You're listening to a stage talk titled covering gender -based violence. This week we spoke to Dr. Alexa Koenig from UC Berkeley. Alexa has spent much of her recent career looking into how digital investigations could aid accountability for war crimes and gender -based violence. Within this discussion, we focused on her work covering gender -based violence and what lessons she's learned from the various groups she's spoken to and worked with, from legal professionals
to survivor groups. This talk was hosted by me, Shaila Ma, on Thursday the 20th of February 2025 in the Bellingcat Discord server. Greetings all, and welcome back. We're here with another Sage Talk. I know, so quick after last week. I wanted to bring you a series of talks on the topic of covering traumatic events. Many of you in this server have reported on or witnessed traumatic
events in the past year. It's important that when we're in that research mode that we remember those that feature in the images or videos you might be focusing on are real human beings with thoughts and lives. But how do we continue to do our jobs and remain respectful and true to those featured in these atrocities? Well, Dr. Aleksa Konig next to me is co -faculty director of UC Berkeley's Human Rights Center and a research
professor at UC Berkeley School of Law. She co -founded the Human Rights Center Investigations Lab and is a director of the investigations program. She's also an author, including of such titles, Graphic Trauma and Meaning in Our Online Lives, Hiding in Plain Sight and Digital Witness. She is a very busy lady. She's covered extensively how to report on potential war crimes and gender -based violence with open sources and today we're going to dive into her research on gender -based
violence. Please note today's discussion will potentially reference abuse towards men, women and children. If you'd like to ask a question please do so in the chat as I mentioned it's in the top right hand side of your screen as we talk. Please note in your question, if you don't want me to mention your name, because this is being audio recorded. And please remember that there may be people within this space who have lived experience of what we're discussing.
So please remain respectful. That all being said, Alexa, I'll pass on to you. Please introduce your work within this topic. And thank you again for joining us. Thank you, Charlie, for having me and thank you to all of Bellingcat and to everyone for being here today to talk about this issue that has become incredibly near and dear to my heart. As Charlie mentioned, our background at the Human Rights Center has really focused a lot on digital open source investigations.
I started at the Human Rights Center at UC Berkeley back in 2007 when I was a graduate student and kind of stuck on in 2012 becoming the executive director and now co -faculty director. During the time that I was there, starting around 2010, one of the things that we noticed in war crimes investigations more generally was the deep need for better understanding how sexual violence
happens in conflict. And we launched a sexual violence and conflict program that my colleague in Selinger, who's most recently been with the International Criminal Court, ended up directing. We also had begun working very closely with the International Criminal Court and other justice and accountability actors to think about the ways that new and emerging technologies were helping to establish evidence of potential crimes, particularly international crimes, that needed
some sort of reconciliation. Back in 2010, 2011, 2012, the International Criminal Court was coming up on its 10 -year anniversary. And unfortunately, the Office of the Prosecutor was losing a lot of its cases. One of the things that a research team from our center determined was that the main complaint of the judges was that the prosecution was not bringing in corroborating information that would help support what survivors were saying they were experiencing in their communities.
And so that really began this parallel quest for the last decade and a half of thinking about how things like information posted to social media sites, information captured on smartphones and disseminated might be useful as evidence in courts of law. Those two trajectories, though, on sexual violence and conflict and technology and war really did not converge until fairly recently. And I'd say in the broader field of war crimes investigations more generally, they
still have not converged. When I did training in war crimes investigations with the Institute for International Criminal Investigations, we certainly had a unit on sexual and gender -based violence. But those specialized topics are kept very separate from digital open source And it's only been recently that we've begun to see a real hunger to try to begin to bring those worlds together because of the incredible utility of digital open source methods for helping us make
sense of what's happened in the world. We now have a professional training program where we do try to train people on the careful use of these methodologies. There are really two, I think, big challenges in the space that today open source investigators are really trying to grapple with to figure out how we make this work as ethical, as efficient, and as effective as possible. One is the gross under prosecution and accountability for sexual and gender based
crimes. We know that historically those crimes have been deeply under prosecuted. And even most recently at the International Criminal Court where there was a successful conviction of a gentleman by the name of Al Hassan who was accused of egregious international crimes in Timbuktu and Mali. He ended up being convicted of almost all of the charges brought against him with the exception of sexual and gender -based violence.
This is despite abundant evidence. This is despite a lot of the incidents overlapping for other kinds of crimes like religious persecution. So there really, I think there is a recognition that we really do need to think through the evidentiary foundations of such cases and begin to identify the gaps that are allowing these kinds of crimes to go under addressed. At the same time, I think there's a real skepticism and in some cases a
real concern. with bringing digital methodologies into these investigations and these prosecutions. So I work very closely with a few colleagues from around the world, Juan Ulick Egan. He and I've done a number of different studies of what is, how do we begin to fill that gap between the hunger to get more justice and accountability, the deep knowledge of methods that have been pioneered by groups like Belincat. to think about the use of these digital methods and actually
getting some form of success. And so we've talked with three different sets of individuals. One group have been people who are trained in investigating sexual and gender -based violence through very traditional methodologies like face -to -face interviewing or collecting physical evidence, getting medical records, et cetera. Second have been digital open source investigators. What are the challenges that they've seen and that
they're experiencing? And then the third are gender experts who really have kind of a broad sense of the ways that different gendered crimes can play out. What we have consistently found is that on the side of digital open source investigators, there's been sort of a response of I don't mean
that in a negative sense. I just mean lack of awareness of some of the specialized ways that different kinds of terminology, different kinds of communication patterns happen, specifically with these deeply stigmatized crimes that may be different than other things that we're detecting through open methods, like attacks on hospitals
or the bombing of buildings. On the side of the sexual and gender -based violence investigators, I'd say that their experience or their response has ranged from skeptical to even hostile to having digital investigators get involved. And they have some really good reasons for that. One of the things that we've seen in the sexual and gender -based violence community is that over the last 20 to 30 years, I pioneered a lot of new resources and tools for investigating
these crimes. Those have all been, and I think rightfully so, putting the survivor at the center of those methodologies and trying to think about how to do no more harm with the investigative process. They've also really been trying to bolster the response of judges to the evidence that comes
into courts. So getting the judges to not have a quote unquote he said she said type of situation where they basically are saying we need additional evidence to break that tie, but to believe survivors and to have that testimony have sufficient weight.
