Evidence About Harm: Dual Status Victim Participant Testimony at the International Criminal Court and the Straitjacketing of Narratives About Suffering

Although victims at the International Criminal Court (ICC) are not parties, they can apply to become “victim participants” and may be authorized by an ICC Chamber to directly and orally express their views and concerns in court. Most ICC Trial Chambers, however, have preferred allowing legal representatives of these victim participants to call victims as witnesses to give testimonial evidence about the harm they suffered. Our article focuses on the practical-epistemological challenges that come with forcing accounts of harm into this testimony-format. We draw upon ethnomethodology and conversation analysis to elucidate the discursive techniques by which legal actors in the Ongwen trial manage these challenges. These include eliciting accounts that “exhibit” suffering, posing questions that transform the “inner self” into an object of inquiry, and approaching witnesses as “informal experts”. Furthermore, while questioning related to establishing criminal liability typically proceeds in a granular fashion, testimony-taking about harm is accompanied by a tolerance for extended answers and an orientation to narrativity.


I INTRODUCTION
On 4 February 2021, the ICC's Trial Chamber IX found Dominic Ongwen, a former commander of the Lord's Resistance Army (LRA), guilty of crimes against humanity and war crimes, committed in Northern Uganda between 1 July 2002 and 31 December 2005. 1 The ICC's role in the conflict in Northern Uganda received a great deal of scholarly attention, focusing on the ICC's impact on the peace-making process between the Ugandan Government and the LRA and the complex, at times conflict-fraught interactions with local reconciliation initiatives. 2 Moreover, there is growing awareness among researchers and practitioners that the ICC represents one among a range of transitional justice mechanisms, including truth commissions, reparations programmes, and institutional reforms. 3 In this context, victim participation is generally recognized as a necessary requirement for transitional justice mechanisms such as the ICC. 4 Victims of international crimes under the ICC's jurisdiction can intervene in ICC proceedings in two procedural roles: as witnesses (and give testimony), 5 or as victim participants (and present their views and concerns). 6 If the accused is found guilty, they can also claim and receive reparations from the convicted. 7 For clarity, this paper reserves the expression victim participation for referring to the victim participant role sensu stricto, and uses victim intervention as a cover term for victims' involvement in the proceedings in general, either in the witness or victim participant role. Importantly, victims may be involved in the process in both roles at once, as victim participants can also be allowed to give testimony (in the witness role). In that capacity, they are referred to as dual status victims (DSV).
This double procedural role illustrates the influence of restorative justice at the ICC, which is clearly reflected in the Court's mandate. It is because the restorative justice paradigm regards victims of crimes as subjects rather than objects, 8 that victims hold this ''enhanced'' procedural role (beyond that of witness) and the rights associated with it in an international criminal justice institution such as the ICC. 9 This is, in turn, consistent with a growing awareness that transitional justice processes need to acknowledge victims' needs, 10 counteracting critiques of earlier international criminal tribunals that did not allow victims to tell their own stories. 11 Academics have acknowledged that transitional justice can be victim-centred. 12 In societies transitioning from mass atrocities, victims' active roles in (international) criminal justice proceedings may illustrate restorative and reparatory practices, guided by specific goals concerning victims and their communities. 13 However, empirical studies on the perceptions and opinions of victims and their communities in Uganda and in other African countries related to ICC's investigations and cases reveal a mixed picture about the way victims' procedural roles are implemented at the ICC. 14 Moreover, partial mishandling of these procedural roles has also led to tensions with the accused's rights. 15 In this article, we examine how victims' different procedural roles at the ICC have been implemented in the recently concluded Ongwen trial. Like in other ICC trials, Trial Chamber IX interpreted and implemented victim participation rather restrictively, and it categorically refused to authorize (even a limited number of) victim participants to present their views and concerns directly and orally before the Court. In May 2018, the Legal Representatives for Victims (LRV) in Ongwen filed a request for: (i) two community leaders (arguably representing the diversity of victim participants in Ongwen) to present their views and concerns orally; and (ii) five DSVs to give testimony on harm. 16 In its decision, Trial Chamber IX allowed three DSVs to give testimony on harm, but rejected the requested oral presentation of views and concerns. 17 With this decision, the Chamber followed the ICC's established preference for witness testimony over victim participants' views and concerns, as the large majority of Trial Chambers have systematically disallowed victim participants to present their views and concerns directly and orally before the judges, except Trial Chamber III in Bemba. 18 Victim participation must always be balanced against competing interests and rights, and on this ground certain authors have argued that, mainly because of the very high numbers of victim participants at the ICC, restricting victim participation enables the Court to better ensure procedural efficiency and offers stronger guarantees for the protection of the accused person's rights. 19 Nevertheless, completely disallowing the direct and oral presentation of views and concerns also has questionable consequences, which many authors appear to have overlooked so far. In this paper, we document some of these consequences.
As a result of Trial Chamber IX's decision, DSV intervention in Ongwen was restricted to: (i) presentation of written views and concerns (in the victim participant role), of which the LRV presented an oral summary in court; and (ii) oral testimony (in the witness role), 16 See ICC, Ongwen, Trial Chamber IX, Decision on the Legal Representatives for Victims Requests to Present Evidence and Views and Concerns and related requests, ICC-02/04-01/15-1199-Red, 6 March 2018, paras. 9, 70 (hereinafter Ongwen Victims Decision). 17 Ibid., p. 26. 18 Bemba, Trial Chamber III, Decision on the supplemented applications by the legal representatives of victims to present evidence and the views and concerns of victims, ICC-01/05-01/08-2138, 22 February 2012, paras. 15-17. 19 See generally Salvatore Zappala`, ''The Rights of Victims v. the Rights of the either called by the Prosecutor to give evidence on the charged crimes and criminal liability or by the LRV to give evidence on harm inflicted on victims. In this article, we focus on the three DSVs that Trial Chamber IX authorized to testify on harm inflicted on the victims, as it is here that the tension between the witness and victim roles is most outspoken. In its decision, Trial Chamber IX gave the LRVs permission to examine each DSV in relation to a specific aspect of harm: The LRVs furthermore called five expert witnesses, who also testified in May 2018. Before that, the Prosecutor had already called several DSVs to testify on matters related to the charged crimes and Ongwen's liability (in 2017 and January 2018). However, since they were not called to give evidence on harm inflicted on victims, they are less centrally relevant to our study.
