Immigration trials and international crimes: Expressing justice and performing race

This article examines the performative collisions between the wrong of genocide and the invocation of this international crime as a means to secure carceral control of borders. Drawing on courtroom observations, legal transcripts and the media coverage of an immigration trial in the United States, the article explores the performative relationship between international criminal law and immigration law. It argues that this relationship helped to construct and racialize the category of the ‘criminalized migrant’ while establishing the perceived ‘civility’ of criminal law as a primary means of enacting domestic border control. While race was never made explicit in the trial, it emerged in a fractured but significant way, as the horror of the Rwandan genocide against the Tutsi reinforced the wrong of violating immigration law.


Introduction
It is 11 March 2019 and I'm sitting in the United States District Court in Boston, Massachusetts. It's a criminal trial. The family of the accused are seated to my right, well dressed, huddled together. They are the only black people in the room, with the exception of one of the 12 jurors. Suits, white hair and leather law books order the spacious courtroom. The charge is immigration fraud and perjury with a potential prison sentence of up to eight years. The jurors are listening intently to the white academic on the stand, an expert witness brought by the prosecution to explain the context of the 1994 genocide against the Tutsi in Rwanda. The central issue for the court is whether the accused Jean Léonard Teganya lied about his alleged involvement in the genocide, in his application for asylum. This municipal prosecution for immigration offences based on allegations of involvement in an international crime fits into an emerging pattern around the world (Bolhuis et al., 2014;Rikhof, 2017). Focusing on cases against Rwandan nationals in the US draws attention to seven similar decisions. Four individuals have been deported to Rwanda after completing their sentences, and three others, namely Prudence Kantengwa, Beatrice Munyenyezi and Gervais Ngombwa, are still serving time in US jails for immigration offences with sentences of up to 15 years' imprisonment. Looking beyond the US, an independently generated dataset (Palmer, 2020) shows at least 120 cases undertaken in 20 countries that have addressed the immigration status, extradition, deportation or trial of individuals suspected of involvement in the Rwandan genocide. These have included the high-profile cases of genocide suspects Léon Mugesera and Jean-Claude Henri Seyoboka deported from Canada to Rwanda and, among others, the extraditions of Charles Bandora from Norway and Jean-Baptiste Mugimba and Jean Claude Iyamuremye from the Netherlands. Thirty-six of these cases have been decided on the basis of immigration law. Teganya's case is thus one among many. It is part of an expanding set of legal activities and offers the opportunity for a close reading of what is at stake when international criminal law (ICL) is re-nationalized into domestic immigration law. To do so, this article deploys performative theory, to examine how ICL and immigration law worked together to further animate the 'criminalized migrant' as the central subject of border control (Bowling and Westenra, 2020: 164).
Drawing on courtroom observations, a close reading of the legal transcripts and the media coverage of the Teganya trial, this article examines the collisions between expressing the wrong of genocide and the invocation of this international crime as a means to secure carceral control of borders. It shows how the language and legal categories used in the trial helped to bring into being the racialized 'criminal migrant'. The article argues that the legal relationships between ICL and immigration law can be usefully understood as performative (Butler, 1997a(Butler, , 1999. Together these legal regimes reinforce and communicate particular migrant identities that are infused with racially informed notions of who violates ICL law and where these types of crimes are committed.
The article is structured in three parts. It opens with a methodological reflection on writing on race as a white scholar, its necessity and its necessary constraints. In doing so, this section highlights how, despite many years of collaborating with Rwandan colleagues working and writing on the country, my research and access to that Boston courtroom are still racially inflected. This recognition establishes the need to make visible the relevance of race while at the same time deploying the resultant analysis to pursue an explicitly anti-racist agenda.
The second part draws on performative theory to examine the collisions and compatibilities between ICL and immigration law in Teganya's case. It shows how immigration offences offer a route through which international crimes are rendered legally relevant in the US and highlights how Teganya's alleged involvement in the 1994 genocide against the Tutsi is given centre stage in the trial. It argues that the performative collision between ICL and immigration law gives rise to a legal relationship that renders a severe punishment for an immigration offence seemingly lenient.
The third part then draws on critical writings on race in ICL to illuminate how this performative relationship reinforces and enhances the increasingly racialized construction of the 'criminalized migrant'. It shows how race emerged in a fractured but significant way in the trial, as the horror of the genocide in Rwanda helped to reinforce the wrong of violating immigration law. This horror is intentionally distanced from the Boston courtroom in which it was being assessed. It is infused with racialized stereotypes of African violence and poverty that work together to establish the perceived civility of criminal law as a primary means of enacting domestic border control. In conclusion the article reflects on how these performative relationships prompt some radical thinking, drawn from the Black Lives Matter (BLM) movement, on what it might mean to defund international criminal justice.

