I. The Role of Witnesses at the International Criminal Court and the Protective Measures Offered to Them

In March 2013, one month before the trial was set to begin, the Prosecutor of the International Criminal Court […] withdrew all charges against Mr. Francis Kirimi Muthaura, former head of the Kenyan civil service who had been jointly charged with Kenyan President, Mr. Uhuru Kenyatta. The prosecution alleged that, due to critical problems with key eyewitnesses and the loss of potential evidence, it was no longer able to establish the case against Mr. Muthaura beyond reasonable doubt.6

Many academic discussions centered around the importance of witnesses in an ICC proceeding, however, I believe that the above account best exemplifies this principle in practice. The evidence of witness testimony plays a critical role in fact-finding carried out by the ICC. As of 2010, the ICC had engaged with around 520 witnesses;7 and, as of 2013, at least 199 witnesses had testified before the Court in The Hague.8 Those witnesses and their testimony allowed the ICC to reach its decisions and bring about what many believe to be justice.

Witnesses can assist Court in two main aspects. First, contrary to domestic judges, most ICC judges have never visited the countries they are discussing.9 Hence, they have never seen with their own eyes the nature of the crimes alleged and the cultural nuances. Hearing the testimony of witnesses can help the Court understand those complex situations and fill in those gaps in its knowledge.10 The second potential assistance regards fact finding that determines the holding of the case. As described in the Francis Kirimi Muthaura case above, the outcome of a trial is heavily related to the existence and quality of the testimony of witnesses.11

The ICC, in contrast to domestic courts, does not have any legal authority to force individuals to testify or take any part in its proceedings.12 Even if an individual is a key witness whose testimony is essential to the deliberations of the ICC, if the witness did not voluntary agreed to testify, there is no legal way to force the witness to do so.13

The power to investigate and to question the suspects, victims, and witnesses is within the authority of the ICC’s Office of the Prosecutor (OTP).14 The process of finding witnesses and deciding which one to bring to The Hague has several stages. First, OTP investigators travel to the relevant country;15 or, in a case of a potential witnesses that has escaped to another country, at the witness’s new location.16

After finding the potential witness and receiving his or her agreement, the investigator and the potential witness have a preliminary meeting to prepare for the questions and topics that will be discussed during the full interview.17

After the first meeting, the two sides meet again for the full interview. At this meeting, the potential witness is given information about the Court and the expected process if he or she becomes a witness.18 Together, the investigator and the potential witness fill out a witness statement. This statement is forwarded to a joint team. The team includes investigators, prosecutors, and cooperation staff that decide which potential witnesses are the most suitable to be brought to testify in The Hague.19 Only after the team makes its decision—and the potential witness agrees to testify—does the status of the potential witness change to witness.

The unique voluntary requirement mentioned above also means that the ICC must act and treat its witnesses in a way that will make them want to voluntarily cooperate with it. One of the most important issues is assuring the personal safety of witnesses, as a person might fear for his or her personal safety or have second thoughts about testifying before the Court.

To assure witness safety, the ICC has a fairly developed mechanism. Within the Court, the Registry operates the Victims and Witnesses Unit (the unit).20 As described in the Rome Statute, the role of this unit is to provide protective measures and security arrangements to:

[W]itnesses, victims who appear before the Court, and others who are at risk on account of testimony given by such witnesses.21

The protective measures and security arrangements provided by the unit, are divided into three categories on the ICC official website. The first category is operational protective measures, that can be implemented where witnesses reside.22 The second one is procedural protective measures that can be ordered by the Court.23 Stating that it’s a last resort protective measure, the ICC also offers a third category, relocation of the witness and his or her close relative from the source of threat. This solution can be offered on a temporary or permanent basis, and the relocation can be within the borders of the witness country or outside of it.24 Selection of the right measure to be chosen will be determined after an interview with the witness.25

II. The Performance Indicators made by the Court

A. The First Report of the Court on the development of performance indicators for the International Criminal Court

As described in the Introduction, in November 2015 the ICC released the Report of the Court on the development of performance indicators for the International Criminal Court.26 The report was created as a response to a request from the Assembly of States Parties to:

[I]ntensify its (the Court) efforts to develop qualitative and quantitative indicators that would allow the Court to demonstrate better its achievements and needs, as well as allowing States Parties to assess the Court’s performance in a more strategic manner.27

In this report, the Court set four key goals for assessing its performance and measuring its success.28 These goals were:29

1. The Court’s proceedings are expeditious, fair and transparent at every stage;

2. The ICC’s leadership and management are effective;

3. The ICC ensures adequate security for its work, including protection of those at risk from involvement with the Court; and

4. Victims have adequate access to the Court.

As mentioned, the key goal that is relevant for the purpose of this comment is the third one that regards the security of the persons involved with the ICC. In the report, it is explained that measuring the Court’s performance in the contexts of security will help it in assessing its effective operation.30 The ICC delineates two relevant performance indicators that, in regards to witnesses, will be measured against ones that are in court protection programs.31 Those indicators were:32

1. % of security incidents with witnesses, staff or information that led to actual harm compared to the total amount of incidents.

