ChapterPDF Available

African Languages in International Criminal Justice: the International Criminal Tribunal for Rwanda and Beyond



This paper examines the various roles played by African languages, along with the practices and policies that were developed to respond to these roles, at the ICTR and SCSL, and, to a lesser extent, the ICC. Through interviewing judges, prosecutors, defense counsel, investigators, outreach officers, and language service specialists from the three courts, as well as exploring existing literature that touches on language use in these settings, the author attempts to draw a picture of how African languages fare in critical domains of international criminal justice. These domains include formal criminal investigations, courtroom proceedings, outreach to the regions where the international crimes have been committed, and communication with victims of these crimes.
African Languages in International Criminal Justice:
The International Criminal Tribunal for Rwanda and Beyond
By Leigh Swigart1
1. Introduction
In a 2013 lecture, renowned Kenyan author Ngugi wa Thiong’o, a long-time promoter of
the expanded use of African languages and author of literary works in his native Kikuyu,
bemoaned the absence of local languages in the justice systems of African countries. 2
Ngugi pinpoints a challenge common to countries across the African continent. A 2010
UNESCO publication reports that only 63 African languages (out of an estimated 1,000-
2,000 spoken on the continent) are used in justice systems, and only 26 countries allow
African languages in legislation.3 With only an estimated 10-15% of the African
1 Director of Programs in International Justice and Society, International Center for
Ethics, Justice and Public Life, Brandeis University, Waltham MA, USA. Swigart has a
Ph.D. in cultural anthropology with a focus on African sociolinguistics. Since 2002 she
has overseen the Brandeis Institute for International Judges (BIIJ) and other programs for
judges and international law practitioners worldwide (See The author first came to know
Hassan Jallow while he was serving as judge of the Special Court for Sierra Leone and
continued to collaborate with him on various programs after he became Chief Prosecutor
of the International Criminal Tribunal for Rwanda. This chapter has benefited greatly
from the feedback of the volume editors, Prof. Linda Carter of the University of the
Pacific McGeorge School of Law, and Ms. Alexandra Tomić, Head of the Court
Interpretation and Translation Section of the Registry of the International Criminal Court.
2 ‘The Language of Justice in Africa,’ lecture delivered by Ngugi wa Thiong’o on 16
September 2013 at the Taj Pamodzi Hotel, Lusaka, Zambia, available at
3 Why and How Africa Should Invest in African Languages and Multilingual Education:
An Evidence- and Practice-Based Policy Advocacy Brief, UNESCO Institute for Lifelong
Learning, 2010, available at
populace fluent in official European languages,4 the potential for linguistic alienation in
formal domains of law and justice is clear.
Ngugi described the scene in a typical African court, where the applicable law is written
in a European language and the officials of the court conduct proceedings in a European
language. If the accused can speak only an African language, he or she will require an
interpreter. ‘The defense, prosecution and the judge occupy a linguistic sphere totally
removed from the person whose guilt or innocence is on the line…This was the way it
was in the colonial era; this is the way it is in the postcolonial era.’5
When one transfers this scenario into the ‘higher order’ of international criminal courts
and tribunals, the difficulties are compounded. The stakes of the trials are very high:
defendants are charged with crimes deemed the most serious by near universal agreement
– namely war crimes, crimes against humanity and genocide – as expressed through
various treaties and conventions. The official languages of the courts are those spoken in
the world’s most powerful states, past and present. These are, not coincidentally, the
same states that had previously imposed their control over large swathes of the globe
through an imperial enterprise, or those that continue today to exert influence through
less direct means, including foreign assistance. Persons participating in an international
criminal justice procedure must either use a language of the court – with which they may
have had limited educational or practical contact, if any at all – or communicate through
an interpreter.
Communication challenges are not, however, limited to the individuals participating in
trials. International criminal courts and tribunals have important institutional mandates
4 Ibid.
5 Supra note 2. Ngugi’s description clearly refers to Western-style judicial proceedings
and not customary or religious “courts” or dispute resolution processes. The latter are
frequently used to resolve family and other disputes across Africa and the use of local
languages in such contexts would be the norm.
that reach beyond the determination of the guilt or innocence of an accused person. These
mandates involve outreach to victims and communities affected by the crimes under
examination and may extend to lofty aims, such as those of the International Criminal
Tribunal for Rwanda (ICTR) to ‘contribute to the process of national reconciliation and
to the restoration and maintenance of peace.’6 These mandates cannot be successfully
pursued without meaningful contact with the victims and communities in question,
contact that may be hindered by the absence of a shared language or even a common
understanding of what might constitute justice in the wake of widespread violence and
societal rupture. International courts and tribunals thus face the significant challenge of
finding effective ways to educate victims and affected communities about their
institutional aims and procedures, inform them of the outcome of trials, and contribute in
at least some measure to societal healing.
Questions about the role of language in international criminal courts and tribunals are
particularly important given the prominence of Africa in their work over the past two
decades. With the trials of the ICTR and the Special Court for Sierra Leone (SCSL)
essentially finished, it is a good moment to take a step back and look at how these two
institutions have dealt with the languages of their ‘constituents’. Furthermore, with the
International Criminal Court (ICC) now into its second decade of operation, and with all
of its current ‘situations’ in Africa, two critical questions emerge: 1) How is the ICC
handling the enormous challenges posed by its necessary contact with multiple African
languages, many of which are neither habitually written nor widespread by territory or
function? And 2) how might the strategies it adopts to linguistically accommodate the
6 SC Res. 955 (1994) on the establishment of the ICTR. See also Tim Gallimore, ‘The
ICTR Outreach Program: Integrating Justice and Reconciliation,’ paper presented at the
Conference on Challenging Impunity, Kigali, 2006, available at ‘By
prosecuting those who committed genocide and other serious violations against human
rights, the ICTR intends to break the cycle of impunity by reestablishing the fundamental
rule of law, under which the guilty are held accountable for their offences. It is expected
that the outcome of these prosecutions will also promote national reconciliation and
restore peace in Rwanda.’
persons most closely connected with its proceedings enhance both access to its
procedures and its legitimacy in the eyes of its constituents?
This paper is based on ongoing research about the various roles played by African
languages in institutions of international criminal justice, and the practices and policies
that have been developed to respond to these roles. Through an initial set of interviews
with judges, prosecutors, defense counsel, investigators, outreach officers, and language
services specialists from the three courts mentioned above,7 along with an exploration of
existing literature that touches on language use in these settings, the author attempts to
draw a picture of how African languages fare in critical domains of international criminal
justice. These domains include formal criminal investigations, courtroom proceedings,
outreach to the regions where the international crimes have been committed, and
communication with victims of these crimes. The aim of this paper is to sketch out some
of the situations and challenges that emerge when African languages come into contact
with institutions of international criminal justice, and to suggest some lessons that the
ICC might take from the experiences of the ICTR and SCSL as it moves ahead.8
2. Breaking ground: the International Criminal Tribunal for Rwanda
A. In the Beginning
Any examination of how African languages fare in the sphere of international criminal
justice must start with the ICTR. When the Tribunal was established to prosecute persons
responsible for genocide and other serious violations of international humanitarian law
7Approximately 20 individuals were interviewed during this phase of the project. In order
to ensure their anonymity, the interviewees are referred to throughout this paper by
function or title rather than name. All individuals spoke in their personal capacity, thus
their comments and evaluations do not necessarily reflect the official views of their
8 ***For a description of language issues in international courts and tribunals more
generally, see Daniel Terris, Cesare P.R. Romano, and Leigh Swigart, The International
Judge: an Introduction to the Men and Women who Decide the World’s Cases (Lebanon,
NH: University Press of New England; Oxford: Oxford University Press, 2007).
committed in 1994, the institution was confronted with numerous challenges of both a
legal and practical nature. Like the International Criminal Tribunal for the former
Yugoslavia (ICTY), the ICTR had to rapidly develop the rules and procedures that would
regulate its operation at all levels. One challenge to be met was how to ensure that a
largely non-Rwandan staff – working in the principal organs of the Office of the
Prosecutor (OTP), Chambers, and Registry – could effectively communicate with
Kinyarwanda-speaking accused persons and witnesses.9 Due process rights required that
the accused be able to understand the charges against them and participate fully in the
trial proceedings. At the same time, investigators had to carry out their work in the field
and take statements from witnesses, usually in Kinyarwanda; OTP staff needed to
construct a prosecutorial strategy based on translations of those statements; judges had to
follow trial proceedings, during which an estimated 90% of witnesses testified in
Kinyarwanda, through consecutive and later simultaneous interpreters;10 and the Registry
had to undertake the training of these interpreters as well as check the translations of
trial-related documents between the two official languages of the tribunal, French and
English, and also those from Kinyarwanda into official languages.11
While such tasks might seem daunting, it should be understood that Rwanda is
exceptional among African countries for the simplicity of its language situation. Not only
is one local language, Kinyarwanda, spoken by almost the entire population, but more
than half the population can read and write it as well, a very high literacy rate for an
9 The absence of Rwandan staff members characterized in particular the early years of the
tribunal, when it was believed that their participation might introduce a bias into the
proceedings. Over the years, more and more Rwandan nationals were hired across the
ICTR staff with the exception of Chambers. Interview with a former ICTR Judge and
President (27 March 2013).
10 Estimates of the percentage of Kinyarwanda-speaking witnesses are from the
interview, ibid., and an interview with a defense counsel who represented accused
persons at the ICTR, and later at the SCSL and ICC (16 April 2013).
