Yvonne McDermott Rees is a Professor of Law at Swansea University and an expert in international criminal law and procedure, evidence, and human rights. She is Principal Investigator on OSR4Rights, an interdisciplinary research project which uses open source research to transform the discovery and documentation of human rights violations.
Dan Anlezark is an investigator, specialising in open source intelligence (OSINT). Dan is currently the Deputy Head of Investigations at Myanmar Witness, an open source accountability project run by the Centre for Information Resilience (CIR). Formerly, Dan has filled investigative roles with the UN’s Group of Eminent Regional and International Experts on Yemen (GEE Yemen) and the UN Investigative Team to Promote Accountability for Crimes Committed by Da’esh/ISIL (UNITAD). Dan is a graduate of the Hillary Rodham Clinton Global Challenges program at Swansea University.
The second edition of GRC’s unique Starvation Training Manual released in 2022 includes a brand new section on OSINT, debunking the myth that this technique can only lend itself to certain types of crimes but is of limited use when it comes to the crime of starvation. The inputs from Yvonne and Dan were invaluable in the inception and creation of this section.
Q1: Can you tell us about the OSR4Rights project?
Yvonne: OSR4Rights ran from 2018-2021, funded by the UK’s Economic and Social Research Council. The team and I explored the extent to which open source investigative techniques are being used in human rights investigations, and the challenges and opportunities posed in dealing with open source material in practice. There was also a technical aspect of the project, where we developed some tools for the analysis of open source information (currently hosted at https://osr4rightstools.org) to help human rights investigators in analysing the huge volumes of information generated through open source investigations. Since July 2022, we’ve been working on the TRUE Project at Swansea University, a five-year project which examines the impact of deepfakes on trust in user-generated evidence of human rights violations.
Q2: Since OSINT does not require confidential access, should a regular member of the public or a non-specialised organisation participate in the documentation of possible human rights violations?
Yvonne: By their nature, digital open source investigations can be conducted by anyone with an internet connection. But that doesn’t mean that everyone should jump into the documentation of human rights violations without a thorough risk analysis – assessing both why they wish to engage in an open source investigation, and how they propose to do it. Thankfully there are excellent trainings and resources available for those looking to get started in this space – I would particularly mention Amnesty International’s online course, the Berkeley Protocol, and trainings carried out by the IICI/Berkeley and Bellingcat, amongst many others.
Dan: There is a need to view OSINT as both a process and a product. An individual can apply OSINT techniques to identify content of interest, but it is the analysis of such where real value is gained, particularly when done so comparatively and holistically. As Yvonne has highlighted, there is a difference between ‘can’ and ‘should’ in these situations, and I would encourage both individuals and organisations to consider where best they may be placed to achieve positive impact prior to utilising open source techniques.
Q3: In your experience, how would OSINT supplement, replace or otherwise interact with information collected through the traditional means, especially in investigations in areas such as Yemen?
Dan: The use of OSINT and open source investigative techniques almost always complements more traditional methods of investigation for accountability of international crimes. Of course, they can both stand on their own, but mutual implementation assists in building a compelling and grounded narrative. In the Yemen context, for example, the collection of witness, victim and/or survivor testimony can be strengthened through the conduct of geolocation and chronolocation exercises. Specially, in the context of an airstrike, comparative assessment of User Generated Content (UGC) collected from social media can be viewed alongside satellite imagery to identify the exact location and time of an airstrike. No single investigative method covers all bases in these investigations, highlighting the need for researchers, investigators, activists, and advocates to be versed across both traditional and more developed investigative methods in the conduct of their work.
Yvonne: I think it’s very important to avoid viewing OSINT as a replacement for traditional investigations. There as a risk that doing so could drive control over the narrative out of the hands of survivors, as we have written about here, and as Libby McAvoy has convincingly shown. As Minogue and Makumbe have argued in the context of Yemen, ‘The use of OSINT does not have to mean driving accountability efforts out of the hands of Yemenis and into those of international investigators. Its analysis is far better carried out by, or involving, those with local knowledge.’
Q4: You both are helping the GRC with the preparation of a manual on starvation for collaboration with GRC. How do you see OSINT lend itself to starvation investigations?
Dan: The key to applying OSINT techniques to any investigation is the use of a critical mindset and adherence to lateral thinking. When assessing the accessibility of Objects Indispensable to Survival (OIS), OSINT can play a major role in understanding the on-the-ground situation from a distance. For example, the diligent scanning of content posted to SnapMap may provide insight into food, clothing, bedding and other OIS item costs and availability. This is but one example of how UGC can provide a glimpse into the life of individuals affected by conflict. Additionally, in contexts where UGC upload is limited, the sourcing of satellite imagery also assists in the conduct of change-detection, showing potential destruction of crop fields or blockades against civilian access to food and other essential items.
Yvonne: I think there is a perception that open source investigations do not lend themselves well to certain types of violations, including starvation. But as Alexa Koenig and Ulic Egan have shown in the context of sexual and gender based violence, sometimes it’s not that the information doesn’t exist, but just that investigators don’t know how to find it. Dan’s examples are great, and show how a bit of lateral thinking can uncover relevant information.