2. For the incidents that led to actual harm, % due to the Court’s error.

From indicator (a) it can be concluded that the group of people that would be included in the measurement are the ones referred to in part I as witnesses and Court staff. That conclusion also means that any potential witness would not be included. Moreover, from the combination of the two indicators it can be concluded that the measurement will only include security incidents and would not be based on the witness’ subjective feelings and thoughts regarding his or her safety. Those two conclusions are extremely important to establishing my argument and will be discussed extensively below.

B. The Second Report of the Court on the development of performance indicators for the International Criminal Court

One year later, in November 2016, the ICC released a follow-up report. In this report, the four key goals remained the same, but some measurable criteria and initial data were added.33 As to the third goal, the ICC divided the term security into two subdivisions—physical and asset security and information security.34 In every subdivision three general questions were asked:35

1. Determination of the general threat level that the Court is facing;

2. Has the Court implemented an appropriate/proportionate threat management programme?;

3. When a risk manifests itself, has the Court’s security framework proven adequate in the circumstances?

Like the performance indicators in the First Report, those questions and criteria would be measured against witnesses (in contrast to the group defined as potential witnesses), that are included in Court protection programs. Moreover, none of the new factors or questions stated in the Second Report includes a wider range of elements other than security incidents.

III. Selection Bias in the Performance Indicators made by the Court

As discussed above, the indicators laid by the ICC contains two main criteria:

1. The pool of people that will be examined—This group will include witnesses and staff (from this point on, the term group will refer to the witnesses and not to the staff.) As defined above, witnesses would be individuals that voluntarily agreed to cooperate with the Court and that were chosen by it.

2. The character and scope of the data examined—This criterion includes only security incidents or incidences that led to actual harm.

My argument is that those two current criteria were drawn incorrectly and too narrowly, causing the indicators to be inaccurate and therefore could and probably would lead to non-accurate results. This phenomenon is referred to in the research and statistic literature as “Selection Bias.” The dictionary definition of selection bias is:

Any source of bias that arises when a supposedly random sample has been chosen in ways that allow cases to drop out, or cases to be chosen with unequal likelihoods, in unintended ways.36

Selection Bias might “lead to erroneous conclusions about a phenomenon because it weakly represents the true nature of a population.”37

Before I try to prove why I think that the two criteria of the ICC indicators suffer from Selection Bias, I would like to explain what I think the right criteria should be:

1. The pool of people that will be examined—The right pool of people should not only contain the actual witnesses, but it should be broader and include potential witnesses. As described above, those potential witnesses would be on any stage of communication with the court, until the final diction of the joint team and their final voluntary agreement to become witnesses.

2. The character and scope of the data examined—The data examined should include subjective characters in addition to the current more objective ones. This way the scope of the data would be broader. What I refer to as subjective characters could be the potential witness or the actual witness lack of sense of security or security concerns. Those concerns might lead a potential witness to decide not to become a witness, or for a witness to hesitate before deciding to become one.

The first stage in proving the selection bias nature of the current version of the indicators is to understand their aim: what is it that they are trying to measure? As described in the First Report, security as an indicator is important for the effective operation of the Court, its risk management, strategic planning, and operation.38 A narrow interpretation of those aims would be the one the Court chose in its reports. An effective Court will be the one that gives the highest security protection to its witnesses.

I disagree with this interpretation. I believe that in order to really evaluate and measure the Court’s effectiveness, the security of a broader group of people needs to be measured. In part I, the importance of witnesses to the ICC was discussed. As mentioned there, charges can be dropped in case of inadmissible or unsatisfying testimony or if there are no witnesses available to Court. From that it can be concluded that an effective and well-functioning court is one that has wide access to witnesses and can allow itself to pick the most qualified among them.

