Despite the words of the Preamble (paragraph 6), the Rome Statute never imposed a legal obligation on States to investigate and prosecute international crimes perpetrated in the State’s territory, though it proclaimed the newly constituted Court as a ‘court of complementarity to national criminal jurisdictions.’ The principle of complementarity instituted by the ICC reiterated the notion that there is no shield to the prosecution of international crimes and, if States are unable or unwilling to prosecute, then the ICC can take over the task. However, the ICC is not a Court of unlimited resources, both human and financial, and this reality can have an impact on the number of situations it can effectively investigate. Moreover, the State’s duty to cooperate, as set out in Part 9 of the Rome Statute, is the greatest stumbling block to its effectiveness, as prosecutions cannot take place if the State is unable or unwilling to arrest and surrender the relevant individuals to the ICC. On this basis, it is contended that the international criminal justice vision, and its corresponding maxim to end impunity for international crimes, should be mostly about enabling the State to become accountable. This was indeed the principle underlying the concept of ‘positive complementarity’, which effectively encapsulates the catalysing effect of the ICC on national criminal mechanisms. The ultimate objective is about instilling the aspiration, at a domestic level, to pursue the rule of law and ending impunity for international crimes. But it is acknowledged that in order to achieve this objective a change of domestic policies, practices, laws and ideology may take some time, and for some States this transformative process may indeed take longer. And this is why the ICC fills a very important gap when States are unable or unwilling.

Having said that, the international community, together with the appropriate ICC organs and officials, must work towards a more realistic model of positive complementarity, especially in States where the concepts of international crimes and impunity have not yet reached a level of political maturity that provokes a domestic transformative process. A typical example of positive complementarity in action can be seen in the case of Colombia, where the ICC effectively assisted the country in its effort to investigate and prosecute international crimes perpetrated in that territory over the past few decades. But the question is why the same approach could not be reciprocated with Gabon (or indeed any other self-referral States). Surely, the fact that Gabon referred itself to the ICC reflects a certain level of willingness to see the perpetrators of these crimes investigated and prosecuted, otherwise it would not have opted for this procedure.Footnote 1 But what made Gabon unable to investigate and prosecute crimes perpetrated before and during the August 2016 presidential elections? And should the objective of positive complementarity be about enabling the State to become effectively equipped (with domestic laws, policies and procedures) throughout its membership to the ICC – from the moment the State ratifies the Rome statute – rather than waiting for international crimes to be perpetrated and then finding out that the State is not equipped to carry our investigations and prosecutions?

The Gabon situation is a very familiar one. Following the 2016 disputed Presidential elections, President Bongo appeared to have been re-elected as president, but Mr Jean Ping, the opposition leader, declared the election invalid and demanded a recount. Typically, violence ensued, leading to Gabon’s claim that Mr Ping had engaged in incitement to genocide, and referred the situation to the ICC Prosecutor. In the meantime, Mr Ping also submitted evidence of crimes against humanity allegedly perpetrated by Gabon officials (on behalf of President Bongo) to the ICC Prosecutor. At the same time, France started an investigation of alleged crimes against humanity in relation to the treatment of a French national in Gabon at the relevant time,Footnote 2 and although no information can be found in relation to the outcome of this investigation, the ICC Prosecutor terminated the investigation due to lack of subject-matter jurisdiction.Footnote 3 However, even though the thresholdFootnote 4 needed to begin an investigation was not met, there may be lessons to be learnt with respect to the promotion of the principle of complementarity in the context of self-referrals.

The contention here is whether this self-referral practice has become too entrenched within African States and whether the ICC acceptance of such practice undermines local justice and accountability. And the issue that really needs some consideration is not about the evaluation of the legal basis for the application of the self-referral policy by the ICC Office of the Prosecutor (OTP),Footnote 5 but more about how this practice can be reconciled with the objective of positive complementarity in African States. There has indeed been some controversy regarding the self-referral practice, which has led to politized opinions about the choice of situations investigated by the Court. For example, in reference to the Uganda situation, Schabas, one of the strongest critics, contends that the self-referral practice is an ‘invention of the Office of the Prosecutor rather than the result of creative interpretation of the Rome Statute…’.Footnote 6 Schabas is also of the opinion that such procedure was never contemplated in the travaux préparatoires of the Court,Footnote 7 a sentiment shared by other prominent academics. Footnote 8 Ambos, in particular, refers to the use of the self-referral mechanism as an unexpected application of the State referral procedure, and considers the actions of the first ICC prosecutor, Luis Moreno Ocampo, to ‘actively’ seeking such self-referrals, as ‘questionable’.Footnote 9

Robinson, on the other hand, is quite critical of the assumptions made by a number of academics regarding the legitimacy, or even the correct interpretation, of the State referral mechanism.Footnote 10 He swiftly disposes of the main objections put forward by other academics regarding the creative interpretation of Article 14 Rome Statute,Footnote 11 as well as the claim that the Rome Statute negotiations delegates never intended such procedure to be included.Footnote 12 This position is also supported by Nouwen,Footnote 13 and indeed even the ICC Appeals Chamber reasoned that the State’s decision to ‘relinquish its jurisdiction in favour of the Court may well be seen as complying with the “duty to exercise [its] criminal jurisdiction” as envisaged in the sixth paragraph of the Preamble’.Footnote 14

Therefore, although the policy and practice of self-referrals is not the central issue here, this paper seeks to address a more theoretical question regarding the ongoing use, and effect, of the self-referral practice of African States, and whether the ICC can influence this practice to reinforce the objective of the complementarity regime, or more specifically the positive complementarity policy, so that States investigate and prosecute international crimes effectively.