Writing on race as a white scholar
Self-reflexive ethnographers and sociologists have long alerted us to the co-construction of knowledge through the inter-subjective relations between observing and observed subjects (Bourdieu, 1988;Das, 1998). In recognizing the position of the researcher in relation to their subject of study, Bourdieu reminds us that such reflexivity is not an end in itself but rather 'genuinely aims at improving this practice' (Bourdieu, 1988: 775). In this article, I will pay particular attention to the subjective relationships at play in that Boston courtroom, namely the legal relationships between immigration law and ICL and the social relationships among the people present. Describing this legal and social space in which Teganya's trial occurred offers a means to interrogate the construction of race in a courtroom that was so visibly white. It alerts us to the racial entanglements and limitations placed on the relationships between people in a given 'social situation' (Gluckman, 1940: 10). As a starting point, such self-reflexivity prompts a more explicit recognition of the place of race in my own inter-subjective connections.
On that particular day in the Boston courtroom, I immediately recognize one of the suited men, as he acted as a lead defence counsel on a case on which I worked at the United Nations International Criminal Tribunal for Rwanda (ICTR). The professional circles of lawyers, law enforcement officers and expert witnesses still involved in litigation concerning the Rwandan genocide are small and well known. I am married to the expert witness on the stand and it is this white relational tie that first opened up my access to that court and its legal actors. As a result, some of the courtroom connections are immediately visible to me but others less so, seen obliquely or oddly ascribed. I recognize the presence of the tight ties among the three Rwandan women who were hurriedly introduced to my husband and me at the entrance to the courthouse. They had travelled from Kigali to Boston to stand as both prosecution witnesses and genocide survivors. Their testimony in the days to come is harrowing and raw.
I am aware that the two blue-suited men who walked in quietly and sat at the back of the room halfway through the day's testimony are Rwandan government officials. These proceedings, occurring around the world, are often initiated by Rwandan-issued arrest warrants and are hugely significant to the current Rwandan regime. They offer a crucial route to keep the experience of the 1994 genocide against the Tutsi alive in the international sphere and at the same time play a key role is disciplining the Rwandan diaspora, with some individuals accused of involvement in international crimes also fulfilling very prominent roles in diaspora political opposition groups (Betts and Jones, 2016). Yet, while visible, it is less easy to access the role of these relational ties in constructing knowledge about Rwanda and its genocide in that courtroom on the east coast of America.
My relationships and my research, to use Toni Morrison's (1992: 11) term, are racially inflected. In recognizing this, my first impulse as a white South African who grew up during the last decade of apartheid is to assert my commitment to establishing racially equal research relations in Rwanda. For over 14 years, I have worked to establish ongoing reciprocal ties with Rwandan researchers and study participants. At best, these ties are underpinned by a transparent and fair distribution of research funds and should, to draw on Tshepo Madlingozi's (2010: 226) challenge, be informed by both the theory and experience of African scholarship. Yet my efforts in this regard, while ongoing, are imperfect and the racialized inequalities are real. Despite continuing to work closely with Rwandan colleagues, writing on the country and building a range of hugely valued personal and professional connections, at this juncture I still have greater access to the white ordering of that Boston courtroom. I can't escape my own whiteness.
If I am to write explicitly about race, it requires a necessary reckoning with whiteness. South African theorist Samantha Vice (2010) argues that the most morally appropriate response to white privilege is to feel ashamed and that the actions that should follow that shame, in post-apartheid South Africa, involve listening more, speaking less and stepping back so that others may step forward. I agree with this call to epistemic and public humility. Yet in the white space of a Boston courtroom, Steve Biko's challenge to the white South African 'liberal' in 1978 returns to me. For Biko (2004Biko ( [1978 is: 27), the job of this 'liberal' was to challenge white racism by applying himself 'with absolute dedication to the idea of educating his white brothers . . . that we may live in "a country where colour will not serve to put a man in a box"'. I do not presume, or perhaps I am not brave enough to pursue a didactic function here, but I will attempt to explicitly analyse how race both enabled my access to that Boston courtroom and informed how people made sense of the immigration trial that occurred in that space.