An ineffective court would be one that does not have this privilege. A reason for that can be related to qualified potential witnesses deciding not to became witnesses because of security-related fears and worries. Those individuals would rather not cooperate with Court because they do not trust it to be able to protect them from the threat they are under (regardless of the existence of such threat).

This is why I believe that having indicators that solely measures the security of witnesses based on security incidents would “allow cases to drop out” as mentioned in the Oxford Dictionary’s definition of selection bias. Those cases are the potential witnesses’ cases and they are essential to the evaluation of Court effectiveness in terms of security. Not including the potential witnesses and their subjective views on Court performance security-wise will in fact lead to erroneous conclusions and would not give the ICC the full picture it’s looking for.

To illustrate my argument, I examine three hypothetical scenarios—three make believe worlds—and apply on them both the current version of the indicators and the offered version of them.

1. No security provided—In this scenario, the level of security that the ICC provides is so low that no person is willing to become a witness.

   1. Under the current version of the indicators—the performance of the Court in terms of security is spotless, since no security incidents or incidences that led to actual harm was made. Of course this outcome is extremely erroneous and does not give the full picture of the Court performance and effectiveness security-wise.

   2. Under the offered version of the indicators—the potential witnesses would be included and their fear of safety is legitimate data to be used. This data provides the ICC with the full picture regarding its ineffectiveness, and its problematic performance in terms of security.

2. Perfect security provided—In this scenario, the level of security that the ICC provide is so high that there are no security incidents, incidences that led to actual harm, or potential witnesses that decided not to become witnesses because of security concerns. This is the only scenario where using the offered version of the indicators will not be needed, since using the current version of them leads to accurate results and reflects the true effectiveness and performance of the court in terms of security.

3. Diverse levels of security provided—In this scenario, the levels of security provided by the ICC are not fixed or equal, but are influenced by different factors, like the conflicts regimes, the scope of the potential threat, the identity of the witness or potential witness, and more. Unlike the first two scenarios that were rather imaginary, the last one is probably similar to the real level of security provided by the ICC.

   1. Under the current version of the indicators—As to the witnesses, the ICC will receive a full picture regarding its ability to handle objective threats. It will not receive any subjective information that will indicate whether those witnesses had any doubts whether to become ones. Since this scenario is not identical to the second scenario (Perfect security provided), the data regarding fears and doubts could have helped the Court evaluate its performance even when it comes to witnesses that agree to cooperate with it.

      The main issue regarding this scenario and the current version of the indicators is involving potential witnesses. At this point I would like to perform a mental exercise. In terms of security, the main—and perhaps the only—reason for a potential witness to decide not to become a witness is because he or she will think that the security offered to them by the Court is not adequate. The potential reasons for that will be discussed extensively below, but it could be over relatively proven facts or over a gut feeling.

      From this perspective, the potential witness from scenario 1 (No security provided) and the potential witness from this scenario are very similar. The main difference between these two is the instance of the Courts shortage of protection and the percentage of those potential witnesses out of the population approach by the Court as a whole. The lack of trust in the Court’s ability to supply adequate security remains the same.

      This is why I think that, just as it was obvious on the first scenario that the current version of the indicators would lead to an extremely erroneous outcome, it should be obvious that this third scenario leads to a similarly erroneous outcome.

   2. Under the offered version of the indicators—Once the Court includes potential witnesses alongside the witnesses in its measure of security and effectiveness, it would be able to evaluate its true performance (or at least receive a better evaluation of it).

Up to this point, I’ve addressed selection bias as one term; but, in fact, there are several sub-categories of optional biases.39 In my opinion, at least two of those sub-categories are relevant to the case of the ICC security-performance indicators. I will shortly address those sub-categories.

1. Sampling selection bias40—This kind of bias is the classical one. Its definition is identical to the one given above.41 In this case and as discussed, the bias arises as a result of the Court’s decision regarding who to sample (witnesses) and who to leave outside of the sampling group (potential witnesses).42 The outcome of this bias will be, as described, a partial result, and therefore an incorrect one.

2. Self-selection sampling bias—As it might be implied from its name, this kind of bias can be caused as a result of the behavior or the decisions of the subjects themselves.43 The dictionary definition of self-selection sampling bias is:44

   A form of non-random selection for a research project in which the subjects have responded to a general invitation to take part in a survey, focus group, or other form of research in which the general area of research may be known to them in advance and towards which they have an opinion.

   This case might not seem like the classical form of self-selection sampling bias, since the witnesses did not choose to be included in the Court’s indicators. Yet I think that the act of self-selection occurred prior to the stage of current indicators examination: it occurred when that witness chooses (voluntarily) to became one.

