1. Introduction
On 4 May 2019, Jean Leonard Teganya was found guilty of two counts of immigration fraud and three counts of perjury by the District Court of Massachusetts in Boston, based on his failure to disclosure his alleged involvement in the 1994 genocide against the Tutsi in Rwanda.Footnote 1 In immediate response to the judgment, Special Agent in Charge, Peter C. Fitzhugh from Homeland Security Investigations, described the verdict as a victory in preventing the exploitation of ‘America’s historic hospitality for immigrants’.Footnote 2 This use of domestic prosecutions for immigration offences based on allegations of involvement in an international crime fits into an emerging global practice. In the United States of America (USA), the Teganya decision was preceded by three similar decisions concerning Prudence Kantengwa, Beatrice Munyenyezi, and Gervais Ngombwa. Following these verdicts, USA Attorney Carmen M. Ortiz said:
These cases should send a strong message to all those who would seek to cheat the immigration system by lying about their background or otherwise deceiving U.S. immigration authorities. The United States will not be a safe haven for those who conceal their past in order to gain the privilege of living in this country.Footnote 3
All three individuals will face deportation to Rwanda after serving their sentences, joining four other Rwandans deported from the USA on suspicion of their involvement in the 1994 genocide.Footnote 4
In the USA, border control is at the forefront of the justifications for these criminal convictions. In Rwanda, these cases fit into a much wider set of activities pursued by the Rwandan Genocide Fugitives Tracking Unit (GFTU), a specialist unit of the National Public Prosecution Authority (NPPA). As of 2018, the GFTU had issued 911 indictments for individuals suspected of involvement in international crimes.Footnote 5 To date, these proceedings have seen 19 individuals deported or extradited to Rwanda to face genocide charges before the Rwandan national courts. Inside Rwanda, these cases are ascribed a very different public meaning to the USA’s explicit focus on border control. They are explained in terms of the latest victories in the ongoing global fight against impunity for international crimes. This is echoed in the international press, which has provided supportive commentary on the high-profile deportation of Rwandan genocide suspects Léon Mugesera and Jean-Claude Henri Seyoboka from Canada and, among others, the extraditions of Charles Bandora from Norway and Jean-Baptiste Mugimba and Jean Claude Iyamuremye from the Netherlands.Footnote 6
Read together these accounts of domestic legal proceedings concerning an international crime express both a universal condemnation of genocide that extends beyond the nation state and simultaneously are being drawn on to reinforce a state’s control of its borders. This article examines this close relationship between national interests in border control and the universal claims of international criminal law. It draws on an independently generated dataset of 122 cases in 20 countries concerning the immigration status, extradition, deportation or trial of individuals suspected of involvement in the Rwandan genocide. In addition to the 19 individuals returned to Rwanda, in these proceedings 31 individuals have had their extradition to Rwanda denied, 36 have faced domestic criminal trials outside of Rwanda and 29 people have had their refugee protection, residency permits or citizenship revoked or have been prosecuted for immigration offences on the basis of an allegation of their involvement in an international crime.Footnote 7 This article looks across these cases to describe the relationship between extradition and immigration related proceedings and then engages in a close reading of cases in France, the United Kingdom (UK), the Netherlands, the USA, Canada, and South Africa to offer a detailed analysis of the expressive work of these judgments, illuminating the role of international criminal law in domestic border control.
The article argues that these cases communicate not only an ongoing and important international commitment to recognizing the wrong of genocide, but also more ambiguously what constitutes a fair trial in Rwanda, who constitutes a ‘criminal migrant’ and, to a Rwandan audience, the transnational penal reach of the Rwandan state. In doing so, the article uses expressivism as a regulatory principle in international criminal law. It is in disaggregating the public meanings of these cases that we are then able to ask the difficult question; are these legal proceedings the right way to pursue the laudable goals of deterrence, retribution or restoration following international crimes?
The argument of the article is structured around three points: first, it places the Rwandan cases within the wider literature on the intersection of international criminal law and international refugee law, showing a strong focus in this scholarship on ensuring the compatibility of these bodies of international law which are understood to uphold shared values. Drawing on insights from discussions in domestic criminology and transnational criminal law, which adopt a more cautious approach to the use of criminal law when dealing with the mobility of people, this section argues for the need to make visible the entanglements of this international law within domestic immigration regimes. In doing so, it questions the comfortable assumption that the major expressive function of international criminal law at the domestic level is to reinforce the universal condemnation of international crimes.
The second section of the article focuses on the Rwandan cases, allowing us to look across jurisdictions to see the domestic legal entanglements of international criminal law. A close reading of a selection of jurisdictions where the majority of cases have been decided show how immigration proceedings regularly foreshadow extraditions, deportations, and criminal proceedings against the same individuals. In these immigration proceedings, the domestic orientation of the law is very apparent, with international law drawn on to differing extents and given domestic interpretive content. This section argues that recognizing this domestic orientation of the law is the first step to examining the potential for plurality in the expressive functions of these cases.
The third section pursues an expressivist assessment of these cases, examining what they communicate about the criminal justice system in Rwanda, about notions of who is an acceptable immigrant and about how the Rwandan state engages with its diaspora. In doing so it shows the plural expressions of these cases both in terms of the audiences they address and the norms they communicate. The article demonstrates how proceedings relating to immigration law either offer more favourable readings of fair trial rights in Rwanda than those described in extradition proceedings or bypass an assessment of fair trial altogether. In addition, this expressivist review highlights how the intentions underpinning the use of international criminal law in immigration-related litigation centre on communicating ideas around criminal migrants, in addition to the more established expressions regarding the legal norm criminalizing genocide. The article argues that it is in acknowledging these plural expressions that scholars will be best placed to assess in a more circumspect and critical fashion the ways in which the future of international criminal law may lie in its domestic application.