11 Interview with an interpreter and translator in the ICTR Language Section (15 April
African language.12 Consequently, the ICTR was required to accommodate the speakers
of a single language instead of the multiplicity of local languages that would have been
found in almost any other African country.13 Kinyarwanda was able to become, as one
interviewee phrased it, ‘a de facto working language of the tribunal’.14
The various legal instruments of the ICTR show its official stance on both the status of
languages and the need for linguistic comprehension. Rule 3 of the ICTR Rules of
Procedure and Evidence notes the following:
(A) The working languages of the Tribunal shall be English and French. (B) The
accused shall have the right to use his own language. (C) Counsel for the accused
may apply to a Judge or a Chamber for leave to use a language other than the two
working ones or the language of the accused. If such leave is granted, the
expenses of interpretation and translation shall be borne by the Tribunal to the
extent, if any, determined by the President, taking into account the rights of the
Defence and the interests of justice. (D) Any other person appearing before the
Tribunal, who does not have sufficient knowledge of either of the two working
languages, may use his own language. (E) The Registrar shall make any necessary
arrangements for interpretation and translation of the working languages. 15
Article 20 of the ICTR Statute establishes that the accused has the right:
12 Figures are from the Fourth Population and Housing Census, conducted in 2012 by the
National Institute of Statistics of Rwanda. The overall national literacy rate is 68%, with
49% literate in Kinyarwanda alone, and another 14% literate in Kinyarwanda and either
French or English or both.
13 It was suggested that there might have been occasional testimony given in Swahili but
this was not verified.
14 Interview, supra note 9.
15 Rule 3 ICTR RPE.
4(a) To be informed promptly and in detail in a language which he or she
understands of the nature and cause of the charge against him or her; 4(f) To have
the free assistance of an interpreter if he or she cannot understand or speak the
language used in the International Tribunal for Rwanda.16
It will be seen that the phrase ‘to be informed promptly and in detail in a language which
he or she understands’ is a critical element in the rights of the accused and is subject to
further refinement in the development of the ICC Rome Statute.
B. Before the Trials
The need to address language issues in an international criminal court or tribunal starts
long before a trial opens. Many interviewees for this project spoke at length about the
efforts made by the ICTR Registry to codify Kinyarwanda terminology for important
international criminal law concepts, for example ‘indictment’. Sometimes new terms
were coined, and other times existing lexical items were imbued with new meanings. An
ICTR judge expressed admiration for the outcome, which ensured a consistent use of
terminology across trials and witnesses.17 An interpreter from the Registry Language
Services Section who helped develop this glossary explained that it was not a static
product – it evolved over the years as translators and interpreters ran across terms that
were hard to render into Kinyarwanda.18 An ICTR defense counsel observed that such a
glossary might not, however, solve all the problems of conveying the meaning of a
specialized legal term to people unfamiliar with the notion; ‘a concept like “burden of
proof”,’ he said, ‘doesn’t need to be interpreted as much as explained.’19
16 Art. 20 ICTRSt.
17 Interview, supra note 9.
18 Interview, supra note 11.
19 Interview with former ICTR and ICTY defense counsel, currently a judge of the ICTY
and ICC (18 April 2013).
The Language Section also took on the training of Kinyarwanda translators (for
documents) and interpreters (for live speech). An interviewee explained that all official
translations – for example of the documents that circulate between the parties to a case
had to be certified by the Language Section. As for interpretation, there was some
‘experimentation’ in how to do it in the early days of the tribunal. At first it was done
consecutively, as opposed to simultaneously, sometimes using a chain of interpretation
from Kinyarwanda to French to English and then back again; this process did not require
specialized training but rather individuals who spoke two of the relevant languages well.
Meanwhile, the tribunal sought out people with skills in languages, law and also
international relations as potential simultaneous interpreters. They underwent an intensive
in-house training over six to nine months to acquire the necessary skills, with continued
training provided afterward. Most simultaneous interpreters worked either between
English or French and Kinyarwanda, with only exceptional individuals able to work in all
three target languages.20
The OTP and defense unit also required interpreters, which they hired independently of
the Registry’s Language Section. Several interviewees had been employed as field
interpreters, that is, language specialists who worked alongside investigators to take
statements from victims and witnesses, almost always in Kinyarwanda. Since only
English and French could be used for official trial documents, a convoluted linguistic
procedure had to be followed to verify a witness statement. An OTP language assistant
described how investigators would ask questions in either English or French, depending
on their nationality or personal language skills. The field interpreter would translate the
questions into Kinyarwanda, listen to the Kinyarwanda responses, and then translate them
on the spot into English or French written notes. The investigator would subsequently
compile a witness statement in English or French from these notes, which would be read
back to the witness through a spontaneous Kinyarwanda translation. If the witness found
20 Interview, supra note 11.
the statement accurate, he or she would sign the European language version.21 One
defense counsel from another region of Africa expressed regret that he had not learned
Kinyarwanda himself, which would have greatly facilitated his work with witnesses.22 A
former field interpreter reported that investigators who spoke other Bantu languages
could, however, sometimes follow witness remarks in Kinyarwanda or at least recognize
Several interviewees mentioned the difficulty of discussing sensitive topics with
witnesses due to cultural rather than linguistic reasons. Asking women to relate the
precise details of an act of sexual violence, for example, was particularly painful for
Rwandan men acting as field interpreters, especially when, as one reported, ‘the victim
was the age of my mother’.24 He went on to say that the victims themselves were often
more comfortable with the process, perhaps because they had been asked to tell their
stories so many times that they were no longer embarrassed. The OTP eventually hired
women investigators and interpreters to work with victims of sexual violence.
Cultural particularities were also obvious in the style that many Rwandan witnesses used
when responding to investigator questions. Long and indirect answers to short and direct
questions were frequent, and witnesses would also respond in, by Western standards, an
imprecise manner to questions concerning time and distance. One interviewee related that
Western investigators would sometimes become exasperated by the circuitousness of
witness responses, exclaiming something like, ‘If I ask for your name, just give it!’ The
question for an interpreter, he continued, was how much of the witness narrative to
reproduce for an impatient investigator; sometimes decisions had to be made about what
21 Interview with a language assistant in the OTP and former field interpreter for OTP
investigators (8 April 2013).
22 Interview, supra note 10.
23 Interview, supra note 21. There are hundreds of languages in central, east and southern
Africa belonging to the Bantu ‘family’, with varying degrees of mutual intelligibility. See
24 Ibid.
to leave out. ICTR investigators from countries like Senegal, Mali and Côte d’Ivoire were
familiar with such narrative patterns, however, and tended to take them in stride.25
C. Reaching Out to the Affected Community
Outreach to the Rwandan population was also an activity for which communication in
Kinyarwanda was critical. A major accomplishment for the ICTR Outreach Program was
the establishment of the Umusanzu mu Bwiyunge26 Information and Documentation
Centre in Kigali in 2000, with a number of branches subsequently created around the
country. The goal of the Centre was to inform the local population about the aims of the
Tribunal and help them keep abreast of judgments. An ICTR Outreach Officer described
the overall program this way in 2006:
The ICTR Outreach Program includes a series of pro-active projects,
complementary to the main institutional communications of the Tribunal.
Particular attention is given to mass media and interpersonal communication in
order to convey efficient and persuasive messages to targeted audiences inside
and outside Rwanda. The Outreach Program makes available information about
the Tribunal in Kinyarwanda using printed materials, radio broadcasts and
speakers. Targeted visits to the ICTR by Rwandan and other audience groups are
organized in order to improve the understanding and perception of the work of the
Tribunal. Apart from informing the public about its work, the Tribunal’s Outreach
Program also delivers training sessions and professional workshops to strengthen
the judicial system in Rwanda.”27
25 Ibid.
26 “Contribution to reconciliation” in Kinyarwanda.
27 Gallimore, supra note 6.
Despite the ambitiousness of the activities described above, it has been noted that the
outreach program of the ICTR was ‘quite anemic … and arrived late in process’.28
Perhaps the most effective outreach would have consisted of live broadcasts of ICTR
trials so that the Rwandan population could follow the proceedings regularly. OTP
officials suggested in the early days of the tribunal that trials be screened in sports
stadiums, and they believed that funding could have been found for such an undertaking.
The Rwandan government did not support this outreach strategy, however.29 In later
years, the Outreach Program collaborated with the NGO Internews to create videos of the
highlights of ICTR trial proceedings and judgments that were subsequently screened
around the country and followed by discussion sessions in Kinyarwanda.30
The ICTR also created and maintained a website with information about all aspects of the
institution and its work, with sections in both official languages as well as
Kinyarwanda.31 A quick perusal of the site quickly shows, however, that the English
section is by far the most detailed and complete. While most judgments and other official
documents are available in French, there are still a number of links that return to
documents in their original English versions. The Kinyarwanda section appears to be
essentially symbolic, with limited material available in that language at all and most links
leading back to the English language site. Several interviewees attributed the incomplete
offerings on the French and Kinyarwanda sites to a lack of resources32 and the fact that
28 Stuart Ford, ‘How Special Is the Special Court’s Outreach Section?’ In Charles
Chernor Jalloh (ed.) The Sierra Leone Special Court and Its Legacy: The Impact for
Africa and International Criminal Law (Cambridge: Cambridge University Press, 2014),
at 505-526.
