Was There a Famine in Gaza? Alex de Waal Interview Fact-Check

⬅️ Sources, Books & Media (MOC) · Arguments: Blockade in Gaza · Gaza Casualty Figures · Related audit: UN admits Hamas is stealing aid in Gaza; Natasha Hausdorff considers implications

Bottom line

There are three different historical claims, and collapsing them creates most of the confusion.

1. 2024: the IPC Famine Review Committee (FRC) projected imminent famine in northern Gaza, but no retrospective famine classification was confirmed. When FEWS NET later classified famine for April, the FRC declined to endorse it because the evidence did not converge. That means “not established,” not “the FRC proved there was no famine.”
2. 2025: the IPC/FRC did classify Gaza Governorate in Famine (IPC Phase 5) for the July–15 August period with reasonable evidence. Food deprivation and acute malnutrition crossed their thresholds directly; the non-trauma mortality threshold was inferred from the broader evidence, not directly measured. This was a narrower claim than “all Gaza was in famine throughout the war,” but it was a real IPC famine classification.
3. Deaths: no reliable study yet supplies a famine-attributable death count for the 2025 classified period. The best household mortality survey covers only through 5 January 2025. It estimated 8,540 excess non-violent deaths (95% CI 4,540–12,500), but did not classify those deaths as starvation deaths. The often-repeated 62,413 starvation deaths estimate was generated by an invalid inference from IPC household phases and should be retired.

The transcript is strongest when it distinguishes acute malnutrition from food insecurity, demands mortality evidence, rejects the 62,413 estimate, and emphasizes how quickly food access changed when policy changed. It is weakest when de Waal lets famine’s social meaning substitute for the IPC’s technical mortality condition, and when the host treats missing direct mortality data as affirmative proof that no famine occurred.

Source and evidence boundary

• Episode: “Was there a famine in Gaza? The top expert answers — Alex de Waal”, The Comedy Cellar: Live from the Table, published 24 September 2026.
• Participants: Noam Dworman (host), Periel Aschenbrand (producer), Alex de Waal (World Peace Foundation / Fletcher School).
• Transcript supplied: approximately 81 minutes of substantive conversation. The podcast listing reports a longer runtime, so this note audits the supplied transcript rather than assuming it covers every second of the uploaded program.
• Method: claims were checked against IPC/FRC and FEWS NET reports, UNICEF/PCBS/WFP data, the peer-reviewed Gaza Mortality Survey, the speakers’ cited articles, and the aid-diversion sources discussed in the interview. Advocacy and belligerent sources are identified as such.
• Terminology: “Famine” below means the IPC area-level classification unless quotation marks or context indicate ordinary-language usage. Household Catastrophe (IPC Phase 5) is not itself an area-level famine declaration.

The answer in one table

Timeline: warning, classification, recovery

Claim-by-claim fact-check

1. “Pre-war Gaza was very well nourished” (4:16)

Verdict: substantially accurate for acute child malnutrition; misleading if expanded into “no serious food problem.”

The 2019–20 Palestinian MICS measured Gaza under-five wasting at 0.8%, underweight at 2.1%, and stunting at 9.0%—low acute malnutrition, but non-zero chronic nutritional deprivation. PCBS summary · MICS report. UNICEF later described global acute malnutrition as not having been a pre-war public-health concern in Gaza.

The host is also right that food insecurity is not synonymous with starvation or measured wasting. But his analogy to ordinary U.S. food insecurity goes too far. Before the war, WFP reported that two in three Gazans struggled to afford food, while the 2022 Global Report on Food Crises counted about 1.19 million acutely food-insecure people in Gaza. WFP, May 2023 · GRFC 2023. These were mainly problems of poverty, purchasing power, diet quality, and dependence—not evidence that half the population was already physiologically starving.

Debate-safe formulation

“Pre-war Gaza combined low measured acute malnutrition with high poverty, food insecurity, poor diet quality, and structural dependence on imports. ‘Well nourished’ is defensible for calories and low wasting; ‘swimming in food’ is not.”