The remainder of the paper charts the consequences of curtailing DSV interventions in Ongwen in this way. To this end, we examine in detail those interventions that DSVs were allowed to make, in the witness role, when called by the LRV to give testimony about harm. To chart what happens when these DSV narratives of suffering are subjected to the question-answer format of testimony-taking, we provide an in-depth analysis of the court transcripts available on the ICC's website. These transcripts provide a detailed record of the trial actors' interactional conduct as the trial progresses, thus offering a window into the ''black box'' of what goes on inside the courtroom that would be unavailable if one only examines judicial decisions. For this, we draw on Ethnomethodology and Conversation Analysis (EMCA), 21 an approach to spoken interaction that seeks to elucidate the discursive methods and language resources that speakers rely on 20 Ongwen Victims Decision, para. 9 and p. 26. 21  for maintaining intersubjectivity in interaction. In this case, we resort to EMCA for uncovering the methods that LRVs, judges and witnesses use for co-producing accounts of suffering that meet the requirements of the testimony format. EMCA scholars have a long tradition of looking at the law as a practical, intersubjective accomplishment, 22 but our research is one of the first to use this approach in studying international criminal tribunals.
The article starts with a legal-normative analysis into the nature and content of the categories ''testimony'' and ''views and concerns''. In ''Views and Concerns Versus Testimony: A Normative Demarcation'' we show that, despite the apparent clarity with which judges and practitioners apply these notions, remarkably few attempts have been undertaken to delineate these categories in positive terms. In the remainder of the paper, we redirect attention from judicial decisions (output) to the interactions over the course of trial hearings (process) and examine the consequences of this indeterminacy in actual ICC trial practice. In the section ''Delineating the Scope of DSV Testimony as a Practical Problem'', we show how for the trial actors in Ongwen, determining the precise scope of testimony about harm constitutes a ''local'' problem that requires case-by-case consideration and for which they have to negotiate an ad hoc solution. In the next two sections, we turn to the actual presentation of testimony about harm in court and examine how this was done in the Ongwen's trial. In ''Methodologies for Establishing Harm'', we start with reviewing the discursive techniques used by the LRV, and at times also by the presiding judge, for eliciting accounts of suffering that are consistent with the testimony format. In passing, we point out how they differ from questioning strategies used for eliciting evidence about facts related to criminal liability. In ''Narrativity and Granularity'', we show that testimony-taking about harm goes hand-inhand with a tolerance for extended answers and a preference for ''narrativity''. Again, this is much less the case in testimony meant to elicit evidence related to criminal liability. In the ensuing ''Discussion'', we argue that the discursive techniques used for eliciting testimony about harm, and the tension between narrativity and granularity that such testimony exhibits, are very much indicative of 22  Before being allowed to testify as a witness, victim participants also need judicial authorization. Here, ICC jurisprudence has clarified that ''the Chamber will only grant applications on behalf of victims whose testimony can make a genuine contribution to the ascertainment of the truth'', 26 and provided that there are sufficient guarantees for the accused person's rights. 27 ICC Chambers have thus rejected accepting victims appearing at the ICC automatically as witnesses, 28 and they have also taken into account whether holding dual status would negatively affect the accused person's rights. 29 In any event, victims who ''fail to reach the threshold to be authorised to give evidence may still be permitted to express their views and concerns in person''. 30 At the same time, however, the ICC legal instruments lack (clear) definitions of ''witness'', ''testimony'', and ''views and concerns'' and refrain from making an unambiguous distinction between these notions based on their specific contents. The travaux pre´paratoires of the ICC instruments provide no explicit or detailed definition of ''views  ICC jurisprudence gives some indications of how the Court understands these notions, but it does not provide precise definitions and fails to clearly differentiate between ''testimony'' and ''views and concerns''. Concerning ''testimony'', for example, Trial Chamber VII in Bemba established that: ''Not every conversation a person has or every communication provided by the person qualifies as ''testimony'' -it is rather only those where persons are questioned in their capacity as witnesses in the context of or in anticipation of legal proceedings''. 33 In turn, Trial Chamber II in Katanga and Ngudjolo Chui determined that: [a] key factor in determining whether an out-of-court statement qualifies as testimony in the sense of article 67(l)(e) and rule 68 is that the person making the statement understands, when making the statement, that he or she is providing information which may be relied upon in the context of legal proceedings […] It is important […] that the statement is formalised in some manner and that the person making the statement asserts that it is truthful and based on personal knowledge. A unilaterally prepared affidavit may thus also qualify as testimony if the person making it clearly had the intention of making factual assertions for the purpose of future or ongoing legal proceedings.  ICC jurisprudence has only to a limited extent fleshed out the meaning of views and concerns, mainly by reiterating the intrinsic connection to the victim participants' personal interests established in Article 68(3): ''general interests of the victims are very wide-ranging and include an interest in […] being allowed to express their views and concerns'' 35 or ''any views and concerns of victims in relation to issues which affect their interests''. 36 Apart from noting that they are the ''equivalent of presenting submissions'', 37 ICC Chambers have not defined views and concerns in positive terms. 38 It has been established that they are not evidence, not delivered under oath, and cannot be subject to crossexamination. 39 Other than that, however, jurisprudence has not clearly explained how DSV's views and concerns differ from DSV's testimony, particularly testimony on harm inflicted on victims.