29 Interview with former Chief Prosecutor of the ICTY and ICTR (11 March 2013).
30 The archive of these films can be accessed at:
(visited 3 September 2014).
31 See
32 Interview with Language Assistant in the OTP (9 April 2013), and interview, supra
note 9.
such translation was considered a ‘low urgency task’ compared to the translation of
documents needed for trials.33
D. In the Courtroom
The challenges associated with the use of African languages in international justice
processes are perhaps most apparent in the courtroom. As noted before, most of the
witness testimony during ICTR trials was given in Kinyarwanda. Accused persons, on the
other hand, who generally came from the Rwandan elite and thus were educated and
francophone, often preferred to testify in French. But some accused chose to testify in
Kinyarwanda as well, perhaps because they felt most comfortable or articulate in their
native language. In the early years, when courtroom interpretation was still consecutive,
this choice might even have been strategic, the delay necessary for interpretation from
one language to another giving the accused that much more time to consider his response
to a question.34
Interpretation in any multilingual court of law may be imperfect but it is clearly
indispensable; even theorists of simultaneous interpretation, which was used for the first
time during the Nuremberg trials, have called it ‘a necessary evil’.35 Some of the
challenges faced by the ICTR in providing the best possible interpretation system during
its trials have already been mentioned, namely the absence of both trained interpreters
and familiar terms in Kinyarwanda to denote concepts in international criminal law. Even
after these challenges were addressed through training and glossary development,
however, persons closely involved in ICTR trials as well as observers remained critical of
33 Interview, supra note 11.
34 This strategy was identified by an interviewee who, during his early years of legal
practice, used to recommend that his clients speak local languages in the courtrooms of
his native country for this reason. Interview, supra note 29.
35 Cited in Ruth Morris, ‘Justice in Four languages or “Interpreters and Mistresses”,’
review of The Origins of Simultaneous Interpretation: The Nuremberg Trial by
Francesca Gaiba, 1998,’ International Association of Conference Interpreters, December
1999 (1), available at
the ambiguities and inaccuracies that continued to raise questions about the proceedings.
As Nancy A. Combs has written about international criminal proceedings in non-Western
settings, ‘cultural divergences between witnesses and courtroom personnel, along with
linguistic and conceptual divergences, magnify the distortion wrought by language
The first case at the ICTR, The Prosecutor v. Jean-Paul Akayesu,37 involved the
prosecution of the mayor of Taba on charges relating to genocide, crimes against
humanity and violations of Article 3 common to the Geneva Conventions. The case is
widely known for handing down the first conviction for genocide in an international court
and furthermore establishing that rape can constitute an act of genocide. The Akayesu
judgment openly recognized and referred to the difficulties raised by the mix of
languages and cultural practices that came together in the courtroom:
The majority of the witnesses in this trial testified in Kinyarwanda. The Chamber
notes that the interpretation of oral testimony of witnesses from Kinyarwanda into
one of the official languages of the Tribunal has been a particularly great
challenge due to the fact that the syntax and everyday modes of expression in the
Kinyarwanda language are complex and difficult to translate into French or
Another section of the Akayesu judgment reads:
36 Nancy A. Combs, Fact-Finding without Facts: the Uncertain Evidentiary Foundations
of International Criminal Convictions (Cambridge: Cambridge University Press, 2010),
at 68.
37 Akayesu (ICTR-96-4-T).
38 Judgment, Akayesu (ICTR-96-4-T), 2 September 1998, para. 145.
According to the testimony of Dr. Ruzindana,39 it is a particular feature of the
Rwandan culture that people are not always direct in answering questions,
especially if the question is delicate. In such cases, the answers given will very
often have to be ‘decoded’ in order to be understood correctly. This interpretation
will rely on the contact, the particular speech community, the identity of and the
relation between the orator and the listener, and the subject matter of the question.
The Chamber noted this in the proceedings.40
Testimony about rape ended up being particularly problematic during testimony.
Witnesses and victims used a number of different Kinyarwanda words and phrases to
refer to sexual relations and sexual violence,41 partly because of cultural taboos
surrounding the topic. The Akayesu judgment notes,
[t]he word gusambanya means ‘to bring (a person) to commit adultery or
fornication’. The word kurungora means ‘to have sexual intercourse with a
woman’. This term is used regardless of whether the woman is married or not, and
regardless of whether she gives consent or not. The word kuryamana means ‘to
share a bed’ or ‘to have sexual intercourse’, depending on the context. It seems
similar to the colloquial usage in English and in French of the term ‘to sleep
with’. The term gufata ku ngufu means ‘to take (anything) by force’ and also ‘to
The variety and ambiguity of terms used by witnesses to signify the act of rape raised a
problem for international criminal law procedure, which requires that specific criteria be
met in order for the charge of a particular crime to be satisfied. Combs has described this
39 Dr. Mathias Ruzindana was an expert witness on linguistics in the trial.
40 Judgment, supra note 38, para. 156.
41 Ibid. para.152-154.
42 Ibid. para.152.
conundrum in her scholarly work on how fact-finding happens in international criminal
Kinyarwanda has no one certain word for ‘rape,’ and victims will frequently use
phrases that might be considered euphemisms to Western ears but that clearly
signify rape to Rwandans. Tribunal prosecutors, seeking to make out their best
case, cannot be satisfied with euphemisms, though, so they typically require
witnesses to describe their rapes in more graphic detail.43
An OTP language assistant spoke of working with witness statement transcripts that
showed that victims had been pushed hard by investigators to provide details about their
rapes, sometimes needing to be asked repeatedly, or in subsequent interviews, before they
supplied the required information.44 Combs observes, ‘a Western victim’s credibility
would be in shreds if she failed to mention that she had been raped until a late stage of
the investigation,’ but these omissions were not considered problematic, as seen in ICTR
judgments.45 Such witnesses would be left, of course, open to harsh cross-examination by
the defense about why the act of rape was not mentioned in the first place.
Linguistic and cultural particularities in the courtroom were not confined to testimony
about sexual violence, of course. Some cases were themselves centered on the way in
which accused persons allegedly used language in a culturally encoded manner to
promote or incite genocide. An important element in The Prosecutor v. Ferdinand
Nahimana, Jean-Bosco Barayagwiza and Hassan Ngeze46– also known as the Media case
was the role played by particular Kinyarwanda terminology and its local understanding
in the incitement of genocide. The prosecution argued that the use of certain terms in
radio broadcasts and in the written press – like inyenzi (‘cockroach’) to denote the Tutsi
43 Combs, supra note 36, at 87.
44 Interview, supra note 32.
45 Combs, supra note 36, at 87.
46 Judgment, Nahimana (ICTR-99-52-A), 3 December 2003.
and gukora (‘to work’), alleged to signify in this context ‘to kill’ – served to desensitize
the Hutu population and incite them to murder the Tutsi population in 1994. In The
Prosecutor v. Simon Bikindi,47 the accused, a well-known Rwandan singer, was alleged to
have used his musical compositions and public speeches to incite and promoted hatred
and violence against the Tutsi. In The Prosecutor v. Tharcisse Muvunyi,48 it was alleged
that the accused used Kinyarwanda proverbs – for example, ‘If a snake rounded itself on
a calabash you can’t kill it without breaking the calabash itself’ – to direct Hutus to
exterminate any remaining Tutsis in their area.49 Courtroom testimony during these cases
saw various linguistic and sociolinguistic experts called to the witness stand to explain
how certain usages were understood by the Rwandan public and why they were so
powerful. The trial chamber judgment in Bikindi notes, for example, ‘Prosecution Expert
Karangwa testified that Bikindi’s songs were couched in poetic language, referring to the
realm of the spoken and unspoken in Kinyarwanda.’50
Another common problem in ICTR witness testimony, potentially associated with the
multilingual character of the trials, was inconsistency. A former OTP field interpreter
noted that, while the OTP had done a good job overall in its language services, there were
always problems that remained. ‘Sometimes a statement taken before trial in the field
from a witness – which is verified and signed – is then rejected in the courtroom by that
same witness, who might attribute the disparity to bad translation.’ However, he
continued, the reality might be that the witness had simply forgotten what he or she said
47 Judgment, Bikindi (ICTR-01-72-T), 2 December 2008.
48 Judgment, Muvunyi (ICTR-00-55A-T), 11 February 2010.
49 See ‘ICTR/Muvunyi – Ex-Officer Allegedly Used Ethnic Proverbs to Incite Tutsi
Killings in 1994,’ Hirondelle News, 17 June 2009, available at
50 Bikindi, supra note 47, para. 197.
before, given the lapse of time. He also suspected that witnesses sometimes heard what
others had testified and consequently changed their stories.51
In his research on the ICTR, anthropologist Nigel Eltringham has documented both of
these phenomena, that is, alteration in witness testimony over time and the ‘co-produced
redescription’ of events as experienced by individual witnesses.52 ICTR defense counsel
Beth Lyons has suggested that the tribunal practice of housing protected Prosecution
witnesses together in Arusha might have contributed to the latter:
…logically, it is more than likely that at least some information is exchanged
while these witnesses are waiting to give their testimony, especially since this
may be the first time the witness is outside Rwanda and away from family and
friends. At a minimum, this living situation is not optimal for preserving each
witness’s separate memory of events, as opposed to a ‘composite’ one that may be
based on both the witness’s observation as well as those of friends or
Finally, there is the problem of simple inaccuracy in interpretation. Lyons writes of her
experiences during several ICTR trials:
Not only are judges at the mercy of the interpreters, but almost everyone else in
the courtroom is as well. At the ICTR, it is routine and too frequent that one of the
Kinyarwanda speaking defence team members or the client has to correct the
51 Interview, supra note 21.
52 Nigel Eltringham, ‘ “Illuminating the Broader Context”: Anthropological and
Historical Knowledge at the International Criminal Tribunal for Rwanda,’ 19(2) Journal
of the Royal Anthropological Institute (2013) 338-355, at 348.
