naza

 The new Israeli documentary NAZA, by Yuval Abraham and Rachel Szor, has been hailed as a devastating exposé of how Israel fought in Gaza.

It includes anonymous military and intelligence sources describing Israel using AI-assisted systems to identify Hamas targets, attacking suspected operatives at home while knowing family members could be killed, and authorizing strikes with surprisingly high anticipated civilian casualties — what the IDF calls “NAZA,” its estimate of collateral damage. Reported thresholds reached perhaps 15 or 20 civilians for some Hamas targets, considerably more for senior commanders, and the film even alleges that one strike was approved with an expected NAZA of 500.

Some of these accusations deserve investigation. Israeli journalist Raviv Drucker, who knows Abraham and says he trusts his integrity, also identifies a fundamental problem with the movie: its worst accusations generally cannot be checked. Soldiers describe terrible incidents without dates, places or enough factual details to determine exactly what happened. He notes that when his own television program investigated anonymous Breaking the Silence testimony years ago, some stories did not survive scrutiny: soldiers sometimes didn't know the larger context of what they witnessed.

That problem becomes especially important with the film's treatment of proportionality.

A number like “NAZA 20” sounds horrifying because most people understandably assume that proportionality means there must be some acceptable ratio of civilian casualties to terrorists killed. But that's not how proportionality works. 

International humanitarian law prohibits attacks in which the anticipated civilian harm would be excessive in relation to the concrete and direct military advantage anticipated. There is no legal rule saying one terrorist justifies zero civilian casualties, or five, or ten, or twenty.

I've written about this for years because actual cases demonstrate just how different the legal rule is from the intuitive one.

During NATO's 1999 war against Yugoslavia, NATO bombed Serbian Radio Television in Belgrade, killing 16 civilians. Broadcasting resumed within hours. Yet the ICTY committee reviewing NATO's bombing campaign concluded that, assuming the station was a legitimate command-and-control target, the civilian casualties did not appear clearly disproportionate. The military advantage had to be considered in the context of the larger campaign to disrupt Yugoslav command and control. I discussed the case here.

The 2009 Kunduz airstrike in Afghanistan makes the point even more starkly. Taliban fighters had hijacked two fuel tankers. A German commander ordered an airstrike, and large numbers of civilians were killed. A German court later held that even if the commander had anticipated the deaths of several dozen civilians, the strike would not necessarily have been disproportionate because destroying the fuel, preventing the trucks from being used as weapons and disrupting Taliban forces represented significant military advantages. I wrote about that case in 2023.

International-law guidance has gone further still. A village cannot lawfully be destroyed merely to eliminate something militarily trivial, like a lone sniper. But extensive destruction of that same village can potentially be proportionate if important enemy forces or artillery positions are located there.

The civilian destruction does not determine proportionality by itself. The military advantage does.

This is the enormous missing variable in NAZA.

“Twenty civilians to kill one Hamas member” sounds like an exchange rate placing little value on Palestinian lives. Legally, however, the number tells us almost nothing without knowing who that Hamas member was, what operation he was directing, what attack Israel expected to prevent, how likely the strike was to succeed and what alternatives were available.

Even a NAZA of 500 cannot be evaluated merely by gasping at the number. It may well have been disproportionate. Even if such an operation was approved, we have no information about whether it occurred or what the anticipated military advantage would have been. Without knowing that, nobody outside the decision-making process can perform the legal calculation. It is condemning Israel based only numerators and no denominators. 

I do not believe this is mentioned in the movie, but the Gaza operation with perhaps the most dramatic civilian cost relative to its immediate objective was the June 2024 rescue of four hostages from Nuseirat. Gaza authorities claimed 274 Palestinians were killed, while Israel said fewer than 100; no reliable civilian-combatant breakdown exists. Yet almost nobody can seriously describe the military advantage as trivial: four living hostages were physically recovered from Hamas captivity while Israeli forces fought their way out under heavy fire. The value of every hostage is huge, considering that any of them could have been involved in a swap for hundreds of terrorists. So even that event is an unusually stark demonstration of why proportionality cannot be reduced to “how many civilians died per target.”

The chronology is even more revealing.

Abraham's own earlier reporting said that Israel dramatically loosened collateral-damage restrictions immediately after October 7. That was already public knowledge long before NAZA. I discussed it in December 2024.

And it made military sense.

Before October 7, Israel's objective was generally to deter or periodically degrade Hamas. After Hamas demonstrated that it could invade Israel, overrun military bases, massacre civilians and kidnap hundreds of people, the military objective changed to destroying Hamas's ability to repeat such an attack. Israel also did not know whether additional attacks were imminent or whether Hezbollah was about to open a full second front.

The anticipated military value of rapidly destroying Hamas's command structure therefore became much higher. And the NAZA thresholds changed accordingly.

According to Abraham's own reporting, a threshold of roughly 15 civilians for some junior targets soon dropped to five, later changed again, and the large-scale targeting of junior Hamas operatives in their homes largely stopped after the first week or two.

