Argument · heard on both sides

“Israel is starving Gaza — famine has been confirmed.”

An authoritative declaration of famine in August 2025 is taken as proof of mass starvation deliberately imposed by Israel; questioning the numbers is treated as supporting the starvation of children.

Made publicly as a argument by International Criminal Court, Pre-Trial Chamber I

Where it's been said

“reasonable grounds to believe that [they] each bear criminal responsibility for … the war crime of starvation as a method of warfare”
International Criminal Court, Pre-Trial Chamber I, court — Decision issuing arrest warrants for Benjamin Netanyahu and Yoav Gallant, 2024-11-21 ✓
The mirror, from the other side
“There is no policy of starvation in Gaza, and there is no starvation in Gaza.”
Benjamin Netanyahu, Prime Minister of Israel — Statement, 2025-07-27 ✓

Traps in play:◆ Proxy presented as outcome◆ The authority endpoint◆ "It's not complicated"

  1. Why people find it convincing

    • Aid into Gaza was severely restricted for long periods: the IPC's Famine Review Committee records that 'Between 2 March and 19 May 2025 no food entered the Gaza Strip by any actor or supplier (humanitarian or commercial)' and that the food needed 'has not been met since April 2025', and on 9 October 2023 the then defence minister, Yoav Gallant, announced a 'complete siege': 'There will be no electricity, no food, no water, no fuel, everything is closed' (as quoted by Human Rights Watch). What entered often didn't reach people: '87 percent of UN trucks reportedly intercepted', and 'the number of households reportedly facing potential safety risks while searching for food has more than doubled to around 80% of households' (Famine Review Committee).
    • In November 2024 the International Criminal Court issued arrest warrants for Israel's prime minister and former defence minister, finding reasonable grounds to believe they bear criminal responsibility for 'the war crime of starvation as a method of warfare', for the period from October 2023 to at least May 2024.
    • UN agencies reported that 'in July alone, more than 12,000 children were identified as acutely malnourished – the highest monthly figure ever recorded' (UN News, 22 August 2025), and Gaza's Health Ministry had recorded 269 malnutrition deaths by 20 August 2025, a count the famine researcher Alex de Waal says is 'always a major undercount'.
    • On 22 August 2025 the IPC Famine Review Committee — the UN-backed global standard — classified famine in Gaza Governorate, the first famine classification in the Middle East ('the first time a famine has been officially confirmed in the Middle East region', UN News).
    • UN leaders put it in the strongest terms. The relief chief Tom Fletcher said the famine could have been prevented 'if we had been allowed', that 'food stacks up at borders because of systematic obstruction by Israel', and that 'It is a famine openly promoted by some Israeli leaders as a weapon of war'; Secretary-General Guterres called it 'a man-made disaster, a moral indictment – and a failure of humanity itself' (UN News, 22 August 2025).
  2. What's true

    Severe hunger and acute malnutrition were real, and aid restrictions were a cause.

    Acute malnutrition admissions peaked at nearly 17,000 children in August 2025; the Gaza Ministry of Health recorded about 440 malnutrition deaths by September 2025 (including 147 children), 185 of them in August alone. Restrictions on aid — including the spring 2025 blockade — contributed directly. The restrictions began as stated policy: on 9 October 2023 the then defence minister Yoav Gallant announced a 'complete siege' ('There will be no electricity, no food, no water, no fuel, everything is closed'), and, in Human Rights Watch's account, Israeli authorities 'cut off all water and blocked fuel, food, and humanitarian aid' that day 'and for weeks thereafter'. In November 2024 the International Criminal Court issued arrest warrants for Netanyahu and Gallant that include 'the war crime of starvation as a method of warfare': warrants, not a verdict. None of that depends on the famine label, and hunger imposed on civilians is grave whatever it is called. The famine warning used Israel's own figures: FEWS NET cites COGAT data that no food entered northern Gaza from March to May 2025 and 'only 151,059 MT of food was approved for entry from May 1 to August 17', against a UN estimate of at least 62,000 MT needed each month.

    dataUNICEF / Nutrition Cluster, Gaza SMART Survey 2026 (2026) aggregate ✓

    Published 24 August 2026. Fieldwork 31 May–15 June 2026, accessible areas only (about 17% of the population, ~358,000 people, excluded for access and security). Release: 'Acute malnutrition among children under five was low'; stunting 12.2%, overweight 4.0%. No GAM percentage in the release.

    Open source ↗

    dataGaza Ministry of Health malnutrition death counts, as reported (Al Jazeera, Aug–Sep 2025) (2025) aggregate ✓

    Al Jazeera, 2 September 2025 (read), citing the Ministry of Health: 185 deaths 'due to malnutrition' in August 2025; cumulative 361 including 130 children since October 2023. The report uses the phrase 'Israel's genocidal war' in its own voice.

    Open source ↗

    primary documentFEWS NET, 'Gaza Alert August 22, 2025: Famine is now occurring in northern Gaza and expected to expand to southern Gaza' (2025) finding ✓

    Full alert read. US-funded early-warning network; classification reached jointly with the IPC and reviewed by the FRC. Uses COGAT's entry data and CATI phone surveys (not statistically representative, argued to be biased downward). Quotes re-checked against the alert 2026-10-12: COGAT 'reported that no food entered the north from March to May'; 'The U.N. estimates at least 62,000 MT of food is required monthly in the Gaza Strip'; 'No new supplies entered the Gaza Strip in March and April, and only 151,059 MT of food was approved for entry from May 1 to August 17. A minimum of 372,000 MT of food is required over a six-month period'; 'direct quantitative evidence of mortality is unavailable'; 'Gaps in mortality data do not imply a low level of mortality on the ground'; the classification of 'Famine with reasonable evidence'. Also: 'The majority of U.N.-coordinated aid is intercepted by desperate civilians or looted by armed actors.'