53 Beth S. Lyons, ‘Enough is Enough: the Illegitimacy of International Criminal
Convictions: a review essay of Fact-Finding Without Facts, the Uncertain Evidentiary
Foundations of International Criminal Convictions by Nancy Amoury Combs,’ 13(3)
Journal of Genocide Research (2011), 287-312, at 290.
interpretation in the record during the proceedings. The court reports, on whose
written transcripts the judgments are based, obviously play a crucial role in the
preservation of the accuracy of the proceedings.’54
Other defense counsel remembered the same act of correction by their clients, although
one added that prosecution teams would also intervene sometimes to correct the record.55
A judge remembered in particular that defendant Ferdinand Nahimana, a former
university professor, would sometimes object to what he considered an inaccurate
translation.56 When such interventions were made in the courtroom, the issue would be
resolved immediately and the correction would go into the official transcript of the
proceeding, in French and English.
The linguistic and cultural issues that arose in the ICTR courtroom are particularly
significant because of one central truth – none of the ‘fact-finders’, that is the judges,
have been of the same nationality as the persons on trial. Consequently they have had
neither the linguistic nor cultural knowledge normally available to judges who adjudicate
criminal cases in their home countries.57 The result is that expert witnesses on issues of
language, culture, and history became important contributors to the judges’ understanding
of certain issues during ICTR trials. Eltringham captures the recognition of the role of
expert witnesses in this statement by an ICTR judge: ‘We need experts because we don’t
belong to the community. At home, experts are only technical, but here, we need
assistance. An expert explains what Kinyarwanda terms mean. We would not need that if
we were Rwandan judges.’58 However, Eltringham also found that ICTR judges who had
54 Ibid. at 289.
55 Interviews, supra note 10 and note 19.
56 Interview, supra note 9.
57 Eltringham notes that not only have no Rwandans appeared on a list of candidates for
the ICTR bench, but also no candidates from countries ‘implicated’ in the Rwandan
situation, that is Belgium, France, the United Kingdom and the United States. Supra note
52, at 350.
58 Ibid. at 350.
served for years felt that they had accumulated a large store of knowledge about
Rwandan language and culture and that the testimony of expert witnesses had a
diminishing value.59
A former ICTR judge interviewed for this project related that there also developed over
time an informal ‘in-house transfer of knowledge’ about the particularities of witness
testimony. From sitting in hearings day in and day out, this judge had learned how to ask
questions of witnesses in such a way as ‘to elicit the desired information in the end’. For
example, since Rwandans in the post-genocide era were no longer allowed to identify
themselves by ethnicity, he learned to ask witnesses instead what ethnic group their ID
cards indicated at the time of the genocide. Strategies like this were passed on to newer
members of the bench so that they could function more efficiently in the courtroom.60
E. The Legacy of ICTR Language Practices
The foregoing description of language issues at the ICTR shows both the attempts made
by the ICTR to work with Kinyarwanda speakers and the sociolinguistic challenges that
emerged in the course of this multifaceted justice operation. The challenges were many
and varied: the absence of Kinyarwanda terminology for critical legal terms; the need to
identify and train skilled interpreters; the mismatch between statements taken from
witnesses in the field and their later testimony in the courtroom, whether due to the loss
of accuracy that results from multiple ‘retranslations’ or problems of witness recollection;
cultural taboos and indirect speech patterns on the part of witnesses that clouded the clear
picture of events required to attribute guilt under international criminal law; inaccurate
interpretation in the courtroom and the questions this raised about the official record; and
finally the basic lack of linguistic and cultural knowledge possessed by the fact-finders
59 Ibid. at 351.
60 Interview, supra note 9.
How, then, might one begin to evaluate the legacy of the ICTR’s pioneering work in both
accommodating and getting what is needed from speakers of an African language during
international criminal justice proceedings? How effective were ICTR attempts to educate
and communicate with the Kinyarwanda-speaking community? What were its successes
in these areas and what might it have done better? Is it possible to determine what other
international criminal courts that interact with speakers of African languages have been
able to draw from the ICTR’s experiences as they developed their own policies and
Although it might be too early to find definitive answers to these questions, the viewpoint
of Rwandans who have been employed by the Tribunal as translators and interpreters is
telling. While clearly believers in the value of the ICTR’s work, the language experts
interviewed for this project were critical of what they felt was the insufficient attention
paid to Kinyarwanda and its potential to contribute to national reconciliation and the rule
of law in Rwanda. These interviewees regretted that the Tribunal had not directed
adequate resources toward the translation of all ICTR judgments and other materials into
Kinyarwanda. Given that literacy in Kinyarwanda is relatively high, and that the language
furthermore plays a central role in the Rwandan administration, including the judiciary,
this was seen as an enormous oversight and one that contradicts the Tribunal’s declared
aim of informing the Rwandan population about its acheivements.
The big work has not yet been done. Judgments are rarely translated into
Kinyarwanda. Instead, on the day of a judgment, there is local press that
summarizes the judgment and this is then put in the local newspapers and on the
radio. But this is not enough if there is to be any reconciliation in Rwanda. This is
not enough to ensure the tribunal’s legacy. Domestic courts cannot cite the
jurisprudence of the ICTR effectively because it is not available in
61 Interview, supra note 21.
This interviewee may be overstating the problem; in fact, domestic courts in Rwanda do
not exclusively operate in Kinyarwanda but also make use of the nation’s other two
official languages, English and French.62 ICTR judgments are thus not entirely
inaccessible to Rwandan lawyers and judges. He also seems to ignore the fact that the
thousands of genocide-related trials carried out in the national courts of Rwanda resulted
in Kinyarwanda language judgments, which, some might argue, are locally produced and
consequently better records of what transpired in the country in 1994.
However, this sense of disappointment in the ICTR’s lukewarm efforts to provide
materials in Kinyarwanda for the benefit of the Rwandan people was expressed by other
interviewees as well. ‘Most of the materials are in English, a little bit less in French, very
few in Kinyarwanda,’ explained a second language expert. ‘Donors complain that the
Tribunal is costly, so they say they cannot afford translation.’63
A third interviewee described the imbalanced allocation of resources across the ICTR,
which he felt had undermined the overall success of the institution. He furthermore laid
out the ICTR experience as a cautionary tale for other international courts and tribunals:
The ICTR has lost the chance to make a big impact by not following through on
translation and dissemination in Kinyarwanda. The important thing is the willingness
of an institution to invest in language experts. International courts pay to fly
witnesses in to testify, and pay the fees of international lawyers. But it is hard to get
them to pay for the accuracy of language. Languages are the poor relatives that no
one wants to spend money on.64
62 Written communication with a Rwandan lawyer and educator, 2 July 2014.
63 Interview, supra note 32.,
64 Interview, supra note 11.
With all of its trials completed and only some appeals procedures ongoing as of this
writing, the ICTR has now moved into its ‘residual’ phase. Along with its sister tribunal,
the International Criminal Tribunal for the former Yugoslavia (ICTY), the ICTR is now
part of the Mechanism for International Criminal Tribunals (MICT).65 The MICT website
provides information in both tribunals’ official languages, English and French, as well as
in Kinyarwanda and Bosnian/Croatian/Serbian, the primary language used by accused
persons and witnesses during ICTY proceedings. The Kinyarwanda section of the MICT
website already appears more complete than that of the ICTR website, with fewer links
back to English documents and pages.66 It will be interesting to follow the MICT in the
years to come and see if and how it continues to accommodate its Kinyarwanda-speaking
constituents, despite the cessation of formal outreach activities. More generally, the
passage of time is clearly required to measure the impact of the ICTR’s work on
Rwandan society and the level of legitimacy it ultimately attains.
3. The Experiences of a Hybrid Court: the Special Court for Sierra Leone
A. Different Conditions, Similar Problems
The SCSL was set up in 2002 to try those bearing the greatest responsibility for serious
crimes against civilians and UN peacekeepers committed during Sierra Leone’s civil war
65 According to its website, ‘[The MICT] was established by the United Nations Security
Council on 22 December 2010 to carry out a number of essential functions of
the International Criminal Tribunal for Rwanda (ICTR) and the International Criminal
Tribunal for the former Yugoslavia (ICTY) after the completion of their respective
mandates. The establishment of the Mechanism is a key step of the Completion Strategies
of the two Tribunals. It is a new small, temporary and efficient body, tasked
with continuing the “jurisdiction, rights and obligations and essential functions” (UNSC
Resolution 1966) of the ICTR and the ICTY; and maintaining the legacy of both
institutions.’ See
66 However, an exploration of the MICT site’s link to the ICTY/ICTR Case Law
Database, which ‘contains extracts of judgements and decisions rendered by the Appeals
Chamber of the ICTY since 1997 and by the Appeals Chamber of the ICTR since 2004,’
shows that it is currently only available in English. See
between 1991 and 2002.67 A brief description of the language situation at the SCSL will
show that the Court faced many of the same challenges as the ICTR in accommodating
persons whose primary mode of communication was an African language. There was,
however, a significant difference in how the two institutions were staffed, a difference
that served to alleviate some communication problems at the SCSL, especially in the
courtroom. The SCSL was a ‘hybrid’ court, with Sierra Leonean judges and personnel
working alongside their international counterparts. This meant that local language
speakers and culture bearers were active participants in all facets of the Court’s work.
