Paras Shah: On a dark night in October 2019, three men were traveling down a remote road in northwest Syria. Suddenly, they felt something hit their van. The men got out. They were confused and afraid. Then a volley of airstrikes hit them. Just a few hundred yards up that deserted stretch of road, the U.S. military was raiding the hideout of ISIS leader Abu Bakr al-Baghdadi. Baghdadi died and former President Donald Trump called the raid “impeccable.” The Defense Department says that no civilians were harmed in the operation.
But reporting from NPR determined that harm did occur. NPR says those men in the truck were civilians. Two of them were killed and a third lost his arm from U.S. airstrikes. The Defense Department disagrees and dismissed NPR’s claims as “not credible” based in part on a lack of metadata in images that NPR and Airwars, a watchdog group that monitors the civilian impact of military actions, provided.
Until now, the U.S. military appears to have never required metadata for images to be considered as evidence. If this new standard continues, it will be much harder for researchers and journalists to collect and submit evidence of civilian harm in U.S. military operations, which erodes accountability when harm occurs.
This is the Just Security Podcast. I’m your host, Paras Shah.
Joining the show to discuss the Baghdadi raid and the U.S. response to claims of civilian harm are Airwars Director Emily Tripp and Conflict Researcher Anna Zahn.
Hi, Emily. Hi, Anna. Welcome to the show.
Anna Zahn: Happy to join.
Paras: Emily, can you help us understand what's happening with the strike on Abu Bakr Al Baghdadi? Who was he?
Emily Tripp: Sure, so Baghdadi was the head of ISIS. So he was really the leader of the caliphate from the beginning. So back in 2013, he was kind of part of the group that were instrumental in setting up the Islamic State. He also became a kind of very important target for the US and allies. He's been compared, kind of, the similar level of importance to bin Laden, for example, for Obama, as a kind of key target.
So essentially, he was leading the Islamic State, and then after the defeat of the Islamic State as a territorial entity in March 2019, there was a big hunt to try and find him, which the US did. So in October 2019, there was a raid in northwest Syria, and Trump kind of declared a very extensive victory, and gave quite a few different details about the incident.
Paras: Right, and when the US military tells the story of that raid, they tell one version, a version in which no civilians were harmed, and when Syrians tell the story of that night in October, they tell a different story. And what did Air Wars and NPR learn from talking to local Syrians, especially a man named Barakat Barakat?
Emily: Yeah, so I mean, this is also something that we've seen in many different cases, is that there are these two versions of events. And this case was very similar. So NPR, actually, they managed to find Barack Barakat, as well as quite a few different family members of another two individuals who were driving with Barakat Barakat at the time. And their version of events is that they were on their way home from work, and that route took them through the village where Baghdadi was, unbeknownst to them.
There were kind of ensuing chaos, airstrikes, and as they were driving this van, essentially, their version of events is that they were then hit by US strikes. Barakat Barakat lost a hand and later an arm. And his two friends were killed.
Paras: So what is the US’s explanation for why they hit that vehicle? What did they say?
Emily: So they basically said that the van was driving with what they call hostile intent. So essentially, that the van had been driving a certain speed. There were some warning shots. And then the van didn't do the thing that they expected it to do.
Anna: Yeah, a bit about the morning shots. So it’s something that’s actually determined by NPR based on the time periods that were provided, and so from NPR’s analysis, the warning shots were carried out, and there was only a few seconds between the time that the warning shots were carried out and the time that the airstrikes began. So in kind of under their assumptions and their determination, there wasn't much time for the vehicle to react. And warning shots — a bit about warning shots that Air Wars has kind of looked into is the efficacy of warning shots and how they communicate to the targets what behavior to carry out.
For example, in this situation, the warning shots were carried out around the vehicle to tell the vehicle to stop driving but from you know, during a raid that's carried out late in the night, it's dark outside, the vehicle’s driving. It's really — from the perspective of the vehicle, it's really hard to understand what the expected behavior is supposed to be, whether they're supposed to stop, whether they’re supposed to turn, whether they're supposed to keep going. And in this case, they had very little time to make that determination.
Paras: Right, so the driver of that car has to make these split second decisions in a really quick time frame, in a really confusing situation. And the Pentagon conducted this civilian casualty assessment report, after the incident — what are those reports? And what did they find here?
Anna: Yeah, so, within the US military, there's different types of civilian casualty assessments that are done, depending on what type of detail they want to get into and how much evidence they have. In this case, they carried out a civilian casualty assessment report, and in this case, it involves a bit of investigating. The assessment goes through the evidence provided by both NPR and Airwars, and kind of goes line by line and determines the veracity and the credibility of the information provided.
One thing about these reports is they provide the reasoning that they have, and then they provide a final conclusion. It's hard for us as a reader of this report to know what part of the assessment was used to determine that the incident was not credible. But one thing that really stuck out to Airwars was the use of metadata. It was mentioned four times in the document. And each time the analysts referred to it, they cited, you know, videos that were provided by Air Wars and videos provided by NPR images. And they noted that there was no metadata alongside these images, even though there's a lot of images that basically do support the US version of the story. So in these details, the US has accepted that a vehicle was driving in the evening, and that they carried out strikes on the vehicle, and that’s what the images depict. However, they do cite that the metadata is not included in these images.
