Eliot Higgins on Bellingcat and taking on Russian spies
Read the transcript of our Science Focus Podcast with Eliot Higgins – scroll down to listen to the episode.
Daniel Bennett OK, so first of all, for someone who might not have heard of Bellingcat or perhaps they've heard the name of it mentioned in news reports, could you kind of explain what is Bellingcat?
Eliot Higgins So today we're a small nonprofit NGO. We have about 20 staff members. And what we do is something called online open source investigation. So that's using material that's available online from social media posts to satellite imagery on services like Google Maps to investigate all kinds of different incidents, from war crimes in Syria to Russian poisonings to stolen animals, to wildlife crimes, all kinds of different subjects.
And we've existed now since 2014 where it began. It was basically my blog then, and we had like £60,000 of crowdfunding, myself and some volunteers. And it's just kind of grown from there.
Daniel Bennett And so you you know, you say you're involved in investigations, but I mean, you've broken huge stories over the past years. I mean, this obviously people might be familiar with Flight MH 17. What are some of the kind of crucial pieces of evidence that you've been able to shine a light on over the last few years?
Eliot Higgins So our first big investigation would have been into MH 17, which is really where our kind of investigation team formed, which is a group of volunteers, most of which have become staff now. But first of all, we tracked the missile launcher that was believed to have shot it down through eastern Ukraine through separatist-held territory to the launch site where the missile was believed to be launched from and found evidence that that was the place it was launched from.
We then identified the same missile launcher in a convoy in Russia a few weeks earlier that had headed to the Ukrainian border. Then we started identifying individuals who were on phone calls published by the criminal investigation, the joint investigation team and by the Ukrainian security services who didn't have names. But we figured out who they were based off the contents of the calls, and they turned out to be Russian military officers and intelligence officers along with other people.
So that was kind of showing that Russia was involved with killing 298 people in this attack on this aircraft. Other big stories have been looking at the use of chemical weapons in Syria. We've identified the real identities of the people involved with the Skripal assassinations, as well as other people involved with that assassination, another Russian intelligence assassinations in Europe.
We've also identified Russia's secret nerve agent program through that investigation. And that then led us to the FSB domestic Russian intelligence team who tried to assassinate Navalny, the opposition leader in Russia, in August last year.
Eliot Higgins And that's led us to even more assassinations and attempted assassinations by the same FSB team, seemingly using the same nerve agents that were used in the Skripal poisonings and other poisonings in Europe. We also do other subject. It's not just Russia.
We've looked into things like border pushbacks by Frontex in the Mediterranean, which is now part of a EU investigation. We've published about the illegal wildlife trade in Dubai and a whole range of different subjects, in particular the far right in Europe and in the US.
More like this
Daniel Bennett It’s a jaw-dropping spectrum of investigations. But just to pull back for a minute. Although you’ve become experts in what you do, you’re not professional investigators are you? You’re not what people would traditionally think of when they think of an intelligence agency like the CIA or MI6.
Eliot Higgins Yeah, we're kind of just really keen amateurs who had hobbies get out of hand. So it's like the [00:04:22]Russia stuff – all [0.5s] these Russian poisoning [investigations] are basically the work of one person who's just really focused on this kind of evidence and working in that particular way on those stories.
I mean, our investigations, you know, we might be 20 people in our staff, but there might be one or two people working on each investigation. But also, we're part of a broader community that's both kind of community of experts – you know, journalists, people work in NGOs, maybe military and arms experts, chemical weapons experts – but also members of the public who we connect to through social media.
I'm a very kind of online person, so I always had that community around me. But when we started doing this work, because it's using open source information, anyone can join in and look at it. It also means that anyone can be part of the investigation because it's a very clear and transparent process of how you come to your conclusions.
So around Bellingcat's work, there's been this community growing that were often using, in investigations, kind of crowdsourcing answers to certain questions or helping them find certain material, or some people might figure out where something was filmed and then be able to explain it. Then we can kind of use that as part of the investigation. So for us, collaboration is really core to what we're doing, at every single level.