At the same time, the SCSL had to operate in a more complex multilingual context.
Unlike the ICTR, which could concentrate its energies on Kinyarwanda alone, the SCSL
needed to provide translation and interpretation between the official language of the
court, English, and a number of Sierra Leonean languages. These included Krio, a creole
language that serves as the national lingua franca, as well as the other three widely
spoken languages, Mende, Temne, and Limba. These languages occupy a different status
in their society than that of Kinyarwanda in Rwanda; they have a limited role in the
formal education system, are not a usual medium of written communication, and do not
play a role in administrative domains. They are, nonetheless, dynamic modes of oral
communication, their level of importance in this sphere perhaps bolstered by the low rate
of literacy found in Sierra Leone.68 According to an SCSL judge interviewed for this
project, those standing trial came from across the spectrum of national language and
culture groups. ‘The war in Sierra Leone represented a breakdown of law and order, and
every ethnic group was involved.’69
67 See
68 Different sources provide a range of literacy rates for Sierra Leone, but they tend to fall
somewhere between 35% and 45%. Although English, Mende, Temne and Arabic are all
indicated as languages of literacy, the author could find no further breakdown on literacy
according to specific language. See
69 Interview with a Judge and former President of the SCSL (28 March 2013).
At look at the SCSL Statute shows the same concern as that of the ICTR to guarantee the
linguistic rights of the accused while still privileging a European language with official
Article 17(4)
In the determination of any charge against the accused pursuant to the present
Statute, he or she shall be entitled to the following minimum guarantees, in full
a. To be informed promptly and in detail in a language which he or she
understands of the nature and cause of the charge against him or her.
f. To have the free assistance of an interpreter if he or she cannot understand or
speak the language used in the Special Court.
Article 24
The working language of the Special Court shall be English. 70
An interview with members of the SCSL Registry and the Office of Outreach and Public
Affairs provided a picture of how the Court ensured communication between speakers of
English and African languages.71 It was reported that most of the witnesses were
‘ordinary people’ who did not speak English; even those who spoke English often
preferred to testify in Krio or another Sierra Leonean language. Their testimony thus
needed to be interpreted for the official court record, kept in English.
After the Court was established in 2002, the Court Management Office identified
potential interpreters and then trained them rigorously for work in the courtroom. Some
had already worked as interpreters in the domestic courts of Sierra Leone, which also use
70 Art.17(4) SCSLSt, Art. 24, SCSLSt.
71 Joint interview with three members of the SCSL Registry and Office of Outreach and
Public Affairs (5 April 2013).
English as an official language. As at the ICTR, interpretation between local and official
languages began as a consecutive process and became simultaneous only when
interpreters acquired the necessary skills.72 Also similar to the ICTR was the Special
Court’s development of a standard vocabulary in various Sierra Leonean languages for
international criminal law terminology. A former SCSL Prosecutor identified some of the
concepts whose precise meanings needed to be conveyed to non-English speakers during
investigations and trials; these included ‘widespread’, ‘systematic’, ‘rape’, ‘pillage’,
‘murder’, ‘surrender’, ‘protected person’, and ‘war crime’. He explained:
When there are no words for particular legal concepts, one has to be very
cautious. If certain terms are not understood by the witness or accused, it is a
challenge to international criminal justice. The prosecution has to demonstrate
that the elements of the crime in question are there. The record may not end up
being correct, and a faulty record is a potentially appealable record.73
B. The Centrality of Outreach
In contrast to the ICTR, extensive outreach programs at the SCSL started soon after the
establishment of the Special Court and continued throughout its lifetime. Interviewees
reported that outreach officers, trained in how to explain the Court’s mandate and
conversant in at least three local languages, traveled all over Sierra Leone to make sure
the local populations understood the SCSL’s work.74 A variety of formats were used to
convey information, including ‘town hall meetings’, visits to schools, or public lectures.75
Visits to observe proceedings at the Court were organized for the public and civil society
organizations as well.76 The Court also had close relations with Sierra Leonean news
outlets and organs; local journalists were trained in legal terminology and how to report
72 Ibid.
73 Interview former SCSL Chief Prosecutor (3 April 2013).
74 Interview, supra note 71.
75 Ford, supra note 28.
76 Ibid.
on court proceedings, often in local languages. Over the years, journalists became quite
used to ‘the SCSL beat’. In remote areas of the country without access to television and
radio, outreach officers equipped with generators would screen videos of trial summaries,
prepared by the Court, where weeks of testimony might be edited into 45 minutes. Voice-
over was provided in Krio, and for the trial of Charles Taylor, there was voice-over in
Liberian English as well.77 The SCSL Outreach Program has often been touted as an
enormous success, and was even described as ‘the crown jewel of the Special Court’ by
independent expert Antonio Cassese, a former ICTY Judge and its first President, who
was appointed by the United Nations to review the Court’s operation.78
C. In the Courtroom
As for SCSL courtroom proceedings, interpretation of testimony could constantly be
checked for accuracy due to the presence of local language speakers. Not only were
Sierra Leonean legal assistants represented on the prosecution and defense teams,
nationals played other roles among the large courtroom staff. Perhaps of particular
importance was the presence of Sierra Leonean judges who could easily stop the
proceedings in order to correct the official record if an interpretation was deemed
inaccurate. As one Sierra Leonean judge reported, ‘This correction is something that an
Austrian or even a Nigerian judge could not do,’ although he added that the international
judges often ended up understanding some Krio after years of listening to testimony in
that language.79
Furthermore, the courtroom had an interpretation supervisor whose role was to listen to
the words of interpreters – both from English to Sierra Leonean languages and vice-versa
77 Interview, supra note 71. Liberian English refers to a group of English language
varieties, creoles and pidgins spoken in Liberia, which are distinct from Sierra Leonean
Krio. See Tom McArthur, The Oxford Guide to World English (Oxford: Oxford
University Press, 2003).
78 Cited in Ford, supra note 28, at 513. Cassese’s report was submitted 12 December
79 Interview with former SCSL Judge and President (22 March 2013).
– to check for accuracy. An interviewee from the SCSL Registry noted that there were
many ‘checks and balances’ in the interpretation process, which was one of benefits of a
hybrid court with local staff.80 A former Prosecutor expressed this in stronger terms,
saying that ‘the hybrid nature of the court was a blessing’ – not only could local staff
check on interpretation, but having local judges was important from ‘a diplomatic point
of view’.81
Despite the built-in advantages of a hybrid court, some observers have noted that trials at
the Special Court were nonetheless characterized not only by linguistic but also by
cultural divergences between Court personnel and those on the witness stand. A former
Prosecutor stated that the accuracy of the English language record at the Special Court
seemed doubtful due to errors of interpretation that had never been corrected.82 Combs
has documented that some of the indirect discourse styles and taboos around certain
topics that impaired communication and fact-finding at the ICTR were also common at
the SCSL.83 She also noted the particularly low level of literacy and educational
attainment of those who testified at the SCSL, and suggests that this complicated
witnesses’ ability to understand what was expected of them while giving testimony.84
Scholar Tim Kelsall followed the Special Court’s Civil Defense Forces trial closely and
analyzed the transcripts of its proceedings.85 He provides another perspective onto how
international criminal tribunals function in non-Western societies, describing in detail
‘some of the challenges posed … by the fact that the Court is surrounded by an
unfamiliar social and legal culture, in which the way people think about human rights,
80 Interview, supra note 71.
81 Interview, supra note 73.
82 Ibid.
83 Combs, supra note 36.
84 Ibid.
85 Tim Kelsall, Culture under Cross-Examination: International Justice and the Special
Court for Sierra Leone (Cambridge: Cambridge University Press, 2009). The full name
of the case is The Prosecutor vs. Sam Hinga Norman, Moinina Fofana, and Allieu
Kondewa (SCSL-04-14-A).
human agency and appropriate social conduct often differs radically from the way
international lawyers think about these things.’86 In particular, Kelsall described how
SCSL judges, even those from Sierra Leone or neighboring African countries, seemed
unwilling to engage with the non-Western beliefs held by witnesses about supernatural
powers, patterns of authority and responsibility, and conceptions of childhood. These
beliefs, Kelsall argues, had bearing on charges against the accused that dealt with
command responsibility and enlistment of child soldiers, even though they did not square
with the rationalist tradition underlying Western law and its practice.87
D. The Legacy of the SCSL in Africa
What is the upshot of the collision of languages and cultural beliefs that occurred at the
SCSL? How successful was the Court overall in fulfilling its mandate to try the persons
most responsible for grave crimes, educating Sierra Leonean citizens about its work
along the way? As for the ICTR, there are widely divergent views about the Court’s
success, depending upon one’s professional point of view or connection to the institution
itself. Even persons closely involved with the SCSL’s work recognize that potential
linguistic and cultural miscommunications were not a trivial matter for a court that must
judge the guilt or innocence of persons charged with war crimes and crimes against
humanity. A former Prosecutor felt confident, however, that if the official court record
was inaccurate due to poor interpretation, then it never ‘rose to the level of an injustice. If
there had been big language issues, then they would have come out on appeal.’88 An
administrator voiced the same belief, saying that there were too many measures in place
at the Court for serious miscommunication to have altered the outcome of a trial.89 Of
course, persons who worked in the SCSL Defense Office or served as counsel for the
accused might well have a less confident view of the trial process and its just outcomes.