In order to make sense of the issue raised in this article about the practice of self-referrals, it is important to re-visit the complementarity principle to understand the limited function of the ICC. The Court is tasked with the prosecution of international crimes only when States are unable or unwilling. It is, therefore, within the context of these limitations that the admissibility proceduresFootnote 15 provide the framework for the Court’s ability to exercise its jurisdiction. This is at the core of the complementarity regime encapsulated in the Rome Statute. As mentioned above, paragraph 6 of the Rome Statute Preamble recalls that ‘it is the duty of every State to exercise its criminal jurisdiction over those responsible for international crimes’. This is not, however, a legal duty as such, in the sense that the State party, upon ratification of the Rome Statute, is not under an obligation to incorporate its provisions into domestic laws. In itself, this lack of obligation can be interpreted as a contradiction, both in relation to the overall context of international criminal justice and the aims of the Rome Statute. If we examine, for example, the Geneva Conventions, a clear legal duty is imposed on States, either to prosecute the alleged crimes or to extradite to another State.Footnote 16 The same can be said with regard to the Genocide Convention; States have a clear obligation either to prosecute these acts or let an international penal court prosecute them, or extradite the perpetrators to another State.Footnote 17 Such obligation is also contained in Article 10 of the Draft Articles on the Prevention and Punishment of Crimes Against Humanity.Footnote 18

Having said that, the Court has expressly stated that the lack of national legislation concerning the investigation and prosecution of international crimes would not automatically constitute a justification for an admissibility challenge.Footnote 19 The rationale for this standpoint is that the Rome Statute does not distinguish between domestic and international crimes, but only considers the conduct of the accused.Footnote 20 And by adopting this wider concept States are afforded discretion with regard to the implementation of the Rome Statute crimes. In fact, according to the Case Matrix Network, the ‘ICC Statute does not specify when – or indeed how – its provisions ought to be implemented. A margin of discretion is therefore afforded to this end’.Footnote 21 But more to the point, the ‘decision to implement the crimes listed under Article 5 of the ICC Statute – genocide, crimes against humanity, war crimes, and aggression – remains at the discretion of the State.’Footnote 22

However, given the clear mandate for States to take on the primary jurisdiction for international crimes, would one still expect a clear obligation to incorporate such crimes at the domestic level in order to enable such primary jurisdiction effectively? The Rome Statute, in fact, only contains a reminder of this duty in the Preamble. Preambles can set the purpose and context for the interpretation of the treaty itself Footnote 23 but it would be too contentious to claim that preambles possess any legally binding authority.Footnote 24 Therefore State parties are not under any obligations to implement the Rome Statute in order to exercise their jurisdiction, though it has been argued that the domestic implementation of the Statute actually reinforces complementarity. For example, according to the Case Matrix Network the domestic implementation of the Rome Statute is ‘…the first step towards applying complementarity…and bringing the primacy of national courts closer’.Footnote 25 It further asserts that ‘…in order for a State to be able to investigate or prosecute crimes under its jurisdiction, it is highly advisable that States adequately incorporate core international crimes into their national legal systems’.Footnote 26 A number of commentators support the view that domestic implementation is in fact a necessary step in order to ensure the primacy of the State over any investigations and prosecutions, thus fulfilling their duty under the Preamble,Footnote 27 a view shared by Amnesty International:

In most cases, a State party to the Rome Statute will have to enact implementing legislation in order to fulfil its obligations under this treaty. In accordance with the principle of complementarity incorporated in the Rome Statute, such implementing legislation offers States an excellent opportunity to enable their prosecutors and courts to fulfil their primary role in ensuring accountability for [international crimes]…Effective implementing legislation will demonstrate that the State is aware of its primary responsibility under international law to ensure accountability for these crimes and will make certain that national courts, not the Court, will undertake these tasks.Footnote 28

According to Robinson, domestic implementation has an ‘intrinsic value’ as it forces States ‘to strengthen their legislative capacity to prosecute’ international crimes.Footnote 29 However, he also points out that there has been an ongoing misinterpretation of Article 17 of the Rome Statute admissibility procedure because commentators have failed to notice that this is a two-stage test rather than simply stating that if a State is unable or unwilling the case becomes admissible.Footnote 30 Robinson himself declares that such confusion is ‘puzzling’ and has led to the development of a ‘slogan’ concept of complementarity, which in turn has prompted a negative stance towards the Court because it is seen as applying a different interpretation to the admissibility criteria set out in the Statute.Footnote 31 What is more surprising, given the importance of this mechanism, is the fact that prominent academics seem to have adopted this incorrect interpretationFootnote 32 and, on one occasion, even the ICC Trial Chamber.Footnote 33 While the reasons for such mis-interpretation are not very clear, Robinson admits that failure to clarify the mechanism has led to more distrust towards the Court and its legitimacy.Footnote 34

However, what is undeniably certain is that the only legal obligation imposed on States is to cooperate with the Court.Footnote 35 It follows that all we are left with is a moral obligation to perform this ‘duty’ – to investigate and prosecute – which has been captured in the first ICC Prosecutor’s policy paper:

The system of complementarity is principally based on the recognition that the exercise of national criminal jurisdiction is not only a right but also a duty of States. Indeed, the principle underlying the concept of complementarity is that States remain responsible and accountable for investigating and prosecuting crimes committed under their jurisdiction and that national systems are expected to maintain and enforce adherence to international standards.Footnote 36