   As described above, those witnesses choose to became ones because they felt that court can protect them and provide them with the level of security they need. Therefore, those individuals have an opinion regarding this subject, as described in the definition above.

   Moreover, I do not have any information about the profile of a common ICC witness, but there is a possibility that those who agree to testify are not the weakest or the most threatened ones. This possibility is reasonable because those individuals are willing to face a potential threat that will be caused despite the Court’s protection.

   Those reasons could lead to a problematic outcome:

   The outcomes can be very similar to selection bias, in that the sample may be insufficiently representative of the population of interest, and in the testing of hypotheses or the evaluation of an experiment it makes it difficult to separate cause-and-effect.45

   In this case, if the ones who are included in the indicators are the ones described in the last paragraph, then it would be hard to separate cause-and-effect from the data that would be supplied regarding their safety.

IV. A Proposal for a Collecting Data Tool

A. The collection of data challenge

So far, I’ve been discussing the biases and the downfalls that can be found in the measuring indicators suggested by Court. My suggestion to include the potential witnesses and their subjective sense of security or security concerns, may cause a new challenge to the Court.

The data examined under the Court’s current version of the indicators is fairly easy to collect. First, it usually can be found easily, since those security incidents or information that led to actual harm are documented in the Court’s database. Moreover, and maybe more important, this data is objective and therefore pretty easy to measure.

In contrast, the data that the offered version of the indicators is based on is quite hard to collect. First, this information cannot be found in the Court’s database, since it deals with potential witnesses that did not became witnesses. Moreover, and again maybe more importantly, this data is of the subjective kind (as mentioned above, it relates to the sense of security or security concerns of potential witnesses).

Subjective data is hard to collect since there is no other way to receive it other than from the individual that experienced it. Also, it is a problematic source of information to relay on, since there is no way to verify the validity of it.

B. The surveys solution

Due to the challenge mentioned above, I believe that the right collecting data tool for the Court to be using is anonymous and confidential surveys. If the Court wants to know what are the thoughts and fears of those potential witnesses, there is no other way to receive this information than to ask them about it. In order to receive the full picture regarding the Court’s performance security-wise, those surveys should include both witnesses and potential witnesses. Including both groups will enable Court to understand the frequency and severity of those fears and feelings of lack of security among individuals that have been in touch with the court at some stage.

Of course, this solution is not flawless. First of all, the Court needs to convince those people to participate in this survey. There probably won’t be any problem with the witnesses’ cooperation, since they cooperated well with the court so far. But potential witnesses might not be willing to participate in those surveys. This lack of cooperation can be a result of numerous reasons, including lack of faith in Court, fear of being identified with it, or lack of documents and other relevant material to back up their statements.

In order to overcome the above obstacle, the Court needs to convince those individuals that they will benefit from filling out the survey without causing this benefit to influence their response to the survey. Moreover, the surveys supplied should be accessible in the language(s) spoken and written in the conflict country and should be phrased in a clear and simple way.46 In cases of illiterate persons, a technology that would be able to overcome reading and writing difficulties needs to be considered.47

The other potential flaw in surveys is that they do not overcome the lack of ability to verify the validity of the information given mentioned above. There could be many reasons that can cause witnesses to be dishonest in their response. It could be a will to be liked or to be considered normal, that cause witnesses to reply with answers that they believes that show them in a better light. This is another potential bias known as “social desirability response set,”48 which is:

A predisposition or readiness to respond to items of a multiple-choice questionnaire with responses calculated to present oneself in a favorable light according to perceived social norms and values.49

It could also be because the potential witness fears the potential threat that caused him not to became a witness in the first place to and he or she does not want this threat to be exposed.

Notwithstanding the foregoing, I do think that the advantages of having a survey and the unique information that can be received only through it, outweighs its potential disadvantages.

C. Highlights regarding the content of the survey

I do not think that the scope of this comment, or my limited understanding in research methods, will enable me to fully describe the structure or the content of the perfect survey to evaluate the Court’s performance. Nevertheless, I do want to mention some topics that I think it’s important for the Court to ask about and clarify:

1. The first question to be asked in the survey should eliminate those potential witnesses who decided not to became witnesses for reasons other than security concerns.50 While the general willingness of potential witnesses to cooperate with the Court is important in order to understand its performance, the aim of this survey is to reserve information regarding the Court’s security performance.

