« An Open Letter to Our… | Home | Syria-sly? or: the Me… »

Syria and the 100,000 Dead in the Civil War

If we had to retitle On V, I would probably call it, "The Skeptical Soldier". If you gather all of our posts, from Clausewitz to memoirs to intel is evidence to Iran to, most recently, data, constant skepticism about the conventional wisdom of the U.S. military ties them all together.

This applies to the media coverage of America's wars and its military. For example, the media--prompted by interest groups--loves big, impressive, round numbers. As the Syrian death toll slowly crept up, so did my guard. When I first heard that the civil war in Syria had killed 100,000 people, I threw my BS flag. (If you are curious, I always keep it in my back pocket.) I even tweeted @OnTheMedia asking, “Any chance the "100,000 dead in Syria #" doesn't hold up a la sex drugs and body counts?”.

To understand why, listen to this On The Media story on the uber-excellent, Sex, Drugs and Body Counts, a collection of essays analyzing the conventional wisdom behind widely cited numbers in various criminal and war-related international stories. In short, bigger and rounder is better:   

"We're not only more likely to remember the first number that we've heard, but we're likely to remember a big round number. So, for example, when the United Nations announces, as it did several decades ago, that the global drug trade was worth 500 billion dollars, that’s a memorable number. They later lowered that estimate to 400 billion, and an economist within the UN started questioning why that number. And apparently they rounded it up from 365 because it would, well, play better in the media and be more memorable.”

The biggest "debunking" in terms of war casualties comes from the number of dead cited by the media during the conflict in Bosnia. Initial reports from the U.N. placed the number of dead at over 250,000. Later, social scientists and historians dropped the number down to half that, 100,000. However, the first number remains lodged in the collective conscious. Eric C used to tell people that five million people died in the generally ignored Congolese Civil War. Researching the civil war in Congo, he found out that the real number was probably two million, if not much, much lower.  

Combining my skepticism with the knowledge that the media exaggerates death counts, naturally, I had my guard up when U.N. Secretary General Ban Ki Moon told the world that the U.N. Commission on Human Rights had determined that over 100,000 Syrians had died since fighting started. The press later repeated this figure without citation or reference, only referring to the U.N. Secretary General, not the scientists, data analysts, journalists or officials who created the number.

That number just didn’t sit right with me.

I decided to find if anyone could verify that 100,000 number. The shocking, good news: We can trust this number. Most bad statistics come from poorly informed or unplanned estimates or the high end of a range of estimates. Not this number.

This number comes from the UN Office of the High Commissioner on Human Rights, who relied on research from the Human Rights Data Analysis Group (HRDAG). This group actually enumerated the number of dead, meaning each person has a name and as much other information as possible. If anything, 100,000 is probably too low, since it doesn’t capture missing or nameless bodies. Though the exact number of dead will clearly fluctuate over time, on first blush, this number is not wildly inaccurate.

I don’t want to undersell HRDAG’s efforts. They are spending a significant amount of time in a rigorous process to identify duplicates, mistakes and fraudulent entries. The researchers also explain their methodology, along with its flaws, limitations and weaknesses. As they write in The Pacific Standard, they strive to be apolitical, guided only by good social science practices. Or read the report itself. 

Unfortunately, not all the numbers from this conflict are created equal. I am still skeptical of the number of dead Syrians by the chemical gas attacks. From The Economist:

“Médecins Sans Frontières, a charity, said three clinics it supports in the area treated 3,600 patients in a matter of hours, 355 of whom died. The Violations Documentation Centre, a Syrian organisation meticulous in its compilation of reports of death and injury, now puts the death toll at 457 or more. Other credible estimates range as high as 1,300. Harrowing videos—a man begging his two dead children to get up and walk; a girl repeating in wonder “I’m alive, I’m alive”—brought the atrocity home to the world.”

Which number do you think proponents of war will cite? Secretary of State John Kerry put the number as high as 1,400, which has since been endlessly repeated. If it eventually turns out to be much, much lower, few Americans will learn or remember that.

I am also skeptical on the numbers on Syrian refugees. The previously cited Sex, Drugs and Body Counts cites multiple examples of U.N. agencies using exaggerated numbers to gain political support. The U.N. Commission on Refugees itself tends to rely on huge, round numbers. Thus the Syrian conflict has created 2 million refugees, which seems too high. Apparently, the numbers come from individually registered Syrians, but I have a feeling that isn’t the case, and some estimation is in play. They also cite that 5 million Syrians have been “displaced” within Syria, which is very high and round too. Also, their use of the term “7 million refugees” is designed to mislead potential donors into thinking all 7 million Syrian refugees have left Syria, which isn’t the case.

As with all intelligence, trust but verify. In this case, we can rely on estimates that at least 100,000 Syrians (from both sides) having died, because they come from science and meticulous work. Great job, Human Rights Data Analysis Group.

four comments

My experience with such organisations (FAO for example) is that they are glad to compile existing information, and then they ant to publish a complete report, so they plug the holes in their data with what’s basically made-up numbers.


I find that the problem really comes with estimates and ranges. So one NGO says, “We think about 1 million people died, but the range could be as low as 500k and as high as 2 million.” Then another NGO or government agency says, “2 million died!”. They then take multiple high ranges to produce a new, high range of estimates. Suddenly, the numbers are dramatically high, and sometimes public affairs rounds them up even higher.

Thankfully, this 100,000 doesn’t have those flaws, though it is by no means a perfect number.


It concerns me that you would focus on Bosnia’s and Syria’s casualty lists of all the miscalculated casualty lists, namely the Vietnam War’s. Are miscalculated lists of civilian deaths, not the dead civilians, your biggest concern in such wars?


Austin, the casualty lists are by no means my biggest concern. That is why we published an open letter on Monday.

However—and this is a huge however—numbers and facts in the run up to war definitely matter. I haven’t seen a single media member explain to their audience where the 100,000 number dead came from. Its origins and validity—in this rare case, they are high—indisputably matter.

To see why, look at the situation in Bahrain. They have had an ongoing protest movement, but the casualties are no where near the same league as Syria. They are also a country of 1.2 million people. So the total number matters.