Data Reportby The Hanover InstituteNovember 16, 2026

Is Israel using starvation as a weapon? The crossing counts and the court record

Israel’s own crossing record counts 32,946 aid trucks to May 2024 and a 79-day nil period in 2025. What the court record and intent standard hold.

Is Israel using starvation as a weapon? The crossing counts and the court record

No court has issued a final ruling on whether Israel used starvation as a weapon of war in Gaza. The one interstate case, the Republic of South Africa’s Application filed at the International Court of Justice on 29 December 2023, is pleaded under the Genocide Convention and has produced interim orders, not a judgment on the merits.

Israel’s own crossing record, published by COGAT, a party to the events, counts 32,946 aid trucks between 21 October 2023 and 31 May 2024. Its live dashboard states that “no aid entered the Gaza Strip between March 2 to May 19 2025.” The record of what entered is the restricting party’s own. The legal dispute turns on the standard for inferring intent, which Nimer Sultany’s peer-reviewed 2024 article and Marko Milanovic’s 2025 commentary address from opposite directions.

Key Findings

  • The Republic of South Africa’s Application (Republic of South Africa, 2023) was filed on 29 December 2023 as General List No. 192. At paragraph 17 it founds jurisdiction on Article IX of the Genocide Convention. It is a state party’s pleading and records what South Africa alleged. No merits judgment exists in the case.
  • COGAT’s truck-level record counts 32,946 trucks between 21 October 2023 and 31 May 2024, 22,281 of them classed as food, on 193 operating days within a 224-day span (COGAT, 2026). This is a party’s own operational count and is not independently verified. The sums are the Institute’s arithmetic on the published rows.
  • COGAT’s dashboard states: “According to the directive from the political echelon, no aid entered the Gaza Strip between March 2 to May 19 2025” (COGAT, 2026). On the party’s own account, that is a nil period of 79 days.
  • Jamaluddine and colleagues (2025, peer-reviewed) estimate 64,260 deaths from traumatic injury to 30 June 2024 (95% CI 55,298 to 78,525). Their estimate excludes deaths from food insecurity. The published lists of the Ministry of Health in Gaza, a party’s own count, carry no cause-of-death column (Ministry of Health, Gaza, 2024 to 2026).
  • Sultany (2024) concludes that “the threshold of intent is met” only on a reading that relaxes what he calls the “high legal requirement” of the only reasonable inference standard. Milanovic (2025, not peer-reviewed) writes that fact-finders outside criminal trials “are not bound to apply a high standard of proof or ORI.”

Has any court ruled that Israel starved Gaza’s civilians?

No court has ruled on the merits. The only interstate case is the Republic of South Africa’s Application, filed on 29 December 2023 as General List No. 192. Paragraph 17 states that “pursuant to Article 36 (1) of the Court’s Statute and Article IX of the Genocide Convention, the Court has jurisdiction.” A starvation allegation before that Court is therefore argued as conduct under the Genocide Convention.

The Application is a pleading. It records what one state alleged, and at paragraph 147 South Africa “reserves the right to revise, supplement or amend the terms of this Application, as well as the grounds invoked.” Paragraph 11 records that “Neither South Africa nor Israel has entered any reservation to Article IX.”

The International Court of Justice has since issued three orders, dated 26 January, 28 March and 24 May 2024. The second is the one worded around hunger. On 28 March 2024 the Court indicated, unanimously on a bench of sixteen, that the State of Israel shall “take all necessary and effective measures to ensure, without delay, in full co-operation with the United Nations, the unhindered provision at scale by all concerned of urgently needed basic services and humanitarian assistance, including food, water, electricity, fuel, shelter.” The Court framed that measure “in view of the worsening conditions of life faced by Palestinians in Gaza, in particular the spread of famine and starvation.”

Several details travel with that order:

  • Judge ad hoc Barak, chosen by Israel, was among those voting in favour.
  • The 28 March order is authoritative in French, so the English quoted here is a translation of the governing text.
  • The orders do not settle what happened. At paragraph 30 of the 26 January order, the Court states that it “is not required to ascertain whether any violations of Israel’s obligations under the Genocide Convention have occurred,” a finding it says could come “only at the stage of the examination of the merits.”
  • At paragraph 49, the 28 March order states that it “is without prejudice to any findings concerning the Respondent’s compliance.”