2. Egypt could have supplied unlimited food and Israel could do nothing (9:14)

Verdict: false as stated.

Rafah is a Gaza–Egypt crossing, but the post-2005 crossing regime did not make it an unlimited commercial-goods corridor independent of Israel. Goods normally moved through Kerem Shalom under Israeli inspection, while Rafah was principally a passenger crossing; Egypt also restricted the border and destroyed tunnels. The host is right that Egypt has agency and imposed its own restrictions. He is wrong that Israel had no capacity to affect Gaza’s formal food-import system.

3. De Waal admitted that human-rights advocates use alarmist labels and that “the mud sticks” (13:10)

Verdict: authentic quotation, overextended inference.

In “Writing Human Rights and Getting It Wrong” (2016), de Waal describes wrongly reporting a man’s death, tactically using “genocide” in Sudan, and the partial, agenda-setting character of human-rights narratives. The host accurately identifies this self-critique. But the essay does not show that every later de Waal claim, IPC analysis, or human-rights report is therefore unreliable. Its fair use is as a source-audit warning: identify evidence standards, uncertainty, incentives, and later corrections rather than substituting biography for a claim-specific check.

4. In March 2024 de Waal predicted the most intense famine since World War II and thousands or tens of thousands of deaths (25:38)

Verdict: the quotations are accurate; the forecast was too categorical and its death magnitude was not verified.

De Waal wrote that Gaza was about to witness “the most intense famine since the second world war” and told The New Humanitarian it was too late to avert a disaster that would kill “thousands, if not tens of thousands”. The Guardian column also moved impermissibly from people in household Catastrophe to a mechanical death-rate calculation, much like the later Sidhwa error.

The evidence at the time justified an urgent warning: the FRC projected famine as imminent, food deprivation was extreme, acute malnutrition was rising, and mortality surveillance had collapsed. But the wording was more categorical than the underlying FRC finding. Later increases in aid and access corresponded with rapid improvement. That supports de Waal’s causal point that policy response may avert a forecast outcome; it does not empirically validate the quoted death total. A warning can be useful and still be overstated.

5. “The FRC found no famine” in 2024 (28:32)

Verdict: false.

The June FRC said it could not endorse FEWS NET’s April classification and was unable to determine whether or not famine thresholds had passed. The report explicitly distinguished this evidentiary non-determination from a negative finding. It also noted that rapidly increased food availability had altered the March trajectory. The transcript’s later phrasing—“according to their metrics there was no famine”—again overstates the record.

6. The Spagat household survey proves there was no famine (31:55)

Verdict: important evidence against very high 2023–24 mortality claims, but not a test of the 2025 famine classification.

The peer-reviewed Gaza Mortality Survey sampled 2,000 households / 9,729 people between 30 December 2024 and 5 January 2025. It estimated:

• 75,200 violent deaths (95% CI 63,600–86,800), above the contemporaneous Ministry of Health total;
• 16,300 non-violent deaths (95% CI 12,300–20,200);
• 8,540 excess non-violent deaths above the pre-war baseline (95% CI 4,540–12,500).

The survey covered accessible areas and used displaced populations to represent inaccessible governorates. Its small number of observed non-violent deaths makes subgroup patterns imprecise: Table 2 records only five non-violent deaths among people under 18 and 54 among adults. That adult-heavy profile is a legitimate challenge to claims that a classic child-dominated famine mortality pattern was already occurring. It does not show what caused each excess death, and the paper does not label the 8,540 as starvation deaths. PubMed · open-access paper. PubMed also flags an August 2026 correction; any quotation of a corrected table should use the final version.

Most decisively, the survey ends on 5 January 2025, before the March blockade and the July–August 2025 classified famine. It can test earlier mortality claims, not the later period.

7. The 62,413 starvation-death estimate in the 2 October 2024 clinicians’ letter (40:33)

Verdict: unsupported and methodologically invalid.

The open letter by 99 U.S. clinicians claimed a total death toll above 118,908; its appendix generated 62,413 starvation deaths. The calculation treated populations in household IPC Phase 5 Catastrophe as though the area-level famine mortality threshold of 2 deaths per 10,000 per day had already been observed. IPC explicitly says Catastrophe is a household condition and cannot be used to infer an area mortality rate.