So I think that one of the fears that we have heard repeatedly from those traditional investigators is If it begins to be that these cases bring in more video and photographic evidence or other stuff that may be pulled from social media, eventually the judges may demand that kind of evidence and actually have it weaken or in some ways undercut the oral testimony that these survivors may give
in court. I do think that based on the research that I've been doing with my colleagues, one of the things that we're trying to basically communicate across these areas of practice is actually Even if that is the concern, there are still so many ways that digital open source information can be helpful. For example, digital open source information can help prove the crime base that sexual and gender based violence has taken place.
And whether you're looking for that affirmatively or you stumble across it in the course of your investigative work, it is really important evidence that could potentially be helpful if it's handled. to forensic and ethical standards. The second is it can provide really valuable indirect evidence that sexual and gender -based violence has occurred and who may have perpetrated it. So everything from identifying the units who may have swept through a town where we believe these things
have happened. to identify what are called proxy indicators of sexual and gender -based violence, I think can be really helpful. So by proxy indicators, if any of you aren't familiar with that term, I know there's a great piece that was written by Hannah Bagdazar, who used to be with Bellingcat
on this. But war crimes investigators have identified a number of different phenomena that you might see in a conflict context that indicate that sexual and gender -based violence has taken place, even if you don't have the direct evidence itself. So some examples of that would be things like the separation of women and girls from men and
boys, things like the burning of villages. I do think there's tremendous possibilities there, particularly with the advent of artificial intelligence -based technologies and machine learning, to begin to find the signal in the noise in ways that humans may not pick up on, particularly if you have a number of these proxy indicators that have all converged at a particular time
and a particular place. The third category where I think digital open source information can be deeply helpful is on providing the context or the contextual evidence around a particular incident. This is especially relevant for when you're trying to establish it in an international crime like genocide or crimes against humanity or war crimes
have taken place. In that case, you're not just trying to prove that a rape or other forms of sexual violence have occurred, but you need to establish that there's a nexus to war if you're establishing war crimes. You need to prove that this kind of pattern of behavior was systematic
or widespread. If you're going to establish this as an international crime against humanity, and if you want to establish this as a piece of genocide, to establish that the intent behind the perpetrators was to destroy in whole or in part a particular group on the basis of their ethnicity and nationality, religion, et cetera. So I think overcoming that and making sure that more traditional investigators have a broad understanding of the full range of utility of these methods is deeply important.
At the same time, I think there's some real opportunities for training among this community of practice to understand the fears and the concerns of both survivors and traditional investigators and to think through kind of the complementary function that this sort of information can share. The other thing that I might say is that We've heard from a lot of traditional investigators who are also just deeply concerned about anyone investigating
these issues without specialized training. In part, it's because of the social stigmas and harms that can come from the revelation that someone has experienced sexual and gender -based violence. This obviously can have very differential harms depending on the community or culture from which the survivor's been impacted. So everything from honor killings, if it turns out that someone has been raped, to other forms of stigmatization and ostracization are really important to think
through. I've certainly heard stories of open source investigators either trying to crowdsource who the perpetrators were for a particular rape and having that really create a negative backlash
for them and for the survivor. I've also heard of information getting posted online or amplified online by members of the digital open source investigations community where survivors who were going to be testifying in court recanted their testimony when it came out more publicly that they had experienced sexual and gender -based
violence. because in their communities and their cultures, the backlash would be so great that they may be expelled from their families or their communities, and they were not understandably willing to take that risk. So I think there's a few tools that are beginning to emerge that I would also just want to make this community
aware of if you're not already. As part of this phenomenon of trying to make investigations and investigators really sensitive to the phenomena that are unique to this area of practice, There are a number of tools that have been created for more traditional investigations. One of those is an international protocol on investigating sexual and gender -based violence. I believe
that came out initially around 2014. I do think even for the digital investigations community, it's good to review that set of guidelines, even though there's very little mention of digital methodologies, because a lot of the fundamental concepts of how to be sensitive in your investigations hold true. The second is something called the Maraud code. It is named after Nadia Murad, who won the Nobel Peace Prize for her activism around the treatment of sexual and gender -based violence
survivors. She is herself a Yazidi activist who unfortunately suffered under the genocide perpetrated against the Yazidi community. And I think one of the things she became very overwhelmed with as this genocide was being investigated was how often journalists or war crimes investigators or trying to get information from her and really overwhelming individuals like herself and others
who've experienced these kinds of crimes. And she really wanted to give back to the broader investigations community about how to and how not to handle these kinds of things, both confidentially and when putting information out publicly. So the Maraud Code was a partnership that was informed by a global consultative process. was coordinated by the Institute for International Criminal Investigations and can be accessed for free in a number of languages
on their website. It was also in partnership with Nadia Murad's organization, and it was informed by the input of a number of survivors. The challenge with the Maraud code, however, is that while it does in a few places mention digital open source investigations, the primary focus really does assume that you are face to face with the survivor and that most of the evidence that you're
collecting is testimonial evidence. So a number of us from a number of different open source investigative organizations have kind of worked together to begin to translate the Maraud code for digital open source investigative practice. This is a document. It's basically been called the Open Source Researcher's Guide to Systematic and Conflict -Related Sexual Violence. It will go up publicly in about two weeks, we're estimating.