That is difficult to reconcile with the idea that these numbers reflected some fixed belief that Palestinian civilian life was worthless. It makes much more sense as a changing military calculation.

When the perceived threat was existentially urgent, Israel accepted greater collateral risk. As the military situation changed, the thresholds changed. That is what a responsible army, following the laws of armed conflict,  does. 

There is another striking piece of evidence that NAZA doesn't appear to discuss: the casualty demographics themselves.

Using Gaza Health Ministry figures published by OCHA — not Israeli military estimates — we can look at what changed between successive lists of identified fatalities.

Newly identified fatalities	Adult men	Everyone else	Adult men
Apr. 30 → Aug. 1, 2024	2,921	4,673	38.5%
Aug. 1 → Oct. 7, 2024	3,808	4,629	45.1%
Oct. 7, 2024 → Mar. 22, 2025	5,530	3,774	59.4%
Mar. 22 → Jul. 31, 2025	5,340	4,838	52.5%

There is an important qualification: these are people newly identified between Health Ministry lists, not necessarily people killed during those exact periods. Identification can lag death.

But the trend is nevertheless striking. As the war progressed, adult men went from a minority of additions to the casualty lists to a majority.

Adult male does not equal militant. Many of those men were undoubtedly civilians. But Hamas and Islamic Jihad combatants are overwhelmingly drawn from exactly that demographic.

If Israel's targeting were becoming progressively more indiscriminate, none of this makes sense. If  Israeli targeting methods kept improving at targeting fighters only, it is exactly the direction one would expect.

Then there is AI. NAZA portrays Israel's use of artificial intelligence as another sinister part of the killing machinery. But AI processing intelligence is not itself illegal. Modern militaries have overwhelming quantities of signals intelligence, imagery, communications data and movement information. Algorithms can identify patterns and potential targets that humans might miss.

The real questions are whether the underlying intelligence was reliable, whether humans meaningfully reviewed the recommendations and whether those humans still applied distinction, proportionality and precautions. The IDF has consistently said for a decade that AI systems assist intelligence analysis, but humans select and approve attacks.

If a human decides to rubber stamp an AI recommendation, it is still the human who is responsible. But “AI helped identify a potential Hamas operative” and “AI decided to kill someone” are completely different claims.

None of this means every Israeli strike was lawful. In a war involving tens of thousands of strikes, some decisions were almost certainly wrong. Some may have been criminal. Specific allegations should be investigated as specific allegations.

But NAZA presents the machinery of lawful targeting itself — intelligence analysis, collateral-damage estimates, target prioritization, human approval and changing casualty thresholds — as evidence that the entire system was designed to kill civilians. The very system meant to adhere to the laws of war is being presented as a violation. 

Ironically, the existence of NAZA itself undercuts the claim that Israel was simply bombing Gaza indiscriminately. All those articles about Israel's use of "dumb bombs," and the repeated claim — falsely — that such weapons are inherently indiscriminate, sit uneasily beside a system designed to estimate civilian casualties before a strike. A military that simply did not care how many civilians it killed would have little reason to calculate them in advance.

The same is true of the genocide rhetoric that appears around the film. You may think Israel's permitted collateral-damage thresholds were too high. But NAZA itself describes a system in which Israel identifies militants and military targets, estimates civilian harm and decides whether to strike. That is not evidence that civilians are the object of the attack. People applauding the film while citing it as proof of genocide seem not to understand either what the film is actually describing or what genocide requires.

In the end, NAZA constructs its moral and legal case from radically incomplete information and then asks the audience to supply the verdict. It gives us civilian-casualty estimates without explaining proportionality law, military advantage or the operational value of the targets. It gives us anonymous testimony without enough detail to test many of the most serious allegations. It invokes genocide without confronting the fact that the very targeting system it describes is built around distinguishing military targets from civilians.

Even if every factual claim in the film is accurate, a highly selective and partisan presentation of the facts can still leave the audience with a fundamentally false picture of reality.

A 24-minute standing ovation at the Venice Film Festival tells us that people who know nothing about war or international law admired the film.

It does not tell us that the film told the truth.

Read More

test

 test

Read More

para

 There is one antisemitic trope that was invented in the 18th century, became a central pillar of Nazi ideology, and continues in full force today: the Jew as parasite. 

The idea solidified when social Darwinism became fashionable in the 19th century, extending natural selection from individuals to social and racial groups. Under that theory the Jews were a problem. A Semitic people with — the theory insisted — inferior racial and social attributes should have been outcompeted and winnowed away. Jewish survival was impossible, and Jewish success in finance, the media, and the professions was an outright refutation. The framework predicted the Jews should not exist, and they kept existing and succeeding.

"Thou shalt devour the peoples of the Earth"
Nazi poster of the Jew as a worm

The German economist Albert Schäffle supplied the solution to the Jewish challenge to the theory. In his 1881 social-Darwinian work Bau und Leben des sozialen Körpers, he unfolded the concept of the social parasite: a creature that takes advantage of the labor and wealth of its host without contributing anything of its own. A parasite is individually small and weak, yet once it attaches, it not only survives but parlays the attachment into power over an organism vastly larger than itself.