Among legal scholars, only very few have attempted to clearly define views and concerns. According to Donat-Cattin, they represent broad terms ''inclusive enough to allow victims (or their representatives) to bring before the Court elements of evidence upon which they wish to express views or concerns''. 40 Bachvarova refers to views and concerns as victim participants' entitlement ''to voice their stance, convey their emotions and put forward their preoccupations on the issue at stake''. 41 McDermott comments that views and concerns involve ''anything related to the proceedings, from relaying personal experiences to discontent related to the speed or efficiency of the trial''. 42 35  In addition to the already quoted jurisprudence, we would also like to draw attention to an ICC booklet published in 2006, which defines a witness as ''a person who gives evidence before the Court by testimony. A witness is normally called by the Prosecutor, who is trying to prove the criminal case against an accused, or the defence''. 43 The ICC website furthermore identifies different kinds of witnesses: expert witnesses testify about matters within their expertise (e.g., forensic experts), overview witnesses help determine contextual facts (e.g., NGO representatives), insider witnesses possess a direct connection with the accused, while fact witnesses ''have knowledge and testify about what happened. They can be crimes-based witnesses when they have suffered harm and testify as witnesses about what happened to them. Some of these witnesses can also hold the status of participating victims before the Court; they are called dual-status witnesses''. 44 The quote indicates that victims testifying as witnesses are of fundamental importance to the ICC. Before the trial, their testimony enables the Prosecutor to gather information about crimes committed, so that an investigation can be started and a strong case can be submitted. 45 During the trial, they can be called by the Prosecutor, the defence and the Chamber to give testimony about the charges, and by victim participants or their LRV to give evidence on the harm they suffered, 46 as happened in Ongwen. 47 Victims who appear as witnesses only intervene when they are called to testify, 48 and usually they are not represented by lawyers. 49 They provide testimony (evidence) by answering questions posed to them, 50 and hence their intervention is subject to tight constraints: 43  […] the process of victims ''expressing their views and concerns'' is not the same as ''giving evidence''. The former is, in essence, the equivalent of presenting submissions, and although any views and concerns of the victims may assist the Chamber in its approach to the evidence in the case, these statements by victims (made personally or advanced by their legal representatives) will not form part of the trial evidence. In order for participating victims to contribute to the evidence in the trial, it is necessary for them to give evidence under oath from the witness box. There is, therefore, a critical distinction between these two possible means of placing material before the Chamber. 51 Victim participants' views and concerns do not constitute evidence, and consequently victims who intervene in the victim participant role cannot be questioned by the parties. 52 If a victim (including a DSV) intervenes as a witness, however, the general framework on witness testimony is applicable: the victim must solemnly declare to speak the truth; his/her statement constitutes evidence; and he/she can be subject to questioning by the parties. 53 Bachvarova's work, arguably one of the few doctrinal analyses offering an in-depth discussion of the differences between DSV's witness testimony and DSV's views and concerns, makes it clear that the two can be distinguished regarding their nature and contents. 54 First, regarding their nature, witness testimony is per se evidence, as established in Article 69(1) of the ICC Statute, 55 and helps to assert facts under judicial consideration. By contrast, victim participants' views and concerns may affect the judicial evaluation of facts and evidence relevant to the case (e.g., by pointing out local customs), but they are not considered evidence. 56 They may enhance the ICC's grasp of the circumstances at hand, but they do not constitute material placed before the court to establish the case outcome (the trial judgment). Second, concerning content, testimony involves matters about which the witness possesses personal knowledge, per- sonal experience, and/or recollection of circumstances within the subject-matter of the respective case. 57 Generally, testimony consists of a narrative of events that took place in the past, prior to the criminal proceedings, and that are subject to examination in court. 58 Conversely, views and concerns are not statements of facts concerning past events but convey victims' stance, preoccupation, or opinions concerning specific issues arising in the proceedings that have a bearing on their personal interests and, thus, they do not (dis)prove facts under adjudication. 59 In this section, we presented an analytical summary of the normative and legal framework that establishes the grounds for delineating and distinguishing testimony (the witness role) from views and concerns (the victim participant role). In the upcoming sections, we adopt a different strategy and pursue the issue from a different angle. Instead of looking at the distinction between testimony and views and concerns ''on paper'' or ''in theory'', we will analyse trial transcripts of the DSV testimony called by the LRVs in Ongwen to examine how the issue is managed in everyday courtroom practice, that is, in the nitty gritty of question-answer exchanges between the LRV and the DSV witnesses giving testimony, including the interventions by the judge and the objections of the parties. We will show that Trial Chamber IX's decision not to allow the direct and oral expression of views and concerns and allow only DSV testimony on harm (which in turn reflects the established practice that the ICC seems to have developed around victim participation) resulted in the creation of an interactional ''playing field'' where the boundaries of what is acceptable as testimony are continually tested and where neat normative definitions turn out difficult to uphold. In the next section (''Delineating the Scope of DSV Testimony as a Practical Problem''), we show how Trial Chamber IX's instructions as to the limits the LRVs must observe concerning the questions that may posed to the DSVs about the harm they suffered poses a ''local'' problem that cannot be settled once and for all but must be decided on a case-bycase basis. The two subsequent sections (''Methodologies for Establishing Harm'' and ''Narrativity and Granularity'') review the discursive techniques to which the trial actors resort for eliciting evidence about harm in the testimonial format imposed by the Court. The latter is essentially a technique for eliciting testimony about facts, 57 Ibid., 169-170. 58 Ibid., 170. 59 Ibid., 169. and the interactional elicitation of ''evidence'' about harm thus poses particular epistemological challenges. Our analysis shows that the discursive techniques that trial actors in Ongwen have come up with for attending to these challenges de facto amounts to a further blurring of the procedural roles and witness and that of victim.
Of course, considered in isolation the analysis offered below does not yet allow any definitive normative conclusion as to the form that victim participation at the ICC should ideally take, and any proposals in this direction should be balanced against other pivotal rights and interests that play a role at the ICC, in particular the rights of the accused person and the need for procedural efficiency in light of the often massive number of victim participants that present themselves before the Court. Nevertheless, these empirical observations about trial actors' attempt to transcend the limitations of the testimonial format and the resulting blurring of procedural roles do represent an element that deserve to be taken into consideration in the debate about how to further flesh out victim participation. By drawing attention to these (maybe unforeseen) consequences of restricting victim participation, we hope to prompt further reflection among scholars and practitioners (including ICC Judges) as to whether ICC Trial Chambers would do better to leave at least some room for the direct and oral expression of views and concerns by at least some of the victims (including DSV), rather than completely banning this pivotal modality of participation.

III DELINEATING THE SCOPE OF DSV TESTIMONY AS A PRACTICAL PROBLEM
Trial Chamber IX in Ongwen thus only permitted one type of direct oral intervention: the presentation of testimony about harm. As we can see below, this restriction by no means relieves the trial actors of the practical task of precisely delineating the witness role determining the limits of what is acceptable as testimony about harm. The excerpt below is taken from the opening stage of the lead examination of the first DSV called by the LRV and documents an exchange between the LRV (MR MANOBA in the transcript) and the Presiding Judge. In its authorization to testify, the Chamber had determined that the witness was allowed to testify on the stigma experienced by abductees after their return. The exchange below demonstrates that, underneath, the seemingly clear instructions by the Chamber hide a practical problem that trial actors have to attend to locally. The episode shows how in the initial phase of the hearing, both the LRV and the judge orient their actions to the institutional conditions of tellability 61 and work together to determine the purpose and the content of the interaction. This negotiation about the scope of the interaction starts with the LRV acknowledging the judge's prior remarks. Yet, rather than proceeding to the questioning of the witness, the LRV directs a comment to the Judge, setting up a contrast between the purpose of the hearing (lines 3-4) and the questions he will have to ask the witness for contextualizing the harm he suffered (lines 4-5). The Judge acknowledges this contrast (''Yes, of course''; line 6), but builds the rest of his response as a [yes] + [but] construction, a standard format for expressing disagreement and continuing an argument in a mitigated fashion. 62 This disagreement is again softened, however, as the remainder of his response implies a shared agreement about the temporal focus of the questions the LRV may pose to the witness (''we are, I think, both saying…'', lines 8-10), which the LRV acknowledges in line 11.