    Open source ↗

    dataOCHA, Humanitarian Situation Update #326, Gaza Strip (September 2025) (2025) aggregate ✓

    Read 2026-10-01: 'according to MoH in Gaza, as of 19 September, 440 malnutrition-related deaths, including 147 children, were documented since October 2023'.

    Open source ↗

    advocacy / viewHuman Rights Watch, 'Israel's Crime of Extermination, Acts of Genocide in Gaza' (19 December 2024) (2024) finding ✓

    Read (2026-09-28): 'On October 9, then-Defense Minister Yoav Gallant announced a “complete siege” of Gaza, stating: “There will be no electricity, no food, no water, no fuel, everything is closed.”'; 'That same day, and for weeks thereafter, Israeli authorities cut off all water and blocked fuel, food, and humanitarian aid'. HRW also records that the 7 October attacks 'amounted to war crimes and crimes against humanity'.

    Open source ↗

    law / courtInternational Criminal Court, press release on the arrest warrants for Benjamin Netanyahu and Yoav Gallant (21 November 2024) (2024) finding ✓

    'the war crime of starvation as a method of warfare' (co-perpetrators), 8 Oct 2023 – at least 20 May 2024.

    Open source ↗

  3. More context

    What the declaration actually said — the fine print the headline dropped.

    The IPC classified 'famine with reasonable evidence': two of three thresholds measured directly, the third judged 'likely' from wider evidence. Food consumption was measured. Acute malnutrition was assessed through the MUAC (arm-circumference) route: at least 15% of children aged 6–59 months 'have a MUAC below 125 mm or oedema and evidence shows a rapid deterioration of underlying drivers of acute malnutrition' — an existing route, in the IPC Technical Manual since version 3.0 (2019) and in 3.1 (2021), by the IPC's own account and used for South Sudan (2020) and Sudan (2024); the claim that 'they changed the threshold for Gaza' does not hold as stated, since the route is in the manual and had been used before; but the manual (version 3.0, and version 3.1 of 2021 alike) allows the 15% MUAC route only for the category it calls 'Famine Likely', while for a full Famine classification 'Evidence for Nutritional Status only includes reliable data on GAM based on WHZ or oedema'. Version 3.1 has no category called 'Famine with reasonable evidence', the label used for Gaza. On the manual's own terms, then, the evidence the IPC describes fits 'Famine Likely': a classification for when 'minimally adequate evidence available indicates that a Famine may be occurring', which 'should trigger prompt action by decision-makers … while calling for urgent efforts to collect more evidence', with requirements 'less strict … for areas with limited or no humanitarian access', as Gaza was. Read that way, the IPC followed its manual's lower-evidence route, and the headline word 'confirmed' claimed more than that route does (unless later IPC guidance, not read here, changed the categories); though whether 15% MUAC matched 30% weight-for-height in Gaza's population is a fair methodological question — one the IPC's own fact sheet names: MUAC estimates are assessed 'to determine if a MUAC rate at or above 15% signals the same level of concern as a 30% acute malnutrition rate based on WHZ'. Mortality was not measured: 'Direct evidence on nutrition-related mortality is not available.' The analysis team 'could not conclude on mortality evidence and therefore was further analysed by the FRC' — the Famine Review Committee made that judgment itself, from indirect evidence, reasoning that the Health Ministry's counts 'are those that die in a health facility or whose body is taken to a health facility, missing deaths in the community', and concluding that this provides 'reasonable evidence that mortality thresholds for famine have been passed'. What the page said afterwards is set out below.

    primary documentIPC, 'Gaza Strip: Famine confirmed in Gaza Governorate, projected to expand' (Famine Review Committee, August 2025) (2025) finding ✓

    IPC Gaza page read: 'As of 15 August 2025, Famine (IPC Phase 5)—with reasonable evidence—is confirmed in Gaza Governorate'; over half a million people in Phase 5, 1.07 million (54%) in Phase 4, 396,000 (20%) in Phase 3; projected to expand to Deir al-Balah and Khan Younis by end of September (641,000 in Phase 5). Page also lists the FRC addendum on acute-malnutrition data, the IPC's response to criticism, and the postponement of the October 2025 analysis. Page structure (saved copy): the content-block IDs show the response to criticism, the addendum and the postponement notice were created after the original statement; the latter two in yellow 'warning' boxes. The phrases this site quotes from it were read on 26 September 2026 (from copies the editor supplied and from the IPC site, which now refuses this site's fetch tool, 403 on 13 Oct 2026; no Wayback copy); they appear in the notes and are not all repeated on this card.

    Open source ↗

    primary documentIPC Manual — Famine Classification Special Additional Protocols (MUAC route; Technical Manual 3.1, 2021, p. 88) ✓

    Read 2026-10-14 (the interactive page 'IPC Famine Classification – Special Additional Protocols', IPC Manual 3.0, printed by the editor on 14 October 2026). 'Famine classification requires R2 direct evidence on all three outcomes'; for Famine, 'Evidence for Nutritional Status only includes reliable data on GAM based on WHZ or oedema'. 'Famine Likely classifications can be performed when evidence requirements for a Famine classification are not met but there is at least R1 (+ or -) direct evidence'; for Famine Likely, 'Evidence for Nutritional Status includes GAM based on WHZ or MUAC, including oedema. The cut-off for GAM based on WHZ for Famine Likely classification is 30 percent, whereas for GAM based on MUAC the cut-off is 15 percent'. Mortality: 'Death rates need to be directly attributable to outright starvation or to the interaction of food consumption deficits and disease; all deaths due to trauma should therefore be discounted'. Version 3.1 (2021) says the same (card 'ipc-manual-3-1').

