The Cyber Evidence Question — Comments

Change Sort Order
Now: Oldest First

Comment on the Cyber Evidence Question: “To what extent can cyber evidence repositories, and digital and open-source evidence, facilitate the work of the OTP, and the ICC more generally?”

The Promise and Problems of Open Source Evidence in ICC Investigations and Trials

I. Introduction

On August 15, 2017, the Pre-Trial Chamber of the International Criminal Court (ICC) issued an arrest warrant for senior Libyan military commander, Mahmoud Al-Werfalli.1 Al-Werfalli is accused of personally committing or ordering thirty-three murders in Benghazi between June 2016 and July 2017. Importantly, this is the first ICC arrest warrant that relies significantly on evidence that was uploaded and later collected from social media. Specifically, the warrant describes events captured in several videos—which were posted on social media sites such as Facebook—that appear to show Al-Werfalli shooting individuals and ordering others to shoot.2

The ICC’s reliance on these videos is telling. More and more, today’s conflicts see large amounts of videos and photos that document crimes find their way online.3 Sometimes bystanders are able to record events with their cellphones and share the recordings as a way of alerting the wider world to the state of affairs in areas of conflict. Perhaps surprisingly often, the incriminating material is posted by the perpetrators themselves.4 This new mine of incriminating evidence is both promising and puzzling for courts the world over, and the ICC is no different. Although potentially powerful, open source evidence continues to be plagued with doubts as to authenticity.5 Obtaining evidence from online sources is one thing, but then next figuring out if and how to use the evidence in a truth-finding forum such as a trial is another. In the case of Al-Werfalli, it remains an open question how the videos relied upon in his arrest warrant would be used by the Office of the Prosecutor (OTP) at trial, or if the Trial Chamber would even find the videos admissible.

In this comment, I explore the benefits and limitations of the use of open source evidence by the ICC, particularly at two different procedural stages: investigation and trial. I argue that open source evidence has the potential to alleviate some of the considerable investigatory constraints the ICC continues to face, and also supplement the Court’s over-reliance on live witness testimony during trial proceedings. However, the Court’s flexible and amorphous evidentiary standard—which favors permitting evidence but attributing it little to no weight to counteract doubts as to reliability—threatens to undermine the potential benefits of open source evidence. To counteract this, I propose development of and reliance on external verification methods which have been used by international tribunals in the past to ensure only the most reliable open source evidence is used in criminal proceedings.

II. The Need for Open Source Evidence at the ICC

The ICC is mandated with investigating and prosecuting the most serious crimes known to the global community: genocide, war crimes, and crimes against humanity.6 Since it was established in 2002, the ICC has indicted forty-four individuals.7 However, only nine have seen their cases proceed to trial, and the Trial Chamber has returned a guilty conviction for only five.8 The OTP’s struggle to advance cases to a successful outcome is due to many factors. But one of the main challenges the OTP faces is collecting and presenting compelling evidence that more definitively links the accused to the crime.9

Digital open source evidence holds a potential solution to many of the barriers that exist between the OTP and quality evidence, both at the investigative and trial stages. At the investigation stage, the emerging availability of open source evidence posted online provide researchers at the ICC an incredible and growing mine of documentary evidence. On top of that, the type of documentation posted on open source platform has the potential to be highly probative and, in some instances, it is generated by the defendant themselves. This makes open source evidence possibly very useful in a trial setting. Both of these attributes—the growing online availability and probative nature of the pieces of evidence—make the use of open source evidence by the OTP in the investigatory and trial settings a promising new source of evidence to explore.

A. Availability

Because of the wide-spread and growing availability of technology across the globe, open source information can provide courts with access to conflict zones or protected spaces that are otherwise inaccessible to the reaches of more traditional criminal investigation and evidence collection.10 This is because civilians on the ground—be they bystanders or participants in a particular event—can increasingly capture photos and videos that create a window through which the wider world can view an event, sometimes even in real-time.11 Many users are then uploading the photo and video documentation they create to online platforms, such as social media sites like Facebook or YouTube, or sending the documentation to online news sources.

In addition, a growing number of activist-lead databases are popping up that have sifted through the open source documentation specifically in search for those that capture human rights violations. For example, the Syrian Archive, an organization that preserves evidence of human rights violations in the Syrian conflict, has gathered videos and photographs from hundreds of sources that document chemical weapon attacks in Syria between 2012–2018.12 The goal of the Syrian Archive is to curate and analyze the documentation to create a tool that can be used for reporting and accountability purposes.13 Similarly, Amnesty International launched a Digital Verifications Corps in 2016, which is aimed at training researchers to verify online images for potential future accountability efforts.14 The intersection between international criminal investigations and this emerging evidence source is clear, and open source information has already been used as a tool in investigations into crimes in conflict zones across the globe.15

From the perspective of the ICC, the growing pervasiveness of this open source evidence is powerful, particularly as an investigatory mechanism. The ICC continues to face many challenges when it comes to performing thorough and timely investigations of potential human rights violations. One of the main constraints on the ICC’s investigatory ability is access. Traditional on-the-ground investigation teams typically reach sites of potential war crimes or atrocities weeks, months, or sometimes years after the events occurred.16 As a result, critical evidence is often lost or destroyed. Because open source evidence in conflict zones is largely user-generated, and the Internet enables users to share images with the world instantaneously, the traditional constraints of criminal investigation can be alleviated by open source evidence.

Moreover, the ICC relies heavily on states participation in evidence collection and has few tools in their toolbox to compel state participation if the state proves uncooperative. For instance, the OTP faced major problems with state non-cooperation while investigating post-election violence in Kenya.17 The Prosecutor was ultimately forced to withdraw the charges against Kenyan President Uhuru Kenyatta because the investigation failed to produce the evidence necessary to proceed. This failure has come to epitomize the immense investigatory challenges the OTP faces and the lack of leverage available to remediate the challenges.18 Not only does stonewalling by state actors thwart the truth-seeking and accountability efforts that the ICC was designed to create, but even more dangerously, consistent non-cooperation from states under ICC jurisdiction could undermine the legitimacy of the Court and pave the way for future non-cooperation in other conflicts.19

In light of these challenges, open source evidence made freely accessibly on the Internet provides a potential solution to some of the investigatory constraints the ICC experiences. Due to the nature of open source platforms which provide access to material from anywhere in the world, some of the roadblocks and pitfalls that the ICC is prone to with on-the-ground investigations can be circumvented. The OTP has recognized this, and in recent years has increased its capacity to conduct cyber-investigations and analyze digital evidence.20 In addition, the OTP has continuously tried to foster external partnerships with both national law enforcement and transnational organizations that have more sophisticated frameworks for collecting and analyzing digital evidence.21 This demonstrates the OTP’s awareness of the critical role that open source, digital evidence will play in ongoing and future investigations.

B. Probative Value

Moving beyond the investigation stage, open source evidence provides another significant benefit to the OTP at trial. One of the challenges that the Prosecutor faces at trial is presenting evidence that sufficiently demonstrates to the judges the criminal intent or state of mind of the accused. Much of the traditional evidence that is typically relied upon is testimonial or circumstantial, and this has led to difficulties for the Prosecutor in successfully convicting defendants at trial. For instance, the Trial Chamber recently acquitted Laurent Gbagbo, former president of Côte d’Ivoire and his close aid Charles Blé Goudé.22 The two were accused of crimes against humanity for their role in the post-election violence that erupted in the Côte d’Ivoire following the 2010 presidential election.23

The acquittal was a major blow for the OTP, which spent several years and a large amount of its resources bringing the case to trial. Critics have been quick to point out that the ICC cannot succeed at its mission if the OTP is failing to select and present successful cases. A significant evidentiary issue that the Prosecutor continues to face at trial is a heavy reliance on live witness testimony. Live witness testimony generates a host of issues and research illustrates that witness interference has been alleged in nearly every case before the ICC.24 The OTP’s 2016–2018 Strategic Plan noted that witness interference may have occurred in eight of the nine first cases heard by the ICC.25 In light of this, the Court’s over-reliance on witness testimony—especially in terms of comparative probative value to other forms of evidence—compound the concerns about witness reliability. Witness interference can undermine the truth-seeking purpose of the Court’s proceedings and lead to wrongful convictions or acquittals. On a practical level, protecting the integrity of witness testimony and ensuring the safety of those who do testify can be a resource drain on the Court and preoccupy the proceedings. From an internal perspective, the OTP has recognized the limitation of traditional bodies of evidence in international criminal proceedings. The 2016–2018 Strategic Plan recognized the need to diversify the body, and the types of evidence it collects and presents at trial, and laid out the strategic steps necessary to develop internal capacity and foster external partnerships in order to do so.26

Historically, some of the most valuable and probative types of evidence available has been the evidence that is generated by the perpetrators themselves. For instance, during the Nuremberg trials, the Chief Prosecutor was able to move away from relying on witness testimony and focus on documentary evidence that demonstrated that genocide was the modus operandi of the Nazi party.27 Because the Nazi’s were such prolific and detailed record keepers, the Chief Prosecutor was able to use those records to link defendants to specific atrocities and acts.28 In large part because there was this substantial paper trail, the Chief Prosecutor avoided victim testimony that would be open to cross-examination and criticisms as to its reliability.29 Because the records used had been created by the perpetrators themselves, the probative value of the documents was immense as it demonstrated the defendant’s state of mind and criminal intent. The choice to rely heavily on the documentary evidence, as opposed to witness testimony, led in large part to the success of the Nuremberg trials at accomplishing one of its main goals, ensuring accountability.

While not without criticism, the Nuremberg trials certainly laid the groundwork for the use of evidence in international criminal proceedings. In the following years, the International Criminal Tribunal for Yugoslavia and the International Criminal Tribunal for Rwanda (ICTR) continued the legacy and pioneered new mediums of evidence. While both tribunals relied far more on witness and expert testimony than the Nuremberg trials had, there were similarly also opportunities to successfully advance evidence that had been created by the perpetrators themselves.30 For instance, in one case before the ICTR, the Prosecution relied on radio broadcasts to successfully convict three influential media executives based on the executives’ role in airing broadcasts which explicitly called for violence against Tutsis during the Rwandan genocide.31 Similarly to the Nazi records in the Nuremberg trials, it was the broadcasts themselves—evidence generated by the perpetrators—which was invaluable in proving the criminal intent and purposeful behavior of the defendants.32

Continuing in this vein, open source digital material is a prime example of how a different type of evidence can be leveraged in court to corroborate, supplement, and enhance traditional evidence sources. The arrest warrant of Mahmoud Al-Werfalli, discussed in the Introduction, is an example of how the Court can use perpetrator-generated photos, videos, and other documentary evidence to advance cases against those that commit human rights violations. And, unlike a witness, a piece of video evidence “can’t be cross examined. It can’t forget. It can’t be bribed.”33

As the Al-Werfalli case has yet to proceed to trial—the suspect is currently still at large—it remains to be seen how the ICC reacts to the use of social media evidence in that case. But national courts have already had success in utilizing social media evidence to obtain convictions against Syrian war criminals. For instance, in 2016, a Swedish national court convicted a former Syrian soldier of war crimes.34 This was a landmark ruling in the wake of the Syrian civil war, as it was the first time a member of the Syrian military had been adjudicated for war crimes. The conviction also demonstrated how evidence collected from social media can (and cannot) be used in prosecutions. A key piece of evidence was a picture the defendant had posted on Facebook, showing himself standing over a pile of bodies, smiling.35 German national courts have also leveraged the abundance of social media evidence to generate convictions, paving the way for the ICC to admit similar types of evidence in the Al-Werfalli case, and potential cases yet to come.36

III. Limitations of Open Source Evidence

Despite the many benefits that open source evidence would seem to offer, the ICC and other courts have been slow to embrace the use of it. This is largely due to concerns about the potential limitations and inequities that open source evidence forces courts to consider. While scholars have advanced many concerns and criticisms of open source evidence in the context of criminal proceedings, this comment focuses primarily on those implicated in the investigative and trial procedural stages.

A. Source Integrity and Reliability

Critics of the use of open source evidence in a criminal context are generally wary of the reliability of such types of information. One major issue is that open source information that is publicly available online is often not easily traceable to an identifiable source. This impacts the ability of the investigators to authenticate the evidence and assess the sources’ reliability. A particularly potent threat emerging at the forefront of technology is that of “deepfakes”—videos that are manipulated to make it look like people said or did something they did not.37 The alarming possibility that deepfakes can and are used to potentially spread falsehoods to news sources, criminal investigators, or government officials is a scary new reality. For instance, in May 2019, a doctored video of US Congresswoman Nancy Pelosi appeared on Facebook, and quickly amassed more than 2.5 million followers and was shared to YouTube, Twitter, and other online platforms.38 The video, in which Pelosi appeared to be stammering and slurring her words during a press conference, captured national attention in the US and highlights how emerging technology can blur the lines between truth and distortion and how open source platforms can facilitate the rapid dissemination of lies.

Because photos and videos can be uploaded to open source platforms relatively anonymously and shared rapidly, verification of open source information in the process of collecting criminal evidence is of paramount importance. Clearly, the possibility of false or unreliable evidence making its way into an international criminal investigation is antithetical to the truth-seeking function of the courts. Currently, the ICC does not internally have the capacity to conduct digital forensics at the scale today’s investigations require, so the Court relies heavily on partnerships with human rights NGOs, states, and other external partners to locate, preserve, verify, and analyze digital evidence.39

Additionally, there are of course implications on the rights of the accused when the reliability of a type of evidence is at issue. Further complicating that fundamental point is the reality that the sheer amount of resources required to mine and verify pieces of open source evidence for use in a criminal investigation creates an inequality of arms based on a defendant’s inability to weaponize open source evidence to the same degree as prosecution teams.40 The inherent inequality between the use of open source evidence to incriminate rather than exculpate, as well as the deep uncertainty about the reliability of open source evidence, will likely lead to heavy challenges by defendants to its growing use by the ICC.

B. Admissibility and Evidentiary Weight

The potential promise of the growing new repositories of user generated evidence is of little value to the OTP if it cannot be used in trials. Luckily, perhaps, the evidentiary standard which the ICC is built around affords the Court a large degree of judicial discretion in admitting evidence into consideration. Rule 63(2) of the ICC’s Rules of Procedure and Evidence gives judges the authority to “assess freely all evidence submitted in order to determine its relevance or admissibility.”41 Somewhat surprisingly, the Rules do not indicate what, if any, evidence would be prima facie inadmissible. The Rules merely require the judges to consider the evidence in light of their relevance, probative value, and prejudicial impact.42 This flexible and open-ended evidentiary standard tends to suggest that the Court will continue down the path of admitting open source evidence that is presented by the OTP.

The ICC does have some protocol in place to guide the submission of pieces of digital evidence, which will apply to open source evidence as well. The Court has adopted an e-Court Protocol, which is designed to ensure authenticity, accuracy, confidentiality, and preservation of the record of ICC proceedings.43 This protocol requires that the chain of custody of a piece of evidence be established using metadata, and the identity of the source be disclosed, among other things. The development and implementation of this Protocol suggests that the Court can adapt to the ways in which technology is changing the way evidence is captured, which is a promising sign for acceptance of open source evidence.

However, although many types of open source evidence may likely be deemed admissible, that does not necessarily mean that they will be viewed by ICC judges as particularly probative. In practice, the ICC has generally admitted most evidence that it is presented with. Any remaining questions regarding the quality of the evidence has generally detracted from the weight that the judges place on the evidence to inform their decision.44 The ICC generally favors live, in-person witness testimony, although the OTP has been committed to diversifying its evidence base due to the many constraints that large scale witness testimony presents.45

Although, on one hand, the flexible evidentiary standard gives the ICC plenty of leeway, it also makes it difficult to predict how judges of the ICC will interact with open source evidence as it becomes more prevalent. One example of an evidentiary issue implicated by the use of open source evidence is exemplified by the hearsay doctrine—evidence of facts outside the personal knowledge of the testifying witness. Many national courts have common law rules that have developed a strong sense of how to deal with hearsay evidence. The ICC on the other hand has no explicit rule against hearsay evidence and lacks clear guidance on how to grapple with such evidentiary problems that hearsay evidence may provoke. This could prove challenging in the context of assessing the probative value of a piece of open source evidence, pulled from an anonymous source on an online platform. The ICC has admitted into evidence anonymous e-mails presented by the Prosecutor that would raise hearsay objections in many common law traditions, and, indeed, these generated strong objections regarding their underlying veracity from the defense team.46 Such objections to the admission of similar types of evidence will likely grow louder as the prevalence of online open sources continues to grow.

Admission of a piece of evidence is just one piece of the equation of how open source evidence will be viewed by the Trial Chamber. When it comes to questions of probative value, open source evidence seems particularly vulnerable to attack. Although it remains an open question, some insight into how judges will perceive open source evidence can be gleaned from the case against Thomas Lubanga. Lubanga, a warlord from the Democratic Republic of Congo, was tried and convicted of conscripting child soldiers. During the trial against him, the OTP relied on ten videos to demonstrate that Lubanga was aware that children were being conscripted as soldiers into his ranks.47 The videos in this case did not come from open source platform, but rather were collected by NGOs on the ground during the conflict.48 The videos were also accompanied by a witness who provided testimony as to what was occurring in the videos.49 Lubanga was ultimately convicted, but on appeal his defense team raised many challenges to the use of the videos by the OTP.50 These arguments were rejected by a majority of the judges in the Appeals Chamber—however, the opinion of dissenting Judge Anita Ušaka provides insight into reservations that still exist around the probative value of digital evidence such as video.

Judge Ušaka’s dissent argued that the Prosecutor’s use of video evidence—offered to demonstrate to the judges that the soldiers conscripted by Lubanga’s forces were children—asked the Court to make impressionistic and speculative judgments that were not based on objective criteria.51 Some of these concerns would likely be exacerbated by the presentation of digital evidence procured from open source platforms. Much open source evidence being proliferated online may not be easily traced back to a source or could be presented by the Prosecutor without corroboration from a witness testifying alongside of it, such as in the Lubanga trial. As such, in the process of evaluating the weight of such a piece of evidence, the Court may necessarily be forced to make impressionistic judgments regarding the evidence’s reliability and truthfulness. Depending on the circumstances of a case, ICC judges may find themselves more in line with Judge Ušaka’s opinion on the Lubanga decision.

The Court also must consider the prejudicial impact of open source evidence on the defendant. Scholars have been quick to point out that the use of open source evidence in courtrooms raises substantial concerns regarding the rights of the accused.52 Concerns about digital manipulation, data security, and the integrity of the investigation process, as discussed above, all provoke valid questions about whether the Court can admit evidence pulled from online sources.

In practice, the Court seems particularly wary of evidence presented from anonymous sources, which could prove problematic in the context of open source platforms. For instance, in the Gbagbo case, the Prosecutor presented evidence captured from an anonymous source.53 The judges noted that this piece of evidence was “highly problematic” and without information regarding a source, it was “impossible to determine what probative value to attribute the information.”54 It appears that the ICC is wary of relying on information from anonymous sources without an accompanying witness or expert to articulate what the evidence discloses.55 This could prove challenging as the OTP seeks to introduce evidence derived from open source platforms, which comes largely from online sources that may be difficult to trace. Ultimately, the pending case against Mahmoud Al-Werfalli is the first-time open source evidence has played such a significant role, so it is yet to be seen how the ICC will react to its use at trial.

IV. The Way Forward

Despite the challenges associated with open source evidence in international criminal proceedings, the reality is that it is likely not going away any time soon. Rather, the OTP has recently re-dedicated itself to utilizing open source and other types of digital evidence in the course of building its cases. In 2013, the OTP created a Digital Forensics Team and the growing case around Al-Werfalli is a strong indication that digital and open source evidence will continue to play a prominent part in the OTP’s investigatory and evidentiary strategy moving forward. As such, Court investigators and lawyers must strategize on how to maximize open source evidence’s potential while mitigating the inherent limitations, particularly as to its reliability and probative value.

The questions around open source evidence’s reliability and probative value in criminal investigations and trials are of course not completely distinct. A strong sense that a piece of evidence is reliable naturally engenders a greater degree of weight that it will carry in the courtroom. Reliability is not necessarily the only factor involved in assessing the probative value of a piece of evidence. But creating a culture that is confident that the most reliable evidence is being curated and advanced by the OTP will feed into the weight that the judges can place on the evidence they are presented with at trial.

One option to combat the creation and distribution of deepfakes or other types of digital misinformation is to impose legal liability on the bad actors who make them. Many legal scholars have already contemplated the usefulness of such a liability scheme. And some individual states have already begun to respond to this new landscape of digital technologies with new laws meant to discourage deepfakes. For instance, a bill passed in the UK created penalties for pornographic deepfakes.56 In the United States, similar federal legislation has been introduced in Congress, and several states are likely to follow suit in the near future.57 These legislative pushes could potentially make the creation of certain types of deepfakes a criminal offense.

However, other scholars have argued that this is approach is too piece-meal and doesn’t address the reality of our digital world which exists largely independent of state borders.58 Ultimately, criminal investigators and lawyers need to become more adept at detecting deepfakes and, conversely, verifying credible sources. This will likely involve multiple methods of information corroboration before determining that a particular piece of open source information is reliable. For instance, metadata can be used to check the time, date, and place that a digital video or photograph was created.59 This type of information is already required by the ICC in order for a piece of digital evidence to be considered admissible.60

That said, metadata can be faked and should not be relied upon exclusively to verify the truthfulness of a piece of digital information. Additionally, access to metadata can prove challenging in its own right when dealing with third-party open source platforms. While many of these platforms, such as Facebook, YouTube, and Twitter, do store the metadata of information posted on their platforms, they have so far been hesitant to collaborate with international criminal investigators due to user privacy and other liability issues.61 However, additional techniques allow investigators to supplement the use of metadata to verify the source of a piece of information.

Best practices in digital forensics indicates that investigators should seek to corroborate open source evidence based on external factors. For instance, investigators can use satellite images to help verify locations, weather records can be used to verify consistency at the particular time and date the metadata alleges the information was created at, etc.62 Many of these best practices are already being incorporated into the frameworks of investigatory NGOs, and have been in practice in international criminal proceedings in the past. For instance, the ICTR tried a case that involved video evidence of a rally along with a transcript of a radio broadcast covering the rally.63 The ICTR found that the combination of these pieces of evidence reliably supported a finding that the accused was in attendance at the rally.64 In another case, the ICTR relied on a similar combination of video footage and transcript to find that the defendant was exercising military control.65 These provide examples of how international courts can triangulate multiple pieces of evidence to create greater assurances of reliability.66

In addition to focusing on external factors to verify reliability, a strong chain of custody can increase the reliability of a piece of open source evidence. A chain of custody is the history of persons who have had a piece of evidence within their control.67 In court proceedings, establishing chain of custody is generally done by testimony of the persons who controlled the electronic evidence, verifying that the piece of evidence as presented in court is substantially the same as it was at the beginning of its lifetime.68 This ensures transparency around how the evidence was collected and stored, which can in turn be protective of fair trial rights of the defendant and the reliability of the evidence presented to the judges.

There are many additional techniques available to corroborate, verify, and authenticate open source digital evidence. The OTP continues to develop its own internal capacity as well as rely on critical external partnerships to conduct the necessary digital forensics to ensure the evidence used in criminal proceedings is accurate and reliable. As technology changes, it’s likely that different and new challenges and possibilities to digital verification will emerge.