2. International criminal law, refugee law and the use of exclusion, deportation, and extradition
The use of criminal law in border control has gained increasing and warranted scholarly attention.Footnote 8 International criminal law is no exception, although the orientation of the debates in international law is different from that at the national level. While the scholarship on domestic border control is characterized by a deep scepticism of the expansion of immigration related offencesFootnote 9 and the use of deportation as a means of sanction,Footnote 10 the focus in international criminal law has been on the exclusion of individuals suspected of international crimes from the protective sphere of international refugee law. This exclusion is generally seen as justified, legitimate, and important for the integrity of both of these bodies of international law as it reinforces the condemnation of international crimes.Footnote 11 The outcome is that in domestic discussions of border control, the examination of criminal law is from a position of caution, while in international criminal law, it is more expansive, with attention on the harmonization of two bodies of international law. Yet the domestic and the international legal regimes are not hermetically sealed. As laid out below, what is often neglected in the scholarship on refugee exclusion, is that the law being applied at the national level forms part of a state’s wider immigration regime. In focusing on the Rwandan cases pursued around the globe, this article illuminates these domestic legal entanglements.
To some extent it is unsurprising that international criminal law and refugee law have historically and currently been widely considered to reinforce shared values regarding the moral wrong of international crimes. In the aftermath of the Second World War, the Refugee ConventionFootnote 12 and the Charter of the International Military TribunalFootnote 13 created the foundational agreements upon which much of the subsequent international legal apparatus of both international criminal law and international refugee law has been built. These two bodies of law were co-constitutive and understood to support one another. The first criminal charges at Nuremberg were brought just as the Refugee Convention was being drafted. The exclusion of suspected war criminals was seen as a way of protecting the integrity of the asylum system while simultaneously reinforcing the international condemnation and universal sanction of the crimes themselves.Footnote 14 At this early juncture, the key expressive work credited to international criminal law, discussed at length in this article, already extended beyond the simple communicative function of a particular punitive sanction handed down in a specific case, to wider claims about its role in signalling shared moral outrage.Footnote 15
It was as a result of this assumed compatibility of these newly developing areas of international law that there was very little debate or disagreement over the drafting of Article 1F(a) of the Refugee Convention, which excludes individuals from the scope of the Convention where there are ‘serious reasons for considering’ that they have ‘committed a crime against peace, a war crime, or a crime against humanity, as defined in the international instruments drawn up to make provision in respect of such crimes’.Footnote 16 The exclusion of individuals from refugee protection supported the emerging legal norms regarding appropriate conduct both during war and towards civilians and protected groups as envisaged in international humanitarian law and the nascent field of international criminal law. These mutually reinforcing dimensions are further visible in the expansion of international criminal law in the 1990s. The establishment of the ad hoc Tribunals for Rwanda and for the former Yugoslavia, the hybrid courts in Sierra Leone and Cambodia, and the International Criminal Court (ICC) were contemporaneous with the Office of the United Nations High Commissioner for Refugees’ (UNHCR) first set of public guidelines on Article 1(F).Footnote 17
These concurrent developments of international criminal law and international refugee law is not to say that critiques of this area of law have been neglected but rather to draw attention to the fact that these critiques have predominantly focused on enhancing the compatibility of these two areas of international law. In particular, important concerns have been raised around the consistency across jurisdictions of the standard of proof set by the benchmark of ‘serious reasons for considering’Footnote 18 and debates around the use of international criminal law to determine what constitutes ‘committed’ within Article 1F(a).Footnote 19 Much of the work in this area has offered suggestions as to how international criminal law could help with the harmonization of the interpretations of Article 1F(a) across domestic jurisdictions. Generally, the punitive impulse is very apparent, with studies on individual refugee-receiving states focusing on examining the possibilities for domestic prosecution of international crimes if return to the home country is deniedFootnote 20 and the difficulties that arise when refugee status has been revoked but return is prohibited.Footnote 21
Within the more critically-oriented literature there is an important acknowledgment that different objectives underpin these bodies of law,Footnote 22 with scholars such as Catherine Dauvergne raising concerns that, if imported wholesale into refugee law, international criminal law could exponentially expand the area of refugee exclusion.Footnote 23 It is notable that Dauvergne’s critique is published as part of an edited collection that is focused on domestic concerns about the use of criminal law in border control where the turn to criminal law has been greeted with much more scepticism. Within this literature, Mary Bosworth coined the phrase ‘penal humanitarianism’, describing how calls to humanitarianism in areas of criminal law and migration allow penal power to traverse national boundaries and travel globally.Footnote 24 Building on this work, Kjersti Lohne examines international criminal law as an instance of penal humanitarianism ‘disembedded from the nation state altogether’.Footnote 25 Looking at the refugee exclusion, trial, extradition or deportation of Rwandan nationals builds on this work, providing an example of how this notion of international penal humanitarianism is then ‘re-nationalized’ and enmeshed in domestic practices of immigration, extradition, and criminal law. Bringing these criminological insights to bear on the debates in international criminal law asks us to turn our attention to the domestic entanglements of international criminal law within the wider set of immigration practices.