And the idea is that we'll have it open and available for people to use for approximately a six -month period and to give input. So we'll have an input form where, as you're using it, one thing we'd greatly appreciate is feedback on what works for you and what doesn't. It is probably the hardest project I have ever contributed to, to draft, even though it is again informed by global consultation of different units around the world
and different individual investigators. in part because we're trying to bring those sexual and gender -based violence to very experienced investigators together with digital open source investigators. And I think from an ethics perspective, we've
really needed both a push and a pull. So a push to encourage people to think about the ways this information can be useful as evidence and to not overlook it or undervalue it in the bigger arc of an international criminal investigation, but to also pull back a little bit and to think about sort of the how to keep the survivor centered,
how to keep them in the driver's seat. I think one of the biggest challenges we've had is, you know, if rape is a crime that is characterized by a lack of consent of the survivor to what is done to them. The last thing we want is a lack of their own consent for how the information about their own experience is being utilized
for investigation or court processes. So how we keep them in the driver's seat and don't gamify some of these investigations, but make it a true partnership where it's appropriate to do so. So that will be coming up. I think the last couple of things that I'd want to mention is I did write with two colleagues, Anthony Galli and Simone Levine, a piece that was again informed by a number of different people. input on the role of consent in digital open source investigations
of sexual and gender -based violence. That came from a bunch of work that had been done by incredible researchers and investigators who would write about this topic and say, of course, we always need consent to use this material. I think one of the questions that I had as a practitioner on the side of we also run investigations in addition to doing kind of more the academic research was whose consent are we even talking about when
we're in the digital context? Is it the person who captured the information on their smartphone? Is it the person who posted it? Is it the bystanders who may be in the background? Is it just the victim? Is the victim alive or dead or can they even be located? What is the best way to reach out to them when All we may have is a handle or a digital profile, and we may not know who
was behind that or managing that profile. All of these, I think, were things that we didn't mean to make more complicated on purpose, but I think were very real questions that we were having as we were working on these things. And so I think one of the things that we ended up putting together was a typology of asking, as the open source investigator, what is the intended
use of this information? So for example, the consent we may need if we want to use this as crime -based evidence, meaning evidence that this kind of crime has taken place, may be very different than just getting leads to who may have been present at the time or who may have taken additional video or may be able to testify. Whose consent are we really talking about? Are we trying to use this almost more for intellectual property reasons to illustrate something that
we're putting out as an article? Or is this something that would actually come into a court of law to try to establish a fact at issue in a trial? And then, of course, the nature of the content. Is this direct evidence, indirect evidence, contextual
evidence, et cetera? I've probably been talking for way too long, so maybe the last thing I'll mention that will be coming out that I hope is helpful is my colleague, Lulik, and I are working on a book right now that will hopefully be out next year from University of California Press that is specifically on all the ways that digital open source information. can hopefully help to strengthen these investigations and prosecutions of some of these deeply underappreciated or sometimes
under -investigated crimes. Lynn, maybe I'll turn it back to you. Amazing. Thank you so much for running us through basically your entire career in 15 minutes. I really appreciate it. There's been a lot of people really liking some of the references that you've made. The Murad Code is actually pinned in our gender -based violence channel as well. So if you are interested in this topic and you want to dig deeper within the server, we actually have a space for that.
It's called hashtag gender -based violence misogyny, and it's linked in the chat. We've also got a few people very excited about your two -week publication and your book, which is great. I can see that people are typing questions as we speak. Please make sure that you're asking your questions in the chat. But first, I wanted to ask, Alexa, and this is a little bit more personal before we delve into the gender -based part. What really made you start looking into this
particular topic? Was it the gaps as you spoke about? Is that why you felt that this particular topic needed more research on it? I think part of it's a little, what is it? I think I enjoy the tricky gray zones where there doesn't seem to be much light being shown. And I do think it was sort of the unique confluence of working
at the Human Rights Center. I was a graduate student researcher on the sexual and gender -based violence, sexual and conflict, sexual violence and conflict program when it was started in 2010. And so I had had the incredible experience of getting to... Kim and her work in that area and going places like Kenya and hearing from lawyers,
from judges, from survivors. Particularly one of the things that we first started looking at was the 2007 -2008 elections in Kenya where there was unfortunately a tremendous amount of conflict that broke out and it was characterized by rampant sexual and gender -based violence. So hearing how that was playing out in various communities.
and just the strength of survivors in demanding better accountability for what they had experienced and identifying the gaps in the law where the law was not well situated to actually address
those challenges with illuminating. So one of the examples was that Kenya had a very progressive new set of laws to combat sexual and gender -based violence, but one provision at the time said that any woman, and they assumed that the survivor was a woman, not a man or a child, but any woman who reports or basically accuses a man of perpetrating sexual and gender -based violence. If she loses her case, she will be subject to the same punishment as the man would have been had he been convicted.