The recognition that the national criminal jurisdiction is a fundamental duty of the State is clearly connected to the traditional Westphalian model of State sovereignty, whereby the exercise of such jurisdiction rests on the State/Government authorities. Therefore, it is inherent to the function and nature of the sovereign State that investigations and prosecutions are carried out as part of the State’s duty. It is, however, paradoxical that it is generally the State or government officials, or individuals exercising control over the State organs (or the State territory), who tend to be the perpetrators of these despicable crimes. On this basis, it is therefore difficult to reconcile the declaration in paragraph 6 of the Rome Statute Preamble, or the OTP policy, with the contextual reality of international crimes. Perhaps, this is more about what the practice ought to be rather than what it is,Footnote 37 a concept that is closely associated with the interpretation of sovereigntyFootnote 38 and with the ongoing transformative process of the principle of sovereignty to bring it in line with the modern setting of the human rights discourse. This is, however, an issue to be discussed elsewhere, and whatever incongruities exist within the existing realities and constraints of the international criminal law machinery, the analysis must start with the provisions set out in the Rome Statute.

The discussion begins with an analysis of Article 17, whereby the ICC will investigate and prosecute international crimes within the jurisdiction of the Court.Footnote 39 According to this procedure, a case is admissible when the State is either unable or genuinely unwilling to investigate or prosecute the crime.Footnote 40 If a State has carried out (or is carrying out) genuine prosecutions, then the Court is barred from starting its own investigations and prosecutions, making the case inadmissible and thus confirming the primacy of the State in the prosecution of international crimes. The reach of the court was further clarified in the Katanga case:Footnote 41

  1. 1.

    Under article 17 (1) (a) and (b) of the Statute, the question of unwillingness or inability has to be considered only (1) when there are, at the time of the proceedings in respect of an admissibility challenge, domestic investigations or prosecutions that could render the case inadmissible before the Court, or (2) when there have been such investigations and the State having jurisdiction has decided not to prosecute the person concerned.

  2. 2.

    Inaction on the part of a State having jurisdiction (that is, the fact that a State is not investigating or prosecuting, or has not done so) renders a case admissible before the Court, subject to article 17 (1) (d) of the Statute.Footnote 42

Furthermore, the admissibility procedure should be applied in conjunction with Article 20 (3) of the Rome Statute, which reaffirms the ne bis in idem principle, according to which the Court cannot try an individual who has already been tried by a national court for the same conduct, thus setting out the two necessary limbs of an admissible case: same individual and same conduct. This is further re-affirmed in Article 17 (1) (c) of the Rome Statute. The test has been applied in a variety of judicial decisions, though in the Ruto Admissibility Judgment the phrase ‘substantially the same conduct’ was adopted,Footnote 43 a formulation criticized for its lack of clear legal origin and analysis,Footnote 44 though Rastan concludes that ‘substantially denotes the degree of variation that is permissible where the conduct in two cases is not identical’.Footnote 45 The same conduct also appears to require or include specific incidents, where ‘one or more crimes within the jurisdiction of the Court seem to have been committed’,Footnote 46 or same facts, which ‘must be understood as “criminal acts that occur in a particular location and at a specific time and in the framework of a course of conduct and series of events”’.Footnote 47 A slight deviation to this interpretation was adopted in the Gaddafi Admissibility Decision, where pre-Trial Chamber I widened the scope of ‘specific incidents’, noting that the events did not ‘represent unique manifestations of the forms of criminality alleged against Mr Gaddafi’ but rather ‘samples of a course of conduct’.Footnote 48 However, the Appeals Chamber rejected such interpretation and re-affirmed the necessity to refer to the specific conduct of the defendant, noting that the “conduct” that defines the “case” is both that of the suspect, Mr Gaddafi, and that described in the incidents under investigation which is imputed to the suspect’.Footnote 49

In light of such level of specificity with regard to the interpretation and application of the admissibility criteria, the State’s ability to actually investigate and prosecute international crimes may become more restricted and, at the same time, may widen the ICC complementary nature. Nouwen puts this quite plainly, stating that

…the same conduct test, and particularly the same-incident test, makes it practically impossible for a State to win on grounds of complementarity when faced with an ICC Prosecutor determined to be or stay involved: the domestic analogy fails to recognise the different context in which international crimes are typically committed.Footnote 50

There is also further ambiguity about the interpretation of the genuineness of any proceedings undertaken (or to be undertaken) by the State, especially as the traveaux préparatoires are generally silent on this issue.Footnote 51 Indeed, it is perhaps surprising that during the Rome Statute negotiations delegates did not choose to set some specific parameters to the word genuine, as contained in Article 17 of the Rome Statute. However, this is not surprising, as the adoption of the word genuine was in itself problematic and, as the delegates could not easily find agreement, eventually opted for the ‘least objectionable word’ – genuine – after rejecting words/phrases, such as ‘apparently well-founded’, ‘effectively’, ‘ineffective’, ‘good faith’, ‘diligently’ and ‘sufficient grounds’.Footnote 52 In accordance with the Vienna Convention on the Law of Treaties,Footnote 53genuine must be interpreted on the basis of its ordinary meaning, and one such meaning is ‘freedom from hypocrisy and pretence - sincere’.Footnote 54 For example, if we look at the Colombia situation, the punishment dispensed by the domestic courts is a reasonable indicator that the proceedings are genuine, as well as the lack of unjustified delays and the certainty that national proceedings are not aimed at shielding a defendant.Footnote 55 However, in the context of punishment, and specifically considering the domestic social and political contexts, there can be many subtle variations, and there is therefore room (and indeed a need) for the development of a clear framework for genuine proceedings in order to promote positive complementarity.Footnote 56