The Spagat paper later called the estimate empirically unfounded and found a much lower all-cause excess non-violent total through January 2025. This does not discredit the clinicians’ direct observations of malnutrition, infection, hospital collapse, or preventable illness. It discredits the numerical extrapolation.

De Waal is correct in the interview that Sidhwa’s inference was invalid. His December 2024 blog also said the available datapoints could not yield an accurate mortality estimate and urged extreme caution with the high-end estimates. But it then proposed 10,000 or more starvation/health-crisis deaths as “credible” from analogies rather than representative Gaza mortality data. That was an explicitly uncertain inference, not a measured result; the later survey’s 8,540 excess non-violent estimate is not directly equivalent to de Waal’s cause-specific language.

8. The 2025 IPC classification did not require a directly observed mortality threshold (49:00)

Verdict: accurate only with the omitted qualifier “with reasonable evidence.”

IPC defines area famine as all three conditions occurring together: widespread starvation, acute malnutrition, and excess non-trauma mortality. IPC famine facts. It permits two evidence strengths:

• solid evidence: direct, clear and compelling evidence for all three conditions;
• reasonable evidence: direct, clear and compelling evidence for two, plus a judgment from the broader evidence that the third threshold is likely met.

Thus de Waal is right that the 2025 classification did not depend on a direct death-rate measurement. He is wrong when he says the IPC “made no comment on mortality” or implies famine can technically exist without the mortality condition. IPC did make a mortality judgment—it inferred that the threshold was likely crossed. The host is right that this is a consequential evidentiary distinction; he is wrong to describe it as no evidence at all.

The arithmetic in the transcript—roughly 8,300 deaths if 940,000 people experienced 2/10,000/day for 46 days—is a useful implication check, but not a direct refutation. The classification asserts that a rate threshold was likely crossed in an area during a period; it does not necessarily establish the exact population continuously exposed at that rate for every one of the host’s chosen 46 days. A retrospective mortality survey is the proper test.

9. Israel could rapidly reverse the crisis by allowing more food (29:29; 64:44)

Verdict: strongly supported as an access-and-supply claim; overstated if presented as a single-cause model.

The 2024 FRC documented sharply increased food availability after March; the October 2025 ceasefire and aid expansion were followed by rapid declines in acute malnutrition admissions and the December offset of famine conditions. De Waal’s August 2025 New York Times essay correctly emphasized Israel’s capacity to change entry conditions.

But entry is not delivery to households. Distribution also depended on military access, convoy safety, functioning roads and warehouses, cash and prices, humanitarian coordination, armed-group activity, desperate crowd interception, criminal looting, and market incentives. Israel controlled a critical upstream variable, not every downstream one.

10. Flour prices were evidence of scarcity in spring/summer 2025 (59:48)

Verdict: supported; the visual’s exact “Israeli scholars” source is not recoverable from the transcript.

Official Palestinian price data show Gaza’s April 2025 consumer price index rising 75.59% in one month, including white flour up 85.27% from March. PCBS release. UNICEF’s June cash bulletin recorded flour 2,614% above the February ceasefire level, and IPC/FEWS NET later reported staple prices hundreds to thousands of percent above pre-war levels before late-July aid increases drove flour prices down. UNICEF June 2025 bulletin.

Price spikes support severe scarcity and access failure, but do not isolate cause. Supply cutoff, expected future scarcity, hoarding, disrupted distribution, cash shortages, monopoly power, and armed interception can all raise prices. The exact chart shown in the episode was not named and should not be attributed to a particular paper without the on-screen citation.

11. Hamas systematically diverted or hoarded the aid (61:12)

Verdict: documented incidents, but no demonstrated basis for treating systematic Hamas theft as the dominant explanation for 2024–25 hunger.