Well, given the fact that it is so difficult to win sexual and gender -based violence cases, the standards of proof are very high, the available evidence is often very thin, particularly if people didn't go right away to a medical facility or that information wasn't collected in a forensic manner, that is a huge bar to ask any survivor to go through. And so it was little things like really seeing the leadership from these Kenyan survivor communities to push back against the
way these laws have been formed. in ways that really didn't take into consideration their lived experience. So I became deeply dedicated to really thinking through some of these issues and what, as someone based in Berkeley, California, what we could be doing to survive these ground -up movements to make things better and to shine a light on these underlooked crimes. The second was... I was put almost like back in 2010, 2011 on the International Criminal Court related work
and that I became obsessed. We hosted a first workshop at the International Criminal Court in 2012 where we brought together people who were pioneering new uses of satellite imagery, people who were really thinking about the spread of the smartphone globally and how information could be captured. how information you capture on your smartphone, what tweaks needed to be made to make this have maximum weight in courtrooms.
So things like capturing the 360 -degree pan so it was easier to geolocate this information or chronolocate this data. It was little tips and tricks that we began to work with a number of different NGOs to try to disseminate out to activist communities and conflict areas. That became my whole bread and butter was the new technologies piece, looking at drones, looking at big data sets and data analytics, the rise of machine learning processes and how we could
harness this in more sensitive ways. Kim was running one program and I was running the other and they were very different cultures and communities. So her community of practice was almost all women led. The communities that I was part of, I was often the only woman in the room. I remember going to a meeting of legal investigators at one point, and there were probably 45, 40 to 50 people in the room. They were seated in an O with kind of the investigative team at the
front. And there was one woman, and I forget which country in Africa she was from, and there was me, and they put us at the top end of each U, and then the rest were all men around this giant circle. And I remember feeling like I was there to represent all kind of Western Northern women, and she was there to represent all women from the global South. And it was very, it was just a very different culture and climate. And then you take war and conflict investigations,
and you add on top of that the tech layer. And it just was a different way of thinking. And I think very much we were leaning into war, conflict, more visible aspects of that, particularly because social media and smartphones had really helped to begin to build this open source investigative space. So it was very visually oriented. And we're talking about sexual and gender -based
violence. Well, I had done a dissertation on the experiences of men in Guantanamo, many of whom unfortunately experienced sexual and gender -based violence themselves. But the thing that I remember from my interviews with them is that when issues of SGBV would come up, it would be very much looking down. It was much more subtle ways of communicating that this had happened. And it may be phrases like, I was tortured like
all the rest. Oftentimes when we talk about men, The terminology that's used is very different than it is for women. It's almost always identified as torture and not a sexual and gender -based violence, which is why the end of prosecution and investigation for them is even greater than for women oftentimes. And just, it's often little
physical things that you may not notice. There is a longtime war crimes investigator named Romeo Delaire, and he spent decades boots on the ground in places like the former Yugoslavia, I believe Rwanda, etc. He's written in his autobiography that it took him years to begin to even just recognize the subtle visual cues that come up
when SGDB has happened. So noticing the ways that someone's legs may be apart, and I'm sorry if that gets too graphic, or there may be broken off, you know, shotguns or broom handles nearby. He would see that someone had been killed, but he wouldn't pick up these kind of more subtle visual indicators that this may also be a bigger
set of crimes. So I became obsessed with how do we begin to bring those two worlds together in a way that can have greater impact understanding the gross under -prosecution of SGDV and knowing the deep dedication of this open source investigative community to bring the best that everyone can to these efforts to pursue justice. That is super interesting. Maybe towards the end of the talk, if you could just write Romeo's full name, because I had a little Google, but can't find him. And
I think that research is incredible. And I think that book would be something that a lot of people would like to read in the server. It's interesting you mentioned judiciary, and it's really interesting that globally, there's this problem of convicting gender -based violence crimes globally. It's not just one country problem, it's everywhere. Malarika asked, is there a similar project underway to help establish guidelines for judges in analyzing
different forms of evidence? You mentioned not wanting to devalue testimonial evidence by using too much digital OSINT evidence, but where else is this dialogue happening and is it codified anywhere? How does that work? Oh, it's such a great question. I think about eight or nine other NGOs got together a couple summers ago, and we have now just launched a judge's guide to digital
open source investigations. And the idea is that it would be kind of a way to help explain some of the basics of open source investigative methodologies like geolocation and chronolocation, ways this information can come in as crime -based evidence or as lead or linkage information. So that is available. It's now available in a number of different languages, I think like seven or eight languages. If anyone wants a copy, please feel free to reach out. We can get that to you. It's
all free. But it's not specific to this particular set of crimes. And I do think there's an extraordinary amount of work that still needs to go into strengthening every link in the chain of accountability. So what I deeply appreciate about the Bellingcat community and others like it, and kind of the broader ecosystem that we're all a part of, is that we put so much work into pioneering methodologies and thinking through how we can maximize the evidentiary value from the information that's
put up in public spaces online. But then how do we think about the ways that trial attorneys are going to use it in court? Oftentimes, if they don't have a facility with how to explain these methodologies, then they're likely to either scrap them or not be able to defend them well. I think defense attorneys really need a lot of support and even knowing what the weaknesses
are of this. My career has all been about pushing out these methods and trying to get them more mainstreamed, particularly in legal process. But it was probably about five or six years ago, triggered in part by the really phenomenal response to the Albert Folley explainer that was put out
by Bellingcat. It's still one that I use in every class that I teach because it's such a good overview of how satellite imagery can be triangulated with video, with other bits of information to try to help establish the who, what, when, where, why, and how. What something in me was just the broad -based, oh my gosh, Bellingcat has proven it. it's so clear now legally what's happened.