A second forum has acted on the specific charge of starvation. On 21 November 2024 the International Criminal Court’s Pre-Trial Chamber I issued arrest warrants for Israel’s prime minister and its former defence minister. The Chamber stated that it found “reasonable grounds” to believe they bore responsibility for the war crime of starvation as a method of warfare. That wording tracks the warrant test in article 58 of the Rome Statute of the International Criminal Court, which asks whether “there are reasonable grounds to believe that the person has committed a crime.” A warrant precedes any trial. It is not a conviction.

Who argues Israel used starvation as a weapon, and since when?

Human Rights Watch stated the claim on 18 December 2023, ten weeks into the war, under the headline “Israel: Starvation Used as Weapon of War in Gaza.” The fullest peer-reviewed argument came later: Nimer Sultany’s “A Threshold Crossed,” published in the Journal of Genocide Research on 9 May 2024.

The claim has been restated on the record by other named bodies since then:

  • The BMJ published a 2024 piece under the headline “Israel is using starvation as a weapon of war in Gaza.”
  • In July 2025, Amnesty International published “Gaza: Evidence points to Israel’s continued use of starvation to inflict genocide against Palestinians.”
  • Also in July 2025, the UN human rights office issued a press release by UN experts titled “Thirst as a weapon.”

So the claim is held by named organisations and scholarship, dated, from December 2023 onward.

Sultany is a Reader in Public Law at SOAS University of London. His article includes a section on starvation, humanitarian law and intent. In a section on how “military goals” can conceal genocidal intent, he argues that the vocabulary of “collateral” harm and proportionality “masks indiscriminate attacks.” His conclusion is explicitly conditional: “If this interpretation is applied consistently, without double standards, the conclusion should be that the threshold of intent is met, and Israel violated the Genocide Convention by committing acts with the ‘intent to destroy, in whole or in part, a national, ethnical, racial or religious group, as such.’”

The interpretation he invokes is not his own coinage. Six states set it out in their Joint Declaration of Intervention of 15 November 2023 in The Gambia v. Myanmar: Canada, Denmark, France, Germany, the Netherlands and the United Kingdom. Sultany describes their argument as urging that the Court lower the “high legal requirement” of “only reasonable inference” and “instead assess genocidal acts and intentions in a more holistic way.”

Two limits attach to the article:

  • It carries no measurement of its own. The casualty figures it uses are relayed from other bodies.
  • Its conclusion concerns genocidal intent under the Genocide Convention. The war crime of starvation as a method of warfare is a separate charge under a separate treaty.

The opposing position is published by Israel’s Coordinator of Government Activities in the Territories (COGAT). Its document of 15 December 2025 is titled “Aid to Gaza: Common Misleading Claims vs. Facts.” In it, COGAT states that “Prior to October 7th, only two crossings operated for transferring goods into the Gaza Strip” and lists five inspection points and three transfer points operating since, which it counts as seven because Kerem Shalom appears on both lists. These are a party’s own statements about its own crossings.

How many aid trucks did Israel’s own crossing record count?

COGAT, the Israeli Ministry of Defense unit that runs the crossings, counted 32,946 aid trucks between 21 October 2023 and 31 May 2024 in its published “Humanitarian Aid” workbook, one row per truck. The figure is a party’s own operational count, tallied during inspection, and is not independently verified.

The content column divides that total exactly, so the breakdown can be summed. On the Institute’s arithmetic, 22,281 trucks (67.6% of the total) were classed as food.

Source: COGAT, 2026, “Humanitarian Aid” workbook; a party’s own count, not independently verified. Institute arithmetic on the published rows.

By crossing. The workbook records 18,371 trucks at Kerem Shalom, 13,359 at “Nitzana\Rafah”, 1,111 at Erez and 105 at JLOTS. These are not four entry points into Gaza:

  • The “Nitzana\Rafah” label fuses an Israeli inspection point with the Egypt-Gaza crossing, so it cannot show where a truck entered.
  • JLOTS was the temporary United States pier, not an Israeli crossing.

By day. Trucks moved on 193 days within a 224-day span. Of the 31 dates with no rows, 24 are Saturdays. That gives 170.7 trucks per operating day, or 147.1 per calendar day. The two averages differ by 16%, and COGAT publishes neither. The record’s highest day was 9 April 2024, with 467 trucks. Its lowest was 8 May 2024, with 9.

