Home > News > A closer look at the Gaza casualty data
449 views 26 min 0 Comment

A closer look at the Gaza casualty data

Casualty counts can be a political tool – and how we report the data has real consequences.

- December 14, 2023
Children picking their way through rubble in Gaza. Gaza casualty data
Palestinians inspect the damage after an Israeli airstrike in Gaza, Oct. 10, 2023 (cc) Naaman Omar, Palestinian News & Information Agency (Wafa) in contract with APAimages, via Wikimedia Commons

Two months after the Oct. 7 Hamas attacks on Israel, Israel’s pledge to destroy Hamas has left much of Gaza in ruins. Aerial bombings and Israel Defense Forces (IDF) ground operations have left more than 18,000 people dead. To help us go beyond the numbers and the headlines, Marc Lynch reached out to Sarah E. Parkinson, Aronson Assistant Professor of Political Science and International Studies at Johns Hopkins University, who studies organizational behavior and social change in contexts of war and disaster. 

Marc Lynch: Hi Sarah. You’ve been studying the politics and the mechanics of casualty counting for years, across multiple conflict zones. What should people be aware of when they hear casualty reports? 

Sarah Parkinson: When discussing the politics of counting casualties, we are really talking about how we recognize large-scale loss of life. Advocacy projects such as We Are Not Numbers in Palestine and Lebanon exist precisely because mortality statistics such as casualty counts almost necessarily erase the nuances of people’s lives. 

Casualty counts can thus become a political tool, creating distance from the horrors and tragedies of war and disaster. The word “casualty” often includes both dead and wounded – but “casualty count” can mean different things. In reference to combatants, it can mean individuals who can no longer fight due to death, injury, illness, disappearance, or imprisonment. Sometimes, you’ll see casualties reported in separate categories of deaths, wounded, and “missing” or “unaccounted for.” 

That suggests accuracy isn’t always possible.

That’s true. We also see these terms used to report on large-scale tragedies such as migrant boat sinkings. Particularly when rescue operations are ongoing and counts are repeatedly updated, casualty counts can include “likely” deaths, or presumed deaths that simply cannot be confirmed because a body was not recovered. In situations where there is widespread structural collapse – bombing campaigns that target entire neighborhoods or earthquakes that collapse whole buildings, for example – people buried under the rubble may not be counted or may be listed as missing. 

So casualty counts – both for the dead and for the wounded – are often incomplete. That’s because war or disaster makes data collection incredibly difficult and often dangerous. In Gaza, as in many conflicts that involve civilian mass casualties, numbers include only the deaths of people who have been taken to health facilities or morgues (often the same location). If someone dies at home and is buried by their family, if they die in an area cut off or far from medical facilities, or if their body is interred in a mass grave, they may not be counted at all, or they may not be counted until much later. A death might also be counted without the person being positively identified. 

Numbers for the wounded can skew towards counting only more serious injuries, something that would generate a hospital admission. During Israel’s Lebanon War in 1982, which I have studied extensively, people who received relatively simple care (like straightforward suturing) at hospitals and clinics were not included in the counts of the wounded, nor were people who were treated at home. 

Who does the actual counting? 

Different organizations may participate in casualty counting: a country’s health ministry, a militant group’s health bureaucracy, a state military, international non-governmental organizations such as the International Committee of the Red Cross, human rights advocates, or journalists. If a country doesn’t have a well-established health bureaucracy, casualty counts may be the cumulative production of journalists and/or advocacy groups who call around to health facilities (and potentially other institutions, such as houses of worship) to get the number every day. And the counts might be affected by urban bias, or any other number of biases, as scholars such as Sophia Dawkins have explored in detail

Some casualty counts break out statistics for combatants and noncombatants/civilians; those may be based on confirmed numbers, estimates, or best guesses given, for example, whether someone is wearing a uniform or insignia. These two categories reflect the “principle of distinction” between combatants and civilians, which is enshrined in contemporary international law. This distinction has existed in different forms, according to political scientist Helen Kinsella, since at least the Middle Ages. Kinsella expressly notes that this combatant/civilian line is highly unstable – and that it is both gendered and racialized.