V. Conclusion

Today, citizens and journalists are documenting human rights violations on the frontlines of emerging conflicts and sharing that information online. Social media, online news sources, and other media sharing sites have the power to make everyday people journalists, documentarians, and investigators. The OTP must harness the powerful changes technology is bringing to how the world witnesses, experiences, and communicates around conflict and international crimes. Open source evidence provides investigators the possibility to gain access to inaccessible places and curate information on a level that has never been possible before. Open source evidence also provides international criminal prosecutors the opportunity to tell the story of victims more effectively than traditional evidentiary sources can on their own.

Despite the limitations that open source evidence might pose, and the challenges that may come with convincing the Court to accept it, the benefits will likely make open source evidence more and more popular in international criminal investigations. The future case law, beginning perhaps with the Al-Werfalli case should it advance to trial, will prove pivotal in how evidence is generated, collected, assessed, and ultimately used as a tool for justice at the ICC.

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

  1. 1.

    The Prosecutor v. Mahmoud Mustafa Busayf Al-Werfalli, ICC-01/11-01/17, Warrant of Arrest (PTC I, Aug. 15, 2017), available online.

  2. 2.

    Id.

  3. 3.

    Rebecca J. Hamilton, User-Generated Evidence, 57 Colum. J. Transnat’l L. 1, 17–18 (Nov. 7, 2018), available online; See also Tolu Ogunlesi & Stephanie Busari, Seven Ways Mobile Phones Have Changed Lives in Africa, CNN, Sep. 14, 2012, available online.

  4. 4.

    Emerson T. Brooking & P.W. Singer, War Goes Viral: How Social Media is Being Weaponized Across the World, The Atlantic (Nov. 2016), available online.

  5. 5.

    Rebecca J. Hamilton, The Hidden Danger of User-Generated Evidence for International Criminal Justice, Just Security (Jan. 23, 2019), available online.

  6. 6.

    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, as amended [hereinafter Rome Statute], Art. 5(1), available online.

  7. 7.

    International Criminal Court Fast Facts, CNN, Apr. 7, 2020, available online (last visited May 28, 2020).

  8. 8.

    Id.

  9. 9.

    UC Berkeley Human Rights Center, Digital Fingerprints: Using Electronic Evidence to Advance Prosecutions at the International Criminal Court (Feb. 2014) [hereinafter Digital Fingerprints], available online.

  10. 10.

    Nikita Mehandru & Alexa Koenig, ICTS, Social Media, and the Future of Human Rights, 17 Duke L. & Tech. Rev. 129, 145 (2019), available online.

  11. 11.

    See Jeff Deutch & Hadi Habal, The Syrian Archive: A Methodological Case Study of Open-Source Investigation of State Crime Using Video Evidence from Social Media Platforms, 7 St. Crim. J. 46 (2018), paywall, doi; Hamilton, supra note 3.

  12. 12.

    Deutch & Habal, supra note 11.

  13. 13.

    Id.

  14. 14.

    Conor Fortune, Digitally Dissecting Atrocities: Amnesty International’s Open Source Investigations, Amnesty Int’l (Sep. 26, 2018), available online.

  15. 15.

    See, e.g., How a Werfalli Execution Site Was Geolocated, Bellingcat (Oct. 3, 2017), available online; See, e.g., Anna Banchik, Leenah Bassouni, Michael Elsanadi, Hannah Ellis, Sonia Hamilton, Natalia Krapiva, Danny Lee & Andrea Trewinnard, Hum. Rts. Investigation Lab, Chemical Strikes on Al-Lataminah, March 25 & 30, 2017 (Jan. 18, 2018), available online; Mehandru & Koenig, supra note 10, at 133.

  16. 16.

    See Caroline Buisman, Delegating Investigations: Lessons to be Learned from the Lubanga Judgment, 11 Nw. J. Int’l Hum. Rts. 30, 52 (2013), available online

    (describing one-month delay after the collapse of the Ghadafi regime before investigators from OTP arrived in Libya).

  17. 17.

    The Prosecutor v. Uhuru Muigai Kenyatta, ICC-01/09-02/11, Second decision on Prosecution’s application for a finding of non-compliance under Article 87(7) of the Statute (TC V, Sep. 19, 2016), available online.

  18. 18.

    Hamilton, supra note 3.

  19. 19.

    See Thomas Obel Hansen, Masters of Manipulation: How the Kenyan Government is Paving the Way for Non-Cooperation with the ICC, Open Democracy (May 30, 2012), available online.

  20. 20.

    See Digital Fingerprints, supra note 9, at 4.

  21. 21.

    Id.

  22. 22.

    The Prosecutor v. Laurent Koudou Gbagbo and Charles Blé Goudé, ICC-02/11-01/15, Delivery of Decision (TC I, Jan. 15, 2019) (Oral Decision), available online.

  23. 23.

    Situtation in the Republic of Côte d’Ivoire, ICC-02/11-26, Warrant of Arrest for Laurent Koudou Gbagbo (PTC III, Nov. 23, 2011), available online.

  24. 24.

    See, e.g., Open Society Justice Initiative, Witness Interference in Cases Before the International Criminal Court (Nov. 2016), available online.

  25. 25.

    Office of the Prosecutor, ICC, Strategic Plan 2016–2018 (Jul. 6, 2015), available online.

  26. 26.

    Id.

  27. 27.

    Patricia M. Wald, Dealing with Witnesses in War Crime Trials: Lessons from the Yugoslav Tribunal, 5 Yale Hum. Rts. & Dev. L.J. 217 (2002), available online.

  28. 28.

    Id.

  29. 29.

    Jennifer Stanley, Note: From Nuremberg to Kenya: Compiling the Evidence for International Criminal Prosecutions, 49 Vand. J. Transnat’l L. 819 (Aug. 8, 2016), available online.

  30. 30.

    Lindsay Freeman, Digital Evidence and War Crimes Prosecutions: The Impact of Digital Technologies on International Criminal Investigations and Trials, 41 Fordham Int’l L.J. 283 (2018), available online.

  31. 31.

    Id. at 301.

  32. 32.

    Wald, supra note 27.

  33. 33.

    Megha Rajagopalan, The Histories of Today’s Wars Are Being Written on Facebook and YouTube. But What Happens When They Get Taken Down?, Buzzfeed, Dec. 22, 2018, available online.

  34. 34.

    Anne Barnard, Syrian Soldier is Guilty of War Crime, a First in the 6-Year Conflict, N.Y. Times, Oct. 3, 2017, available online.

  35. 35.

    Id.

  36. 36.

    Cathrin Schaer, Prosecuting Syrian War-Crimes Suspects from Berlin, The Atlantic (Jul. 31, 2019), available online.

  37. 37.

    See, e.g., Jon Christian, Experts Fear Face Swapping Tech Could Start an International Showdown, The Outline, Feb. 1, 2018, available online.

  38. 38.

    Doctored Nancy Pelosi Video Highlights Threat of “Deepfake” Tech, CBS, May 25, 2019, available online.

  39. 39.

    See, e.g., Alexa Koenig, “Half the Truth is Often a Great Lie”: Deep Fakes, Open Source Information, and International Criminal Law, 113 Am. J. Int’l L. Unbound 250 (Aug. 19, 2019), available online, doi.

  40. 40.

    See Hamilton, supra note 3.

  41. 41.

    International Criminal Court, Rules of Procedure and Evidence, ICC-ASP /1/3, R. 63(2) (2d ed. Sep. 9, 2013), available online, archived.

  42. 42.

    The Prosecutor v. Jean-Pierre Bemba Gombo, ICC-01/05-01/08, Decision on the Prosecution’s Application for Admission of Materials into Evidence Pursuant to Article 64(9) of the Rome Statute of 6 September 2012 (TC III, Oct. 8, 2012), available online.

  43. 43.

    The Prosecutor v. Callixte Mbarushimana, ICC-01/04-01/10, Decision amending the e-Court Protocol (Pre-trial Chamber I, Apr. 28, 2011), available online.

  44. 44.

    Aida Ashouri, Caleb Bowers & Cherrie Warden, The 2013 Salzburg Workshop on Cyber Investigations: An Overview of the Use of Digital Evidence in International Criminal Courts, 11 DEESLR 115, 118 (2014), available online, doi

  45. 45.

    Id. at 3.

  46. 46.

    The Prosecutor v. Thomas Lubanga Dyilo, ICC-01/04-01/06, Decision on the confirmation of charges (PTC I, Jan. 29, 2007), available online.

  47. 47.

    The Prosecutor v. Thomas Lubanga Dyilo, ICC-01/04-01/06 A 5, Judgment on the appeal of Mr Thomas Lubanga Dyilo against his conviction (AC, Dec. 1, 2014) [hereinafter Lubanga Appeals Judgment], available online.

  48. 48.

    Elizabeth O’Shea & Kelly Matheson, From Camera to Courtroom: Using Video Evidence for International Prosecutions, Int’l Just. Monitor (Oct. 4, 2016), available online.

  49. 49.

    Id.

  50. 50.

    See Lubanga Appeals Judgment, supra note 47, ¶ 216; see also The Prosecutor v. Thomas Lubanga Dyilo, ICC-01/04-01/06, Mr. Thomas Lubanga’s appellate brief against the 14 March 2012 “Judgment pursuant to Article 74 of the Statute” (AC, Dec. 3, 2012), available online.

  51. 51.

    The Prosecutor v. Thomas Lubanga Dyilo, ICC-01/04-06 A 5, Dissenting Opinion of Judge Anita Ušacka (AC, Dec. 1, 2014), available online.

  52. 52.

    See Mehandru & Koenig, supra note 10.

  53. 53.

    See Hamilton, supra note 3.

  54. 54.

    The Prosecutor v. Laurent Koudou Gbagbo, ICC-02/11-01/11, Decision adjourning the hearing on the confirmation of charges pursuant to article 61(7)(c)(i) of the Rome Statute (PTC I, Jun. 3, 2013), available online.

  55. 55.

    See Mehandru & Koenig, supra note 10.

  56. 56.

    See, e.g., Dan Sabbagh & Sophia Ankel, Call for Upskirting Bill to Include “Deepfake” Pornography Ban, The Guardian, Jun. 21, 2018, available online.

  57. 57.

    DEEP FAKES Accountability Act, H.R. 3230, 116th Cong. (Introduced Jun. 12, 2019), available online.

  58. 58.

    See e.g., Koenig, supra note 39.

  59. 59.

    See Prosecutor v. Mbarushimana, supra note 43.

  60. 60.

    Id.

  61. 61.

    See Mehandru & Koenig, supra note 10.

  62. 62.

    See Koenig, supra note 39.

  63. 63.

    The Prosecutor v. Édouard Karemera and Matthieu Ngirumpatse, ICTR-98-44-T, Judgement and Sentence (ICTR TC III, Feb. 2, 2012), available online.

  64. 64.

    Id.

  65. 65.

    The Prosecutor v. Théoneste Bagosora and Anatole Nsengiyumva, ICTR-98-41-A, Judgement (ICTR AC, Dec. 14, 2011), available online.

  66. 66.

    See Ashouri, Bowers & Warden, supra note 44.

  67. 67.

    Id. at 11.

  68. 68.

    Id.

Digital Evidence Repositories and Vulnerable Populations: How the Accumulation of Digital Evidence May Interact with the Privacy of Sexual Assault Survivors

I. Introduction

With the rise of digital technology, it is likely that the future of evidence collection at the International Criminal Court (ICC) will increasingly rely on digital evidence. This may take the form of videos, images, etc. (media) taken by victims or civil society organizations, or evidence uploaded to social media platforms by victims themselves. We have already seen a recent arrest warrant heavily rely on digital evidence.1 The collection of digital evidence would aid in the documentation of mass atrocities; the international community would no longer have to wait to document events because it is deemed too unsafe to send people into active conflict zones. It would probably be faster, cheaper, and easier to collect evidence.

However, the collection and storage of digital evidence does not come without concerns over individual privacy. It is unclear who would control such a repository: The ICC itself? Civil society groups? International governments? Private companies? The answer to this question would also likely dictate whether and to what extent it would be accessible to the public. Keeping it from the public could prompt criticism of a lack of transparency. On the other hand, opening it up to public viewing would increase concerns around the individual privacy of those whose likenesses or other personal information could be gathered from the media. Even if the repository is intended to be private, the likelihood that it would stay that way is questionable, as centralized storage of this sort of data could garner public attention and run the risk of being vulnerable to hacking and publication. There have already been multiple hacks of government data around the world.2 Whether the repository is public or private, therefore, it is possible that the identities of victims could eventually be made public.

While the possibility that the details of an investigation, including victim information, could be leaked to the public has always been a concern with any kind of evidence, digital or not, the sheer mass scale of a digital repository makes it a unique consideration. Once information is publicly available, it could be seen by virtually anyone, at any time. Today, an estimated five billion people, or about 66% of the world’s population, report having a smartphone.3 Each of these five billion people could have access to sensitive information about victims. Additionally, individual victims usually have the option to testify or cooperate with an investigation or trial. Using media in which they are featured, which may or may not have been collected with their consent, strips this level of voluntariness from the ICC-victim/survivor relationship.

These privacy concerns are all the more important when we consider the identities and private information of vulnerable victim populations. This comment will deal with specific considerations that affect the evidence of rape or sexual assault survivors. Issues around privacy affect every victim population. However, I chose to focus on this particularly sensitive population because I believe it highlights the gravity of the potential consequences that may result from a violation of privacy rights. The Court is effectively one of few systems in the international community which can prosecute mass rape and sexual violence. As such, it should consider the potential consequences of storing evidence regarding this particular crime as the survivors of sexual violence may face specific harms which may not apply to any other victim/survivor population. Though I do not claim that this is the only vulnerable victim population or that the privacy concerns highlighted throughout this comment only affect sexual violence survivors, the reputational harms impacting this population should factor heavily into the Court’s considerations when establishing a digital repository.

In Section II, I look at the specialized risks surrounding the publication of the identity of a rape or sexual assault survivor. These risks include social, psychological, and financial risks—how the social stigma placed on survivors can be detrimental to their overall well-being once they are identified as survivors. In breaking down these concerns, I touch on how it affects both the individual survivors as well as their family and the general community structure. Section III provides an overview of the current victim protection protocol in place at the ICC. In Sections IV and V, I address potential challenges to the current protocol in light of digital evidence and suggest some ways in which the Court could address these concerns.

II. Risks Associated with Identification as a Survivor

A. Socio-Psychological

Sexual violence can have a severely negative effect on the psychological and social well-being of a survivor. With limited resources, it is often difficult for humanitarian organizations to provide comprehensive post-rape psychological care.4 Even after receiving counseling, the World Health Organization reports that up to 50% of women retain stress symptoms after surviving sexual violence.5 A sexual attack can result in an inability to perform every-day tasks, nightmares, flashbacks, and suicidal ideation.6 There may also be long-term psychological effects including depression, anxiety, psychotic episodes, and PTSD.7 These issues can be chronic and debilitating. A Human Rights Watch report showed that survivors of gang rapes during Kenya’s 2007–2008 post-election violence reported chronic depression, anxiety, and unexplained pain that interfered with their ability to work, care for their children, and be full members of their communities.8

One of the most commonly experienced responses to rape is the stigmatization of the survivor.9 One survivor reported:

I was collecting firewood for my family when three armed men on camels came […] raped me one after the other. When I arrived home, I told my family what happened. They threw me out of our home and I had to build my own hut away from them. I was engaged to a man and […] he did not want to marry me anymore and broke off the engagement because he said that I was now disgraced and spoilt.10

There is a belief that the survivor is “disgraced, dishonored, or otherwise ruined by the violation.”11 This can result in the survivor, already traumatized and lacking institutional support, being “forced out of their families, isolated from their communities, and left to deal with their trauma alone.”12 Another report cites the percentage of female survivors accompanied by their husbands to a hospital after an assault as 1%, because most have been abandoned upon their family learning about their attack.13 In conflict settings, where the protection of a community can be especially necessary, abandonment by a social group could expose the survivor to subsequent attacks.

B. Financial

The economic lives of survivors may also be fundamentally altered after their attack is known. In societies where marriage can be a crucial part of a woman’s financial security, divorce or abandonment can result in poverty.14 Additionally, the psychological impacts can hinder a survivor’s ability to return to their previous life or job which would similarly harm their economic stability.15 Women may become too afraid, of re-victimization or social stigma, to leave their homes to work, thereby contributing to the poverty levels of both the family and community.16

C. Family

The breakdown of family structures in communities can also be seen following mass rape. One victim recounts:

[M]y husband insults me every day, calling me the wife of the militiamen who raped me […] [and] when I ask my husband to give me some food he replies to me, why don’t you go and ask for food to your husband in the forest?17

Domestic abuse beyond the initial assault has become increasingly common, a by-product of the breakdown of culturally accepted marriage norms coupled with the stigma placed on a survivor’s family.18 This may result in communal trauma. In interviews with Mozambican women, survivors reported that “when one woman was raped, the rape was felt by the whole community.”19

The effects of rape and sexual assault may also extend past the individual to their children. Children “born of wartime rape constitute a vulnerable group, who are often labelled by communities as the ‘bad blood of political, ethnic, or religious enemies’” and are subsequently stigmatized at birth and lack “access to community resources, family protection and education or livelihood activities.”20 In Rwanda, there have been reports of abuse, neglect, and infanticide of babies born of sexual violence during the genocide.21 A community refusal to recognize the child as legitimate may also result in administrative challenges in birth registration or citizenship rights.22 This lack of formal citizenship can have economic impacts if the child is denied basic rights such as education and can impact the child’s ability to receive asylum.23 The children, as stated in a UNIFEM report, “become the symbol of the trauma the nation as a whole went through, and society prefers not to acknowledge their needs.”24

III. Current ICC Practices

The current witness protection protocol of the ICC, while robust and generally highly rated by past witnesses, may not be enough to fully protect survivors and witnesses in the digital space. The ICC states that:

The court has a number of protective measures that can be granted to witnesses, victims who appear before the Court, and other persons at risk on account of testimony given by a witness. The court’s protection system is based on best practices which are aimed at concealing the witness’s interaction with the Court from their community and from the public in general.25

Article 68 of the Rome Statute provides that the Court “shall take appropriate measures to protect the safety, physical and psychological well-being, dignity and privacy of victims and witnesses.”26 In doing so, the Statute stipulates that the nature of the crime, in particular, where the crime involves sexual or gender violence, will be regarded.27 Article 43(6) stipulates the creation of a Victims and Witnesses Unit (VWU) which will provide, in consultation with the Office of the Prosecutor, protective measures and security arrangements, counseling and other appropriate assistance for witnesses, victims […] , and others who are at risk on account of testimony given.28 The VWU has indicated that protecting the identities of witnesses and victims through anonymization is key to successful protection.29 Article 54(3)(f) states that the Prosecutor may “take necessary measures, or request that necessary measures be taken, to ensure the confidentiality of information [or] the protection of any person.” 30, 31 Finally, Article 57 provides that where necessary, the pre-trial chamber will provide for the protection and privacy of victims and witnesses.”32

A. Pre-Trial Witness Protocol

The VWU first engages with witnesses after they have been identified, have agreed to testify, and have been accepted by the relevant Chamber.33 The VWU then provides a Witness Information Form which serves to provide operation and logistical support information and document any particularized issues in order to assist in preparing a specific witness plan.34 Protective measures in the witness’s home countries are aimed at a reduction in threat exposure or providing responses to an identified threat.35 These may include an assisted move, though witness relocation is “only used as a last resort, due to the immense burden put on witnesses and families.”36 Concealing the interaction between the witness and the Court from the public is often the primary goal of the VWU, as identification can carry significant risks to potential witnesses.37 The VWU may help in creating a cover story for the witness to provide to their employer or, if appropriate, confirm the witness’s participation in a trial in order to secure a leave of absence.38 The goal is to select the “least intrusive method” of protection.39 Once the witness is confirmed and transferred to the Court, they are part of the process of “courtroom familiarization” during which they are showed the Court and have the procedural elements of a proceeding explained.40

B. In-Court Testimony

Trial protective measures include expunging the individual’s name or identifying features from the public record, prohibiting the parties and participants in a proceeding from disclosing the same, using electronic presentation of evidence, identity-altering technology, pseudonyms, and in camera proceedings.41 They may also include allowing a psychologist to monitor the witness’s well-being, allowing a support person to sit next to the witness, adapting the setup of the Court to prevent eye contact between the witness and the accused, and using video-links to allow the witness to testify outside the courtroom.42 The ICC states that these protective measures do not affect the fairness of the trial and apply equally to all witnesses, whether called by the Prosecution, Defense, the legal representatives of the victims, or the Judges.43 In the Lubanga trial, the Court ordered the implementation of protective measures including withholding witness identities from the public, using pseudonyms, distorting voices, screens separating the accused and witnesses, and conducting parts of the proceedings in closed sessions.44 Voice distortion and pseudonyms were also used during the Katanga trial.45 The Bemba trial, which had a high number of witnesses who claimed sexual violence, also used pseudonyms and face distortion.46

Witnesses have reported feeling safe during their testimony, with eight of nine survivors of sexual violence choosing the highest safety rating with regard to their courtroom experience.47 Women, particularly, stated that they felt a sense of safety and gave slightly higher satisfaction ratings to the VWU support during testimony than their male counterparts.48 Nevertheless, some witnesses reported unease regarding their participation with the court when planning their return home stating that they “wanted to talk to someone about my safety when I return home” while another witnesses said they feared there was no guarantee of safety because they felt “that the accused knows me, even though he could not see me.”49

C. Post-Testimony Protocol

The fears of participating witnesses do not end when their testimony is over. Said one witness, “now, after my testimony, I will have a bigger need of protection.”50 Another stated that their “life will look like the life of somebody who has enemies.”51 The comprehensive protection measures offered by the ICC do not necessarily extend for an indeterminate amount of time. Other than relocation, it is not likely that a witness will obtain a sort of “life protection” post-testimony.52

D. Weaknesses

Despite comprehensive measures, some witnesses continue to be concerned for their own safety. Witnesses in the Lubanga and Katanga cases revealed that, despite an intricate system of protective measures, they fear a backlash from their communities:

I was worried about my security […] and that the protective measures were not enough. People can easily read transcripts of the proceedings on the internet and identify me.53

This fear may not be enough to result in protective protocols after a survivor testifies, as qualification for the ICC Protection Program must be based on objective grounds, such as actual threats and “personal preferences or subjective fears of individual witnesses are not enough.”54 Even when protections are granted, they may not be entirely successful. During the Lubanga trial, a transcript for a hearing between defense counsel and a judge was unintentionally broadcast and the Court subsequently refrained from holding private sessions until it could be certain that the issue would not reoccur.55 The report highlights that sexual assault survivors were not necessarily part of the positive responses to questions regarding post testimony experiences, which may have skewed the results in favor of witnesses who provided testimony in less stigmatized trials and who had support from community members or family prior to testifying.56

IV. Possible Issues when Applying Current Protocol to a Digital Repository

While the current ICC witness protection protocol is comprehensive and, despite some challenges, generally succeeds in offering protections to the survivor-witnesses it engages with, it might not be enough to uphold the Rome Statute’s mandate of “protecting the safety, physical and psychological well-being, dignity and privacy of victims and witnesses.” The resources of the Court are limited and providing protection to the witnesses during the pre-testimony, testimony, and post-testimony stages of witness cooperation will become exponentially more expensive and difficult if the Court has to contend with massive amounts of media and other data. While the current protection protocol for in-chambers testimony, including voice or face distortion, will probably continue to be effective for those witnesses who are providing live testimony, new policies may have to be implemented with a digital evidence repository. These potential new policies should address issues of identification/anonymization, consent, and storage.