Daniel Bennett You’ve been involved in some incredibly sophisticated projects, but just to rewind a little, can you explain how you got involved in this, and at the core of that, explain what is open source investigation is?
Eliot Higgins So really going back to when I started this, before I started doing this kind of work, I was basically just, you know, I working in admin jobs. I had no specialisation in investigation or journalism or anything, really.
Eliot Higgins But I was spending a lot of time on the internet arguing with people about what was happening in the world. And at the time, in 2010, that was the Arab Spring. So in 2011, I was kind of on the Guardian Middle East liveblog arguing with people every day and looking for links and just being interested in what was happening. But something that came up was people would share videos and everyone would argue about whether they were real or not. But no one actually tried to figure out if they were genuine, if they were where they were being filmed.
So I had a video, it was in a place called Tiji in Libya, supposedly. The rebels said they had just captured the town and the video was basically a tank rolling down a road with two wide lanes of traffic, a mosque next to the road and various buildings. So I thought, well, maybe I can find this road and mosque on satellite imagery. So I went to Google Maps, found the town very easily. Tiji Libya, it was the first thing that came up.
And when I looked at the satellite map, it was very clear there was a major road running through the middle of it. So I zoomed in. It was two lanes of traffic. It was divided by a divider, just like the one in the video. And I followed it along, and there was a mosque on the road with a dome in a minaret that was identical to the one that I could see in the video.
And then I watched the video again. I started looking at smaller details like the walls that were around the mosque and the curve of the road and the utility poles that were visible. You could see them on the satellite imagery casting shadows so you could see where they were. And by comparing these smaller and smaller details, I could be more and more sure that this was exactly the location they claimed it was. So then I could kind of go back and win the internet argument about where this video was actually filmed.
So, it kind of started there, but I just found it really fascinating. You could do this. And I was just really interested in what was happening in Libya and just frustrated that the reporting was so focused on the kind of perspective of the journalist on the ground. Whilst there was so much information that was being shared online from a range of different sources, that was just being ignored because people felt they couldn't verify it, or they were like, well, this is a YouTube video, what does that tell us? It's not news report from a reporter on the ground.
But if you actually examine them and analyse them and put them into context with other it information and verified what you could see, it actually gave you a much more granular view of the conflict. So you can actually see where the front lines were, where the fighting was occurring on each day.
Eliot Higgins So I just kept doing it. And in early 2012, I decided to start a blog, not with any intent of it being anything more than a place where I could put my kind of thoughts and write stuff down.
Eliot Higgins But I also was seeing so many people who were using these videos already, basically being conspiracy theorists. They were turning up on these conspiracy websites, where there'd be some like some white guy in the background of the video, and they'd be saying there's a CIA agent who's working with Libyan rebels to overthrow Gadhafi. And usually it was like some journalists or something who had just wandered into shot.
But I wanted to write about what I could see not what my opinions were. So I started writing about videos from Syria showing the weapons that were being used. Because I didn't speak Arabic, there was no point in me listening to what they were saying because I wouldn't know. But I could see the weapons.
And then I used online resources to identify the weapons and I posted about that. And then people were interested in weapons starting to talk to me. People from NGOs. When I post about cluster bombs, Human Rights Watch started asking questions like, where did you get these videos from? Can you find more of them?
Then it just kind of slowly progressed step by step until 2013 when I came across weapons that I'd never seen before from the former Yugoslavia. And I was in contact with New York Times journalist. I shared it with them and they went off and said they'd spoken to US officials and they're saying this is part of a secret Saudi smuggling operation that I had stumbled across on YouTube videos posted by the rebels themselves.
And that led to me getting a lot more kind of attention in the media because this was such a new thing that someone could use YouTube videos to expose the Saudi arms smuggling operation. And it just kind of grew and grew from there. And then in 2014, that's when Bellingcat was launched.
Daniel Bennett So when did you quit your day job, and what convinced you that it was time to take this full time?
