The most useful way to evaluate the claim that the Israel Defense Forces (IDF) has shown unusual restraint is to separate three questions:
How deadly has the campaign been? How many civilians have been killed, how much of the population has been displaced, and how extensive is the destruction?
How has the IDF conducted the war? What do the civilian-to-combatant death estimates, targeting practices, siege measures, and restrictions on humanitarian aid tell us?
How does Gaza compare with other wars? Is the level of harm unusually high, broadly typical of urban warfare, or lower than one might expect from a military trying to minimise civilian casualties?
There is no single statistic that conclusively answers all three. A military can take some precautions while still conducting a campaign that causes enormous civilian harm. Equally, a high death toll alone does not establish that every attack was unlawful.
The fairest assessment therefore compares several indicators, distinguishes verified facts from contested estimates, and applies the same standards to Israel as to other belligerents.
The clearest conclusion is that Gaza has experienced an exceptionally destructive campaign, with civilian harm on a scale that is difficult to characterise as ordinary or modest by the standards of recent urban wars.
The figures below are not all directly comparable: some count deaths over a few months, others over years, and definitions of civilian deaths vary.
Gaza · reported Palestinian fatalities
Reported by Gaza's Ministry of Health as of 4 June 2025, according to the UN's humanitarian reporting. This is a historical benchmark, not the current total.
Scale of displacement
Palestinians had been forcibly displaced by May 2024, according to the UN Commission of Inquiry. Later displacement continued.
For context, the UN Commission of Inquiry noted in May 2024 that the deaths in the Gaza campaign had already greatly exceeded those of previous rounds of Israeli-Palestinian hostilities. The scale of destruction was not simply a repetition of the 2008–09 or 2014 Gaza wars.
A useful, if imperfect, comparison appeared in a 2024 BMJ article examining the humanitarian consequences of recent conflicts.
Approximate deaths shown in the article's comparative chart; these are estimates assembled from different sources, not a harmonised database.
Approximate civilian and combatant deaths presented in a 2024 BMJ comparison. Different time spans and methods make direct rankings uncertain.
| Civilian deaths | Combatant deaths | |
|---|---|---|
| Gaza 2023–24 | 37000 | 12000 |
| Mariupol 2022 | 10300 | 10000 |
| Mosul 2016–17 | 11000 | 11000 |
| Raqqa 2017 | 3000 | 0 |
| Fallujah 2004 | 6000 | 2000 |
| Sarajevo 1992–96 | 11500 | 4500 |
| Grozny 1994 | 35000 | 5000 |
Source: BMJ, “When is enough, enough?” 2024. The chart's civilian and combatant categories, underlying sources and periods are not consistently defined. Raqqa's zero combatant figure means the comparison did not supply one, not that no combatants died.
Three things stand out.
The pace and concentration of civilian deaths are alarming. Gaza reached tens of thousands of reported deaths in less than a year. That is a different scale of destruction from a short, tightly bounded military operation.
Gaza is not literally without historical precedent. Grozny, for example, suffered a devastating civilian death toll; Sarajevo endured years of siege and shelling. Some older conflicts were comparably catastrophic, although the evidence is much less consistent.
Mosul and Raqqa are relevant comparisons, but not clean controls. They involved Western-backed forces fighting Islamic State in dense urban environments. Their civilian death tolls were substantial, yet the estimates shown in the BMJ comparison are considerably lower than Gaza's civilian estimate for 2023–24.
The International Committee of the Red Cross has also documented how urban offensives in Iraq and Syria produced disproportionately high civilian losses. Urban warfare itself is extremely dangerous, but that fact does not explain away the differences between campaigns. The weapons used, intelligence quality, rules of engagement, evacuation options and willingness to suspend attacks all matter.
My assessment: Gaza belongs among the most destructive urban campaigns of recent decades. It is not accurate to say it is unprecedented in all history, but neither is it persuasive to present its death toll as an unremarkable consequence of urban combat.
One important update to the historical comparison: by 1 October 2026, Gaza's Ministry of Health was reporting 74,040 Palestinian deaths and 175,168 injuries since the start of the war. It reported 1,439 deaths since the October 2025 ceasefire. These are reported totals, not a fully independently verified count of every death; the UN has explicitly distinguished figures supplied by local authorities from figures it has verified itself.