86 Ibid. at 2.
87 Kelsall, supra note 85.
88 Interview, supra note 73.
89 Interview, supra note 71.
Published critiques of the Court from the perspective of the defense tend to focus,
however, on issues other than linguistic ones.90
The legacy of the Special Court’s language practices and policies for the local population
must be evaluated differently from that of the ICTR. Given that local languages do not
play a formal role in the Sierra Leonean judiciary, the translation of SCSL judgments into
Krio or other national languages would do little to ensure the impact of its jurisprudence.
But the domestic judiciary has still benefited from the work of the SCSL, a judge pointed
out – Sierra Leone now has a cadre of interpreters who have been trained specifically in
international criminal law and can use new interpretation technologies.91
In the end, it is perhaps the Court’s efforts to educate the population of Sierra Leone
about its work via the spoken media and in-person formats that will be its biggest legacy.
The SCSL seems to have set a new standard in this area, one that other criminal courts
can use as a model.92 The newest international criminal tribunal established in Africa is
the Extraordinary African Chambers (EAC), inaugurated in Senegal in 2013 to try the
former Chadian dictator Hissène Habré and others for crimes against humanity, war
crimes and torture.93 This institution appears to have paid close attention to the outreach
goals and strategies of the SCSL, or at least to have the same view on the necessity of
90 See, e.g., Vincent O. Nmehielle, ‘The Defense Office of the Special Court for Sierra
Leone: a Watershed in Realizing the Rights of Accused Persons in International Criminal
Justice,’ in Charles Chernor Jalloh (ed.) The Sierra Leone Special Court and Its Legacy:
The Impact for Africa and International Criminal Law (Cambridge: Cambridge
University Press, 2014), 527-549.
91 Interview, supra note 69.
92 Its apparent success notwithstanding, there are those who remain skeptical about the
on-the-ground impacts of such wide-ranging outreach activities. Stuart Ford has
documented that the SCSL reported 4,600 distinct outreach activities over 2007-2008,
compared to the approximately 170 activities carried out by the ICTY during the same
period in the former Yugoslavia. Subsequent surveys do not show, he argues, that the
Sierra Leonean citizenry understood the work of the Special Court any better than their
counterparts in the Balkans understood that of the ICTY. Supra note 28.
93 See
effective communication with those who have suffered from the crimes under
consideration. The EAC Statute makes provision for the audiovisual transmission of trial
proceedings except when the security of witnesses and other participants is in question.94
According to Human Rights Watch:
The approved budget [of the EAC] calls for a robust outreach program to ensure
that the trial is meaningful to the people of Chad, arguably those most interested
in and those most affected by this trial. The anticipated outreach activities
include: transmitting trial proceedings to Chad; translating proceedings into local
Chadian languages; producing audio and video summaries and written materials
with regular updates about progress in the case; and bringing Chadian journalists
and civil society leaders to Senegal to observe the court proceedings.”95
Given that the trials of the EAC are still in the investigative stage as of this writing, it is
unclear how this ambitious outreach program will play out in practice. Looking back at
the experience of the ICTR, one cannot help but wonder whether the fact that most of the
principal EAC personnel are Senegalese – including judges, investigative judges, and the
prosecutor – will hinder the institution’s ability to communicate with the Chadian public
or carry out its fact-finding effectively.
4. Into the Future: the International Criminal Court
A. New Needs, New Standards
94 Art. 36, EACSt.: ‘Enregistrement des audiences: Les audiences devant les Chambres
africaines extraordinaires, sous l’autorité du Procureur général, sont filmées et
enregistrées afin d’être diffusées sauf si cela contrevient aux mesures nécessaires à la
protection des témoins et autres participants.’
95Q&A: The Case of Hissène Habré before the Extraordinary Africa Chambers in Senegal, Human
Rights Watch,’ available at
extraordinary-african-chambers-senegal. The EAC is funded largely by Western
countries along with Chad and the African Union.
If the language challenges faced by the ICTR and SCSL as described above have
appeared daunting, they pale in comparison to those that the ICC has encountered in its
first decade and will continue to encounter in the years to come. As the world’s first
permanent international criminal court, the ICC has no territorial or situational
specificity. It cannot foresee the locations of the crimes it might be called upon to
investigate nor the languages in which persons accused of those crimes might choose to
communicate. Neither can it count upon having speakers of key languages among its own
staff. From the very outset, the ICC is thus confronted with challenges unknown to the
international criminal courts and tribunals that were established to address crimes in a
particular zone of conflict.
To date, all ICC ‘situations’ are in Africa – the Central African Republic, Côte d’Ivoire,
Democratic Republic of Congo (DRC), Kenya, Libya, Mali, Sudan and Uganda. Each of
these situations has given rise to multiple cases, which are in various phases of activity:
investigation, pre-trial, or trial. Only two judgments of conviction, in the Lubanga and
Katanga cases, and one judgment of acquittal, in the Ngudjolo case, have been delivered
as of this writing, with Lubanga and Ngudjolo currently under appeal.96 The diversity of
ICC cases currently in process means that a number of African languages, termed
‘situation languages’, have become critically important for various organs and offices of
the Court, in addition to its working languages, English and French.
The ICC also differentiates itself from the ICTR and SCSL in having made provision in
its statute for the participation of victims in its proceedings:
One of the great innovations of the Statute of the International Criminal Court and
its Rules of Procedure and Evidence is the series of rights granted to victims. For
96 Judgment, The Prosecutor v. Thomas Lubanga Dyilo (ICC-01/04-01/06), 14 march
2012; Judgment, The Prosecutor v. Germain Katanga (ICC-01/04-01/07), 7 March 2014;
Judgment, The Prosecutor v. Mathieu Ngudjolo Chui (ICC-01/04-02/12), 18 December
the first time in the history of international criminal justice, victims have the
possibility under the Statute to present their views and observations before the
Court… The victim-based provisions within the Rome Statute provide victims
with the opportunity to have their voices heard and to obtain, where appropriate,
some form of reparation for their suffering. It is this balance between retributive
and restorative justice that will enable the ICC to not only bring criminals to
justice but also to help the victims themselves rebuild their lives.97
The involvement of victims in the work of the Court adds a new level of need for direct
contact with affected regions and those who have allegedly suffered from the crimes of
accused persons, contact which can be most effectively made through the use of local
Interviews with persons working across the ICC revealed that the Court understands the
importance of accommodating speakers of its various situation languages. Even in the
early days when the Rome Statute was being negotiated and written, there was
recognition that, although European and other world languages would occupy a
privileged institutional role, the bar should be raised vis-à-vis the comprehension by
accused persons who speak other languages of the charges against them. Whereas both
the ICTR and SCSL statutes state that an accused person has the right ‘to be informed
promptly and in detail in a language which he or she understands (emphasis added) of
the nature and cause of the charge against him or her,’ the ICC Rome Statute goes
further, specifying that this communication must be in a language that the accused ‘fully
understand and speaks.’
Article 50 – Official and working languages
97 See http://www.icc-
1. The official languages of the Court shall be Arabic, Chinese, English, French,
Russian and Spanish …
2. The working languages of the Court shall be English and French. The Rules of
Procedure and Evidence shall determine the cases in which other official
languages may be used as working languages.
Article 67(1) - Rights of the accused
(a) To be informed promptly and in detail of the nature, cause and content of the
charge, in a language which the accused fully understands and speaks.
(f) To have, free of any cost, the assistance of a competent interpreter and such
translations as are necessary to meet the requirements of fairness, if any of the
proceedings of or documents presented to the Court are not in a language which
the accused fully understands and speaks. 98
An interviewee for this project suggested that this slight alteration in the text outlining
rights of the accused indicates a significant increase in the institution’s commitment to
providing the best language services possible.99 An ICC judge confirmed this, declaring,
‘the Court takes Article 67(1)(a) very seriously.’100
Interestingly, in his commentary on Article 67 of the Rome Statute, legal scholar William
Schabas notes that informing an accused of the ‘content’ of a charge against him or her,
and not just ‘the nature and cause’, represents an advance over the International Covenant
on Civil and Political Rights (ICCPR).101 Reference to being informed of charges ‘in a
language which the accused understands’ is also drawn from the ICCPR, but Schabas
notes that ‘[t]he purpose served by requiring not only that the accused understand the
98 Art. 50(1)&(2) ICCSt.; Art.67(1)(a)&(f) ICCSt.
99 Interview with official of the Court Interpretation and Translation Section, Division of
Court Services, in the ICC Registry (8 April 2014).