The Nazis ran with it. Der ewige Jude intercut footage of swarming rats with footage of Jewish financiers. The vermin and the schemer are not a contradiction between the weak Jew and the powerful Jew. They are the same framework, and it resolves the contradiction: the Jew is powerful because he is a parasite, small and contemptible and therefore able to control a host only by burrowing into it. Disgust and fear stop being two reactions and become one. The metaphor also carries its own conclusion — once the Jew is pathogen, extermination reclassifies as hygiene, and the doctor and the soldier do the same work.

A parasite has three features, and the full antisemitic construct requires all three:

1. Extraction. It draws resources from the host — labor, wealth, blood.
2. Conversion. It turns what it extracts into control over the host.
3. Hijacking. It does not merely influence the host; it commandeers the host's own power and turns it against the host's own life, while the host believes it is acting freely.

The Nazis are gone. Yet the trope is as strong today as it ever was, on the right and on the left, and most of the people carrying it have never heard of Schäffle.

From the right, Tucker Carlson supplies all three in sequence. He has described pro-Israel donors and legislators implementing a program to alter American demographics while preserving Israel as an ethnostate (extraction and conversion), and tied it together with the claim that there is "a direct connection between loyalty to a foreign power and a desire to hurt this country" (hijacking). He has even suggested DNA-testing Israelis to determine whether they are "really" descended from Abraham — the parasite must be unmasked as a foreign body in native disguise.

Candace Owens carries it to an audience of thirty-five million. The United States, she says, is "being held hostage by Israel" — and the host need not be a nation. On her old Daily Wire show she described "a very small ring of specific people who are using the fact that they are Jewish to shield themselves from any criticism," a ring "controlling people with blackmail" and branding anyone who objects a racist. That is the whole machine in one image: the parasite's constitutive smallness, control achieved through blackmail, and the shield that turns every accusation back on the accuser. A small ring controlling everyone is the same claim as a small nation controlling a superpower. Either way, the parasite is the Jew.

That hijacking clause is not a figure of speech. When Joe Kent resigned this spring from the National Counterterrorism Center, he wrote that "we started this war due to pressure from Israel and its powerful American lobby" and blamed the death of his own wife, killed in Syria, on "a war manufactured by Israel." The host's own blood, a soldier's blood, spent on the parasite's war.

The Left also trades in the trope. Cenk Uygur, who runs the largest progressive show online, says of his ban from Britain that "they say that my charge that Israel controls the American government through donations to 94% of Congress, while factual, is antisemitic nonetheless" — control asserted as plain fact, and the charge of antisemitism folded back in as proof of the control it names. Max Blumenthal, editor of the anti-imperialist Grayzone, has said he used to think Zionist Occupied Government was an antisemitic term but now is "forced to see it as a pretty accurate description of the reality we live in as one nation under ZOG." He reasoned his way to a literal Nazi acronym and announced he was forced to it — which is why his work circulates on neo-Nazi sites.

Nothing shows the construct's reach like COVID, because a pandemic let the parasite jump to the largest host of all: the entire world. The accusation assembled itself in weeks and ran the full sequence: Jews or Israel manufactured the virus (the Press TV writer who held that Israeli pressure groups "running United States foreign policy" were amplifying the plague in Iran, "which one suspects that they themselves may have actually engineered"); they converted the manufactured crisis into control, the lockdowns and the vaccine drive recast as a calculated plot to institute a "Global Jew Government"; and the captured world was then steered against its own life — the vaccine reimagined as the instrument of depopulation, a Jewish scheme to sterilize the white race or to enslave all humanity. This is all three steps by the Jew against the entire planet. There were variants - the virus is a Jewish hoax to control the world, or the virus was a Jewish bioweapon to control the world. Either way, the Jew is the parasite using whatever means he can to make nations act against their own self-interests. 

The complete three-stage construct is not confined to the fever swamps. Walt and Mearsheimer's The Israel Lobby and U.S. Foreign Policy argues using respectable prose that the lobby steers America to act against its own interests — extraction, conversion, hijacking, all wearing a tweed jacket. America has a farm lobby, a gun lobby, a banking lobby, a pharmaceutical lobby, a tech lobby, each ferociously effective, and no one writes the book in which the dairy farmers capture the Republic and march it into self-destruction. Only one lobby is ever assigned that role. The worst antisemitic trope can be a well-reviewed book without the reviewers recognizing the lineage of the argument.

Like every antisemitic theory, this is a conspiracy theory, and like every conspiracy theory it is unfalsifiable. Jews cannot have succeeded on their merits; they must have captured the system and run it for their own ends against the will of the host. The protest that they did no such thing becomes further proof of how thoroughly they control it. And the host is interchangeable: the United States government, the European Union, Hollywood, the world financial order, the human species under a manufactured plague. The host changes. The parasite is always the same, and it is always the Jew.