And that made me nervous as an attorney because as a lawyer, some of you have probably seen this piece, but there's one still from a video where there's a number of dark spots on the soil. Like as an attorney, just saying that these dark spots in the soil are likely blood or whatever is not going to be sufficient for actually establishing that this is in fact what you claim it to be. Like you would want soil samples, you would want testimonies. And so really helping Defense, understand
what questions to even ask. If our goal is to get to the truth and full transparency, we really need robust teams on either side to help bring that to the surface. And this is kind of betraying my Western lawyer tendencies to even say that in the first place. So I do think there's a powerful role of storytelling and journalism in this entire ecosystem to drive up attention to things that need to be paid attention to and to motivate lawyers to even pick up these cases in the first
place because people care. But then there's the legal evidentiary side where the standards are much higher and you're not necessarily telling. just one story, but you're having to really interrogate that story to hopefully make sure the right person goes behind bars and that this is in fact something where, you know, sometimes I think there's a lot of great research out there that says, as human beings, we over rely on what we can see
with our eyes. And we tend to fill in a lot of gaps in our understanding with assumptions that we're not even aware we're making. And there is a lot of research that suggests that there's not a one -to -one correlation between confidence and accuracy in analyzing visual material. And as open source investigators, I think it's really important that we understand those weaknesses and potential fallacies, that we can do what we can to potentially offset them if what we
really want is the truth. And I think I've drifted way too off stream. So that's a big piece. I do think one thing we would love is, as we're putting out this open source investigators guide to investigating sexual and gender -based violence, in addition to really hoping to crowdsource feedback about what works, what doesn't, where there's gaps that you wish could be filled for the final
version. Any citations to great resources that you think should have more attention from this broader community of practice, we would love to make sure that we're integrating that, understanding that I think even among kind of the global consultation that we've been doing, that's obviously going to be a biased perspective and there may be some great resources that we've overlooked. Somebody in the chat has already found Romeo Delaire for me, so thank you so much. They've already put
the links in there. Speaking of kind of recognizing science, Sarah, one of our lovely mods, has written, Hi Alexa, recently there have been articles about sexual violence occurring in the women's area of Muzanzi prison in Goma during the seizure of the city by the M23 rebels. There are a lot of open source videos of the interior of the prison. The articles which document this incident just use the traditional, there are reports,
versus an examination of the scene. Based on Hannah's articles, which by the way I've linked in the chat, I can identify some indicators. Burning the scene down, burned bodies, some bottles. I feel like there's enough to talk about in an article, but frankly I don't feel good about writing it. I feel I'm... not educated enough to say, yes, this is a good example. Do you know of any further resources or subject matter experts I could talk to to help me evaluate what I've
seen? Thanks. Such a good question also. Absolutely. I do think that one thing I really appreciate, Charlie, about your work and what you've all been leading on in Bellingcat is putting together a subgroup of people who are really interested in these particular issues that hopefully can be peer reviewers and resources for each other. Certainly, I think that if you wanted to reach out to us or to the Institute for International Criminal Investigations, we can always give insight
and input. One of my extraordinary colleagues, Andrea Richardson, she worked as a war crimes investigator for a long time and her specialized area was open source investigation. She had graduated from our program alongside Hia and really kind of deployed those digital open source skills to work with this international investigative body looking at sexual and gender based violence
crimes. But I think she's one of the best people that I can think of that really has a deep dedication to this intersection and can think that through. So yeah, I really, we then, and this is something I'd also love to bring back to the Institute for International Criminal Investigations. So they are working with us on developing these guides, but they also work and lead classes and trainings on sexual and gender -based violence. And they have an extraordinary roster. of experts
in that space. What I'd love to see is more coming together of digital open source investigators with those traditional investigators, with the gendered crime experts, to be able to even see if we could create kind of a sub like group of all of us to help each other grow and learn about how to make these investigations as robust as
possible. There's someone I had interviewed for Elix in my book who said, You know, one of the big challenges in the space, for example, is that when you learn a foreign language, you learn how to ask how to order food in a restaurant or where the bathroom is or how to find a particular location. But you're not taught the words for sex, certainly not the slang terms for sex. You might pick that up kind of colloquially. You're also just not taught the terms for parts of the
body. And so there's a lot of kind of a lack of that understanding of language and how it plays out. I'm hopeful that machine learning processes can help us identify these proxy indicators that you're talking about in ways that we may not have a tremendous degree of confidence, but maybe this is a great way for machines to sort of complement humans in helping us make sense of what actually we may have and fully mining that and helping us understand the reliability
and trustworthiness. I also think machine learning processes could help us begin to identify terminology, the way it's being used, the ways that subtle visual cues may be popping up that correlate with sexual and gender -based violence so that we just have an easier time as investigators not overlooking this. Definitely not alone in feeling like there's not enough to publish on
this. I have talked with major media publications who've been very nervous about even saying, For example, in a very famous conflict that SGVB has happened, they're very confident in saying hospitals have been attacked or schools have been attacked. Not that, because the indicators are often so subtle. So I do think getting a couple of these experts to say, yeah, these things are subtle, but this makes so much sense and
it's so common. And what we've seen repeatedly is that particularly in detention facilities, this is rampant. And here's what we know. I think that could be brought together. Yeah, the reason we wanted this talk and what we've been talking about quite a lot in the gender -based violence channel is the need for training, more training on this and reporting in general. And hopefully this discussion is very nice base layer for something that we can potentially pursue going forward.