Eliot Higgins Well, in 2012, I had this blog called the Brown Moses blog, which is named after Frank Zappa song. And I've been using that name as a pseudonym online. Then when in 2013, I published the story about the arms, I got loads of media attention. The Guardian, interviewed me first and then I had like CNN come along talking about the stay at home Mr Mom who was at home looking after his kid and finding weapons in the conflict in Syria, which is not [inaudible] I enjoyed. But nevertheless, I was still working full time then.
But I was just coming to a period where my basically the company I was working for was having redundancies and I was kind of the chopping block. And I had a company approached me – it was like a business intelligence company finding out if your oil workers can be attacked by al-Qaida and that kind of thing – and they said, we'd like you to work for us. And they offered a fairly decent wage, more than I was being paid before. But they said you have to stop doing your blog.
And at that time, I was getting more and more kind of media attention and I thought, well, I need to pay my mortgage, so I'm just going to tell Twitter I can't do it. And I said that I've got to stop doing what I'm doing, but then lots of people said why don't you try crowdfunding? So I thought that was a good idea, which it probably wasn't. But I did it anyway. And it was you know, I scraped like £12,000 together for that crowdfunding, and that allowed me to start working on it, like, all full time.
And I then started meeting more and more activists who were like really like blown away with what you could do with this work. People from NGOs being asked to speak at events about my work and then that led me to the 21 August 2013 Sarin attacks, where I kind of found myself in the position of having way more information about what happened than anyone else, because I just was watching the YouTube videos and figuring out where all the rockets had landed and the munitions that were used.
And I recognised the rockets as being used previously by the Syrian government. And this kind of came into sharp focus when Seymour Hersh did an article for the London Review of Books where he basically said it was a false flag, it was Jihadi rebels using Sarin from Turkey to do a false flag to draw the US into conflict. And I was looking at these videos, saying that's clearly just complete rubbish. So I wrote about that.
Seymour Hersh was not very happy, but a lot of journalists saw that as a kind of clash between old journalism versus new journalism. But for me, it was never really about being against something, but having a new way to investigate things that could complement traditional forms of journalism. And then I just kind of got more and more well known.
And then in July 2014, I crowdfunded a launch of a new website Bellingcat, where it would give people a place they could publish articles using open source investigations, but also offered resources to people to learn how to do it. And yeah, and then three days later, Malaysian Airlines Flight MH 17 was shot down. And that became our first really kind of huge story and a massive catalyst for Bellingcat and open source investigation in general.
Daniel Bennett There’s a ton there that I just want to tease apart. So first off I just one to pick up on one thing that probably get asked a lot. Bellingcat – where does that name come from?
Eliot Higgins It's from Belling the Cat, which is a fable about a group of mice who are very scared of a large cat, and they come up with the idea of putting a bell around its neck, but they don't have a plan to put the bell on the cat's neck. So we're kind of teaching people how to bell the cat.
Daniel Bennett And you know, you gave us short version there of the full sequence of events that are in the book. And I just want to clarify that everything you were able to do – like identify the missiles and the location they were sent from, wasn’t through any kind of specially trained hacking or computers skills. These discoveries were made via information that was out there on the internet. You just had to find it.
Eliot Higgins It was a combination of things. I mean, initially when [MH 17] was shot down, these videos and photographs emerged online – basically just a through loads of people searching for videos and photographs related to it – of BUK missile launchers.
Eliot Higgins So you had this kind of online Twitter community of people who just want to find stuff about it like you do about any event nowadays. And they showed this BUK missile launcher. But the question was, where were these photographs and videos taken? So one example is there's a picture of this missile launcher on a low loader going through a town, a photograph taken from like a garage forecourt. And a guy called Arik Toler came to me saying, I think I know where it is.
Eliot Higgins And he explained he has used the shop sign in the background in Russian, basically Googled it and the name of towns in eastern Ukraine. He then found a match, which was a court document where there'd been a fight in the shop, which gave the full address, which then pointed him to this location, but also discovered a dashboard camera video someone had uploaded online of them driving around eastern Ukraine past this same location.
So you then not only have the satellite imagery, but actual video footage from the ground. Just because someone had a hobby of filming stuff on their dashboard camera, putting it on YouTube and listing the streets they'd gone down. So we are then able to find video footage of the same location.