A USAID Bureau for Humanitarian Assistance internal review examined 156 reported losses or thefts of U.S.-funded supplies from October 2023 through May 2025 and found no incident it could affiliate with a U.S.-designated terrorist organization; the underlying slides were reported by Reuters. That finding is limited to reported losses in U.S.-funded partner operations; it cannot prove that Hamas never benefited from aid, taxed commerce, intimidated distributors, or diverted other supplies.

Conversely, the UN’s 13 July 2026 statement directly condemned Gaza’s “de facto authorities” after armed personnel entered a WFP warehouse and distribution point, assaulted drivers, and obstructed operations. That substantiates specific Hamas-authority interference in 2026. It does not retroactively quantify Hamas’s share of 2024–25 losses. The best conclusion is mixed: interference and diversion occurred; systematic diversion on the scale needed to explain the famine classification has not been demonstrated in the cited evidence.

12. June 2026 UNICEF findings show children “sprang back” (65:19)

Verdict: partly accurate, but the host overreads a post-ceasefire snapshot.

The UNICEF SMART survey found combined global acute malnutrition at 1.3%, WHZ-based GAM at 0.5%, and severe acute malnutrition uncommon in accessible areas. It also found stunting at 12.2%, minimum acceptable diet for only 22.4% of children aged 6–23 months, acute maternal malnutrition, high diarrhoea and fever prevalence, and major micronutrient/diet-quality concerns. The report records treatment admissions peaking near 17,000 children in August 2025 and falling below 3,000 by March 2026 after humanitarian assistance expanded.

This strongly supports rapid nutritional recovery after access improved. It does not show there was no acute-malnutrition emergency in August 2025. UNICEF explicitly warns that current acute malnutrition can change quickly and that the 2026 survey cannot reinterpret the past.

13. Low recorded “malnutrition deaths” means thousands of indirect deaths could not have occurred (68:27)

Verdict: a serious discrepancy that demands investigation, not a clean disproof.

The Gaza Ministry of Health’s strong administrative series is primarily for violent trauma deaths. Non-trauma deaths outside functioning hospitals, without certification, or coded to immediate causes such as infection, dehydration, cardiac arrest, or chronic disease are less completely recorded. Famine mortality is usually excess non-trauma mortality from the interaction of hunger and disease, not only death certificates listing malnutrition.

At the same time, the host’s scale check is legitimate: a sustained rate of 2/10,000/day over a population near one million would imply thousands of deaths. A later representative mortality survey that found substantially fewer non-trauma deaths in Gaza Governorate for July–August 2025 would directly challenge the IPC inference. Until that exists, both “zero because the registry is low” and a precise thousands-dead total overstate the evidence.

14. Famine is also a social experience of humiliation and dehumanization (55:05)

Verdict: analytically valuable but not a substitute for the IPC classification test.

De Waal’s work emphasizes famine as a destruction of social relations, dignity, mutual obligation, livelihoods, and the capacity to provide for family. His invocation of the Warsaw Ghetto, Leningrad, Auschwitz, Elie Wiesel’s Night, Primo Levi, the Dutch Hunger Winter, and Syrian sieges concerns this social-demographic literature. The host fairly accepts the moral reality while insisting that the ordinary listener hears “famine” as a claim about mass starvation deaths.

Both points can be true. Social degradation belongs in a full history of famine. But when defending the IPC’s technical 2025 classification, de Waal cannot answer a mortality-evidence challenge merely by widening famine into humiliation. The classification still asserts that all three outcome conditions were present, with the third inferred.

15. The Israeli hostages were intentionally starved (79:07)

Verdict: relevant and supported, but analytically separate.

Released hostages, medical assessments, recovered footage, and accounts of captivity document deliberate deprivation and severe weight loss among hostages. De Waal agrees that this should be named. It is a distinct Hamas crime and belongs with Oct 7th Hostages. It does not reduce the protection owed to Gaza’s civilian population or establish that civilian deprivation was caused by Hamas.

What each participant gets right—and where each overreaches

Noam Dworman

Strongest points

• Separates food insecurity, poor diet quality, acute malnutrition, and famine.
• Correctly challenges the 62,413 estimate and asks for a death-rate implication check.
• Uses the Spagat survey as a real empirical constraint rather than repeating projections as observed deaths.
• Concedes that severe Phase 4 suffering matters even if the famine threshold is disputed.
• Recognizes that Israel had capacity to increase food supply and may have crossed legal or moral lines in 2025.