Trucks, not tons. The trucks are counted; the weights are declared. The weight column is headed “Estimated weight (tons)” and takes only eight discrete values, with 746 rows left blank. COGAT’s report of 9 July 2026 states that its tonnage “is based on the self-declaration of humanitarian organizations and private sector providers.” By contrast, “pallet number are counted” during security clearance.

A frozen series. The truck-level workbook ends on 31 May 2024. It was still linked from the portal’s home page in September 2026, beside a dashboard updated that same day. Truck counts after May 2024 therefore come from the dashboard, not the workbook.

How much fuel did Israel record entering Gaza?

COGAT’s published “Fuel” sheet records 335 fuel tanker-trucks entering Gaza between 1 November 2023 and 31 May 2024, over a contiguous 213-day daily series. This is a party’s own count and is not independently verified. It is also a floor: the sheet codes nil days with both dashes and empty cells, so a blank cannot be read as a zero with certainty.

The same sheet records 598 gas tanker-trucks over the same series. Across 213 days, the fuel figure works out to about 1.6 tanker-trucks a day on the Institute’s arithmetic. COGAT does not publish that rate. The sheet gives no volume per tanker, so the count cannot be converted into litres or set against a consumption figure.

Source: COGAT, 2026, “Humanitarian Aid” workbook, Fuel sheet; a party’s own count, not independently verified. Institute arithmetic on the published rows.

The published file needs repair before it can be read. Days 1 to 12 of each month were stored with day and month transposed, so the first row reads as 11 January 2023, a date before the war. Days 13 to 31 survived as text. Reversing the transposition restores a series with no missing dates, which is how the 213-day range was established. The fuel column also carries 77 dashes alongside further blanks, which is why both totals are floors rather than measurements of the whole.

Fuel is counted apart from everything else COGAT publishes. Its live dashboard states that “Totals do not include fuel,” so the dashboard’s truck totals measure goods other than fuel.

Fuel also appears in the court record. The 28 March 2024 measure of the International Court of Justice names fuel in its list of required supplies, between “electricity” and “shelter”.

When did Israel’s own record show no aid entering at all?

COGAT’s Aid Data dashboard, read on 10 September 2026, states: “According to the directive from the political echelon, no aid entered the Gaza Strip between March 2 to May 19 2025.” Counted inclusively, that is a nil period of 79 days, recorded in the restricting party’s own words and published on its own portal.

The sentence carries three facts:

  • A total. It records that no aid entered, not reduced entry.
  • A cause in the party’s own terms. It attributes the halt to a “directive from the political echelon,” the dashboard’s phrase for a decision above COGAT.
  • Dates. The halt runs from 2 March to 19 May 2025.

The sentence records entry only. It carries no figure for stocks already inside Gaza during those weeks, and no COGAT publication in this record supplies one.

Source: COGAT, 2026, Gaza Aid Data portal; a party’s own record, not independently verified.

The nil period changes how every longer figure must be read. Any cumulative or average truck figure spanning 2025 crosses 79 days on which, by COGAT’s own account, the count was zero. On 10 September 2026, the dashboard’s route table summed to 110,815 trucks over whatever date range its filter held when read:

  • 52,825 via Egypt
  • 38,544 via Israel
  • 9,947 via Jordan
  • 9,418 via Ashdod Port
  • 77 via “Israel - BNG Airport” and 4 by airdrop

A war-long daily average computed from that total, without the nil weeks stated beside it, describes no actual period of supply.

The workbook cannot test the halt, because its truck-level series ends in May 2024. The dashboard is the only COGAT publication in this record that reaches 2025. The entry standard, and how COGAT’s counts compare with other bodies’ measures, are read in a separate report on aid access and the enacted standard.

How many people lived in Gaza when the war began?

The Palestinian Central Bureau of Statistics (PCBS) projected Gaza’s mid-2023 population at 2,226,544. The figure is projected forward from the 1 December 2017 census rather than counted. It is the Bureau’s own estimate, published by the statistical office of a party to the events, and is not independently verified. It is the population that the supply recorded in the sections above had to reach.

The census base itself recorded 1,899,291 residents, by PCBS’s own account. Everything after 2017 is projection, including the age structure.

Age structure. PCBS projects that children under 18 made up 47% of Gaza’s mid-2023 population, or 1,046,400 people. The comparable shares were 41% in the West Bank and about 44% across Palestine. A separate PCBS release gives 40% for Gaza. That figure is the share under 15, a different cut-off, not a competing estimate. The 15-to-17 band between the two contains 147,693 people.