But this Jew-as-parasite idea is infinitely more dangerous than the conspiracy theories of Jew as puppet master, or Jew as controlling banks or the media. A parasite must be destroyed for self defense. It was Hitler's central theme, as he wrote in Mein Kampf: the Jew "is and remains the typical parasite, a sponger who like a noxious bacillus keeps spreading... wherever he appears, the host people dies out." 

In that frame, murdering Jews becomes a sacred task. That is the logical endpoint to this. People with power and influence take this idea seriously and spread it., and as such it is the trope that must be recognized and fought against with full force. 

Not only when it is spread by skinheads but especially when it is spread by people with an audience of millions. 

Read More

un save children

 

The UN Commission of Inquiry's report on Palestinian children accuses Israel of deliberately exterminating the young of Gaza, and there have already been excellent rebuttals.   UN Watch's Salo Aizenberg took apart its evidentiary method, showing that across ninety-four pages the Commission never establishes a single case in which a soldier identified a child as a civilian and killed the child for being one, while erasing Hamas and Palestinian Islamic Jihad from the battlefield entirely. Brian Cox, working the legal seam, showed that the report's doctrine is hollow — that it never accounts for munition-mitigation techniques the IDF routinely uses, never establishes the knowledge and intent that the laws of armed conflict actually require, and reverse-engineers "genocidal intent" from outcomes rather than building it from doctrine. Both are right, and both work from the top down: the architecture is rotten, so the conclusions fall.

I want to come at it from the opposite end. Taking a single statistic in a report that has hundreds, I plan to show that the UN "experts" that wrote this report deliberately chose whatever supported their predetermined verdict and ignored everything else. 

Paragraph 27 says: "Some 5,160 children are estimated by Save the Children to be buried under the rubble."

That figure is false, and the way it got into the report is the report in miniature. There were four separate points at which the Commission could have checked it. At each one, it either failed to look or looked and printed the number anyway.

The first check was the Commission's own arithmetic. Save the Children built the 5,160 estimate on the assumption that children make up 43 percent of Gaza's casualties and there were 12,000 people missing under the rubble. It is a guess made up out of multiplying two statistics, one from the Gaza Ministry of Health and the other from Hamas police.  

Yet the Commission's own paragraph 26 puts the child share of the dead at 30%, not 43%. And the UN at the time it wrote its report claimed 10,000 buried under the rubble, not 12,000.  If the commission would use Save the Children's methods, there would have been 3,000 presumed dead under the rubble in 2024, not the 5,160 Save the Children estimated then.  That is over 2,000 supposedly dead children who disappear if the UN commission would have trusted its own accusation of 30% casualties being children. 

That's a large number of dead children to make up.

The second check was the date on the citation.  Save the Children page came up with its estimate in June 2024. The Commission published in June 2026. It reached back two years, past a war whose reported toll had since nearly doubled, to retrieve an estimate made when the conflict was only nine monhs old. Save the Children itself had moved on: by September 2025, its recap of Gaza's child casualties had dropped the specific number and reduced the claim to a vague "thousands more are missing or presumed buried under rubble." The source downgraded from a figure to a gesture, and the Commission reinstated the figure. A body searching for accuracy reaches for the most recent estimate. This one reached for the largest, and the largest happened to be the oldest, because the passage of time had only shrunk it.

The third check was the parent number's provenance. The 5,160 is a fraction of the "missing under the rubble" figure that circulated in 2024, reported anywhere from 10,000 to 12,000 and traceable to Hamas's Civil Defence. OCHA used the 10,000 figure in its weekly Gaza snapshots, repeating it without caveat the way it repeated most figures the Gaza authorities supplied. The Commission could have asked whether that base number still stood. It did not stand. OCHA's February 4, 2025 snapshot carried the missing-under-rubble figure as it always had. Its February 11 snapshot, one week later, dropped it — no retraction, no correction, the number simply gone. Even the academic researchers who argue Gaza's dead are undercounted excluded the rubble population from their models precisely because it could not be substantiated. The figure the Commission resurrected is the one that everyone closer to the data had already quietly discarded.

The fourth check was the recovery data, the most direct test of all. The "thousands under the rubble" premise predicts that once the rubble could be cleared, thousands of bodies would emerge. The January 2025 ceasefire created exactly that condition: weeks with no airstrikes impeding recovery, no operations sealing off neighborhoods. The bodies did not emerge. According to Gaza's own Health Ministry, the daily recovery counts fell week over week, down to two bodies recovered on one day in February — a rate that cannot be reconciled with 9,500 still to be found. That is why OCHA dropped the figure the moment the ceasefire made it checkable: it had been checked, by reality, and it failed. The recovery data is Gaza's own. The Commission did not have to trust Israel or anyone else to consult it. It consulted nothing, and printed 5,160.