Eliot Higgins And when we published that journalists who were on the ground saw that and actually went to the same location and interviewed the locals who confirmed that there was a missile launcher there and someone even took a photograph from exactly the same spot, retracing the scene.
So that's always been an interesting interplay now between what we're doing and what people on the ground are able to do with the information, because we're so transparent and we're using open source evidence, people can look at what we're doing and saying, oh, actually, I might have a go looking into this bit of it myself and expand on our work.
And that's kind of always be the nature of what we're doing with Bellingcat, because I understand that it's often it's about networks of people and it's not just about connecting to people to work together, but also putting the information out there that if someone sees what you're doing, they're able to pick it up and do something themselves with it.
And they might do that publicly. They might do it privately. But in a way, it keeps that information kind of alive after you've published it. It's always about what can be done with that information, not what we've done with it.
Daniel Bennett And so I'm curious because as we talk about it, you know, you talk about all the big findings in the big moments, but I'm wondering what it's like day to day, as somebody doing this investigation, you know? Is it, are we talking about days and days where you're just, you know, sifting through footage and you're looking for something specific?
What is it like if you're one of these, you know, one of your contributors or even yourself in the early days where you've got a puzzle that you're trying to crack? I mean, is it the glamorous idea that you find this and you find this and, you know, like a movie, by night time the mystery solved? What's that process like?
Eliot Higgins I mean, it's often like very kind of intensive and it takes a long time. You're constantly kind of digging through material, but you're kind of looking for those kind of eureka moments where you find the one thing that he matches and that might take you looking for like 1,000 kind of, you know, Facebook pages or social media profiles or photographs or videos just endlessly digging through stuff.
So it can be extremely time consuming. But on the other hand, you know, it is extremely rewarding when you find those bits of information you need that kind of help you make your cases. It's just kind of digging through kind of the internet haystack, looking for needles.
Daniel Bennett So someone listening to this might think they might want to try their hand at a bit of open source investigation. But what are the risks of that?
Eliot Higgins Well, I mean, for one thing, when we're kind of crowdsourcing stuff, we do use that as a technique. But we've got a big audience of people who like to investigate, but we want to give them simple tasks, because when you crowdsource an investigation, if it's complicated, you end up having it kind of groupthink happening.
It happened with the Boston Marathon bombing where they actually had a group of people on Reddit, hundreds of people who identified the wrong person. But because there was this kind of groupthink about what was valuable and what wasn't, they went down the wrong path.
You're seeing the same thing happen with 6 January and identifying suspects there, because there's been several individuals misidentified by groups who are 100 per cent sure they found the right person. It's that same groupthink again. But if you give people a simple task, for example, when we've done the Europol 'Trace an Object - Stop Child Abuse' campaign where people are being asked to identify individual objects taken from abuse imagery like a bottle of shampoo or a bag, that's a lot easier because it's a simple task. It's like, do you know what this object is, yes or no? So that's kind of better to do. So you kind of have to use that kind of audience in the right way.
You also, you know, if you're dealing with footage from conflict, you will see horrific stuff and vicarious trauma can be a big issue. So when we're working with that, we have to be kind of very aware of that. And, you know, I would strongly discourage people from jumping into an investigation of a war crime, because if you're not prepared to work with that kind of material, you see stuff that will stay with you for a long time.
So you also have to kind of educate people around those kind of dangers as well. And also for people, sometimes people can, you know, who aren't good investigators will just kind of build up a kind of house of cards of mistakes and, you know, come to really unsolid conclusions. And that kind of also leads into kind of the, you know, other communities investigate stuff online. Conspiracy communities.
The way they operate is they if they come to a conclusion, they find evidence that contradicts what they're saying because of their kind of almost cult like addiction to it, they'll find a reason to dismiss it or run another experiment that shows that the last experiment wasn't accurate. There's a clear example of that is in the film Behind the Curve of how the Flat Earthers were at the end of the film.