Source: Palestinian Central Bureau of Statistics, 2023; a party’s own projection, not independently verified.

Post-war revision. PCBS has since revised its own projections downward:

  • For 2024, it gives 2,129,724, which it describes as “a decrease of 6%”. Against its own pre-war table row of 2,287,558, the decrease computes to 6.9%.
  • For mid-2025, it gives 2,114,301, which is 10.0% below its pre-war projection of 2,349,301.

PCBS attributes the revision to deaths, missing persons, out-migration and reduced births together. It is not a casualty count.

Economic baseline. The World Bank’s September 2023 report, “Racing Against Time,” sets out Gaza’s labour market in the second quarter of 2023:

  • 46.4% of Gaza’s labour force was unemployed.
  • Labour force participation was 39.9%.

Every Gaza figure in that report is relayed from PCBS, as the Bank’s own source lines state. The Bank’s framing of the territory is that “Two million Palestinians live in Gaza, which has been under a near-blockade and effectively cut off from the West Bank since 2007.”

Do Gaza’s published death counts include deaths from hunger?

Jamaluddine and colleagues, writing in The Lancet in 2025 (peer-reviewed), estimate 64,260 deaths from traumatic injury in Gaza between 7 October 2023 and 30 June 2024 (95% CI 55,298 to 78,525). The estimate counts traumatic injury only. It leaves out deaths from food insecurity, inadequate water and sanitation, and disrupted health services.

The authors state that their rate comparison holds “even when ignoring non-injury excess mortality.” In other words, the hunger toll sits outside the figure, not inside it.

The method. The study combines three lists:

  • Ministry hospital morgue records
  • The ministry’s online mortality survey
  • Social media obituaries

The Ministry of Health in Gaza reported 37,877 deaths as of 30 June 2024. That is its own count, published by a party to the war, and not independently verified. The gap of 26,383 deaths is 41% of the study’s estimate, not 41% of the ministry figure.

Source: Jamaluddine and colleagues, 2025, peer-reviewed (95% CI 55,298 to 78,525); Ministry of Health, Gaza, 2024, a party’s own count, not independently verified.

The study and the ministry are not independent counts. The ministry’s list to 30 June 2024 divides into two strata:

  • 22,368 records labelled as ministry records
  • 5,817 records labelled as notified by relatives

The first matches the study’s count of 22,368 hospital-identified decedents for the same update. The second stratum corresponds to the ministry’s online survey, but the study worked from the raw survey file, which yielded 7,581 unique decedents after duplicates and missing persons were removed.

What the lists record. Three of the four ministry list documents in this record carry the same fields: serial number, name, identity or document number, date of birth, sex, age and source. The April 2024 list drops date of birth and source and adds an address column. None carries a cause-of-death column, and none carries a date of death.

On 7 January 2024, the ministry stated five exclusions of its own:

  • The missing and those under rubble
  • The unidentified who reached hospitals
  • Unidentified bodies handed back by Israel
  • Burials by families outside hospitals
  • Deaths in Gaza and North Gaza after its information system stopped in November 2023

The result. The peer-reviewed estimate is not built to measure deaths from hunger, and the ministry’s published lists carry no field that could separate them. A related report reads how the ministry’s registry has been tested.

What does the ceasefire’s mandated report say about aid access?

The Board of Peace’s report to the Security Council, document S/2026/418 of 15 May 2026, states that violations “continue to occur on a near daily basis.” It lists “continued impediments to humanitarian access” among consequences that “cannot be minimized.” It gives no number for either.

Whose account it is. The Board of Peace is the transitional administration whose establishment Security Council resolution 2803 (2025) welcomed. Its report is the Board’s own account of its own implementation. It was transmitted by the United States, a guarantor of the arrangement, and covers 17 November 2025 to 14 May 2026.

What the paragraph on violations contains. It gives no total, no rate, no attribution to a party and no definition of what was counted. Its phrase “civilians killed” also carries no figure.

Why the report gives no count. The resolution requests “a written report on progress,” and a progress report is what the Board delivered. The absence of a violations series follows from that request.

The diversion estimate. The report gives the Board’s own estimate that diversion of UN-facilitated in-kind aid fell from an estimated 90% at its peak during hostilities to approximately 1% since November 2025. The report states no method for either figure, and nowhere in the document is the 90% peak sourced. Both hedges belong with any use of the figures.

Commitments, not deliveries. The report’s figure of 4,200 trucks a week is a commitment, not a delivery count.