    Open source ↗

    primary documentIPC, Response to Issues Raised Following IPC's Famine Classification in the Gaza Strip (2025) finding ✓

    Read in full (30 August 2025). Seven questions: MUAC route and the 1.9 WHZ/MUAC ratio from a regional study; programme-based screenings; July split into halves (12.5% whole month; 10.1% → 16.4%); MoH 240 deaths by 15 Aug and 'expert judgement'; two phone surveys (36% vs 12% very severe HHS); projections and aid surges; parties excluded. The phrases this site quotes from it were read on 26 September 2026 (from copies the editor supplied and from the IPC site, which now refuses this site's fetch tool, 403 on 13 Oct 2026; no Wayback copy); they appear in the notes and are not all repeated on this card. Re-read in full 2026-10-14 from a copy the editor supplied: 'The IPC has not changed its standards for famine classification in Gaza'; MUAC route: 'At least 15% of children aged 6–59 months have a MUAC below 125 mm or oedema and evidence shows a rapid deterioration of underlying drivers of acute malnutrition'; 'famine is not classified based on MUAC alone'; 'a regional study (Leidman et al.), indicating a WHZ-to-MUAC ratio of 1.9'; July in Gaza Governorate 'stood at 12.5%. However, calculating a monthly weighted average can mask rapid changes', 'from 10.1% to 16.4% within three weeks'; 'By 15 August 2025, the Ministry of Health had reported 240 deaths officially attributed to starvation'; the two food-security sources: 'Both sources were based on telephone-based surveys' (its own Table 1 lists Source 2's modality as 'Mixed method'), Source 1 'considered more robust, due to sampling and its provision of longitudinal data'; very severe HHS '12% in Gaza Governorate according to Source 2, and 36% for Source 1. However, it is important to note that the prevalence of poor food consumption, as measured by the FCS, was higher in Source 2 (100%) than in Source 1 (87%)'; 'to safeguard impartiality, parties to the conflict were not involved in the analytical process'. Its opening: 'These issues should not divert attention from the most critical matter of stopping the Famine'.

    Open source ↗

    primary documentIPC, Addendum to the FRC report of 22 August 2025 (published 30 August 2025) (2025) ✓

    Read in full: seven modifications — table titles 16–21 retitled to state data dates and users; table 18(b) (MUAC results reported up to 12/08/2025, utilised by the FRC) added as 'missing'. The phrases this site quotes from it were read on 26 September 2026 (from copies the editor supplied and from the IPC site, which now refuses this site's fetch tool, 403 on 13 Oct 2026; no Wayback copy); they appear in the notes and are not all repeated on this card.

    Open source ↗

    primary documentIPC Global Famine Review Committee, 'Gaza Strip' report (22 August 2025) (2025) finding ✓

    Read in full (text extracted). Summary table: for Gaza Governorate 'The analysis team could not conclude on mortality evidence and therefore was further analysed by the FRC'; 'Direct evidence on nutrition-related mortality is not available'; for Deir al-Balah and Khan Younis, acute malnutrition and mortality thresholds not met in the current period. The phrases this site quotes from it were read on 26 September 2026 (from copies the editor supplied and from the IPC site, which now refuses this site's fetch tool, 403 on 13 Oct 2026; no Wayback copy); they appear in the notes and are not all repeated on this card. Re-read 2026-10-14 from a copy the editor supplied (two-column layout): 'Direct evidence on nutrition-related mortality is not available'; 'The analysis team could not conclude on mortality evidence and therefore was further analysed by the FRC'; 'The MoH mortality counts are those that die in a health facility or whose body is taken to a health facility, missing deaths in the community'; malnutrition deaths reported by the Ministry of Health, 'providing reasonable evidence that mortality thresholds for famine have been passed'. Also (read 2026-10-14): 'Between 2 March and 19 May 2025 no food entered the Gaza Strip by any actor or supplier (humanitarian or commercial)'; the food required to feed the population 'has not been met since April 2025'; 'the number of households reportedly facing potential safety risks while searching for food has more than doubled to around 80% of households'.

    Open source ↗

    primary documentIPC Global Partners, Integrated Food Security Phase Classification Technical Manual Version 3.1 (2021), famine protocols, pp. 85–87 (2021) ✓

    Read 2026-10-14 (full PDF supplied by the editor). 'Version 3.1 should be applied to all analyses going forward'. Figure 61 and notes, p. 86: 'Direct evidence for Acute Malnutrition includes GAM based on WHZ. The Famine threshold for GAM based on WHZ is 30 percent and above. In the absence of data on WHZ, for Famine Likely classifications, data on MUAC can be used together with an understanding of the relationship between WHZ and MUAC in the area of analysis supported by at least two indicators with R1 evidence on the immediate causes of malnutrition to confirm MUAC findings.' P. 87, for Famine: 'Evidence for Nutritional Status only includes reliable data on GAM based on WHZ or oedema'; for Famine Likely: 'Evidence for Nutritional Status includes GAM based on WHZ or MUAC, including oedema. The cut-off for GAM based on WHZ for Famine Likely classification is 30 percent, whereas for GAM based on MUAC the cut-off is 15 percent'. 'The IPC permits the classification as Famine Likely when all regular and special protocols are met, except for the existence of reliable evidence for the three outcomes. Areas can be classified as Famine Likely if minimally adequate evidence available indicates that a Famine may be occurring or will occur. When an area is classified as Famine Likely, it should trigger prompt action by decision-makers to address the situation while calling for urgent efforts to collect more evidence.' Requirements are 'less strict … for areas with limited or no humanitarian access'. The manual contains no category called 'Famine with reasonable evidence' (the phrase 'Reasonable evidence' appears only in its evidence-reliability scale). Later IPC guidance, if any, is not read here.

    How the IPC page changed, in order.