100 Interview with ICC judge (26 March 2013).
101 UN General Assembly, International Covenant on Civil and Political Rights, 16
December 1966, United Nations.
language but also speak it seems unclear.’102 A linguist or interpreter would have
recognized the significance of the addition. The knowledge required to actively use a
language is superior to that needed to passively understand it; the new standard thus
offers a fuller guarantee that an accused will fully comprehend the charges. Schabas does
indicate, however, that ‘insisting on this detail helps to exclude some Strasbourg case law
by which the norm is respected if the language is understood by the accused’s lawyer,
and not necessarily by the accused personally.’103
B. A Diversity of Language Services
The ICC’s Court Interpretation and Translation Section (STIC) is responsible for
numerous language-related activities and services at the ICC, including translation of
Court documents and interpretation during courtroom hearings. The latter is of particular
interest vis-à-vis African languages. To date, simultaneous interpreters have been trained
in Acholi (Uganda), Swahili Congolese variant (DRC), Lingala (DRC), Sango (Central
African Republic), and Zaghawa (Sudan). Consecutive interpreters are available for other
African languages as well, such as Alur (Uganda) and Lendu (DRC), and it is planned for
Fur (Sudan) and Sudanese Arabic.104 It was reported that although simultaneous
interpreters are recruited in their own right, some have been accredited previously for
consecutive field interpretation for the OTP. Potential interpreters are given a test to see,
as one interviewee phrased it, ‘if they have that particular switch in their brain that allows
them to hear one language and with a short delay translate it orally.’105 She added that,
although it is too early to establish definitively, especially given that the number of
interpreters is still small, there is some indication that the most successful ICC
interpreters are those who have been trained as lawyers and have the situation language
as their native one.
102 William A. Schabas, ‘Article 67: Rights of the Accused,’ in Otto Triffterer (ed.),
Commentary on the Rome Statute of the International Criminal Court: Observers’ Notes,
Article by Article (München, Germany: Beck; Portland, Or.: Hart, 2008), at 1257.
103 Ibid.
104 Written communication with STIC official (7 April 2014).
105 Interview, supra note 99.
Like the ICTR and SCSL, the ICC has developed legal terminology in situation
languages, as well as established clear definitions in both English and French for key
concepts in the Court’s work, such as ‘victim’ and ‘reparations’.106 One of STIC’s most
impressive accomplishments is its preparation of a glossary of legal phrases used in the
ICC courtroom, translated from English into French, Lingala, Sango, Arabic, and both
Congolese and standard variants of Swahili (with a mirror French version glossing the
same phrases into English and the other languages). This ‘Phraseology in the
Courtroom’107 features dozens of predictable phrases used at various stages of ICC
proceedings, setting a standard for their interpretation. Following are some examples:
A person shall not be compelled to incriminate himself or herself or to confess
guilt. (Pre-trial stage: investigation and prosecution);
Could the officer read out the charges? (Pre-trial stage: confirmation of charges);
The accused is presumed innocent until proven guilty before the Court. (Trial
stage: commencement of trial);
I solemnly swear that I will speak the truth, the whole truth, and nothing but the
truth. (Trial stage: examination of witnesses);
All parties and participants to the proceedings have the right to deliver their
closing arguments in the hope of swaying the judges’ decision in their favour.
(Trial stage: closing arguments);
After the closing statements, the Chamber shall retire to deliberate, in camera.
(Trial stage: deliberations);
In determining the sentence, the Chamber shall take into account the gravity of
the crimes and the individual circumstances of the convicted person. (Trial stage:
106 Ibid.
107 Copyright STIC, 2012. The Phraseology, prepared by a team of STIC interpreters over
several years, was kindly made available to the author.
Despite the vigorous efforts of STIC to deliver the best interpretation, questions of
accuracy are bound to arise in the context of the courtroom. It was reported that whereas
simultaneous interpretation usually aims for 75% accuracy, at the ICC they are trying for
90-95% accuracy, a percentage initially but informally established at the ICTY.108 This is
a high standard but it is necessary since the transcripts of interpreted testimony, and not
the audio recordings of testimony, serve as the official court record. The rate of accuracy
is verified by checking a random 10% of interpreted transcriptions in French and English
against the audio version of testimony and also against one another. A 20-year veteran of
interpretation services at international criminal tribunals also noted that, in her
experience, an inaccurate interpretation usually becomes obvious. ‘No matter who is or is
not checking, it comes out.’109
One of the most challenging languages that the ICC has had to accommodate is Zaghawa,
a situation language for the Banda case.110 Zaghawa is a language spoken in the Darfur
region of Sudan and parts of Chad, with only an estimated 169,000 speakers total.111
Banda has chosen to testify in Zaghawa and STIC has accordingly undertaken the
identification and training of interpreters. To prepare for the trial, important legal
terminology work has been done with Zaghawa trainee interpreters in Arabic as well,
which is an official language of Sudan and widely spoken in Darfur.112
The use of Zaghawa and other so-called ‘languages of lesser diffusion’ at the ICC raises
an important question: when only the person testifying and the interpreter are conversant
in the language, how are inaccuracies to be detected? At both the ICTR and the SCSL,
108 Interview, supra note 99.
109 Ibid.
110 The Prosecutor v. Abdallah Banda Abakaer Nourain (ICC-02/05-03/09). The case
originally involved two accused but proceedings against the second were terminated after
the Court received evidence pointing towards his death.
111 See
112 Interview, supra note 99.
there were always native speakers of the various African languages in the courtroom – if
not serving as judges then as counsel or in support positions – who were able to identify
mistakes in the English or French interpretation and ask for the official court record to be
corrected. While some inaccuracies still made it into the official record, as noted above,
the informal oversight of African language speakers still served as an important check.
An ICC judge expressed some concern about this dilemma at the Court, as well as about
the dearth of interpreters available for certain minority languages. She noted that at one
point, the ICC Chambers and OTP were each calling for the services of the one available
Zaghawa interpreter.113 The STIC policy is to establish a roster of interpreters and then
split them between the OTP and the Registry (for courtroom work) to avoid any conflicts
of interest that might arise if the same interpreters work for both. But sometimes an
exception has to be made in order for a case to move forward.114 More Zaghawa
interpreters have since undergone training, anticipating the opening of the Banda trial.115
C. Reaching Out in Multiple Languages
The ICC Outreach Unit also comes, by necessity, into frequent contact with African
language speakers. The Court describes the work of the Outreach Unit thus:
In order for the Court to fulfil its mandate, its role and judicial activities must be
understood by a variety of audiences. In this respect, the Court’s outreach
programme has been created to ensure that affected communities in situations
subject to investigation or proceedings can understand and follow the work of the
Court through the different phases of its activities.116
113 Interview, supra note 100.
114 Interview, supra note 99.
115 As of this writing, the trial is scheduled to open on 18 November 2014.
116 See http://icc-
The Outreach Unit engages with both specialized audiences (e.g. legal practitioners,
mainstream journalists, students and educators) and non-specialized audiences (e.g. grass
roots populations and community radio journalists), and chooses its language of
communication accordingly. While French or English can be used with the former, local
languages are de rigueur with the latter. Outreach coordinators, who generally hold a
number of desirable skills – knowledge of international law and communications,
management experience, etc. – are not nationals of the country where they work, for
reasons of both security and neutrality. Outreach assistants, on the other hand, are
nationals of the situation country and need to have, among other skills, knowledge of
local languages and cultures. One of the most important audiences for outreach is
victims, who are often assisted by local intermediaries in understanding and going
through the judicial and reparation process. The Outreach Unit has identified 20 African
languages that are critical for its work – in Uganda, DRC, Sudan, Central African
Republic, Kenya, Cote d’Ivoire, Libya and Mali – although some of its activities in these
languages have yet to begin. Almost all of its communications in these languages are
strictly oral.117
D. In the Courtroom
How have African languages fared to date in the ICC courtroom? With only a couple of
trials completed, all from the DRC,118 there is not yet enough experience from which to
draw conclusions. An ICC judge who sat on the Lubanga trial observed that it was an
experiment in many ways, including in how the timing worked in terms of interpretation
and translation. Although the Court was prepared to accommodate the defendant with his
native language, Lubanga chose to testify in French. Had he chosen an African language,
the judge continued, the proceedings would have been more complicated and taken much
longer.119 While this view from the bench is interesting, it may not truly reflect the inner-
workings of the language services department; when asked, a STIC official did not agree
117 Written communication with official of the ICC Outreach Unit (19 April 2013).
118 Supra note 96.
119 Interview with ICC judge (17 April 2013).
that language issues or a lack of interpreters would ever prevent a trial from taking place
or require the postponement of a hearing.120
Some of the same issues documented during ICTR and SCSL trials also arose, not
surprisingly, during the Lubanga trial. The defendant would occasionally correct the
interpretation of a witness’ testimony from Swahili into French, for example. Lubanga’s
lawyers would then halt the proceedings to see if the problem could be corrected
immediately for the record. Otherwise, the parties were instructed to discuss it for later
resolution. There were also moments of ‘cultural confusion’ when the judges and
courtroom staff did not understand some witnesses’ reluctance to testify. In one incident,
a young witness was unwilling to give his testimony before that of his father, out of
respect for the latter’s age and authority. A judge reported that much diplomacy and care
were necessary to convince him to testify in the requested order.121
Although it is unusual, there is also sometimes at call at the ICC for translated and
transcribed documents based on African language recordings. A judge spoke of the
Mbarushimana122 case where the defense team requested that multiple audiotapes in
Kinyarwanda be transcribed and translated, asserting that they were critical for the case.