My aim then is to make race explicit, informed by the reflexive reality that to oppose racism one must necessarily recognize how one wields that racist power (Butler, 1997a: 17). Pursuing this aim starts with this account of my own white relational ties and a recognition of the accounts to which I have less access. In this endeavour I am building on the wider writings in law and anthropology that both recognize the 'othering' work of law and the need to turn the western ethnographic gaze to its own ways of knowing the social world (Abu-Lugod, 2013;Goodale, 2017). It was my husband's involvement in the case that helped to secure my access to the courtroom, opening up informal discussions with the lawyers and Immigration and Customs Enforcement (ICE) officers involved in the case. This would later lead to the provision of the legal transcripts of the sentencing hearing. As previous studies on immigration have shown (Kalir et al., 2019) understanding the routes through which we can access people and institutions that exercise control over others tells us something about how these institutions work. Making explicit my own racial positioning in that courtroom that felt so unavoidably white, informs what conversations I had access to and how I analysed the empirical material presented in this article.
There is, of course, another account of what occurred in that Boston courtroom that focuses on the actions and agencies of the Rwandans in the room. These are sets of relationships I had less immediate access to. This account could tell of the dedication of the survivors and their support networks to pursue accountability around the world for the horrors of the genocide (Ndahinda, 2011). This includes the bravery required to testify publicly to genocide and to sexual assaults (Mukagasana, 2019) and the perlocutionary acts that were instrumental in enabling them to speak in that courtroom (Butler, 1997a: 44). It could explain the role of the Rwandan state in pursing these cases and deploying criminal law, domestically (Chakravarty, 2016;Ingelaere, 2016) and now through transnational penal networks, to constrain political opposition. This could include looking more closely at how Rwandan witnesses drew on difference as a resistive mechanism against Rwandan state power, as seen fleetingly in the reported words of one of the witnesses: Being here in America it's very hard to understand that it's a very different system . . . here, we expect people to hold their leaders accountable. In Rwanda, it's the other way around. The leaders hold people accountable so people have to comply with what the government wants them to accomplish. (Cramer, 2019c) These are important accounts that speak to the potential for performative resistance in the trial (Ba, 2015;Modiri, 2015). However, they are not the accounts that are told by the predominantly white people I observed and read arguing, adjudicating and reporting on Teganya's actions, the white bodies around which the world of that courtroom cohered (Ahmed, 2009: 41). These white accounts are the focus of this article, drawn from my own uncomfortable positioning in that courtroom. In adopting this focus, the words participants deployed and the sets of performative relationships established between international criminal law and immigration law make visible the racialized work of these trials.

The performative collision of immigration and international criminal law in the trial of Jean Léonard Teganya
Jean Léonard Teganya has been entangled in a range of cross-border immigration-related proceedings for many years. In 1994, he was a medical student and intern at Butare University Hospital in southern Rwanda. He is accused of having directly taken part in atrocities against Tutsi civilians at the hospital. In the wake of the genocide, he fled Rwanda, first to the Democratic Republic of Congo (DRC), then Kenya and India. In 1999, he arrived in Canada, where he met his wife, started a family and twice unsuccessfully sought asylum.
In 2014, Teganya crossed into the US, entering through Houlton, Maine in an effort to evade a Canadian order of deportation to Rwanda. There he was taken into custody by US Customs and Border Control officers and once again requested asylum. In the case against him before the District Court in Boston, it was alleged that on the application for Asylum and Withholding of Removal, he made false statements by failing to disclose both his affiliations and activities with the Mouvement révolutionnaire pour le développement (MRND), the political party in power in Rwanda before and during the genocide and his direct involvement in perpetrating the genocidal violence aimed at destroying, in whole or in part, the Tutsi people as a recognized ethnic group.
It is the allegation of Teganya's involvement in an international crime that distinguished it from the usual workload of the US Attorney's Office in the District of Massachusetts and ICE's Homeland Security Investigations (HSI). As the Public Affairs Specialist in the US Attorney's Office said to me over the phone as I negotiated access to interviewing the legal participants in the trial and reviewing the transcripts, 'We do want to facilitate access especially if it enables activism on these types of cases.' This idea of activism taps into a rich history of human rights advocacy for criminal trials for genocide, war crimes and crimes against humanity (Sikkink, 2011), yet the charge in this case was one of immigration fraud. In the realm of border control, far from a notion of advocating for criminal trials, there has been sustained critique of the growing use of criminal law to secure state borders (Aliverti, 2013;Stumpf, 2006). This article draws on performative theory to explore this tension at the heart of Teganya's trial, endeavouring to unpick the relationship between ICL and immigration law. In doing so it highlights a continual shift in the trial between the charge of an immigration offence and evidence of Teganya's direct participation in the genocide. As discussed below, the outcome of this performative relationship is that it makes a severe punishment for an immigration offence seem lenient while tapping into deeply racialized ideas about where and by whom international crimes are committed. At the time of writing, Teganya's case is on appeal and therefore the legal actors involved were not willing to offer public commentary. They did, however, facilitate full access to the transcripts of the sentencing hearing. This text, along with the media coverage of the case and my courtroom observations, allows for an examination of how the wrong of genocide is included in an immigration offence and the resultant performative effects.