But this negotiation on the need to delineate the scope of questioning and mutual acceptance of the temporal restrictions imposed by the purpose of the hearing does not imply that the issue is now once and for all resolved or, indeed, that it has to be resolved conclusively at this preliminary stage. This becomes clear from how the LRV designs his response to the judge in line 11 (''I'll do my best'') and his preface to the witness before beginning to question him in line 19 (''we'll try to cover some areas''), both of which commit the speaker to an effort rather than the intended result. In this way, the LRV aligns with the judge without having to agree with him about how the hearing will unfold. Later on, the judge reiterates that he will decide the boundaries of questioning on a case-by-case basis, in response to a defence objection that the first witness's testimony is establishing an additional criminal liability not part of the confirmed charges. 63 In both cases, the participants do not solve the problem, but rather make it explicit and establish the grounds for further negotiations in later stages of the hearing. Negotiations like these frequently reoccur in the course of the examinations of witnesses called by the LRV, and they illustrate how the participants are struggling with the temporal scope of a testimony on harm. When witnesses are allowed to talk about the consequences and the effects that potentially criminal events may have had on them, they will invariably also produce descriptions related to the actions that precipitated this harm. In the ICC's daily practice, accounts of harm are difficult to separate from the descriptions of the criminal event that precipitated that harm.

IV METHODOLOGIES FOR ESTABLISHING HARM
To assess the impact of the Chamber's decision to restrict DSV intervention to testimony about harm, it does not suffice to point out the indeterminacies of the regulatory framework or the absence of clear instructions delineating the testimonial playing field. Instead, one should also consider how, following the ICC's general practice of completely disallowing the direct and oral presentation of victims' views and concerns, Trial Chamber IX's decision to straitjacket narratives of suffering into the testimony format affected the shape and substance of the question-answer sequences through which it is delivered. Therefore, this section documents three recurrently used techniques for making harm visible. As the analysis shows, these techniques are collaboratively implemented by the questioning attorney, the witness, and at times also by the Judge (when stepping in as questioner).
Before proceeding to the analysis, we would like to insert a short note concerning our selection of excerpts and their overall representativeness. Our account of these practices for eliciting testimonial evidence about harm is based on an exhaustive analysis of the testimony of all three DSV witnesses called by the LRVs (mentioned in introductory section). This equals 9 hours and 17 minutes of testimony and 157 pages of transcript. The practices described in the following pages are thus representative of those that we identified in the corpus, in the inductive fashion that is characteristic of EMCA style research. 64 As is also customary in EMCA, our exposition contains multiple, sometimes lengthy excerpts of actual interaction, allowing the reader to repeat this inductive process and to decide for 64 See, for example, Paul ten Have, Doing Conversation Analysis: A Practical Guide (second edition) (London, Sage, 2007) 36-38. him/herself whether the analysis is convincing. 65 There are, however, obvious limitations to this presentational strategy, as it is impossible to include every instance that we found. Hence, the extracts below were selected because they somehow epitomize the practices that we identified. In addition, we also included slightly lengthier episodes in which two or more practices occur in close proximity. These lengthier episodes illustrate how these practices ''work together'' to overcome the problems trial actors encounter in eliciting accounts of harm within the limitations of the testimonial format, and as such they provide further support for the validity of our analysis. 66 Furthermore, one may also ask to what extent this analysis of DSV testimony in Ongwen is representative of similar DSV testimony in other trials before other chambers. First, we would like to reiterate that our analysis seeks to identify recurrent practices through which these testimonies are produced. As such, it focuses on the ''hows'' and ''whens'' of interaction rather than on the question ''why''. It seeks to account for the regularities observable in courtroom talk in terms of the characteristics of the ''language game'' that renders talk meaningful, 67 rather than resorting to the psychological motivations of individual participants. 68 Of course, we do recognize that because of its specific nature, the ICC recruits judges from a wide range of backgrounds, versed in different legal traditions and not necessarily sharing the same professional qualifications. 69 We do not deny that trial chamber judges who, for example, served as domestic prosecutors or investigative judges and who possessed hands-on experience in examining witnesses prior to their appointment at the ICC bench might proceed differently than judges with prior careers in academia or diplomacy. In this sense, the composition of the bench may indeed have had an effect on the way the examination of DSV witnesses before Trial Chamber IX was conducted. 70 But still, even if we only reviewed one particular case and even if the questioning practices that we observed may have been affected by the personal backgrounds of Trial Chamber IX's judges, the fact remains that the epistemological tensions to which these practices respond are of a structural nature. It is hypothetically possible that other chambers respond to them in a different way, but they will nevertheless still have to find a solution for the tensions reported here. In this sense, our analysis is clearly also relevant beyond Ongwen.

Establishing Harm via Descriptions of the Perpetrators' Conduct
In its most basic form, evidence on harm arguably consists of descriptions of actions and conduct by the perpetrators that would precipitate suffering. This is the case in the following extract, in which the LRV is questioning the DSV about what happened immediately after his abduction: Extract 2 71 1 Q. [9:56:00] How were you treated when you were taken to the bush by the rebels 2 that you mentioned? How did they treat you? 3 A. [9:56:22] They did not mistreat us, the younger people, but when you were 4 older, you had it rough with them. 5 Q. [9:56:34] How long did you have to walk on the day you were abducted before 6 you could probably take your rest? 7 A. [9:57:09] That day when we were abducted, we walked throughout the night. 8 And even when it was daybreak we continued walking up to about 2 p.m., then we 9 rested beside a water stream. 10 Q. [9:57:29] Were you given any food to eat? 70 Investigating the impact that the judges' professional background has on their questioning techniques would require a different form of analysis, which also uses traditional ethnographic techniques in addition to focusing on the discursive realization of practices in context. 11 A. [9:57:39] Being young, I was being given things to eat, but the rest of the 12 abductees were not being given anything.
In this episode, the trial actors' orientation to make suffering palpable is visible right from the beginning, as the LRV solicits an account about the treatment the DSV had received from the rebels in the bush (lines 1-2), which projects an answer explicitly describing distressing actions inflicted upon the witness. The DSV is unable to come up with a description of harmful treatment that would align with this expectation (''They did not mistreat us, the younger people'', line 3). Yet, he nevertheless indicates that he indeed understood the LRV's question in this way, by implicitly connecting the receiving of damaging treatment and the abductee's age group. Hence, he explains that as a younger person the rebels did not mistreat him, but he immediately contrasts this with the treatment older people received (''but when you were older, you had a rough time with them'', lines 3-4). Despite the witness's orientation to the expectation that he should produce accounts of suffering-implicative treatment, the LRV treats the first response as not yet sufficient and unpacks his initial inquiry into two further questions. In lines 5-6, he poses a question containing an ''embedded presupposition'' 72 that the victim suffered (he had to walk a long time before being allowed to rest), followed by another question in line 10 that strongly favours a ''no'' response. 73 Overall, this fragment shows that the LRV's questions are designed to elicit descriptions of scenes related to the abduction of which the understandability relies on common sense notions of suffering (e.g. walking throughout an entire night until two in the afternoon), that is, descriptions which are ''accountable'' as versions of suffering. The responses elicited in this way thus ''exhibit'' harm rather than ''claim'' it: the descriptions imply suffering, and therefore merely eliciting such descriptions is sufficient for demonstrating and establishing the harm inflicted upon the victim. 74 Extract 3 provides further evidence of the intrinsically sufferingimplicative nature of certain descriptions of perpetrator-conduct. Immediately prior to it, the LRV had asked the witness whether he 72  In lines 1-3, Judge Schmitt takes the floor and asks the witness to categorize the victims in demographic terms (age, gender), suggesting possible descriptions that might aggravate the atrocity of the execution and add to the complicity of the witness. In lines 21-22, the LRV subsequently invites the witness to re-categorize the persons he killed in ''relational terms'' (''Was this person or these people related to you or close family members that you were forced to kill?''), thereby underscoring the implications this forced complicity had on the witness's psychosocial wellbeing. The fact that the LRV in line 26 initiates a new line of questioning on an unrelated topic indicates that the witness's description of the executed fellow LRA member as 75 Ongwen 1 May 2018 Transcripts, pp. 12 (line 17) -13 (line 19).