Paras: Interesting. So what is metadata?
Emily: So metadata is essentially information that's contained within the original copies of images. So it could be the location, the date, or the time. It's essentially any data that gives additional information other than what you see in that video or image.
Maybe to know, it's not inherently reliable. So it can be manipulated fairly easily, or just kind of incorrectly captured if the device wasn't picking up the right date or set to the right time, for example. Importantly, also, most social media platforms strip this data during uploads for quite understandable privacy reasons.
Paras: Right, so if I go outside right now, and I take a picture on my iPhone, that image would contain metadata — it would talk about the location and the timing of that photo. But if I upload it to Facebook or Instagram, then that metadata would be stripped away. So it would be harder to authenticate exactly when that photo was taken or where it was taken. Is that right?
Emily: Exactly. I mean, it would be harder to authenticate just based on that one image itself. But authentication takes a number of different forms, for example, you might be able to geo-locate that image and cross check that image with a number of other images that were taken at the time, which is kind of the common practice in an open source investigation. So I would kind of warn against using it as a kind of direct mechanism for authentication.
Paras: What are the implications of the US citing this lack of metadata in images that NPR provides? What does that tell us about the investigations process?
Emily: I mean, there were a couple of things. One is a kind of purely logistical concern. In order to get metadata, at scale, in a war zone is, is a very mad requirement, I would say. You have to essentially go to that original person, you have to find the person who captured that image, you have to ask them to send you that version of the image, and you have to ask them to send it kind of safely and properly and using the right processes. And then you yourself have to make sure that you can check all of that metadata and store all of that information. So it's a kind of a level of evidence, which is very, very high. And in war zones, particularly, that's just really not the reality.
So I think, on the kind of one hand, we have this logistical concern with actually getting that information. But then on the other, you have a kind of more conceptual concern. I think this is what really stood out to us at Air Wars, you know, we've captured thousands and thousands, of tweets, Facebook posts, where individuals have just kind of quickly captured the after effects of an airstrike or posted about certain, you know, cases that have happened of violence. And I think it's important to note that those cases, and the ways that local citizens document harm, is inherently spontaneous. I mean, really, if you think about the information environment in a conflict zone, these are often individuals capturing, you know, what might be some of the worst tragedies of their lives, and uploading them in order to hope that people pay attention and listen.
And so I think, for us, the concern with this kind of use of reference of metadata is — I mean, quite apart from the logistics, and the technicality of it — to us kind of signals a certain level of distance between the US investigators who are looking at civilian harm, and the reality of that information environment in which the harm is reported. And that's really worrying, especially given how extensive those allegations are, and how ostensibly the US wants to be the kind of leader in this space, when it comes to civilian protection and mitigation.
Paras: As far as we can tell, is this a one off instance of requiring this metadata? Or could this be part of a new trend in the way that the US assesses civilian harm?
Emily: I wish we knew. I wish we knew the kind of all of the processes that they assessed civilian harm so that we could talk about trends in that way. But really, the release of these types of assessments just doesn't happen unless the US government is pushed.
So in this case, the only reason we really know about this reference to metadata, for example, is because NPR sued for that assessment. There's other major cases led by Azmat Khan in the New York Times where they have also kind of spent years litigating for these documents. And when these documents are released, I mean, we're talking about thousands and thousands of pages of inconsistently processed, unpredictable redactions, which you know, really change and affect the readability of this work. We're actually in the middle of this project with some Canadian academics over at the University of Ottawa to try and figure out kind of how to even approach these documents to understand trends like that.
What we do know is that Air Wars referrals, the organization I work for, they actually account for kind of the majority of the US’s own civilian harm assessments. And about a quarter of those were relayed as credible by the US military. And I can say that, you know, there has never been a time where we have submitted metadata to the US military. So there have been cases that they have conceded as credible allegations, without this, ostensibly, without this requirement. We've also had quite a bit of back and forth with the US military over the course of the war against ISIS, where they would come back to us on these referrals and they would request additional information. Mostly they would ask for things like additional information on the accuracy of locations, because something that we do is kind of say, okay, here was an allegation of civilian harm, and it took place in this street or this neighborhood. And they might come back to us and ask, okay, but you know, exactly what corner of the street did this happen? So we had this kind of relationship with the US in that way. Not one of those requests that I could find going through all of our documents and records, not one of those requested information for metadata.
Maybe just a final point on an overall trend. While I'm not sure about this kind of use of metadata being something that may be a new thing, I think what I can say is a kind of clear, consistent trend is inconsistency. We've seen time and time again that civilian harm has been accepted or rejected, based on kind of an assessment criteria that is quite unclear to outside parties. So much so that the Department of Defense, actually, last year, created this new civilian harm action plan, and they themselves wrote that in the kind of opening section to the civilian harm assessment, part of this action plan that these assessment processes had been applied inconsistently across the department.