They do an experiment, prove themselves wrong, and then really complicate the experiment because they can't accept the results. And then they say, oh, well, it's too complicated to find an accurate result. Therefore, we're still right. So you've kind of got to you know, use the kind of resources as a community very carefully. You've got to make sure people aren't being exposed to stuff that can be damaging to them.
Daniel Bennett You know, I was wondering, do you see it that way? Do you see that? You know, there's a lot of there's a lot right now, even we report on it, a lot of talk about the failings of the internet. And one of those is obviously the conspiracy theories that come out of it. I'm going to name one: QAnon, where people are basically, I think, manipulating audiences with a with a set of dots and saying 'Here's some things' and then letting the public at large join them themselves and form their own conspiracies. Do you see yourselves as almost the opposite of that, in a way?
Eliot Higgins In a sense, yeah. Because I think what there is, though, there's a fundamental distrust in traditional sources of authority. Lots of people have that in the media, the government, medical professionals. Now, if you have that, you go online and you're looking for alternative kind of source of authority, there are groups who will give you that. Now, those groups may be ones who focus on conspiracy theories. You might find like the alternative health community.
That's not to say these people are all conspiracy theorists, but it's kind of the first step onto finding more people who have that way of thinking. If you're, say, someone who is particularly against war or conflict, you'll find communities online who are saying that Assad has never done any chemical weapons attacks in his life or actually Ukraine shot down MH 17. But they're all basically the same.
If it's the Earth being flat, if it's QAnon being real, if it's coronavirus being Bill Gates's conspiracy to put microchips into people, if it's MH 17 being shot down by Ukraine, if it's chemical weapons attacks didn't happen in Syria, there's that fundamental distrust of some form of authority that they then reject. They then get drawn into communities that reinforce those opinions. They have websites, bloggers, personalities, podcasts, who will tell them that, actually, you're right and everyone outside of our community is being deceived and they're wrong. We're the only people who know the truth.
So they start building up almost this heroic sense of themselves that actually we're the ones who know the truth and the people outside it are either poor, misguided fools or they're part of the conspiracy. And so they start becoming detached from reality in these bubbles and once they're there, it's very hard to reach them.
So I think what we do have to do more as a kind of society and community is look at ways how we actually engage people who are looking for alternative sources of authority in developing their own sources of authority through evidence-based investigation, which is why we do so much training that we're looking at kind of training people who are more of a school age, university age to do investigations.
You know, tell them that actually you're not powerless. You can actually do stuff with a laptop, because if I can, say, a laptop and expose Russian spies left, right and center, anyone can do this. There's nothing special about what we've done. There's no special spy tool that we've used. I've used Google Earth and Google searches and Google stuff, but YouTube, Twitter, Facebook, this is kind of our tools. It's not kind of some super James Bond spy machine that we're using.
Daniel Bennett And that's interesting because, you know, when I was reading the story and the story of how you started out, it sort of resonated really quite a lot, the kind of where you started as an avid gamer. And you were obviously doing something I was doing at the same time, which was spending long, long evenings playing online video games with groups of people and having a great time.
But, you know, there it reminded me of the thing that we both shared there, which is kind of solving puzzles with groups of people and figuring out how to crack a problem. Is that something you found is similar in the people you've worked with since in Bellingcat? Is there a sort of, a type of person, I suppose, that is drawn to this kind of stuff?
Eliot Higgins I think so. I mean, if you're the kind of person who can kind of find four hours in an evening to raid a World of Warcraft raid, and you keep killing the same monster over and over again until you get it right, that same kind of obsession, I would say, is something that transfers very well over into investigations because you need to, you know, sit down and do stuff that is often very kind of relentless and, you know, tough and tiring and be prepared to do it.
But be, you know, have that reward at the end of it, finding what you need. It's a bit more open ended because there's no kind of monsters to slay or anything like that. But I think you still need that kind of level of obsessiveness.
And I think as well, being part of those kind of online communities, you know, not just the game communities, but, you know, these early online communities coming from the something awful forums, which kind of spawned 4chan and 8chan and all these terrible things. You do have a better sense of how the internet works as well, because I deal along with, kind of, we do a lot of disinformation in our work because we do get approached by kind of policy makers who are like saying, oh, can you talk to us about disinformation.
