The Media Today

Q&A: Omar Ferwati on Forensic Architecture’s probes of the present and the past 

May 29, 2024
Palestinians look for survivors after an Israeli airstrike on the Gaza Strip. (AP Photo/Mohammed Al Masri)

Last week, the International Court of Justice, an arm of the United Nations, ordered Israel to stop its offensive in the Gazan city of Rafah, where more than a million Palestinian refugees had amassed during more than seven months of war. Many have since fled. The ICJ lacks the means to enforce its rulings. Since this one was handed down, an Israeli strike on a camp for displaced people in Rafah killed dozens of Palestinians. Benjamin Netanyahu, Israel’s prime minister, insisted that the strike had targeted two senior figures in Hamas, and described the killing of civilians as a “tragic mishap.” 

The ICJ ruling followed an original series of orders that the court released in January after weeks of proceedings in a case brought by South Africa accusing Israel of genocide in its response to Hamas’s attack on October 7. During the case, lawyers representing Israel defended the country against such allegations, including by presenting a series of visual materials—annotated images, videos, maps—which they claimed offered proof that Hamas fighters were embedded among civilians in Gaza. (Israel has also described the claim of genocide as “false, outrageous, and morally repugnant.”)

Israel’s case drew heavily on what Omar Ferwati calls “the aesthetics of civilian documentation.” Ferwati works at Forensic Architecture, a London-based research agency established in 2010 by Eyal Weizman, a British Israeli architect, to model and analyze human rights violations, not least in Israel. The widespread availability of satellite data and crowdsourced footage has increasingly allowed research groups like Weizman’s to scrutinize conflict zones remotely, using architectural principles and digital models as investigative tools. Ferwati, whose role includes developing new analytical techniques, told me that Israel’s defense team leveraged some of the same open-source methods that Forensic Architecture uses itself—reflecting, in his view, a growing understanding of how digital artifacts are shaping public perceptions of conflict. 

About a month after the ICJ released its initial rulings in the case, Forensic Architecture published a report in which it examined Israel’s legal defense, argument by argument. The group ultimately challenged the way that Israel’s lawyers used even legitimate photographs and videos, alongside what it called “incorrect annotations and labeling, and misleading verbal descriptions.” Forensic Architecture concluded that the defense distorted facts to its advantage. “It’s not just about what you show,” Ferwati said. “It’s about how you describe it.” 

In addition to investigating active conflict zones, Forensic Architecture has reconstructed historical sites of conflict, including in recent reports on massacres in 1893, perpetrated by German colonial forces in Namibia, and 1948, by Israeli forces in the Palestinian village of Tantura. The group’s work has been cited in legal cases, including at the ICJ; it has also been showcased in museums and galleries around the world, raising questions as to whether it also constitutes art. Last week, Ferwati and I spoke about the aesthetics of digital investigation, the interdisciplinary possibilities of open-source journalism, and the potential impact of revisiting historical events with contemporary tools. Our conversation has been edited for length and clarity.

YRG: Open-source investigative techniques—such as the analysis of satellite imagery, cellphone footage, and other geospatial data—are growing in popularity as journalistic tools. But Forensic Architecture focuses on “spatial analysis.” What is this technique? And how does it draw on the principles of architecture?

Sign up for CJR's daily email

OF: We’re interested in situating media and events in space and time. In a 3D model, we can re-create the space of an event in question; then we can place within it footage that we find [on social media] or that is sent to us directly. Synchronizing pieces of footage and then watching them in relation to one another begins to unravel unanswered questions about an event.

For example, when we investigated the murder of Shireen Abu Akleh [a Palestinian American journalist with Al Jazeera who was killed while covering an Israeli raid in the West Bank in 2022], we found one piece of footage which showed Shireen walking up at the moment that she was shot, and we have another piece of footage that shows us where the Israeli military vehicles were stationed. Neither of those on their own give us a full understanding of the spatial relationship between where the Israeli soldier that shot her was and where Shireen was. But by putting both in a model, we can position where she was standing and where the vehicle would have stood. We can then draw the direct trajectory and assess if it was possible that she could have been shot from there; was there a direct line of fire? It also begins to tell us a little bit about the intentionality: How narrow was that line of fire? These are all things that we can deduce by combining that media.

Forensic Architecture’s work sometimes blurs the lines between investigation and art. It has been presented in museums and galleries around the world. Do you think of what FA does as art?

This is a contentious issue. Our work as a research agency is geared toward being admissible as evidence in court, in addition to being [presented] in [other] forums, including people’s tribunals, exhibition spaces, and cultural spaces of discourse. I personally don’t see the work that we do as art, though. It enters an aesthetic practice because it engages with the senses, with tools of communication that are primarily audiovisual. So it can overlap with the mediums used in museums and galleries. The evidence that we investigate sits somewhere between disciplines. 