International criminal law is somewhat of a late-comer to these legal realist interests in the background rules, in this instance immigration laws, that are at work in any legal regime.Footnote 26 For some time debates in transnational criminal law, particularly those around the criminalization of human trafficking, have prompted scholars to acknowledge that treaty development that requires states to criminalize trafficking, smuggling, and modern slavery simultaneously restrict the capacity to claim asylumFootnote 27 and have actively reinforced immigration control measures.Footnote 28
Yet within discussions of Article 1F, the domestic immigration law dimensions are often explicitly separated from the analysis undertaken by international lawyersFootnote 29 and, where they are included, they are examined in terms of the compatibility between refugee exclusion and extradition law. Here the bodies of law are once again assumed to reinforce shared values with, for example, Geoff Gilbert noting that ‘denying the status of refugee … makes transfer to stand trial [for international crimes] more viable’.Footnote 30 Extradition is then framed as a way of upholding the fundamental guarantee of non-refoulement while ensuring ‘non-impunity’. In this vein, the Rwandan extradition cases in the Netherlands following the revocation of refugee status have been examined as a preferred outcome.Footnote 31 In short, the combination of refugee exclusion and extradition law are understood to work together to both fight impunity and protect the integrity of the refugee system.Footnote 32 However, this international criminal law literature fails to take sufficient account of how refugee exclusion is embedded in wider domestic immigration regimes. Insights from debates in criminology on crime and border control and in legal realist discussion in transnational law on the use of criminal law to constrain global mobility prompt a closer examination of the domestic work of international criminal law in controlling borders. In doing so, this empirical enquiry is set within a wider trajectory of scholarship raising critique of the use of criminal law as a means of realizing and protecting human rights.Footnote 33
A broader examination of the cases relating to Rwandan nationals, all of which concern the allegation of involvement in international crimes, provides an opportunity to look across jurisdictions to describe the domestic immigration and extradition laws. The domestic orientation of these laws has been largely neglected due to the focus in the literature on the harmonization of international law and the strength of the moral impulse of ‘fighting impunity’ so deeply embedded in international criminal law. The focus on the Rwandan cases enables an analysis of the role of international criminal law in domestic immigration regimes, including, but not limited to, the incorporation of Article 1F of the Refugee Convention. In doing so, it draws attention to how, in the Rwandan cases, immigration proceedings regularly foreshadow extraditions, deportations, and criminal proceedings against the same individuals and asks for a closer interrogation of the different types of expressive work that these proceedings are doing.
3. Cases concerning the immigration status, deportation, extradition, and trial of individuals suspected of involvement in the 1994 Rwandan genocide
The importance of expressing the ongoing condemnation of the Rwandan genocide and establishing the individual responsibility of those involved, has been a central driver in the extensive efforts to locate and arrest Rwandan genocide suspects around the world.Footnote 34 To understand the reach and extent of this transnational phase of post-genocide accountability-seeking, that has run concurrently with, and now extends beyond, the operation of the United National International Criminal Tribunal for Rwanda (ICTR) and the localized gacaca courts,Footnote 35 it is necessary to start with the formation and work of the GFTU. Since 2007, the GFTU has endeavoured to investigate allegations against genocide suspects living outside of Rwanda and to co-operate with national prosecution services and international judicial bodies to either prosecute the accused in the host state or extradite or deport them to Rwanda. The establishment of the GFTU was preceded by an initial partnership formed in 2004 among Interpol, the ICTR, and the NPPA. This led to the issuing of 300 Interpol Red Notices against Rwandan nationals,Footnote 36 including the then nine remaining ICTR fugitives.Footnote 37 While these Interpol warrants have been a centrepiece of the work of the GFTU, the number of indictments issued by the Rwandan government goes well beyond this, with the official GFTU report from April 2018 stating that 911 indictments have been issued.
The number of completed and ongoing cases is more difficult to determine and as a result this article draws on an independently generated dataset that maps, as comprehensively as possible, all of the jurisdictions in which cases concerning Rwandan genocide suspects have been or are currently being heard. Within this dataset a case constitutes a set of legal proceedings, including an initial decision and any subsequent appeals. These cases have been heard in 20 different jurisdictions, with completed or ongoing litigation in Sweden, Denmark, the Netherlands, France, Italy, the UK, Switzerland, Finland, Belgium, Germany, Norway, the USA, Canada, the Democratic Republic of Congo, Uganda, Zambia, Malawi, Kenya, South Africa, and New Zealand. From the total of 122 cases, 60 judgments were publicly available. These cases were heard in 12 of the 20 countries in which proceedings have been undertaken.Footnote 38 Where the judgments were not publicly available, cases were only included where information could be triangulated through three separate sources, including media reports and key informant interviews. What is crucial is that across the different jurisdictions these cases have been decided under three possible legal regimes: extradition proceedings, immigration proceedings, and domestic criminal trials.
In the analysis below, the dataset as a whole is drawn on to identify overarching trends regarding the outcomes of the decisions and the relationship between immigration, extraditions decisions, and domestic trials. The jurisdictions in which written judgments were available are focused on building the arguments regarding the domestic orientation of the applicable laws and the expressive roles of these decisions regarding fair trial in Rwanda and the suspect’s immigration status. Overall, the disaggregation of cases concerning immigration, extradition or domestic trial makes it possible to identify the differences in legal reasoning, across jurisdictions and across different areas of law. It is this comparative undertaking that illuminates the mutually reinforcing relationship between national interests in border control and universal claims regarding the wrong of international crimes.
3.1 Domestic entanglements
To make sense of the role of international law in the diverse jurisdictional settings in which Rwandan genocide suspects have been pursued, it is important to acknowledge the domestic orientation of the relevant law that is being applied. These domestic entanglements challenge the current scholarship’s general focus on international criminal law’s compatibility with international refugee law, showing the varied roles that international law plays in what is fundamentally legislation relating to the immigration law of the state concerned. Within the dataset all of the refugee-related decisions are based on domestic immigration law, only some of which include the wording reflected in the 1951 Refugee Convention. The extent of this domestic orientation is immediately evident when looking at one of the questions that has featured heavily in the literature on international criminal law and Article 1F generally, that of the standard of proof required to support the allegation of involvement in an international crime. Looking across the cases, domestic immigration law is hard at work. It is this initial observation that suggests the need to look at the expressive work of these decisions beyond their role in singularly supporting the moral condemnation of international crimes.
The majority of the cases in the dataset have occurred in France (31), the Netherlands (14), and the USA (12) and 33 of these decisions were available and reviewed. A brief discussion of the law as it has been applied in these jurisdictions shows the differing level of influence of international law in these domestic legal immigration regimes. In addition, the Canadian cases have been included as they highlight how judicial interpretation contributes to the domestic orientation of the laws. Together, describing these domestic entanglements suggests that these cases may be oriented towards different audiences and may be doing different types of expressive work.
