compelled to have recourse, as a last resort, to rebellion against tyranny and oppression, that human rights should be protected by the rule of law’. The formulation is rather enigmatic and shrouded in ambiguity. Indeed, it could be interpreted as an exhortation to states to avoid rebellion.Footnote ³⁹ Moreover, one should be cautious not to deduce too easily a right to rebellion from these lines, in view of the explicit resistance against recognition of such a right by country delegates during the drafting process of the Universal Declaration.Footnote ⁴⁰ Nonetheless, the Preamble suggests that the international society of states allows people to rise against their oppressing rulers, as an ultimate measure. The situation must have become unbearable and there should be no other method available to escape the ordeal, as is clearly expressed in the words ‘as a last resort’.

The notion of rebellion as ‘ultimate remedy’ resonates in the grand tradition of political philosophers. Calvin's acknowledgement of the right to resist the monarch had – unlike what one might have expected – clearly political, rather than religious, connotations. If the king renounced his primary duty, to wit the protection of the liberties of the people, selected persons entrusted with power and authority within the realm would be allowed to disobey and, if necessary, depose him in order to preserve order:

Calvin predicated this right – or even duty – to disobey the unfaithful king on the premise that the relation between the ruler and citizen ‘was not a direct one, but occurred through the mediating agency of the law’.Footnote ⁴² Ultimately, the resistance therefore served to vindicate the primacy of the law.

For John Locke the right to revolt was a logical sequel of his postulating the predominance of society over politics, in which we already discern the traces of Rousseau's political discourse. The monarch only rules by the grace of the will of the community and its role was that of ‘image, phantom, or representative of the commonwealth, acted by the will of society, declared in its laws’.Footnote ⁴³ If the king strayed from the right course, blatantly abused his powers and oppressed the people, society had the right to depose him by forceful means, in order to restore the ideal state of nature. Locke suggested that the fierceness of the reaction that was visited upon him was proportionate and reciprocal to the initial violence:

Nonetheless, Locke immediately qualified his position by pointing out that rebellion would only be appropriate when ‘the inconvenience is so great that the majority feel it, and are weary of it, and find a necessity to have it amended’.Footnote ⁴⁵

It is not difficult to understand the relentless emphasis on ‘last resort’, because an unchecked right to revolt would be a ‘perpetual foundation for disorder’.Footnote ⁴⁶ Besides, for states the curtailment of the right to rebel is a question of sheer self-preservation. Domestic criminal codes contain criminal provisions, outlawing insurrection and overthrow of incumbent governments and such conduct is generally threatened with severe punishment. Article 94 of the Dutch Penal Code, for example, punishes the assault, undertaken with the intent to destroy or unlawfully change the constitutional government or the order of succession of the throne, with life imprisonment or a prison sentence of maximum of 30 years (or a fine of the 5^th category).Footnote ⁴⁷ Such regulations belong to the sovereign realm of states and international (human rights) law does not deny them the right of self-defence against the threat of annihilation. In this context, it is noteworthy that Article 4(1) of the International Covenant on Civil and Political Rights embodies the state's authority to derogate from certain human rights in emergency situations:

Commenting on this provision, Bertil Dunér observes that ‘it goes without saying that this possibility given to the state contradicts a right to rebellion: it is for the survival of the state that it was introduced’.Footnote ⁴⁸ Even those states that recognize a right to resist official power make the necessary qualifications. Article 20(4) of the German Constitution (Grundgesetz) acknowledges such a Widerstandsrecht: ‘All Germans shall have the right to resist any person seeking to abolish this constitutional order, if no other remedy is available’. The provision is essentially conservative, as it can only be invoked in order to preserve, not to change the constitutional order. Moreover, the notion that the right to resist is a measure of last resort is manifestly included in the addition that there should not be any other remedy available.

It is interesting to compare the rather modest recognition of the right of resistance, as acknowledged by political philosophers and embodied in the German Constitution and the Universal Declaration of Human Rights, with the proposed provision on unconstitutional change of government in the Malabo Protocol. In all serious pleas for a right to resist, including the German constitutional provision, the requirement that any resort to force should be an ultimate remedy is a steady concern. It seems to suggest that rebellion is only allowed if a tyranny or oppressive regime chokes all political opposition and makes life unbearable. The Malabo Protocol only prohibits assaults on democratically elected governments. The implication seems to be that in those situations a revolution would not be an appropriate ultimum remedium, because there would be other (democratic) means to oust the incumbent powerholders. In this sense, the provision in the Malabo Protocol would not be in contravention of a restricted – and generally acknowledged – right to revolt, because the former clearly does not encompass resistance against tyranny. The acknowledgement of the citizens’ right to resist attempts to abolish the constitutional order, as incorporated in the German Constitution, is especially interesting. It might arguably govern situations of powerholders clinging to their position and acting in clear violation of the constitution. This comports with the ‘static’ variety of unconstitutional change of government, as envisaged in Article 28E of the Malabo Protocol. By making such practices a crime, the Malabo Protocol rather seems to sustain (a limited use of) the right to rebel than to counter it. We might therefore come to the intermediate conclusion that the criminal provision on unconstitutional change of government does not infringe the right to rebel.