Casualty counts may be incomplete – but also can be deliberately underreported, overreported, or censored. Military organizations might have a vested interest in portraying their success or respect for international law through a low casualty count. Groups may hide some deaths, or specific deaths, for tactical or reputational reasons. 

States may overreport to make an opposing regime seem more hostile. Or administrations might try to insulate themselves from public backlash to war, by minimizing media coverage of fallen soldiers. The Dover Ban policy, for example, barred journalists from photographing the caskets of deceased U.S. soldiers arriving at Dover Air Force Base. The Dover Ban was instituted during the Gulf War and reaffirmed after 9/11 in part because U.S. administrations worried that the images of rows of coffins would decrease public support for U.S. campaigns in Afghanistan and Iraq. In 2009, the Obama administration lifted the ban.

Okay, that’s a really great background on the challenges of casualty counting. So how does that apply in Gaza? We’ve seen reports that more than 18,000 Palestinians have been killed and 50,000 wounded. How reliable are those numbers, and what do they really tell us?

The main source for casualty statistics in Gaza is the Gaza Ministry of Health, which is run by Hamas. According to AP reporting, the ministry is staffed by “a mix of recent Hamas hires and older civil servants affiliated with the secular nationalist Fatah party.” Health care facilities in Gaza document deaths and injuries and send them to a central Ministry of Health team, which compiles those numbers daily in an electronic spreadsheet. The ministry spokesperson, Ashraf al-Qidra, then reports the numbers publicly. The U.N., the World Health Organization, and Human Rights Watch, along with a number of Palestinian factions, have long treated these numbers as reliable. 

Recent studies published in The Lancet, a highly respected medical journal, by independent teams at two different universities – one published Nov. 26, 2023, by coauthors at the London School of Hygiene and Tropical Medicine, and one published Dec. 6, by scholars at the Johns Hopkins Bloomberg School of Public Health – have found the Gaza Ministry of Health casualty counts to be reliable. Given that Gaza is one of the most surveilled places on earth, it would be hard to artificially inflate mortality statistics with a large number of fake deaths; the tracking process includes both the names and ID numbers of those who have died. There have been arguments about who is responsible for some casualties, specifically related to events like the Oct. 17 ordinance explosion at al-Ahli Hospital, which killed several hundred people. 

So is independent verification possible in Gaza?

News organizations have noted that it is impossible for journalists or humanitarians to conduct comprehensive, independent casualty counts because Israel has blocked them from entering Gaza. EuroMed Human Rights Monitor, a Swiss-registered human rights organization with an on-the-ground network, reports higher numbers than the Ministry of Health, but they also note that their numbers include people trapped under the rubble

The Gaza Field Office of the United Nations Relief and Works Agency for Palestine Refugees in the Near East (UNRWA) independently tracks casualties among their staff, reports them to headquarters in Jordan, and publishes them in their Situation Reports as they are confirmed. Both of the recent studies in the Lancet note that the rate at which UNRWA staff are being killed generally matches the overall mortality rate reflected in the ministry’s statistics. So that’s an indication that the Gaza Ministry of Health statistics do not reflect an inflated mortality rate or fabricated data.

As a grim side note, casualty counts might increase at different rates in the reported statistics. It’s possible to confirm a death, yet confirming age, gender, or identity can take considerably longer. This was true of the identification process in Israel after the Oct. 7 Hamas attacks; some bodies were so badly burned or disfigured that they could be confirmed as part of an overall casualty count but were not positively identified for weeks. 

So when we see social media accounts posting that the U.N.-published Gaza Ministry of Health numbers indicated a larger change in the number of dead children than newly reported (overall) deaths for the same day, it could be because already-reported deaths were later confirmed to be children. That’s what would make the ratio of children killed rise without the overall mortality count rising by the same number. 

A recent Israeli study claimed that 61% of the casualties in Gaza were civilians. Is that plausible?