Therefore, whilst it is the Court that determines whether a case is admissible or not,Footnote 57 the narrow contours of the same conduct requirements and the lack of clear parameters applicable to the genuiness of proceedings may jeopardise the objective of complementarity. And this is especially the case when we consider the self-referral practice. Instinctively, as it is States (or rather their officials) that perpetrate these crimes, one could argue that it is in fact better for the ICC to define the contours of complementarity in order to achieve more effective justice. But what is the role of the State? In its simplest form, the Rome Statute creates the first ever International Criminal Court and sets out the Court’s procedures to make it operationally effective, including the cooperation procedure between member States and the Court. In other words, the creation of the Court does not change anything at the domestic level; it is not transformative, despite the high expectations envisaged at the time of its creation. In fact, at the domestic level, governments or government officials can still choose to do what they want: they can legislate and demonstrate genuine intentions to investigate and prosecute international crimes, not just to avoid the jurisdiction of the Court but also to endorse and reinforce a human rights-centric model of sovereignty. On the other hand, States can continue to perpetrate international crimes against their citizens, which may ultimately result in sham prosecutions or no prosecutions at all. In this context, the surrendering of the perpetrators to the ICC may be problematic, and they could still be allowed to hold prominent positions in government. They can also travel around friendly StatesFootnote 58 without incurring the risk of being arrested and surrendered to the Court. It may indeed become a waiting game for the ICC Prosecutor, waiting for a change of government,Footnote 59 or waiting for a previously friendly State to start cooperating with the Court. In essence, however, nothing much has changed. States (though not all of them) remain, as ever, interested in protecting and affirming their sovereignty status,Footnote 60 whereas the ICC is a Court setup with the mandate to prosecute international crimes – to complement States’ traditional jurisdiction – and end the culture of impunity.


If we think of complementarity as a ‘process of norm internalization’,Footnote 61 then all member States should, in theory, be empowered to participate equally and achieve a similar level of norm internalization in order to establish coherent and effective domestic implementations of international criminal law. The prevailing, shared and agreed meaning of complementarity is that primary responsibility for the investigation and prosecution of international crimes rests with the State; therefore ‘…their capacity to do so should be strengthened.’Footnote 62 This notion may be interpreted as the need for a structured and methodological approach to examine the specific circumstances of the State and provide appropriate assistance for that State to meet its demands under international law. The idea that all States are equal is not the same as the notion that all States are sufficiently developed and evolved to meet the standards and demands of international criminal justice. This is even more so in the case of States that, through the principle of self-determination, have acquired independent sovereign status. States emerging from long periods of European governance have had to conform to the State’s model created by international law, a model that is heavily influenced by the European tradition, and could potentially clash with local cultures and customs.Footnote 63 Moreover, a number of States still hold on quite tightly to a classical conception of sovereignty, where any notion of interference in the internal affairs of their own State is practically inconceivable, even if the nature of the interference is to protect fundamental human rights.

As the traditional Westphalian concept of State sovereignty has for far too long been used as a shield to protect alleged perpetrators of international crimes, it is vital not to continue this perverse ideology. However, the international criminal justice mechanism established by the Rome Statute should aim at enabling all States to achieve the status of being ‘able and willing’. It is acknowledged from the previous discussion that this may not have been the original objective of the Rome Statute, but the quest for the fight against impunity would benefit exponentially if more States were able to investigate and prosecute such crimes.

In addition, the principle of sovereign equality established in the UN Charter allows States to acquire a specific set of juridical rights. At the time, the principal aim was the establishment of the juridical principle of sovereign equality. It was the conviction that all States should be able to participate in international law making, either multilaterally or bilaterally, and that no State should interfere in the internal affairs of other States. Nevertheless, the objective of the principle of sovereignty should also be to promote a fair and just international community.Footnote 64 In other words, it is not just the international community, as an entity, that should be capable of offering optimal conditions to achieve stability and social well-being, necessary for peaceful and friendly relations among States,Footnote 65 but also for each State to achieve stability and well-being within its own territory. Cristina Pellandini, head of the International Committee of the Red Cross (ICRC) Advisory Service on international humanitarian law, has long recognised the importance of national committees in order to effectively realise the political will to adhere to international humanitarian law at domestic level:

Pushing forward the enactment of such legislation requires close cooperation between many different entities, both within the government and civil society. National committees for the implementation of international humanitarian law, because they are inter-ministerial or inter-institutional working groups, bringing together various national agencies with responsibilities in the field of international humanitarian law, have proved to be a very useful mechanism. Their main purpose is to advise and assist the government in implementing and spreading knowledge of international humanitarian law.Footnote 66

On this point, the ICRC does enable States to set up such national committees and provides ongoing support.Footnote 67 And notwithstanding the fact that the ICRC is a completely different entity from the ICC, there are clear common objectives. Therefore, the ICRC does not exclude the possibility that the activities of such national committees could extend to the implementation of other treaties’ obligations, such as the ones that arise from the Rome Statute.Footnote 68 And this in turn would embed a culture of prevention and prosecution of international crimes at the domestic level. If the ICC is serious regarding the principle of complementarity, then a more dynamic model could be put in place to raise awareness of the duty to investigate and prosecute, enabling States to internalize international criminal law effectively, but remaining sensitive to the principles of State sovereignty and non-intervention. This approach to complementarity is proactive,Footnote 69 in the sense that it positively encourages national jurisdictions to investigate and prosecute international crimes. Back in 2004 the ICC Prosecutor supported the concept of proactive or positive complementarity and clarified that one of the ICC strategies was in fact that ‘rather than competing with national systems for jurisdiction, we will encourage national proceedings wherever possible.’Footnote 70