6. The crime of unconstitutional change of government in the Malabo Protocol and the principle of non-intervention

The essential question in this article is to explore whether there are good reasons for African states to elevate the repression of unconstitutional change of government to a supra-national (regional) level. In that context, it is necessary to inquire whether any criminalization of this conduct does not contravene essential rights and prerogatives – as has been done in the previous paragraph. However, that does not suffice. While the previous section has indeed demonstrated that the offence of unconstitutional change of government does not violate the right to rebel and that states are entitled to restrain and, if necessary, suppress insurrections, it is by no means clear why this offence should come under the jurisdiction of a regional court. The proper yardstick for assessing whether this crime qualifies for ‘promotion’ is, I submit, whether criminalization surpasses merely parochial interests of the state and epitomizes the idea that it has become the business of a wider community of states.Footnote ⁴⁹

In the realm of international criminal law, the universal interest in the repression of international crimes can be gleaned from the fact that they qualify as violations of customary international law.Footnote ⁵⁰ Could the argument be made that, by analogy, a regional organization like the African Union can legitimately claim jurisdiction over a crime the prohibition whereof has solidified in (regional) customary law? Such a line of reasoning is (implicitly) defended by Ademola Abass who, after expressing concerns that not ‘all crimes [within the jurisdiction of the African Court] are, in fact, “international” and “serious” enough to warrant international prosecution’, contends that the crime of unconstitutional change of government would certainly meet those criteria:

The idea that regional customary law could serve as an appropriate legal basis for the selection of crimes that qualify for subject matter jurisdiction of a regional criminal court is interesting.Footnote ⁵² But a thorough research into whether unconstitutional change of government indeed meets this standard is beyond the scope of this article. Moreover, I would hold that the category of crimes under the jurisdiction of the African Criminal Court is not exhausted by those offences that demonstrably belong to the realm of regional customary law. All of the offences mentioned in Article 28A of the Malabo Protocol are subject of (regional) treaties in which states commit themselves to criminalize the conduct, render mutual assistance in criminal matters and pledge to either prosecute or extradite (aut dedere, aut judicare) those suspected of those offences that are found on their territory.Footnote ⁵³ The very fact that African states conclude treaties with a view to the common criminal law enforcement in respect of certain offences is proof that they share an interest in their suppression. Any decision to outsource such law enforcement to a regional court is a logical next step that is facilitated by the prior enactment of such treaties. After all, states are not prohibited by international law to establish a regional criminal court and equip this court with jurisdiction over crimes of common concern, provided that such offences are criminalized under their domestic law and the regional court does not apply the law retroactively.

Yet, intuitively, the crime of unconstitutional change of government seems to be of a different nature than the other crimes featuring in Article 28A of the Malabo Protocol. Civil unrest that can turn into rebellion or insurrection is a typical internal affair. Article 2, section 7 of the UN Charter precludes the United Nations from intervening in affairs that essentially belong to the domestic jurisdiction of a state and regional instruments confirm that states are under a duty to abstain from such interference.Footnote ⁵⁴ The argument could be made that a transfer of criminal jurisdiction in respect of internal political crimes to a regional court would amount to such an intervention which is prohibited under international law. On closer scrutiny, however, this assumption is far-fetched and even incorrect. It is generally acknowledged in international law that states are allowed to seek the assistance of other states in order to suppress rebellion or insurrection. Conversely, states are not entitled to support insurgents.Footnote ⁵⁵ It should be recalled in this context that in the famous Nicaragua v. US case the International Court of Justice found that the US assistance to the contras, such as financial support, training and supply of weapons constituted ‘a clear breach of the principle of non-intervention’.Footnote ⁵⁶ The construction obviously serves the preservation of world order and favours established governments, to the detriment of rebels.Footnote ⁵⁷ The asymmetric interpretation of the principle of non-intervention raises the question whether this principle, rather than precluding criminalization of unconstitutional change of government, not actually supports such an initiative. After all, it could be argued that the assistance of states to the hard-pressed government might consist of bringing the rebels to criminal justice before a regional court. However, this position is also untenable for two reasons. For one thing, the free consent of the state is a prerequisite for the legitimacy of the foreign assistance.Footnote ⁵⁸ Besides, absent a treaty, no state is under an obligation to render assistance to the established government during insurgency.Footnote ⁵⁹ Secondly, as soon as the insurgency has ‘matured’ into belligerency, a foreign state is under an obligation to remain neutral and forfeits its right to support the incumbent government.Footnote ⁶⁰ Obviously, states would be allowed to take sides and become a co-belligerent with one of the two factions. For the purpose of our research, this entails that foreign states would not be allowed to bring belligerents to criminal justice before a regional court on account of their revolt, as soon as they have succeeded in achieving that status.