A. Identification/Anonymization Issues

As stated in Section III, it is a primary goal of the VWU to conceal interactions between individuals and the Court, which may prove more difficult with the continued use of digital evidence. In 2017, the Office of the Prosecutor issued an arrest warrant for Libyan military commander Mahmoud Mustafa Busayf Al-Werfalli which heavily relied on open source information, specifically media that depicted several victims being shot and killed.57 When describing the videos in the warrant, the victims are generally referred to as “unidentified men,” while others are described as hooded or otherwise not identifiable.58 This is a notable weakness in the ICC’s victim protection protocol going forward. An inability to identify survivors does not mean the presentation of their data in court shields them from privacy risks. By its nature, digital evidence will mean that it should be easier to provide evidence for crimes in the Court; victims will no longer have to be individually identified and the Prosecutor will have to expend less resources, both in money and time, to gather evidence. However, the fact that the survivors are unidentified to the Court does not mean they are unidentifiable .

The VWU may successfully keep survivors from being identified now as it is as able to contact them directly in a way that may no longer be possible when the evidence comes from digital media sources which may or may not be traceable. There is little that can be done to offer any of the pre-trial protections currently in place at the ICC when the “witness” in question is not really a witness at all, at least not in the traditional sense, but rather an unidentified individual in a given conflict zone. A witness quoted in Section III spoke of their fears that someone could read court transcripts and find out who they are, possible because they fear that members of their community or others would take retributive action against them when they return from testifying. This fear would go from a concern that someone could find a transcript to a fear that someone could find the identifiable video or photograph of their attack. This identification, as explained in Section II has the potential to irrevocably alter an individual survivor’s life. This fear may make potential witnesses wary of the evidence collection methods of the ICC.

The ICC will have to look at the protection of witness privacy as something to be done on the backend of evidence collection. Witnesses will likely no longer first be identified and then require protections from the Court. In fact, when dealing with digital evidence, it is possible that a witness may never directly engage with the Court. Therefore, the Court has to consider how it will effectively and adequately protect an exponentially larger number of witnesses than it has ever had before and do so while possibly never being able to identify the individuals in the manner required by the current protection protocol.

B. Consent

The Court should also consider issues of consent, which likely have not previously been primary concerns when engaging with witnesses. Every witness that has provided testimony or otherwise engaged with the Court has ostensibly done so willingly. This may or may not be the case when using digital evidence. Studies have shown that whether or not disclosure of traumatic events serves a therapeutic purpose may depend on the reactions of recipients to the disclosure.59 Some jurisdictions around the world allow effectively forced testimony by allowing law enforcement to threaten survivors of sexual violence who are unwilling to testify with jail time.60 Victim’s rights groups have denounced this practice. Colby Bruno, of the Victim’s Rights Law Center highlights the importance of survivor consent to testimony, saying:

[I]f the more private victims or victims who are emotionally fragile do not want to report and go through a lengthy process that is sure to re-traumatize them, then they should not have to, and that should be their choice.61

The previous positive experiences witnesses have had while testifying have always inherently respected a survivor’s right to choose to testify. Even when confronted with difficult cross examination, witnesses still ultimately had agency in testifying.62 Forced testimony, in the jurisdictions which allow it, can lead to a fraying of the relationship between a survivor-witness and law enforcement. For example, in the United States, where material witness warrants allow District Attorneys to jail potential witnesses for refusing to testify against a defendant, witnesses who the warrants have been used against have reported being unlikely to engage with law enforcement in the future.63

This example is not to suggest that the use of digital evidence means the Court would perpetrate this same kind of coercion of witnesses. However, the relationship between potential witnesses and the Court is one of the most important tools in bringing successful cases and it may be strained if survivors do not sufficiently feel as though they have agency and choice in having their attack used in Court.

C. Storage

The Court should also consider the length of time for which it will store data. The protections offered by the ICC to cooperating witness reasonably come to an end at a given point, specifically when the witness is no longer a participant in a trial or investigation or is otherwise not at risk.64 Given this termination of protection, the question arises of when an individual is no longer “at risk” with a digital repository and when they are considered to be cooperating with the Court? Will the Court keep the digital evidence stored past the end of a trial and appeals? If so, for what purpose? Prolonged storage would expand the time during which an individual could be considered a witness or otherwise engaged with the Court. If data is kept after it is no longer in active or contemplated use by the Court, individuals would practically be in continued cooperation with the Court and therefore at risk for the sorts of reputational harms I highlighted in Section II. Whether it is kept private or public, as highlighted in Section I, there might be a risk of publication as a result of common data-breaches. The Court should consider how long it intends to keep media in order to adhere to established protection principles.

V. Possible Solutions

The ideas I propose are mostly geared towards the protection of witnesses before they are identified by the Court. As reported in Section III, the current protection system is rated positively by past witnesses and has, by most markers, succeeded in the goals of the Court to protect them. There is not, however, a robust system in place to protect witnesses who are depicted in media used as evidence by the Court but do not ultimately engage directly with the Court so as to avail themselves of the in-chambers protections or relocation.

The issues of consent and anonymization may be dealt with together since they both center around a survivor being publicly identified. There is no practical way to ensure informed, explicit consent universally. For example, a third-party may document someone’s attack and provide that evidence to the Court without the survivor’s knowledge. The person depicted in the evidence cannot give explicit consent so ensuring maximum anonymization is the next best solution to both the consent and anonymization issues faced by the Court. Censoring evidence to anonymize the identities of those in the video or image should be the default in the Court’s evidence collection. This would also help digital evidence collection processes adhere to the VWU’s standard of attempting to keep cooperation with the Court anonymous. It is of benefit to the Court that some organizations have already begun to engage with this issue and some of their solutions may prove helpful to the ICC. One organization of note is WITNESS, a leader in human rights documentation.65 Their technology has been previously used by the Court.66 The organization provides video equipment and training to activists and has continued to evolve as the sources of digital evidence have changed. With their creation of ObscuraCam in collaboration with the Guardian Project, WITNESS advocates that those who document human rights abuses “stop, think, and be empowered to control the media they are capturing.”67 ObscuraCam allows for the blurring or pixelization of people captured in images and videos.68 The intent behind the technology, of allowing people to “make purposeful choices about when they speak out and what degree of anonymity they hold for themselves” should be the Court’s goal as well.69

The Court should work with NGOs toward comprehensive witness education on the use of privacy protective tools such as ObscuraCam when documenting human rights abuses. Working with those organizations to educate potential witnesses on the use of this kind of technology could relieve some of the concerns around consent and anonymization, thereby potentially increasing a feeling of relative security within the survivor population. Though the universal use of this technology might not be achievable due to issues of access, demonstrated efforts by the Court to ensure privacy to the largest extent possible may reflect positively on the overall feeling of witnesses towards the Court’s evidence repository.

With regards to the storage question, the Court should consider principles of storage limitation and potentially place a time limit on media and data retention. The concept of storage limitation is an established principle in the world of privacy law. Article 5 of the General Data Protection Regulation, a European Union regulation on data protection and privacy, states:

Personal data shall be: kept in a form which permits identification of data subjects for no longer than is necessary for the purposes for which the personal data are processed; personal data may be stored for longer periods insofar as the personal data will be processed solely for archiving purposes in the public interest, scientific or historical research purposes, or statistical purposes […] subject to implementation of the appropriate technical and organizational measures required […] to safeguard the rights and freedoms of the data subject.70

Additionally, one of the foundational principles of privacy by design, a philosophy geared towards the creation of databases that consider data privacy during each stage of development, is purpose specification. 71 The purpose for which a specific piece of data is kept must be identified and, usually, communicated to the individual data subject at, or before, the collection of the data.72 Privacy by design principles highlight that:

[T]he use, retention, and disclosure of personal information shall be limited to the relevant purposes identified to the individual […] Personal information shall be retained only as long as necessary to fulfill the stated purposes, and then securely destroyed.73

The issue of when to delete should be considered in relation to the length of a trial. If the Court has no anticipated need for evidence past the end of a preliminary investigation, trial, or appeal, i.e. the lifecycle of a Court proceeding, it should consider the balance between retaining evidence and the potential risks to individuals depicted in the media. As highlighted in Section II, the harms which could come to identified sexual violence survivors are severe and life-altering. They should be weighed against the potential future use of the evidence. This balancing might be helpful in the creation of a specific time period after which evidence would be deleted.

It is possible that the Court may decide that the balance between an individual privacy right and the overall public good that may come from data retention falls in favor of indefinite retention and chooses to permanently store evidence. If so, this policy should be clearly communicated to the public through witness outreach and education efforts in order to ensure that individuals who provide evidence to the Court are not left feeling abused or misled regarding the use of their evidence. If it is the Court’s intention to store evidence past the end of a proceeding, it should attempt to ensure that those providing it are aware that they are effectively in cooperation with the Court indeterminately.

VI. Conclusion

In light of the creation of a digital repository, the ICC should consider the risk of individualized harms which may come to those portrayed in digital media used at trial and stored by the Court. Whether the repository is kept public or private, risk of disclosure and publication inherently exists as a result of the nature of data storage. Possible publication should be of specific concern to the Court due to the nature of the crimes which it investigates, specifically crimes of sexual violence including mass rape and sexual assault. Identification as a survivor of sexual violence carries severe, often negative, and possibly life-altering consequences which include psychological trauma, familial breakdowns, and financial instability. Due to these risks, the Court should consider its current witness protection protocols which, although shown to be successful in offering protection to past witnesses, might not be enough as a result of new issues of identification, consent, and storage length. In order to successfully protect survivors, the Court should consider implementing a witness education program which informs the public of privacy protective tools when documenting attacks as well as consider setting a time limit for data retention.

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

  1. 1.

    Emma Irving, And So It Begins… Social Media Evidence in an ICC Arrest Warrant, Opinio Juris (Aug. 17, 2017), available online; The Prosecutor v. Mahmoud Mustafa Busayf Al-Werfalli, ICC-01/11-01/17, Warrant of Arrest (PTC I, Aug. 15, 2017) [hereinafter Werfalli Arrest Warrant], available online.

  2. 2.

    See Center for Strategic and International Studies, Significant Cyber Incidents Since 2006, available online (last visited May 26, 2020).

  3. 3.

    Laura Silver, Smartphone Ownership is Growing Rapidly Around the World, but Not Always Equally, Pew Res. Center (Feb. 5, 2019), available online.

  4. 4.

    Skye Wheeler, “My heart is always scared”: The Simmering Mental Health Crisis for Rape Victims in War, World Pol’y (Apr. 4, 2017), available online, archived.

  5. 5.

    Médecins Sans Frontières, Shattered Lives: Immediate Medical Care Vital for Sexual Violence Victims 13 (2d ed. Jun. 2019), [hereinafter Shattered Lives], available online.

  6. 6.

    Id. at 13.

  7. 7.

    Id. at 13, 28.

  8. 8.

    Wheeler, supra note 4.

  9. 9.

    See generally, Missale Ayele, Public Health Implications of Mass Rape as a Weapon of War, Ga. St. U. (Public Health Thesis, May 7, 2011), available online.

  10. 10.

    Shattered Lives, supra note 5, at 13.

  11. 11.

    Id. at 18.

  12. 12.

    Nicola Jones, Janice Cooper, Elizabeth Presler-Marshall & David Walker, The Fallout of Rape as a Weapon of War 3, ODI (Jun. 2014), available online.

  13. 13.

    Id. at 3.

  14. 14.

    Id. at 3.

  15. 15.

    Id. at 3.

  16. 16.

    Cassandra Clifford, Rape as a Weapon of War and its Long-Term Effects on Victims and Society, 7th Global Conference: Violence and the Contexts of Hostility (May 7, 2008), available online.

  17. 17.

    Id.

  18. 18.

    Id.

  19. 19.

    Tina Sideris, Rape in War and Peace: Some Thoughts on Social Context and Gender Roles, 43 Agenda 41, 42 (2000), paywall, doi.

  20. 20.

    Report of the Secretary-General, Conflict Related Sexual Violence, UN Doc. S/2019/280, at 8 (Mar. 29, 2019), available online.

  21. 21.

    Charli Carpenter, War’s Impact on Children Born of Rape and Sexual Exploitation: Physical, Economic and Psychosocial Dimensions, University of Pittsburgh Working Paper (Sep. 21, 2005), available online.

  22. 22.

    Conflict Related Sexual Violence, supra note 20, at 8.

  23. 23.

    Carpenter, supra note 21.

  24. 24.

    Id.

  25. 25.

    Witnesses, ICC, available online (last visited May 26, 2020).

  26. 26.

    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, as amended [hereinafter Rome Statute], Art. 68, available online.

  27. 27.

    Id.

  28. 28.

    Rome Statute, supra note 26, Art. 43.

  29. 29.

    Protection and Support, VRWG, available online (last visited May 26, 2020).

  30. 30.

    Rome Statute, supra note 26, Art. 54.

  31. 31.

    (There has been previous litigation to establish the respective roles that the Office of the Prosecutor and the Victims and Witnesses Unit play in witness and victim protection but since the intricacies of this division of duties lie beyond the scope of this comment, I only provide this general overview).

  32. 32.

    Rome Statute, supra note 26, Art. 57.

  33. 33.

    Stephen Cody, Alexa Koenig, Robin Mejia, & Eric Stover, UC Berkeley HRC, Bearing Witness at the International Criminal Court: An Interview of 109 Witnesses 19 (Jun. 2014), [hereinafter Bearing Witness], available online.

  34. 34.

    Id. at 19.

  35. 35.

    Witnesses, supra note 25.

  36. 36.

    Id.

  37. 37.

    Coalition for the International Criminal Court, Is Enough Being Done to Protect ICC Witnesses? (May 18, 2015), available online.

  38. 38.

    Bearing Witness, supra note 33, at 19.

  39. 39.

    Witnesses, supra note 25.

  40. 40.

    Id.

  41. 41.

    International Criminal Court, Rules of Procedure and Evidence, R. 87 (2d ed. Sep. 9, 2013), available online, archived.

  42. 42.

    Witnesses, supra note 25.

  43. 43.

    Id.

  44. 44.

    International Bar Association, Witnesses Before the International Criminal Court 30 (Jul. 2013), available online.

  45. 45.

    Id. at 31.

  46. 46.

    Id. at 31.

  47. 47.

    Bearing Witness, supra note 33, at 47.

  48. 48.

    Id. at 47–48.

  49. 49.

    Id. at 50.

  50. 50.

    Id. at 53.

  51. 51.

    Id. at 53.

  52. 52.

    See generally, Tom Maliti, Expert: “Relocation of a witness does not entail life protection by the ICC, Int’l Just. Monitor (Jul. 29, 2011), available online

    (reporting an interview with Mariana Pena, the Permanent Representative at The Hague of the International Federation for Human Rights who stated that relocation of a witness and possibly their families could mean the witness remains in the secondary location for the rest of their lives but that “participation in the program will need to come to an end at a certain point […] because it is costly, but also because security risks will necessarily decrease after some time.”).

  53. 53.

    Bearing Witness, supra note 33, at 23.

  54. 54.

    Jennifer Easterday, Witness Protection: Successes and Challenges in the Lubanga Trial, Int’l Just. Monitor (Jun. 26, 2009), available online.

  55. 55.

    Id.

  56. 56.

    Id. at 62.

  57. 57.

    Nikita Mehandru & Alexa Koenig, Open Source Evidence and the International Criminal Court, 32 Harv. Hum. Rts. J. (Apr. 15, 2019), available online; See also Werfalli Arrest Warrant, supra note 1, at 8–11.

  58. 58.

    Werfalli Arrest Warrant, supra note 1, at 8–11.

  59. 59.

    Sarah E. Ullman, Correlates and Consequences of Adult Sexual Assault Disclosure, 11 JIV 554, 555 (Dec. 1, 1996), paywall, doi.

  60. 60.

    See Samantha Michaels, Should Rape Victims Have to Spend Time in Jail for Not Testifying? Louisiana’s Prosecutors Think So, Mother Jones, Apr. 19, 2019, available online; See also Henry McDonald, Alleged Rape Victim Arrested in Ireland for Refusal to Testify, The Guardian, Feb. 12, 2012, available online.

  61. 61.

    Bronwyn Isaac, Why Forcing Women to Testify Against their Assailants Enforces Victim Blaming, Bustle (Apr. 21, 2017), available online.

  62. 62.

    Bearing Witness, supra note 33, at 49

    (reporting that a witness felt uncomfortable with some questions from counsel, saying “I was very angry because the lawyer twists my answers around and around.”).

  63. 63.

    See Sarah Stillman, Why Are Prosecutors Putting Innocent Witnesses in Jail?, The New Yorker (Oct. 17, 2017), available online

    (reporting that a witness who refused to testify against her abuser and was subsequently jailed saying that she would “probably not call the police again, as long as it wasn’t life-threatening.”).

  64. 64.

    Maliti, supra note 52.

  65. 65.

    Our Story, WITNESS, available online.

  66. 66.

    Id.

  67. 67.

    Announcing ObscuraCam v1: Enhance Your Visual Privacy!, Guardian Proj., available online (last visited May 26, 2020).

  68. 68.

    Id.

  69. 69.

    Id.

  70. 70.

    Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the Protection of Natural Persons With Regard to the Processing of Personal Data and on the Free Movement of Such Data, and Repealing Directive 95/46/EC (General Data Protection Regulation), 2016 O.J. (L 119), available online, archived; See Lydia F. de la Torre, What Does “Storage Limitation” Mean Under the EU Data Protection Law?, Medium (Jan. 22, 2019), available online

    (explaining the principle of storage limitation further).

  71. 71.

    See generally, Ann Cavoukian, Privacy by Design: The 7 Foundational Principles, IAPP (Nov. 2, 2010), available online.

  72. 72.

    Id.

  73. 73.

    Id.

Digital Evidence and the Use of Artificial Intelligence

I. Introduction

With the advent of new advancements in the technological world, it should come as no surprise that the primary source of information gathering and documentation is within the grasp of every individual. Internet and smartphones have virtually created journalists and investigators on every corner. International human rights and criminal prosecution organizations have had their eyes for quite some time on the potential benefits of using the open web and social media platforms to gather information and evidence for investigation, prosecution, and averment of crimes. It has already been seen that International Tribunals such as the International Criminal Tribunal for Rwanda (ICTR), the International Criminal Tribunal for former Yugoslavia (ICTY) and even the International Criminal Court (ICC) have initiated investigations and prosecuted accused persons on the basis of digital evidence. With the benefits of having such vast amounts of information at hand there are bound to be challenges as well. The key challenges in using digital evidence are with regards to identification of relevant evidence and establishing their authentication in order to satisfy the Court’s threshold of admittance.

As we have seen in the case of Sri Lanka’s soldier’s 2009 video capturing execution of prisoners, forensic analysts have been used to verify the authenticity of digital evidence,1 but there has been very limited success in identifying the correct evidence. The relevance of evidence has been established only through manual sorting and this potentially leaves a lot of ground to be covered. Artificial Intelligence (AI) can be employed in such scenarios to effectively scour a dump of data and comb out the relevant information. In this comment, I first briefly discuss the threshold for admissibility, and standards for assigning evidentiary value to digital evidence. Then I discuss the potential of AI in gathering and sorting data to be used for investigation and prosecution by the ICC.

II. Admissibility of Digital Evidence

The ICTR, the ICTY or the ICC do not have a separate criterion for regulating digital evidence and it comes under the broad definition of documentary evidence.2 The ICC derives its admissibility criteria for evidence from Article 69 of the Rome Statute which requires that the Court takes the probative value of the evidence into account while deciding on its admissibility. At this stage, the evidence needs to only prove its prima facie credibility. However, the weight of the evidence gets decided at the conclusion of trial in light of all the evidence.3

These tribunals have generally relied upon supporting eyewitness testimony to establish the credibility of digital evidence. In the Milan Milutinović trial, the chamber relied on the testimony of camera operators to establish the reliability of footage provided by the BBC.4 However, this has not stopped the tribunals from allowing admission of evidence even if it is not supported by witness testimony. The ICTY has held that there is no blanket ban on the admission of documentary evidence simply due to the lack of supporting witness testimony.5 A similar stand was taken by the ICC when it ruled:

[N]othing in the Statute or the Rules expressly states that the absence of information about the chain of custody […] affects the admissibility or probative value of Prosecution evidence.6

Even though the threshold for evidence at the admissibility stage is quite low, the ICC still has certain guidelines for digital evidence. It has been ruled that digital evidence must conform to “e-Court Protocol”7 to “ensure authenticity, accuracy, confidentiality and preservation of the record of proceedings.”8

III. Evidence Gathering

The question of the admissibility or evidentiary value arises only after the relevant evidence has been collected. The collection of evidence proves to be the most challenging part as these cases revolve around gross violation of human rights and it sometimes takes years even to start the investigation. In such situations, it becomes imperative that any form of evidence be collected and preserved as soon as it becomes available. Many initiatives, by NGOs and private organizations, have been started in order to collect, investigate, and preserve digital evidence such as video files or images. Some of these are:

  1. BellingCat: They are an independent team of investigators, journalists and researchers who use open source information such as images or video on social media sites to investigate conflict or human rights violation. They have already worked on the chemical attack in Syria and the downing of the MH17 flight.9

  2. Syrian Archive: They too gather information from open source social media platforms and other websites and preserve the relevant data for further use.10

  3. WITNESS: They support and train people in conflict zones to “capture and preserve footage to improve the chances of it being used in courtroom.”11 They also operate the WITNESS Media Lab which works towards verifying and authenticating the videos.

  4. ACLU Mobile Justice App: The American Civil Liberties Union has launched the Mobile Justice app which can be used to record and upload any unwarranted interactions with law enforcement.12 These videos are then anonymously sent to the ACLU staff.

These initiatives are definitely a step in the right direction but, given the fact they are still very limited in scope and availability, have not been able to fulfill the ICC’s requirements for credible sources of digital evidence. These can only go as long as to provide evidence as amicus curiae or in the capacity of a friend to the court, and their evidentiary value would be decided in light of all other pieces of evidence available with the Court. If the ICC is to truly become a pioneer in using digital evidence to investigate and prosecute crimes at the international level, then there is a need to form its own archive or repository where such websites or individuals can submit any form of videos or images against human rights violations. There would certainly be challenges to such an initiative, but I believe benefits will outweigh these challenges. The challenges faced by the ICC would include satisfying non-violation human rights as per Article 69(7) of the Rome Statute, privacy concerns, safekeeping of the materials, establishing authenticity, objectively weeding out crucial evidence, etc. In this comment, I only discuss meeting the challenge to weed out important pieces of evidence amongst an overwhelming amount of data and how AI can be employed for such purposes.