Performative theory, building on the pioneering linguistic work of JL Austin, draws our attention to how language can consummate an action. In the legal field this is often invoked through the example of a judge pronouncing a verdict or handing down a sentence. These linguistic insights into speech as an act have been extended through examining how language constitutes and constructs a sense of self in relation to others (Butler, 1999;Hall, 1996). Anthropologists have usefully deployed this literature to show how performative claims are often established in relation to the performativity and subjectivity of claims made by others (Mai, 2018). Bringing these insights to bear on the relationship between ICL and immigration law shows how the language used to describe and to enact Teganya's trial established a performative relationship between these two legal realms. In doing so it enabled and drew on a language that racialized the 'criminalized migrant' as the legitimate subject of border control, creating a 'constitutive constraint' (Butler, 2007b: xi) on who migrants are. To examine this performative relationship requires a close analysis of the relevance of the crime of genocide in this immigration trial.
The opening of Teganya's trial was reported in the Boston Globe in the following terms: 'The defendant had a problem,' Assistant US Attorney Scott L. Garland told a jury in US District Court in Boston . . . 'His problem was that his application for asylum would be denied if the US found out what he had done in Rwanda, because persecutors cannot claim asylum.' (Cramer, 2019a) The notion of being a persecutor is drawn from the Immigration and Nationality Act 1952, section 208, in which an individual can be excluded from asylum protection in the US 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'. The standard of proof required to meet this 'persecutor's bar' is that there is sufficient evidence if it 'raises the inference' (Alvarado v. Gonzales, 449 F.3d 915, 930 (9th Cir. 2006)) of participation in persecution.
Teganya was charged under section 1546 of the US Code with fraud and misuse of visas, permits and other documents and under section 1621 with perjury on the basis that he knowingly provided false statements material to the consideration of whether he should be excluded from refugee protection on the basis of his participation in persecution. In this way, the international crime of genocide-defined in the Genocide Convention of 1951 as a series of designated acts to be committed 'with the intent to destroy, in whole or in part, a national, ethnical, racial or religious group, as such'-is incorporated into a trial of immigration fraud.
In the trial, in bringing together immigration law and ICL, the violence of the Rwandan genocide was given centre stage. The testimony, like the many trials for direct participation in the violence heard before the ICTR, the Rwandan national courts and the localized gacaca courts, was absolutely harrowing. The media offered the following account: A Rwandan woman described the brutal attacks by Hutus against two young Tutsi children and their mother, who had been dragged outside the hospital where they had sought refuge from the conflict. 'The mom was asking "Please, at least leave me with one child,"' said Isabelle Mukankusi, who was 14 and a patient at the hospital at the time. All three were killed, she said. (Cramer, 2019b) This incident was not directly connected to Teganya but was drawn on to set the stage of the violence in Butare hospital in which his conduct was to be understood. The two witnesses who followed Ms Mukankusi were given the most significance by the prosecution in their later arguments. One witness testified that Teganya raped her while six months pregnant and facilitated the rape and murder of her cousin. The next witness stated that Teganya delivered her to a group of men who took her to a field where she was repeatedly raped. One of these women is now in her late 60s and the media coverage recounted how she cried as she recollected the murder of her closest family members in those horrendous 100 days following the downing of President Habyarimana's plane on 6 April 1994. Teganya was not always directly connected with the violence; it was sometimes perpetrated by individuals the witnesses saw as his associates. The horror is breath-taking, defining for those who have lived through it, hard to listen to or to read. This violence, however, was also inextricably enmeshed with the wrongs associated with crossing borders and disclosing or failing to disclose information in asylum applications.
While the charge of genocide formed the bulk of the testimony against Teganya, immigration laws performed various types of background work in the trial. For example, discrepancies on a passport were used to challenge the reliability of a defence witness. On the stand the witness was challenged about her nationality and admitted that she lied on her passport when stating she was born in Burundi and not in Rwanda. In a side room to the court, one of the ICE officials casually told my husband and me that witnesses who come forward to testify in these trials often find themselves facing immigration-related charges when they leave.