It's so much focused on the idea that this kind of stuff comes from the outside, it's coming from Russia or being confused by Russia, but really it's kind of generated by these online communities. And they don't get that because they haven't spent their lives on the internet.
They've spent their lives being serious people, doing serious things. I've spent my time kind of posting memes on internet forums and trying to think of funny tweets. But if you're part of those communities and you come from there, you actually have a much better understanding of where this is coming from, why you have Q, why you have kind of all these kind of far right groups appearing on places like 4chan.
All this stuff is kind of second nature to you. But if you come from outside of that, you're trying to understand the problems. It just doesn't seem to make any sense. So you start thinking it must be coming from the outside because what these people believe is completely mad. So it's really hard to explain that to policymakers as well. Telling them these Q people, they actually believe this stuff.
They think it's real and they think you're the bad guy because they can't fix the problem otherwise, because then they think, oh, we'll put some more money into countering disinformation with fact checking websites. It's like, no, you'd need to get it at a much earlier age in a more systematic way, where you can teach people, 16 to 18-year-olds, how to basically investigate stuff and you can frame that as journalism.
But really, investigation isn't something that belongs to one particular field or another, and it doesn't belong to experts. It belongs to all of us and kind of open source investigation particularly enables that because the evidence is so transparent, we aren't relying on sources telling us stuff. We're finding videos, photographs and other evidence that can make a very clear case.
You can use it to present a case. And if you can equip people with those skills, then they won't go off finding these alternative kind of ecosystems where they're finding about how actually coronaviruses made up or that, you know, there is no chemical weapons attacks in Syria.
Daniel Bennett Yeah, and that's another I think another point I'd like to touch on, which is, because I suppose as you're native to the internet, you've been around these communities and forums for a long time. I think that comes out in the way that, you know, Bellingcat is so distinctly apolitical or almost not just apolitical, but, you know, you never make assertions or insinuations or anything. You just about shining a light on the evidence. And I think that's bred out of where you've come from. Is that fair to say?
Eliot Higgins Yeah. I mean, everything we do is based on firm evidence. We lay it out and we explain how we came to our conclusions and we try and be quite cautious about what we're saying as well. I mean, even though we do make some quite big stories, it's always based off evidence. I mean, we're always trying to be transparent of sources, even when we've done the Russian investigations, which were actually used basically black market data that is very available in Russia.
And you can kind of find a Web forum where someone's selling someone's phone records to anyone in Russia, including FSB officers, which we found. And you can gather information quite easily. But we explained how we did that in our website. And what that meant is other people actually went off and bought that material to make sure we were actually telling the truth because we were being accused of being MI6 and MI5, CIA, you know, the whole bunch of them, by the Russian government, but what Russian journalists did is they just brought the same information we did and said, no, this is actually just available.
And if you find the right internet forum, you can buy it. But we also don't even trust that by itself because it's not an open source. So we'll cross-reference those data points we find from those data sets against other independent data sources. So we have like two or three points of verification for each claim that we're making. So even when it's not open source, we're being as careful as we can to verify everything that we're using.
Daniel Bennett Most recently you hit the news for tracking down the assassins who had tried to poison opposition leader Aleksei Novalny. And then had Navalny call his own assassins! How do you deal with the idea that you’re potentially annoying some quite traditionally powerful people? Does it worry you, do you worry when you’ve got a delivery at the door?
Eliot Higgins Well, I have to be more cautious now. I mean, we've obviously exposed Russian spies, which I'll go to in a second. But like now when I'm travelling, I won't eat food in hotels. Like, I won't have room service because I just don't know where it's come from.
At one point I was saying a hotel I've been staying at for quite a number of years. And there was a knock on the door quite late at night, at eight o'clock, what could this be? And the door opened and there was a guy in a suit with a name badge on say, Oh, I'm the manager tonight. I would like to thank you for staying 10 times at our hotel. We'd like to give you some cookies and these sweets as a thank you. And I was like, OK, I took them in and thought, great, free cookies.