IV. Artificial Intelligence

Almost all the initiatives I mentioned before, and many other as well, manually sort out the relevant piece of information amongst the data that is uploaded or collected by their respective organizations. This is a very tedious process and requires a lot of man hours. Additionally, there is always the risk of leaving out many crucial pieces of evidence. To put things into perspective, consider the example of the Syrian Archive which had 3,314,265 uploads on December 15th, 2019 and only 5,743 of these were verified.13 Now one can only imagine the number of videos or images that would be uploaded if the ICC decided to create its own repository. In such a scenario, I believe the first order of business would be to objectively weed out the relevant pieces of information and only then initiate work on establishing authenticity, or chain of custody, etc. This is where AI comes into play.

AI can be understood as a software or tool designed:

[F]or automating and improving the accuracy, speed and/or scale of machine decision-making, pattern-recognition and prediction in complex or large environments, with the aim of substituting for, or improving upon, human performance in specific tasks.14

Another major application of AI comes in the form of Machine Learning which tries to replicate the human ability to learn from experience.

It is this ability of Machine Learning which makes it particularly useful for the criminal justice system. The potential uses of AI in order for the ICC to conduct an effective investigation into the crimes under its purview are as follows:

A. Weeding Out Important Evidence

If the ICC is to create its own repository or even use the existing repositories created by several other organizations then it becomes very hard for them to weed out the relevant and crucial piece of evidence. However, this is exactly the scenario where AI and Machine Learning shines. Data has often been described as the lifeblood of AI and the more data the better it is for AI. Many NGOs are already using AI to scour and report child pornography.15 The AI used by these organizations have algorithms designed to recognize features or parameters similar to those found in child pornography.16 Similarly, an algorithm can be developed to identify different patterns of violence or human rights violations. These will not only save countless human hours but will also enhance productivity.

B. Gathering Information from Different Platforms by Recognizing Patterns

AI can also be used even if the ICC does not create its own repository or partner with any other organizations maintaining such repositories. A similar AI as described above can be used to crawl through those websites which allow crawling or scouring and can collect potential pieces of evidence, identifying them on the basis of parameters set by the developers.

C. Detection or Prediction

Another use of AI is the detection or prediction of incidents which violate international law. This might not always come in the general scope of the ICC but can still be useful for gathering evidence. An example of such use is the “Sentry” system which provides Syrian civilians early warning of incoming airstrikes.17 Its developer, Hala Systems, describes Sentry as an “indication and warning system that utilizes a multi-sensor network to generate a credible, real-time, situational awareness of threats.”18 It furthers explains that “Sentry uses AI to instantaneously validate information from multiple sources, allowing stakeholders to detect, identify, and predict threats.”19

V. Challenges of Employing AI

Although the term “Artificial Intelligence” was coined in 1956 by John McCarthy,20 it was not until 2010 that real progress on AI started.21 Being such a new and revolutionary technology comes with its own set of challenges that will have to be addressed in order to reap in maximum benefits of this technology. A few of the key challenges are discussed below.

A. Accuracy and Error

As mentioned earlier, AI uses certain parameters which has been set by the developer to identify the required patterns. Consider, for example, how facial recognition works. Facial recognition “uses biometrics to map facial features.”22 If these features match the features present in its database, then a match can be found. Even if the tool is not looking for match, it will still use certain features to establish whether the input is a human face or not. The problem arises in cases where the image or video might not be clear enough, or the feature of the face might not be consistent with the parameters inputted. In such scenarios, there would be an error in the results.

This would be even more challenging in case of the ICC because it is very possible that the people capturing videos and images in conflict zones might not have access to capable equipment or even a very good camera phone and this is bound to result in very low quality and even distorted videos or images.

In order to counter this challenge, there are two options which can be employed by the ICC. First and foremost is the obvious improvement in the AI’s capacity given its machine learning abilities. Accordingly, the more data that is provided to the AI, the better the results it will produce. The second option would be to follow the works of organizations like Witness who have been training and advising ordinary citizens on how to capture better videos and images of conflicts.

B. Access

In the event that the ICC employs AI to scour or crawl social media websites to gather potential evidence, it is certain that it would need access to major social media sites such as Facebook, YouTube, Twitter, etc. But the issue here arises from the fact that majority of these websites are US headquartered and follow US rules. Now, given the fact that the US administration has openly criticized the ICC and has made it clear that they will not support ICC in any way, and neither will allow its companies to support the ICC,23 makes it impossible for the ICC to get access to these sites. This might be overcome if these companies’ entities are established in countries outside US and they allow access to the ICC but given the administration’s threat of sanctions it is very improbable that the companies will go forward with such schemes.

C. Lack of Resources and Expertise

AI is very sophisticated technology and it has not developed enough to be used by untrained individuals. This basically translates into the requirement of a lot of expertise and resources in order to develop and operate any piece of AI. Given the fact that the ICC’s budget and annual expenditure is constantly being questioned by the member countries,24 it would be quite challenging to undertake a project to develop AI for its own use. This issue could be resolved if the ICC partners with one of the tech companies already possessing AI. Such partnerships have proved very fruitful for many NGOs who received free access to AI tools to curb child sexual content.

VI. Conclusion.

The ICC has been facing many issues with regard to the use of digital evidence and AI is the way by which it can enhance its capability to gather and authenticate evidence for investigation and prosecution. The potential of AI can be understood from the fact that a ton of research and studies are already being conducted on the uses of AI in furtherance of Criminal Justice needs and it is time the ICC too joined the bandwagon. Having said that, I do accept that there are many challenges to overcome and it will take lot of time to iron out all the wrinkles. In conclusion, I would like to point out the fact that the trends of criminal investigation are evolving at a breakneck speed and, in order to stay relevant, the ICC will have to evolve too.

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

  1. 1.

    Louis Charbonneau, Sri Lanka Execution Video Appears Authentic—U.N., Reuters, Jan. 7, 2010, available online.

  2. 2.

    The Prosecutor v. Édouard Karemera, Mathiey Ngirumpatse and Joseph Nzirorera, ICTR-98-44-T, Decision on the Prosecutor’s Motion for Admission of Certain Exhibits Into Evidence (ICTR TC III, Jan. 25, 2008), available online.

  3. 3.

    The Prosecutor v. Jean-Pierre Bemba Gombo, ICC-01/05-01/08, Decision on the admission into evidence of materials contained in the prosecution’s list of evidence (TC III, Nov. 19, 2010), available online.

  4. 4.

    The Prosecutor v. Milan Milutinović, Nikola Šainović, Dragoljub Ojdanić, Nebojša Pavković, Vladimir Lazaarević and Serten Lukić, IT-05-87-T, Judgement, ¶ 896 (ICTR TC III, Feb. 26, 2009), available online.

  5. 5.

    The Prosecutor v. Radoslav Brđanin and Momir Talić, IT-99-36-T, Order on the Standards Governing the Admission of Evidence, ¶ 20 (ICTY TC II, Feb. 15, 2002), available online.

  6. 6.

    The Prosecutor v. Thomas Lubanga Dyilo, ICC-01/04-01/06, Decision on the confirmation of charges, ¶ 96 (PTC I, Jan. 29, 2007), available online.

  7. 7.

    The Prosecutor v. Callixte Mbarushimana, ICC-01/04-01/10, Decision amending the e-Court Protocol (PTC I, Apr. 28, 2011), available online.

  8. 8.

    Id.

  9. 9.

    About, Bellingcat, available online (last visited May 29, 2020).

  10. 10.

    About, Syrian Archive, available online (last visited May 29, 2020).

  11. 11.

    Our Work, WITNESS, available online (last visited May 29, 2020).

  12. 12.

    ACLU App to Record Police Conduct, ACLU, available online (last visited May 29, 2020).

  13. 13.

    Observation Database, Syrian Archive, available online (last visited May 29, 2020)

    (providing no information on the number of videos that have actually been checked).

  14. 14.

    See Matthijs M. Maas, International Law Does Not Compute: Artificial Intelligence and the Development, Displacement or Destruction of the Global Order, 20 Melb. J. Int’l L. 29 (Aug. 2019), available online.

  15. 15.

    Angelin Yeoh, Google gives NGOs the Power of AI to Detect Child Sexual Abuse Images, The Star, Sep. 4, 2018, available online.

  16. 16.

    Id.

  17. 17.

    Danny Gold, Saving Lives with Tech Amid Syria’s Endless Civil War, Wired, Aug. 16, 2018, available online.

  18. 18.

    Home Page, Hala Systems, available online (last visited May 29, 2020).

  19. 19.

    Id.

  20. 20.

    Christopher Rigano, Using Artificial Intelligence to Address Criminal Justice Needs, 280 NIJ Journal 1 (Jan. 2019), available online.

  21. 21.

    Id.

  22. 22.

    Steve Symanovich, How does Facial Recognition Work?, Norton, available online (last visited May 29, 2020).

  23. 23.

    Owen Bowcott, Oliver Holmes & Erin Durkin, John Bolton Threatens War Crimes Court with Sanctions in Virulent Attack, The Guardian, Sep. 10, 2018, available online.

  24. 24.

    Sophie van Leeuwen, Many ICC Countries are not Paying, Just. Hub (Jun. 22, 2015), available online.

What Policies Should the Office of the Prosecutor Adopt in Receiving Cyber Evidence From User-Generated Evidence Gathering Apps to Help Protect Those Providing It?

I. Introduction

The rapid increase in the use of camera-equipped and internet-connected devices has enabled individuals to record far more information about their lives and their surroundings than ever before. This fundamental shift in information gathering is transforming the way that evidence is collected, analyzed, and presented at trial. These changes are already being seen on an international level. In 2017, the International Criminal Court (ICC or Court) issued an arrest warrant for Mahmoud Mustafa Busayf Al-Werfalli that was primarily based on videos of executions found on social media websites.1

While this increased access to information provides the ICC with unique opportunities, such as the real-time recording of human rights violations and the gathering of evidence in areas inaccessible to investigators, it also brings with it notable and pressing challenges. As cellphone users begin documenting war crimes and crimes against humanity, private actors will play an increasingly influential role in international criminal investigations. This new role brings with it the many risks inherent in collecting evidence, including the security concerns of the evidence gatherer.

The act of documenting in a conflict zone or situation may put both the documenter and subject at risk at the time of capturing the images or information, and later, if the information is discovered.2

This risk is even greater for private actors who lack the support and evacuation plans that are available to court-appointed investigators.3 These security risks are also present if the evidence is used at trial. It is yet to be decided whether or not these private actors will fall under the ICC’s existing protection regime, whether as witnesses or as intermediaries.4

International non-governmental organizations (INGOs) have created two evidence-gathering apps that seek to mitigate these risks and preserve the user’s anonymity while also gathering enough information to verify and authenticate the evidence in the hopes that it will be admissible at trial. Because these risks can only be mitigated and not eliminated, both apps also rely heavily on the notion of informed consent to ensure that participants are fully informed about the potential risks of their involvement. However, there are fundamental concerns of just how informed that consent truly is. With the promise of anonymity and the belief that their evidence will be admissible at trial, individuals are more likely to take greater risks.

In this comment, I investigate ways in which the Office of the Prosecutor (OTP) can mitigate the risks that users face when submitting information through user-generated evidence apps. Section II of this comment provides a brief introduction to the spread of information communication technology, the advent of user-generated evidence, and the INGOs operating in this space. Section III discusses the security risks that private actors face when providing user-generated evidence and how INGOs have sought to mitigate these risks through app-design and outreach programs. Section IV introduces ways in which the OTP can mitigate these security risks moving forward by incentivizing meaningful informed consent and adopting guidelines that ensure accountability.

II. Harnessing User-Generated Content as Evidence in the ICC

A. The Spread of Information Communication Technology

Since the advent of social media and the introduction of the iPhone in 2007, information and communication technology has evolved and spread rapidly on a global scale. As of November 2019, there are an estimated 5.15 billion people with mobile devices worldwide, making up over sixty-six percent of the world’s population.5 Ninety-seven percent of the world’s population lives within reach of a mobile cellular signal, and ninety-three percent of the world’s population lives within reach of a mobile broadband internet service.6 This widespread use of cell phones and global internet connectivity has brought with it new opportunities to document crimes and has eliminated the traditional financial barriers of photography and video recording.7

With smartphones in the hands of more people than ever before, there is no longer a need for people to buy a camera, as they now have one in their pocket to use whenever they would like. With mobile phone video becoming such an accessible technology, it has become increasingly important to “harness the potential of citizens documenting crimes, which may be used as evidence in international criminal trials.”8 These advances have ushered in an era of digital evidence, and, more specifically, of user-generated evidence.9 “User-generated evidence is recorded on a device such as a smartphone by an ordinary citizen with the intent to achieve legal accountability for wrongdoing.”10

B. External Partnerships Necessary to Collect, Analyze, and Use Digital Evidence

User-generated evidence provides new opportunities for the ICC, which has faced obstacles in obtaining sufficient evidence in the past. “Considering the Court’s global mandate, coupled with its limitations in funding and staff, obstacles to obtaining evidence are unsurprising.”11 Noting its struggles with heightened evidentiary standards and limited witness protection resources, the OTP emphasized a need for more diversified evidence as an alternative to witness statements in its 2012–2015 Strategic Plan.12 The OTP also recognized the “explosive growth” of digital data as a promising and highly reliable alternative and committed to expanding its own internal capacity as well as its reliance on external partners to work with digital evidence.13

In 2013, at the Salzburg Workshop on Improving War Crimes Investigations, the OTP and the ICC met with specialists in cyber investigations, human rights investigators, foundation representatives, and University of California Berkeley faculty and students to discuss how to gather and analyze digital evidence relevant to serious international crimes.14 Noting that “a lack of resources is the greatest constraint on the OTP’s ability to produce quality digital evidence,” the workshop concluded that the OTP should focus on fostering external partnerships with INGOs and private tech companies already operating in this space.15

In its 2016–2018 Strategic Plan, the OTP noted that “strategic partnerships with the law enforcement community, INGOs, and academic institutions were crucial to keeping up with developments in the field” of cyber investigations.16 With this in mind, the OTP outlined a strategy to build its own internal capacity by recruiting experts and investing in equipment, staff training, and partnerships.17 The success of this approach was noted in the OTP’s most recent strategic plan, which found that the diversification of evidence allows for “better corroboration of the facts, a lesser reliance on witnesses, and a greater resilience in the face of security challenges or witness interference.”18 The OTP has again committed to enhancing this practice from 2019 to 2021 through more strategic coordination with partners:

The Office can expedite its core activities through more strategic cooperation and coordination with partners, including first and early responders. Notably, the investigative landscape has been evolving in different ways: (1) more individuals and civil society actors are collecting relevant information as events unfold; (2) the awareness has grown that police and military contingents deployed by the international community may play a role in the preservation of evidence, and; (3) specific bodies and mechanisms are being created to work on the collection and preservation of evidence to support national and international jurisdictions. […] Where possible, and while emphasising that such cooperation cannot replace actual investigations by the Office, the Office will enhance its interaction and cooperation with partners through a structured dialogue on matters relating to collection and preservation of evidence.19

The “specific bodies and mechanisms” that have been created to collect and preserve evidence are discussed in more detail below.

C. User-Generated Evidence Apps as a Possible Solution

Presently, two apps have been designed with the specific goal of collecting and preserving user-generated evidence for international criminal prosecutions. The eyeWitness to Atrocities app was launched by the International Bar Association (IBA) after consulting with lawyers at the ICC and other international and hybrid tribunals.20 The second app, CameraV, is a joint venture between the video advocacy organization WITNESS and a software technology group, Guardian Project.21 It was funded by the technology company Benetech, a start-up based in Palo Alto, California.22

Both the IBA and WITNESS understood that the development of a technically sophisticated and replicable methodology behind the process of bringing user-generated evidence into courtrooms would be central to the credibility of their claim that what users filmed would be useful in a court of law.23

As such, both apps have built-in features to help address authenticity concerns in the hopes that the user-generated evidence will be admissible in Court.

The ICC has developed specific standards for authenticating video, photography, and audio-based evidence.24

The ICC requires metadata be attached to all submissions, as well as documentation showing chain of custody, the identity of the source, the original author and recipient information, and the author and recipient’s prospective organization.25

Both eyeWitness to Atrocities and CameraV record vast amounts of metadata to meet these admissibility standards:

[I]ncluding GPS coordinates, light meter readings, and nearby cell tower signals […] to enable the location and time of the footage to be verified.26

The IBA and WITNESS have both embarked in outreach efforts to conflict-affected communities to encourage individuals to download eyeWitness to Atrocities and CameraV and to use the apps to secure user-generated evidence.27 “As of July 2017, the IBA had received 1,200 pieces of footage, translating into some seventy hours of potential evidence.”28 “Over 10,000 users have downloaded CameraV.”29 Beyond advocating for users to download and use the apps, the IBA and WITNESS have also sought to educate users on the types of evidence needed to build a criminal case, emphasizing the need for linkage evidence.30

WITNESS has developed a 200-page field guide, which it uses to train those who may witness crimes.31 The field guide, Video as Evidence, explains some of the basic legal concepts of evidence and provides tips for capturing evidence in a way that could be useful in a criminal case while also emphasizing safety concerns.32 The IBA has produced materials to explain how to use the eyeWitness to Atrocities app but has focused primarily on outreach through traditional media and social media and in-person outreach to local human rights groups in Syria, Kenya, and South Sudan.33

While user-generated evidence apps are a promising development in the field of international criminal cases, they also raise a new set of concerns. Most notably, this development marks a shift in the traditional investigatory process from professional, court-appointed investigators to private citizens. As Hamilton noted:

The future of evidence collection is no longer primarily in the hands of investigators employed by a Hague-based court. Ordinary users of cell phones can now influence what crimes and which perpetrators will (and will not) be prosecuted.34

This shift brings with it new security and accountability concerns, which are discussed in detail below.

III. Security Concerns Unique to User-Generated Evidence

There is an inherent amount of risk involved in evidence collection. With the advent of user-generated evidence apps and the OTP’s increased reliance on digital evidence, some of those risks are being shifted “away from court-appointed investigators who have institutional support and an evacuation plan, to people without such protections.”35 These security considerations are a significant factor in both the creation and preservation of digital evidence as well as its use in a criminal trial. This section of the comment will discuss the risks that users face at all stages, and will also examine how effectively INGOs have addressed those risks through app design and education efforts.

A. Security Risks of Evidence Collection

As eyeWitness to Atrocities notes in its FAQs:

[T]here are always risks involved with documenting human rights abuses. There is not only danger from the user’s proximity to a volatile situation, but also the risk of arrest or other repercussions from authorities who do not want information about their actions to be publicised.36

Indeed, many of these security risks were part of the reason the OTP sought to diversify its evidence collection strategies, given its limited resources to provide witness protection.37 “Many users are, in a literal sense, witnesses; they often witnessed the commission of the atrocities they seek to document.” One can get a sense of the security risks that users face by looking at the risks that previous ICC witnesses faced.

While the Court does not publish statistics on the number of its witnesses that face security threats, it is possible to get a sense of the scale of the problem when considering that in the Court’s first case, the OTP, over a ten-week period, referred thirty-two witnesses into the Court’s protection program. That program was established to protect those who are at risk of harm and/or death.38

The ICRC discusses the risks that individuals face when providing information in their Protection Guidelines:

In practice, risks may range from physical and mental harm or threats of harm, discrimination, social marginalization and stigmatization, and are often not foreseen by the individual soliciting the information or the person providing it. Particularly in armed conflict and other situations of violence, people providing information to protection actors may even face reprisals, regardless of the content of the information shared, for merely participating in the data collection process.39

Both eyeWitness to Atrocities and CameraV have built-in safety features to mitigate these risks, allowing users to delete material quickly.40 EyeWitness to Atrocities also has a “panic button” that allows the user to erase all recorded information and the app itself to make it appear as though the app is not even on the phone at all.41 While these features are meant to mitigate the risk of retaliation, they are not infallible.42 Ensuring that users are adequately informed of these risks, and able to provide meaningful informed consent before collecting evidence, is of the utmost importance and will be discussed in Section IV.

B. Security Risks of Evidence Retention

The risk of retaliation does not end after the evidence has been collected. There is further risk that the user can be identified and retaliated against as the evidence is transmitted and stored. Both eyeWitness to Atrocities and CameraV offer built-in anonymity options as a means of risk reduction.

[eyeWitness to Atrocities] offers users anonymity by connecting verification of uploaded images to the phone itself rather than to any user associated with the phone. Everything a user records while inside the app is automatically tagged and encrypted, with a hash value of the pixel count recorded as a marker against subsequent manipulation.43

CameraV also creates a unique hash value identifier connected to the phone rather than the user to protect the user’s anonymity.44

Although both apps encrypt information and send it to a secure location, there is always a risk that these secure sites will be hacked. When discussing the issue of hacking, Mark Ellis, the Executive Director of the IBA, explained the decision to store eyeWitness to Atrocities’s footage in a secure evidence locker based in London:

And the idea I had was, well, go to a group that gets a hundred thousand attempts to penetrate their system every single day. And that’s LexisNexis […] [F]ortunately, LexisNexis works very closely with the International Bar Association, and so we approached them and sat down and said here is what we have, what do you think […] They have built us this backend database on this, and one of the things we had to ask ourselves is where should this backend vault be? Should it be in the United States, or should it be somewhere in Europe? Where do you want to hold that, and it was felt that in the United States, there wasn’t that degree of confidence actually […] and we made the decision, and so that vault now is in Europe. It is in London.45

The IBA now promises that their partnerships ensure that all footage is secure from hacking.46 “Yet with all the best protocols in place, recent experience suggests that highly motivated hackers are able to breach highly sophisticated security systems.”47

C. Security Risks at Trial

This risk of retaliation follows the user through trial as well. The rules of disclosure that apply before and during a trial “may require that both documenters and subjects be named and made available for questioning by the other party in the case.”48 While the user-generated evidence apps are designed to ensure a level of anonymity, preventing adherence to disclosure rules, “there is always the possibility that a motivated defense team will be able to uncover their identity.”49 The protection regime of the ICC has provisions to address this; however, their resources for providing protection may be limited.50 Additionally, there is no guarantee that the user will be categorized as a witness and fall under the ICC’s existing witness protection scheme. Indeed, the goal of these apps is to make the apps themselves witnesses. As Hamilton notes:

[F]or those users who film linkage evidence after the fact, and who never appear in the Hague, their role is arguably more akin to that of an intermediary—someone who connects the Court with a witness. And, for now at least, the Court only offers protection services to intermediaries on a case-by-case basis.51

Because of these many uncertainties, “many resources that encourage the documentation of crimes rightly emphasise informed consent at each stage of creating and sharing such information to ensure that participants are fully informed about the potential risks of their involvement.”52 Whether or not the INGOs deploying user-generated apps succeed in receiving a user’s informed consent is a more complicated question that is addressed in the next section of this comment.