While Teganya's alleged conduct at the hospital in Butare remained the focus of the prosecution witnesses' testimony, the illegality of the border crossing still emerged as significant. The Boston Globe highlights how the jury handed down a guilty sentence for immigration fraud and perjury after less than three hours of deliberation. Following this conviction and prior to Teganya's sentencing, the wrong associated with the border crossing informed one of the juror's assessments. Brian Ross, a 58-year-old recruiter from the South Shore, told local journalists that he was swayed to convict Teganya on two grounds: first that the defendant's testimony that he had not seen bodies of Tutsi victims while working in the hospital seemed implausible; and, second, that, when he entered Canada with a false passport, he told immigration authorities he was desperate not to return to his native country: 'That was big . . . him saying he would do anything not to go back to Rwanda', Ross said (Cramer, 2019d). What is key here is that the performative claims of the wrong of genocide help to establish the wrong of an immigration offence. This performative relationship plays out in the most significant manner in the sentencing decision. The government requested a sentence of 20 years' imprisonment, more than double the sentencing guideline for immigration fraud which is set between 78 and 97 months.
In the words of Assistant US Attorney Scott L Garland: The government's argument for 20 years of imprisonment begins and is underpinned largely by the enormity of the genocide, the enormity of Mr. Teganya's part in the genocide and the enormity of his lies about what he did then and its effect upon the asylum system . . . putting oneself back in May of 1994, playing that out as the refugees are coming in live essentially, while the genocide goes on, I think it's much, much easier to understand how the lies that Mr. Teganya told, his dressing himself up in the garb of the persecuted rather than the persecutor, how awful they were. They weren't just the heartland lies that might be told in immigration court for which we still prosecute people, lies about former marriages, lies about current marriages, lies about who they're related to, presenting false identity documents, but these were instead lies about the most consequential catastrophe that Rwanda has faced. The two wrongs-genocide as an international crime and immigration offences-are thus inextricably linked. The severity of lying about participation in genocidal violence helps to reinforce the prosecution of those who lie to immigration officials about their marital status or their nationality. The judge, in handing down his sentence, drew further attention to the relationship between these two areas of law: The jury convicted him of immigration offenses and perjury. There was testimony that he participated in multiple murders and rapes, and that was the subject of the immigration fraud and the perjury. He was not charged in this court with murder or rape and could not be so charged. Obviously, the jury did not specifically find that he committed or participated in any particular murder or particular rape. They were not asked to and could not legally have been asked to. There is, I think, substantial ambiguity in the verdict as to what he actually did for that reason, but the basic question is do I sentence him as a liar or do I sentence him as a murderer or rapist or genocide participant? (Sentencing Hearing Transcript, 2019: 26) In answering this question, the judge sentenced him as a liar, imposing the maximum sentence of 97 months in jail and staying within the guidelines for a conviction for immigration fraud. However, in doing so, the severity of the wrong of genocide suddenly rendered the heavy sentence of eight years' imprisonment for failing to disclose information in an asylum application seemingly lenient, reinforcing the use of criminal law as a means of sanctioning the provision of false information in an asylum application process. The severity of an eight-year prison sentence for immigration fraud is particularly visible when looked at in light of sentences handed down for direct participation in the Rwandan genocide. In a comparative study of the sentencing practice for those convicted for genocide before the ICTR, the Rwandan national courts and gacaca, the median sentence for the most senior suspects at the ICTR was 25 years, followed by 14 years in the domestic courts' trials and 19 years in the gacaca courts. Looking within the range of Teganya's sentence, that of six to 10 years' imprisonment, 10% of trials with determinate sentences given by both the ICTR and the gacaca courts for those deemed 'the most responsible' for the genocide fell within this range. In the national courts over 71% of cases with determinate sentences fell within the 6-10 year range, although in focusing on determinate sentences this excludes the high number of early death penalty sentences imposed in the immediate wake of the genocide, most of which were later commuted to life imprisonment (Hola and Nyseth Brehm, 2016: 72-75).