It may be interesting to pay some attention to the relationship between the principle of non-intervention and the static form of unconstitutional change of government. Would states be allowed to intervene when defeated dictators cling to their position and refuse to abdicate in favour of democratically elected power contenders? The issue is topical in view of the recent intervention by troops from other West-African countries in the Gambia in order to remove outgoing president Yahya Jammeh who was adamant in his decision to stay in power.Footnote ⁶¹ In his discussion on whether such an intervention for the restoration of democracy is permitted, Malcolm Shaw is quite determined where he holds that ‘apart from the problems of defining democracy, such a proposition is not acceptable in international law in view of the clear provisions in the UN Charter’.Footnote ⁶² One might deduce from this, per argumentum a contrario, that less invasive measures which imply no use of (military) force, like regional criminal law enforcement, would be allowed. However, even if that position can be vindicated, other obstacles may abound.

At this point, it is necessary to briefly discuss the topic of immunities. Article 46Abis of the Malabo Protocol stipulates that:

The provision exhibits an obvious retrogression from Article 27 of the Rome Statute that has abolished all immunities before the ICC. It is not entirely clear whether Article 46Abis only refers to personal immunities (ratione personae) or would also cover functional immunities (ratione materiae), but the explicit mention of ‘serving Heads of State’ and the addition ‘during their office’ suggest that the former option was intended.Footnote ⁶³ The problem of immunities in the context of international criminal law has been widely debated, in particular in relation to the arrest warrants issued against President Al-Bashir.Footnote ⁶⁴ Moreover, the immunity issue has a far wider purport and may affect prosecution and trial of all international and transnational crimes under the jurisdiction of the African Criminal Court. For these reasons, an extensive discussion of the topic is beyond the scope of this article. Nonetheless, it cannot be denied that immunities have special repercussions for the crime of unconstitutional change of government. For one thing, those who succeed in toppling a democratically elected government may invoke immunity as soon as they have come to power. Secondly, those who refuse to make way for victorious and freely elected contenders appear to be shielded against prosecution as long as they remain in office. Immunities serve as a double-edged sword that benefits vested interests and perpetuates unlawful exercise of power. To be sure, there are ways to circumvent such predicaments, by withholding recognition of governments that have seized power by illegitimate means or by stipulating that tenure ends whenever an incumbent head of state refuses to abdicate.Footnote ⁶⁵ Still, it bears emphasis that immunities constitute an impressive bulwark that may impede a successful repression of unconstitutional change of government.

Returning to the principle of non-intervention, it appears that this principle neither prohibits, nor sustains the criminalization of unconstitutional change of government. The principle is entirely subservient to the sovereign will of states, which stands to logic as it serves to protect their interests. States are at liberty to seek assistance of other states against insurgents and such assistance may take the form of criminal law enforcement by an African Criminal Court, but they are not bound by such an arrangement if they are not party to the Malabo Protocol. The freedom that international law bestows on states to seek assistance is even wider than acknowledged under the Malabo Protocol because such prerogatives are not dependent on the democratic quality of the government in power. The only limitation that the principle of non-intervention entails is that belligerents are not to be held criminally accountable because that would imply a breach of neutrality. Furthermore, the interpretation of the principle is understandably skewed in favour of preservation of the status quo. However, any obstacles in the prosecution of incumbent powerholders will probably be caused by inconsistencies in the law on immunities, rather than by the principle of non-intervention itself. Nonetheless, both the principle of non-intervention itself and its exceptions and limitations are primarily inspired by the quest for stability in international relations. And this rationale is of primary importance for the understanding of the introduction of the crime of unconstitutional change of government and its elevation to a regional level in Africa.