''close friend'' (line 24) and his account of how that friendship started are presumed to be sufficiently suffering-implicative.

Establishing Harm via Description of the Inner-self
Eliciting descriptions of conduct is only one among several other ways for producing evidence about harm that lawyers have at their disposal. The episode below documents another recurrently used strategy. It came immediately prior to extract 3 above, and illustrates how the LRV opens his questioning about potential crimes the witness had committed during his time in the bush, such as the forced execution of fellow child-soldiers: person. I was 3 given to kill a person who had escaped and was re-abducted. And I was picked 4 because I was young, so that was meant for me to be used to know what to do. And 5 I was informed if I escape, I would also be killed. So they gave me to kill this person 6 as a way of instilling that fear in me. 7 Q. [10:02:00] And how did this make you feel? 8 A. [10:02:19] Because of the life in the bush, it scared me, but when I did it and 9 after that I forgot about it.
The extract shows how the LRV's line of questioning transforms harm that the victim inflicted on another person into an episode that was harmful to the victim himself. In cases like these, the establishment of harm typically proceeds in two steps. The LRV first asks a polar question to request confirmation from the victim about some new information, in this case, that he had been forced to kill a person. This initial question serves as a preliminary step to focus the interaction on a particular event and its circumstances, 77 which is in turn transformed into ''harm'' by eliciting an account of the impact this 76 Ibid., p. 12 (lines 6-14). episode made on the victim's ''inner self'' (''And how did this make you feel'', line 7). In this case, the impact is also conveyed indirectly, by the way in which the DSV downplays the effect the killing had on him and normalizes what must have been an abhorrent experience (lines 8-9).
The next episode demonstrates the extent to which such inquiries into the impact on the witness's inner self are normatively expected and the pivotal role they play in transforming testimony into testimony about harm. Prior to the excerpt, the witness had been recounting how, in the course of the abduction, a female co-abductee had been slain by these rebels after crying about leaving behind her newborn baby. Defence counsel Taku had objected to the witness's account, arguing that it raised new facts not part of the confirmed charges and was therefore inadmissible. 78 In the end, however, Judge Schmitt dismissed the defence objection on the grounds that harm needs to be contextualized, but he also advised the LRV to maintain a clear temporal focus on the witness's life after his return from the bush. Next, the LRV resumes questioning: Extract 5  I had never 9 seen someone hacked to death in my life.
After Judge Schmitt returns the floor to the LRV, the latter unsuccessfully attempts to move the interaction back to the murder previously described by the victim (lines 1-2) blocks the questioning by suggesting that the victim had already provided a response to that (lines 3-4). Although the judge's intervention prevents the LRV from eliciting a more detailed description of the murder, it implicitly recognizes that the victim indeed witnessed the event and provides an opportunity for the LRV to maintain the focus of the examination on the murder. The LRV responds by issuing an inquiry into the impact of the killing on the witness inner self (''how does this make you feel'', line 6) that transforms the account of the killing into ''harm'' inflicted upon the victim and that does not elicit any further comments. The fact that inquiries into the inner self are specifically tied to testimony about harm is further illustrated by their conspicuous absence in testimony about events related to Ongwen's criminal liability. The following fragment comes from the examination of a DSV called by the prosecution, who is questioned about a similar incident in which an LRA child soldier was allegedly forced to kill another recruit.
Extract 6  there was a 7 young girl, the girl was told to kneel down, they said they wanted to pray for her. 8 And when she knelt down, they hit her behind her head. When they asked why did 9 they kill that girl, they said she was a witch. That was one of the deaths or the 10 killings that I personally saw. They told the person, they told her, ''Kneel down. 11 They are going to pray for you.'' When she knelt down, then they clubbed her at the 80  Throughout the excerpt, inquiries into the DSV's inner self (e.g., feelings, motivations, etc.) are entirely absent. Instead, the description of someone being killed is followed by an inquiry into the chain of command (lines 13-14) and by further questioning about another child-soldier summarily executed by the LRA (starting in line 19), and the entire episode is geared towards establishing criminal liability (as the DSV in question is called to testify by the Prosecution).
Taken together, the instances discussed so far indicate that eliciting testimony about harm is not just a matter of straightforward ''questions'' that are put to the witness, but of inferential orientations employed by the LRV for exhibiting the victim's harm or suffering. In certain situations, harm is inferable ''from the outside'' and can be established through descriptions of conducts and events provided by the suspect, as it is the case in extracts 2 and 3. In other cases, including extracts 4 and 5, harm is conveyed by the response to a question about the victim's inner-self, i.e. ''how does [something] make you feel?''.

Establishing Harm via Local Expert Voice
Harm is multifaceted. Not only does it cross the boundary between (public) fact and (private) feeling (as the previous examples demonstrated), but it also extends across event-sites and across temporalities. The latter is evident from a third strategy that LRVs use for making harm visible. It involves questioning DSVs who, besides having been personally affected, also (i) occupy a position of responsibility within the communities that suffered from the LRA attacks on internally displaced person (IDP) camps (a schoolteacher EVIDENCE ABOUT HARM: DUAL STATUS VICTIM PARTICIPANT and an administrator); and (ii) are therefore able to testify about the ''aggregate'' harm from which the community suffered.