    The famine page on the IPC's site was edited after the headlines went round the world; the later additions, in order: (1) 22 August 2025 — the finding ('Famine (IPC Phase 5)—with reasonable evidence—is confirmed in Gaza Governorate'), the Famine Review Committee's review, the FEWS NET alert and a fact sheet, published together. (2) Afterwards — a 'Response to Issues Raised', because 'several questions have been raised about how the analysis was conducted', set in a highlighted box that adds 'these issues must not distract from the most urgent matter'. (The response document itself words it 'should not divert attention from the most critical matter of stopping the Famine'.) (3) 30 August 2025 — an addendum 'to clarify the data on Acute Malnutrition used by the FRC': it retitles six annex tables to say which screening data they show (data reported up to 31 July, used by the analysis team) and adds a table that 'was missing' — results reported up to 12 August 'and utilised by the Famine Review Committee', so the committee's review rested on a later dataset than the analysis it reviewed. (4) Last — a notice that the next analysis, planned for early October 2025, was postponed because 'it was not feasible to conduct a reliable analysis at this time', placed in a highlighted box at the very top, above the original finding. How the order is known: dates in the documents themselves, and on a saved copy of the page the content blocks' increasing ID numbers, which record the order in which they were created (see 'How this site is made'). The Committee's own words were a warning as much as a finding: 'As this Famine is entirely man-made, it can be halted and reversed', and without a ceasefire and restored supplies 'avoidable deaths will increase exponentially'.

    primary documentIPC, 'Gaza Strip: Famine confirmed in Gaza Governorate, projected to expand' (Famine Review Committee, August 2025) (2025) finding ✓

    IPC Gaza page read: 'As of 15 August 2025, Famine (IPC Phase 5)—with reasonable evidence—is confirmed in Gaza Governorate'; over half a million people in Phase 5, 1.07 million (54%) in Phase 4, 396,000 (20%) in Phase 3; projected to expand to Deir al-Balah and Khan Younis by end of September (641,000 in Phase 5). Page also lists the FRC addendum on acute-malnutrition data, the IPC's response to criticism, and the postponement of the October 2025 analysis. Page structure (saved copy): the content-block IDs show the response to criticism, the addendum and the postponement notice were created after the original statement; the latter two in yellow 'warning' boxes. The phrases this site quotes from it were read on 26 September 2026 (from copies the editor supplied and from the IPC site, which now refuses this site's fetch tool, 403 on 13 Oct 2026; no Wayback copy); they appear in the notes and are not all repeated on this card.

    Open source ↗

    primary documentIPC, Addendum to the FRC report of 22 August 2025 (published 30 August 2025) (2025) ✓

    Read in full: seven modifications — table titles 16–21 retitled to state data dates and users; table 18(b) (MUAC results reported up to 12/08/2025, utilised by the FRC) added as 'missing'. The phrases this site quotes from it were read on 26 September 2026 (from copies the editor supplied and from the IPC site, which now refuses this site's fetch tool, 403 on 13 Oct 2026; no Wayback copy); they appear in the notes and are not all repeated on this card.

    Open source ↗

    primary documentIPC, Response to Issues Raised Following IPC's Famine Classification in the Gaza Strip (2025) finding ✓

    Read in full (30 August 2025). Seven questions: MUAC route and the 1.9 WHZ/MUAC ratio from a regional study; programme-based screenings; July split into halves (12.5% whole month; 10.1% → 16.4%); MoH 240 deaths by 15 Aug and 'expert judgement'; two phone surveys (36% vs 12% very severe HHS); projections and aid surges; parties excluded. The phrases this site quotes from it were read on 26 September 2026 (from copies the editor supplied and from the IPC site, which now refuses this site's fetch tool, 403 on 13 Oct 2026; no Wayback copy); they appear in the notes and are not all repeated on this card. Re-read in full 2026-10-14 from a copy the editor supplied: 'The IPC has not changed its standards for famine classification in Gaza'; MUAC route: 'At least 15% of children aged 6–59 months have a MUAC below 125 mm or oedema and evidence shows a rapid deterioration of underlying drivers of acute malnutrition'; 'famine is not classified based on MUAC alone'; 'a regional study (Leidman et al.), indicating a WHZ-to-MUAC ratio of 1.9'; July in Gaza Governorate 'stood at 12.5%. However, calculating a monthly weighted average can mask rapid changes', 'from 10.1% to 16.4% within three weeks'; 'By 15 August 2025, the Ministry of Health had reported 240 deaths officially attributed to starvation'; the two food-security sources: 'Both sources were based on telephone-based surveys' (its own Table 1 lists Source 2's modality as 'Mixed method'), Source 1 'considered more robust, due to sampling and its provision of longitudinal data'; very severe HHS '12% in Gaza Governorate according to Source 2, and 36% for Source 1. However, it is important to note that the prevalence of poor food consumption, as measured by the FCS, was higher in Source 2 (100%) than in Source 1 (87%)'; 'to safeguard impartiality, parties to the conflict were not involved in the analytical process'. Its opening: 'These issues should not divert attention from the most critical matter of stopping the Famine'.

    Open source ↗

    primary documentFEWS NET, 'Gaza Alert August 22, 2025: Famine is now occurring in northern Gaza and expected to expand to southern Gaza' (2025) finding ✓

    Full alert read. US-funded early-warning network; classification reached jointly with the IPC and reviewed by the FRC. Uses COGAT's entry data and CATI phone surveys (not statistically representative, argued to be biased downward). Quotes re-checked against the alert 2026-10-12: COGAT 'reported that no food entered the north from March to May'; 'The U.N. estimates at least 62,000 MT of food is required monthly in the Gaza Strip'; 'No new supplies entered the Gaza Strip in March and April, and only 151,059 MT of food was approved for entry from May 1 to August 17. A minimum of 372,000 MT of food is required over a six-month period'; 'direct quantitative evidence of mortality is unavailable'; 'Gaps in mortality data do not imply a low level of mortality on the ground'; the classification of 'Famine with reasonable evidence'. Also: 'The majority of U.N.-coordinated aid is intercepted by desperate civilians or looted by armed actors.'

    Open source ↗

    What the IPC answered its critics.