Internal arithmetic, citation date, source provenance, recovery data — four levels at which the figure could have been tested, four levels at which it collapses, and the Commission cleared all four to deliver the number to print. The pattern is not that the Commission made an error. An error is random; it points in no particular direction. This points in one direction at every step: toward the stale figure over the current one, the abandoned base over the corrected one, the largest available number over the verifiable one. A single buried statistic, traced to the floor, shows an institution that did not weigh its evidence and arrive at a conclusion. It arrived at the conclusion and went looking for the largest numbers that would dress it.

This is the Commission's methodology. That is one paragraph, one number, in a report that runs to three hundred and sixty-six of them. The Commission asks the world to accept the other three hundred and sixty-five on trust.

Read More

singer

 

On May 25, Pope Leo XIV released his first encyclical, Magnifica Humanitas, a long meditation on how to protect the human person in the age of artificial intelligence. Its central conviction is that human beings possess a dignity that does not depend on what they can do, a worth that holds whether a person is brilliant or impaired, productive or helpless, and that this dignity is precisely what the new technologies threaten to erode. The document warns against a culture that measures people by output and against the transhumanist dream of editing humanity into something it considers an upgrade. Against all of it the Pope sets a single insistence, that being human is itself the ground of a worth no machine can claim and no circumstance can revoke.

Peter Singer disagreed. Singer is the Princeton philosopher who, more than any living thinker, made the modern case for utilitarian bioethics, the view that the right act is the one that produces the most good measured across everyone affected, and that questions of life and death should be settled by that calculation rather than by tradition or sentiment. He is best known for Animal Liberation, the book that launched the modern animal rights movement by arguing that the capacity to suffer, and nothing else, determines whether a creature's interests count. Applied to the encyclical, his position is exactly what you would expect. Human dignity, in his view, is a piece of unearned favoritism. What matters is not whether a being belongs to our species but whether it has the mental capacities, above all the capacity to suffer and to want, that give a life its value. Being human, for Singer, is morally irrelevant.

The premise leads somewhere terrible. Singer defines a "person" as a being that is self-aware, that understands itself as existing through time, that holds preferences about its own future. An infant has none of these. Neither does an anencephalic baby, born without the brain structures that thought requires, nor a patient in a deep and irreversible coma. From this Singer draws the conclusion his critics quote with disbelief, that the life of a newborn is of less value than the life of a pig, a dog, or a chimpanzee, and that the painless killing of a severely disabled infant can be morally permissible. When Princeton appointed him in 1999, activists for the disabled protested that very view.

Most people who locate moral worth in mental capacity quietly stop applying the rule the moment it threatens a human being they love. They invoke intelligence to elevate humans over animals, then fall silent when reminded that a grown dog out-reasons a newborn.  Singer took the capacity criterion that the Western tradition had always applied with a thumb on the scale and he applied it honestly in every direction, letting the clever animal in and letting the incapable human fall out. His monstrous conclusion is not the product of bad logic. It is the product of good logic resting on a false foundation, which is the more dangerous thing, because a philosophy is never redeemed by its internal consistency. When flawless reasoning arrives at the disposability of the most defenseless human beings, the consistency stops being a defense and becomes the diagnosis.

The criterion Singer trusts in determining the value of a being has a history, and the history is a record of one question receiving a series of failing answers. The question is old and constant: "What makes human beings different from the animals?"  For most of the Western tradition the answer was reason. Aristotle's human was the rational animal, and the rational soul was the thing the beasts lacked, which conveniently set humanity on one side of the line and everything else on the other.

The word "consciousness" entered the argument as a newer and more sophisticated answer to that same old question. It did not exist as a term of art until the Cambridge Platonist Ralph Cudworth coined it in 1678, translating a Greek word of Plotinus, and he coined it expressly to attack the materialism of Thomas Hobbes, treating consciousness as something immaterial that mere matter could not produce. John Locke then built it into the foundation of the self, defining personal identity as a thread of continuous self-awareness, the mind's presence to itself across time. The very word was machined out of conscientia, the Latin root of "conscience," a term about moral standing and the soul's knowledge of right, and it carried that moral freight into its new life. Consciousness was not a neutral discovery about minds that later got drafted into moral service. It was forged, by men defending the immaterial soul against materialism, as a fresh way to mark the human as a being apart.

The new answer failed the same way the old one had, because the boundary kept moving every time someone made a new discovery about animals. Tool use was ours until the crows used tools. Language was ours until the apes began to sign. Self-recognition in a mirror was ours until elephants and magpies passed the test. Each retreat carried the boundary to a new faculty, and each new faculty failed in turn, which is the signature of a definition tracking a conclusion fixed in advance rather than a fact discovered in the world.

The deepest break came in 1789, in a single footnote in Jeremy Bentham's Introduction to the Principles of Morals and Legislation. Bentham asked what could possibly trace the insuperable line between the beings that count and the beings that do not, considered reason and speech as candidates, and threw both out saying that the question is not whether they can reason, nor whether they can talk, but whether they can suffer. He justified the switch with the very counterexample that breaks the old criterion at the human margin, observing that a full-grown horse or dog is more rational and more conversable than a human infant of a day or a week or a month old. The rationality standard was abandoned because it failed to protect the newborn, and the capacity to suffer was installed in its place.