That update matters because the humanitarian consequences did not end when the ceasefire began. Reuters reported on 6 October 2026 that more than 1.2 million people in Gaza were still facing acute hunger, while reconstruction remained minimal.
The strongest argument for restraint is not that civilian casualties have been low. They plainly have not. It is that Israel has faced an unusually difficult battlefield: Hamas operates in a densely populated territory, uses tunnels and operates among civilians, while Israel says it has used warnings, evacuation orders, precision weapons and other precautions to reduce civilian casualties.
Those are relevant considerations. But the question is whether the evidence demonstrates that the IDF's precautions have been sufficient, and how its conduct compares with other militaries.
This is one of the most disputed measures.
The Gaza Ministry of Health's overall death toll does not itself distinguish civilians from combatants. Israel has published its own estimates of Hamas and other fighters killed, but the two sides' figures cannot simply be combined into a reliable ratio: their methods, definitions and evidential standards differ.
A 2024 peer-reviewed study in The Lancet's associated research literature, published in The Lancet Regional Health – Europe? More precisely, the paper is indexed in PubMed and available on PubMed Central, estimated the share of deaths that were combatants using a demographic model. It estimated that combatants accounted for:
62.1% of deaths in the 2008–09 Gaza conflict.
51.1% in the 2014 conflict.
12.7% in the 2023 conflict, with a modelled uncertainty interval of 9.7–15.4%.
If that estimate is approximately right, it would imply that civilian deaths greatly outnumbered combatant deaths in the 2023 conflict, and that the civilian impact was markedly worse than in the previous Gaza wars.
But there is a crucial limitation: the authors infer combatant deaths from demographic patterns and assumptions about age, sex and mortality. They do not have an independently verified identity for every person killed. The model's assumptions are contestable, and the result is not an agreed official civilian-to-combatant ratio.
The appropriate conclusion is therefore not that 12.7% is a definitive count. It is that a published demographic analysis found a very high civilian toll relative to combatant deaths, and that this is evidence against an unqualified claim of exceptional restraint.
The UN Human Rights Office examined six major Israeli attacks between October and December 2023. It verified 218 deaths in those incidents, while warning that the true number could be higher. Its June 2024 report raised serious concerns about the use of large bombs in densely populated areas, the destruction of residential buildings, and whether the attacks complied with distinction, proportionality and precautions.
The issue is not simply that large bombs were used. Large bombs can be lawful against a sufficiently important military objective in some circumstances. The issue is whether the expected civilian harm was excessive in relation to the concrete and direct military advantage anticipated, whether the target was a lawful military objective, and whether feasible precautions were taken.
The UN report also noted that warnings were absent in five of the six attacks it examined. That is a serious finding about those incidents, although six cases alone cannot establish the precise rate of unlawful attacks across the entire campaign.
Israel's use of warnings and evacuation orders elsewhere is relevant evidence in its favour. But a warning does not automatically make an attack lawful, particularly if civilians have nowhere safe to go or the attack itself remains indiscriminate or disproportionate.
There is no reliable global database that assigns every war a comparable percentage of unlawful attacks. So we cannot honestly say that, for example, 30% of IDF attacks violated international humanitarian law and that this was twice the rate in Mosul.
But there is substantial evidence beyond the casualty figures that raises serious concerns about Israel's conduct.
The UN Human Rights Office found that six examined attacks raised serious concerns about distinction, proportionality and precautions. Its assessment was that the pattern indicated the IDF may have systematically violated these principles.
Aid restrictions and the consequences of siege measures have been central to international legal scrutiny. In November 2024, the International Criminal Court issued arrest warrants for Benjamin Netanyahu and Yoav Gallant, finding reasonable grounds to believe they bore criminal responsibility for the war crime of starvation as a method of warfare, among other alleged crimes. These are arrest warrants, not convictions.
The UN Commission of Inquiry documented extensive destruction of civilian infrastructure and large-scale displacement. It concluded in 2024 that Israeli forces had committed war crimes and crimes against humanity. The scale of destruction is relevant evidence, though not every destroyed building was necessarily a civilian object unlawfully attacked.
The same UN Commission found that Hamas and other Palestinian armed groups committed war crimes, including deliberately attacking civilians and taking hostages. Their violations are serious in their own right, but do not remove Israel's separate obligations to protect civilians.