Overall, what this shows is that Teganya's sentence of eight years for immigration fraud and perjury is not wholly dissimilar to those being handed down to individuals convicted for participating as senior leaders in the genocide. Nonetheless, the judge saw himself as sentencing Teganya as a liar not as a murderer or rapist or génocidaire. Yet, these underlying offences informed the case and justified both the prosecution of, and the heavy sentence for, lying on immigration forms. The focus on the allegation of involvement in an international crime obscured how fraud, in relation to documents that enable mobility, is of such central importance because we live 'in an age in which profit rests ever more on the ability to control the long-distance migration of people and things' (Comaroff and Comaroff, 2007: 137). ICL is playing its part in determining who gets to benefit and who carries the costs of these profits. This is further evident in the media coverage of the sentencing in this case, where the prosecutor's allegations of involvement in the genocide provided the entry point. The Associated Press gave the following account: A Rwandan man convicted of hiding his involvement in the country's 1994 genocide in an attempt to win asylum in the U.S. was sentenced Monday to more than eight years in prison. Jean Leonard Teganya participated in at least seven murders and five rapes during the genocide, in which Hutu extremists slaughtered Tutsis and Hutus who tried to protect them, prosecutors said. (Durkin Richer, 2019) It is the allegation of his involvement in the genocide that is being drawn on to explain the sentence. Yet the performative relationship between ICL and immigration law went further than this. The critical writings on race in both ICL and immigration law help to explain how the activities in that Boston courtroom drew on stereotyped notions of both African violence and poverty. In doing so, this language aided in constructing the increasingly racialized category of the 'criminalized migrant' and the 'civility' of criminal law as a primary means of border control. To return to Butler (1997a: 1), ICL offered the 'socially sanctioned forms of address' that helped to established the particular wrong of immigration fraud and infused the category of the 'criminalized migrant' with racialized assumptions of where international crimes are committed and by whom.

The place of race in the trial of Jean Léonard Teganya
Critical studies in ICL have been increasingly characterized by a concern with uncovering who benefits and who carries the costs of the moral, epistemological and empirical assumptions underpinning this area of legal practice (Schwöbel, 2014a). Coupled with empirical and disciplinary contributions from sociologists, anthropologists, criminologists and political scientists, to name a few, this work has shown that ICL was not something 'born' at Nuremburg in the wake of the Second World War (Haslam, 2020;Nesiah, 2016). Instead, it has been painstakingly constructed through the interests of parties at treaty negotiations (Tallgren, 1999) and through case law and interpretation. It has been built on social practices (Kelsall, 2010), emergent common cultural interpretations of legal action (Wilson, 2012), political alliances and power differentials (Nouwen and Werner, 2011), markets of legal expertise (Christensen, 2020), ambitions and coalitions of non-governmental organizations (NGOs) (Lohne, 2019) and through branding and selling (Schwöbel, 2014b). Among these diverse and at times divergent sets of drivers, race plays an important role, both as the basis for the crime of genocide and persecution as a crime against humanity and increasingly as a basis for critique of the current legal practice (DeFalco and Mégret, 2019: 60). Looking at these writings on the place of race in ICL more generally helps to illuminate how the performative relationship between ICL and immigration law, visible in Teganya's trial, brings into being a racialized understanding of who violates immigration law.
The scholarship of Mahmood Mamdani and Makau wa Mutua was pioneering in drawing attention to the intimate relationship between law and race in the fields of human rights and international criminal law. In 2001, Mutua (2001: 207) argued that human rights is marked by a tripartite metaphor of savage, victim and saviour that 'carries racial connotations in which the international hierarchy of race and color is re-entrenched and revitalized'. Matua's insights are particularly marked in ICL. The ideal victim is innocent, vulnerable and generally female, preferably a child or child-like and black-and she is marketed as such. The perpetrator, embodying the racialized category of Matua's 'savage culture', is personified in the black and Arab leaders on trial before the International Criminal Court (ICC) in The Hague. ICL then offers the route to white salvation for both these victims and their 'savage' community (Anders, 2011). This maps onto Mamdani's (2009) critique of the representations of violence in Darfur, where he argues that particular notions of the Arab perpetrator, in the minds of largely American lobby groups, informed their support for both military and criminal legal interventions in Sudan. In the wider international criminal law literature, these concerns are echoed and recast in the increasing attention to the role of international criminal trials in constructing idealized and racialized notions of both perpetrators and victims (Clarke, 2011(Clarke, , 2015Kendall and Nouwen, 2013;Sagan, 2010;Schwöbel-Patel, 2018).
