Beyond Theory – Interview to Yvonne Ng. By Vicenç Ruiz and Juan Alonso
WITNESS is a global human rights non-profit organization. Its mission is to to help people use video and technology to protect and defend human rights. As the Senior Program Manager of Archives at WITNESS, Yvonne Ng is dedicated to supporting people to create and use archives for human rights change and accountability. She assists and trains grassroots partners, produces sharable learning materials, and advocates on issues related to human rights archives.
Why and how was WITNESS founded? What is your mission?
WITNESS helps people use video and technology to protect and defend human rights. We were founded over 30 years ago by the musician Peter Gabriel as a project of the Lawyers’ Committee for Human Rights. Peter saw the impact that the bystander video of the Los Angeles Police Department’s brutal beating of Rodney King had on revealing the reality of police violence and racism to millions of people around the world. Despite the outcome of that case, he foresaw the immense power of video to enable people to document human rights violations and create change.
Since our start in 1992, we have grown into a global nonprofit organization with over 40 passionate team members based in over a dozen countries around the world. Our work remains centered on the people and communities most impacted by human rights violations, and we work alongside advocates in those communities who are using video and technology towards human rights goals. We also share our experience and learning within a broader human rights video landscape, and support our allies and peers working in this space.
Normally when we talk about archiving we imagine that it is something that only takes place in archives, but your approach is very different from that of an institutional archive. In terms of archive management, how is WITNESS different from other archives? What do you understand as “own collection” of WITNESS? Have you considered centralizing third-party fonds management services?
I think the idea that “when we talk about archiving, we imagine that it is something that only takes place in archives” does not serve institutional archives or archivists well. To conceive of archiving in this narrow way, when we are living in an age defined by the creation, control, sharing and processing of information, limits archivists’ relevance and potential to contribute. Archives — records that have enduring value to an individual or group — are everywhere. Archiving, if understood as the work involved in selecting and caring for those records, is being performed by all kinds of people and entities. The archival field has plenty to offer to other contemporary discourses and domains (Take for example, the approaches that computer and data scientists Eun Seo Jo and Timnit Gebru were able to glean from archives to inform their thinking about data collection for machine learning in “Lessons from Archives: Strategies for Collecting Sociocultural Data in Machine Learning”.)
But to answer your question, as a program situated within a human rights organization, we approach archiving fairly differently than an institutional archive whose primary mission is to preserve and provide access to collections. For WITNESS, archiving is more of a means to an end (i.e. a human rights advocacy goal), and not the end itself. We also take an expansive and inclusive view of what constitutes archival practice.
In terms of WITNESS’s own collection, we have not actively collected for many years since shifting away from co-producing short-form advocacy videos with our partners, a strategic transition that reflected changes in the socio-technological landscape for video as well as changes in human rights documentation and advocacy. We do have a valuable and unique collection of over 20 years of raw footage and edited works documenting a wide range of human rights issues globally, that were filmed and produced by our partners and our team. We still maintain the digital collection at WITNESS, and have a relationship with the City of Malmo Archives in Sweden to provide long-term preservation to the whole collection. Some of our collection is also at the University of Texas Libraries at Austin.
Related to the previous question, one of your functions is dedicated to supporting people to create and use archives for human rights changes and accountability. Even on your website you have a special section dedicated to training activists in the management of audiovisual archives in which you share guides and other resources. How do you carry out this assistance? Do you carry out ad hoc consultancies for individuals or organizations that require it? What is the approach to assess the readiness of small organizations on starting their own archival initiatives with limited resources?
The Archives Program is one of WITNESS’s programs, and supports WITNESS’s mission through the lens of archiving and preservation. For example, my archivist colleague Ines Aisengart Menezes and I work with our regional programs to deliver archiving trainings, and we collaborate with community-based partners within our regions to support their archiving practices. These collaborations typically emerge from relationships in which trust and understanding have been built over time by our team, and where we find alignment with the partner in terms of what support we can provide. Our collaborations can take different forms depending on the needs. It can range from providing microgrants, training, and equipment to support their work; co-designing workflows/methodologies or tools; and providing ongoing consultation and hands-on support, and helping to make connections to other allies. Through the process, we have the opportunity to develop and test new approaches to archiving in different contexts, and to learn alongside our partners. We are grateful to have supported partners such as Rohingya Vision on the Rohingya Genocide Archive project and Berkeley Copwatch on the People’s Database for Community-Based Police Accountability.
From what we learn in our collaborations, we create and co-create accessible resources about archiving such as guides, tipsheets, and videos aimed at human rights activists that we share in our Library and on social media in multiple regions. We also try to bridge human rights and archival communities to share useful knowledge between domains and contribute ideas on how archives and archival practices can work in support of human rights.