In the Netherlands, the key piece of domestic legislation is the Aliens Act 2000 within which Article 33(a) provides the administrative means through which a residency permit can be cancelled if information that could lead to the rejection of a refugee application was withheld or incorrectly represented.Footnote 39 It is this administrative measure that, in the cases reviewed, then enables consideration of Article 1(F)(a) of the Refugee Convention and whether information grounding exclusion was withheld or misrepresented. Showing the importance of domestic interpretations of the relevant law, the Convention is then given very specific content, placing the burden on the state to demonstrate that there are ‘serious grounds for believing’ that a foreign national is subject to the criteria and explicitly addressing the threshold of what constitutes ‘committed’, establishing a ‘personal and knowing participation test’.Footnote 40
In France, the court similarly directly applies the standard of ‘serious reasons for considering’ as found in the Refugee Convention. However, in addition, Article L. 712-2 of the Code of Entry and Residence of Foreigners and of the Right of Asylum provides that asylum seekers shall not benefit from subsidiary protection if there are serious reasons to believe that they have committed a crime against the peace, a war crime or a crime against humanity.Footnote 41 The cases in the dataset do not provide details on what is required to meet this standard, although in cases such as those brought against the former Rwandan first lady Agathe Kanziga Habyarimana, it was held that it is sufficient in meeting this legal threshold to rely on secondary sources such as non-governmental organizations (NGO) reports and anonymous testimony before other jurisdictions’ courts, including the ICTR.Footnote 42
In the USA, the cases concerning Rwandan nationals have been decided exclusively under domestic immigration law. The Immigration and Nationality Act 1952, Section 208, establishes that an individual can be excluded from asylum protection if they ‘ordered, incited, assisted, or otherwise participated in the persecution of any person on account of race, religion, nationality, membership in a particular social group, or political opinion’.Footnote 43 The standard of proof required to meet this ‘persecutor’s bar’, is that there is sufficient evidence if it ‘raises the inference’Footnote 44 of participation in persecution and this is then coupled with litigation relating to domestic criminal offences for immigration fraud.
In Canada, the case against Mugesera directs us towards Section 19(1)(j) of the Immigration Act 1985 which, at the time it was in force, established that a person shall not be granted admission to Canada if there are ‘reasonable grounds to believe’ that they have committed a crime against humanity outside Canada. In this case, the Supreme Court established that the standard of ‘reasonable grounds to believe’ that the defendant was involved in an international crime requires something more than mere suspicion, but less than the standard applicable in civil matters of proof on the balance of probabilities. The Court goes on to state that ‘reasonable grounds will exist where there is an objective basis for the belief which is based on compelling and credible information’.Footnote 45 This standard is a domestic one and at this juncture the law did not refer to the ‘serious reasons for considering’ articulated in international law. This Immigration Act has now been replaced by the Immigration and Refugee Protection Act (IRPA)Footnote 46 which directly incorporates Article 1F of the Refugee Convention in Section 98 of the Act. The IRPA was already in force by the time Mugesera’s case came before the Supreme Court in 2005. However, the Court referred to the law as it was at the time of his initial exclusion decision. One of the outcomes of this appears to have been that this domestically generated standard of proof has now been generally applied as the standard of proof set by the wording ‘serious reasons for considering’.Footnote 47
There is a lot that could be unpacked here and, at this juncture, this analysis offers a basic sketch. Nonetheless, what is clear is that the applicable law is embedded in legislation that is fundamentally oriented towards domestic immigration control and it makes use of the standard set through the Refugee Convention to differing degrees. It is not simply a question of drawing on international criminal law as a tool to harmonize the domestic implementation of international refugee law. The intentions underpinning these laws are domestically as well as internationally oriented. An expressivist examination of these cases enables a disaggregation of the potential audiences that are being addressed and opens up the potential to make visible the different normative expressions that are at play. This, however, requires a careful account as to what is meant by expressivism and why it may be analytically valuable for cases drawing on international criminal law that are being decided under a range of domestic laws.
4. Examining the expressive work of the cases concerning participation in the Rwandan genocide
Within the literature on international criminal law, expressivism has been drawn on as a justification not only for the punishment of those convicted of international crimes,Footnote 48 but also for the selection of suspects,Footnote 49 the specific charges,Footnote 50 the trials themselves,Footnote 51 and as a way of assessing the value of dissenting opinions.Footnote 52 The focus on the capacity of international criminal law to generate and legitimate new legal norms has led scholars to describe expressivism ‘as the best justification for the ICC’s work’.Footnote 53
At first glance, this turn to expressivism may seem to relieve international criminal law of some of its anxieties around establishing a deterrent effect insituations of collective violence or offering proportionate punishment to such severe harms. However, as Elizabeth Anderson and Richard Pildes suggest, expressivist justifications for particular legal practices do not stand as goals in and of themselves, the aim is not to ‘maximise the proper amount of expression in the world’.Footnote 54 In their view, an expressivist examination offers a way of interrogating the connection between the use of particular legal processes and the efforts to achieve specific goals. An expressivist approach allows us to ask: are these legal proceedings the right way to pursue the goals of deterrence, retribution or restoration? In this way, an expressivist examination of these legal proceedings functions as a regulative principle for intentions, not as a goal of action.Footnote 55
Examining these intentions requires accounting for what those undertaking these trials are intending to express and what they think their actions do express alongside, and in addition to, an external evaluation of the ‘public meaning of their actions’.Footnote 56 How are these trials being interpreted by others? In Anderson and Pildes’ view this should be coupled with an ‘external normative judgement’Footnote 57 that sets external criteria as to whether these attitudes that are being expressed are appropriate. This approach to expressivist analysis means that it can be usefully drawn on to look at legal proceedings decided under different branches of law. It also asks us to disaggregate the ‘public’ that may be addressed by trials concerning allegations of involvement in international crimes. This disaggregation is particularly pertinent when, as discussed in the previous section, these trials are being conducted in courts applying domestically oriented immigration related laws alongside extradition and criminal law.