But then I thought I have no idea who that person was, you know, he was wearing a name badge and a suit. You can get name badge anywhere. I started getting suspicious. And in the end I left the cookies and sweets in the bin of that room and left the next morning. And then when I was leaving, another manager came up, said, Oh, we hope you enjoyed the cookies and stuff we gave you last night. And I was like, oh no, I've already checked out.
But with the Navalny stuff, I mean, that really started when we exposed the Skripal assassins because they turned up on Russia Today saying they were sports nutrition salesmen. But using this kind of Russian data market, we'd already acquired their passport registration files, which are like stamped with the number of the Russian MOD and all kinds of really suspicious stuff on it.
So we knew there was something going on there. But based off that, we discovered they actually use a systematic way of creating their identities. They have the same first name, the same place of birth and the same birth date. And we have access to all these leaked Russian databases that had just been floating around for years. So we kind of use that to search these databases with those three things. Found the list of a dozen potential people, found eleven of social media, and then that left us with one or two guys who didn't have those profiles, got their registration forms for passports.
And it was the same guys, but it was their real identities. We looked into them and they were both serving GRU officers. That then les us to a third suspect who was involved with another poisoning with eight other GRU officers of an arms dealer in 2015, and that was another nerve agent poisoning. But at the time, the local authorities dismissed it as food poisoning and didn't investigate it properly. But once we discovered this link, they did investigate properly and found all sorts of CCTV footage of some guy walking up to his car the day before his poisoning, fiddling around.
So that then led us to those guys' phone records. And we discovered they've been calling up scientists who worked at a lab that supposedly manufactures sports nutrition drinks. But they didn't have a background in sports nutrition. They had a background in Novichok manufacturing. And it was a group of these scientists with this kind of Novichok experience. So we have that. We have Russian secret chemical weapons program. Then in 2020, when Navalny was poisoned, we checked the phone records of these same people.
They've been calling FSB officers who themselves had been following Navalny for forty times since 2017, including the day he was poisoned and we had phone records of them, travel records, all this leaked information. So we found that figured out they poisoned Navalny. We then had Navalny call one of them up and tricked him into confessing everything in a 50 minute phone conversations by pretending to be his like his boss's boss's assistant. So we did that. That was a big story. And then we discovered at least four more poisonings that we've published about so far, three of which were successful, one that failed.
One was opposition leader. And he was one of Boris Nemtsov's allies. He'd been poisoned twice into a coma in 2015 and 2017. And the team had been following just before his poisoning. We have three people who are successfully murdered, including two quite minor activists in the Caucasus regions who had been followed by the team just before they died, a member of the official opposition.
And we still have four more cases we're still working on. And we think that's just the tip of the iceberg. It's just slow to work for the material. So there could be loads more of these killings. So we kind of, the short version is we found Russia's secret assassination programme that uses a secret nerve agent program.
Daniel Bennett Right. OK, so I want to then quickly just move on to sort of what's next. In particular in the book, you talk about a couple of really interesting developments in your area. One is Mnemonic. Can you just tell tell us a little bit about that and what you're building there?
Eliot Higgins Yes. So we've been working with a number of organisations over the years on tech projects, and we have been working with a group called the Syrian Archive who renamed themselves Mnemonics Labs, who were collecting videos from the conflict in Syria. And they've collected over a million videos along with other contents. And that's a vast amount of information. But what we want to do is turn it into a kind of more organised data, because these are things like YouTube videos that just have like a YouTube description of when they were uploaded.
But we don't have metadata like geolocation information. So what we'd like to do is build a volunteer section, and that's what we're working towards at the moment, where it would be possible to actually put some of those videos from that archive to our volunteers and have them geolocate them. There's also other technology that's being developed that allows videos to be grouped together by similarity.
So if a building is filmed from one direction and then appears in another video, they're connected. So you can actually have kind of networks of videos that are physically related to each other in physical space given to the group of volunteers to geolocate. And because they are similar locations, they actually have loads more material then they can use to locate this one more building.