IV. Recommendations to the OTP to Mitigate Security Risks Inherent in the Collection, Retention, and Use of User-Generated Evidence

A. Mitigating Security Risks in the Collection of User-Generated Evidence

As discussed throughout this comment, security risks are inherent in the role of evidence collection. While the IBA and WITNESS have sought to minimize these risks through strategic app design and education efforts, these efforts are not infallible. Security risk may be impossible to eliminate, but it can be mitigated by securing meaningful informed consent and the management of an individual’s expectations. As Hamilton notes:

The greater [an individual’s] expectations, the more risks they are likely to be willing to take. And the key question for INGOs involved in marketing user-generated evidence apps or otherwise encouraging local populations to document crimes, as well as for courts relying on user-generated evidence more generally, is the extent to which they are confident that users fully comprehend the risks and benefits involved in evidence collection.53

In order to give meaningful informed consent, the user needs to be informed of all risks and voluntarily accept those risks.54 In order for consent to be voluntary, one party cannot “promise, nor even raise the prospect of, a benefit that is not assured.”55 This is especially important in the context of user-generated evidence, where the benefits of gathering the evidence are still unknown. If users believe that the footage they capture will be used to secure a conviction against a perpetrator, they may be willing to take more risks.56 If they are aware that the evidence might not be admissible in Court, or that it may be utilized by the perpetrator himself, they might not take the same risks.57

The OTP could ensure that users are able to give meaningful informed consent moving forward by mandating risk management and expectation management in its contracts with INGOs, as suggested by Hamilton:

While the two organizations behind the user-generated evidence apps that are currently on the market are trying to reduce the risk of unmet expectations in their outreach programs, the situation for users going forward cannot be left to the good faith of these organizations. There is nothing to prevent other organizations entering this market in the future, and—in the absence of a contract or similar mechanism—no guarantee that such organizations will follow the practices that the IBA and WITNESS have been trying to develop.58

If the ICC were to contract with INGOs, it could ensure consistent messaging to users about the risks of evidence gathering and help to ensure that users are not taking those risks on the assumption that it will be used in Court.59

INGOs involved in collecting user-generated evidence are acting as intermediaries by connecting Court officials to evidence, and traditional intermediaries are contracted by the Court and bound by an associated code of conduct.60 Hamilton cites two major concerns that any contract would need to address: “user security and the realistic management of user expectations.”61 To address user security, the OTP should require that INGOs are trained on the Court’s Guidelines on Good Practices on Risk Prevention and that all interactions with users align with the principles of harm minimization.62

To ensure that consent is given voluntarily:

[T]he OTP could usefully draw from the language in its existing Code of Conduct for Intermediaries to specify that an INGO doing outreach on user-generated evidence ‘shall not make commitments to […] (potential) witnesses […] that he/she/it is not in a position to fulfill.’63

This would include any promise that the user may remain anonymous, given the uncertainties discussed previously, and would also require full disclosure that the evidence may never be used by the Court at all.64 If the OTP establishes that it will not accept user-generated evidence provided by any INGO found to violate these contract provisions, INGOs will be highly incentivized to comply.65

B. Mitigating Security Risks in the Retention of User-Generated Evidence

The International Committee of the Red Cross (ICRC) has established Protection Guidelines as a model of professional standards for human rights actors and humanitarian actors working in conflict situations.66 Hamilton reminds us that “INGOs are not legally required to follow these, or any other, guidelines. But Courts seeking to use user-generated evidence in their investigations and proceedings could establish a norm of getting INGOs to do so by stipulating that they will only accept user-generated evidence gathered through organizations that adhere to these guidelines.”67

In 2013, the Protection Guidelines were revised to address advances in information and communication technology, with special consideration given to the ways in which sensitive information should be managed.68 The subsequent and most recent revision, issued in June 2018, further emphasized that “the rapid proliferation of initiatives to use information and communication technology in new ways for protection purposes and the growing body of law on data protection” requires updated guidelines for information management.69

Chapter 6 of the Guidelines establishes standards for:

[T]he processes of handling data and information (the collection, analysis, storage, sharing, use, destruction or archiving of data and information) that are required to enable evidence-informed action for quality protection outcomes.70

While explicitly addressed to human rights and humanitarian protection actors, the Guidelines emphasize the importance of these standards to other actors in the field of information gathering:

The standards outlined in this chapter should also be applied, when appropriate, by professionals working in humanitarian or human rights responses but not in traditional “protection” work—for instance, people working in assistance, media and communication, fundraising and information communication and systems technology—who may also be processing personal data and protection data and information collected from persons affected, witnesses or other sources.71

The Guidelines identify several security risks that individuals face during the collection, retention, and use of user-generated evidence that were discussed in the preceding section, ranging “from physical and mental harm or threats of harm, discrimination, social marginalization and stigmatization.”72 The Guidelines seek to mitigate these risks through a series of directives related to data minimization, data retention, data security, confidentiality, and information sharing.73 Specifically, as it relates to the risk of third-parties accessing the data while it is retained, the Guidelines state:

If an appropriate level of confidentiality and security for personal data or sensitive protection information cannot be guaranteed, the protection actor should refrain from collecting the data, or from transferring them. If security challenges develop, owing to a change in the environment since the data collection, the protection actor should destroy the data if it is not able to mitigate the risks to data security.74

By stipulating that user-generated evidence will only be accepted if these Guidelines are adhered to, the Court could mitigate the security risks inherent in the retention of such evidence and ensure that any future apps adhere to the highest standards of security.

C. Mitigating Security Risks in the Use of User-Generated Evidence

As discussed previously, the risks that users face at trial will depend on the way that user-generated evidence is characterized, if it is accepted by the Court at all. Hamilton discusses the two ways in which the Court might characterize the user:

If they accept the argument that a user-generated evidence app is a witness, then the user who recorded the footage is best analogized to an intermediary—someone who has helped the Court gain access to a witness. If, on the other hand, the judges require the authenticating testimony of a user to admit the video into evidence, then that user is best analogized to a witness.75

If the user is characterized as a witness, their security risks at trial can be mitigated through the Court’s existing witness protection scheme.76 While this characterization might ensure the safety of the user at trial, it also runs counter to the purpose of user-generated evidence apps that arose in response to the need for diverse evidence other than witness testimony.77 “The great hope of those involved in developing user-generated evidence apps, however, is that judges will accept the app itself as the witness.”78 Indeed, the apps were designed with self-authentication in mind.79 If the app itself is accepted as a witness by the Court, the user would not automatically fall under the Court’s witness protection scheme.80 Instead, the user is better analogized to an intermediary. Hamilton recommends two ways in which the Court could mitigate the risks that the user would face at trial if categorized as an intermediary: through an ICC-user contract and the application of the ICC Guidelines on the Use of Intermediaries.81

An ICC-user contract would act in the same way as an ICC-intermediary contract and would provide the user with the same amount of security.

Once someone is formally acknowledged by the Court as an intermediary in this way, then the Court has a degree of responsibility for that intermediary’s security. The Court’s responsibility is a subsidiary one. Per the model contract between the ICC and an intermediary, the primary responsibility for the intermediary’s security lies with the intermediary itself.82

However, the Court must take measures to manage those risks, up to and including the use of protective measures.83

However, “the Court does not generally contract with ‘self-appointed’ or ‘one-off’ intermediaries, which is likely to be an accurate characterization of most users.”84 Therefore:

[W]ritten guidelines, while not legally binding, could serve as a backstop and provide clarity about which actors are responsible for reducing risks to users (as well as what users are themselves responsible for).85

The ICC Guidelines on the Use of Intermediaries stress the responsibility of the intermediary to take precautions for their own safety, but when these measures fail to protect the user, then “appropriate protective measures shall be implemented [by the Court].”86 While the Guidelines apply automatically to any intermediary formally contracted by the Court, their coverage is decided on a case-by-case basis for intermediaries operating outside of a formal contractual relationship.87

As discussed above, users are unlikely to have a formal contractual relationship with the Court, leaving them no guarantee of coverage under the Guidelines. To resolve this uncertainty, Hamilton suggests that:

[T]he Court make a formal determination, ideally in advance of the Al-Werfalli trial, that any user whose footage is used at any stage of the legal process falls within the Guidelines. Making a categorical determination in relation to users whose footage the Court relies on would be a marked improvement in the security situation for users.

V. Conclusion

While some level of risk will always be inherent in evidence gathering, user-based evidence brings with it a set of unique concerns, as these risks are shifted away from institutional actors to private citizens. INGOs have sought to mitigate this risk and maximize the potential admissibility of user-based evidence through strategic app design. The OTP can further mitigate these risks throughout the various stages of user-based digital evidence, from evidence-gathering to trial, in its contracts with INGOs and with users themselves. The OTP should look to the contracts, guidelines, and policies that it already uses for intermediaries to ensure that users are able to give meaningful informed consent and to ensure that INGOs are using best practices in the gathering and storing of sensitive information through their evidence-gathering apps.

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

  1. 1.

    The Prosecutor v. Mahmoud Mustafa Busayf Al-Werfalli, ICC-01-11-01/17, Warrant of Arrest (PTC I, Aug. 15, 2017), available online.

  2. 2.

    International Bar Association, Evidence Matters in ICC Trials 27 (Aug. 5, 2016) [hereinafter Evidence Matters], available online.

  3. 3.

    Rebecca J. Hamilton, The Hidden Danger of User-Generated Evidence for International Criminal Justice, Just Security (Jan. 23, 2019), available online.

  4. 4.

    Rebecca J. Hamilton, User-Generated Evidence, 57 Colum. J. Transnat’l L. 1, 55 (Nov. 7, 2018), available online.

  5. 5.

    Ash Turner, How Many Smartphones Are in the World?, BankMyCell, available online (last visited May 25, 2020).

  6. 6.

    International Telecommunication Union, Measuring Digital Development: Facts and Figures 2019, 8 (Nov. 18, 2019), available online.

  7. 7.

    Hamilton, supra note 4, at 15.

  8. 8.

    Evidence Matters, supra note 2.

  9. 9.

    Hamilton, supra note 4, at 3.

    (The term “user-generated evidence” was coined by Hamilton to refer to the trend of technologists and criminal justice advocates coming together to collect content recorded by ordinary citizens).

  10. 10.

    Id.

  11. 11.

    UC Berkeley Human Rights Center, Digital Fingerprints: Using Electronic Evidence to Advance Prosecutions at the International Criminal Court 3 (Feb. 2014) [hereinafter Digital Fingerprints], available online.

  12. 12.

    Office of the Prosecutor, ICC, Strategic Plan June 2012–2015, 22 (Oct. 11, 2013), available online.

  13. 13.

    Id.

  14. 14.

    Digital Fingerprints, supra note 11.

  15. 15.

    Id. at 9.

  16. 16.

    Evidence Matters, supra note 2, at 21.

  17. 17.

    Id.

  18. 18.

    Office of the Prosecutor, ICC, Strategic Plan 2019–2021, 13 (Jul. 17, 2019), available online.

  19. 19.

    Id. at 21–22.

  20. 20.

    Hamilton, supra note 4, at 17–18.

  21. 21.

    Id. at 18.

  22. 22.

    Id.

  23. 23.

    Id. at 28.

  24. 24.

    Nikita Mehandru & Alexa Koenig, Open Source Evidence and the International Criminal Court, 32 Harv. Hum. Rts. J., n.21 (Apr. 15, 2019), available online.

  25. 25.

    Id.

  26. 26.

    Hamilton, supra note 4, at 18.

  27. 27.

    Id. at 21.

  28. 28.

    Id.

  29. 29.

    Id.

  30. 30.

    Id. at 24–25.

    (Linkage evidence is evidence which connects the crime to the person or people responsible).

  31. 31.

    Id. at 25.

  32. 32.

    Id.

  33. 33.

    Id. at 26.

  34. 34.

    Hamilton, supra note 3.

  35. 35.

    Id.

  36. 36.

    FAQs: How does eyeWitness protect the security of the app users?, eyeWitness, available online (last visited May 25, 2020).

  37. 37.

    Evidence Matters, supra note 2.

  38. 38.

    Hamilton, supra note 4, at 37.

  39. 39.

    International Committee of the Red Cross, Professional Standards for Protection Work 109 (3d ed. 2018) [hereinafter Professional Standards], available online.

  40. 40.

    Hamilton, supra note 4, at 38.

  41. 41.

    Owen Bowcott, eyeWitness to Atrocities: The App Aimed at Bringing War Criminals to Justice, The Guardian, Jun. 7, 2015, available online.

  42. 42.

    Hamilton, supra note 4, at 38

    (quoting IBA director Mark Ellis’s acknowledgment that governments can seize the device and discover it was used to record evidence).

  43. 43.

    Id.

  44. 44.

    Id.

  45. 45.

    Mark S. Ellis, Shifting the Paradigm—Bringing to Justice Those Who Commit Human Rights Atrocities, 47 Case W. Res. J. Int’l L. 265, 269–70 (2015), available online.

  46. 46.

    Hamilton, supra note 4, at 44.

  47. 47.

    Id. at 45.

  48. 48.

    Evidence Matters, supra note 2.

  49. 49.

    Hamilton, supra note 4, at 51.

  50. 50.

    Evidence Matters, supra note 2, at 27.

  51. 51.

    Hamilton, supra note 4, at 38.

  52. 52.

    Evidence Matters, supra note 2, at 27.

  53. 53.

    Hamilton, supra note 4, at 52.

  54. 54.

    Id. at 30.

  55. 55.

    Id.

  56. 56.

    Id. at 30–31.

  57. 57.

    Id. at 31.

  58. 58.

    Id. at 54.

  59. 59.

    Id.

  60. 60.

    Id.

  61. 61.

    Id.

  62. 62.

    Id.

  63. 63.

    Id. at 55.

  64. 64.

    Id.

  65. 65.

    Id.

  66. 66.

    Professional Standards, supra note 39.

  67. 67.

    Hamilton, supra note 4, at 59.

  68. 68.

    International Committee of the Red Cross, Professional Standards for Protection Work 15 (2d ed. 2013), available online.

  69. 69.

    Professional Standards, supra note 39, at 16.

  70. 70.

    Id. at 106.

  71. 71.

    Id. at 110.

  72. 72.

    Id. at 109.

  73. 73.

    Id. at 110.

  74. 74.

    Id. at 131.

  75. 75.

    Hamilton, supra note 4, at 55.

  76. 76.

    Id. at 56.

  77. 77.

    Evidence Matters, supra note 2.

  78. 78.

    Hamilton, supra note 4, at 56.

  79. 79.

    Id.

  80. 80.

    Id.

  81. 81.

    Id.

  82. 82.

    Id.

  83. 83.

    Id.

  84. 84.

    Id. at 57.

  85. 85.

    Id.

  86. 86.

    Id. at 58.

  87. 87.

    Id. at 57.

Protecting Against Deepfakes: How the ICC Can Ensure Trust in the Verification and Use of Open-Source Evidence

I. Introduction

Digital devices have allowed ordinary civilians to become on-the-ground investigative reporters in almost every region of the world. Contemporaneous uploads of footage featuring protestors being tear-gassed in Hong Kong or Syrians fleeing from chemical warfare have allowed viewers all over the world to see, hear, and emotionally feel human rights crises unfold in real-time. As a result, the amount of online evidence and information available to prosecutors and investigators through open-source channels has raised the possibility of reinvigorating—or even changing the narrative—in formerly stalled investigations. The Office of the Prosecutor (OTP) has already begun to utilize open-source information from social media and other online platforms to further investigations. In 2017, the ICC issued a first-of-its-kind warrant for the arrest of Mahmoud Al-Werfalli, a commander within the Libyan National Army, based on open-source videos.1 In the sphere of international criminal law and the fight against human rights abuses, the new and increasing use of open-source information has opened up the possibility for investigations to be better, safer, and more efficient than ever before.

Open-source refers to publicly available online materials, ranging from photographs to documents to audio and visual recordings.2 Such information, commonly housed on social media sites, is often posted by ordinary civilians and is available and growing at a prodigious rate.3 The benefits of open-source information are clear. As Professor Danielle Citron of Boston University School of Law testified to the U.S. House Permanent Select Committee on Intelligence, the use of video and audio recordings obtained through online public records has allowed us to “become firsthand witnesses to events, obviating the need to trust others’ accounts.”4 Content captured contemporaneously by any individual with a capable phone or device has the power to strengthen investigations in ways that documentation and evidence gathered at a later time by investigators plainly do not. Open-source data enables researchers to advance investigations from anywhere around the world rather than risk traveling to volatile regions where they may be severely restricted or put in danger. At the same time, the use of such data in the judicial system has the potential to make investigations safer by corroborating and giving strength to witness testimony or by lowering the necessity for vulnerable witnesses to testify in court in the first place.5 Rebecca Hamilton, formerly of the OTP, suggests open-source evidence has the ability to “democratize” investigations by giving power to local communities and their own ability to bring an end to the abuses they have suffered or witnessed first-hand.6

An increased reliance on and use of open-source information in the judicial system is not without consequences. A growing trust in the authenticity of content posted online has given bad actors the opportunity to alter open-source data to their advantage in order to convey a false or dangerously misleading narrative separate from reality. The term ‘deepfake’ is used to refer to digitally distorted content such as “videos generated via algorithms that make it look like a person said or did something she did not.”7 Deepfakes, originally popularized in the porn industry, are typically sets of data merged together.8 As Professor Citron articulated, “[c]reators of deepfakes count on us to rely on what our eyes and ears are telling us, and therein lies the danger.”9 As of October 2019, at least 14,678 videos considered to be deepfakes exist online and are free for anyone to access—and to erroneously interpret as true—around the world.10

Addressing the problems that go into verifying and authenticating open-source evidence to protect against the rising prevalence of deepfakes is particularly difficult given the ICC’s lack of financial and personnel resources available to tackle the extraordinary amount of open-source information that continues to multiply every day. Thus, this comment argues that the ICC must develop partnerships with third-party organizations already employing trained data scientists with the forensic knowledge to equip the ICC with resources to publicly verify open-source evidence. A failure to adapt and evolve in the digital age could cause the ICC to risk losing relevance and public trust during a time when the collection of key evidence in human rights investigations may be more critical than ever.

This comment does not touch on the complex evidentiary rules governing the Court that likely pose separate challenges to the use of open-source evidence, the difficulties regarding authenticating the chain-of-command when using publicly-sourced evidence, or the use of user-generated applications that seek to protect the anonymity of those who provide open-source evidence in the first place. This comment does, however, focus on the need for the ICC to hold itself to a higher standard as a permanent international tribunal and argues that the Court must establish a blanket verification process for all open-source evidence. A strong and neutral authentication process will safeguard the validity of the OTP’s investigations, the rights of the accused whose freedoms may hinge on the soundness of such evidence, and the ICC as a whole. To do so, this comment will discuss the verification approaches of other non-state actors that offer paradigmatic examples for the ICC and the OTP to learn from. This comment then seeks to expand on arguments and recommendations by other qualified experts in the field.

II. Paradigmatic Verification Models by Non-State Actors

The consequences of failing to authenticate and verify open-source information are obvious and severe. False convictions contingent on doctored evidence present a human rights challenge in itself; allowing a doctored video to enter through the cracks of the judicial system risks undermining trust in prosecutorial investigations at a time when global confidence in the media is critically low.11 As of now, the OTP appears to engage primarily in simple verification schemes focused only on substantiating the time, date, and location of publicly provided evidence.12 But to evolve in an era where bad actors are constantly scheming for ways to outsmart and outmaneuver prosecutors and academics with advanced digital capabilities and new technology, the ICC must do more to advance its verification capabilities beyond being able to identify obvious open-source alterations. Lessons can still be taught from Reuters’ inadvertent use of a digitally altered photo in 2007 of airstrikes during the Lebanon war that appeared to show more smoke rising from military action than in reality.13 The New York Times has also had to publicly confront the aftermath of a similar mistake when they published a photograph distributed by Sepah News, Iran’s state media site, of an Iranian missile test in 2008 that was discovered to have likely been digitally altered to look as if there was an additional missile (note that there is no confirmation of official Iranian government involvement in the doctored photograph).14

Amnesty International has led the way in training volunteers with verification skills through their creation of the Digital Verification Corps (DVC).15 The DVC was created to specifically tackle the authenticity challenges faced by an overwhelming amount of open-source evidence. With roughly 120 student volunteers installed in at least six universities worldwide and growing, the DVC continues to train “highly-qualified video verifiers” in an effort to assimilate digital verification with critical human rights investigations.16 Sam Dubberley, the manager of the DVC, stated:

When we are not on the ground to witness these events as they happen, the onus is on us to prove beyond a shadow of a doubt that they really did happen.17

In 2018, the DVC single-handedly took the lead in verifying video footage that appeared to show soldiers shooting two women and children in the head, despite the Cameroonian government’s denial and resolute claim that the video was a deepfake.18 As a result of the DVC’s analysis, the Cameroonian government finally arrested the soldiers and acknowledged the executions.19 With a growing presence around the world of trained and passionate students, the DVC stands at the ready to analyze and verify open-source content in varying languages and time-zones.20

The UC Berkeley Human Rights Investigations Lab has also played a leading role in digital verification by training and combining resources with the DVC.21 The Investigations Lab has orchestrated and hosted workshops to engage and train students with technology at the forefront of investigative research with the goal of advancing investigative abilities to tackle human rights abuses with reliable open-source information.22 At the same time, the UC Berkeley Human Rights Center has already engaged with the OTP through similar workshops aimed at educating investigators on how to use and verify open-source information in order to overcome evidentiary thresholds.23

Bellingcat is a different kind of open-source investigative site, started by an international blogger, that integrates and equips journalists with new technologies and publicly sourced information.24 Bellingcat prioritizes collaborating with other organizations and contributors in order to grow its network and spread knowledge and investigative skills on how to better use open-source evidence in the investigative context.25 The group has already begun to assist the ICC in the Court’s understanding and use of open-source information.26 Finally, the Atlantic Council’s Digital Forensic Research Lab is another prominent example of a leading nonpartisan organization with a mission to break down misinformation on platforms that host open-source information.27 Housed with digital forensic analysts, the Digital Forensic Research Lab tracks international events and publishes reports that seek to dissect open-source methodologies and shed light on distortion campaigns happening all over the world.28

III. Recommendations

The ICC must adopt a coordinated solution to an international challenge that knows no borders and has no boundaries. Regardless of the inculpatory or exculpatory value of certain open-source evidence the ICC seeks to use in an investigation, the ICC —and in turn the OTP —must use the same consistent and public verification process for authenticating open-source material. A consistent authentication process will go a long way in fostering trust in the ICC’s use of such open-sourced data in the first place. This is particularly important in cases where an investigation hinges on a piece of video or audio evidence obtained on an open platform, as the Court must be particularly mindful in order to protect all parties involved in proceedings before the tribunal—including the accused. As commentators have noted, the use of open-source evidence carries with it many concerns for defendants, such as:

[I]ts proclivity to convey a prosecution-biased picture of the crimes and to exacerbate the inequality of arms that already disadvantages defendants in international criminal proceedings.29

When the burden is often shifted to the defendant to prove reliability—or lack thereof, the Court should do everything it can to make sure proceedings are fair and defendants have the same ability to use open-source evidence as the prosecutors. Investigations particularly dependent on open-source evidence should have an even higher impetus to prove reliability through an immediate and public verification process.

Second, the ICC must establish concrete partnerships with third-party investigative bodies. Experts like Hamilton have acknowledged the need for the OTP to engage new players in understanding, using, and verifying the massive amounts of data relevant to the Prosecutor’s investigations and proceedings before the ICC.30 The DVC’s unique utilization of university students is a model example of how to overcome budget challenges; modern day academics are being outrun by individuals with forensic knowledge and the ICC should take the initiative to learn from—and train—students with the digital know-how to quickly and accurately verify and spot digitally altered open-source data. Involving students in the ICC’s investigative proceedings has the dual purpose of both advancing the ICC’s ability to function and training the next generation of human rights defenders. The ICC should furthermore require the OTP to communicate with experts throughout the investigative process, ensuring such experts are available to speak to the truthfulness and verification of open-source evidence when needed.