Drawing on Matua's work, Robert Knox (2013) has highlighted the specific racializing work of international law more generally, where both settler practices of colonialism and, in contemporary settings, the competition among imperial powers, explain how the development of international law draws explicitly on racialized categories. The binary logic of this racialized 'othering' creates a bright legal line between the 'civilized' and the 'uncivilized'. As Antony Anghie (2005: 4) crucially shows, international law then provides the civilizing bridge, offering the legal 'techniques to normalize the aberrant society'. This is again strongly reflected in ICL, where neo-colonial critiques of the prosecutions targeting African states are responded to by highlighting the number of African states that are signatories to the ICC's Rome Statute and how the situations under investigation in Uganda, the DRC, the Central African Republic (CAR) and Mali were initiated through referrals from these states (Bassiouni and Hansen, 2013). This response is entirely consistent with how international law helps to construct racialized categories then immediately offers the legal techniques through which African states can normalize their 'aberrant societies'. This is not to negate the specific political and legal drivers for these referrals or the relationship between particular African states and the ICC (Clark, 2018). Rather, it is to draw attention to the ways through which the current practice of ICL frames racialized understandings of where and by whom these types of crimes are committed and offers criminal law as the means of redemption.
In Teganya's trial, this heady mix of international crimes, law and race is brought into the sphere of immigration law, where criminal offences are increasingly being used to establish border control. Teganya's trial operates at the intersection of the US engagement with ICL and immigration law-a setting in which immigration-related prosecutions currently form the bulk of federal criminal prosecutions, outnumbering drugs and weapons prosecutions (Chacón, 2012). The host of new criminal offences for immigration violations, of which Teganya's charge of immigration fraud is one among many, has given rise to the notion of 'crimmigration law', capturing the interweaving of administrative and criminal law in responding to the increasing movement of people around the world (Stumpf, 2006). The role of these legal proceedings in constructing a racialized notion of who is criminalized for crossing borders is now being given much needed scholarly attention (Bosworth et al., 2018). At the same time, in the sphere of ICL, the US has been very active in building international criminal justice when applicable to foreign nationals abroad (Kaufman, 2016) while at the same time deeply resistant to its application to American citizens (McGonigle Leyh, 2020), particularly to its mostly white service members and government officials.
The invocation of international crimes as a means of excluding individuals from refugee protection and the resultant prosecution for immigration fraud is entirely consistent with these concurrent developments. International crimes are rendered relevant to immigration offences, offering a means of determining who is on the path to being considered American. At the same time these immigration offences reinforce the notion that international crimes are best understood as those committed by foreigners beyond US borders, while sidestepping any pressure to domestically prosecute individuals for international crimes, potentially including US service members.
These dynamics were all in play in that Boston courtroom as the racializing work of the proceedings emerged in a fractured but significant way, hewing to ideas of African violence and poverty in creating the social category of the 'criminalized migrant'. To return to that first day of the trial, as I sat listening to the expert testimony on the nature of violence during the genocide, the judge explained to the jury that such testimony was intended only to provide background, to create context. In his words, 'It doesn't tell you anything about whether a particular individual is or is not telling the truth.' He suggested that this type of testimony would be entirely unnecessary if the court had to 'address an issue here in America'. The judge endeavoured to frame the proceedings in terms he believed the jurors would find intelligible. To do so, the distance and distinctiveness of the violence in Rwanda was emphasized, drawing on familiar stereotyped motifs of African violence and poverty-motifs that, as discussed above, are reinforced by the wider practice of prosecuting genocide, war crimes and crimes against humanity at the international level.
The centrality of the assumptions about African poverty and the distanced nature of the Rwandan violence became particularly evident in the assessment of witness credibility. Here a concern about the veracity of the Rwandan witness testimony returned repeatedly to the significance of the $71 per diem for food allowance and the $40 daily reimbursement paid by the US government to their witnesses who travelled to give testimony. Teganya's defence argued that these payments would incentivize and distort witnesses' testimony. The same was not asked of the white expert witness heard only a few days before. An assumption about the Rwandan witnesses' socio-economic status was drawn on to challenge the veracity of their account.
Concerns over the veracity of the Rwandan witnesses' accounts in the final sentencing decision also invoked the distance of this trial from Rwanda and the exceptionalism of the violence associated with international crimes, committed outside the US. In the words of the presiding judge: Virtually every atrocity that was described by witnesses in this trial was supported by the testimony of a single witness without much in the way of corroboration, and at least some of the witnesses had some credibility issues, at least as to some aspects of their testimony. It's not to say I don't believe the testimony taken as a whole, but my confidence level is not as high as it might be in, let's say, a murder trial in a modern American court, let's put it that way. (Sentencing Hearing Transcript, 2019: 27) Just as the critical ICL literature alerts us to how this body of law helps to establish racialized categories of black victims and perpetrators, whose salvation lies in civilized criminal processes, turning to this literature helps to explain how the racialized and distanced violence of genocide establishes the civility of criminal law as a primary means of establishing domestic border control. The charge of involvement in an international crime introduces particularly heinous violence associated with actions outside of the US-in Africa-and thus justifies and reinforces the importance of the use of penal sanction in securing American borders. It is the 'otherness' of that 'African violence' that draws on the racializing work of ICL, to place Teganya as the 'criminalized migrant' in his interpretive box. This occurs in a wider context of anti-immigrant sentiment in which, as Achille Mbembe (2017: 24) reminds us, 'entire categories of the population are indexed and subject to various forms of racial categorization that transform immigrant (legal or illegal) into an essential category of difference'. This article has endeavoured to show the role of ICL in this performative endeavour.