What the report cannot be combined with. The Board counts aid distributed by United Nations agencies and humanitarian partners during the ceasefire. COGAT counts entries at its own crossings, on its own records, from 10 October 2025. The two cannot be summed. The Board’s report is the first under the resolution, and a second was not yet published as of 10 September 2026, so no trend across reporting periods can be read from it. The instrument’s own terms are read in a separate report on ceasefire violations and what the instrument defines.

Does proof of deprivation alone establish intent to starve?

Article 8(2)(b)(xxv) of the Rome Statute, adopted in 1998, defines the war crime as “Intentionally using starvation of civilians as a method of warfare by depriving them of objects indispensable to their survival, including wilfully impeding relief supplies as provided for under the Geneva Conventions.” Its first word is an intent element, which sits alongside the deprivation.

Three different standards of proof are now attached to the Gaza question:

  • Reasonable grounds. The International Criminal Court’s November 2024 warrants used the article 58 test of “reasonable grounds to believe.”
  • Plausibility. The International Court of Justice’s interim orders used a plausibility test. At paragraph 54 of the 26 January 2024 order, the Court found plausible “the right of the Palestinians in Gaza to be protected from acts of genocide” and “the right of South Africa to seek Israel’s compliance.” That paragraph concerns rights, not violations, and paragraph 30 of the same order defers any finding of violation to the merits.
  • Beyond reasonable doubt. A criminal conviction requires the Court to be “convinced of the guilt of the accused beyond reasonable doubt,” under article 66(3) of the Rome Statute.

The dispute in the scholarship concerns how intent may be inferred when no order states it. Sultany (2024) describes the “only reasonable inference” requirement as a “high legal requirement.” He sets out the six-state case for a “more holistic” assessment, and his conclusion that “the threshold of intent is met” depends on that reading being applied. His argument addresses genocidal intent, the specific intent the Genocide Convention requires, which is a narrower question than intent under the starvation article.

Milanovic, writing on EJIL:Talk! on 18 September 2025 (a law blog, not peer-reviewed), traces the test to its source: “The notion of an ‘only reasonable inference’ comes to international law from the jurisprudence of the first modern ad hoc criminal tribunals.” He reads it as a consequence of the criminal standard of proof rather than a rule peculiar to genocide, and he states its reach: “Other fact-finders are not bound to apply a high standard of proof or ORI, for genocide or other international crimes alike. This includes formal fact-finding bodies, like a UN CoI, but also governments, NGOs, academics, or ordinary people.”

The record therefore separates two questions. The deprivation is documented in COGAT’s own counts and dashboard. Whether that deprivation was intended, and on which standard of proof, is the matter on which the named authors divide.

Methodology and limitations

Sources. The evidence comes from five kinds of source, each labelled by type:

  • Party records. COGAT’s workbook was downloaded and parsed row by row, and its dashboard was read on 10 September 2026. The Ministry of Health in Gaza’s four list documents and its daily reports were read on the same date.

  • A state party’s pleading. The Republic of South Africa’s Application, dated 28 December 2023 and filed 29 December 2023, was read in full, including paragraphs 9 to 17, 144 and 147.

  • Peer-reviewed research. Jamaluddine and colleagues (2025), in The Lancet, and Sultany (2024), in the Journal of Genocide Research.

  • Named-author commentary. Milanovic (2025), on EJIL:Talk!, the blog of the European Journal of International Law. It is editorially curated and not peer-reviewed.

  • Official statistics and reports. The Palestinian Central Bureau of Statistics’ 2023 releases, the World Bank’s September 2023 report to the Ad Hoc Liaison Committee, and Board of Peace report S/2026/418.

Court orders and warrants. The three orders of the International Court of Justice (26 January, 28 March and 24 May 2024) and the International Criminal Court’s warrants of 21 November 2024 are named by date. Their operative wording is quoted. No figure in this report rests on them. Treaty texts, including the Rome Statute and the Genocide Convention, are quoted as instruments.

Arithmetic. Every sum, share and daily average derived from COGAT’s rows is the Institute’s own arithmetic on the published file and is labelled as such. COGAT publishes none of these derived figures.