The following testimony involved the third witness called by the LRV, a ''local councillor'' elected by members of the local community to mediate with the government and administration. This DSV has been called by the LRV specifically to testify about the interrelated and cumulative nature of different forms of harm. After an extensive series of questions about the conditions in Lukodi IDP camp and about community life prior to the May 2004 attack, the LRV refocuses the examination on present-day issues, particularly on the psychological and economic impacts suffered by people in the village: Extract 7  People had 5 goats, chicken and other things. But all these things have been taken away and it's 6 not easy to gain back. 7 For me, as an example, I had cattle, I had goats because I had worked hard. But now 8 I'm not able to farm enough resources, to farm and raise enough resources to buy 9 more livestock. For that matter, people are economically poor. 10 Most times people appeal to government to send some handouts to the villages in the 11 community. The government is not even able to do that. They try to bring, 12 government tries to bring a few heads of cattle and some goats, but it is not enough 13 for everybody. Sometimes they bring only five and only five people benefit from it. 81  14 For that matter, it is not easy to create wealth among the community. 15 People are desperate, desperately in a poor, living in a poor condition. People are 16 not able to farm and get enough money to pay for the school fees, and for that matter 17 children who should be in school are not going to school because they can't afford 18 school fees. 19 Some of the children who are living in the community try to go and look for money in 20 the town areas and they come back with infections and end up dying. People are 21 getting sick. And you are not able to go and get medical treatment. For that matter, 22 the life, general life condition is very poor, not like it was in the past.
The extract illustrates how this ''local expert voice'' is produced collaboratively, as the DSV responds to a LRV question positioning him as someone able to speak about a quasi-technical topic (the economic situation) on behalf of his community. In his response, the councillor not only reports his observations as a resident of the IDP camp at the time of the attack, but also demonstrates his entitlement to comment on the general state of his fellow community members before and after the conflict. In his answer, he alternates between referring to the community as ''our people'' (line 3), thus invoking a sense of shared identity and positioning himself as a spokesperson, and the more neutral ''people'' (lines 4, 10, 15 and 20), distancing himself from those whose conduct he describes and conveying a sense of impartiality. To produce such evidence on collective harm, he also uses comparisons between past and present. For example, after describing that ''our people are poor these days'', referencing the current state of affairs, he provides a list of items community members used to possess in the past, including cattle, goats and chickens. Throughout his account the DSV displays his knowledge about the community, disclosing information about people contacting government authorities for assistance (lines 10-11) or children contracting diseases and dying after travelling to other towns (lines 19-21). He explains community cultural practices and how they were affected by the LRA attacks, making harm visible by describing their presumed aggregate impact on a broader collective of people.
Nevertheless, DSVs like the Lukodi councillor or the schoolteacher (who testified on 2 May 2018) are not expert witnesses in the legal sense, and their testimony challenges the aforementioned distinction between factual witnesses and expert witnesses who give evidence based on their professional expertise. The next excerpt illustrates how this tension manifests itself in testimony, and how expert voice is (re)negotiated during the hearing. Prior to this extract, the councillor had been recounting how the corpses, which had been buried immediately after the attack, had to be exhumed two days later to collect forensic evidence. In line 1, Judge Schmitt intervenes in the LRV's examination: Extract 8  Many people 20 transferred. Some people went to live in the town areas and never returned. Then In lines 5-6, Judge Schmitt asks the witness how the burial and exhumation ''made him feel'' (thus inviting him to disclose his ''inner self'') and inquires whether he is still affected by the events in the present. The councillor responds with a powerful account of his sensory experiences and bodily responses (in particular to the smell), including his desire for a cigarette (lines 8-9), followed by a description of flashback symptoms that he is still experiencing today. The LRV resumes questioning in lines 16-18, rephrasing the judge's impact-question but refocusing it on how the other members of the community reacted to the events, and hence forcing the witness to come up with an aggregate assessment.
From a legal-normative perspective, the indeterminacy and mixing of interactional positionings or ''footings'' 83 that characterizes such informal expertise is particularly tricky. One troublesome indeterminacy is the tension between ''giving testimony'' and ''representing'' the community on which one is presenting aggregate data. The Lukodi schoolteacher who testified on 2 May 2018, for example, reported to the LRV that on the eve of his appearance before the court, he consulted with the headteacher and someone of the school management committee, where it was decided that he would ''represent'' the school. In the subsequent cross-examination, the defense begins with asking the witness whether they also discussed the content of his testimony (which the schoolteacher denied). 84

V NARRATIVITY AND GRANULARITY
According to Bachvarova's aforementioned effort to establish a normative distinction between views and concerns and testimony based on their content, testimony supposedly consists of a narrative 83  account of events situated in the past, related to the accused person's criminal liability. 85 Analysts of trial discourse, however, have consistently pointed out that such accounts are typically co-authored, as the lead examiner plays a decisive role in eliciting testimony. 86 Typically, testimonies are organized in question and answer sequences, with legal professionals asking the questions and witnesses responding to them. The questioner role puts lawyers in control over the agenda of the interaction. 87 In principle, they are free to elicit extensive narrative responses from the witnesses, but in practice, extended narratives are quite rare in courtroom hearings 88 and the opportunity for witnesses to produce lengthy accounts is limited. 89 Restricting narrative responses seems to be an important resource for controlling the direction of the interaction, 90 and questioners systematically prefer asking a series of relatively constrained questions that elicit fairly short and direct answers. This step-by-step, highly regulated approach to shaping the witness's narrative is particularly suitable for eliciting testimony about facts related to the accused's criminal responsibility, as it allows the party who called the witness to systematically review the constitutive elements of the crime that needs to be established. Here, we shall refer to such heavily co-authored accounts as reflecting a ''granular'' approach to testimony, and the DSV examination by the prosecution in extract 6 above is a good example of this.
Although the LRVs in Ongwen at times also resort to such a granular mode of eliciting testimony, there are also multiple instances where witnesses are invited to recount events in an open, unconstrained narrative, and the latter are often favourably received by the Court. Prior to the excerpt below, the witness had recounted how he had gone to church together with his grandmother, where she created a stir by publicly announcing her grandson's return from the bush after having been kidnapped by the LRA. In lines 2-3, the LRV asks In line 18, Judge Schmitt takes the floor and instructs the LRV to not interrupt the interaction and let the witness continue with his narrative (presumably in response to an embodied attempt by the LRV to take the floor, not captured by the court transcript). Next, he reminds the witness that he should not mention any names of other people. The judge acknowledges that this may have been the topic of the LRV's intervention, but confirms the witness's right to the floor by pointing out that he is the ''owner'' of the ''story'' being told (''it's your story'', line 21). For the next ten minutes, the judge virtually takes control of testimony-taking, alternating pointed questions with extended answers from the witness. In extract 12, the judge returns the floor to the LRV.
