Symposium on Open Source Investigation Labs: Balancing Digital Truths and Human Costs: Witnesses and Open-Source Evidence in Atrocity Trials

[Rossella Pulvirenti is a Senior Lecturer in Manchester Law School (UK), specialising in international criminal law and human rights, with a specific focus on evidence and witnesses’ rights]

The evidentiary landscape of international criminal justice is undergoing a profound transformation through the use of open-source intelligence (OSINT). OSINT marks the third major revolution in evidentiary approaches to prosecuting mass atrocity crimes. While it offers unprecedented opportunities for documenting and verifying international crimes, its use generates complex questions, particularly regarding the role and protection of witnesses.

In this contribution, the term ‘witness’ will be used as a synonym of witnesses of facts, who can give evidence of factual matters within their knowledge. They are not permitted to express their opinion during their testimony, as doing so would classify them as expert witnesses.

Despite OSINT’s growing relevance, there is little guidance on the intersection between OSINT and witness protection. This contribution briefly explores this interaction, which may have adverse consequences for witnesses, and argues for reframing them as rights-holders within international criminal proceedings.

Three Phases of Evidentiary Evolution in International Criminal Justice

The history of international criminal law reveals a clear evolution in evidentiary strategies. The use of documentary evidence characterised the first phase in the aftermath of World War II. For instance, at the Nuremberg trials, the prosecution led by Justice Robert Jackson, chose to rely on documentary evidence to prove war crimes, crimes against humanity and crimes against peace. This prosecutorial strategy was not dictated by the lack of witnesses willing to come forward, but rather a deliberate choice to prove “incredible events by credible evidence”.

The second phase of evidentiary evolution emerged in the 1990s with the establishment of international tribunals, such as the International Criminal Tribunal for the Former Yugoslavia (ICTY) and International Criminal Tribunal for Rwanda, and hybrid tribunals, like the Special Court for Sierra Leone. This marked the shift towards a witness-centric prosecution strategy because these tribunals recognised the value of personal testimony in proving atrocity crimes.

The International Criminal Court (ICC) continued this trend. However, relying on witness testimony in mass atrocity trials presents significant challenges. These witnesses often face threats to their safety and that of their families. If witnesses are threatened, scared and fear for their life, they might refuse to testify or fabricate false testimony. Several ICC cases illustrate this point. For instance, in the Bemba and Others case, Bemba and his accomplices were convicted for crimes against the administration of justice under Article 70 of the Rome Statute after bribing witnesses. Likewise, in the Kenyan cases, several witnesses withdrew their testimony, leading Trial Chamber IV to acquit Ruto, and the prosecution to drop its case against Kenyatta due to the lack of sufficient evidence. Another problem, as highlighted by the Independent Expert Review, is that the ICC Office of the Prosecutor depends heavily on cooperation from States Parties and national governmental authorities. These entities act as the gatekeepers because they control access to crime scenes, witnesses and evidence. A lack of cooperation from local authorities can severely hinder, or even halt, criminal prosecutions. In addition to security issues, witnesses might also experience psychological challenges. After giving evidence, witnesses may feel exploited or abandoned by the justice system. The case of Ogwen’s wives who testified for the ICC prosecutor in the case against their husband illustrates how post-testimony neglect can lead to feelings of exploitation and social rejection.

Witnesses in the Age of OSINT

ICL is currently experiencing an exponential increase in the availability of OSINT, both at the international and national levels. However, this shift is not yet fully reflected in the ICC’s practice, where traditional forms of evidence still prevail. Koenig suggests that OSINT has not been widely adopted because the ICC is investigating older crimes, but this trend is likely to evolve as more recent conflicts come under scrutiny.

Despite the growing reliance on digital evidence, witnesses remain central to atrocity trials. Recent research on the Hassan case shows that OSINT is frequently used to corroborate other types of evidence, including witness testimony. However, the relationship between OSINT and witness evidence is not always unidirectional. For instance, in the ICC case against Mahmoud al-Werfalli, witness testimony was employed to corroborate OSINT, which had already provided “sufficient indicia of authenticity” to justify a second arrest warrant. While it remains unclear how parties decide between relying on testimony or OSINT, the integration of OSINT undeniably increases investigative efficiency and can even enhance the quality and reliability of witness testimony.

OSINT is less vulnerable to the issues which affect witnesses, particularly in international criminal tribunals where investigations occur years or even decades after the crimes occurred. Witnesses testifying at the ICTY and at the Extraordinary Chambers in the Courts of Cambodia, witnesses described events that had occurred nearly twenty years before their testimony. Also, research shows a clear link between trauma and the accuracy and quantity of information given during witness testimony.

Second, within the national context, investigators typically access the crime scenes promptly and, where possible, collect witness statements. However, if any international criminal tribunals investigate ongoing or recent conflicts, access could be restricted due to serious safety concerns. In such cases, OSINT offers a valid alternative because it would not be necessary to physically enter the territory of a country to gather evidence. The situation in Syria is a clear example of this. Although Syria closed its borders to international investigations since 2011, the Independent International Commission of Inquiry on the Syrian Arab Republic has received thousands of videos showing alleged human rights abuses committed in Syria. In such scenarios, states no longer operate as gatekeepers to evidence.

Impact of Digital Reconstructions on Witnesses

While OSINT has become a powerful tool in international criminal investigations, its use raises significant concerns regarding the safety, dignity, and rights of witnesses. Their relationship is characterised by a complex interplay between digital evidence and human vulnerability.