    On 30 August 2025 the IPC answered the main criticisms in writing. Its own answers carry the evidence both sides need. On the malnutrition threshold: the IPC says arm-circumference (MUAC) screening has been an accepted route since 2019 (its manual, versions 3.0 and 3.1, lists that route for 'Famine Likely', see above), used for South Sudan (2020) and Sudan (2024); but in Gaza 'there is no survey data available that would allow for a direct comparison' with the weight-for-height measure, so the committee used a ratio from a regional study (1.9). On the data: the screenings were programme-based, around hospitals, and the IPC argues they 'likely underestimated' severity because malnourished children were taken out of later rounds and others were already receiving supplements. On the crossing itself: across the whole of July, malnutrition by MUAC in Gaza Governorate stood at 12.5%, below the 15% threshold; split into two halves it rose from 10.1% to 16.4%, and the IPC argues that a monthly average 'can mask rapid changes'. On deaths: the Ministry of Health had reported 240 starvation-related deaths by 15 August, and 'analysts rely on available evidence and expert judgement' to judge the mortality threshold. On contradictory data: of two telephone surveys, one found 36% of households in the most extreme hunger category and the other 12%; the IPC relied on the first as larger and more consistent, though the second found poor food consumption higher (100% against 87%). On projections that proved too severe: aid tends to surge after each IPC alert and fall again, 'a recurring pattern'. And neither party to the conflict took part in the analysis, 'to safeguard impartiality'. Read together, the answers show a classification made close to its own thresholds, with each choice made there stated and argued for in the IPC's own words.

    primary documentIPC, Response to Issues Raised Following IPC's Famine Classification in the Gaza Strip (2025) finding ✓

    Read in full (30 August 2025). Seven questions: MUAC route and the 1.9 WHZ/MUAC ratio from a regional study; programme-based screenings; July split into halves (12.5% whole month; 10.1% → 16.4%); MoH 240 deaths by 15 Aug and 'expert judgement'; two phone surveys (36% vs 12% very severe HHS); projections and aid surges; parties excluded. The phrases this site quotes from it were read on 26 September 2026 (from copies the editor supplied and from the IPC site, which now refuses this site's fetch tool, 403 on 13 Oct 2026; no Wayback copy); they appear in the notes and are not all repeated on this card. Re-read in full 2026-10-14 from a copy the editor supplied: 'The IPC has not changed its standards for famine classification in Gaza'; MUAC route: 'At least 15% of children aged 6–59 months have a MUAC below 125 mm or oedema and evidence shows a rapid deterioration of underlying drivers of acute malnutrition'; 'famine is not classified based on MUAC alone'; 'a regional study (Leidman et al.), indicating a WHZ-to-MUAC ratio of 1.9'; July in Gaza Governorate 'stood at 12.5%. However, calculating a monthly weighted average can mask rapid changes', 'from 10.1% to 16.4% within three weeks'; 'By 15 August 2025, the Ministry of Health had reported 240 deaths officially attributed to starvation'; the two food-security sources: 'Both sources were based on telephone-based surveys' (its own Table 1 lists Source 2's modality as 'Mixed method'), Source 1 'considered more robust, due to sampling and its provision of longitudinal data'; very severe HHS '12% in Gaza Governorate according to Source 2, and 36% for Source 1. However, it is important to note that the prevalence of poor food consumption, as measured by the FCS, was higher in Source 2 (100%) than in Source 1 (87%)'; 'to safeguard impartiality, parties to the conflict were not involved in the analytical process'. Its opening: 'These issues should not divert attention from the most critical matter of stopping the Famine'.

    Open source ↗

    The outcome test: famine-level mortality doesn't appear in the counts — including the Ministry of Health's own.

    The famine mortality threshold is at least 2 non-trauma deaths per 10,000 people per day. For the more than half a million people the IPC placed in famine conditions, that implies on the order of 100 deaths a day (2 per 10,000 × 500,000; this page's arithmetic) — thousands over a few weeks. The Gaza Ministry of Health recorded 185 malnutrition deaths across all of Gaza in August 2025 (about 6 a day); OCHA, citing the Ministry, gives cumulative totals of 440 by 19 September and 461 by 7 October 2025, which is 21 in those eighteen days. Monitoring was disrupted and disease deaths aren't always attributed to hunger. How much that matters is the core dispute: famine researcher Alex de Waal (World Peace Foundation) argues that deaths attributed directly to malnutrition 'are always a major undercount by an order of magnitude or more', because disease is the main killer in famines, and puts excess non-trauma deaths at roughly 12,700 to over 50,000, against the Ministry's 269 recorded malnutrition deaths by 20 August 2025. If he is right, the gap can close; the later counts and surveys are the test of it. After the October 2025 ceasefire and a surge of aid, UNICEF's SMART survey (fieldwork 31 May – 15 June 2026, covering the accessible areas where about 83% of Gazans live) reported that 'acute malnutrition among children under five was low' and stunting at 12.2%; the survey's own executive summary puts acute malnutrition by weight-for-height at '0.5 per cent' and combined acute malnutrition (including arm-circumference and oedema cases) at '1.3 per cent'. An earlier version of this page gave '0.2–0.8%, below Egypt and Jordan', from a summary whose source was never traced; neither that range nor the comparison with neighbouring countries is in the survey. The strongest answer to this test comes from the famine analysts themselves. FEWS NET, which classified the famine jointly with the IPC, wrote that 'direct quantitative evidence of mortality is unavailable' and that 'Gaps in mortality data do not imply a low level of mortality on the ground': deaths outside health facilities go uncounted, and in past famines mortality crossed the threshold '4-12 weeks' after malnutrition did and stayed high for weeks after aid was scaled up. That is an argument about what the counts can't see; whether the later counts and surveys are consistent with it is the open question. Where that leaves it: severe hunger and acute malnutrition were measured; famine-level mortality was not — the review committee inferred it, the Ministry's facility counts don't show it, and famine researchers expect those counts not to. No population mortality survey covering July–September 2025 was found in a search on 30 September 2026 (the household survey published in The Lancet Global Health in February 2026 ends in January 2025, and the journal's accompanying commentary discusses that same survey), so whether hunger killed on the scale the classification implies is still an open question, not a settled one either way.

    dataGaza Ministry of Health malnutrition death counts, as reported (Al Jazeera, Aug–Sep 2025) (2025) aggregate ✓

    Al Jazeera, 2 September 2025 (read), citing the Ministry of Health: 185 deaths 'due to malnutrition' in August 2025; cumulative 361 including 130 children since October 2023. The report uses the phrase 'Israel's genocidal war' in its own voice.