This is therefore not just about putting the necessary legislation in place to enable States to investigate and prosecute, but also to encourage States to do so, a policy confirmed by the 2009–2012 Prosecutorial Strategy. This model would also help to dismiss feelings of distrust towards the International Criminal Court, especially in the African continent, which has attracted most ICC prosecutions to date.Footnote 71 Also, the ICC outreach programme, which normally takes place in post-conflict areas, aims at educating and empowering the victims of human rights violations. The aim of the Programme is to disseminate knowledge and understanding of the ICC and its judicial activities amongst a variety of audiences in order to increase victims’ awareness of the procedures and investigations that need to take place to bring perpetrators to justice. Together with a strategy of External Relations and Public Information, it aims at communicating actively with the relevant communities in order to build support and maintain a high level of cooperation. This programme could, in effect, lead to a catalysing effect of the Rome Statute, and its related ICC activities, into the domestic educational, legislative and judiciary processes, and the work of the Registry also contributes to this development at the local level.

However, in order to meet the objective of positive complementarity, this programme could be extended to situations where there is a history of political unrest or in transitional societies, rather than waiting for an event (an international crime) to occur on the territory of such State and then for the ICC to start its investigations and prosecutions because the State is not equipped to do so. But having the correct legislative and judiciary process in place is not synonymous with the ability and willingness to genuinely investigate and prosecute. The aim is to better equip the lawyers, the judiciary, and government officials and to instil concepts of justice and the rule of law in order to deal with impunity for international crimes at the domestic level. Moreover, given that the ICC can only deal with the most serious offenders, other perpetrators could potentially escape criminal liability if the State does not have the adequate domestic laws to investigate and prosecute them.Footnote 72 In this case, the ICC OTP, whilst carrying out their own investigations in a State where violations have taken place, could offer assistance to the local judges and lawyers to enable them to prosecute other perpetrators.Footnote 73 This process should be carried out with sensitivity towards the local customs and working practices, and should lead to a stronger dialogue with local human rights Non-Governmental Organisations (NGOs) and civil agencies in order to gain trust and dispel any doubts or perceived concerns of neo-colonialism.Footnote 74 Such criticism has been quite predominant in some African States. The indictment of Sudan’s President Al Bashir led the Sudanese government to declare that the ICC itself was a mechanism which perpetuated colonial dominance, ‘…a tool for those who believe that they have a monopoly on virtues in this world, rife with injustice and tyranny.’Footnote 75 In Kenya, on the other hand, the neo-colonial narrative was used as a weapon against cooperation.Footnote 76 However, as the Democratic Republic of the Congo (DRC) situation demonstrated, it would be presumptuous to assume that all States would be against closer collaboration with outside agencies and OTP officials in order to improve their domestic situations (including impunity and deterrence). Moreover, this approach supports the development of a dialogue at the national level, thus enabling the processes that are needed to rethink the State’s relationship to its citizens at the local level. Having said that, and despite the rhetoric of positive complementarity at the time, cooperation only took place in one direction.Footnote 77 Indeed, the Rome Statute does provide for the Court’s cooperation and assistance towards an investigating State, though the language here changes from shall to may, denoting discretion rather than obligation.Footnote 78

Proactive complementarity also supports the emerging doctrine of earned sovereignty, an approach that aims to resolve ‘…sovereignty-based conflicts by providing for the managed devolution of sovereign authority and functions from a State to a sub-State entity.’Footnote 79 This approach rests on the notion that the traditional principle of State sovereignty (or the acquisition of sovereignty through self-determination for many former colonies) may not provide the best basis to enable the concerned State to exercise its sovereign powers in keeping with a more contemporary understanding of sovereignty. According to this modern understanding of sovereignty, the development of domestic legislative procedures and judicial processes will be more in line with current conceptions of human rights protection and effective punishment for international crimes. This can be achieved through closer collaboration between the State, the OTP and NGOs. The development of internal legal structures to enable the State to carry out investigations and prosecutions of international crimes could occur through a joint effort, thus fulfilling the State’s primary jurisdiction mandate.

Bassiouni recognised the need to make national prosecutions of international crimes a reality. In a speech at an international law conference in Washington D.C. in 2010 he expressed doubts about the success of international criminal tribunals and the International Criminal Court. In particular, he recognised that the ICC will not be able to meet the expectation of the international community, mainly because of the “exorbitant” economic and administration infrastructure, citing 1,100 staff, a budget of $145 million (2010 only) and only four cases in 7 years since it started operating.Footnote 80 He concluded that the responsibility for the prosecution of international crimes would be taken up by the States.Footnote 81 At the same time, one of the key points that emerged from the ICC Review conference was the realisation that national jurisdictions needed to be strengthened. To this end a new resolution on complementarity was adopted, which, inter alia, reaffirmed that effective prosecutions for international crimes could be ensured by taking measures at national level, and encouraged the Court, State parties, international organisations and civil society to explore ways to enhance the capacity of national jurisdictions to prosecute international crimes.Footnote 82

Given the perceived dynamic and evolving nature of positive complementarity, and the seemingly developing policy that more needs to be done to enable States to prosecute international crimes, in practice – and specifically in the context of achieving positive complementarity in Africa – the policy does not appear to have been very effective. Schabas puts this quite succinctly when he remarks on the situations been prosecuted by the ICC and whether the gravity elementFootnote 83 is indeed been examined properly:

The first six “situations” investigated by the Prosecutor concern geographically contiguous countries in central Africa: Uganda, Sudan, Central African Republic, Democratic Republic of the Congo, Kenya and Libya. Is it really conceivable that an objective application of the gravity criterion, as proposed in materials from the Office of the Prosecutor, leads inexorably to this result? Is this simply a coincidence? There must surely be a strong presumption that some sort of policy determination is involved, absent any convincing explanation to the contrary.Footnote 84

As Schabas made these remarks back in 2012, Mali (a self-referral situation) and Côte d’Ivoire (a proprio motu situation) should be added to the existing list of African States where the ICC is carrying out investigations and prosecutions. A predominant focus on Africa continued until January 2016 when the Court authorised the start of proprio motu investigations in the Republic of Georgia.Footnote 85

In in the midst of this background, we do have a case in point with regard to positive complementarity, namely Colombia. What is important to note here is that, unlike Gabon, Colombia did not refer itself to the ICC, and what appears to have been extremely beneficial for Colombia was the OTP opening of a preliminary examination in its situation. This is clearly insufficient in itself to claim that there is a direct link between the opening of a preliminary examination and the State’s resolution to start domestic proceedings in order to avoid the Court’s jurisdiction.Footnote 86 Academics and practitioners are divided about its effectiveness,Footnote 87 but it may be a more effective tool to use rather than allowing the practice of self-referrals, as discussed below.

Colombia was affected by internal violent struggles between government and paramilitary forcesFootnote 88 for about fifty years, causing widespread killings and serious human rights violations, war crimes and crimes against humanity.Footnote 89 However, Álvaro Uribe’s election as President in 2002 was followed by a successful campaign against the paramilitary groups and marked the start of the transitional justice process. In 2002 Colombia ratified the Rome Statute, adding a declaration under Article 124 that it would not accept the jurisdiction of the Court for seven years with regard to war crimes. Although Colombia started to take some steps towards the investigations and prosecutions of the crimes alleged to have been perpetrated in the country,Footnote 90 the ICC, having received communications about the situation, decided to start a preliminary examination of the Colombia situation in 2006, in line with Article 15(2) of the Rome Statute.Footnote 91

This initial OTP interest in the Colombian situation was followed by the country’s credible attempts to conduct their own investigations and prosecutions, which in turn led the ICC to carefully scrutinise the legislative changes introduced by Colombia. This was followed by years of discussion and interaction between the OTP and Colombia, as well as an evaluation into whether Colombia was (and is) doing enough to secure the prosecution of the alleged perpetrators.Footnote 92 This led to the Deputy ICC Prosecutor, James Stewart, to declare that ‘Colombia, as a State Party to the Rome Statute, has engaged with the Prosecutor in a positive approach to complementarity’.Footnote 93

This level of optimism did have some shortcomings. The sentences set out by the Justice and Peace Law of 2005 ranged from five to eight years. These were indeed very short sentences given the gravity of the crimes committed, but the objective of this particular law was to bring justice and reparations; therefore, the short sentences were to be accompanied by truth telling, reparations to the victims and a promise not to return to unlawful behaviour.Footnote 94 Unsurprisingly, criticism of these lenient sentences ensued.Footnote 95 When the Colombian Constitutional Court was asked to rule on the merits of the Law, it did not comment on the apparent (and much criticised) by-product of this Law, namely that it introduced ‘a system of veiled pardon and covered up impunity’, stating rather that the Law tried to achieve a balance between peace and justice.Footnote 96 In the context of the argument set out in this paper, the Colombia situation provides the first and only example of positive complementarity. It is, however, not without criticism and there are doubts about its authenticity,Footnote 97 but the ICC continues to keep a watchful eye on the situation.

In the Libya situation, on the other hand, the ICC found the case inadmissible in relation to Abdullah Al-Senussi, the Head of Libya’s Military Intelligence,Footnote 98 though admissibility was upheld in the case of Saif Gaddafi, Libya’s de facto prime minister.Footnote 99 In both cases there were some procedural issues. Libya found it difficult to appoint defence lawyers but at least Al-Senussi was in Libya’s custody, whereas Gaddafi was not, partly due to the fact that Libya may not have been in effective control of the prison where he was held. Despite Libya’s on-going efforts to secure Mr Gaddafi’s transfer to Tripoli, to secure witnesses and to appoint a suitable defence counsel, the Chamber reasoned that Libya was unable to prosecute Mr Gaddafi, irrespective of whether it was willing.Footnote 100 The same potential obstacle existed in the case of Mr Al-Senussi – the lack of defence counsel – but the Chamber considered it to be a potential obstacle in the future, but not one that would have an immediate negative impact on the admissibility of the case.Footnote 101 There is also the issue mentioned in the discussion above regarding the different interpretations adopted by the Pre-Trial Chamber and the Appeals Chamber regarding ‘same conduct’, which could lead to uncertainty and possibly unpredictable results regarding the admissibility of a case. Ultimately, however, the Pre-Trial Chamber decided that the Gaddafi case was admissible because it could not determine the contours of the case investigated at the domestic level and therefore Libya had failed to demonstrate that the case being investigated was the same as the one set out in the Arrest Warrant.Footnote 102 It also expressed concerns regarding Libya’s ability to carry out a genuine investigation and prosecution in relation to Gaddafi.Footnote 103 Regarding the Al-Senussi case, on the other hand, it was found that the case being investigated by Libya was ‘substantially the same conduct’ being investigated by the Court,Footnote 104 concluding that the case was therefore inadmissible.