In the established scholarship on expressivism in international criminal law, the potential diversities of the communities are, to some extent, acknowledged. De Guzman describes these communities as ‘states, nongovernmental organizations (NGOs), affected communities, and the global community’,Footnote 58 while other scholars are more circumspect, acknowledging the notion of the international community as ‘figurative’Footnote 59 or more critically as ‘illusory’.Footnote 60 However, as the domestic orientation of the law discussed in the previous section suggests, in these transnational cases the ‘affected community’ includes an immediate domestic audience, alongside and in addition to the affected Rwandan community. As argued in the next section, this requires that the decisions relating to extraditions be looked at alongside those relating to immigration, where the orientation to the host state’s domestic audience is particularly pertinent. In addition, while the existing writings do acknowledge the need to disaggregate the communities to which the expressive value of trials are communicated, less attention is given to the possibility that the norms themselves may be plural and contradictory.Footnote 61 It is only in exploring the potential plurality of the norms expressed that the heterarchical operation of international criminal law starts to become visible.Footnote 62
To engage with this heterarchical structure and the potential plurality of normative expressions, to a variety of communities, it is necessary to disaggregate the expressive functions of these proceedings to examine which courts are signalling what, to whom. As discussed in the next section, disaggregating the expressive functions of these cases goes some of the way in helping to explain the stark divisions in the dataset that has seen five countries refusing to send individuals back to Rwanda and ten countries supporting their extradition or deportation. The legal standards and expressive role of immigration law, rather than extradition law alone, needs to be accounted for when explaining these differences. The next section separates out these different expressive functions, showing what these cases say about the criminal justice system in Rwanda and about the role of international criminal law in domestic immigration law. It is then possible to ask whether these different expressive roles illuminate appropriate intentions underpinning these particular actions. In this way expressivism, as a regulatory principle, offers an analytical tool to assess the use and work of international criminal law in immigration and extradition proceedings, rather than simply falling back on comfortable assumptions about the value of ‘fighting impunity’. As demonstrated below, it helps to explain how and why international criminal law can be drawn on to re-enforce national borders while continuing to make universal claims.
4.1 The criminal justice system in Rwanda
One of the most striking issues across all of the cases in the dataset is that, although decided according to very different laws, they all offer an assessment of the criminal justice system in Rwanda. In doing so, they offer judicial expressions on fair trial rights in Rwanda today. Here the picture is complex as the dataset shows marked differences in appraisals of the Rwandan criminal justice system both across and within different jurisdictions. Within jurisdictions, immigration proceedings offer a more positive reading of fair trial than articulated in the extradition proceedings or they provide a route through which an assessment of fair trial is bypassed altogether. To explore this complexity, it is helpful to break these cases into five categories:
1. Cases where extradition to Rwanda is currently refused
Relevant jurisdictions: France, the UK, Italy, Switzerland, Finland
2. Cases where extradition to Rwanda has been allowed
Relevant jurisdictions: Sweden, Denmark, the Netherlands, Norway, Germany, the DRC;
3. Cases where the immigration decisions alone have resulted in deportation to Rwanda
Relevant jurisdictions: Uganda, Canada, the USA;
4. Cases where only domestic criminal trials have been undertaken
Relevant jurisdiction: Belgium;
5. On-going proceedings
Relevant jurisdictions: Zambia, New Zealand, Malawi, Kenya, South Africa.
With ten states sending individuals back to Rwanda and only five states refusing to do so, an initial assessment would suggest that there has been a shift from the ‘judge-spun web of scepticism’Footnote 63 on fair trial rights in Rwanda, correctly identified by Mark Drumbl in 2007, towards one of increasing confidence in the Rwandan criminal justice system. This change was seemingly ushered in by the ICTR’s second wave of decisions to transfer some of its remaining caseload to Rwanda.Footnote 64 Yet, a closer look at the timing of cases refusing extradition challenges this linear account, most notably with France and the UK refusing extradition in 2016 and 2017, respectively, despite extraditions being upheld in Norway, Sweden and subsequently by the European Court of Human Rights (ECtHR) in 2011, by Denmark in 2013, the Netherlands in 2016, and Germany in 2017.
In the European cases, the legal standard of assessment centres on whether, under Article 6 of the European Convention for the Protection of Human Rights and Fundamental Freedoms (ECHR), there is a ‘real risk of a flagrant denial of justice’. The interpretative approach across the European cases has been broadly uniform regardless of whether the country is deciding to extradite or not, with, for example, the UK cases supporting Lord Phillips’ assessment in RB (Algeria) Footnote 65 that in determining what constitutes a flagrant denial, the Court should consider the outcome of a breach, if it eventuates, such as whether there is ‘a significant adverse outcome, in terms of the nature of conviction and sentence’.Footnote 66 As a result, the importance of these cases has often been framed in terms ‘of a very particularized risk’Footnote 67 facing Rwandan nationals without any immediately visible wider implications of the decisions. However, focusing exclusively on the legal norms expressed through the ECHR obscures the differences in what is being expressed about criminal justice in Rwanda, which is being determined both as being fair and as being unfair and crucially the role that immigration law is playing in the processes overall. As argued above, to take account of this pluralism, it is necessary to disaggregate our expressionist appraisals.
4.1.1 Category 1 - Cases where extradition to Rwanda is currently refused
For states that continue to deny extradition to Rwanda, the court decisions communicate two central concerns regarding fair trial in Rwanda. The first focuses on the domestic legal framework and the second on the capacity to raise an effective defence. Immigration courts, however, have offered a different reading, looking more favourably on Rwanda’s capacity to provide a fair trial.