These are not merely accusations made by Israel's political opponents. They include findings from UN investigative bodies and judicial decisions by an international criminal court. They are also not all equivalent in legal status: an investigative commission's findings, an arrest warrant, and a final criminal conviction are different things.
In September 2025, the UN Independent International Commission of Inquiry concluded that Israel had committed genocide in Gaza. Its report considered killings, serious bodily and mental harm, conditions of life imposed on Palestinians, and measures affecting births, alongside statements by Israeli officials and the overall pattern of conduct.
That is a grave and significant finding by an independent UN investigative commission. However, it is important to distinguish it from a final judgment by the International Court of Justice (ICJ) in the separate genocide case brought by South Africa against Israel. As of October 2026, that case remains in progress; the ICJ's provisional measures are not a final determination that genocide has occurred.
Israel rejects the genocide allegation and argues that its campaign is directed against Hamas, not Palestinians as a group. That position must be considered, but the legal questions turn on the evidence of conduct and intent, not solely on the stated purpose of the operation.
To evaluate the restraint argument fairly, it is worth taking the military context seriously.
Hamas operates in densely populated areas. Tunnels, military positions in civilian neighbourhoods and fighting from built-up areas make identifying and attacking military objectives more difficult.
Israel has issued warnings and evacuation orders. These can be meaningful precautions when they give civilians adequate time and a genuinely safe destination.
The military objective matters. Israel has sought to dismantle Hamas's military capabilities, prevent future attacks and recover hostages. Those are legitimate objectives in principle, although they do not justify every means of pursuing them.
The comparison with Mosul and Raqqa has limitations. The duration of fighting, weapons, intelligence, enemy tactics, evacuation opportunities and how deaths were counted differ. A raw death toll cannot, by itself, prove that one military was more or less careful than another.
These considerations explain some of the danger civilians faced. They do not establish that Israel exercised exceptional restraint.
For that stronger claim, one would want evidence such as independently audited civilian casualties per attack, reliable estimates of combatants killed, records of aborted strikes, investigations into mistaken targeting, and evidence that commanders changed tactics when civilian harm became excessive. Publicly available data do not provide a sufficiently complete, independently verified comparison across all these measures.
There is also a difference between taking precautions and meeting the legal standard. A military may warn civilians, use precision-guided munitions and still conduct an unlawful attack if the target is not a military objective or the anticipated civilian harm is excessive. Conversely, a large number of civilian deaths does not automatically make every individual attack unlawful.
Here is how I would summarise the evidence.
Question | Assessment |
|---|---|
How bloody is Gaza compared with recent urban wars? | Exceptionally destructive. The reported death toll and civilian impact are among the most severe of recent urban conflicts. |
Does the civilian death toll prove every attack was unlawful? | No. Each attack must be assessed in its circumstances, although aggregate patterns can be powerful evidence. |
Is there evidence of serious breaches of humanitarian norms? | Yes. UN investigations have documented serious concerns and found war crimes and crimes against humanity; the ICC has issued warrants for alleged crimes. |
Can we prove the IDF never exercised restraint? | No. Warnings, particular targeting decisions and other precautions may have reduced harm in individual cases. |
Is there convincing evidence that the IDF showed exceptional restraint overall? | Not in the publicly available comparative statistics. The casualty patterns, destruction and humanitarian consequences weigh against that broad claim. |
My conclusion is that the claim of exceptional Israeli restraint is not supported by the overall statistical and investigative record. Gaza's civilian toll and destruction are far too extensive to dismiss as routine urban warfare, and the findings of international investigative and judicial bodies make the concerns about violations of humanitarian law substantial rather than speculative.
That conclusion does not require claiming that every Israeli strike was unlawful, that Hamas bears no responsibility for the danger to civilians, or that every disputed legal allegation has been finally adjudicated. It means that the burden of demonstrating exceptional restraint has not been met by pointing to warnings, precision weapons or the difficulty of fighting Hamas alone.
The most defensible position is that Gaza has suffered an extraordinarily high level of civilian harm, with credible evidence of grave violations of international humanitarian law. Whether the IDF's conduct was more or less restrained than that of particular other militaries cannot be quantified precisely from the available data, but the evidence does not justify treating its overall conduct as a model of humanitarian restraint.