The overwhelming focus on African States, the (so far) successful story regarding Colombia and its domestic prosecutions, and the clarification of the admissibility procedures revealed in the Libya situation lead to question the potential for enhancement of the criminal justice model, the rule of law at the domestic level, and the development of a more tangible dynamic concept of positive complementary. Inconsistencies and lack of clarity with regard to the application of the concept of complementarity can be counter-productive for the establishment of a strong domestic model, but this is mainly the case when the concept of complementarity is interpreted in a narrow legal sense. For example, according to Phil Clark, complementarity should be viewed as a legal and political tool, which leads to four ‘overlapping notions’: a ‘legal framework’, a ‘political conception’, a ‘relational principle’ and a ‘developmental conception’.Footnote 105 The model proposed by Clark conceptualises the dynamic nature of what complementarity ought to be, so that the ICC does not become overloaded and the burden of (effective) investigations and prosecutions rests on the State, in accordance with the objective of the Rome Statute. But this is not the story of complementarity in Africa, where a near-tradition of self-referrals, accompanied by proprio motu prosecutions and Security Council referrals, have led to the adoption of a legal framework that does not resemble the dynamic nature of positive complementarity.Footnote 106


If we consider complementarity as a developing and dynamic mechanism that allows (or possibly enables) States to carry out effective prosecutions for international crimes, the conundrum now focuses on the issue of self-referrals. It is submitted that the self-referral mechanism essentially contradicts positive complementarity and the spirit of the international criminal justice project. It has already been argued above that positive complementarity, specifically in the context of investigations and prosecutions, has so far been successful only in the context of Colombia. The contention at this stage of the discussion is that positive complementarity and self-referrals should be evaluated together for the following reasons: first, it is necessary to determine whether the self-referral mechanism conforms to the policy of positive complementarity; secondly, if it does not, whether it is time to subject this mechanism to a more appropriate level of scrutiny to ensure it is in line with positive complementarity.

In accordance with Articles 13–15 of the Rome Statute, a situation can be referred to the ICC either by a State party, or by the Security Council, or by the ICC Prosecutor using proprio motu powers. A State party referral of a situation to the ICC, as envisaged originally by the Rome Statute drafters, meant that a State party to the Rome Statute would be able to refer another State to the ICC. Territorial or nationality limitations were never envisaged by the Rome Statute draftersFootnote 107 and, although potentially political in nature, this kind of State referral is easily justified on the basis of the gravity of the crimes and the common interests of all States to see these crimes prosecuted. However, this type of referral – State to State – has never taken place, and it may be for this reason that the Rome Statute contains the Prosecutor’s proprio motu powers as another trigger mechanism (as well as the SC referral), just in case no referrals were forthcoming from State parties.Footnote 108

All African situations currently under investigation and prosecution have been referred to the ICC using all three mechanisms, with one significant proviso: a State party referral was never intended to comprise self-referrals. An argument put forward by some commentators is that a self-referral is not a State party referral as envisaged by the drafters and it denotes an unusual underlying line of thinking: why would a State complaint about itself? Given the example above concerning the rogue State, a complaint by another State is expected, but how does a self-complaint fit within the purpose and spirit of the Rome Statute? The immediate answer would be that a self-referral provides the Court with work, and this is in fact how the Court initially became operational. The then ICC Prosecutor, Mr Luis Moreno-Ocampo, aware of the fact that the use of proprio motu powers to initiate investigations were not popular, actively invited voluntary referrals to enhance cooperation between the State and the Court, leading to the self-referrals by Uganda and the DRC.Footnote 109 Addressing the Second Assembly to State Parties in 2003, the Prosecutor, whilst stating that he was about to seek the Pre-Trial Chamber authorisation to start investigating the DRC situation, following a number of communications, he also acknowledged that

Our role could be facilitated by a referral or active support from the DRC. The Court and the territorial State may agree that a consensual division of labour could be an effective approach. Groups bitterly divided by conflict may oppose prosecutions at each others’ hands and yet agree to a prosecution by a Court perceived as neutral and impartial.Footnote 110

The self-referral mechanism enabled the Court to start its operations but what prompted Uganda and the DRC to follow this route? Were they automatically declaring themselves inactive in relation to the alleged crimes? Were there any advantages for them to follow this route, especially as it may have been quite likely that the OTP would have been able to obtain the necessary authorisations from the Pre-Trial Chamber? Whilst waiting for States to refer the DRC situation, Uganda delivered the first ever notification to initiate an investigation into the Uganda situation, indicating that

…the ICC is the most appropriate and effective forum for the investigation and prosecution of those bearing the greatest responsibility for the crimes within the referred situation…the Ugandan authorities have not and do not intend to conduct national proceedings.Footnote 111

It is important to set this line of thinking into the right context. This referral marked the beginning of the ICC operations, and there was clearly an exchange of interests between the Court and Uganda. Nouwen refers to this partnership as a marriage of convenience, with the occasional convergence of interests between the ICC and the Ugandan authorities, but on other occasions different interests led to a difficult relationship.Footnote 112 Difficulties started when Uganda requested the ICC to deal solely with the Lord’s Resistance Army (LRA), only to be confronted by ICC officials declaring such request as impossible because it is the OTP that sets out the scope of the investigations. For the ICC this marriage promised the convenience of Uganda’s full cooperation without the need to seek the Pre-Trial Chamber authorisation in order to start proceedings. Having said that, any investigations and prosecutions in this situation have so far dealt only with LRA commanders, with Otti and Kony still at large and Ongwen’s prosecution only started in 2016. There are, however, two slightly worrying factors regarding the Uganda situation: first of all, the underlying motive for the self-referral was Uganda’s inability to arrest the alleged LRA perpetrators, a point clearly stated in the Arrest Warrants.Footnote 113 However, if Uganda’s officials could not arrest the alleged perpetrators one would wonder how the ICC could possibly succeed in this task. It cannot! And this leads to the next crucial point, namely the realisation that an impromptu marriage of convenience, arranged to satisfy certain interests, may have led to an unintended entrenched practice for the African nations.