The domestic legal framework has been a long-standing point of assessment in all of the cases concerning the transfer,Footnote 68 extradition, and deportation to Rwanda of individuals suspected of involvement in the 1994 genocide. In 2007, the Rwandan parliament passed a bespoke ‘Transfer Law’Footnote 69 to facilitate the reception of cases from both the ICTR and from third country jurisdictions. However, the initial refusal by the ICTR to transfer cases to Rwanda highlighted concerns over the resultant law which established a life-sentence with solitary confinement as the maximum penalty, coupled with issues regarding the willingness of witnesses to testify for the defence. The Rwandan parliament responded to these criticisms through extensive legislative change, removing the solitary confinement condition for cases that fell under the Transfer Law and reforming their witness protection programme. This resulted in the ICTR’s Uwinkindi decision, which granted transfer to Rwanda, stating that, ‘the relevant Rwandan laws must be given a chance to operate before being held to be defective’.Footnote 70 Yet the domestic legal framework remains one of the two reasons that extraditions continue to be refused.
The most recent UK High Court decision denying extradition turned in part on a question of Rwanda’s domestic legal framework. The issue concerned the wording adopted in very recent legislation concerning the completion of the gacaca proceedings in Rwanda. Under Article 8 of Organic Law No 04/2012/OL:
a person extradited to be tried in Rwanda and who has been sentenced by Gacaca Courts should be tried by a competent courthouse provided by this organic law. However, the decision of the Gacaca Courts shall first be nullified by that Court.
One of the respondents in this case, Emmanuel Nteziryayo, had been acquitted by a gacaca court and the UK High Court went on to find that Article 8 only referred to someone who had been sentenced, which the Court took to mean ‘and thus convicted’, leaving Nteziryayo’s acquittal in place and his extradition unlawful on the basis of double jeopardy.Footnote 71
In France, since 2014, the Cour de cassation (Criminal Chamber) has consistently held that there was an insufficiently precise and accessible definition of what constituted genocide and a crime against humanity under Rwandan law, at the time that the incriminating facts were committed.Footnote 72 This position is held despite the acknowledgment of the enactment of the 1948 Convention on the Prevention and Punishment of the Crime of Genocide and the 1968 Convention on the Non-Applicability of Statutory Limitations to War Crimes and Crimes against Humanity, which the Cour de cassation has found to have insufficient indication of the penalty prescribed for these offences.Footnote 73 This reasoning has been criticized but nonetheless appears to be well established within France. Unlike the UK concerns regarding the domestic legal framework, this does not leave any scope for the Rwandan government to amend the applicable law in order to facilitate extradition and these decisions have now been followed by domestic criminal trials for genocide in France against Pascal Simbikangwa, Octavien Ngenzi, and Tito Barahira.Footnote 74 In these extradition cases, the expressive content is not only about an on-going commitment to the universal wrong of international crimes, it is also an expression regarding fair trial in Rwanda. This expressive content differs in the cases concerning suspects’ immigration status.
The French courts have heard the most cases relating to Rwandan genocide suspects, with 20 cases concerning extradition, all of which have been denied, six cases relating to the suspects’ immigration status, and five domestic trials. In contrast to the extradition decisions, all six immigration-related cases have resulted in the revocation of refugee protection or the removal of residency permits. In addition, these immigration proceedings have moved towards offering a more positive reading of fair trial. In the cases before the National Court on the Right of Asylum (CNDA), the Court has stated that the Rwandan judicial system will provide a fair trial.Footnote 75 This is in contrast to a number of the initial investigative chamber decisions, denying extradition on fair trial concerns, including the standard and length of imprisonment and treatment of defence counsel. The concerns regarding defence counsel are mirrored in the UK extradition decision where the courts’ assessment of the standard and capacity of Rwandan defence counsel provided the central pillar for justifying the refusal to extradite.Footnote 76
Looking at these critical expressions on fair trial rights highlights the importance of taking simultaneous account of proceedings adjudicating extradition and immigration status. Immigration and refugee status proceedings cannot be neatly separated from the extradition proceedings, nor do they necessarily reinforce shared values. An expressivist analysis suggests that these proceedings are underpinned by different intentions, one oriented towards the types of people who ‘should’ be granted the right to stay in a country and the other towards an assessment of a foreign state’s criminal justice system. Those focusing on assessing a foreign jurisdiction are more comfortably critical than those focused on determining when a foreign national should legally ‘belong’ to the host state. As a result, the decisions are doing different types of communicative work. These differences in extradition and immigration proceedings are similarly apparent in cases where extradition to Rwanda has been granted.
4.1.2 Category 2 - Cases where extradition to Rwanda has been granted
Where extraditions have been granted the expressions of fair trial rights in Rwanda are more favourable; however, immigration and refugee-related proceedings have also simultaneously enabled an avoidance of any fair trial assessment altogether. In the dataset, immigration-related deportations to Rwanda started in 2005 but it was only in 2011 that the first extradition was granted, following Uwinkindi’s transfer from the ICTR. The Norwegian courts led on this issue, with the extradition of Charles Bandora. The Oslo District Court, relying heavily on the transfer decisions of the ICTR, found that ‘given the changes that Rwanda has made to its laws and legal system, and the guarantee that Rwanda has provided that Mr Bandora will be given a fair trial … there are no longer any grounds for rejecting the application’.Footnote 77 This was followed by a favourable decision handed down by the European Court of Human Rights (ECtHR).Footnote 78 In these decisions and the cases that followed in Denmark and Germany, the legislative framework in Rwanda was read favourably by the courts. Similarly, in the most recent decisions in the Netherlands the capacity of Rwandan defence council to offer sufficient expertise was deemed sufficient.Footnote 79
However, there remain some worrying inconsistencies regarding the decisions made under extradition law and those decided in the immigration courts. This is seen most strikingly in the Netherlands when comparing the deportation of Jean de Dieu Munyaneza with those of the more high-profile extraditions of Jean-Baptiste Mugimba and Jean Claude Iyamuremye. The differences between these three cases suggest that in the Dutch context, Rwandans whose extradition has not been requested but who have immigration proceedings initiated against them on the basis of allegations of involvement in the genocide are left the most legally vulnerable. A more detailed discussion of this case is also illustrative of the extensive and multi-faceted nature of these domestic legal proceedings.