    Open source ↗

    dataOCHA, Humanitarian Situation Update #326, Gaza Strip (September 2025) (2025) aggregate ✓

    Read 2026-10-01: 'according to MoH in Gaza, as of 19 September, 440 malnutrition-related deaths, including 147 children, were documented since October 2023'.

    Open source ↗

    dataOCHA, Humanitarian Situation Update #329, Gaza Strip (October 2025) (2025) aggregate ✓

    Read 2026-10-01: 'according to MoH in Gaza, as of 7 October, 461 malnutrition-related deaths, including 157 children, were documented since October 2023'.

    Open source ↗

    dataUNICEF / Nutrition Cluster, Gaza SMART Survey 2026 (2026) aggregate ✓

    Published 24 August 2026. Fieldwork 31 May–15 June 2026, accessible areas only (about 17% of the population, ~358,000 people, excluded for access and security). Release: 'Acute malnutrition among children under five was low'; stunting 12.2%, overweight 4.0%. No GAM percentage in the release.

    Open source ↗

    advocacy / viewWorld Peace Foundation, 'How Many People Have Died of Famine in Gaza?' finding ✓

    Alex de Waal, World Peace Foundation (updated 2025): malnutrition-attributed deaths 'are always a major undercount by an order of magnitude or more'; excess non-trauma deaths estimated ~12,675 to over 50,000; MoH 269 malnutrition deaths as of 20 Aug 2025.

    Open source ↗

    primary documentFEWS NET, 'Gaza Alert August 22, 2025: Famine is now occurring in northern Gaza and expected to expand to southern Gaza' (2025) finding ✓

    Full alert read. US-funded early-warning network; classification reached jointly with the IPC and reviewed by the FRC. Uses COGAT's entry data and CATI phone surveys (not statistically representative, argued to be biased downward). Quotes re-checked against the alert 2026-10-12: COGAT 'reported that no food entered the north from March to May'; 'The U.N. estimates at least 62,000 MT of food is required monthly in the Gaza Strip'; 'No new supplies entered the Gaza Strip in March and April, and only 151,059 MT of food was approved for entry from May 1 to August 17. A minimum of 372,000 MT of food is required over a six-month period'; 'direct quantitative evidence of mortality is unavailable'; 'Gaps in mortality data do not imply a low level of mortality on the ground'; the classification of 'Famine with reasonable evidence'. Also: 'The majority of U.N.-coordinated aid is intercepted by desperate civilians or looted by armed actors.'

    Open source ↗

    dataState of Palestine Nutrition Cluster / UNICEF, 'Gaza SMART Survey 2026', executive summary (PDF) (2026) aggregate ✓

    Read (2026-09-28): 'Among children aged 0–59 months, Global Acute Malnutrition (GAM) based on weight-for-height Z-score was estimated at 0.5 per cent'; 'Combined Global Acute Malnutrition, based on weight-for-height, MUAC where applicable, and/or bilateral pitting edema, was estimated at 1.3 per cent'; stunting 12.2% (95% CI 10.4–14.2); 7.2% of pregnant and lactating women acutely malnourished by MUAC; the excluded areas' situation 'cannot be inferred from this survey'. No comparison with Egypt or Jordan.

    Open source ↗

    How the numbers were used.

    'Famine confirmed' presented a warning system's classification as an established outcome; figures such as 'every child under five at risk' circulated without denominators or evidence levels; and a retort such as 'so you support starving children?' answers a question about the mortality evidence with a moral charge (see 'Motte and bailey'); how often it was used is not counted here. Intent differs by actor and should not be assumed: the IPC is built to warn before mass death, and a famine averted by aid can look like a false alarm afterwards. The narrower charge the record supports is about the wording: 'confirmed', the IPC's own word, covers a mortality finding the committee inferred rather than measured. The mirror claim, 'the famine was fake', erases the real hunger and deaths just as badly. The denial was just as quick on the other side: on 27 July 2025 Israel's prime minister said 'There is no policy of starvation in Gaza. And there is no starvation in Gaza'; PolitiFact rated the second sentence 'Pants on Fire', citing the WHO's count of 63 malnutrition-related deaths in July, and noted that the Ministry's own figures 'cannot be independently verified'. Where the aid went is a separate question with its own evidence (see 'Aid to Gaza feeds Hamas — they turn anything into weapons.' and 'The GHF aid sites were death traps — aid used as a weapon of genocide.').

    pressPolitiFact (Madison Czopek), 'Netanyahu’s ‘no starvation in Gaza’ statement contradicts extensive evidence of hunger crisis' (29 July 2025) (2025) finding ✓

    Read 2026-10-01. Rating: Pants on Fire, for 'There is no starvation in Gaza.' Netanyahu, 27 July 2025, at a Jerusalem event sponsored by Daystar: 'Israel is presented as though we are applying a campaign of starvation in Gaza … What a bold-faced lie. There is no policy of starvation in Gaza. And there is no starvation in Gaza.' Army spokesperson Effie Defrin told NPR the photos were part of a Hamas campaign to create 'an image of starvation, which doesn’t exist.' Evidence cited: WHO, 63 malnutrition-related deaths in July (24 children under 5); WFP, one-third of Gazans not eating for days. Other half: the Ministry of Health's figures 'cannot be independently verified', though the UN, WHO and casualty trackers 'cautioned against outright dismissal'. PolitiFact contacted the Israeli embassy in the US and received no response.