Extract 10  In returning the floor to the LRV, Judge Schmitt seemingly produces an explicit valorization of the DSV in terms of his capacity to build powerful narratives. The positive appraisal of the witness's intelligence (''a clever person'', line 3) and the characterization of his testimony as ''he's narrating'' (line 3) are invoked as grounds for allowing such extended narrative turns. At another point (not reproduced here), the judge also explicitly brands the witness's narrative as ''emotionally moving''. 93 The judge's allusion to the witness's ownership of ''his story'' in the instruction to the LRV to let the witness continue his narrative (extract 10), and the judge's favourable valuation of emotion (not reproduced here), furthermore suggest that narrativity is associated with a particular kind of testimony. The next extract, again from the DSV local councillor's testimony, makes clear that for other kinds of evidence the narrative format is not deemed suitable. Prior to the excerpt, the witness had been giving ''expert voice'' testimony on living circumstances in the IDP camp prior to the attack, for which he used the narrative format. In the fragment below, the LRV refocuses the testimony on the attack itself. She begins her new line of questioning by asking whether the witness could describe what happened on that particular day (lines 1-3), a relatively unconstrained inquiry designed to elicit an initial account or ''master narrative''. 94 The witness once more responds in a narrative format, again producing an aggregate account of what happened: shouting and 6 blowing whistles and immediately started shooting at people. If they find you, 7 you'll just have to be shot or they will stab you with the bayonet of the gun or they 8 would throw you in the burning huts as they also continued torching the huts that 9 were there. Some people would be looting food items, until when they completed 10 their operations they completely burnt down the camp.
One minute and a half into the testimony, the defence counsel objects: Extract 11 (continued) He wanted a 27 document from me that I should write for him so he could go and buy items. So the 28 rebels entered at that time when I was even there. I was not told. I saw it with my 29 own eyes.
In the objection, defence counsel Taku contrasts what the testimony ought to consist of, i.e., first-hand accounts (''He should say what he saw, what he did, constrain to that'', lines 12-13), with what it actually comprises, i.e., ''a general statement about the charges'' (lines 13-14). The judge overrules the objection as premature based on the assumption that the LRV will be moving into this territory of what the witness actually experienced in her next question, but in doing so, he issues a strong hint to the LRV that her subsequent questioning should indeed be more restrictive (lines 15-17). The LRV accepts this guidance (''I'm grateful'', line 22), and subsequently resorts to a line of questioning that exercises stricter control on the witness's answers. She unpacks her open initial inquiry (''Can you tell us what happened on that day?'', lines 2-3) into a series of questions that re-examine (and reconstitute) the previous narrative in a granular way, gradually eliciting bits of factual information that the DSV witnessed with his own eyes. Note that the witness too explicitly orients to this expectation that he should restrict himself to what he personally witnessed (''I was not told. I saw it with my own eyes'', lines 27-29).
Hence, the granular approach is relaxed for eliciting accounts of harm (when a more narrative approach might even explicitly be favoured), but imposed strictly for eliciting factual accounts (even when the latter only serve as background for testimony on harm).

VI DISCUSSION
The picture that emerges from DSV intervention in Ongwen in view of Trial Chamber IX's decision to completely reject the direct and oral presentation of views and concerns (just like virtually all ICC Trial Chambers) and only accept DSV testimony on harm is a complex and multifaceted one. On the one hand, the category of testimony on harm suffered by victims is not clearly defined in the ICC instruments, and ICC jurisprudence has not sufficiently clarified how it should be differentiated from the closely related category of ''views and concerns'' provided by victim participants in terms of their specific contents. We also saw that trial actors repeatedly struggle with delineating the temporal scope of testimony about harm, and that this delineation represents a ''practical problem'' that requires multiple ad hoc decisions as testimony progresses. Yet, the last two analytical sections (''Methodologies for Establishing Harm'' and ''Narrativity and Granularity'') also demonstrated that, despite these indeterminacies, trial actors nevertheless have access to multiple discursive resources for making harm palpable, thereby demonstrating an astute practical understanding of what testimony about harm amounts to and how it should be presented before the Court. Furthermore, we found that the above-mentioned testimonial evidence about harm significantly deviates from ''regular'' testimony.
At the ICC, testimony typically involves questioning the defendant and/or witnesses about a clearly delineated range of past events to determine (i) whether these constitute crime(s) under the ICC's jurisdiction as well as (ii) whether the accused committed those crimes and under which mode of criminal liability. 96 Questioning is thus designed with an eye on producing evidence that either supports or undermines the case against the defendant. This establishing of criminal liability typically proceeds in a ''granular'' fashion, with successive questions that systematically review its different constituent elements in a step-by-step manner: who did what at which point in time; who knew what; who gave the order for doing something, etc.
The way in which the LRVs submit DSV testimony before the Court vividly demonstrates how, and why, narratives of harm challenge elicitation within such a rigid evidential framework. This, we contend, is directly related to the specific nature of harm. First, harm represents a phenomenon distributed across time and space, and extends from the past into the present and future. As such, it resists the unequivocal spatiotemporal delineation that characterizes crimi-nal charges (and may at times even be hard to disentangle from the charged crimes), as shown in the section ''Delineating the Scope of DSV Testimony as a Practical Problem''.
Second, harm blurs the common-sense distinction between publicly observable ''facts'' (available to any competent onlooker) and private ''feelings'', which is routinely perceived as a separate territory of knowledge to which only the subject-actors involved have access. 97 The DSV testimony that we analyzed shows that trial actors attend to this specificity of harm and suffering. Thus, we saw that the LRV's questioning oscillated between (i) eliciting descriptions that ''exhibit'' suffering (and in this sense ''externalize'' and ''objectify'' it) and (ii) asking ''how did this make you feel''-questions forcing the witness to disclose the impact a particular episode made on them (a technique often used when no direct (physical) harm was inflicted upon the victims themselves, e.g., when they were forced to observe harm done to others). The third technique that we identified, the practice of approaching the witness as informal expert, represents a way to deal with the ''distributed'' nature of testimony on harm, as it allows questioners to elicit aggregate assessments of the impact particular events had on the community over a longer stretch of time. This also includes collective assessments of these events' psychological impact, which (again) externalize subject-actors presumably ''private'' knowledge and transform it into observable facts. The tolerance for narrativity provides additional evidence of the distributed nature of the phenomenon, and further illustrates the extent to which the witness is the ''private owner'' of the experience he is recounting. 98 Eliciting witnesses' accounts of harm and suffering is thus in many ways at odds with the standard procedures for producing testimonial evidence, and the resulting testimony lacks the clearly delineated temporal scope, objectivity standards and granularity that characterize evidence related to criminal liability. In this sense, it can be argued that Trial Chamber IX's decision to completely disallow the direct and oral presentation of views and concerns resulted in an incongruity of modality and content. More specifically, Trial Chamber IX allowed the LRV to introduce certain statements made by a DSV in relation to the charged crimes as ''evidence'', although they required a greater degree of interpretation than evidence on criminal liability and although the extent to which they can effectively be challenged under cross-examination is questionable. This incongruity is probably most palpable in Judge Schmitt's remarks, in extract 10, about the witness's ownership of his testimony and the Judge's insistence that the witness should be allowed to develop his narrative in an unconstrained fashion.