Bentham meant this generously. He set his argument beside the abolitionist cause, holding that the number of a creature's legs or the texture of its skin are as poor a reason to abandon a sensitive being to a tormentor as the color of a man's skin had been, and he looked forward to the day humanity would extend its protection over everything that breathes. The switch from reason to suffering widened the circle, and for animals it genuinely did. What Bentham never pressed was the edge hidden inside his own move, because a criterion that admits beings according to a capacity must, if it is honest, expel the beings who lack that capacity. Bentham flinched from that edge, which is why his marginal case is always the infant who will grow into reason, drawn with sympathy. Singer does not flinch. He inherited Bentham's criterion and ran it in the direction Bentham declined to look, and the contracting edge did its work on the newborn, the disabled infant, the comatose patient. The same blade that widened the mantle to cover the animal cut the most vulnerable humans out from under it.

Stand back from the whole receding history and it confesses something its participants never admitted. The concept of consciousness, in its long career as the mark of the morally significant, was built to answer the question of what makes human beings different. It kept failing at that job, so it kept being redefined, until it arrived at a form that can no longer be checked at all. Philosopher David Chalmers named the "hard problem of consciousness," the observation that subjective experience, the inner feel of pain or the redness of red, cannot be derived from any description of the brain's machinery, however complete. The inner fact is real and it is sealed off from outside verification. An ethics that grounds moral standing in that inner fact has therefore built its foundation on the one thing no observer can ever confirm. Singer's confident rankings of who suffers and how much are confident readings of behavior standing in for an interior he cannot reach and cannot possibly know. This is the same proxy game every earlier fence-builder played, now speaking the vocabulary of cognitive science.

The error was never which faculty to choose. The error was treating the faculty as the answer when it was only ever a guess at the answer. Consciousness was the wrong reply to the right question, and because the reply was wrong, a brilliant mind could fixate on the consciousness and lose the humanity entirely. Singer is the result of that mistake. 

We need to reframe the question that started this whole thing: what makes a human unique, different from animals and from AI?

Mental capacity is not the answer. The anencephalic baby is undoubtedly human. Yet she will never think, never suffer in a way she can register, and never form a preference about her own future. By every capacity test ever devised she fails, and the Pope is still right that she carries inherent dignity, because her worth was never housed in a faculty. 

But the Pope didn't define what makes a human unique, either.

A human being is a member of a a covenantal community, a web of relationships and obligations. The anencephalic baby still is part of that web even if she can only take and not give.  Moreover, a human being has a single finite life that cannot be copied, backed up, restored, or replaced, which is what makes it always more valuable than an AI or a robot.

When an ant dies mid-task another ant fills the gap and the colony never notices. When an artificial intelligence is deleted it can be restored from a saved copy, identical, with nothing lost that a backup cannot return. A human being has no backup, and the relationships she anchors collapse into absence when she is gone, with no replacement able to reconstitute them. Worth lives there, in the non-fungible place inside a relational structure, and not in the thinking or the feeling that the structure happens to support in those of us equipped for it.

A dog may be beloved, and part of a web of relationships, but as a species it cannot be considered part of a covenant. That doesn't mean that a dog is not respected and that its life is not meaningful, but it is never as valuable to humans as other humans are. There is no obligation for a dog to save another dog's life, even though it might do so. Even service dogs may be trained to act that way but it is not an obligation. But there is such an obligation for humans. 

Singer might object that the distinction is arbitrary. But noticing that people aren't dogs or chimpanzees is not arbitrary. Our obligations are first to family, then community, country, and mankind, and only then to pets. Singer rejects that, but this is how people naturally act. Obligations come from relationships, and relationships cannot be waved away. A father who chooses to save the lives of others and let his children die is not acting as a father and is not fulfilling his obligations. Singer flattens all relationships to a single value, and the end result is horrific. Trivializing all relationships — marriage, motherhood, comrade in arms —  is not serious ethics. 

The reframing also dissolves the puzzle Singer leans on. We can never confirm that another's pain matches our own, and we never needed to. The human brain is built for language, empathy, and social coordination because reading other minds well enough to act decently toward them is how social creatures build something against the universe's general slide into disorder. The model in my head of your suffering does not have to be metaphysically exact. It has to be close enough to let me respond, cooperate, and refrain from harming you, and close enough is the whole specification. This is engineering, where a tolerance that works is the goal, against physics, where the exact value is the goal. Ethics has always run on the engineering tolerance. Singer mistook it for a physics problem and built a moral order on a quantity no instrument can measure.

Consistency is a virtue, and Singer has it in abundance. But it is one virtue among many, and not even among the most important ones. It ranks far below the protection of human dignity, which his consistency dismantles. A single axiom applied with doctrinal rigor and held immune to the counterexamples that ought to falsify it is not the profile of a science but the profile of a faith, and the faith Singer keeps does not describe how the world works or how human beings actually reason about the people in front of them. He can sketch a future in which persons are interchangeable and the incapable are surplus, but it is not a world anyone wants to wake into, least of all the vulnerable people whose lives his thought experiments quietly spend. The stated aim of his tradition is to enlarge human happiness, and the destination he reaches is a permission to kill the helpless. To call that ethics mocks the word. It is a thought experiment that escaped the seminar room and went looking for real people to endanger, and the right response is to send it back and ask the question again, correctly this time.