Munyaneza’s residency permit was revoked on the basis of an allegation of his involvement in the genocide, noting his specific mention in an African Rights report. On 21 December 2012, the Court of The Hague upheld his appeal against this decision and his residency permit was reinstated. The case was then appealed to the Supreme Court and overturned, resulting in the final removal of his residency permit.Footnote 80 On 19 January 2015, the Rwandan NPPA sent an arrest warrant and indictment for international crimes to the Dutch Embassy in Kigali. On 13 February 2015, Interpol headquarters sent an international arrest warrant to Interpol in The Hague. The warrant reached the Ministry of Security and Justice on 17 March 2015, four days before Munyaneza’s scheduled deportation. The Ministry did not inform Munyaneza of the arrest warrant or indictment. He was deported on the basis of the immigration decision to Rwanda on 21 March 2015 and was arrested on arrival at Kigali International Airport. He then made an application from Rwanda to the Court of The Hague for a summary proceeding that would require his return transfer to the Netherlands. It was argued that his case should have involved a separate extradition proceeding. His application was rejected on the basis that at the time of deportation, despite the arrest warrants, there was no extradition order and that the state was acting in accordance with its original intention of removing Munyaneza from its territory following the revocation of his residency permit granting him the right to remain.Footnote 81 The intentions underpinning the Dutch immigration laws are made explicit in this decisions, the law is concerned with and prioritizes the removal of individuals from the country. It is this intention and how it differs from cases involving extradition or domestic trial that further supports the central claim of this article that plural expressions are at play in these decisions and must be made explicit.
In contrast to Munyaneza’s deportation, in the cases of Mugimba and Iyamuremye, where extradition orders had been issued, we see an initial extensive set of immigration proceedings resulting in the revocation of residency permits followed by extradition proceedings. In the immigration hearings for these two suspects, the assessment of the Rwandan domestic justice system is immediately and strongly favourable. The extradition proceedings show some of the disagreements on the legal framework and the availability of robust defence more starkly but ultimately allow extradition. In the Iyamuremye case, The Hague Court draws strongly on the decision by the ECtHR upholding the Swedish decision to extradite Sylvère Ahorugeze, finding that since 2009, Rwandan laws have changed, legal practice has improved, and Rwandan courts will act in accordance with the ECHR. However, on 27 November 2015, The Hague District Court reversed the decision, finding that the right to legal aid in genocide cases is insufficiently ensured in practice. This refusal to extradite was overturned on 5 July 2016Footnote 82 and finally a combined appeal from both Iyamuremye and Mugimba was rejected on 12 November 2016.Footnote 83 The two accused arrived in Rwanda on 13 November 2016 and are currently held in pre-trial detention. These two cases continue to be monitored by the Kenyan branch of the International Commission of Jurists who report to the Dutch Embassy in Kigali and are tried under the specialized ‘Transfer Law’, but the same cannot be said of the proceedings against Munyaneza. Once looked at alongside one another, it becomes increasingly clear that the immigration and extradition decision are doing different types of expressive work, both in terms of the audience they address, and the norms they are communicating.
4.1.3 Category 3 - Cases decided under immigration law
In countries where all of the proceedings happen through immigration law alone, without a separate extradition proceeding, fair trial assessments have either been favourable or bypassed altogether. In Canada, within the immigration proceedings themselves, there is an assessment of risks to the deportee. In the Seyoboka case, reviewing the decision by the Immigration and Refugee Board, the Federal Court offered the following account of fair trial in Rwanda:
there is no doubt that for many years, his life and human rights could not be guaranteed in Rwanda. But that is no longer the case. In the past few years, Rwanda has brought its justice system up to international standards, as acknowledged by the Appeals Chamber of the ICTR, and can safeguard Mr. Seyoboka’s right to a fair trial in his country of origin should he be prosecuted. It is now time for him to face his past actions, and let justice run its course.Footnote 84
In the USA, the domestic prosecutions for immigration fraud, followed by deportation, discussed at the outset this article, have enabled a bypassing of any discussion of fair trial. While three of the suspects applied for the withholding of removal and protection under the Convention Against Torture (CAT) on the grounds that, should they be deported to Rwanda, they would be subject to persecution and torture, in all instances these claims were rejected by the Board of Immigration.Footnote 85
What does this tell us? First, the contradictory positioning in the general scholarship on Rwanda, where the dominant concerns regarding the authoritarian trajectory of governance are coupled with an acknowledgement of the country’s significant investment in infrastructure, social welfare and governance institutions,Footnote 86 are reflected in the jarringly different legal decisions within and between different jurisdictions. Second, at least in part, immigration law is driving some of the expressive work here. The effort, not only to protect the authenticity of the refugee system, as generally recognized in the literature, but also to align with the wider immigration practices of excluding the ‘criminal immigrant’,Footnote 87 means that fair trial in Rwanda is read more favourably or simply bypassed in the immigration system. Bringing immigration and extradition decisions into the same frame shows that they are doing different types of expressive work. The assumption in the general international law literature that refugee law and international criminal law are generating shared values is less easily sustained when the former is enmeshed in domestic immigration regimes.