In essence, it appears that Trial Chamber IX inadvertently and partially conflated the procedural roles of witness and victim participants, and that views and concerns (to be presented in the victim participant role) were submitted as testimonial evidence (delivered in the witness role). This conflation was already evident in Trial Chamber IX's initial decision to reject the oral presentation of views and concerns: ''through the questioning of the Prosecution witnesses by the Legal Representatives and the upcoming testimony of the witnesses to be called by the Legal Representatives, it has sufficient information to adequately address all points of the judgment, along with the views of the victims' communities''. 99 Moreover, Trial Chamber IX ''does not deem it appropriate to hear additional submissions, which are not related to the presentation of the evidence or subject to scrutiny through Defence questioning''. 100 From the above it is clear that Trial Chamber IX's rejection is not based on a careful consideration of the nature and merits of views and concerns per se, but on the ''added value'' they might have and on whether or not they can successfully complement testimonial evidence. In advancing testimonial evidence as the relevant yardstick, the Chamber neglects the autonomous nature, contents and relevance of views and concerns and operates under the assumption that they merely represent an ''unsworn'' counterpart of testimonial evidence.
The most dramatic manifestation of this conflation can be arguably found in the questions that Judge Schmitt asked the first DSV, the abducted child-soldier, concerning whether he would benefit from professional assistance to help him overcome the psychological consequences of the harm inflicted on him: The future temporal orientation of Judge Schmitt's second question lies outside of what would normally be considered testimony, and the issue of the victim's current predicament and the measures that could alleviate his situation also aligns more meaningfully with views and concerns.
Admittedly, Regulation 56 of the Regulations of the Court allows the Trial Chamber to hear witnesses and evidence related to reparations during the trial, in order to enhance procedural efficiency. However, in cases where there is no conviction (e.g., Ngudjolo Chui), or where the conviction is overturned on appeal (e.g., Bemba), this ''putting the cart before the horse'' scenario paradoxically has precisely the opposite effect. Hence, Trial Chamber I in Lubanga stated that ''there will be some areas of evidence concerning reparations which it would be inappropriate, unfair or inefficient to consider as part of the trial process''. 102 Moreover, from a normative perspective, and in light of the different ICC procedural stages, it is arguably more appropriate to postpone reparations-related questions until the postconviction reparations stage, because: (i) this procedural stage is specifically devoted to claiming and granting rehabilitative and other measures as part of reparations for victims; (ii) victims are proper parties to such post-conviction reparation proceedings (not merely victim participants); (iii) as conviction is a condition for the ICC to order reparations, counter-productive effects on victims and procedural efficiency are avoided in cases of acquittal or overruling of the conviction by the Appeals Chamber; and (iv) if the accused has been already convicted, there are substantially fewer risks of jeopardizing certain defence rights such as the presumption of innocence.
The LRV follows Judge's Schmitt example in concluding the examination of the local councillor (the third DSV, who testified on 3 May). After inquiring into the community's current situation and whether they have been able to recover from the attack, the LRV proceeds to the types of assistance that might be helpful today, to which the councillor answers that his people need money, food, housing, rehabilitation, etc. 103  Similar questions were posed to DSVs intervening in the victim participant role in Bemba, 105 the only ICC trial to date in which victim participants (including DSVs) were allowed to directly and orally present their views and concerns before the Court.
The above illustrates the extent to which inadvertently conflating the procedural roles of witness and victim participant dilutes the specificity of each individual role, eventually compromising the integrity of both testimony and views and concerns . To be clear, we are not advocating here for the general admissibility of direct and oral presentation of views and concerns by victim participants. Because of the large number of victim participants in ICC trials, this is virtually impossible, would negatively impact the rights of the accused and would seriously undermine procedural efficiency. However, we do consider it problematic that, in practice, this modality of victim participation has been totally banned by Trial Chamber IX and by virtually all the other ICC Trial Chambers. We contend that allowing at least some victim participants to directly and orally express their views and concerns could significantly reduce the risk of conflation and dilution that we documented in this paper.
It could be argued, furthermore, that this practice of systematically disallowing views and concerns is not fully consistent with both the letter and the telos of the normative framework for victim participation at the ICC. While tives of the victims (emphasis added)'', it also lays down that ''the Court shall permit their views and concerns to be presented … (emphasis added)''. As Donat-Cattin has argued, this implies that ''victims themselves have the right (''shall'') to express such views and concerns''. 106 Totally disallowing the direct and oral presentation of views and concerns thus clearly does not sit well with these provisions. With regard to the telos of these provisions, the situation is even more complicated. The ultimate purpose of including these provisions was to guarantee the right of victims of mass atrocities to participate in the ICC proceedings in an active and meaningful manner, 107 which arguably is particularly affected by the fact that virtually all the ICC Chambers have completely disallowed victims to intervene via direct and oral presentation of their views and concerns.
Victim intervention in the victim participant role at the ICC is thus confined to intermediated oral participation via their legal representative(s) or to participation through written submissions. All of this renders victim participation at the ICC largely symbolic. 108 As Ambos has remarked, ''Given that Article 68 (3) ICCS grants victims the statutory right to present their views and concerns, the drafters of the Statute must have believed that victim participation and accused rights can be reconciled''. 109

VII CONCLUSION
It is clear that DSV's testimony on harm suffered by victims (in the witness role) and DSV's views and concerns about such harm (in the victim participant role) are two interconnected categories. The analysis also demonstrated, however, that inadvertently conflating these two procedural roles and straitjacketing narratives of suffering in the testimonial format, as Trial Chamber IX did in Ongwen, dilutes their specificity and compromises the integrity of both views and concerns and of testimony. On the one hand, it may confront the DSV on the witness stand with hostile lines of questioning. On the other, it also stretches the category of testimony to include materials that lack the specificity, externality and granularity expected of testimonial evidence, which may in turn compromise defence rights. If ICC Chambers would allow some victim participants to directly and orally present their views and concerns before the judges during trial (which is now virtually non-existent in the ICC's practice), it might be easier to maintain a clear distinction between the two procedural roles. Ultimately, this would enhance the efficacy of both roles, as it would give victims the opportunity to discuss the harm they have suffered unimpeded by evidentiary standards.
As a potential way forward, then, the ICC Trial Chambers could and should consider moving away from their general practice that completely confines direct and oral intervention of victims (including DSV) to the witness testimony format. With due attention to the respect for the rights of the accused and procedural efficiency, the ICC Chambers could and should instead exceptionally allow victim participants to directly and orally express their views and concerns (victim participant role) before the Court, especially for those victim participants whose views and concerns are particularly representative of the majority of the victims in the respective trial and case. at the International Criminal Court'', 2019-2023). Open Access funding provided by University of Jyva¨skyla¨(JYU).

OPEN ACCESS
This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons. org/licenses/by/4.0/.
Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.