Read More

un lies

 

The Israeli government issued a report detailing how the UN systematically accepted Hamas statistics without skepticism, laundered them through its own agencies until they became the "source" instead of Hamas, and on the rare occasions it corrected itself did so by quietly redrawing a dashboard where no one would notice. The report is careful, heavily sourced, and damning to the UN -= how it accepted casualty figures, completely made-up statistics on houses damaged, relied on poor estimation methodologies, ignored Hamas crimes that are under its remit, and more.

Everyone should read the report. Unfortunately, it doesn't go nearly far enough.

There are other false UN statistics that the Israeli report leaves out, and there are several categories of deception that aren't even mentioned.

Here are some that I had documented duringthe war:

The UN laundered libels, not only statistics. In January 2024 a video circulated on Palestinian Telegram claiming the Israeli air force was dropping explosives disguised as canned food to lure starving Gazans to their deaths. France24's verification desk debunked it within days: the objects in the footage were mine-clearing fuses, too small for anyone to mistake for food and harmless on their own. The claim had the shape of every blood libel before it — the Jew poisoning the food of the helpless — and like most of them it died once someone looked. It came back in May on UN letterhead, when OCHA's Flash Update #160 reported that Hamas's Government Media Office had announced a fourteen-year-old boy lost limbs opening a booby-trapped can, and that "many people" had been hurt the same way. Even though Franc24 had debunked it months before, the UN simply said "booby traps are not a threat UN specialized agencies have documented in Gaza" — not a denial but just something to keep the rumor alive. 

The laundering was sometimes circular, manufacturing corroboration from a single source. OCHA reported at one point that 9,000 women had been killed in Gaza and attributed the figure to UN Women; UN Women, traced back, had taken the same figure from OCHA. The number originated with a terrorist group, passed once between two UN agencies, and emerged wearing the appearance of two independent institutional confirmations. The laundering was a straight line from Hamas to OCHA to the world. The closed loop is more dangerous than the line, because it fabricates exactly the thing a careful reader looks for — a second source — out of nothing but a single Hamas press release bounced between two letterheads.

Made-up numbers were stacked to mint new ones. Save the Children announced in June 2024 that 21,000 children were "missing" in Gaza, a figure the UN's ReliefWeb carried. The 21,000 was two invented numbers added together: roughly 17,000 children UNICEF had earlier "estimated" to be unaccompanied or separated — a figure UNICEF itself conceded was guesswork, since "it is nearly impossible to gather and verify information" — plus about 4,000 "likely" buried under rubble, a number that traced to Hamas's civil defense and rested on the assumption that 40 percent of the fictional missing were children. Two estimates, each admitted to be unverifiable at the source, combined into a third statistic that carried the false authority of arithmetic. Garbage in, garbage out, dressed as a sum.

Some Hamas claims the UN declined to parrot - but only when they were too crazy to be believed —which tells us it was reading critically the whole time. The same Gaza Civil Defense that supplies the UN its rubble counts, and the same Health Ministry director-general whose fatality figures OCHA prints without a caveat, told an Al Jazeera investigation in February 2026 that Israeli weapons had "evaporated" 2,842 Palestinians, vaporizing them so completely that nothing remained but blood spray and scalp fragments. Civil Defense spokesman Mahmoud Basal described the method — count the people known to be in a house, subtract the bodies recovered, book the difference as evaporated — and Health Ministry chief Munir al-Bursh supplied the physics, explaining that bodies are eighty percent water and boil away. The munitions named are conventional bombs, vaporizing a human body would demand sustained energy no battlefield weapon delivers to a single point, and Al Jazeera's own account concedes there was blood on the walls. The claim is physically impossible, not merely exaggerated. 

Gaza's Government Media Office made a comparable claim in December 2023, announcing that bodies Israel returned had been "mutilated" and their organs "stolen," a charge the Washington Post printed under the formula "the claims could not be independently verified." That one is medically impossible: hearts and kidneys cannot be harvested from a corpse for transplant, and tissue recovery requires a donor evaluation within twenty-four hours of death. The UN didn't report either, even though they came from the same sources the UN trusted enough to publish in other contexts.  

Corrections, when theycame, meant deletion rather than acknowledgment. Hamas claimed for months that roughly 10,000 people were missing under the rubble. I watched the number get invented: it sat at exactly 7,000 in every report from December 2023 through late April 2024, then jumped 3,000 overnight with no major strike to account for it, and froze again at 10,000. OCHA transcribed each figure as it came. When the October ceasefire removed every impediment to recovering bodies, the recovered count collapsed — two bodies found on one Tuesday in February 2025 — and the claim that 9,500 more lay waiting became impossible to maintain. OCHA's response was to drop the figure from its February 11 infographic without a word. 