OSINT may inadvertently reveal identifying information about witnesses, negatively impacting witnesses’ right to privacy. Individuals may become identifiable within large data collections. The lack of confidentiality could put their lives in danger. Moreover, OSINT can be used to generate lists of potential witnesses and survivors to be interviewed as potential witnesses.

To address these risks, the collection of OSINT must respect the principle of informed consent. Witnesses should be clearly informed about how OSINT may be used in relation to their testimony and allowed to consent or object to its use. Consent must be voluntary and not coerced. Under international and regional standards on the right to privacy (such as Article 17 of International Covenant on Civil and Political Rights, Article 8 of the European Convention on Human Rights, Article 11 of the American Convention on Human Rights), restriction to the individual’s right to privacy may be permissible only if they are proportionate, necessary in a democratic society and is aimed at the protection of certain interests, such as the interest of national security. However, it is not clear if OSINT fulfils all these criteria.

Additionally, it is essential to consider not only how witnesses influence the reconstruction process, but also how the process affects them. Digital reconstructions often require survivors to relive traumatic experiences in detail. This, together with the repeated exposure to graphic or violent content during investigations or trials, can have lasting psychological consequences. As international criminal law increasingly embraces digital methodologies, it must also uphold witnesses’ dignity and well-being.

Ultimately, there is a real risk that OSINT could overshadow the human element of international criminal trials. This is generally embodied by witnesses, who offer factual recounts enriched with emotions and nuance that digital evidence alone cannot capture.

OSINT and the Normative Gap

The integration of OSINT into international criminal proceedings has outpaced the development of an appropriate normative framework, which addresses the role and the protection of witnesses.

The existing legal and procedural framework on OSINT, including the Berkeley Protocol on Digital Open-Source Investigations (Berkeley Protocol), offers limited guidance on how OSINT intersects with witness protection. Generally, the Berkeley Protocol is geared toward the investigation and prosecution phases of international crimes. However, witnesses often engage with the judicial system beyond these stages, and it remains unclear whether the principles outlined in the Protocol extend to those later interactions. Notably, the Berkeley Protocol explicitly acknowledges that individuals captured in digital content, such as witnesses, have a right to be protected from any type of harm, including digital, financial, legal, physical, psychological and reputational (pg. 34, para 72). While this recognition is significant, other sections of the Berkeley Protocol affirm the importance of fundamental principles. For instance, section E of the Berkeley Protocol is dedicated to the protection of the right to privacy and dignity, although these are not explicitly linked to witnesses. Similarly, paras 28 and 34 of the Berkeley Protocol, respectively, discuss the right to privacy and the right to dignity as key legal and ethical principles, but their application to witnesses remains abstract and underdeveloped.

The Murad Code, which specifically deals with survivors of systematic and conflict-related sexual violence, offers a more survivor-centred approach because it focuses on informed consent, giving control to survivors over whether and how their stories are recorded, used and shared, especially in evidence (pg. 10, principles 1-3). The Murad code is built around the rights to privacy, confidentiality and security (principle 10), the full respect of survivors’ identities from public exposure or unauthorised sharing (principle 5.7), trauma-informed approach (principles 5-6) and a clear ethical stance that survivors should not be treated as mere sources of evidence, but as individuals with rights (principle 2).

While the Murad Code offers a more holistic model specifically designed for safeguarding survivors of systematic and conflict-related sexual violence in international trials, its scope is quite limited and does not cover all the possible case scenarios involving witnesses.

Reframing Witnesses as Rights-Holders

To close the normative gap and ensure that digital evidence practices uphold the rights and well-being of those who testify, practitioners should move towards a more witness-informed approach. At the heart of this witness-informed approach, there is a shift in perspective. Witnesses must be seen not merely as sources of evidence, but as rights-holders whose dignity, privacy, and safety are paramount. This reframing aligns with broader human rights principles and strengthens the legitimacy of international criminal justice.

Article 68 of the Rome Statute might help with this task. Although it was drafted before the rise of OSINT, it provides some foundational principles for witness protection when it states that the ICC needs to safeguard the dignity, privacy, and security of witnesses. These principles offer a valuable starting point for extending protections into digital contexts.

Operationalising this shift could involve either establishing a new protocol to protect witnesses whose testimony may be corroborated or contradicted by digital evidence or expanding the relevant principles of the Berkeley Protocol and Murad Code to address the rights and vulnerabilities of witnesses throughout all stages of proceedings, and not just during investigations. These guidelines should be fully respectful of witnesses’ informed consent and should ensure that witnesses are given agency over how their data is used. Balancing this aspect with the courts’ necessity to use this data for successful prosecutions presents a challenge. Nonetheless, these guidelines should strike the right balance between the technological efficiency of OSINT and human dignity, because the former form of evidence must not replace or diminish the value of live testimony. Finally, a truly witness-centric approach would include post-testimony support mechanisms, including psychosocial care and follow-up, to prevent witnesses’ retraumatisation and further social isolation.

Concluding Remarks

The digital revolution in evidence-gathering offers powerful tools for accountability in international criminal law. However, while OSINT could strengthen the fight against impunity, it should also pay close attention to the human cost, especially for witnesses. If OSINT is not accompanied by robust guidelines, the pursuit of international justice risks harming the very individuals it seeks to protect. The future of international criminal law depends not only on technological innovation but on its ability to protect and empower the individuals at its core.