Paras: Has Barakat said what the US’s assessment and its finding of a lack of credibility mean to him? How do people on the ground in Syria perceive this?
Emily: Yeah, I mean, he gave a very moving interview to NPR, which I would recommend anyone listen to. We work a little with his lawyers, there's only a center who are representing him kind of tirelessly to try and secure compensation. I mean, if you read the reporting on Barakat Barakat, I mean, he's gone through so much. And I think, you know, the idea that there was a way that the US could kind of properly compensate for all that he has lost is, you know, that's never going to happen. And that's been quite widely acknowledged by him. But I think as you see the, you know, the needs on the ground, the processes, the lack of connection between the US military, even in their investigation process, and Barakat Barakat — it's clear that there's a sense of desperation, you know, and real disappointment. And we've seen that time and again, not just from him as an individual, but from kind of Syrians, Iraqis, all over the place who have been really affected by US actions as the failure to acknowledge that harm, is a real source of anger. And, and it is shameful, in many ways.
Paras: And as you mentioned, the US military is in the process of implementing the civilian harm mitigation and response action plan. And it's quite a mouthful, but what are your recommendations going forward? How should the military think about these types of civilian harm assessments in the context of that plan and generally?
Emily: Yeah, I mean, we've put together quite a few different recommendations for the Department of Defense on this. I think, for us, it's all about transparency and due process. The civilian harm action plan itself has a whole section on data management, for example, making sure that the information that they collect is really kind of reflective of that information environment I was discussing before, but also that built into that mechanism for collecting information around the civilian harm. But there's also some kind of tracking or record of how you then respond to those cases. I think that's been a real point of inconsistency across the Department of Defense and between different commands as well. You know, how the US military applies and tracks civilian harm in one part of the world might be very different from how another command structure would apply it in another part of the world. So, our recommendations will center on this kind of consistent approach, a transparent approach, and one that's kind of in line with the reality of the situation on the ground.
Anna: Yeah, and one thing I wanted to just kind of add in on this point is, even while this kind of process of reforming the military and their process with civilian harm is going on, there still continues to be problems with the transparency from the military. There's a pretty high profile strike in May where the US military claimed to be targeting an al Qaeda member in Syria, and there a lot of reporting came out that the person who was killed in the strike was actually a shepherd, a father of 11. And at this point, the US military has publicly stated that they've reopened the case, but there's yet to be any information come out about the victim, who they were targeting, whether they killed the target. And so at this point, organizations like Air Wars are still waiting to hear from the military, when what their result of the strike was and how they saw the strike occur. And at this current time, the last time that the US military, specifically CENTCOM, released a civilian harm report was in March of last year. And that report showed that there are still 37 open cases that have yet to be determined by the military at this point.
Paras: Is there anything else that we haven't discussed yet that you'd like to add?
Emily: I think I might just add, I mean, I think just to extend on what Anna was saying, you know, if, if we think of this case as something that has kind of happened in the past — you know, it took several months for the investigation to take place off the assessment to take place, it then took kind of, it's taken this investigation from NPR to kind of raise the profile again — I mean, this is a process that shouldn't happen, right? I mean, it shouldn't take an NPR investigation or an investigation from the New York Times to prompt the US military to look into these claims. So I think I would kind of just really want to emphasize that this process that we've seen so many times at the DoD reacting to high profile cases of civilian harm, I mean, when you look at it in the context of everything that we see Air Wars, for example, I mean, these cases are devastating, and, and concerning and very worrying, but they're part of a mass number of cases where, you know, the New York Times or NPR hasn't gone and done an investigation. Maybe it's only been, you know, one person who was killed, and maybe they weren't so high up in ISIS, or al Qaeda, or whatever it was. But yeah, those are still lives lost, that should be accounted for by the US military.
Anna: Yeah, and I think just to kind of go off that, to put it in context, The New York Times in their FOIA requests that resulted in the document release —within that release, there are over 1000 non credible cases. So those are also — they've had the same result concluded by DoD that they did not kill civilians. And at this current point, they're not eligible to be reopened unless new evidence comes out. And that's in spite of the fact that the entire CHMR-AP process is an acknowledgment that there have been failures and errors in previous civilian harm assessments.
Paras: There'll be a lot to watch in this area, especially as the US continues to conduct military operations all over the world. Emily, Anna, thanks so much for joining the show and for sharing your findings with us.
Emily: Thank you, thanks.
Anna: Thank you so much.
Paras: This episode was hosted by me, Paras Shah. It was edited and produced by Tiffany Chang, Michelle Eigenheer, and Clara Apt. Our theme song is “The Parade” by Hey Pluto.
Special thanks to Emily Tripp and Anna Zahn. You can read Emily and Anna’s analysis of the U.S. response to claims of civilian harm in the Baghdadi raid on our website. If you enjoyed this episode, please give us a five star rating on Apple Podcasts or wherever you listen.