4.2 Immigration law and criminal migrants
Once immigration and extradition decisions are looked at together, the second major trend across the dataset is that challenges to an individual’s immigration status are generally being upheld, regardless of a country’s decision regarding extradition. Across the dataset, 35 cases address a suspect’s immigration status, 29 of these cases have found against the suspect,Footnote 88 resulting in the removal of residency permits, withdrawal of citizenships, denial of asylum applications or criminal prosecutions for immigration offenses. For cases where extradition has been denied, this leaves individuals in a legal limbo that has concerned a number of scholars writing on Article 1F(a) generallyFootnote 89 or it enables deportation without extradition proceedings as seen in the cases in the Netherlands.
What helps to explain these stark differences in the immigration-related cases? Drawing on expressivist insights and their potential to act as a regulatory principle of intentions, different drivers underpin the extradition and immigration-related decisions. Extradition law focuses on a state’s assessment of another state’s criminal justice system: the norms being generated and asserted in these Rwandan cases relate to a state looking outwards towards another state’s practice, often in the Global South. In contrast, immigration law expresses a state’s views on the types of people considered deserving of citizenship or refugee protection – in other words, looking inwards, asserting who does and does not belong. These types of refugee exclusions and deportations constitute only a very small fraction of the total rejections of refugee applications and deportations on the basis of criminal conduct; yet, as so explicit in the wider international criminal law scholarship, the expressive work matters. These cases provide legal expressions, amplified because of the focus on an international crime, about the types of ‘criminal migrants’ that different countries’ immigration systems seek and are able to keep out.
4.3 Post-genocide justice and political opposition
Finally, it is important to return our focus to Rwanda. The Rwandan survivor communities and their supporters living in the diaspora have played an important role in the sustained focus on genocide suspects living abroad.Footnote 90 However, the global reach of these proceedings has been largely driven by Rwandan government indictments that, as shown above, have fit into the immigration agendas of numerous states. It is important to ask why, from the Rwandan government’s perspective, this policy has been pursued. There is no doubt that one of these drivers is the need to keep the experience of the Rwandan genocide alive in the international sphere while signalling the importance of trying those accused of involvement in violence that was undertaken to destroy in whole or in part the Tutsi ethnic group. The importance of this must be given significant weight and is strongly evidenced through the involvement of diaspora survivor groups.
However, the use of expressivism as a means of regulating the use of international criminal law in immigration, extradition, and domestic criminal law, prompts some wider observations. Through a complex web of transnational co-operation, made up of policing, prosecution, and immigration services, the Rwandan government is successfully extending the punitive reach of the state well beyond its national borders. All of the cases in the dataset deal with members of the Rwandan diaspora and enable the expression of this transnational reach of Rwandan penal power. The potential impact of this penal power on the capacity of the diaspora to politically organize is particularly visible in the cases addressing immigration and refugee status undertaken in South Africa. These cases have focused on individuals connected with the predominantly Tutsi diaspora opposition group, the Rwandan National Congress (RNC).Footnote 91
All of Rwanda’s post-genocide justice processes have been criticized for being strategically used by the government to repress political opposition, particularly within Hutu communities.Footnote 92 In my own work, while acknowledging these critiques, I have emphasized that we must still engage with the wider set of objectives that these accountability processes have been understood to pursue. These include the clear need within Rwandan communities for information about the names of individuals who were killed during the genocide and those responsible for their deaths, coupled with the practical need to respond to the large-scale incarceration of suspected génocidaires that immediately followed the conflict.Footnote 93 However, as the immigration cases show, universal condemnation of genocide is not the only expressive work that this litigation achieves. As argued above, in a period where there is a general crisis narrative around border control, immigration, and refugees, these cases are doing domestic work in countries around the world. They are signalling who is being kept out by immigration and asylum processes while at the same time expressing the reach and penal power of the Rwandan state. With these cases occurring around the world, this final form of legal expression directed to the Rwandan diaspora must remain part of the discussion.
5. Conclusion
An analysis of the extradition, deportation, and domestic trials of individuals accused of involvement in the Rwandan genocide brings attention to the plural expressive work of international criminal law when it is incorporated into diverse domestic laws, including those concerned with individuals’ immigration and refugee status. While the focus in the international law literature has been on ensuring the interpretive compatibility between international criminal law and refugee law, concerns in domestic and transnational legal studies about the role of criminal law in immigration prompt a more cautious analysis.
A focus on cases concerning participation in the Rwandan genocide decided around the world illuminate some of the domestic legal entanglements and the national agendas at play. Drawing on an independently generated dataset of 122 cases concerning 102 individuals decided in 20 countries around the world, this article has argued that we need to make explicit the different types of expressive work that these cases perform. A close reading of the judgments decided through extradition proceedings, immigration proceedings or domestic criminal trials, shows that the decisions express a range of things to different audiences. In addition to the continued statement of the universal condemnation of the wrong of international crimes, they express a verdict on the criminal justice system in Rwanda. In this expressive domain immigration proceedings often precede extradition or domestic criminal trials, offering a more positive reading of fair trial rights in Rwanda or enabling the avoidance of an examination of fair trial altogether. In addition, across the dataset as a whole, immigration-related decisions are most commonly being decided against the individuals concerned. In doing so, these decisions do some expressive work within the country in which they are decided, signalling a nationally-oriented message that these are the types of ‘criminal migrants’ to be kept out. Finally, for Rwandan communities, these cases express both the crucial continued condemnation of the Rwandan genocide against the Tutsi and at the same time, they offer an expression of the penal reach of the Rwandan state into its diaspora communities, inhibiting their capacity to politically organize.
Overall, recognizing this expressive pluralism offers a means to interrogate the intentions underpinning the use of international criminal law in border control. This is particularly important with the future of international criminal law increasingly explained in terms of its domestic application and the value of trials and punishments at the international level repeatedly justified in terms of their expressivist role in generating and establishing new legal norms. Opening up international criminal law to its entanglements with immigration law, offers the possibility of being more circumspect and cautious about the domestic work international criminal law can do.
