Has Israel really killed up to 186,000 people in Gaza? How to understand the numbers war
People injured during Israeli bombardment arrive at Nassr hospital in Khan Yunis on July 9. Photo by Bashar Taleb/AFP/Getty Images
The political contest being waged over the number of civilian deaths in Gaza is making it harder, not easier, for us to actually understand just how many civilians have been harmed through the past nine months of war, and in what ways.
Those who study civilian casualty counts, as I do, understand that they are one of the many metrics needed to understand the true human cost of a conflict. They cannot present the full picture on their own. By emphasizing political objectives over scholarly ones — say, trying to prove that Israel is committing a genocide — the combatants in this war of words are impeding the process that typically produces reasonable estimates of civilian harm.
Often, the discrepancies between the outcomes of those processes and initial estimates are glaring. It was originally estimated that about 200,000 people were killed during the 1992-1995 Bosnian War. Three decades later, the officially accepted number is less than half that amount. The fatalities associated with the Congo wars were initially estimated at 5.4 million by a humanitarian NGO. However, an academic review found that a rigorous study would have only tallied half that number of deaths. In the latter case, the academics ultimately concluded that the numbers were inflated for fundraising purposes.
And yet, rather than learn the importance of being careful and precise with estimated casualties, academics and activists involved on both sides of the Israel-Hamas war are embracing findings that are clearly flawed. The ambiguity inherent in much of international humanitarian law creates a political incentive for those advocating for Israel’s prosecution to generate as large an estimate of civilian casualties as possible. And it places pressure on those defending Israel’s war against Hamas to try to minimize civilian casualties, both in practice and in the papers.
The latest headline-making instance of this conflict: An article in the revered medical journal The Lancet that declared “it is not implausible to estimate that up to 186,000 or even more deaths could be attributable to the current conflict in Gaza” — a number far beyond the Hamas-run Gaza Health Ministry’s count of 37,396 dead.
That paper was quickly disseminated across social media as proof of a genocide conducted against Palestinians. But it has important shortcomings involving data that will be crucial to how we evaluate the true contours of this conflict.
First among those flaws: The authors of The Lancet‘s study made no effort to distinguish between civilian and military fatalities.
Their concern is with the idea of excess mortality: How many more people have died in Gaza in the last nine months than would have been expected to do so under ordinary circumstances, and how many more are likely to die? Under that banner, there is no distinction between combatants and noncombatants; every death is presumed to have equal moral weight.
Indirect deaths are difficult to calculate, and the reasons behind them are difficult to parse. Researchers must take into account conditions during and after conflict, which is almost impossible to do effectively while a war is ongoing. For instance, one study of indirect deaths as a result of the wars in the Democratic Republic of Congo found that the number one cause of death in the country was still malaria, despite years of ongoing armed conflict.
What this means: We can estimate how many more people have died in Gaza since October than normally would have in such a time span. However, we will not have clear data illustrating exactly why, and how such deaths were related to the war, for many years.
By failing to isolate civilian deaths from combatant ones, the Lancet authors fall into a trap set by Hamas, which does not distinguish between civilian and military fatalities in its own reported death toll, and has consistently refused to provide reporters with estimates of its military casualties. (In humanitarian law, this is a violation of the principle of distinction, which requires that militaries distinguish between civilians and combatants.) Estimates of those casualties have generally ranged from 6,000 to 12,000, numbers that reflect imprecise assessments of the operational impact of the war and the significant mixing of civilians and combatants in Gaza.
It should go without saying that a true calculation of the civilian toll of war must delineate between civilians and combatants as clearly as possible; to fail to do so is to further obscure already muddied waters. It can also hamper accountability. In more than a third of the cases considered by the International Criminal Tribunal for the former Yugoslavia, a United Nations war crimes court, officials were unable to identify the civilian or military status of the victims involved.
Also troubling: The Lancet‘s study notes that 30 percent of the reported deaths in Gaza are unidentified, but fails to address what percentage of the reported deaths are unverified.
The inclusion of unverified deaths is an especially important concern, given the Hamas-run Gaza Health Ministry’s increasing reliance on “information from reliable media sources and first responders,” as well as self-reporting by relatives and reports put out by the Gaza Government Media Office. What constitutes a reliable media source, as opposed to an unreliable source, or even a clearly politicized source, is not discussed. The United Nations itself acknowledged in December 2023 that the methodology used by the Gaza Government Media Office to report fatalities “is unknown.”
This is not to say those numbers are wrong — only that it’s a mistake to assume that they’re right without any independent verification. Groups producing casualty counts can have competing interests. It can take years to establish the true human toll of a war, including by examining long-term health outcomes, like nutrition, fertility, and mental health. By baking numbers that have not been established through rigorous, well-understood processes into our shared consciousness, those behind studies like The Lancet‘s will make it harder for the public to accept whatever the long-term outcomes of that research are.
There are plenty of examples about how this has worked in the past — including that of Oct. 7. Initial estimates suggested that some 1,500 people had been killed in the course of the Hamas attack; estimates were later consistently revised down, eventually settling on a number just below 1,200. But the figure of 1,500 still floats around social media: People remember the first reports they see better than subsequent, more factual ones.
Striving for accuracy — including by openly acknowledging how much we still don’t know — is particularly important in cases, like this war, in which allegations of genocide are at stake. Genocide, as defined in international law, hinges on two important requirements: intent and the destruction of a group as a whole or in part. Leaving aside the difficulties of establishing intent, what “in part” means has never been defined. Is it 2% of a group’s population — about the percentage of Gazan Palestinians, including civilians and combatants alike, killed in this war? If the threshold is higher, by how much, and why?
The projected death toll for Gaza laid out in The Lancet is a matter of modeling, not concrete fact. When calculating their projection, the researchers assumed that there would be four indirect fatalities in Gaza for every direct one. In doing so, they assumed that the war in Gaza will ultimately prove more deadly than recent wars in Yemen (which involved a ratio of 1.3 indirect deaths to 1 direct, according to the UN) or Ethiopia (at most a 2-1 ratio). They may have valid reasons for making that assumption, but they currently remain unexplained.
There have been many, many thousands of civilian deaths in Gaza. It is certain that many more civilians have died than combatants. But the truth that those who use these numbers for political ends overlook is that it will take years to establish reliable and agreed-upon figures, if the current politicization of estimates is ever overcome. The Bosnian Book of the Dead, a database of 96,985 war-related deaths from the 1992-95 war, was completed only in July 2006, more than a decade after the war’s end. The expectation that a hard-and-fast number can be created in real time for fatalities in the Israel-Hamas war is both unrealistic and misleading.
A message from our editor-in-chief Jodi Rudoren
We're building on 127 years of independent journalism to help you develop deeper connections to what it means to be Jewish today.
With so much at stake for the Jewish people right now — war, rising antisemitism, a high-stakes U.S. presidential election — American Jews depend on the Forward's perspective, integrity and courage.
— Jodi Rudoren, Editor-in-Chief