You’ve written about how the kind of investigations you do are made possible by the ubiquity of cellphone footage and the proliferation of mapping data. Do you think this shift has changed the nature of war reporting? 

It seems that there’s an increasing awareness [among civilians] that the footage that they’re broadcasting in real time of what has happened to them, or what is about to happen to them, is a kind of act of defense. We see that evidence documented by civilians, and analysis that derives from that footage, is increasingly put forward as evidence in legal and political forums. We also see [the same] realization from the states that are engaged in violations. Confiscating cameras, putting people in places without any connection, cutting off communications—for example, across Gaza repeatedly since October by the Israeli military. This looks to be evidence that they are aware of the power of citizen documentation.

Forensic Architecture analyzed the visual materials that Israel’s defense team used in  January in the country’s trial before the International Court of Justice. What did you find?

[The Israeli legal team] made a series of claims, including with annotations and video clips. We looked one by one at what source material they derived from and did a simple 3D analysis, including when they made claims about the trajectories and the origins of attacks. We also looked at the [question of] timing: Was a site that they made a claim about being used by civilians at the time [of an incident] or not? We found that there were eight instances where the Israeli legal team misrepresented the visual evidence they cited, either through wrong annotations or with misleading verbal descriptions. We found that there was an interesting use of singular cases, even when they were misrepresented, as blanket justifications for systematic and widespread attacks on civilians, shelters, schools, and hospitals. It gives the impression that all attacks on any civilian targets and on any civilians become justified. 

Forensic Architecture is very explicit about its focus on injustice, which raises questions about truth and impartiality. How do you navigate questions of objectivity in this context?

It’s an important question to ask: how someone that is doing analysis engages with the facts on the ground; how they are working in service of truth while being also in service of victims of egregious violations. As I understand it, the claim of objectivity made by some outlets [can actually constitute a refusal] to engage with the implications of their work. We don’t see ourselves as neutral, but we do see our work as accurate. We are very clearly in solidarity with victims of violations. We will listen to victims of violations. And then we will ask questions and think about the claims that they make. It just so happens, I think, that the claims made by many victims are often what will lead you to the truth.

Everything that we [produce comes] from a multidisciplinary team and is checked over and over. But in my view it’s important to highlight, whether in journalism or open-source research, that there is still a choice [when it comes to] what you’re engaging with and the questions you’re asking. It’s not just about whether the analysis is right or wrong, but about where you’re applying that ability to analyze, where you’re applying your lens. 

Forensic Architecture not only analyzes material from unfolding crises but also does retrospective analyses of historical incidents. For example, you conducted an evaluation of the Hornkranz massacre, perpetrated in Namibia by the German colonial administration in 1893, and the massacre in Tantura, Palestine, in 1948 by an Israeli brigade. Both investigations drew on first-person testimony. Given that human memory is notoriously fallible, how do you integrate data-oriented tools with more subjective storytelling?

What we are trying to do increasingly is to use contemporary technology to reconstruct spaces as they were at the times of historical events—in these cases, historical [acts] of violence. In the cases of Tantura and Hornkranz, our findings stem directly from the stories we were told by the victims or by the descendants of the victims. I think we have a choice to make about whether we see the testimonies of people as stories or as leads for new findings. In the case of our work on the Hornkranz massacre, generations removed from [the event] still have the stories of how violence unfolded. We can either decide to let stories go down, generation after generation, removed from the rest of the archive. Or we can understand it as a way to lead research.

In [investigating the Tantura massacre], we really have no way of understanding where to begin without victims’ guidance. After the village was invaded, it was completely destroyed—nothing remains of it today except for two or three buildings. But what we have instead is a rich archive of testimonies of what happened that day. We also have some records, maps, and photographs from the village before the massacre took place. Because of the testimonies, we were able to closely examine aerial images that were taken shortly before and shortly after the massacre in May 1948. We created a “memory sketch,” where the victims themselves drew out the village as it had been. This is an example of what we call “situated testimony,” in which we sit next to a victim and we build that scene; we deposit their testimony not just in writing or orally, but spatially. They can tell us what their houses were like, where the paths were, what they’d seen, and in this case, where people were buried. We found the site of two mass graves, as they were described by the victims. 

What impact do you hope the historical work will have, compared to the more contemporary projects you work on?

I think every historical case we’ve looked at has an urgency in the present. That’s part of the reason why we continue to look at them—victims, or the descendants of victims, are still calling for justice today. For example, the site in Namibia is not accessible by the descendants of the Witbooi Nama people whose ancestors were massacred there. We’ve worked on multiple cases in Namibia in different periods of time, and we find them all interconnected with the suffering of that genocide. 

In revisiting [these cases], we are trying to enrich our understanding—to evidence what they have been telling us all along. We’re not always revealing something entirely new; we’re finding what they’re saying in plain sight. We’re trying to add materiality.

Other notable stories:

New from CJR: AI can’t anthropomorphize its way to empathy

Yona TR Golding is a CJR fellow.