2. In order for the Court to fully understand why a potential witness decided not to become a witness, it needs more than the security title. The survey should narrow down the specific reason that caused potential witnesses to feel that they cannot receive enough protection from the Court. The Court needs to know whether the fear was based on an actual threat on their life or was it a non-tangible fear. Did they expect an immediate or a more drastic protection measure than the ones offered to them?

   These questions—and a lot more—should be addressed to the potential witnesses. Identifying the source of the fear is a first step to understanding the Court performance in terms of security and maybe could even lead to the Court taking measures to improve itself.

3. The survey should include questions regarding the background of the potential witnesses. Those questions are needed in order to build a better database of witness indicators. Those indicators will allow the Court to identify in advance potential witnesses that possibly will subsequently develop security concerns that will lead them to not cooperate with the Court.

   For example, if a substantial number of women between the ages of 25 to 34, who live in remote villages, describe a similar reason for the fear they are sensing, then the Court could develop tools to try and prevent in advance the defection of the next potential witness with similar characteristics.

Conclusion

Formalization of indicators to measure the Court’s performance in terms of security is not an easy task. As I tried to demonstrate throughout this comment, the way in which one interprets the meaning of terms like performance, effectiveness or even security is critical in the process of defining the limits of those indicators.

The way in which the ICC chooses to form its indicators presents a more narrow and limited interpretation of those terms. Although this way might lead to relatively successful results or information if the scope is as narrow as the Court chose, I believe that it does not present the full picture regarding the Court’s performance.

Only after the Court overcomes the different kinds of selection bias in its performance indicators will it receive the full picture mentioned above. As I explained and demonstrated, in my opinion, the right way to do so is to expand both the pool of people and the nature of the data that would be included under those indicators. Only after this necessary redesign of the Court’s indicators can it fully evaluate its true performance.

As claimed in the last part of this comment, the next step after this evaluation would be to use the data received in order to improve the Court’s performance. Whether in the form of preventing security incidents to witnesses or in different methods to predict and prevent potential witnesses from deciding not to cooperate with the Court. The way to shape those steps is up for the Court to decide, but it’s important that, before doing so, it relies on true and accurate information regarding its performance.

Endnotes — (click the footnote reference number, or ↩ symbol, to return to location in text).

1. 1.

   Assembly of States Parties, Strengthening the International Criminal Court and the Assembly of States Parties, ICC-ASP/13/Res.5 at Annex I, ¶7(b) p.47 (Dec. 17, 2014), available online. ↩

2. 2.

   International Criminal Court, Report of the Court on the Development of Performance Indicators for the International Criminal Court, 1 (Nov. 12, 2015), [hereinafter First Report], available online, archived. ↩

3. 3.

   Id. at 6. ↩

4. 4.

   Id. ↩

5. 5.

   International Criminal Court, Second Court’s Report on the Development of Performance Indicators for the International Criminal Court (Nov. 11, 2016), [hereinafter Second Report], available online, archived. ↩

6. 6.

   International Bar Association, Witnesses before the International Criminal Court: An International Bar Association International Criminal Court Programme Report on the ICC’s Efforts and Challenges to Protect, Support and Ensure the Rights of Witnesses, 1 (Jul. 2013), available online ↩

7. 7.

   Chris Mahony, The Justice Sector Afterthought: Witness Protection in Africa, Inst. Sec. Studies, 17 (2010), available online. ↩

8. 8.

   Witnesses Before the ICC, supra note 6, at 14. ↩

9. 9.

   Id. at 12. ↩

10. 10.

    Id. ↩

11. 11.

    Dermot Groome, No Witness, No Case: An Assessment of the Conduct and Quality of ICC Investigations, 3 Penn St. J. L. Int’l Aff. 1, 3 (Apr. 2014), available online.

    (“On my first day as a prosecutor in the Manhattan District Attorney’s Office, my bureau chief, Warren J. Murray Jr., began our training by introducing us to a few of his well-known prosecutorial maxims or ‘rules.’ Rule number one, his first and most important was: ‘No witness—no case.’”). ↩

12. 12.

    Witnesses Before the ICC, supra note 6, at 15. ↩

13. 13.

    Id. ↩

14. 14.

    William A. Schabas, An Introduction To The International Criminal Court, 248–49 (4th ed. 2011). ↩

15. 15.

    Id. ↩

16. 16.

    Carla Ferstman, Video Lecture, Comment on the Victims Question: How can Victims’ Representation at the ICC be Improved and Victims’ Rights be Protected?, ICC Forum, at 23:08 (Feb. 12, 2013), available online. ↩

17. 17.