Just linked to that slightly. Alpha says, so it's kind of touched on culture as a significant issue in pursuing justice in many SGBV cases. What are additional precautions we can take to bypass cultural clashes and the issues they can create so the evidence is compelling whoever
sees it? I think that with this topic and the subject area more than any others that I've personally investigated, it's It kind of calls on the best part of open source investigations, which is the collaborative component of it and the multidisciplinary
component of it. I do think ideally that each of us, if we are working on these issues, kind of begins to build a roster for ourselves and for our teams if we're working as part of a team of like who's a gender expert that we can talk to who if we're not from that community or that culture is someone that has knowledge of sexual and gender -based violence in that community and culture who can help us understand the risks of what we may be thinking of putting out that
also help us pick up the signals and the noise that we may not ourselves be detecting. I think having that one, you know, we're doing a lot right now in gender persecution in Iran. And one of the things that's been so invaluable is just having a couple of Iranian researchers on the team who had to leave the country, who deeply, I mean, And some of them just pick up things that would take me, like they can geolocate something in a matter of seconds because they're like,
oh yeah, I used to walk on that street. I mean, I think we've all had that experience when we're from outside a community, how long it may take versus someone who's familiar with that community. I think particularly with sexual and gender -based violence and the ways that gets wrapped into cultures and relationships. It's important. So ideally, I think you'd have a digital open source
investigative expert. You'd have someone who deeply understands the culture and context and someone who understands sexual and gender based violence and how it plays out. And ideally, that becomes your team or your circle of advisors as you're pulling this together. This community is so generous to each other that I do think that particularly being there for each other as we recognize that this area needs to be built out and built out very carefully and thoughtfully.
I think is kind of an exciting opportunity, but I do realize it's a lot of extra work to begin to build those networks. As the internet grows and we're seeing new forms of media from the arrival of short form videos, for example, the control of your image as a general person has become a lot harder. Even now, we seek credit approval to show people's images and videos and things like that. But we don't necessarily do the same thing for people who have been videoed,
as you mentioned, in these circumstances. You mentioned that you've done a lot of a bit of research on consent and the possibility of seeking consent. What has that research told you? And is there a future in any sense where we might be kind of contacting or reaching out to the survivor in question and asking them for consent to use a video that we possibly have analyzed.
Yeah, that's again a great question. You know, I think when we were doing the research for the consent piece, which honestly took four years because there was such little guidance on this topic, who you are and what you're doing this for I think really does matter as to whether and if so how you should reach out. So one of the places where I think we ultimately landed is that If you're doing this for legal accountability, are you the right person to be reaching out?
Or if you maybe think that you need consent from a survivor or someone who posted the video who may have been a victim in other ways themselves or in the same ways themselves. Should you reach out to someone in your network who is an expert in trauma -informed communication, who understands how survivors work in that community or have deep networks of NGOs? I think NGOs are often an overlooked resource, ones that help survivors, to basically even just call or contact one of
those NGOs and say, Here's the situation. I've got this video. I'd like to write on it. I'd like to get consent. What would be the appropriate way for me to reach out? Or can I reach out through you, depending on what your internal code of ethics for your organization or that you have developed for yourself may be? And really think about if you're the right person and have the
right training. I think that's one thing that we do discourage and the guide is reaching out directly without having someone, maybe you have trauma -informed training and maybe you're an expert in SGBV, in which case I think it's a different scenario. But if your predominant training and expertise is in digital open source investigations, who can I work with who may be the right person
to do this in an informed way? The other thing is I think, you know, being really aware that digital modes of communication can be deeply dangerous. So for example, on Twitter, it used to be back in 2016, 2017, you'd tweet at someone and say, hey, can I use your video? And then I think the norm began to shift. OK, do this
in a private space. Send them a DM instead. And then it began to be growing awareness, particularly when we're dealing with issues like this, since we don't really know who's behind an account. Even if it's a survivor saying that this is their account, it could be their husband or their partner or someone else who has control over them or their trafficker who actually has control, or maybe it is them, but someone else has access
to it as well. And we could be raising risks just by reaching out that we're not aware of. So then I think it became about encrypted modes of communication, but that of course takes, that can be difficult to ascertain how and when. So again, I think this is where I think digital investigations planning is also really, really
important. When we were building the Berkeley Protocol on digital open source investigations, I think the most valuable things in that, because they're just basically foundational principles. Everyone needs to develop their standard operating practices on top of it. They help the International Criminal Court do it, but we realize each team does that, has to do that for themselves, know what technologies and tools and processes they're using. But I think the most valuable things are
the four annexes in the back. And one thing I've really become a huge proponent of is at the outset of any investigation, whether it's because you stumbled on something you think is important to write about or because you're deliberately investigating something, is building out that digital investigations plan and really kind of thinking about who that network of supporters that you want to have around you will be in advance and letting them know, hey, I'm up to reach out
if you have a little bit of time. Invaluable advice. Yeah, I used to cover trafficking for BBC television and getting in touch with a trafficking charity who was very used to speaking to women who had experienced the sort of thing they were speaking to the women about prior to filming and prior to those interviews was incredibly important and really shaped the way that we approached that. conversation and led to much better testimony as well. So I really do encourage you to reach
out to NGOs. I think they're fantastic at being able to guide journalists in that space. And yeah, I wanted to ask this question by Inez because it was asked a while ago. Do you have any experience with covering a violence directed towards trans people and gay people, especially those in countries where their sole existence is illegal. Do you see any differences to this when it comes to
covering gender -based violence? One of the things that we recommend when we do trainings on doing digital open source investigations of sexual and gender -based violence is mapping out the legal situation in whatever country or context you're investigating. So, you know, even if you're from the country, sometimes some of the laws just are unclear or they're rapidly shifting.