In my view, the best and probably only reason why states have a common interest in the repression of unconstitutional change of government is that insurgency is contagious. While it seemingly is restricted to the territory of one state, it cannot easily be contained and has the nasty habit of spilling over to neighbouring countries. The analogy with the crime of aggression is highly appropriate here. There is no consensus among political philosophers and legal commentators that aggression is the most hideous of crimes. Larry May, for instance, contends that the crime of aggression is not clearly the worst of crimes, because state aggression may in some cases be the lesser evil, especially when it creates, on balance, positive effects.Footnote ⁶⁶ Those, however, who consider aggression the worst of crimes often refer to Von Clausewitz’ famous discourse On War in which he argues that war is essentially boundless. According to Von Clausewitz, ‘war is an act of force which theoretically can have no limits’.Footnote ⁶⁷ And he goes on to explain the mechanics of escalation that are inherent to warfare. The logic of war is that ‘each of the adversaries forces the hand of the other. War tends toward the utmost exertion of forces’ which implies increased ruthlessness, since ‘the ruthless user of force who shrinks from no amount of bloodshed must gain an advantage if his opponent does not do the same’.Footnote ⁶⁸ Because the stakes of glory versus defeat are so high, the antagonists hold each other hostage and run into a downward whirl of death and destruction.Footnote ⁶⁹ This vicious circle of ever mounting violence is not restricted to the initial parties. As the adversaries try everything possible to gain the upper hand, they seek the assistance of allies. And so, war expands in space and time, involving more and more participants. It is precisely due to this process of escalation that war does not stop at geographical boundaries. It has prompted states to outlaw war as a method of dispute settlement, because the experience had taught them that it can redound on themselves.Footnote ⁷⁰

In case of civil strife, one would expect that its violent effects would be less volatile, but the experience in Africa in particular has proved otherwise. Examples of insurrections crossing borders and affecting several countries abound. The carnage in the Congo at the turn of the century which has often been qualified as an ‘African world war’ originated from the civil wars in Rwanda and Burundi. Other players soon tuned in, always in pursuit of raw materials, sometimes in order to root out rebels who found refuge in the vast jungle (Museveni's Uganda) and sometimes in retaliation for earlier support of insurgents by Congo (Angola).Footnote ⁷¹ The Lord's Resistance Army, after having partially been defeated by Ugandan forces in the north of the country, moved its camps to neighbouring Sudan and the Central African Republic, creating havoc in those states.Footnote ⁷² The string of events in the ‘Arab Spring’ exhibiting a domino effect in several countries was explicitly mentioned in the warrant of arrest issued against Al Senussi.Footnote ⁷³ And the infamous civil war, fuelled by the hunt for diamonds in Sierra Leone, was initially triggered by incessant tribal warfare in neighbouring Liberia that did not end with the demise of the dictatorship of Samuel Doe, but gave new opportunities to arch-schemer Charles Taylor.Footnote ⁷⁴ Against this backdrop one can understand the urge of African states to join forces and engage in criminal repression of the unconstitutional change of government.

7. Some final reflections

It has been my objective to investigate and understand why a typical domestic political event like unconstitutional change of government has been upgraded in the African context to a ‘supranational crime’ that is supposed to be countered by law enforcement at a regional level. I have argued that criminalization of the conduct as such does not infringe a supposed right of rebellion. Nor would the elevation of the offence to the regional level contradict the principle of non-intervention. However, while these principles do not defeat the prosecution and trial of unconstitutional change of government by the African Criminal Court, it is still not clear why states would be inclined to move criminal law enforcement in respect of this crime to a higher level. One reason that comes to mind is that serious political strife is connected to core crimes in the sense that they generate those crimes. The analogy with the crime of aggression is apposite, in that aggression not only promotes war crimes but also constitutes the sine qua non for their occurrence. The Nuremberg Military Tribunal succinctly expressed this: ‘To initiate a war of aggression, therefore, is not only an international crime; it is the supreme international crime differing only from other war crimes in that it contains within itself the accumulated evil of the whole.’Footnote ⁷⁵ Like inter-state war, insurgencies trigger violence and augment the risk of heinous crimes being committed. Nonetheless, the explanation is not entirely satisfactory, because penalization of unconstitutional change of government does not aim to counter political violence at all costs. Its objective is to suppress assaults on democratic institutions, suggesting that such illegitimate activities in particular are likely to affect peace and stability within a political community. A second – and arguably better – explanation is that insurgencies are not contained to single states but are inclined to spread to other countries. Again, the analogy with the crime of aggression is striking. The Nuremberg Tribunal noted that ‘[w]ar is essentially an evil thing. Its consequences are not confined to the belligerent States alone, but affect the whole world’.Footnote ⁷⁶ These lines lay bare the infectious nature of war. With respect to internal rebellions, this is not essentially different. The African continent in particular has been plagued by internal strife that easily moves from state to state. States have a common interest in suppressing both the dynamic and static form of unconstitutional change of government because the ensuing unrest and violence is not likely to stop at their borders.

The inclusion of this crime in the Malabo Protocol suggests that states are confident that the African Criminal Court may be better equipped to take on the investigation and prosecution of perpetrators of the offence. After all, the relationship between this court and domestic jurisdictions is governed by the principle of complementarity which implies that the Court can intervene when the state, due to internal political tensions, is unable to pursue criminal proceedings. Of course, we have no guarantee that the incorporation of the crime within the jurisdiction of the African Court will reduce or prevent its occurrence. However, the same holds true for the crime of aggression in the Rome Statute. It may be considered, though, as a normative expression of a genuine common concern of African states.