So you can have 50 videos of the same building that can kind of almost be geolocated in a batch. And when you're dealing with many videos, it's a lot quicker to do that than go video by video. So our hope is that we're able then to create a kind of data-rich data set of these videos, which would allow you to basically set a geofenced area on a map, circle a town, set a date, and we'll show you all the videos that have been tagged there on that date with their geolocation data we're providing.
For researchers into the conflicts, human rights, justice and accountability, it could be a very useful discovery tool for people who don't want to have to sort through a million random videos.
Daniel Bennett And and you talk as well about, I suppose, one of the, it's not a threat, but one of the things that, you know, could be a problem in the future is is trying to find ways to save all this data and keep it somewhere, particularly in terms of, you know, you talk about the volume. There's a million videos collected from Syria. So is that a way of sort of preserving the history of this so that that history can be hopefully written in the most accurate way?
Eliot Higgins You know, because this is a massive archive of information and we can't assume that YouTube will keep these videos online forever because we know they've deleted videos from the conflict before. A few years ago, they deleted hundreds of thousands of videos because they started using a new algorithm to detect kind of violent content and jihadi videos. And it picked up loads of videos from Syria.
A lot of them were incorrect, you know, false positives. And these channels got three strikes because they had like 50,000 videos on their channels examined. There would be like 3 that kind of triggered the system and then reviewed by someone who couldn't tell the difference between a jihadi and kind of like a normal Syrian rebel group, because you need contextual information. And, you know, some people, they just don't know the difference, unfortunately. And then it triggers it. The channel gets banned.
And that's, you know, 50,000 videos from the conflict in Syria just gone in a blink of an eye. I mean, in some cases, they would get banned and we'd work with the Syrian archive and talking to YouTube about this, they'd restore them. And then like a day later, they'd be banned again when the algorithm found three more videos of it. There are some channels that were banned several times, even though we had them restored and they were always banned for the same reason.
I had my own personal channel banned. And that included one video that wasn't even listed that just and that included, like all my playlists of like literally thousands of videos sorted by weapon type and all these details just gone in the blink of the eye. And lots of people have been using those for research. So I got it restored. But it was you know, it was still wasn't easy to do.
And there's plenty of people, you know, Syrians who are recording these videos, some of whom, you know, are dead and don't have any way to, you know, look after their accounts in these circumstances. And those videos could be lost forever. So that really showed to everyone, I think, how important archiving this stuff off these platforms is for the future of kind of understanding, conflict and analysis.
Daniel Bennett It's an incredible story from where you were just over a decade ago to where you find yourself now. What would you say to someone who is perhaps where you were 10 years ago, who is in a job that they didn't like, and had this passion or hobby? What would you say to yourself even? What has it kind of about passion and, you know, doing these kinds of things?
Eliot Higgins Find something that interests you, you know, give it a go, because I just started with kind of no idea what I was actually doing and I figured this stuff out for myself. But now you have loads of resources online and you don't have to always do a huge investigation and it doesn't always have to be something a million people read. It can be just, you know, do it for yourself, write something because you're interested in it, and you will learn more and think of it a way to build your skills.
You don't have to start geolocating 10,000 videos and figure out who killed who. You can just take one video and say, can I figure out where it was filmed? And just that process itself and writing that up as well. Make a blog. People don't have to read it, but just give you a chance to kind of have that process, it makes you think about the process itself and how you explain that process to other people.
And always be careful not to make leaps of logic. Only write about what you can definitely see and definitely say not what you think you're seeing. And then, you know, you'll be a lot more accurate and you'll be producing useful information that other people might come across and start using themselves in their own work. You can be part of a community that's doing that, and I think that can be very positive to anyone who wants to try and do it.
Let us know what you think of the episode with a review or a comment wherever you listen to your podcasts.
Listen to more episodes of the Science Focus Podcast:
- Dr Julia Shaw: Why do we do bad things?
- Marcel Danesi: Why do we want to believe lies?
- Project Discovery: Could computer games help find a cure for COVID-19?
- Chris Lintott: Can members of the public do real science?
- Lara Martin: Meet the computer scientist teaching an AI to play Dungeons and Dragons
- Rana el Kaliouby: What if computers could read our emotions?