Inside the Difficult, Dangerous Work of Tallying the ISIS Death Toll

Trying to calculate casualties in a war zone where counting means risking death.

ISIS militants posing outside of Damascus, Syria, in April.Balkis Press/Sipa USA/AP


The week after ISIS launched its catastrophic November attacks in Paris, the Institute for Economics and Peace (IEP) published its annual Global Terrorism Index.” It contained an unexpected finding: By killing 6,664 people in 2014, the Nigerian Islamist extremist group Boko Haram was the world’s deadliest terrorist group, beating out ISIS by nearly 600 victims. But the report, with its exact-sounding figures, raises a question: Is it really possible to know how many people ISIS has killed?

As in many conflicts, assessing the actual toll of the Syrian civil war is a difficult, potentially dangerous business. And when it comes to putting a precise number on ISIS’ death toll, researchers who track these grim statistics are skeptical. “My gut instinct is, we don’t know,” says Megan Price, the executive director of the Human Rights Data Analysis Group (HRDAG). “I don’t think those are knowable numbers.”

“What we know is just a part of what is going on,” says Bassam al-Ahmad, the spokesman for the Violations Documentation Center in Syria (VDC), a monitoring organization that gathers information from inside the country to maintain a running count of how many people have died in its nearly five-year-old conflict. Using a three-stage documentation process, the VDC has confirmed the deaths of 4,406 people at the hands of ISIS so far. But Ahmad says that is by no means the total number. “What is the percentage of what we know?” he asks. “Maybe around 50 percent.”

“They came to his house and abducted him,” says Ahmad. “Until now, we don’t know anything. We have sources who say he was beaten and tortured.”

And uncovering the information needed from ISIS territory presents even greater challenges. Media and NGO access is all but barred. In the group’s Syrian capital, Raqqa, internet cafés have been forced to close, and in other areas ISIS militants have put them under surveillance, making it difficult for activists to share information with the outside world. There are consequences for those who dig up information that the insurgent group doesn’t want publicized, like kidnappings, torture, and executions. In January 2014, ISIS militants kidnapped one of the VDC’s reporters in Raqqa. “They came to his house and abducted him,” says Ahmad. “Until now, we don’t know anything. We have sources who say he was beaten and tortured.”

That reporter, as well as four other VDC employees who disappeared from the Syrian city, Duma, in December 2013 and have not been heard from since, are not included in the group’s tally of the dead. The VDC does not know if they’re alive, but it also has no way to confirm their deaths. It’s not an uncommon problem. “We hear many stories about people who are kidnapped by ISIS and are executed. But sometimes you have no information about them. If you don’t have information about an incident, that doesn’t mean something hasn’t happened,” Ahmad says. The number of people who have simply vanished in Syria is in the tens of thousands. They are not included in the VDC’s count.

“Violence can be hidden,” says Price. “ISIS has its own agenda. Sometimes that agenda is served by making public things they’ve done, and I have to assume, sometimes it’s served by hiding things they’ve done.” For example, after Kurdish forces recaptured the northern Iraqi city of Sinjar from ISIS last month, as many as 16 previously unknown mass graves were uncovered.

In determining ISIS’ 2014 death toll at 6,073 people, the IEP drew its data from the National Consortium for the Study of Terrorism and Responses to Terrorism at the University of Maryland, which uses publicly available materials like news articles and legal documents to establish its estimates. It’s a common documentation method, but it can be limiting because more newsworthy events, like bombings with a high number of casualties, make it into the data, but under-the-radar deaths may not.

“Violence can be hidden,” says Price. “ISIS has its own agenda. Sometimes that agenda is served by making public things they’ve done… sometimes it’s served by hiding things.”

“Traumatic events that cause multiple deaths get very well communicated—and maybe even exaggerated—in their frequency and occurrence compared to more mundane, violent acts that result in death,” says Les Roberts, a professor of population and family health at Columbia University and a former Centers for Disease Control and Prevention epidemiologist who has led dozens of surveys on mortality in war zones, including in Iraq. (Read more about the controversy surrounding his 2004 estimate that 100,000 civilians were killed after the American invasion.) “A bomb is newsworthy. A bomb is big enough that the people in the morgue when you call them will say, ‘Oh yeah, we had four deaths in here,’ because they came in together and they create a psychological image, even though they had 42 dead bodies come in from a variety of other causes, mostly gunshot wounds.” In ISIS’ case, this may mean that public mass executions and suicide bombings get counted while other atrocities are overlooked.

Getting a grip on ISIS’ impact is just part of the larger struggle to get an accurate picture of the carnage in Syria. The VDC has confirmed approximately 200,000 casualties by name, photos, or videos. Their number is dwarfed by the Syrian Observatory for Human Rights’ most recent estimate of as many as 330,000 casualties. The Office of the UN High Commissioner for Human Rights stopped providing public enumerations of Syrian casualties after January 2014, saying it could no longer guarantee the accuracy of its source material.

According to Price, the HRDAG intends to publish new numbers in early 2016—nearly two years after the last public UN count. These figures will be based on not only cases gathered by the VDC and three other groups documenting deaths in Syria, but an extrapolated estimate of unreported deaths. Even if it’s the most accurate account to get presented, however, the new number will still likely be off. “Having looked at the data in this particular conflict, as well as several others, the safe answer always is: that number is too low,” Price says.

Roberts notes that it can be hard to collect reliable casualty data even in a country that’s at peace. “In the United States, which I think is a near-optimal environment for us investigating murders,” says Roberts, “our best estimate is that something in the ballpark of one-third of murders are never detected.” Sometimes a death is misclassified as an accident or a suicide, or a body is never found. Now imagine trying to get this information with ISIS breathing down your neck. “If in the United States we have a gross undercount of murders, how can we expect in the anarchy of Syria that it’s going to be even comparable?”