Lastly, the ICC and the OTP must be forthright in acknowledging the consequences that come with too heavy of a reliance on open-source evidence. As technology makes it even easier to create deepfakes, hostile state and nonstate actors may be incentivized to forge open-source evidence as leverage. At the same time, a growing awareness of the prevalence of deepfake content might cast doubt on open-source evidence in a perverse way; furthering what Professor Citron calls the “liar’s dividend,” where liars try to avoid culpability by twisting the narrative and claiming a real video is in fact fake. 31 As aforementioned, the Cameroonian government tried to get away with doing just that.32 Calls claiming open-source evidence is ‘fake news’ may discredit the ICC’s efforts to integrate open-source evidence into formal proceedings, further justifying the need for a coordinated and neutral verification system to shield against attempts to discredit the use of open-source information.

IV. Conclusion

The ICC and the OTP must welcome the entrance of ordinary civilians as players in human rights investigations. It is in the OTP’s best interest to prioritize the successful use of open-source information in order to ensure investigations may proceed past evidentiary thresholds.33 The Court should remain hesitant to over-rely on any one piece of open-source content, while, at the same time, adopting a verification scheme with the assistance of third-party organizations that prioritizes and safeguards the rights of all parties subject to the tribunal’s investigative arm. A failure to do so would be a failure of the Court’s critical obligation to evolve to the modern age.

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

  1. 1.

    The Prosecutor v. Mahmoud Mustafa Busyf Al-Werfalli, ICC-01/11-01/17, Warrant of Arrest (PTC I, Aug. 15, 2017), available online; Alexa Koenig, Open Source Evidence and Human Rights Cases: A Modern Social History, in Digital Witness: Using Open Source Information for Human Rights Investigation, Documentation, and Accountability 32, 40 (Sam Dubberley, Alexa Koenig & Daragh Murray, eds., Feb. 19, 2020), paywall, archived.

  2. 2.

    See Alexa Koenig, “Half the Truth is often a Great Lie”: Deep Fakes, Open Source Information, and International Criminal Law, 113 Am. J. Int’l L. Unbound 250 (Aug. 19, 2019), available online, doi.

  3. 3.

    Id.

  4. 4.

    See, e.g., Danielle Keats Citron, Professor of Law, House Intelligence Committee Hearing on the National Security Challenge of Artificial Intelligence, Manipulated Media, and “Deep Fakes”, 2 (Jun. 13, 2019) [hereinafter Citron Testimony], available online, archived.

  5. 5.

    Id.

  6. 6.

    Rebecca J. Hamilton, User-Generated Evidence, 57 Colum. J. Transnat’l L. 1 (Nov. 7, 2018), available online.

  7. 7.

    Koenig, supra note 2, at 252.

  8. 8.

    Id.

  9. 9.

    Citron Testimony, supra note 4, at 2.

  10. 10.

    Rachel Metz, The Number of Deepfake Videos Online is Spiking. Most are Porn, CNN, Oct. 7, 2019, available online.

  11. 11.

    Edelman Trust Barometer 2018—UK Findings, Edelman (Jan. 22, 2018), available online.

  12. 12.

    Hamilton, supra note 6, at 18.

  13. 13.

    Paul Holmes, Reuters Toughens Rules after Altered Photo Affair, Reuters, Jan. 18, 2007, available online.

  14. 14.

    Mike Nizza & Patrick J. Lyons, In an Iranian Image, a Missile Too Many, N.Y. Times, Jul. 10, 2008, available online.

  15. 15.

    Conor Fortune, Digitally Dissecting Atrocities: Amnesty International’s Open Source Investigations, Amnesty Int’l (Sep. 26, 2018), available online.

  16. 16.

    Paola Verhaert, Amnesty International’s Digital Verification Corps: New Networks and Methods for Human Rights Research, The Engine Room (Jun. 19, 2017), available online.

  17. 17.

    Fortune, supra note 15.

  18. 18.

    Verhaert, supra note 16.

  19. 19.

    Id.

  20. 20.

    Id.

  21. 21.

    Koenig, supra note 1, at 44.

  22. 22.

    Id.

  23. 23.

    Id. at 32.

  24. 24.

    About, Bellingcat, available online (last visited May 25, 2020).

  25. 25.

    Id.

  26. 26.

    Koenig, supra note 1, at 44.

  27. 27.

    Digital Forensic Research Lab, Atlantic Council, available online (last visited May 25, 2020).

  28. 28.

    Id.

  29. 29.

    See Nancy Amoury Combs, Investigative Delegations: Predictable Predicaments, 113 Am. J. Int’l L. Unbound 267, 270 (Aug. 19, 2019), available online, doi

    (citing Hamilton, supra note 6, 39–42).

  30. 30.

    Id.

  31. 31.

    Bobby Chesney & Danielle Keats Citron, Deep Fakes: A Looming Challenge for Privacy, Democracy, and National Security, 107 Cal. L. Rev. 1753, 1758 (2019), available online, doi.

  32. 32.

    Verhaert, supra note 16.

  33. 33.

    Hamilton, supra note 6, at 12.

The “Coming Storm”: Possibilities for Preserving Evidence of War Crimes Published on Major Online Service Providers

I. Introduction

There should be no greater windfall for criminal prosecutors, national or international, than the growth of the Internet as a tool for communication and networking. Potential evidence against perpetrators abounds: photos and videos posted by offenders, victims, and witnesses alike; personal communications that can establish a wrongdoer’s motive or intent or uncover a conspiracy; even perpetual, personalized tracking devices in the form of cell phones that have the potential to pinpoint a person’s whereabouts down to a specific address whenever turned on. Legal systems the world over have recognized the importance of utilizing this type of online data and evidence and the International Criminal Court (ICC) is no exception. From an institutional perspective, the ICC has clearly identified digital evidence as a potentially vital source of data while carrying out investigations. As early as 2012, the Office of the Prosecutor (OTP) has consulted on workshops in which the growth in importance of digital evidence was recognized as a “coming storm.”1 The ICC has also recently begun using digital data more frequently in practice. The investigation of the conflict in Darfur, culminating with the 2015 cases of Banda, Jerbo, and Abu Garda, utilized satellite imaging from private parties, including Google Earth, to demonstrate village destruction and follow population and troop movements.2 In the Al Mahdi case, the Court was presented evidence from YouTube and other audio posted online.3 Evidence from YouTube and social media sites was presented in the Al-Werfalli case, and evidence from Facebook helped prove witness tampering in the Bemba case.4 It is reasonable to infer, therefore, that the ICC recognizes the potential for digital evidence to be utilized in its investigations.

That being said, the challenges presented by the vast expanses of potential data stored on major online platforms have not been entirely dealt with on an international level, and there are aspects of the use of digital evidence that have not been completely reconciled by the ICC. Amongst the myriad of challenges presented by the rapid growth in the prevalence and relative importance of digital evidence is that a significant percentage of the evidence that could be used by the ICC in its investigations is posted to major online service providers (OSPs) like Twitter, Facebook, or YouTube. Each of these OSPs have some form of overarching community guidelines, all of which disallow—and promise to remove—posts that depict images of violence or exploitation of humans. While those guidelines ostensibly seem altogether reasonable, they produce an unfortunate consequence in that the automatic removal process for posts flagged for violations of community guidelines sometimes removes valuable evidence of crimes falling under the purview of the ICC. In recent years, both Facebook and YouTube have removed significant amounts of content depicting atrocities perpetrated during the Syrian Civil War, and Facebook has also removed evidence of further potential crimes against humanity committed in Myanmar against the Rohingya population.5 This practice poses an issue for the ICC. When content is removed by the platform it is published on, the data itself as well as all of its authenticating markers is no longer accessible through the site. Unless it has been saved elsewhere, the only way to retrieve the data is to get it from the firm’s servers, which comes with significant barriers to an international tribunal like the ICC.

Given the issue laid out above, this comment addresses whether the ICC has a reliable means to work with the OSPs, and the countries in which the OSPs are incorporated, in order to access data that has been removed from public view for violating community guidelines and, barring that, whether there are alternative measures the ICC can take to ensure the reliable and real-time collection of evidence posted online. Further evidence will first be presented that major online service providers use community guidelines to remove potentially relevant evidence posted on their sites. Following that, the discussion turns to the potential ways through which the ICC can work with the American and Irish governmental and legal systems specifically, as the main OSPs are American firms with Europe, the Middle East, and Africa (EMEA) Headquarters in Ireland, and whether those avenues allow for enough access for the ICC to retrieve any relevant evidence from an OSP server. Finally, two alternative actions that could be undertaken by the ICC to protect its interest in online data are presented and analyzed; first, potential bilateral cooperation agreements with OSPs, and second, the establishment of a new online forum under the purview of or otherwise attached to the ICC for people with evidence of atrocities to submit their evidence directly. In the end, no one potential action is sufficient to solve the issue: this comment proposes that the most effective manner through which the ICC can ensure that it has access to any relevant digital evidence is to establish its own private forum dedicated to the collection and storage of online data, maintain dialog with the necessary OSPs until such a time as the ICC program becomes viable, and pursue what national and international legal or governmental cooperation is necessary.

II. The Problem: Facebook and YouTube Removes Evidence of War Crimes

Although the ICC has yet to formally open an inquiry into the various accusations of atrocities committed during the conflict in Syria, a preliminary examination into the actions taken against the Rohingya population in Myanmar was opened in 2018 and the OTP was granted the right to investigate in 2019.6 Both conflicts present a real possibility that crimes against humanity, whether war crimes or an outright genocide charge, have occurred and the ICC has been called upon to investigate the actions of multiple parties in the Syrian conflict through the same procedural stance by which it is now investigating the conflict in Myanmar. In addition, both conflicts illustrate the potential for the use of digital evidence to corroborate the occurrence of atrocities and the problems that stem from posting such evidence on major OSPs.

A. Facebook

Despite its effective use in other human rights contexts (i.e. the Arab Spring), Facebook has established some community guidelines that have produced negative consequences for activists seeking to shed light on ongoing crimes against humanity in both Syria and Myanmar. Facebook’s community standards have specific provisions against posts depicting or containing references to violence and criminal behavior, the general exploitation of humans, and sexual exploitation specifically. Under the violence and criminal behavior standard, any posts that show or allude to organized hate, mass murder, and organized violence or criminal activity are considered a violation and are subject to removal.7 Moreover, Facebook’s guidelines against posts depicting human exploitation requires the removal of posts that show or incite the use of child soldiers8 and any posts that allude to or portray sexual exploitation.9

While those policies may be reasonable and even necessary, they have already been used to remove potentially relevant digital evidence that could be used by the ICC in ongoing or potential future investigations. The story of one Syrian man, Abdulsalam, who witnessed and recorded the bombing of a bakery in al-Bab, illustrates the issue perfectly. In January 2014, the Syrian army dropped a barrel bomb on a bakery near Abdulsalam’s mosque (the Syrian army had a pattern of using barrel bombs on bakeries in particular) in what he described as “a particularly horrific bombing.”10 He took a few pictures of the helicopter that dropped the bomb, then went directly to the scene, where he took pictures of the destruction of the bakery in which he captured images of the civilian victims of the bombing. He uploaded those images to his Facebook account promptly following the attack, believing it to be the most effective way to preserve the pictures and spread the word about the true toll of the attacks.11

Avi Asher-Schapiro, writing in the Intercept regarding Abdulsalam’s posts depicting the al-Bab bombing and its aftermath, details Facebook’s response to his posts and highlights the main issue with Facebook’s policy:

Seven months later, Abdulsalam got an automated email from Facebook notifying him that the images had been removed. Other users had complained that his photos were too gory. By the time he got the email, Abdulsalam’s other copies of the pictures were gone; his hard drive had been burned, along with his small office, when the Islamic State stormed al-Bab and Abdulsalam fled across the border to Turkey.12

In addition to the primary issue of losing direct access to Abdulsalam’s firsthand evidence, Facebook’s deletion of his pictures compounded the issues of evidence collection in Syria particularly. Specifically, the Islamic State had begun to move into Aleppo by the time that Abdulsalam took his pictures. While that sounds tangential, human rights NGOs on the ground in Syria, including Human Rights Watch (HRW), which had been documenting evidence of potential war crimes in Syria, had been pulled out prior to the January 2014 barrel bomb attack.13 One of HRW’s researchers on the ground, Ole Solvang, recognized the potential importance of Abdulsalam’s removed evidence when he said, “if there ever is a trial, this is the stuff that could become important evidence.”14

Moreover, the same situation played out in a more general sense when, starting in September 2017, Facebook began removing images posted by Rohingya Muslims regarding their treatment in Myanmar. Rohingya activists, including the Irish-based Mohammed Rafique, reported that their Facebook accounts and pages were suspended together with the removal of content they posted depicting images and videos of torture and killings in Rohingya villages.15

As evidenced by the removal of significant evidence of potential crimes against humanity in both Syria and Myanmar, Facebook’s community guidelines establish a system through which pertinent data posted online is removed from access by the general public before ICC investigators have a chance to find it. Despite amending its guidelines in 2016 to allow violent or graphic content that is “newsworthy, significant, or important to the public interest,” the reporting structure, through which users individually flag content that they feel violates Facebook guidelines which are then reviewed by Facebook employees, still allows important potential evidence to slip through the cracks. Moreover, there is every potential that individuals who have a Facebook account and the desire to protect or support individuals committing war crimes can flag evidence against those perpetrators and have it removed, thus making it far more difficult to ever find the evidence in the first place. Therefore, Facebook’s current structure makes the potential for evidence to be removed before the ICC can open investigations (as the Myanmar evidence was removed in 2017, prior to the OTP opening a preliminary examination in 2018) all too real, especially as people likely still believe, as Abdulsalam did, that Facebook is an effective tool to memorialize data.

B. YouTube

In a similar vein to Facebook, YouTube, through its parent company Google, has also established a series of policies meant to safeguard its site from violence, graphic content, and actions taken out of hatred for a specific group. YouTube’s policy specifically states that it does not allow any footage, audio, or photos of the aftermath of war, the aftermath of a terrorist attack, sexual assaults, torture, protests, riots, or corpses with significant injuries like detached limbs.16 YouTube also removes content promoting violence or hatred against groups based on a series of group identifiers (e.g. race, religion, gender).17 Both policies require the removal of any offending content.

While the issues with Facebook’s structure for enforcing its community guidelines causes issues due to its user-driven nature, YouTube’s system of using artificial intelligence to comb through videos posted on its site that violates its community guidelines against extremist or disturbing content has yielded similar negative consequences for the preservation of evidence of war crimes posted online. In 2017, the same year that Facebook was removing data and suspending accounts associated with images depicting the violence against the Rohingya in Myanmar, YouTube’s artificial intelligence system shut down the pages of 900 groups documenting atrocities in Syria. Included in those groups whose information was removed from the site were organizations focusing on analyzing images coming out of conflict zones, organizations tracking the toll of air strikes in Iraq and Syria, and local Syrian media organizations like the Idlib Media Center, which was one of the last groups producing content from rebel-controlled territory at the time.

When evidence of crimes against humanity is posted on their site, Schapiro writes:

[S]ocial media companies have over and over again chosen to ignore, and, at times, disrupt the work of human rights groups scrambling to build cases against war criminals.18

This is borne out through both the Syrian and Burmese examples, which show that, whether intentional or not, online service providers are likely going to remove evidence of the types of crimes under the purview of the ICC posted by their users because of policies put in place to combat other disturbing global trends. Moreover, OSPs are not allowed to share content data, only metadata, which only includes general account information and the location from which the evidence was posted. The content of the post is sequestered in the company’s servers and becomes all the more difficult to access once it has been removed from the platform. Given the competing forces that pressure OSPs to remove content that could otherwise be used as evidence of crimes against humanity at trial in the ICC, and given the difficulty of accessing said content once it is removed, the ICC needs to develop means of cooperating with either the OSPs or the countries in which those OSPs are incorporated to ensure that critical data is not lost in the name of community safety online.

III. Working with National Governments to Regain Removed Online Evidence, Potential and Pitfalls

In order to recover data that has been removed under the community guidelines of any OSP once it has been removed, ICC investigators would have to gain access to the provider’s servers or have that provider send them the data from the server on which it is backed up. As the ICC at current does not have any agreements with major OSPs to protect information that could be valuable, and because the major OSPs are American firms (and the US is not a state party to the Rome Statute which mandates member state compliance with ICC investigations), it may have to negotiate with the national authorities holding jurisdiction over the location of the server in order to obtain the removed data. This section outlines what that process could look like and the drawbacks to the approach generally.

A. United States

In order to understand the regulatory framework the ICC would have to navigate in order to obtain American help in retaining evidence posted online and removed under OSP community guidelines, it is necessary to examine the Stored Communications Act (SCA), which is found in Section II of the Electronic Communications Privacy Act.19 Under the SCA, which limits governmental and institutional access to stored electronic records, US domestic law enforcement may request information through an order or after the issuance of a subpoena or a warrant, depending on the type of information requested.20 With regard to foreign or international law enforcement officials, the Act is silent.21 Importantly, there is a distinction made between private OSPs and OSPs that serve the public—public-facing OSPs, the category under which Facebook, YouTube, and the like fall, are always subject to the provisions of the SCA and compliance is mandatory with regard to its rules that generally prohibit the disclosure of content.22

As was stated above, the SCA is silent on the matter of foreign or international law enforcement in favor of only explicitly allowing domestic law enforcement to request court actions to view stored data. In addition, longstanding US legal tradition holds that foreign law enforcement actions on its soil must be announced to and accepted by the government, even when a Mutual Legal Assistance Treaty (MLAT) exists between it and the foreign country conducting the investigative action. Because it would appear that the ICC would have to get help from US entities when searching for digital evidence removed and sequestered to an OSP’s server, we next have to analyze whether the ICC can effectively ask US legal and governmental institutions for aid when US-based OSPs remove potentially pertinent evidence posted online.

The most prevalent law governing USICC interactions since its adoption in 2002 is the American Servicemembers’ Protection Act (ASPA). In general, the ASPA sets out rules prohibiting US support for the ICC, including specific provisions against responding to a cooperation request from the ICC, transmitting a letter rogatory from the ICC, and using certain funds to aid the Court.23 Moreover, the Act bars the ICC and its “agents” from investigating within the US.24 Though those provisions seem to wholly remove the prospect of any potential aid from US institutions for the ICC, let alone help from the US in order to gain access to the servers of one of its companies, a last minute amendment by Senators Chris Dodd and Patrick Leahy (the “Dodd Amendment”) opens up the possibility of US aid in certain scenarios. The amendment states in relevant part:

[N]othing in this title shall prohibit the United States from rendering assistance to international efforts to bring to justice […] foreign nationals accused of genocide, war crimes or crimes against humanity.25

In addition, there is the potential for a Presidential waiver in order for the ICC to involve itself in an investigation in the US, though none of these waivers have been given out as of yet.26 This provision, on its face, allows US entities to cooperate with ICC investigations of non-US citizens and, as argued in Access Denied, opens up two interesting options for the ICC to work directly with US institutions on investigations, including when the Court needs access to an American OSP’s servers as a result of the removal of evidence posted online due to the OSP’s community guidelines. Those options are:

  1. a court-to-court approach through a request for judicial assistance from the US District Courts, and

  2. a court-to-country based, diplomatic approach.27

1. Requests for Judicial Assistance from US District Courts

Under Section 1782 of Title 28 of the U.S. Code, a section outlining proper assistance to foreign and international tribunals, foreign tribunals are permitted to send a letter rogatory to the district court attached to the location of the individual or entity they are pursuing, but only that court.28 The district court then is empowered to order the production of documents (amongst other rights), including for the purpose of foreign criminal proceedings conducted prior to formal accusation.29 Though the plain meaning of the Statute, which states that “foreign or international tribunals” can utilize this direct-to-court letter rogatory, should effectively qualify the ICC to utilize Section 1782, further proof of the right of the ICC to use this method was outlined in the Supreme Court case Intel Corp. v. Advanced Micro Devices, Inc., which established that the Statute may be used by any “interested person” and that the proceedings in question need not be imminent as long as they are “within reasonable contemplation.”30 This would seem to indicate that the Statute is interpreted by the US court system as broadly applicable and would allow for inquiries to be made during the preliminary examination phase.

The most significant benefits of addressing the issue of evidence being posted online and removed through a direct appeal to the US district courts are the timeliness and legitimacy factors, as well as removing the specter of noncooperation from the executive branch. American courts have among the best infrastructure of any legal system in the world, so the process will be as quick as possible and the affirmation of the relevance of the ICC investigation by an American court could lend it additional credence. Moreover, courts will look at an issue through a mostly legal lens (ideally), and, as such, political and security concerns will play less of a role.31 However, content data is not accessible through Section 1782 and the investigator would have to pinpoint the location of the data so that he or she could bring suit in that court, making it difficult to follow through on in practice.32 If content data is unavailable through this channel, it does not provide sufficient coverage to be a viable option through which to retrieve removed data.

2. Request for Judicial Assistance through Diplomatic Channels

Another potential route for the ICC to work directly with the American Government is through diplomatic letters rogatory sent to the State Department. Under 28 U.S.C. § 1781, the State Department is empowered to transmit letters rogatory to the addressed tribunal in the United States from foreign entities.33 Those are executed based on treaty obligations or “international comity and courtesy” and sent through the Justice Department’s Office of International Affairs (OIA). From there, OIA can proceed through the § 1782 process alluded to above, or it can use 18 U.S.C. § 3512 (the Foreign Evidence Request Efficiency Act or FEREA), which allows for the appointment of an Assistant US Attorney to act as an investigative intermediary and collect the necessary evidence.34 In addition, the Statute allows a federal judge to issue orders for appearance of witnesses or production of documents, and even allows federal judges to issue warrants to aid in international criminal investigations.35 FEREA differs significantly from § 1782 in that the request for assistance does not have to be filed in the specific court of jurisdiction and it opens up the possibility of access to content data as well as the metadata accessible under § 1782.36

The potential benefits for the ICC in utilizing letters rogatory through the State Department and the OIA are that it could extend the reach to content data, which would make it an actually viable option for the retrieval of data stored on servers located in the United States but removed from public view by OSPs for violating community guidelines. Rights under FEREA would need to be held to extend to international tribunals, as the Statute addresses requests from “foreign authorities” (though it is reasonable to believe that it would be, as criminal investigatory bodies are included under that moniker),37 but the extension of the right of foreign tribunals to access stored content data would essentially nullify the issue at hand. Despite that, these letters rogatory depend more heavily on the cooperation and discretion of the American executive branch, which does not bode well for their success rate, and international requests involving the discretion of multiple American branches of government tend to take significant time to complete. While diplomatic letters rogatory could potentially provide a direct solution to recovering removed digital data from OSP servers located in the United States, it would be necessary to wait for the right American administration to begin the process, the accused could not be American or from a US strategic ally, and the data in question would have to be stored in the US. In a situation which satisfies all of the above-mentioned conditions, they could be valuable.