So understanding what's illegal in that country, whether it's homosexual behavior has been criminalized, if there's a death penalty attached to it, if trans people are particularly under attack, legally and socially, I think Mapping that out so you're aware of the risks of the scope of your reporting, I think that can help you to design an investigation and reporting plan that's sensitive to these issues. We are, in fact, doing work right now
on looking at attacks on trans populations. I'd say the way that we have pivoted is to really try to be careful about whose story we feature. So we may, for example, have evidence of 30 different attacks. We will, of course, prioritize if we're going to do anything public facing. using materials from people who have said, I want my name attached to this. I want to be public. But we won't also, of course, just take that at face value. We'll
assess, are they out of the country? Are they in a country now that is much friendlier to the kinds of issues and challenges that they're facing and much more supportive? Do they have a support circle around them? How can we even know that? I do think there's a heightened care. And a lot of times it comes down to the level of abstraction by which we'll report about these things. So will we represent 30 different people with a
flower icon as opposed to their face? We'll really try and do that threat modeling and that risk modeling. Of course, if we have direct communication with that person, we'll also have a conversation with them. We just put out a report. It was co -authored by a colleague at the Atlantic Council and Andrea Richardson from our team on gender persecution in Iran. And we did a really careful assessment of all of the case studies that we
feature. We, of course, had digital open source information relevant to a vastly greater number of incidents. featured the ones that were already very publicly known, where the risks to people's families were less because the families have left the country. There was one incident where we had identified a number of attacks on women who were accused of wearing improper hijab. And we blurred their faces, of course, and we didn't
use their names in those instances. Their videos had gone viral, but they were not people who had come out publicly and said, we want greater amplification of our story. We did make the assessment that we could keep it enough obscured that we would not be doing additional harm. But we also went to a number of journalists who we knew had reported extensively on these issues on the ground in Iran and asked them, do you know where these survivors are that you reported on that we pulled
information about? Now, do you feel that they're at risk? And so we did a number of checks of this with people that we knew were from frontline communities towards the people that we work with. So I do think it just takes a heightened level of care, heightened level of planning. I would get peer review on any threat modeling before you put anything out to just see if there's anything
that you may be missing. And of course, if you're in contact with the survivors and they can review what you're going to put out before you do, I think this is one of those instances where I would really want their understanding. There's one time where I did an interview of someone and I have it recorded and on the record what this individual said they'd experienced. And when I put it in the book, I was so great in
my book graphic. I was so grateful that I had run the quotes by every person that I quoted, because when I sent her those quotes, she was like, I never would have said that. There's just no way. It's there recorded. I have the transcripts.
But from an ethical perspective, I scrapped that from the book because I realized what was probably happening was she realized that what she had shared, well, from the journalistic perspective, was something you want to put in the book because it just grabs attention that she had realized the risks were too great to have that piece of
her story known. So yeah, I got scrapped. Yeah, I can't count on my hand how many times that's happened to me where I've interviewed someone And then after, when going back with their comments, they've been a bit shocked by what they'd said to me. And that informed consent is then taken away. And it's really important that you respect
that decision. Talking about hostile environments, Sarah actually has to follow up question from her other one earlier and says, any tips for dealing with an audience who is actively hostile to the concept of sexual violence occurring, namely the side of the perpetrator? Often I hear things like, burn bodies are the result of munitions, for example. There are bottles all over and don't mean anything. Clothes come off when you drag
a body, for example. Sexual violence will always have subtle indications and how do you push back? Yeah. That is calm. It's fascinating in this area of practice to see the deep psychological resistance to acknowledging sexual and gender -based violence, even when it may be right in
front of you. I do think as an investigator, it's important to know what those pushbacks are, like those really powerful illustrations you just gave of like, you drag a body and the clothes come off, or the bodies are burned, you know, through ammunition, as opposed to deliberately
set on fire, et cetera. I think this is where I see such tremendous potential and power in bringing some of the sexual and gender -based violence experts together with this community of practice, because I think they have the years of experience, the decades of experience and evidence that burning of these bodies after sexual and gender -based violence is common, that rape in prisons is almost always happening in these detention facilities and kind of overcoming some
of the stereotypes and assumptions. And I think that's where we get in trouble as humans is we do have psychological resistance. At the same time, you are going to have to be able to defend your analysis. And sometimes it may be even getting ahead of those critiques and saying, here's what some people say, you know, here's what the pushback may be that bodies are dragged and clothes come
off. However, I spoke with these experts and what they have communicated to me is and being really and I think this is the power of transparency, which I knew this community in particular has really promulgated is being fully transparent about the potential critiques, the potential
limitations, the potential. flaws in analysis that could exist, but why you remain confident and really being for, and I think this is the lawyer training again of being able to counter those and almost counter them before they're even made. I used to train first year law students. And it was, you need to be able to make the other side's argument even better than they can if you really want your arguments to hold up in
a court of law. Because you need to have thought through what is the response to each of those. And sadly, I think this is one area where we really do need to be ahead of that curve and be anticipating all the pushback and continuing to push back. And I think this is where it takes a spine. And I know everyone here has one. I have never received more pushback than on this topic, whether it's from the SGBV investigators, the traditional investigators themselves, who
basically stay out at times. If you don't have specialized training, you shouldn't be touching this with a 10 -foot pole. to others who have seen so much of this and it's so underreported and undervalued that they're like we need to be doing everything we can to help people understand the full scope and scale and impact of this work. This is where peer groups become I do think invaluable in encountering and sharing resources. One more
thing on assumptions. One of the first projects I ever worked on we had just founded the investigations lab at Berkeley in 2016. And we were working with a team of lawyers and there were pieces including potential SGVV that they had been investigating in a detention complex in a particular country that I have to keep meaningless because the case
is still ongoing. But there was someone who'd smuggled a bunch of footage out of those detention facilities and fled to a third country and now had a hit man after him because the government was very unhappy with what this person smuggled out. So I flew to that third country. I brought the hard drive home. Our task at that time was, could we verify the videos that had been collected?