The Gaza Ministry of Health numbers do not distinguish between civilian and combatant deaths or injuries. That leaves scholars, human rights organizations, and the news media looking for ways to estimate civilian and combatant deaths – emphasis on the word “estimate.” The London School of Hygiene and Tropical Medicine study estimated that at least 68.1% of the deaths in Gaza have been noncombatants. And a study by Israeli sociologist Yagil Levy that was published in Haaretz estimated the civilian death toll at 61%. Both studies get to that number in mostly the same way; they use the Gaza Ministry of Health data from October 7 to 26. Both studies place children (those younger than 18), adult women (ages 18-59), and the elderly (those 60 and over) into a “noncombatant” category (the Lancet correspondence calls them “groups that probably include few combatants”). Levy discusses men (ages 18-59) as adults who he did not include in the noncombatant category; the study in the Lancet is more vague, with the unstated implication being that adult men (those not in “groups that probably include few combatants”) may constitute “potential” combatants. 

Wait – these studies count all adult men as combatants? That seems problematic. 

Well, yes – it’s a very real problem. The underlying premise is that these are “military-age men” or “potential combatants.” You’ll see these terms repeated or quoted in the media, and I’ll come back to that later. 

In terms of mortality calculations, both studies are deliberately conservative in their approach. Both make their assumptions clear. Levy explicitly notes that even if one excludes all adult men (aged 18-59) in Gaza from the noncombatant category, that some number of Palestinian deaths are due to Palestinian rocket misfires, and that the number of non-combatant adult males counted as combatants cancels out some number of child combatants who are counted as civilians (to be clear, children are a protected group under international humanitarian law), the percentage of recent civilian deaths in Gaza is still incredibly high. 

Levy’s conservative estimate was that civilian deaths accounted for 61% of deaths related to Israel’s October bombing campaigns. By comparison, in past Israel-Gaza wars, civilians accounted for 33% to 42% of casualties from aerial attacks. Levy goes on to underscore that given Israel’s capacity for distinction and technological capabilities, the level of civilian harm is almost certainly the result of explicit decisions taken by Israeli officials. This claim jibes with multiple public statements by members of the Israeli government as well as with reporting from Israeli media outlets

Neither of these studies was conducted with evidence of the actual combatant status of deceased adult men. Yet, we know from independent polling by Arab Barometer that the majority of adults in Gaza do not trust Hamas and would not vote for a Hamas leader for president of the Palestinian Authority. We are talking about a population where only around a third of adults articulate political support for Hamas. It’s consequently empirically unrealistic that all confirmed adult male deaths in Gaza are combatant deaths. 

These studies – conducted with blunt measures and stated assumptions given a lack of more detailed information – demonstrate that even with the most conservative estimate possible, the civilian mortality rate in Gaza is extremely high in both real and comparative terms. But the studies published in the Lancet and Haaretz do not engage with polling data or with political realities on the ground; they are based on broad demographic proxies. 

Why are these assumptions so dangerous? 

It’s a very imprecise measure to assume that all male Gazan mortalities are combatants and that adult women, children, and the elderly are civilians. It’s worth underscoring that this division ties directly into the gendered assumptions that undergird the contemporary principle of distinction, which scholars like Kinsella, who I mentioned above, and Cynthia Enloe, a scholar of international relations who works on gender and war, have extensively critiqued

Publishing combatant-civilian casualty numbers without fully explaining the caveats behind the data can have life-and-death consequences. Sharing casualty estimates that posit all men in Gaza as Hamas combatants, for instance, feeds into a normative narrative that actively denies their human rights and endangers them. Indeed, in 2014, anthropologist Maya Mikdashi critiqued dehumanizing and disenfranchising narratives where “boys and men are presumed guilty of what they might do if allowed to live their lives.” Political scientist Charli Carpenter has clarified that in some settings, civilian men are at extreme risk of harm but lack the protection of presumed innocence afforded to women and children. 

Given the embedded assumptions in these studies, media and human rights organizations should always clarify the assumptions behind these casualty numbers. Reporters can state up front that the percentage of confirmed deaths that are also civilian deaths is likely significantly higher than the statistics in either the Lancet article or the Haaretz one. Out of respect for basic human rights and humanity, media reports should emphasize that Palestinian men can be, and mostly are, civilians. 