B. Ireland

The public-facing OSPs that are most frequently at risk of removing pertinent digital evidence of crimes against humanity are all multinational organizations that host servers in multiple countries, despite being founded in the US. Therefore, some relevant data that is removed will be stored on servers in other countries necessitating involvement with nations outside of the US, and moreover the ICC could potentially look to partner with a state party to the Rome Statute that hosts a significant number of OSP servers to retrieve evidence, as those state parties would be more amenable to cooperation with the ICC than non-state parties. Ireland, a state party to the Rome Statute, is the EMEA headquarters for both Google and Facebook, as well as a host of other major OSPs. In 2006, Ireland passed the International Criminal Court Act (ICCA), which, in part, allowed for the issuance of warrants to access records or documents when the ICC requests state cooperation.38 It also has a law on the books, the Criminal Justice Act of 2011 (CJA), which imposes liability on officers or directors of OSPs that, out of willful neglect or a more culpable mens rea, delete or remove evidence from online access.39 On its face, Ireland seems like a reasonable venue for success in obtaining evidence stored on OSP servers.

However, the Irish legal setup leaves some gaps that make it ineffectual for use in attempting to retrieve data removed from public view. The CJA, primarily, does not list violations falling within the jurisdiction of the ICC among the crimes for which online evidence must be preserved,40 so it would be difficult to utilize coordinated threats under that Statute. Moreover, when OSPs have removed data pertaining to ongoing or potential ICC investigations, the process was either automated or driven by user complaints, not done at the direction or due to the negligence of employees at either company. Finally, while it is helpful that the ICCA calls for information sharing between Ireland and the ICC, and that channel should be explored where necessary, it only provides access to data stored on those servers located in Ireland and would not prove persuasive in an American domestic setting or any other national domestic setup. The root issue is the same as it was in the US context: access through national legal or governmental means alone is not sufficient to cover the totality of data stored on multinational OSP servers.

C. Mutual Legal Assistance Treaties

Although attempting to find a backdoor to solidify access to evidence posted online and removed from public access but stored on OSP servers through a national legal setup was unsuccessful at addressing the problem of data lost from public view in its entirety, perhaps the ICC could utilize existing bilateral MLATs where possible to access data where it was previously unable to. While the ICC cannot enter into MLATs or make original requests through an MLAT regime, as it operates on the traditional treaty basis of reciprocity, many states parties to the Rome Treaty do have existing MLATs with the United States. The ICC can ask any state party to request information on its behalf, and the Dodd Amendment would most likely legitimize any requests by the ICC pertaining to investigations of non-US citizens despite a provision of the ASPA that forbids the use of MLAT requests for ICC purposes.41 This method is particularly appealing where applicable because most major OSPs have policies that allow for freer information sharing through MLAT requests. Google, in particular, will produce the same information to a foreign agency requesting through the MLAT procedure as it would to a requesting US agency.42 Therefore, much like the diplomatic letter rogatory approach, the MLAT procedure would provide the ICC with the entirety of the data it seeks if it is successful.

Much like the purely national means of accessing data that has been removed by OSPs, the MLAT approach leaves gaps in the access available to the ICC. Significantly, these requests carry the most weight when they come from a state that referred a situation to the ICC themselves and is gathering their own evidence for a similar domestic criminal proceeding.43 This considerably narrows the scope through which MLAT requests could reasonably be expected to produce the desired evidence. In addition, many MLATs require permission from the state receiving the request before any information can be used for purposes beyond the scope of the treaty,44 so even though the Dodd Amendment opens up US MLATs for use by the ICC when investigating non-US citizens, it can likely still exercise discretion in whether or not to comply with the request. Much like the purely national means of access to data stored on OSP servers after being removed from public view, access through MLATs is effective only insomuch as a particular scenario allows it. When a state party has an MLAT with the appropriate country, self-referred, and is also looking for the same data as the ICC, and that data has been removed by an OSP and is being stored on the servers of a non-state party, the MLAT process provides a possible, albeit time intensive and discretionary, route to recovering the necessary data. Otherwise, it is likely less effective.

On the whole, the existing legal systems that govern national and cooperative bilateral legal systems do provide potential openings through which the ICC can access OSP servers in which removed pieces of evidence still reside. Given a situation where the content itself is removed from public consumption, the most effective means of accessing that stored data would be if that data happens to be stored in a country that is a state party to the Rome Statute, like Ireland. Short of that, when dealing with US firms and US servers, the most effective existing legal means to access stored data would be through diplomatic letters rogatory, which provide the most wide-ranging possible access despite the diplomatic process being lengthy and not completely reliable. If, as is the case with the Syrian conflict, Americans are involved and data is removed, there is a massive gap in the ability of the ICC to access data stored on US servers, as the Dodd Amendment only allows ICC action within the US when it is needed to investigate a foreign national. In addition, these potential solutions fail to address the fundamental issue of the problem presented: when data is removed, unless it is copied and reproduced somewhere else, the public at large is not able to identify it. Essentially, the existing legal processes through which the ICC can access data stored on OSP servers are only effective where the Court knew about the existence of the data beforehand and can pinpoint its location and the server it would have been backed up to. The question remains, therefore, as to what actions the ICC can take to prevent the loss of data as a result of blanket community guidelines established by OSPs in the first place.

IV. Alternatives to the State-Based Model: Multilateral, Bilateral, and Unilateral Options for the ICC to Protect and Maintain Evidence Posted Online

This section outlines and examines two potential solutions to the issue of losing digital evidence of war crimes to wide reaching community guidelines that are enforced by user submission or automated process. The first potential solution is the option for the ICC to seek to enter bilateral cooperation agreements with OSPs that host a significant amount of data from areas of investigation. More pertinently, however, the other solution examines projects to create a database of verifiable, protected digital evidence of human rights violations and suggests a potential ICC-run forum for online evidence collection.

A. Bilateral Cooperation Agreements with Online Service Providers

While the ASPA limits the possible ICC cooperation US governmental entities can undertake, there are no specific barriers to direct cooperative agreements between the ICC and private corporations.45 Superficially, a bilateral cooperation agreement between the ICC and the OSPs with community policies that risk the removal of pertinent digital evidence of war crimes, could potentially designate backup servers for such evidence. That would be an incredibly significant step in terms of evidence collection, and there are no legal barriers to such an agreement.

That being said, company policy is entirely a matter of discretion, and as such any agreement would be purely voluntary and unreliable.46 While such an agreement is possible, it has never occurred mostly because in would involve direct negotiation with companies that have widely variant approaches and policies, that all behave differently, and that each take a myriad of factors beyond the simple push for justice into account when making decisions. In addition, direct relationships between the Court and certain OSPs could give rise to questions of arbitrariness.47 This option also, much like direct letters rogatory to US district courts, limits access to metadata as opposed to content data because content data cannot be legally disclosed.48 Once again, metadata, while valuable in other contexts, is not helpful when the content is what needs to be recovered. Therefore, while, in theory, bilateral agreements between ICC and OSPs would be the most effective means through which the Court could guarantee access to data, they are unlikely to actually come to fruition.

B. Establishment of a Private Entity to Collect and Store Evidence Posted Online

The final solution proposed by this comment is one that is purely a preventative step, as opposed to working to preserve and/or access data once it is removed. Multiple organizations have undertaken relatively successful Internet or app-based projects which are focused on collecting and storing digital evidence of multiple crimes against humanity. The International Bar Association (IBA), for example, created the eyeWitness to Atrocities project, a mobile app purely for the purpose of collecting and securing online video evidence of crimes against humanity in response to difficulty obtaining information regarding Sri Lankan troop actions against Tamil prisoners.49 eyeWitness, while effective for its purpose, does not provide a great model for an ICC for a few reasons. First, it was only meant to record and store video footage. It is also its own separate entity incorporated in England and Wales, which would open up potential national legal issues with regard to privacy and corporate protections. eyeWitness also partnered with LexisNexis for use of its servers.50 If the ICC follows a similar path, it would certainly be helpful in terms of cost and logistics, but that could open up issues about the location of servers. Attachment to an OSP for use of its servers or the creation of a separate, private entity would necessitate some form of perpetual agreement between the ICC and that entity—whether that is possible, let alone likely, is unclear.

Perhaps a more robust model to follow stems from Benetech, a firm that created the Martus technology, which was a free, open source, public application that was built to collect, store, backup, and analyze data from activists and journalists documenting potential human rights violations. Essentially, it was a cloud-based service that allowed those documenting evidence of crimes against humanity to store and protect that evidence.51 The technology was active for upwards of fifteen years and was only deactivated due to fears that it was out of date, not for lack of use or need.52 The main obstacles to the ICC undertaking a project like this would clearly be cost and logistical—the Court would have to set aside significant amounts of its budget to undertake a project like this on its own.

The major drawback, aside from cost, to working to establish an ICC specific or a Court-connected private database of digital evidence is that, generally, social media platforms have more name brand value.53 It would take significant time and advertising to grow the general knowledge of a new ICC program. If that is the main issue, however, and the upside is a database through which the ICC can avoid dealing with online service providers entirely, then the potential benefit far outstrips the cost needed to get there.

V. Conclusion and Policy Recommendation

In sum, the best possible solution for the ICC to combat the problem of OSP community guidelines indiscriminately removing from public view caches of potentially pertinent digital evidence would be to work to establish its own private forum for the uploading, safekeeping, and analysis of evidence of war crimes posted online. The best model for this would be the Martus model, but, if the Court does not have the time, funding, or logistical knowledge to establish its own application, an expansion of the IBA model with a perpetual cooperation agreement would be the next best option. While those are explored further, maintaining dialog with OSPs regarding the effect of their policies on the Court could potentially move certain firms, especially those like Facebook which have received copious negative publicity for its handling of such evidence on its site. But changing the policies of OSPs is not a guaranteed solution. Short of all of that, the best method for reaching evidence removed from public view but stored on OSP servers would be to go through states parties to the Rome Statute, or through diplomatic letters rogatory in the case of servers located in the United States.

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

  1. 1.

    Peggy O’Donnell, Alexa Koenig, Camille Crittenden & Eric Stover, UC Berkeley HRC, Beyond Reasonable Doubt: Using Scientific Evidence to Advance Prosecutions at the International Criminal Court (Jan. 7, 2013), available online.

  2. 2.

    Róisín Á Costello, International Criminal Law and the Role of Non-State Actors in Preserving Open Source Evidence, 7 Cambridge Int’l L.J. 268 (Dec. 1, 2018), paywall, doi.

  3. 3.

    Id.

  4. 4.

    Id.

  5. 5.

    Avi Asher-Schapiro, YouTube and Facebook are Removing Evidence of Atrocities, Jeopardizing Cases Against War Criminals, The Intercept, Nov. 2, 2017, available online.

  6. 6.

    ICC Approves Probe into Myanmar’s Alleged Crimes Against Rohingya, Al Jazeera, Nov. 14, 2019, available online.

  7. 7.

    Community Standards: Violence and Criminal Behavior, Facebook, available online (last visited May 29, 2020).

  8. 8.

    Community Standards: Safety, Facebook, available online (last visited May 29, 2020).

  9. 9.

    Id.

  10. 10.

    Asher-Schapiro, supra note 5.

  11. 11.

    Id.

  12. 12.

    Id.

  13. 13.

    Id.

  14. 14.

    Id.

  15. 15.

    Julia Carrie Wong, Michael Safi & Shaikh Azizur Rahman, Facebook Bans Rohingya Group’s Posts as Minority Faces “Ethnic Cleansing,” The Guardian, Sep. 20, 2017, available online.

  16. 16.

    Violent or Graphic Content Policies, YouTube, available online (last visited May 29, 2020).

  17. 17.

    Hate Speech Policy, YouTube, available online (last visited May 29, 2020).

  18. 18.

    Asher-Schapiro, supra note 5.

  19. 19.

    Tommy Umberg & Cherrie Warden, The 2013 Salzburg Workshop on Cyber Investigations: Digital Evidence and Investigatory Protocols, 11 DEESLR 128 (Nov. 21, 2014), available online, doi.

  20. 20.

    Id.

  21. 21.

    Id.

  22. 22.

    Id.

  23. 23.

    Alexa Koenig, Keith Hiatt & Khaled Alrabe, Access Denied? The International Criminal Court, Transnational Discovery, and The American Servicemembers Protection Act, 36 Berkeley J. Int’l L. 1 (May 2018), available online.

  24. 24.

    Id.

  25. 25.

    Id.

  26. 26.

    Id.

  27. 27.

    Id.

  28. 28.

    Id.

  29. 29.

    Id.

  30. 30.

    Id.

  31. 31.

    Id.

  32. 32.

    Id.

  33. 33.

    Id.

  34. 34.

    Id.

  35. 35.

    Id.

  36. 36.

    Id.

  37. 37.

    Id.

  38. 38.

    Costello, supra note 2.

  39. 39.

    Id.

  40. 40.

    Koenig, Hiatt & Alrabe, supra note 23.

  41. 41.

    Id.

  42. 42.

    Umberg & Warden, supra note 19.

  43. 43.

    Koenig, Hiatt & Alrabe, supra note 23.

  44. 44.

    Id.

  45. 45.

    Id.

  46. 46.

    Costello, supra note 2.

  47. 47.

    Koenig, Hiatt & Alrabe, supra note 23.

  48. 48.

    Id.

  49. 49.

    Costello, supra note 2.

  50. 50.

    Id.

  51. 51.

    Overview, Martus, available online (last visited May 29, 2020).

  52. 52.

    Collin Sullivan, Martus Sunset, Benetech (May 15, 2018), available online.

  53. 53.

    Costello, supra note 2.

Admissibility of Illegally Obtained Cyber Evidence at the ICC

I. Introduction

The International Criminal Court (ICC) is already grappling with questions about what types of cyber evidence will be admissible to the Court. Even more difficult questions about the admissibility of illegally obtained cyber information are on the horizon. The law governing the admissibility of illegally obtained cyber evidence contains several ambiguities, and new types of evidence are likely to expose the ambiguities. This comment attempts to answer the question: What types of illegally obtained cyber evidence would be inadmissible to the Court?

While not conclusive, this comment asserts that a few categories of evidence are inadmissible. Untailored seizures of individual data acquired directly by the Prosecutor and cyber evidence unlawfully acquired by intermediaries at the specific instruction of the Prosecutor are likely inadmissible. Illegally obtained privileged cyber evidence and evidence obtained by means of torture are also clearly inadmissibly, but because of the underlying content of the evidence rather than the means of delivery to the Office of the Prosecutor (OTP).

This is an important and relevant issue because many ICC prosecutions have failed due to weak evidence. The OTP has relied heavily on witness testimony which is notoriously unreliable. International criminal prosecutors appear to increasingly rely on quantity of evidence rather than quality.1

Illegally obtained cyber evidence has the potential of shifting the trend toward admission of fewer exhibits which are more probative, and will be an attractive option for the OTP moving forward. This type of evidence is already making its way into courts. In December 2018, the US Department of Justice successfully convicted suspects, involved with the firm Mossack Fonseca, largely on the strength of the illegally obtained hacked evidence derived from the Panama Papers and hacked data provided to the prosecutor by a third party.2 This type of evidence will become increasingly accessible to prosecutors, including the ICC OTP.3 Clarifying the law regarding admissibility is critical.

II. Statutory Interpretation and Case Law

The threshold for excluding evidence obtained by means of illegal seizures is unclear. Some scholars have focused on the ambiguity of the terms “obtained by means of”.4 Others have focused on the causal link between prosecutors’ involvement and the violation itself.5 Still others have considered the impact of human rights law on the Court’s interest in this area.6 This comment attempts to interpret the recent jurisprudence of the ICC and ad hoc tribunals, applying it to likely factual scenarios.

A. Relevant Rome Statute Provisions

Article 69(7) of the Rome Statute is lex specialis with regard to the question of admissibility of illegally obtained evidence.7 The Lubanga Trial Chamber held explicitly that Article 69(7):

[W]as a special provision, under the lex specialis principle, prevailed over other general [69(4)] provision on the inadmissibility of evidence in proceedings and constituted an exception to the general principle of the admissibility of evidence.8

But several other provisions bear on questions of admissibility generally, and have been considered by courts in their analyses of illegally obtained evidence in particular.

1. Article 69(4) of the Rome Statute

Article 69(4) lays out the general threshold for admissibility of evidence, specifying that:

The Court may rule on the relevance or admissibility of any evidence, taking into account, inter alia, the probative value of the evidence and any prejudice that such evidence may cause to a fair trial or to a fair evaluation of the testimony of a witness, in accordance with the Rules of Procedure and Evidence.

The Lubanga Trial Chamber established the three-fold test based on Article 69(4) which asks judges to determine admissibility of evidence on the basis of the exhibit’s 1) relevance, 2) probative value, and 3) whether the probative value outweighs its prejudicial effect.9 The consideration of probative value is notably absent from the admissibility analysis for illegally obtained evidence under Article 69(7) however.

The explicit mention that admissibility be decided in accordance with the Rules and Procedures and Evidence (RPE) in Article 69(4) is also worth noting. While Article 69(4) refers to the RPE explicitly, Article 69(7) only refers to the Rome Statute itself. One view is that since the RPE isn’t mentioned in Article 69(7), it shouldn’t be considered in interpreting the article. Another view is that inclusion of RPE in Article 69 broadly suggests it should bear on the admissibility analysis in each of the subsections, Article 69(4) and Article 69(7). The drafting history does not appear to provide clarity on the reason for the mention of the RPE in Article 69(7), but because the RPE does not include rules specific to the admissibility of illegally obtained evidence, its inclusion might have been deemed unnecessary to the drafters at the time.

The more general analysis of Article 69(4) still applies to all admissibility decisions, so if the RPE is amended to provide a factor bearing on the question of illegally obtained evidence, it should be applied by judges in admissibility decisions.

2. Article 55 of the Rome Statute

Under Article 55, violation of rights that would require exclusion of evidence include the right not to be subject to torture or other cruel punishment, the right to be free from inhuman or degrading treatment, the right to counsel, and the right not to self incriminate.10 None of these rights relate to the question of mandatory or discretionary exclusion of cyber evidence because of the illegal means by which it was gathered. No court has considered these rights in relation to such evidence.

3. Article 21(3) of the Rome Statute

Article 21(3) is a broad catch-all provision that requires application and interpretation of the Statute to be “consistent with internationally recognized human rights.”11 The term “internationally recognized human rights” actually first appeared in the drafting history in Article 69(7) and was later added to Article 21(3).12 ICC judges must interpret and apply Article 21(3) in rendering admissibility decisions as well. This mandatory consideration could be viewed as conflicting with Article 69(7), which permits judges to admit evidence which violates internationally recognized human rights so long as admission of the evidence not violate subsection (a) reliability or (b) integrity of the proceedings.

Courts have reconciled this inconsistency in two ways. One interpretation is that under ICC jurisprudence, admission of evidence in some cases might not be considered to be a violation of internationally recognized human rights where the violation is proportional to the objective sought. Investigations of serious crimes under the Rome Statute can render what would otherwise be a violation of human rights by a human rights court consistent with human rights where the conduct is proportional to the interest of investigating serious crimes.

Another interpretation, adopted by the Lubanga Trial Chamber, implicitly acknowledges the inconsistency between Article 69(7) and Article 21(3), while asserting Article 69(7) is lex specialis with regard to admissibility so is therefore the controlling provision.13 Despite the argument by some scholars,14 no decision has explicitly considered Article 21(3) in its admissibility analysis so it will not be considered in this analysis.

B. Narrowing the Scope

Breaks in the chain of custody do not factor into determination of admissibility at the confirmation stage, according the Lubanga Trial Chamber.15 Chain of custody issues also do not factor into the analysis of violations to the right to privacy as an internationally recognized human right.16 Issues of authenticity are also separate from this analysis. Authenticity is determined by a separate standard, applying a different burden of proof. The International Criminal Tribunal for the former Yugoslavia (ICTY) emphasized multiple times that the question of admissibility of illegal obtained evidence is assessed separately from general authentication analysis.17

Commentary on the issue also clarifies two relevant points that help narrow the scope of relevant factors. First, the admissibility analysis is the same regardless of which side enters the evidence and whether it be in favor or against the accused. Piragoff makes this clear in his commentary.18 Ambos argues the same.19

Second, whether or not the search or seizure was in violation of national law is not a relevant consideration for admissibility at the ICC, according to Article 69(8).20 In Lubanga, the Pre-Trial Chamber rejected the defense request to exclude evidence acquired by authorities in violation of Congolese law.21

C. Article 69(7) of the Rome Statute

Article 69(7) of the Rome Statute provides:

Evidence obtained by means of a violation of this Statute or internationally recognized human rights shall not be admissible if:

  1. The violation casts substantial doubt on the reliability of the evidence; or

  2. The admission of the evidence would be antithetical to and would seriously damage the integrity of the proceedings.

The case law of the Article 69(7) suggests four relevant and separate analyses can be extracted from this provision.

1. “In Violation of this Statute

Evidence obtained in violation of the Rome Statute would trigger analysis of whether Article 69(7) subsections (a) or (b) were satisfied. The relevant protections provided for in the Statute include rights of the accused22 and rights of witnesses and victims.23 However, no court has held that any of these protections were implicated in the type of unlawful acquisition of evidence that is relevant for the cyber evidence discussed in this comment.

Notably, explicit protections against improper search and seizure is absent from the Statute, as is the right to privacy. Provisions providing for these protections were removed from preliminary drafts, well before the final negotiated drafts.24

Scholars disagree about whether “violations of this statute” was meant to include the Rules of Procedure and Evidence. Piragoff argues it was;25 Safferling argues it was not.26 Regardless, it is not clear that the issue is relevant to the analysis here, because the RPE do not provide for protections relevant to cyber evidence discussed in this comment. The analysis of the courts has focused on violations of internationally recognized human rights. The analysis of this comment does the same.

2. Internationally Recognized Human Rights

To determine what qualifies as an “internationally recognized human right,” the ICC has looked broadly at International Human Rights Law (IHRL) jurisprudence. The primary source cited thus far has been the European Court of Human Rights (ECHR). The Lubanga Pre-Trial Chamber held, that “only serious violations of human rights may lead to exclusion of evidence.”27

For this analysis, the relevant internationally recognized human right is the right to privacy. Privacy is the most common human right asserted in admissibility cases at the ICC and other ad hoc tribunals. It also appears to be the only relevant human right related specifically to admissibility of cyber evidence.

Privacy was held to be an internationally recognized human right early on in the ICC jurisprudence in the Lubanga Confirmation Decision. This followed consistent rulings at the ICTY.