Overreaches

• Converts the FRC’s 2024 non-determination into “found no famine.”
• Treats a survey ending in January 2025 as if it could decide a July–August 2025 classification.
• Treats the absence of a direct mortality survey as evidence that the mortality threshold was not crossed.
• Calls pre-war Gaza “swimming in food,” erasing severe poverty, access insecurity, and diet-quality problems.
• Says Egypt could bring unlimited food through Rafah without Israeli constraint.

Alex de Waal

Strongest points

• Correctly explains why low counts explicitly coded “malnutrition” can miss hunger–disease mortality.
• Correctly rejects the Sidhwa inference and acknowledges the Spagat survey’s puzzling demographic profile.
• Correctly stresses that early warning is meant to trigger action and that an averted forecast is not necessarily a false warning.
• Correctly identifies the exceptional speed with which Gaza’s nutritional conditions could worsen and improve as access changed.
• Appropriately distinguishes documented incidents of aid interference from proof of systematic Hamas diversion.

Overreaches or muddles

• His March 2024 public language was more categorical than the evidence and joined household Catastrophe to a mechanical mortality rate.
• He understates the force of the Spagat result against his own December 2024 suggestion of 10,000-plus starvation/health-crisis deaths.
• He says the 2025 IPC classification “did not depend on mortality” and “made no comment” on it. More precisely, it inferred the mortality threshold under the reasonable-evidence protocol.
• When pressed on technical mortality evidence, he shifts into famine’s social meaning. That explains suffering but does not validate the IPC threshold.
• “Israel could have stopped it at any time with no military disadvantage” is a strong causal and strategic judgment; the rapid response to access supports the first half more directly than the “no disadvantage” counterfactual.

Debate-safe synthesis

Short answer

“In 2024, authoritative monitors warned that famine was imminent, but the FRC never retrospectively confirmed one and later could not endorse FEWS NET’s April classification. In 2025, the IPC/FRC did classify famine—with reasonable, not solid, evidence—in Gaza Governorate for July to mid-August. It directly measured extreme food deprivation and acute malnutrition, while inferring that excess mortality likely crossed the threshold. We still do not have a representative death survey for that period, so the honest position is: the formal classification occurred; its mortality component remains an inference awaiting retrospective testing; and precise famine-death totals are not established.”

On the mortality numbers

“The 62,413 starvation estimate is unusable because it converts household IPC categories into observed deaths. The Spagat survey is the best mortality evidence through 5 January 2025: 8,540 excess non-violent deaths, not 62,413 starvation deaths. But it ends before the 2025 blockade and famine period, so neither side can use it to settle July–August 2025.”

On causation

“Israel’s entry restrictions, military operations, and choice of distribution architecture were major upstream causes; Hamas interference, criminal looting, desperate crowd interception, market breakdown, and cash scarcity also impaired access. The evidence does not support either monocausal slogan—‘Israel alone caused every shortage’ or ‘Hamas theft explains the famine.’”

Source map: every identifiable reference invoked in the transcript

Episode and media clips

1. Comedy Cellar episode / video — primary transcript source.
2. De Waal pre-war Gaza clip shown at 4:16 — original program is not named in the supplied transcript; the clip itself is auditable in this episode, but its earlier context remains unidentified.
3. Norman Finkelstein’s prior Comedy Cellar appearance referenced at 5:23 — exact episode not named in the transcript and not confidently matched.
4. De Waal / Glenn Greenwald clip shown at 32:18 — exact original episode not named and not confidently recovered; the quoted excerpt is preserved in the supplied transcript.
5. Benjamin Netanyahu interview with Bari Weiss referenced at 69:47 — exact interview/date and the number quoted are not specified, so this allusion is not used as evidence.
6. De Waal Fox News appearance referenced at 79:42 — insufficient identifying details to recover confidently.