    Sheila Vélez, Katanga and Ngudjolo Chronicle #2: Head of the OTP Investigation Team, Aegis Trust (Nov. 25, 2009), available online. ↩

18. 18.

    Id. ↩

19. 19.

    Id. ↩

20. 20.

    Rome Statute of the International Criminal Court, Adopted by the United Nations Diplomatic Conference of Plenipotentiaries on the Establishment of an International Criminal Court, Jul. 17, 1998, UN Doc. A/CONF.183/9 [hereinafter Rome Statute], Art. 43(6), available online. ↩

21. 21.

    Id. ↩

22. 22.

    International Criminal Court, Understanding the International Criminal Court 40 (Aug. 22, 2013), available online, archived.

    (“[T]he Initial Response System is a 24/7 emergency response system that enables the Court, where feasible, to extract witnesses to a safe location should they be targeted or in fear of being targeted. Other operational protective measures include educating witnesses on the importance of confidentiality and cover stories or agreeing on an emergency backup plan.”). ↩

23. 23.

    E.g., id.

    (“Such measures may consist of face/voice distortion or the use of a pseudonym. Separate special measures can be ordered by the Court for traumatised witnesses, a child, an elderly person or a victim of sexual violence. These can include facilitating the testimony of witnesses by allowing a psychologist or family member to be present while the witness gives testimony or the use of a curtain to shield the witness from direct eye contact with the accused.”). ↩

24. 24.

    Id. ↩

25. 25.

    Id. ↩

26. 26.

    Strengthening the ICC and ASP, supra note 1, at ¶ 7(b). ↩

27. 27.

    First Report, supra note 2, at 1. ↩

28. 28.

    Id. ↩

29. 29.

    Id. ↩

30. 30.

    Id. at 6. ↩

31. 31.

    Id. ↩

32. 32.

    Id. ↩

33. 33.

    Second Report, supra note 5, at 1. ↩

34. 34.

    Id. at 13. ↩

35. 35.

    Id. at 14. ↩

36. 36.

    Mark Elliot et al., A Dictionary of Social Research Methods, Oxford Ref. (2016), available online (last visited Jul. 20, 2017). ↩

37. 37.

    Noel Castree et al., A Dictionary of Human Geography, Oxford Ref. (2013), available online (last visited Jul. 20, 2017). ↩

38. 38.

    First Report, supra note 2, at 6. ↩

39. 39.

    David Collier & James Mahoney, Insights and Pitfalls: Selection Bias in Qualitative Research, 49 World Pol. 56 (1996), available online. ↩

40. 40.

    James J. Heckman, Sample Selection Bias as a Specification Error, 47 Econometrica 153, 154–155 (Jan. 1979), available online. ↩

41. 41.

    Id. ↩

42. 42.

    Id. at 153.

    (“Sample selection bias may arise in practice for two reasons. First, there may be self selection by the individuals or data units being investigated. Second, sample selection decisions by analysts or data processors operate in much the same fashion as self selection.”). ↩

43. 43.

    Collier & Mahoney, supra note 35, at 59. ↩

44. 44.

    John Duignan, A Dictionary of Business Research Methods, Oxford Ref. (2016), available online (last visited Jul. 20, 2017). ↩

45. 45.

    Id. ↩

46. 46.

    Ferstman, supra note 16, at 33:47.

    (Carla Ferstman describes how a difficult application process, as a result of unclear forms or the lack of forms in the victim’s language, can cause the victim to not cooperate with the Court). ↩

47. 47.

    Id. ↩

48. 48.

    “Social Desirability Response Set” in Andrew M. Colman, A Dictionary of Psychology, Oxford Ref. (2015), available online (last visited Jul. 20, 2017). ↩

49. 49.

    Id. ↩

50. 50.

    Kelly Dawn Askin, Can the ICC Sustain a Conviction for the Underlying Crime of Mass Rape Without Testimony from Victims?, in Contemporary Issues Facing the International Criminal Court (Richard H. Steinberg ed., 2016), earlier version (Jun. 26, 2012) available online; Ferstman, supra note 16.

    (There are numerous reasons that can cause potential witnesses to decide not to testify. Among them can be found: the lack of trust in the ICC as a whole, the cultural differences that could lead to a different perception of justice, the trauma that accompanies the testimonial process, and the distant and culturally different location of the court that could be intimidating to some. This is just a short list; there are many more). ↩