The right to privacy is codified in multiple human rights instruments, including Article 17 of the International Covenant on Civil and Political Rights, Article 8 of the European Convention on Human Rights (European Convention), and Article 11 of the American Convention on Human Rights. Each of these instruments provide an exception—for “lawful” exceptions to the right to privacy.28

The Lubanga Pre-Trial Chamber did find that evidence submitted for admission by the OTP was acquired in violation of the accused’s internationally recognized human rights.29 The Court identified two factors as weighing toward a conclusion that there was a violation of internationally recognized human rights—first, that the interception was not proportional to the objective sought; and second, that the “indiscriminate nature of the search and seizure involving hundreds of items.”30 Specifically, the Court found a violation where Congolese authorities provided the prosecutor with “hundreds of [confiscated] documents […] including correspondence, photographs, invitations, legislation, reports, diaries, and other ‘personal information.’”31 The Trial Chamber agreed, emphasizing that “hundreds of documents were seized that were unrelated to the purpose of the search.”32

The ECHR held in Klass v. Germany, that surveillance was compatible with Article 8(II) of the European Convention.33 National security was the justification for the surveillance in this case, however.34 It is unclear what other justification would convey such a broad privacy exception. In A v. France, prevention of crime was recognized by the ECHR as proper justification for the tapping of phones.35 The ECHR found no violation of Article 8 privacy protections—the critical distinction being that phone tapping was legal under French law.36

The lawfulness of the privacy intrusions proves to the critical factor in ECHR holdings of what qualifies as an Article 8 human rights violation of privacy. In France, the surveillance was lawful so there was no human rights violation,37 whereas the same conduct in the UK, where the same surveillance was not legal, amounted to a violation of Article 8.38

The ICC does not apply the same analysis as human rights courts to determine whether a human rights violation occurred. Lawfulness of surveillance under domestic law is not relevant to ICC analysis, as the Court is generally not permitted to consider domestic law in admissibility analysis.39

The ECHR also had the opportunity to address the implications of a third party providing evidence to the Prosecutor. In Schenk v. Switzerland, a private individual illegally recorded the accused and provided the recording to national authorities.40 The ECHR admitted the evidence without articulating any adverse impact from the fact that the recording was provided by a third party.41

Finally, one question that may arise related to this fact is: does it matter whose human rights are violated in acquiring the evidence? The Court has answered this definitively—it does not. The Lubanga Trial Chamber held that:

[T]he fact that a violation involves the human rights of a third person, other than the defendant, is not relevant when deciding whether the first step under Article 69(7) is satisfied.42

3. Article 69(7)(a) Reliability

Most of the admissibility challenges at the ICC have related to intercepted cell phone conversations. In each of these cases, the Court held that the illegal nature of its acquisition did not cast substantial doubt on the reliability of the evidence. This is true of the jurisprudence at the ICC and the ad hoc tribunals.

In Lubanga, the illegal search and seizure of the accused did not substantially undermine the evidence’s reliability according to the Pre-Trial Chamber.43 Similarly, the Mbarushimana Court held that the illegal nature of intercepted communication did not “have any impact on the reliability of the evidence thereby obtained.”44

The jurisprudence strongly suggests that the nature of violation of human rights does not meaningfully impact the Article 69(7) analysis. The nature of human rights violations could affect the reliability of evidence, but are more likely to be assessed by the judges under authentication analysis or as a question of the evidence’s “weight.” It is not relevant for Article 69(7) admissibility analysis.

4. Article 69(7)(b) Integrity

The construction of Article 69(7)(b) sets the bar for exclusion of evidence relatively high. Evidence satisfying the first of Article 69(7) in violation of human rights would only be inadmissible under Article 69(7)(b) if its admission “would be antithetical to and would seriously damage the integrity of the proceedings.” This sets out a conjunctive requirement—must be “antithetical to” and “would seriously damage” the integrity of the proceedings. A human rights violation that is antithetical to but would not seriously damage the proceeding’s integrity would still be admissible.

This raises the question: what would seriously damage the integrity of the proceedings? Klamberg argues that the fundamental values of the Rome Statute are relevant to the “integrity of the proceedings.”45 The Lubanga Pre-Trial Chamber invoked these values when it cited “the need to respond to victims’ and the international community’s expectations.”46

The ICC considered the jurisprudence of the ad hoc tribunals in analyzing this question and ruled consistently with it. The Lubanga Pre-Trial Chamber followed the approach of the ICTY in Brđanin, concluding that the privacy violation did not adversely impact the proceedings.47

Similarly, the Karadžić Trial Chamber held that illegally obtained phone intercepts were in violation of the human right to privacy, but that their admission did not conflict with the integrity of the trial.48 The Court held that the “fundamental right to privacy is not absolute and may be derogated from in times of emergency.”49 It is unclear what the Court meant by “emergency.”

The ICTY had the opportunity to rule on some contextual factors that may affect the impact on the integrity of the proceedings. Whether the intercepted communications (that were concededly obtained in violation of the accused’s privacy) were obtained in the context of an ongoing armed conflict or in peacetime did not affect the analysis. In Stanišić and Župlijanin, the Court held that the admission of intercepts illegally obtained during wartime did not adversely affect the integrity of the proceedings.50

This was consistent with an earlier holding in Kordić, where the Court held that admission of illegally intercepted phone calls by a soldier during armed conflict “clearly […] did not damage the integrity of the proceedings.”51 In Haraqija, the defendant distinguished their facts on the basis that illegally intercepted phone calls and text messages were obtained during peacetime, well after the armed conflict ended. The Trial Chamber dismissed the significance of the distinction, noting it as an issue of domestic law which the Court would not consider.

In Haraqija, the Appeals Chamber cited the national authority’s “good faith” effort to protect witnesses even while it illegally intercepted accused’s text messages and phone calls.52 In that case, national authorities provided recording equipment to witnesses in a “good faith” interest of protecting them in an “atmosphere where witnesses felt unsafe.”53

Moving forward, the OTP (or defense counsel) might argue by analogy that obtaining cyber evidence illegally is a safer alternative than sending investigators into areas of armed conflict or other dangerous situations. It would be an overstatement to claim that these factors would cause an ICC judge to find an Article 69(7) violation. These are factors which previous courts have noted the absence of which contributes to a negative finding. The logical assumption is therefore that they matter. In the aggregate, and considering likely admissibility scenarios moving forward, this comment attempts to infer likely judicial outcomes.

In summary, Article 69(7) provides for a two-step proportionality test: First, a search or seizure may violate privacy but still be legal as a lawful interference if it is proportional. Second, even if the search or seizure would be a violation of international human rights jurisprudence as analyzed under IHRL, it may still be admissible so long as:

  1. the violation cast substantial doubt on its reliability, or

  2. admission of the evidence not be antithetical to or cast substantial doubt on the integrity of the proceedings.

III. Three Categories of Admissibility

A. Category 1—Clearly Admissible Evidence

Seizures in violation of domestic or international law, but which were conducted pursuant to a warrant, will be admissible. The jurisprudence at the ICC and the ad hoc tribunals strongly suggests that where any court issues a search warrant, the integrity of the proceedings would not be seriously damaged even if privacy violations occur in the execution of the warrant.

The ICTY addressed this issue specifically in Delalić, where Austrian authorities obtained a search warrant for the accused. Multiple irregularities and violations occurred during the seizure, causing a violation of his right to privacy. Still, the Court admitted the evidence, holding that:

[I]t would constitute a dangerous obstacle to the administration of justice if evidence which is of relevant and probative value could not be admitted merely because of a minor breach of procedural rules.54

The ICC would similarly admit cyber evidence obtained under similar circumstances.

B. Category 2—Clearly Inadmissible Evidence

Based on analysis of the jurisprudence, two types of cyber evidence are clearly inadmissible under Article 69(7). First, privileged information is inadmissible. Article 69(7) analysis is triggered by violations of the Rome Statute and Article 69(5) of the Rome Statute calls on the Court to observe privileged communication. Consequently, the OTP (and defense counsel) should be careful to mark privileged documents in their eDiscovery searches.

Second, cyber evidence obtained by means of torture is clearly inadmissible. The ICTY held in Martić that “statements which are not voluntary, but rather are obtained by means including oppressive conduct cannot be admitted.”55 Evidence obtained by torture is the classically cited example of the type of evidence which would seriously damage the integrity of the trial.56 Ambos argues that evidence obtained by torture is absolutely inadmissible, whether used in favor of or against the defendant, and whether introduced by the prosecutor or defense.57

The clear inadmissibility of evidence obtained by torture should signal to the parties that there may be a human rights violation prior to the unlawful interception of the information. In that case, admission of that piece of evidence would be analyzed for each alleged violation of internationally recognized human rights or violation of the Rome Statute under 69(7).

C. Category 3—Gray Area

The limited jurisprudence and vague text of the Rome Statute leave substantial ambiguity about hypothetical admissibility rulings. Each of the following proposed types of illegally obtained cyber evidence could be viewed as admissible or inadmissible under Article 69(7). Considering the factors tribunals have applied in the past, this comment considers the most likely admissibility decisions.

This comment asserts that the role played by the OTP in acquiring the evidence—whether active acquisition or passive—is instructive in predicting admissibility decisions. While not a dispositive factor in its analysis justifying admission, the Court in Lubanga focused on the limited role of the OTP investigator in the illegal seizure conducted by Congolese authorities.58 In dismissing defense arguments that illegally obtained evidence damaged the integrity of the proceedings at the ad hoc tribunals, courts have consistently implied that violations of human rights perpetrated by national authorities did not impugn the proceedings to the same degree that direct violations by the Prosecutor might.59

1. Third-Party Acquired Evidence to a Passive OTP

Three categories of evidence are likely to be introduced, which are characterized by the passive role of the OTP.

i. Seizures Provided to the OTP by a Government Agency

Illegally obtained data provided to a passive OTP is likely admissible. ICC jurisprudence suggests that the “integrity of the proceedings” analysis should weigh the competing goals of the Court in this analysis, including ending impunity, fairness to the accused, and deterrence of future crimes.60 If states were to start cooperating with Court by providing evidence, this would represent a major boost to the cooperation and complementarity objectives of the Court—an area where the Court has acknowledged a problem. Any damage to the integrity of proceedings weighed against this benefit would not require exclusion of the evidence.

ii. Seizures Provided by Victims and Defectors

Evidence illegally obtained by victims and defectors are likely admissible, because there is a lower risk of pervasive human rights abuses by way of their seizure. The fact that these sources are individuals, without the resources of a state, suggests the degree of human rights violation would be limited, therefore the likely damage to the integrity of the proceedings is also limited. By considering evidence from these sources, the Court advances its anti-impunity goal. In these limited cases, the advancement of this goal outweighs the potential damage to the integrity of the proceedings.

An example of this information is the 55,000 photos and documents recorded by Syrian authorities smuggled out of the country by the military police defector known by the pseudonym “Caesar”.61 This type of evidence, if the Court were able to exercise jurisdiction over crimes perpetrated in Syria, would greatly bolster the goals of the Court while only minimally damaging the integrity of the proceedings.

iii. Seizures of Stolen (i.e. Hacked) Personal Data by Intermediaries Acting as OTP Agents

Seizures by intermediaries acting as agents are likely to be admissible. The factual inquiry of determining the existence of an agency relationship between the OTP and intermediaries is likely to complicate admissibility analysis. It may require an evaluation of the Prosecutor’s mens rea as to the nature of the human rights violation caused by the solicitation of the evidence.

In a U.S. Court of Appeals case in the 4th Circuit involving illegally obtained evidence provided to prosecutors by a Turkish hacker, the Court asked if the hacker acted as an “agent” of the Prosecution when he hacked the data.62 Finding that the agency test was satisfied, because the “government knew of and acquiesced in the searches conducted by [the Turkish hacker] and that [his] actions were motivated solely by an interest to further law enforcement efforts,” the Court excluded the evidence as a 4th Amendment violation.63 At the ICC, the OTP will have broader powers to employ intermediaries as agents.64 Violations of national law are not considered in the analysis, and the OTP is, by necessity, more reliant on intermediaries for evidence than American prosecutors, considering the disparity in prosecutorial and enforcement powers.

Intent vis-à-vis the human rights violation perpetrated by the agent might likely lead to inadmissibility, but negligence would not. Recklessness could go either way. More likely, the mens rea of the OTP with regard to a violation committed by the agent intermediary would not be the dispositive inquiry. This is a procedural question of admissibility rather than a question of criminal liability. The judge will likely evaluate the damage to the integrity of proceedings holistically, balancing the OTP’s reasonable reliance on intermediaries against the damage to the Court’s integrity by a transparent attempt to deflect blame onto intermediaries.

iv. ICC Direct Solicitation for Specific Information

Direct solicitation of illegally obtained evidence impugns the integrity of the Court more than passive receipt of evidence, and is more likely to lead to inadmissibility.

a. Where Instruction Causes a Source to Steal Information

Similar to “agent” analysis, but assuming a greater control of the intermediary by the OTP, evidence stolen at the specific instruction of Prosecutor is likely to be inadmissible. While speculative, this outcome is consistent with the Trial Chamber’s concern in Lubanga that the OTP investigator not have been directly responsible for the privacy violation perpetrated by the national authorities.65 Specific instructions by OTP prosecutors and investigators that a source illegally obtain cyber evidence would make the OTP directly responsible for the violation. This responsibility weighs more heavily toward damage to the integrity of the proceedings, requiring exclusion of the evidence.

b. Untailored Seizures of Individual Data Acquired Directly by the Prosecutor

Wide-reaching and untailored illegal seizures of private data submitted for evidence are likely to be inadmissible. ECHR jurisprudence viewed untailored seizures as violations of the accused’s internationally recognized human rights because these seizures failed the proportionality requirement.66 These substantial violations of human rights would have a more damaging effect on the integrity of the proceedings. The OTP should be cautious to avoid submission of evidence acquired on the basis of such substantial privacy violations.

c. Where Information is Stolen Prior to the OTP Request

When the OTP actively solicits specific stolen information and where that information had already been stolen by intermediaries prior to the Prosecutor’s request, the evidence will likely be admitted. The fact that violation occurred prior to the OTP’s involvement and not at its request, reduces the likelihood of damage to the integrity of the proceedings. This hypothetical is more analogous to the OTP investigator’s mere presence at the illegal seizure in Lubanga and the passive receipt of illegally seized evidence by the Austrian authorities. In both cases, the evidence was admitted. The same outcome is likely here.

X. Conclusion

In summary, the issue of admissibility of illegally obtained cyber evidence is a timely one, and requires further clarification. Article 69(7) jurisprudence suggests that untailored seizures of individual data acquired directly by the Prosecutor, and cyber evidence unlawfully acquired by intermediaries at the specific instruction by the Prosecutor are likely inadmissible. The Court should consider adopting more specific guidelines through the Rules and Procedures of Evidence to clarify this area of law.

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

  1. 1.

    Christopher Gosnell, The Changing Context of Evidential Rules, in Principles of Evidence in International Criminal Justice 221 (Karim A. A. Khan, Caroline Buisman & Christopher Gosnell, eds., Jan. 15, 2010), paywall

    (noting that the ICTY admitted 386 exhibits in the Tadić case and 375 exhibits in Delalić, while admitting many more in the later cases—3,086 exhibits in Brđanin [2002–2004] and 3,800 exhibits in Krasijsnik [2004–2006]).

  2. 2.

    Victor L. Hou, Lisa Vicens, Rahul Mukhi, Martha E. Vega-Gonzalez & Christine Jordan, U.S. Criminal Prosecution Based on Panama Papers Hack Raises Novel Legal Issues, Cleary Cyberwatch (Jan. 17, 2019), available online.

  3. 3.

    See e.g. Mirja Gutheil, Quentin Liger, Aurélie Heetman, James Eager & Max Crawford, Directorate-General for Internal Policies, European Parliament, Legal Frameworks for Hacking by Law Enforcement: Identification, Evaluation and Comparison of Practices 8 (Mar. 2017), available online.

  4. 4.

    Amal Alamuddin, Collection of Evidence, in Principles of Evidence in International Criminal Justice 232, 241 (Karim A. A. Khan, Caroline Buisman & Christopher Gosnell, eds., Jan. 15, 2010), paywall.

  5. 5.

    Andrea Ryan, Toward a System of European Criminal Justice: The Problem of Admissibility of Evidence 27 (Apr. 24, 2016), paywall.

  6. 6.

    Wojchiech Jasiński, Admissibility of Illegally Obtained Evidence in Proceedings before International Criminal Courts, in Prosecuting International Crimes: A Multidisciplinary Approach 201 (Bartłomiej Krzan, ed., Jul. 11, 2016), paywall, doi.

  7. 7.

    Kai Ambos, The Transnational Use of Torture Evidence, 42 Israel L. Rev. 362, 375 (2009), available online.

  8. 8.

    The Prosecutor v. Thomas Lubanga Dyilo, ICC-01/04-01/06, Decision on the admission of material from the “bar table,” ¶ 34 (TC I, Jun. 24, 2009) [hereinafter Lubanga Bar Table Decision], available online.

  9. 9.

    The Prosecutor v. Thomas Lubanga Dyilo, ICC-01/04-01/06, Corrigendum to Redacted Decision on the defence request for the admission of 422 documents, ¶ 94 (TC I, Mar. 8, 2011), available online.

  10. 10.

    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, as amended [hereinafter Rome Statute], Art. 55, available online.

  11. 11.

    Id. Art. 21(3).

  12. 12.

    Donald K. Piragoff & Paula Clarke, Article 69 Evidence, in The Rome Statute of the International Criminal Court: A Commentary 1712, ¶ 65 (Otto Triffterer & Kai Ambos, eds., 3d ed. Jan. 14, 2016), available online, paywall.

  13. 13.

    Lubanga Bar Table Decision, supra note 8, ¶ 34.

  14. 14.

    See e.g. Alexander Zahar & Göran Sluiter, International Criminal Law: A Critical Restatement 367 (Dec. 12, 2007), paywall

    (arguing for Article 21(3) balancing).

  15. 15.

    Lubanga Bar Table Decision, supra note 8, ¶ 43.

  16. 16.

    Alamuddin, supra note 4, at 297.

  17. 17.

    Prosecutor v. Slobodan Milošević, IT-02-54-T, Preliminary Decision on the Admissibility of Intercepted Communications (ICTY TC III, Dec. 16, 2003), available online; Prosecutor v. Radoslav Brđanin, IT-99-36-T, Decision on the Defence “Objection to Intercept Evidence,” ¶ 66 (ICTY TC II, Oct. 3, 2003), available online.

  18. 18.

    Piragoff & Clarke, supra note 12.

  19. 19.

    Ambos, supra note 7, at 375.

  20. 20.

    Rome Statute, supra note 10, at Art. 69(8).

  21. 21.

    The Prosecutor v. Thomas Lubanga Dyilo, ICC-01/04-01/06, Decision on the confirmation of charges, ¶¶ 62–63 (PTC I, Jan. 29, 2007) [hereinafter Lubanga Confirmation Decision], available online.

  22. 22.

    Rome Statute, supra note 10, at Arts. 55, 67.

  23. 23.

    Id. Art. 68.

  24. 24.

    George E. Edwards, International Human Rights Law Challenges to the New International Criminal Court: The Search and Seizure Right to Privacy, 26 Yale J. Int’l L. 323, 350 (2001), available online.

  25. 25.

    Piragoff & Clarke, supra note 12, ¶ 63.

  26. 26.

    Christoph J. M. Safferling, Towards an International Criminal Procedure (May 8, 2003), paywall.

  27. 27.

    Lubanga Confirmation Decision, supra note 21, ¶ 86.

  28. 28.

    Mark Klamberg, Evidence in International Criminal Trials 303 (Feb. 21, 2013), paywall.

  29. 29.

    Lubanga Confirmation Decision, supra note 21, ¶ 82.

  30. 30.

    Id. ¶ 81.

  31. 31.

    Id. ¶ 80.

  32. 32.

    Lubanga Bar Table Decision, supra note 8, ¶ 38.

  33. 33.

    Klass and Others v. Germany, Judgment, ECHR 5029/71, ¶ 50 (Plenary, Sep. 6, 1978), available online.

  34. 34.

    Id.

  35. 35.

    A. v. France, Judgment, ECHR 14838/89 (Chamber, Nov. 23, 1993), available online.

  36. 36.

    Id.

  37. 37.

    Id.

  38. 38.

    Malone v. The United Kingdom, Judgment, ECHR 8691/79, ¶ 72 (Plenary, Aug. 2, 1984), available online.

  39. 39.

    Rome Statute, supra note 10, at Art. 69(8).

    (“When deciding on the relevance or admissibility of evidence collected by a State, the Court shall not rule on the application of the State’s national law.”).

  40. 40.

    Schenk v. Switzerland, Judgment, ECHR 10862/84, ¶¶ 41–46 (Plenary, Jul. 12, 1988), available online.

  41. 41.

    Id.

  42. 42.

    Lubanga Bar Table Decision, supra note 8, ¶ 37.

  43. 43.

    Lubanga Confirmation Decision, supra note 21, ¶ 85.

  44. 44.

    The Prosecutor v. Callixte Mbarushimana, ICC-01/04-01/10, Decision on the confirmation of charges, ¶ 71 (PTC I, Dec. 16, 2011), available online.

  45. 45.

    Klamberg, supra note 28, at 405.

  46. 46.

    Lubanga Confirmation Decision, supra note 21, ¶ 86.

  47. 47.

    Id. ¶ 25–46.

  48. 48.

    Prosecutor v. Radovan Karadžić, IT-95-5/18-T, Decision on the Accused’s Motion to Exclude Intercepted Conversations, ¶ 8 (ICTY TC III, Sep. 30, 2010), available online.

  49. 49.

    Id.

  50. 50.

    Prosecutor v. Mićo Stanišić and Stojan Župljanin, IT-08-91-T, Decision Denying the Stanišić Motion for Exclusion of Recorded Intercepts (TC II, Dec. 16, 2009), available online.

  51. 51.

    William A. Schabas, The UN International Criminal Tribunals: The Former Yugoslavia, Rwanda and Sierra Leone 460 (2006), paywall, doi; citing Prosecutor v. Dario Kordić and Mario Čerkez, IT-95-14/2-T, Oral Decision of Judge May, ¶¶ 13693–94 (ICTY TC III, Feb. 2, 2000), available online.

  52. 52.

    Prosecutor v. Astrit Haraqija and Bajrush Morina, IT-04-88-R77.4-A, Judgement, ¶ 24 (ICTY AC, Jul. 23, 2009) [hereinafter Haraqija Judgment], available online.

  53. 53.

    Id. ¶ 28.

  54. 54.

    Prosecutor v. Zejnil Delalić, Zdravko Mucić, Hazim Delić & Esad Landžo, IT-96-21, Decision on the Tendering of Prosecution Exhibits 104–108, ¶ 20 (ICTY TC II, Feb. 9, 1998), available online.

  55. 55.

    Prosecutor v. Milan Martić, IT-95-11-T, Decision Adopting Guidelines on the Standards Governing the Admission of Evidence, ¶ 9 (ICTY TC I, Jan. 19, 2006), available online.

  56. 56.

    Alamuddin, supra note 4, at 238.

  57. 57.

    Ambos, supra note 7, at 375.

  58. 58.

    Lubanga Confirmation Decision, supra note 21, ¶¶ 77–78.

  59. 59.

    Id.; Haraqija Judgment, supra note 52, ¶ 24.

  60. 60.

    See Klamberg, supra note 28.

  61. 61.

    Anne Barnard, Inside Syria’s Secret Torture Prisons: How Bashar al-Assad Crushed Dissent, N.Y. Times, May 11, 2019, available online.

  62. 62.

    United States v. William Adderson Jarrett, 229 F. Supp. 2d 503 (E.D. Va., Nov. 1, 2002), available online.

  63. 63.

    Id. at 519–20.

  64. 64.

    International Criminal Court, Guidelines Governing the Relations between the Court and Intermediaries 2 (Mar. 2014), available online.

  65. 65.

    Lubanga Bar Table Decision, supra note 8, ¶ 46.

  66. 66.

    Malone v. The United Kingdom, supra note 38, ¶ 72.