Limitations of the record:

  • COGAT’s data. The truck-level series ends on 31 May 2024. Tonnage is a banded, shipper-declared estimate, so trucks are cited and tons are not. Dashboard figures reflect whatever date filter was active when read. The fuel sheet’s date column required repair, and both fuel totals are floors.
  • PCBS’s population figures. Every population figure after the 2017 census is a projection. The Bureau has revised its post-war estimates downward by 6% to 10%.
  • The Lancet estimate. It covers traumatic injury only, carries a wide confidence interval, and is built partly on the ministry’s own lists.
  • The ministry lists. They carry no cause-of-death field, so no count of deaths from hunger appears in the record read here.
  • The Board of Peace report. It gives no violation count, and its diversion figures state no method.
  • Party sources. COGAT, the Ministry of Health in Gaza and PCBS are each a party to the events, or the statistical office of one, and none of their figures is independently verified.

Conclusion

So, has Israel used starvation as a weapon of war? On the record, the deprivation is documented and the intent remains in dispute.

The documentation of deprivation comes largely from the restricting party itself. COGAT’s own crossing record, a party’s count that is not independently verified, shows:

  • 32,946 trucks in the first seven months of record, about two thirds of them classed as food
  • Fuel arriving at an average of roughly 1.6 tanker-trucks a day
  • A 79-day period in 2025 when, by COGAT’s own statement, “no aid entered” under a directive from the political echelon

The legal record has moved, but no court has ruled on the merits:

  • The International Court of Justice unanimously ordered Israel to enable food, water and fuel “in view of... the spread of famine and starvation,” while stating that it was not yet ruling on violations.
  • The International Criminal Court issued warrants on “reasonable grounds,” which is a test for arrest, not for guilt.

Two gaps remain in the record. The only peer-reviewed death estimate in it, 64,260 traumatic deaths, leaves hunger outside its count by design. The ministry lists, a party’s own count, carry no field that could separate deaths from hunger. The ceasefire’s mandated report describes impediments to aid as occurring on a “near daily basis” and attaches no number to them.

What remains open is how intent is to be read. Deprivation recorded in the restricting party’s own ledger is one kind of evidence. A stated directive halting all entry for eleven weeks is another. Whether those facts allow only one reasonable inference, or whether a body outside a criminal trial may reach its view on a lower standard, is the point on which Sultany and Milanovic part. The question extends beyond this war: what any society asks of a state that records, in its own words, the closing of supply to two million people, and at what standard of proof it chooses to judge that record.

Frequently Asked Questions

Why do the vote majorities fall across the three Court orders?

The bench shrank from seventeen judges in January 2024 to sixteen in March and fifteen in May. Judge ad hoc Moseneke ceased to sit after a South African national was elected to the Court, and Judge Tomka did not sit in May. Measured against each bench, the shares barely move: thirteen of fifteen on the Rafah measure in May is 87%, against fifteen of seventeen on four of the six January measures, 88%; the other two January measures passed sixteen to one.

Did the Court direct any measure at Hamas?

None of the measures across the three 2024 orders is addressed to any party other than the State of Israel. The 28 March order explains at paragraph 44 that the Court cannot indicate measures addressed to third States or other entities, citing Article 59 of its Statute, and no Palestinian party is before the Court. In its reasoning, the Court called on 26 January 2024 for the “immediate and unconditional release” of hostages “held since then by Hamas and other armed groups”, and repeated that call in both later orders.

Did the May 2024 order require Israel to halt the Rafah offensive?

The 24 May 2024 order indicated, by thirteen votes to two, that Israel shall “Immediately halt its military offensive, and any other action in the Rafah Governorate, which may inflict on the Palestinian group in Gaza conditions of life that could bring about its physical destruction in whole or in part.” Whether the relative clause limits the halt to action of that character is disputed. Two judges appended dissenting opinions and three appended declarations to the order.

Which crossings did COGAT say were operating after the war began?

COGAT’s document of 15 December 2025 lists five inspection points: Kerem Shalom, Ashdod Port, Nitzana, Sha’ar Efrayim and Tarkumiya. It lists three transfer points: Kerem Shalom, Kisufim and Zikim. These are a party’s own statements. COGAT’s 2024 workbook names Erez rather than Zikim, and no COGAT document read here equates the two.

How many patients did COGAT record leaving Gaza for treatment?

COGAT’s Medical response page states that “Close to 5000 ill and wounded individuals, along with escorts” left Gaza for treatment. This is a party’s own figure that combines patients and escorts with no split. The same page gives conflicting accounts of the crossing used, and it carries no date.

What share of medical evacuation requests does COGAT say it approved?