    Open source ↗

    Reading the numbers

    The argument leans on one word, 'famine', and the figures behind it can be read several ways. (1) Classification against measurement: the IPC measured food consumption and acute malnutrition directly, and judged the mortality threshold 'likely'; its own review committee wrote that 'direct evidence on nutrition-related mortality is not available'. (2) What the threshold implies: famine mortality means at least 2 non-trauma deaths per 10,000 people per day. For the roughly 500,000 people placed in famine conditions, that is about 100 a day (this page's arithmetic). Normal mortality in that population would be about 4 a day, at the pre-war rate of about 3 per 1,000 a year (World Bank), so famine level means roughly 96 extra non-trauma deaths a day. (3) What was recorded: the Gaza Ministry of Health counted 185 malnutrition deaths across all of Gaza in August 2025, about 6 a day, roughly a sixteenth of that excess, and for a larger population. (4) The worst case for each side. For those who say famine-level death happened: the recorded deaths fall short unless the true toll was about sixteen times higher; Alex de Waal argues that malnutrition-attributed deaths 'are always a major undercount by an order of magnitude or more', and the threshold counts all non-trauma deaths, including disease, not only those attributed to hunger. The one household survey on this site gives a measured check on how far counts fall short. The Lancet Global Health survey (2,000 households, fieldwork ending 5 January 2025) estimated 16,300 non-violent deaths from October 2023, 8,540 of them above what pre-war rates would predict. That is about 19 excess deaths a day across all of Gaza over 456 days, already about 19 times the Ministry's 440 malnutrition deaths counted by September 2025 (this page's arithmetic): non-trauma deaths, which the famine threshold counts, run far above the hunger-attributed count, as de Waal argues. But as a rate it is under 0.1 excess deaths per 10,000 people a day (on a population of about 2.3 million), roughly a twentieth of the famine threshold's roughly 1.9 excess (this page's arithmetic), for an earlier period and the whole population rather than the famine areas in August 2025. So famine-level mortality is not shown by the counts, and not ruled out by them either: the survey shows the counts are a large undercount of non-trauma deaths, and also that, on average up to January 2025, those deaths ran far below famine level. For those who say there was no starvation: nearly 17,000 children were admitted for acute malnutrition in August 2025, and the Ministry counted 440 malnutrition deaths by 19 September, 147 of them children; that claim fails on any reading. (5) Afterwards: a UNICEF survey in May–June 2026 found acute malnutrition of 1.3% among young children in the areas it could reach; acute malnutrition recovers within months, so that shows recovery, not what happened in 2025. The slower measure tells more: stunting, low height for age, builds over months of poor nutrition and is slow to reverse. The 2026 survey found 12.2% (95% range 10.4–14.2%) of young children stunted in Gaza; before the war, UNICEF's 2019/20 survey found 8.7% for the West Bank and Gaza together. Even the low end of the 2026 range is above the pre-war figure, though the two surveys differ in area and the 2026 one left out about a sixth of the population it could not reach. That points to a real and lasting period of undernutrition; it does not measure deaths. Holds on every reading: severe hunger and malnutrition were real, stunting rose above its pre-war level, and the Ministry counted children among those who died of it; 'there is no starvation' is false; the recorded deaths are far below the famine mortality threshold. Holds only if non-trauma deaths were undercounted by about sixteen times or more: that famine-level mortality occurred. Not supported by any reading: that 'famine confirmed' meant mass death was measured, or that the hunger was invented (see 'Reading war numbers' and 'Proxy presented as outcome').

    What the labels (Holds, Open, Arguable …) mean

    primary documentIPC Global Famine Review Committee, 'Gaza Strip' report (22 August 2025) (2025) finding ✓

    Read in full (text extracted). Summary table: for Gaza Governorate 'The analysis team could not conclude on mortality evidence and therefore was further analysed by the FRC'; 'Direct evidence on nutrition-related mortality is not available'; for Deir al-Balah and Khan Younis, acute malnutrition and mortality thresholds not met in the current period. The phrases this site quotes from it were read on 26 September 2026 (from copies the editor supplied and from the IPC site, which now refuses this site's fetch tool, 403 on 13 Oct 2026; no Wayback copy); they appear in the notes and are not all repeated on this card. Re-read 2026-10-14 from a copy the editor supplied (two-column layout): 'Direct evidence on nutrition-related mortality is not available'; 'The analysis team could not conclude on mortality evidence and therefore was further analysed by the FRC'; 'The MoH mortality counts are those that die in a health facility or whose body is taken to a health facility, missing deaths in the community'; malnutrition deaths reported by the Ministry of Health, 'providing reasonable evidence that mortality thresholds for famine have been passed'. Also (read 2026-10-14): 'Between 2 March and 19 May 2025 no food entered the Gaza Strip by any actor or supplier (humanitarian or commercial)'; the food required to feed the population 'has not been met since April 2025'; 'the number of households reportedly facing potential safety risks while searching for food has more than doubled to around 80% of households'.

    Open source ↗

    primary documentIPC, 'Gaza Strip: Famine confirmed in Gaza Governorate, projected to expand' (Famine Review Committee, August 2025) (2025) finding ✓

    IPC Gaza page read: 'As of 15 August 2025, Famine (IPC Phase 5)—with reasonable evidence—is confirmed in Gaza Governorate'; over half a million people in Phase 5, 1.07 million (54%) in Phase 4, 396,000 (20%) in Phase 3; projected to expand to Deir al-Balah and Khan Younis by end of September (641,000 in Phase 5). Page also lists the FRC addendum on acute-malnutrition data, the IPC's response to criticism, and the postponement of the October 2025 analysis. Page structure (saved copy): the content-block IDs show the response to criticism, the addendum and the postponement notice were created after the original statement; the latter two in yellow 'warning' boxes. The phrases this site quotes from it were read on 26 September 2026 (from copies the editor supplied and from the IPC site, which now refuses this site's fetch tool, 403 on 13 Oct 2026; no Wayback copy); they appear in the notes and are not all repeated on this card.