De Waal’s writings and institutional analyses

7. Alex de Waal, “Writing Human Rights and Getting It Wrong” (2016).
8. IPC FRC, Gaza Strip conclusions and recommendations (18 Mar 2024).
9. Alex de Waal, “We are about to witness in Gaza the most intense famine since the second world war” (21 Mar 2024).
10. Alex de Waal interview, “Thousands will now die of starvation and disease in Gaza” (21 Mar 2024).
11. FEWS NET April 2024 IPC-compatible analysis, reviewed in the FRC’s 4 June 2024 report.
12. Alex de Waal, “How Many People Have Died of Starvation in Gaza?” (17 Dec 2024).
13. Alex de Waal, “Netanyahu Is Choosing to Starve Gaza” (NYT syndication; original 1 Aug 2025).
14. IPC/FRC, 2025 Gaza famine analysis portal, snapshot, and IPC response to methodological objections.
15. IPC/FRC, December 2025 “famine conditions offset” update and FRC statement.

Mortality, nutrition, prices, and diversion

16. Spagat et al., “Violent and non-violent death tolls for the Gaza conflict”, Lancet Global Health 14(4), 2026.
17. Feroze Sidhwa et al., 2 October 2024 open letter and appendix.
18. UNICEF, Gaza SMART Survey 2026 and executive summary.
19. PCBS, April 2025 Consumer Price Index; UNICEF, June 2025 Gaza cash-transfer market bulletin. These corroborate the episode’s flour-price visual; the exact visual source remains unidentified.
20. Reuters, USAID review found no evidence of systematic Hamas theft in 156 reported U.S.-funded aid-loss incidents (25 Jul 2025). The internal slide deck itself was not publicly located.
21. UN/OCHA, statement condemning obstruction by Gaza’s de facto authorities (13 Jul 2026).

Historical and literary comparisons invoked—not Gaza measurements

22. Dutch Hunger Winter (1944–45) and mortality in Syrian sieges — invoked as possible urban-famine demographic analogies. No particular study is named in the transcript, so they should remain hypotheses, not evidence that Gaza shared the same profile.
23. Elie Wiesel, Night; Primo Levi’s Auschwitz writings; Warsaw Ghetto; Siege of Leningrad — invoked to describe famine’s social and moral degradation. They are analogies, not empirical tests of Gaza’s IPC thresholds.
24. Sudan — invoked as a contemporaneous famine of much larger magnitude. This contextual comparison does not settle Gaza’s classification.

Open questions and falsification tests

• Conduct or obtain a representative mortality survey covering Gaza Governorate from 1 July through at least 15 August 2025, with baseline mortality, migration, missing-person, and cause-of-death sensitivity analyses.
• Recover the exact on-screen source for the episode’s flour-price graph and verify whether it was WFP/PCBS data, a later Israeli re-analysis, or an academic price paper.
• Identify the original de Waal–Greenwald clip and pre-war-nutrition clip to test whether editing removed qualifications.
• Separate in all future arguments: aid authorized to enter → aid physically crossed → collected → transported → warehoused → distributed → affordable/accessible → consumed.
• Quantify aid loss by actor and mechanism: Hamas or other armed groups, criminal gangs, desperate civilian self-distribution, corrupt subcontractors, Israeli military action, spoilage, or unknown.
• Check whether the final corrected version of the Spagat paper changes any table values used in debate; PubMed lists an August 2026 correction.
• Do not infer a famine death total by multiplying IPC household-phase populations by the area-famine mortality threshold. Do not infer zero indirect mortality from the administrative count of deaths explicitly coded “malnutrition.”

Integration notes

• Mortality categories and the Spagat time-window rule feed Gaza Casualty Figures.
• The 2024/2025 access chronology and causal boundary feed Blockade in Gaza and Israel - United States Relations.
• The IPC classification-versus-validity distinction and Hamas-diversion evidence directly complement the opposing-source audit in UN admits Hamas is stealing aid in Gaza; Natasha Hausdorff considers implications.
• The hostage-starvation point belongs in Oct 7th Hostages and should not be used as a moral offset against harm to Gaza civilians.