And so we did that. And at the end of this, we asked the partner organization, do you want us to see what else might be out there in digital open sources? And they said, you can try, but you're not going to find anything. Like, this person had to smuggle this footage out of the
country. Like, there's just no way. We found so much evidence on YouTube of similar patterns of sexual and gender -based violence and torture and other crimes that when we handed over all the digital open source information, what we were told was, oh my goodness, this may actually now no longer be a domestic -level crime, but an international crime against humanity because it's so widespread and systematic, or appears
to be. And I think that's something that we've just seen is so hard to know what we don't know as investigators. But I think the SGBV investigators, traditional investigators, often don't understand the full depth and breadth of potentially invaluable information that may be now posted to online spaces where people are communicating. And I think the online investigators may not understand the full range of subtle ways that you can detect that this is happening so that you can pick it
up as an investigator. I think we need these merged kind of research teams and find the right people who really have a passion for this to help grow this area of practice because I do think there is so much potential. Absolutely. I just wanted to highlight someone in the chat said thank you so much for this discussion. Just a reminder that this is being recorded so you
can listen back as well. We've mentioned a lot of amazing resources so please do take the time to listen back and investigate those said resources. Talking about resources, just to end off, and we always ask this question when we're talking about particularly traumatic events and covering traumatic events. What practices can organizations or newsrooms put in place to protect their employees or to protect researchers from vicarious trauma and other harms? And what do you do personally
when it gets too much? That was, again, some of my favorite questions. I think one of the most basic, and I'm guessing most of the people in this room do this already, is that when you're compiling information across large data sets, whether it's on a spreadsheet or on some other kind of particular platform, having a mechanism for flagging. For us, on every spreadsheet we do, it flags, yes, graphic, if it's graphic material,
and it makes that whole line red. so that you know that this is going to be stuff you don't want to be doing late at night when you're by yourself alone in your bedroom, but it's stuff that you may prioritize for early in the morning or may have someone else work on. The other thing that I really like to do is have a column next to it that just without graphic detail says what
kind of graphic violence it is. So if it's sexual and gender -based violence or it's decapitation or it's murder of children, the things that we know are big triggers for people. I like working as part of a team or even sometimes what we've done is created partnerships among people within teams. So there's two people who are always kind of watching for each other and can be saying, hey, you're suddenly not seeming like yourself.
Are you doing okay? Or I've noticed you've missed a lot of our meetings, or maybe even flagging for me or some other person who's a manager or supervisor to say, hey, I'm worried about so -and -so. They're just not doing their normal stuff. So that we can keep an eye out for each other. But for example, I have a colleague who used to be really affected by violence against children. And for me, it has always been gunshots.
And I was in what was a mass shooting. And there's this moment where you realize that this is guns and not backfiring of a car. That gets me every single time. So little things like I could take the children pieces. And she may take the active gunfire pieces. And we figure out how, as a partnership, we kind of protect each other, minimize the risks
and maximize the value of what we can do. I also realize I may not have the greatest objectivity, ones where I'm already being triggered from whatever traumatic experiences I've gone through in my lifetime. So thinking constructively about where your lived experience is an asset because you understand the culture or context or language or whatever, and where it may be a liability because it's too close and it is something that you may have a difficult time with the objectivity
of. I think there's three things that Andrea and I tried to point out in graphic that are become my touchstones. One is that awareness for myself and for others of what may be triggering and figuring out ways if there are ways to kind of mitigate some of those harms. Here's the tips and tricks. So things like, I never watch a video with the sound on the first time if I know something
awful happens in that video. Or actually, any time I'm watching a video and I don't know what's coming, I'll keep it really low or I'll keep it off, just because I know the auditory content is what gets to people most of the time. You know, in the gray scale, sometimes speeding it up or slowing it down. I've personally found all those tactics really helpful. And then the
third bucket is that community piece. I do think, you know, when working on really upsetting content, having a small community around you or doing it with someone else who understands what you've just looked at can debrief with you, or you don't talk about it all, but you know someone else has shared that experience. There is a camaraderie
and power in that. I remember doing investigations and research in northern Uganda where the Lord's Resistance Army had unfortunately taken a number of people and used particularly women and young girls as in slavery contexts. And we had curfews every night at 6 p .m. and we would basically give money to the driver and say, could you buy us like, you know, some beers? And we would sit and we went over a drink, but once we were locked
into our building at 6 p .m., we would kind of debrief what we'd seen and we'd have a beer or two and kind of talk through that. There's something about that boots on the ground camaraderie that we know protects military units, et cetera. We need to replicate that, I think, in this community. And that's, of course, an area on which Bill and Kat has truly led. Gosh, I could talk about this forever. And thank you so much for taking
the time out to unpack all of this. I'm sure we'll ask you back in a couple of months once we've had a few other speakers on this topic. But thank you again for taking the time and for answering all these really important questions. You know, thank you for having me and thank you to this whole space for your passion for this topic or your interest in this topic. We need more people who are deeply skilled. Really thinking through these issues and carve you a future space.
Amazing. All right, remember there is the hashtag gender based violence channel if you want to take this discussion further. But anyway, thank you everybody. I'm going to stop the recording now. Thank you for listening to the stage talk. If you'd like to catch a stage talk live where you can ask the guests questions, Join the Bellingcat Discord server by visiting www .discord .gg slash Bellingcat. The music you've heard is titled Dawn by Newer Self and is courtesy of Artlist.