One of the skills in which you train activists is digital preservation of video, something that requires human and financial resources as well as planning that considers, for example, changes in format and support. What happens when the activist cannot take on this task because, for example, they don’t have time? Do you think it is a good idea to work in collaboration with other institutions to guarantee good management and preservation of this audiovisual material? What cases do you know or recommend in this regard?
Archiving, even pared-down approaches, requires time, capacity, and resources. Getting started can be a big hurdle. It can be challenging for activists or grassroots human rights groups to add archival workflows on top of the urgent and careful documentation work they already do, and it is even harder to sustain archiving over time, especially in precarious contexts. When we train or create guidance, we present archival practices as steps that can be incorporated into documentation workflows to strengthen the ability to use video collections, and not as an “all or nothing.” We emphasize principles and good basic practices that can be implemented using free/open-source or at least widely available tools.
Not all of our collaborations include an archiving element. Often it is not the highest priority for a partner or it is beyond their current capacity, and so our collaboration may center on other areas where WITNESS can provide support such as filming, storytelling and advocacy, or video as evidence. In these instances, we might still share information about archiving, as well as other good documentation practices to help keep videos intact, identifiable, and retrievable in the meantime.
To your question, it can be a good idea to collaborate with an institution or other organization to help manage and preserve audiovisual material. I think it depends on the situation, the needs, and the alignment of values and goals. A more resourced entity could provide services and infrastructure that might otherwise be unavailable, or be situated in a location that provides more security and protection. One of our close allies, Mnemonic, for example, hosts a small number of ongoing archiving programs and projects, and also does ad hoc rapid response to provide infrastructure for documentation initiatives in situations where human rights violations are taking place but the documentation ecosystem is underdeveloped.
In addition, in digital preservation we not only have to consider the aspects that affect the integrity of the audio-visual document file, but also all the hardware and software that allow digital video management. In 2020, organizations such as Human Rights Watch or Mnemonic were affected by the inaccessibility of the youtube-dl code. This happened almost a month before Human Rights Watch published the report “Video Unavailable.” What exactly happened? Why was the recovery of youtube-dl so important?
Online videos have emerged as an important source of information and evidence about human rights violations. They are especially valuable in hard-to-reach contexts like conflict zones, where investigators and the media need to rely on citizen-shot video for information. Youtube-dl is one of the primary open-source tools that human rights groups, tool developers, and others use to download videos documenting violations from sites like YouTube, Facebook, and Twitter.
In 2020, the Recording Industry Association of America submitted a Digital Millennium Copyright Act (DMCA) request to GitHub to remove all public code repositories of youtube-dl, spuriously claiming that the tool circumvented technological protection measures and that its primary purpose was to enable downloads of commercial music videos and sound recordings. Github initially complied with the takedown request, but reinstated the repository after they determined, with the help of EFF, that the tool did not in fact violate the DMCA. Human Rights Watch, Mnemonic, and WITNESS issued a joint public statement in support of Github’s decision to reinstate youtube-dl, citing its important and non-infringing application in human rights research and investigations.
Unable to cow Github, the recording industry has, sadly, sought out less visible and powerful actors to try to get its way. In 2022, a group of commercial record labels sued a small web hosting provider, Uberspace, in German court. Uberspace does not even host youtube-dl, but simply the homepage for the youtube-dl project. Fortunately, a German civil rights defense group, Gesellschaft für Freiheitsrechte (GFF), has been helping Uberspace defend itself in this case, which is still pending.
Your approach to the archive is interesting because, beyond the fact that the material can accompany reports or illustrate your activities through corporate videos, image galleries, publications, and social networks, it condenses one of the essential aspects of the archive: an evidentiary document that enables accountability to people or governments that are violating international standards of human rights. How can one guarantee the authenticity of these documents in a trial? To what extent is it dangerous to manage this information and apply security and confidentiality protocols? How can we help WITNESS on this matter?
Courts and fact-finders want to know that a document is what it purports to be, and that it hasn’t been tampered with or manipulated. Authenticity speaks to the reliability, and therefore the probative value, of documentary evidence. Traditionally, authenticity can be established with external indicators, like testimony or information about the identity of the source, or internal indicators, such as metadata or other characteristics of the document.
Authentication becomes more challenging with digital open source information, which is increasingly being used before courts and by fact-finding bodies. Digital open source information includes content like photos and videos from social media, satellite imagery, and other published data. This kind of information can be more challenging to authenticate when it is disassociated or decontextualized from its original source, or if its original source is not known, and because digital content can be easily modified. When traditional indicators of authenticity are unavailable, investigators can rely on certain analysis techniques to try to determine integrity, identify sources, and verify where and when the document originated. Open source investigation tools and techniques have developed and matured in recent years, and we are interested in ways of democratizing access to them, so that activists and practitioners in communities most impacted by human rights violations can use digital open source information effectively in their own advocacy.