    Open source ↗

    dataWorld Bank, World Development Indicators: 'Death rate, crude (per 1,000 people)', West Bank and Gaza aggregate ✓

    Read via the World Bank's data API, 2026-10-10: 2018 3.217; 2019 3.134; 2020 3.391; 2021 3.768; 2022 2.992 (last pre-war year); 2023 7.198 (includes the war). Covers the West Bank and Gaza together, not Gaza alone.

    Open source ↗

    dataGaza Ministry of Health malnutrition death counts, as reported (Al Jazeera, Aug–Sep 2025) (2025) aggregate ✓

    Al Jazeera, 2 September 2025 (read), citing the Ministry of Health: 185 deaths 'due to malnutrition' in August 2025; cumulative 361 including 130 children since October 2023. The report uses the phrase 'Israel's genocidal war' in its own voice.

    Open source ↗

    advocacy / viewWorld Peace Foundation, 'How Many People Have Died of Famine in Gaza?' finding ✓

    Alex de Waal, World Peace Foundation (updated 2025): malnutrition-attributed deaths 'are always a major undercount by an order of magnitude or more'; excess non-trauma deaths estimated ~12,675 to over 50,000; MoH 269 malnutrition deaths as of 20 Aug 2025.

    Open source ↗

    dataOCHA, Humanitarian Situation Update #326, Gaza Strip (September 2025) (2025) aggregate ✓

    Read 2026-10-01: 'according to MoH in Gaza, as of 19 September, 440 malnutrition-related deaths, including 147 children, were documented since October 2023'.

    Open source ↗

    dataState of Palestine Nutrition Cluster / UNICEF, 'Gaza SMART Survey 2026', executive summary (PDF) (2026) aggregate ✓

    Read (2026-09-28): 'Among children aged 0–59 months, Global Acute Malnutrition (GAM) based on weight-for-height Z-score was estimated at 0.5 per cent'; 'Combined Global Acute Malnutrition, based on weight-for-height, MUAC where applicable, and/or bilateral pitting edema, was estimated at 1.3 per cent'; stunting 12.2% (95% CI 10.4–14.2); 7.2% of pregnant and lactating women acutely malnourished by MUAC; the excluded areas' situation 'cannot be inferred from this survey'. No comparison with Egypt or Jordan.

    Open source ↗

    dataUNICEF State of Palestine, 'Children in the State of Palestine': child development data from the 2019/2020 Multiple Indicator Cluster Survey (Child Rights Brief, December 2021) (2021) aggregate ✓

    Read, 2026-10-10. 'There were noticeably more stunted children under five years old in 2019/20 compared to 2014. 8.7% were significantly below average height for their age, compared to 7.4% in 2014'; stunting among the poorest children 10.5% against 6.4% among the wealthiest. The stunting figure is for the State of Palestine as a whole (West Bank and Gaza); a separate Gaza figure is not in the passages read here.

    Open source ↗

    data'Violent and non-violent death tolls for the Gaza conflict: new primary evidence from a population-representative field survey', The Lancet Global Health (2026) (2026) aggregate ✓

    Spagat, Pedersen, Shikaki, Robbins, Bendavid, Hegre, Guha-Sapir, The Lancet Global Health 14 (April 2026). Read in full: 2,000 households, 30 Dec 2024 – 5 Jan 2025; 75,200 (63,600–86,800) violent deaths; 56.2% women, children, over-64s; 12,200 missing; 16,300 non-violent deaths, 8,540 excess; MoH 34.7% below the central estimate. Abstract re-read via OpenAlex 2026-10-12: 'The MoH figure for this period (49 090 violent deaths) was 34·7% below our central estimate'; 'violent deaths have substantially exceeded official figures whereas the demographic composition of casualties aligns with MoH reporting. Non-violent excess deaths, although substantial, are lower than some projections have suggested.' From the abstract: women, children and over-64s were '56·2% (95% CI 50·4-61·9) of violent deaths, totalling 42 200 deaths (95% CI 33 100-51 300)'. The age-and-sex figures used on counting-the-dead (32,900 men and 16,600 women aged 18–64 killed; 8,470 of the 12,200 missing being men of that age) come from the earlier full reading; the full text is walled today (2026-10-12) and they are not re-checked.

    Open source ↗

    How to talk about this.

    Acknowledge first: hunger in Gaza was real and severe, children were malnourished, and restrictions on aid were a cause. Ask one question: 'what did the famine classification measure, and what did it project?'; it separates the suffering from the headline. Share one checkable thing: the classification's own fine print, which distinguishes measured food consumption and malnutrition from inferred mortality. What not to say: 'the famine was fake', which denies the hunger, or 'Israel is starving Gaza on purpose' as a settled fact, which skips the questions of intent and of where the aid went. From the other side: grant the documented looting of aid and the reported diversion of dual-use goods (see 'Aid to Gaza feeds Hamas — they turn anything into weapons.' for what is and isn't sourced), then ask what an aid system that feeds people despite it would need.

  4. Where it leads if pressed

    Judge headlines by outcomes.

    Both sides can accept: civilians went hungry, aid restriction caused harm, and the headline 'Famine confirmed' carried more certainty than the classification's own fine print, which rests its mortality finding on inference. The lesson for readers is general — ask whether a figure is a projection, a proxy or a measured outcome, and check it later against what actually happened.

The usual reply — and its own traps

Arguments travel in pairs. Here is what the other side typically answers. Check whether both lean on the same trick.

heard from pro-Palestinian voices“Israel is committing genocide in Gaza.” Both lean on: ◆ Proxy presented as outcome ◆ The authority endpoint ◆ "It's not complicated"

Where this argument points

This argument comes up in the debate about these solutions — for, against or around them:

On the timeline

Keep exploring

From the archive

How newspapers reported related events at the time — suggested by the archive's own tags, not yet reviewed for this argument. Open a scan and judge for yourself.

Scans: kvetch-of-the-day news archive.