The wrong of the genocide becomes the way of understanding the wrong of violating immigration law. This relationship brings into being a hugely powerful image of who these 'criminal migrants' are and the need to locate their actions beyond the US. The trial proceedings made use of racialized understandings of African violence and poverty that obscure the reality of the visible racial dimension in that Boston courtroom, which coheres with the US's own racialized criminal justice system (Owusu-Bempah, 2017): a criminal justice system increasingly preoccupied with prosecuting immigrants. The court's focus is clearly on the exceptionality of the violence 'over there'. The outcome is that when ICL is re-nationalized through domestic immigration offences, it both enables and reanimates a racialized notion of the types of people who commit immigration offences and ties their conduct to the most egregious international crimes.

Conclusion
In this article, performative theory has been drawn on to show the ways in which the language of ICL, when incorporated into immigration trials, is used to justify severe penal sanctions for immigration offences. Bringing ICL into the realm of immigration law helps to construct an increasingly racialized social category of difference, 'the criminalized migrant'. Yet to return to the methodological challenge raised at the start of this article, prompted by my own positioning in that Boston courtroom, what is the wider significance of making race explicit for the practice of ICL and immigration law?
While there is increasing recognition among scholars of the role ICL in constructing racialized ideas around who commits international crimes and where they occur, there is very little recognition of these critical insights within the legal institutions enacting this law. Most recently, as Owiso (2020) has highlighted on social media, in the much anticipated 378-page independent Expert Review Report of the International Criminal Court (ICC) chaired by Justice Richard Goldstone, the words 'race' and 'racism' are never mentioned. In endeavouring to explain the absence of any racial reckoning by the ICC, DeFalco and Mégret (2019: 60) have drawn on critical race theory to argue that '"racism" has been treated as something external to ICL itself, a phenomenon that tribunals are called on to judge and ultimately condemn, rather than something that they might be implicated in'. They suggest that this has particular effects on those practitioners working inside of these institutions, suggesting that 'international criminal justice actors see themselves as all the more innocent of racism because they are engaged in the prosecution of the worst and most pathological excesses of global racial politics' (2019: 61). A performative assessment of how ICL is renationalized into immigration offences furthers this critique by drawing attention to how the racializing role of ICL is reanimated in the particular context of carceral immigration control. This process of making race explicit in both the research and practice of ICL at the international and national levels offers a new entry point to pursuing an anti-racist agenda.
In 2020, it was the BLM movement, initiated in the US and globally resonant from France to Nigeria, that has given particular prominence to the racializing work of criminal justice. The BLM movement's proposed reform to systemic racism in criminal justice is to 'defund the police'. It is a call for the re-ordering of resources away from penal coercion and towards social responses (Loader, 2020). In the realm of transnational crimes such as drug trafficking or human trafficking, this could take the form of a public health response to drug abuse (Hughes and Stevens, 2010) or a labour law response to exploitative work practices (Shamir, 2012). In the ICL domain, the anti-impunity push is so strong that alternatives to criminal law are seldom articulated. While there is an emerging critique of the use of criminal law as a means of realizing human rights (Engle, 2015;Pinto, 2020), the way forward is less clear. There are arguments for more domestically and regionally based courts, for the legal recognition of both truth commissions and localized forms of accountability for international crimes or, in the context of border control, that border regimes themselves might constitute a crime against humanity (Kalpouzos and Mann, 2015;Mann, 2020). Yet, if the racializing work of ICL and its renationalized forms in immigration regimes is to be taken seriously, the BLM movement may have a lot to offer ICL. It can prompt the question of what might defunding international criminal justice look like? Part of the answer might be that we start to think about responding materially to large scale violence through reallocating resources to violence-affected populations and actively enabling their movement across borders.

Funding
The author(s) received no financial support for the research, authorship and/or publication of this article.