The OTP objected, saying it would take months, if not years, for this work to be
completed and would delay the trial terribly. The defense team eventually dropped the
request.123 It is possible that future calls for the transcription and translation of African
languages may emerge in the trial of Joshua Arap Sang who, as head of a Kalenjin-
language radio station, is accused of inciting ethnic hatred in 2007-2008 following a
contested Kenyan presidential election.124
120 Personal communication with STIC official (26 May 2014).
121 Interview, supra note 119.
122 The Prosecutor v. Callixte Mbarushimana, ICC-01/04-01/10.
123 Interview, supra note 100.
124 The Prosecutor v. William Samoei Ruto and Joshua Arap Sang (ICC-01/09-01/11).
E. Looking Ahead
At this point in the life of the ICC, it would be difficult to assess the success of its various
language policies and practices. However, it would appear that the Court is fully
committed to ensuring fair trial rights and fulfilling its complex communications
mandate, both of which require intensive work with situation languages. Indeed, the
Court already does a better job than most African justice systems in incorporating
African languages into the formal legal process and treating them as full languages,
capable of expressing complex legal principles. This, of course, takes significant
resources, something that domestic courts in Africa rarely have.
The lessons of the ICTR can prove instructive for the ICC. As the first international
criminal tribunal to confront the need for communication with African language speakers,
it had to act quickly and innovate often. But as already noted above, some ICTR language
staff ultimately felt frustrated by the institution’s lack of attention to follow-through in
Kinyarwanda. One ICTR interpreter offered this advice to the ICC, paraphrased below:
Make the necessary investment in language services and it will save money in the
end. Every time a trial is delayed because a necessary document has not been
translated, it costs money. When a trial takes 30% longer because there is no
trained simultaneous interpreter, only a consecutive one, it brings with it
enormous costs. People at the top need to understand that investment in language
skills will shorten the proceedings and lower the costs of the Court. Creating a
highly professional language staff with proper infrastructure makes sense, both
practically and fiscally.125
As for learning from the experiences of the SCSL, the global court seems to have fully
embraced the outreach philosophy of the hybrid court, judging from the ICC’s diverse
range of programs targeting affected communities and victims in multiple languages. The
125 Interview, supra note 11.
question now is whether such a plan can be sustained from the point of view of both
human and financial resources as new situations and cases arise.
5. Conclusion
There are many contradictory views about the success and even appropriateness of
contemporary international criminal justice. Its advocates may tout it as the most
effective tool for redressing humanity’s worst crimes and putting an end to impunity. Its
critics may point out its enormous cost, its slow pace, or the fact that it represents a form
of ‘white man’s justice’ for its largely non-Western defendants. After closely studying its
proceedings, some may even assert that its very methods are flawed and resulting
judgments unfair.126 This paper has not weighed into this fractious debate, but rather
examined a single but critical facet of the work of international criminal courts and
tribunals – their communication with participants in the justice process who do not speak
an official institutional language. Without this communication, the international justice
‘project’ could not effectively or fairly render justice to defendants or to victims.
Whatever one’s views on this project, the reality is that the ICC is here for the
foreseeable future and that its current cases center on African language speakers. As
Kelsall observes of the ICC, ‘witnesses and defendants in these trials will come from
societies with very different cultures or cultural mixes to those that predominate in the
West, with varying ideas about morality, responsibility, evidence and truth. International
justice, because of this, needs to learn the lessons of working with unfamiliar cultures
fast.127 The corollary is that it needs to learn the lessons of working with unfamiliar
languages fast.
This chapter has aimed to shed light on how policies and practices related to the use of
African languages have been approached and how they have evolved from the early days
126 Combs, supra note 36, and Lyons, supra note 53.
127 Kelsall, supra note 85, at 3.
of the ICTR till today. By bringing forth the voices of individuals who have worked on
the front lines of interpretation, translation, and outreach, as well as in the roles of judge,
prosecutor, and defense counsel, a certain narrative about African languages has begun to
emerge. These languages – grammatically complex, rich in tradition and expressive
capacity, functionally marginalized by historical twists of fate – are now taking center
stage in some of the most innovative and forward-looking institutions to be found in the
early 21st century. Testimony given in these languages, despite the inevitably imperfect
process of interpretation, serves as the building blocks of judgments that advance global
thinking about international humanitarian and human rights law. It can only be hoped that
those who have suffered at the hands of the accused have found and will continue to find
some sense of resolution and peace in hearing their languages on the witness stand, in the
mouths of outreach officers, and in radio broadcasts about trial proceedings and
Language experts at the ICC seem very much aware of the potential pitfalls that come
with their work, amply illustrated by experiences from the ICTR and SCSL and
increasingly from their own institution. They are also diligently trying to overcome them,
and have clearly thought strategically about the best way to increase linguistic access to
the Court, and in the process enhance its perception of legitimacy and fairness. If the full
mandate of the ICC is to be achieved – that is, ‘to not only bring criminals to justice but
also to help the victims themselves rebuild their lives’128 – then the Court must pay
careful attention to its communication with participants and audiences in both the
courtroom and in the regions impacted by the crimes under consideration. This
undertaking may be formidable, but it is clearly well worth the cost and effort.
128 Supra note 97.
... It also faced challenges unknown to the ICTY, namely those arising from the linguistic and cultural disparities between international judges, prosecutors, defense counsel, and other staff on one hand, and the persons who came before the ICTR as accused persons and witnesses on the other. These challenges were many and varied; they included the absence of Kinyarwanda terminology for fundamental legal terms; the need to identify and train skilled interpreters; cultural taboos and indirect speech patterns on the part of witnesses that clouded the clear picture of events required to attribute guilt under international criminal law; inaccurate interpretation in the courtroom and the questions this raised about the official record; and finally the basic lack of linguistic and cultural knowledge possessed by the international judges themselves (Combs 2010; Eltringham 2013 Eltringham & 2014 Swigart 2015). All of these challenges had to be met with innovative strategies, and it was predominantly interpreters, expert witnesses, and diverse professionals from both Rwanda and other Africa countries who led their development. ...
... It has been described how SCSL judges, even those from Sierra Leone or neighboring African countries, had difficulty engaging with the non-Western beliefs held by witnesses about supernatural powers, patterns of authority and responsibility, and conceptions of childhood, and squaring them with the rationalist tradition underlying international law and its practice (Kelsall 2009). These challenges called for a constant rethinking of practices during investigations, in the courtroom, and notably for the SCSL, in its multi-faceted outreach program (Anders 2011; Jalloh 2013; Kelsall; Swigart 2015). In short, the need to accommodate African realities in institutions of international criminal justice, as well as persons connected to trials, greatly shaped the processes established to address crimes in both Rwanda and Sierra Leone. ...
... It also faced challenges unknown to the ICTY, namely those arising from the linguistic and cultural disparities between international judges, prosecutors, defense counsel, and other staff on one hand, and the persons who came before the ICTR as accused persons and witnesses on the other. These challenges were many and varied; they included the absence of Kinyarwanda terminology for fundamental legal terms; the need to identify and train skilled interpreters; cultural taboos and indirect speech patterns on the part of witnesses that clouded the clear picture of events required to attribute guilt under international criminal law; inaccurate interpretation in the courtroom and the questions this raised about the official record; and finally the basic lack of linguistic and cultural knowledge possessed by the international judges themselves ( Combs 2010;Eltringham 2013Eltringham & 2014Swigart 2015). All of these challenges had to be met with innovative strategies, and it was predominantly interpreters, expert witnesses, and diverse professionals from both Rwanda and other Africa countries who led their development. ...
Full-text available
When discussions about international law and justice in Africa arise, the narrative is generally predictable. Africa is commonly characterized as a continent that is unable to meet international standards in this domain and thus requires intervention from the outside to insure accountability for serious crimes and the protection of human rights. Commentary on the relationship between the International Criminal Court and its African member states tends to exemplify this narrative. This paper argues that the African continent should instead be seen as the source of many innovative strategies for applying globally accepted legal principles at the local level. This can be seen in the domain of international criminal justice, as well as in the work of regional human rights bodies and courts of regional integration, the latter proving themselves surprisingly activist in the sphere of human rights protections. In order to better analyze this phenomenon, the framework on the variable authority of international courts, developed by Karen Alter, Laurence Helfer, and Mikael Rask Madsen, is applied to the African justice context. This paper concludes that the sometimes exceptional judicial configurations, practices and interpretations found in Africa should be recognized internationally for what they are: expressions of a significant consensus on the part of African legal practitioners, jurists, and activists – if not always the authorities – about the need to protect fundamental human rights and challenge impunity on the African continent.
Full-text available
The trial of Jean-Paul Akayesu is by far the most well known and widely discussed case at the International Criminal Tribunal for Rwanda (ICTR), a distinction that can be attributed to the fact that it was groundbreaking for several reasons. However, with regard to the importance of this trial both as a precedent for subsequent ICTR cases and within the broader context of international jurisprudence, its most significant contribution has undoubtedly been the recognition and prosecution of rape as a means of perpetrating genocide. The task of collecting admissible evidence to that end was heavily impacted by the necessity of interpreting and translating witness testimonies from their original language of expression, Kinyarwanda, into the working languages of the Tribunal, French and English. The multiple challenges associated with this translation process concerned not only questions of semantic equivalence of specific lexical terms. They also highlighted the considerable ‘cultural collision’ that occurred in the courtroom between the Rwandan witnesses and the international judiciary. This article elucidates the complex task faced by the courtroom interpreters in navigating these linguistic and sociocultural considerations underpinning the evocations of rape within the evidentiary framework of the Akayesu trial.
Legitimacy of Unseen Actors in International Adjudication - edited by Freya Baetens August 2019
ResearchGate has not been able to resolve any references for this publication.