But then there is the issue of admissibility, which appears to lack the appropriate level of scrutiny that would be required in the context of complementarity. If, as mentioned above, complementarity is branded as a dynamic principle, then it should be discussed and debated at every stage of the proceedings, and in every case. This should be done to evaluate the State’s ability and willingness to investigate and prosecute any given individual at any given time. A correlation of this argument is that the Court maintains transparency throughout the proceedings, reminding itself that at all times it remains a Court complementing the State parties’ jurisdiction and not a Court of primary jurisdiction. In fact, despite the self-referral, the ICC admissibility has been disputed. When the Arrest Warrants for Kony and Ongwen were issued back in 2005, they simply stated that the cases ‘appear to be admissible’Footnote 114 on the basis that Uganda had not initiated any proceedings yet. The Prosecutor’s observations in 2008 merely confirmed the same fact,Footnote 115 giving an identical rationale, and Pre-Trial Chamber II confirmed that the admissibility issue was not in dispute because of the ‘total inaction on the part of the relevant national authorities’.Footnote 116 Appealing against the Pre-Trial Chamber decision on admissibility – because of a lack of proper legal representation at the admissibility hearing – it was stated that the decision would stand and that the lack of legal representation, though a procedural error, did not affect the substance of the case.Footnote 117

In fact, Uganda established the International Crimes Division within the Ugandan domestic courts in 2009, which could be interpreted as a catalysing effect of the self-referral to the ICC and a desire to seek peace in the region. It can be considered as an indicator of the transformative effect that the self-referral menchanism may have had on the domestic criminal justice system.Footnote 118 The reality, however, is not so ingenuous. The trial of Thomas Kwoyelo, the only LRA commander tried by the International Crimes Division so far, has been marred by ‘drama, intrigue and politics’.Footnote 119 There have been several attempts to hold a Pre-Trial hearing to confirm the charges for Kwoyelo in the past eighteen months, but for a variety of reasons the charges remain unconfirmed and the case remains at the Pre-Trial stage. It is difficult to evaluate at this stage whether the Ugandan efforts are genuine, although the Amnesty Act granting amnesty to any Ugandans involved in acts of war since 1986 may not help this state of affairs,Footnote 120 and Kwoyelo’s defence lawyers have already challenged the trial proceedings on the basis that Kwoyelo should be granted amnesty, in line with the Amnesty Act of 2000. Unusually, the OTP has not challenged the genuineness of these proceedings, despite the fact that they have been going on for ten years, the trial has not yet started, and the victims in this conflict are still waiting for justice. Given that the Uganda situation was a self-referral, it would be a welcome development for such a self-referral to have led to reinforced domestic procedures in order to prosecute effectively. But it is clear that the mere establishment of a War Crimes/International Crimes Division in Uganda is not enough to prepare a State to carry out effective prosecutions, despite the support offered by the OTP.Footnote 121 Oola’s account concludes with some stark words:

…the Ugandan government has hidden under the veil of complementarity to prosecute one side to the conflict, shy away from truth-seeking, deny immediate reparatory measures to victims and avoid acknowledgement of its responsibility and needed institutional reforms. All of this has contributed to silencing a majority of the victims of Uganda’s conflicts. The ICC’s intervention, in part, has enabled this one-sided focus.Footnote 122

At the same time, we can also question why Ongwen was arrested and surrendered to the ICC, rather than facing the prosecution in Uganda, given that at the time of his arrest the Uganda International Crimes Division was (theoretically) operational and, even more to the point, the main reason for the self-referral was Uganda’s inability to arrest the perpetrators.


Altogether, there have been five self-referrals by African States (Uganda, DRC, twice from the Central African Republic, Mali and Gabon) since the Court became operational. And in every self-referral case, apart from Gabon, the ICC Prosecutor started an investigation. The fact that Gabon’s self-referral did not end up in the list of the ICC investigations is not the real issue here. The issue is why do African States continue this practice, and why the ICC allows this practice to continue. This is not to suggest, of course, that the examination of a situation following a self-referral should not be conducted with the utmost scrutiny of the relevant legal criteria, and there are times when a self-referral may be the only feasible way forward in order to achieve justice. But there must always be some consideration for the development of dynamic complementarity. A self-referral could become just another excuse for ‘dumping’ a situation on to the ICC.Footnote 123 If we want to see dynamic complementarity in action, then African States, just like any other State in the international community, deserve the same opportunity to achieve national accountability for international crimes and, at the same time, strengthen their domestic criminal justice systems. In other words, we should be concerned about the potential catalysing effect of the Rome Statute within (African) national jurisdictions. The facilitation of this process should lead to the practice of self-referrals as exceptional and empower African States to strengthen their domestic processes of criminal justice and accountability, in line with the vision to end impunity for international crimes. The issue is not, therefore, about the ability – financial and human – of African States to conduct effective investigations and prosecutions. It is about the OTP’s adoption of a notion of complementarity that goes beyond the strictly legal framework and adopts a more dynamic model in order to encourage African States to carry out such prosecutions, through discussions and even political pressure.Footnote 124 Lastly, there is also an overwhelming need for a clear OTP policy about the self-referral mechanism, its relationship to the principle of proactive/positive complementarity and the State’s primary duty to investigate. Such policy should be followed by a clear programme of recommendations and support in the event – as in the Gabon situation – that the preliminary examinations do not lead to any investigations and/or prosecutions by the ICC.