The Israel Defense Forces claims a 2:1 kill ratio, meaning that they have killed two civilians for every Hamas militant. Setting aside the point that this isn’t something to brag about, does it sound plausible? 

That’s unclear – the IDF might have its own internal numbers; it might be basing that claim on the Gaza Ministry of Health numbers, or on a combination of data. The original AFP report notes that the 2:1 kill ratio was shared in a briefing for foreign reporters and that the IDF “believes the overall Gaza death toll claimed by Hamas is fairly accurate.” 

The percentages published in the Lancet and Haaretz do roughly map along a 2:1 ratio of civilian to militant deaths. But to build on what I said above and the fact that independent studies have found the Ministry of Health numbers credible, it’s a very specific political choice to label all Palestinian men between 18 and 59 “military-age men” (as Israeli spokesperson Eylon Levy, among others, has done). This equates the constructed category of “potential combatants” with viable targets, and functionally assumes Palestinian men’s culpability based on demographics alone.

But let’s also recognize that using a ratio effectively masks the scale of killing. It’s a big psychological difference to read “2:1 civilian to militant ratio” versus, at the time the ratio was released, ”Israeli airstrikes and ground attacks have killed 15,899 Palestinians, at least 10,599 of them civilians.” 

The raw numbers speak directly to the massive power imbalance, the disproportionate use of force, and a lack of application of the principle of distinction during this war. And the use of a ratio also plays into a specific military mindset that shapes U.S. policy decisions related to weapons sales. At what point is the ratio so high that the U.S. government will balk at sending Israel more weapons? 

Let’s broaden the discussion. Your work has pointed towards the less-visible human toll – starvation, and so much else in conflict zones. Could you talk a bit about the broader range of suffering that this kind of war entails?

My work and plenty of others’, to be clear. My interviewees in Lebanon talked to me extensively about the lasting effects of starvation, dehydration, disease, and disability. They frequently cited witnessing starvation and dehydration, for example during Lebanese militias’ siege of the Tel al-Za’tar refugee camp in East Beirut, as motivating their later mobilization (both violent and non-violent). 

Death due to denial of medical care – because people cannot reach medical facilities or because those facilities have been destroyed – is of course a factor as well. This is tragically evident in parts of Syria, especially, for example, in the aftermath of the Feb. 6, 2023 earthquake. Many of the deaths attributed to illness in war are in fact due to the destruction of medical infrastructure. The Lancet piece published by the London team notes that these kinds of casualties – e.g., deaths from untreated obstetric complications – are likely to increase in Gaza in the context of the IDF ground operation.

We should also emphasize the cumulative effects of exposure to toxic chemicals, pollution, and other environmental hazards. Anthropologists Omar Dewachi and Kali Rubaii, for example, both study the legacies of the U.S. occupation of Iraq; Rubaii has demonstrated the severe environmental effects that burn pits had for the population of Fallujah. From a public health perspective, some of these effects will be intergenerational, for example, because environmental hazards cause birth defects. Some researchers have even found that trauma has intergenerational consequences. These are all likely to be factors in overall health in Gaza for decades to come.

What should we be looking for going forward, then, in terms of evaluating the impacts of this war?

People need to understand studies like those in the Lancet and Haaretz as rough cuts that do not have evidence for or provide proof of combatant status. These studies were conducted for limited, specific goals and offer a starting point for further inquiry into casualty numbers. In the long run, more sophisticated techniques that address the vast issues with missing data could help produce a more accurate estimate of total casualties – and the full consequences of this war. 

In the short term, any assumptions about men’s combatant status, rather than an acknowledged lack of more detailed information, contributes to a narrative that assigns guilt to Palestinian men based on demographics alone, compounds their vulnerability, and facilitates indiscriminate violence. Media outlets need to caveat these statistics and to demand sources for the raw numbers and labeling criteria that underlie any “kill ratio” presented to them.

Casualty statistics are an incomplete, inherently flawed measure of the scale of destruction caused by war. They are hard to accurately compile, subject to biases, and constantly politicized. While they have their place, these numbers never present a complete story.