COGAT’s June 2026 FAQ states that it approved “over 98% of all submitted medical evacuation requests.” That is an approval share of requests that reached COGAT, not a count of people who left. COGAT’s own FAQ makes departure depend on a third country agreeing to receive each patient.

Was Gaza’s economy shrinking before the war?

The World Bank reported in September 2023 that Gaza’s economy contracted by 2.6% year on year in the first quarter of 2023, relaying PCBS data. The Bank attributed the contraction chiefly to agriculture, forestry and fishing, which “shrank by almost 30 percent after the GoI restricted the sale of Gazan fish in the West Bank in August 2022.”

Sources

  • BMJ, 2024. Israel is using starvation as a weapon of war in Gaza. The BMJ, 385:q1018. https://www.bmj.com/content/385/bmj.q1018. Journal commentary.
  • Board of Peace, 2026. Implementation of United Nations Security Council Resolution 2803 (2025): Report of the Board of Peace through the Office of the High Representative for Gaza. S/2026/418, 15 May 2026, transmitted by the Permanent Representative of the United States. https://documents.un.org/api/symbol/access?s=S/2026/418&l=E&t=pdf. Official record (a party’s own account of its implementation).
  • COGAT, 2026. Gaza Aid Data portal: “Humanitarian Aid” workbook (Crossings and Fuel sheets) and Aid Data dashboard; with “Aid to Gaza: Common Misleading Claims vs. Facts” (15 December 2025), “Humanitarian Situation Report During the Ceasefire, October 2025 to June 2026” (9 July 2026), the Medical response page and “FAQ - Humanitarian efforts into Gaza - June 2026”. Coordinator of Government Activities in the Territories, Israel Ministry of Defense. https://gaza-aid-data.gov.il/mainhome/. Official record (a party’s own count, not independently verified).
  • Convention on the Prevention and Punishment of the Crime of Genocide, 1948. Article IX. Treaty text.
  • Jamaluddine, Z., Abukmail, H., Aly, S., Campbell, O. M. R., and Checchi, F., 2025. Traumatic injury mortality in the Gaza Strip from Oct 7, 2023, to June 30, 2024: a capture-recapture analysis. The Lancet, 405(10477), 469 to 477. DOI 10.1016/S0140-6736(24)02678-3. Peer-reviewed.
  • Milanovic, M., 2025. Proving Genocide. EJIL:Talk!, blog of the European Journal of International Law, 18 September 2025. https://www.ejiltalk.org/proving-genocide. Named-author law blog commentary, not peer-reviewed.
  • Ministry of Health, Gaza, 2024 to 2026. Published lists of martyrs with complete data (to 29 March, 30 April, 30 June and 31 August 2024), daily statistical reports, and statements of 7 January and 29 July 2024. Ministry media office channel, https://t.me/MOHMediaGaza. Official record (a party’s own count, not independently verified).
  • Palestinian Central Bureau of Statistics, 2023. Population and labour indicator tables; “H.E. Dr. Awad, highlights the Palestinian children’s situation on the occasion of the Palestinian Child Day” (5 April 2023); “Presents the Conditions of the Palestinian Population on the Occasion of the World Population Day” (10 July 2023). https://www.pcbs.gov.ps/. Official statistics (a party’s own projections, not independently verified).
  • Republic of South Africa, 2023. Application instituting proceedings and Request for the indication of provisional measures, Application of the Convention on the Prevention and Punishment of the Crime of Genocide in the Gaza Strip (South Africa v. Israel). Filed 29 December 2023, General List No. 192. Primary record (a state party’s pleading).
  • Rome Statute of the International Criminal Court, 1998. Articles 8(2)(b)(xxv), 58 and 66(3). Treaty text.
  • Sultany, N., 2024. A Threshold Crossed: On Genocidal Intent and the Duty to Prevent Genocide in Palestine. Journal of Genocide Research, published online 9 May 2024. DOI 10.1080/14623528.2024.2351261. Peer-reviewed.
  • United Nations Security Council resolution 2803 (2025). Operative paragraph 10. Instrument.
  • World Bank, 2023. Racing Against Time: World Bank Economic Monitoring Report to the Ad Hoc Liaison Committee. September 2023. https://documents1.worldbank.org/curated/en/099638209132320721/pdf/IDU0e8b2e87e098b004a7a09dcb07634eb9548f4.pdf. Official report (Gaza figures relayed from PCBS).

Related research