As you mention, there are of course security and confidentiality concerns with managing human rights evidentiary material, which can contain sensitive information. Organizations that undertake this kind of work should be clear about why they are collecting this material, assess the risks carefully, and develop criteria and protocols for how they collect, handle, store, and provide access to their collections. We do not currently collect or analyze significant quantities of this kind of information at WITNESS, but we have allies and peers who are thoughtfully engaged in this work.
Both in dictatorships and, increasingly, in democracies (i.e. in Spain the so-called “Gag” Law), documenting abuses by security forces is a risk, both for those who film and for those who are filmed suffering the abuses. When archiving is illegal, how can people fight for their right to record?
The right to record law enforcement is essential to civil society’s ability to expose violations and hold authorities to account. However, this right is not recognized nor protected in many countries around the world, and in some cases (as you mention in the case of Spain) it may be prohibited. Even in places that have laws that protect the right to record, the reality on the ground is often different. Restricting the right to record and disseminate recordings of police officers and security forces performing their official duties in public is not only problematic, but also ripe for abuse by authorities.
In terms of what people can do, knowing your rights is the first step. Since laws and enforcement trends vary from place to place, it is important to research the laws and analyses by lawyers and legal activists in your own locale. Before you film, evaluate the risks to yourself and the people you will be recording (Remember that filming may not always be the best option). If you do film, for your own safety, stand at a distance, and do not obstruct or interfere with the police. Film with intention to make your effort count, and then protect yourself and your recordings with good mobile phone security and automatic cloud backup.
Another example of a current threat is the malicious use of so-called “deepfakes” and other forms of AI-generated “synthetic media”. In this context, how do we push back to defend evidence, the truth and freedom of expression? How can we work together to detect AI-manipulated media? Is it also an opportunity for us?
While malicious deepfakes and AI-generated synthetic media are a real concern, our overall message about their threat is to “prepare, don’t panic”. That is, we should be proactive about addressing the ways that truth is being undermined, but not escalate the hype and rhetoric which in itself can cause harm. One dangerous consequence of deepfakes hype, for example, is that authoritarian regimes can now cash in on the so-called “liar’s dividend” — the yield from mounting public distrust and skepticism about the authenticity of all videos that enables them to deny any real evidence of their abuses as “fake news.”
Deepfake detection is one approach to addressing media manipulation, but it has some limitations. Detection is resource-intensive and needs to endlessly keep up with generative technologies in a sort of “arms race”. Even when detection is possible, the tools, skills, and capacities to perform detection or analysis on complex cases are not distributed equitably or even appropriately to where needs are around the world. Addressing the problem of malicious deepfakes and other misinformation therefore requires a more comprehensive and inclusive approach than just detection.
The key theme in our current strategic vision is “Fortifying the Truth” — meaning that instead of just focusing on detecting lies, we aim to strengthen the credibility, integrity, and power of human rights defenders’ truthful accounts of violations, to amplify their voices and narratives, and to help them in creating trustworthy video that can be effectively used. We do this through supporting activists with skills and capacity to create reliable media, helping to develop new documentation tactics, and through influencing technology providers towards creating rights-protecting infrastructure that supports authentication and provenance.
Shoshana Zuboff popularized the term “surveillance capitalism” to define a new stage of capitalism based on the commodification of personal data, that is, on the transformation of personal information into merchandise subject to sale and purchase for profit. It seems a paradox that, in the so-called information society, citizens have so little control over our personal information; that it is in the hands of private companies. Recently, HRW published a report titled Governments Harm Children’s Rights in Online Learning warning that governments of 49 of the world’s most populous countries harmed children’s rights by endorsing online learning products during Covid-19 school closures without adequately protecting children’s privacy. In addition to the exploitation of personal data, the data managed by private companies is not manageable or exploitable by the citizens and initiatives such as Documenting the Now for seeking a user-friendly means of collecting and preserving this type of digital content as it arises. What do you think of it? What role do you think the institution’s archives should play in this scenario? What is WITNESS doing?
This isn’t an area that WITNESS works on, but it is of course a big problem that laws and regulations don’t seem to have caught up with yet. It is increasingly impossible to opt out of providing personal data to access everyday and essential services. I don’t think the models we have for thinking about or getting “consent” are sufficient for the ways that data is collected and used today, and how intertwined we are with our personal data. Groups like Internet Democracy Project talk about taking an embodied approach to data, and understanding the impact of data practices not just in terms of “data protection” or invasions of “data privacy” but in terms of real bodily autonomy and integrity, and physical well-being, especially for the most marginalized.
As for what institutional archives, WITNESS, of others could do, even if we don’t work specifically on these issues, I think we could all examine our own data practices and identify threats and harms; model responsible data usage; play a role in advocating for legal and regulatory frameworks that protect people, especially the most vulnerable; and promote data literacy and self-protection among our own teams and constituents.