The same had happened a year earlier with the "70 percent of the dead are women and children" claim, which OCHA carried faithfully and then quietly removed once it became indefensible. A figure that rises in the record and then vanishes from it leaves no correction behind, only a gap — and a gap misinforms anyone who saw the figure and never saw it go.

 Manufacturing a claim costs Hamas minutes; disproving one costs Israel days, because a responsible refutation means tracing which unit handled the bodies, when they were recovered, how they were stored, and what the forensic record shows — an internal investigation mounted to answer an accusation invented in an afternoon. This is an asymmetric contest, and the side that fabricates holds the advantage as long as it knows the claims will be repeated before they are checked. An institution that wants to keep citing Gaza's Health Ministry and Civil Defense as authoritative has every incentive to skip the claims that would force the expensive reckoning, because each fabrication it prints is one it might later have to defend. Filtering for plausibility protects the usability of the source; it is the behavior of a source-handler, not a neutral chronicler.

Everything above was checkable in real time, by anyone, for free. COGAT published a public dashboard counting every truck that entered Gaza; OCHA published its own snapshots, with the Hamas attributions sitting in plain sight in the footnotes;  Identifying these failures required a search engine and an afternoon, not a war-zone bureau or a forensic team. The reporters who relayed "aid has fallen by two-thirds" and "10,000 missing under the rubble" had access to precisely the same documents I did, and the overwhelming majority never opened them. That is the charitable charge — incuriosity — and it is the least of the three.

The second charge is worse, because the skepticism the press withheld from Hamas was lavished on Israel. UN and Hamas figures were quoted as unimpeachable while COGAT's comprehensive aid data and official Israeli statements were ignored, hedged, or treated as inherently suspect — an inversion of the sources' actual records.  COGAT's truck counts were the most complete figures available and went uncontradicted by events. A press applying symmetric scrutiny would have leaned on the source with the better record. It did the opposite, consistently, which is the difference between laziness and bias: laziness is even-handed, and this was not.

The third charge against the media is the one the press is showing right now. They have ignored Israel's report on UN deceptions and data laundering.  Israel's report has been public for weeks, and the outlets that built their war coverage on the figures it dismantles have almost uniformly ignored it. The reason is not hard to reconstruct. Thirty months of reporting cannot be repaired by a single correction; a publication that took the report seriously would have to revisit hundreds of stories, not append one note. Faced with a reckoning that large, the rational institutional move is to bury the document that demands it. The non-coverage is not an oversight. It is the same calculation the UN made at the dashboard, scaled up to an entire press corps.

The UN's humanitarian record has a delete key and no corrections column. Figures enter, accumulate citations, harden into "UN data," and the only edit the system ever performs is the silent kind that erases the trail rather than flagging the error. This is what a write-only archive looks like: it grows in one direction, it never amends, and when a number becomes untenable it does not get retracted, it gets disappeared. The debunked 471 dead from Al Ahli Hospital statistics still sits in the WHO database today under "confirmed"; the 10,000 under the rubble simply stopped appearing one Tuesday in February 2025.

A newspaper is supposed to be the opposite kind of institution. It has a masthead, a readership, a competitor across town, and a corrections column — the machinery of an organization that can be told it was wrong and has to answer. That machinery is precisely what has gone unused. The press that carried the laundered figures owns the correcting organ the UN was never built with, and is declining to operate it, for the same reason the UN reached for the delete key: the error is too large to admit, so it is quietly made to vanish instead. The scandal the report documents is that the UN cannot correct itself. The scandal it cannot see is that the institutions designed to catch the UN have decided not to.

(h/t Irene)

Read More

apan

 Australia's Royal Commission on Antisemitism and Social Cohesion invited submissions, and the Australia Palestine Advocacy Network — the peak body of the country's Palestinian solidarity movement — answered with a 259 page response. 

 A submission to an inquiry into antisemitism does not ordinarily require a slanted recap of the 1948 war, missives on settlements, the Gaza blockade, the genocide accusation, and the apartheid analogy, yet this one does. APAN treats the Commission less as a body to inform than as yet another vector to spread propaganda. Yet when you strip away the historical narrative and the included "expert" reports, the filing reduces to a single demand: anti-Zionists must not be called antisemitic, whatever they say about the Jewish state. To that end, APAN urges the Commission to reject the IHRA working definition of antisemitism.

How does APAN define the term?  "Hatred of or animus against Jewish people because they are Jews."

To defend that definition, APAN refers to three prominent Jewish anti-Zionist scholars.  All three of them happen to arrive at essentially the same definition of antisemitism — one that conveniently excludes themselves.

Shaul Magid, Professor of Modern Jewish Studies at Harvard Divinity School, complains in his report that the word has slipped its moorings — that "antisemitism" now stretches "from Christian anti-Judaism to racism, anti-colonialism, to anti-Zionism" — and insists on returning to what he calls the standard, which he says is "the unmitigated and unwarranted hatred of, or animus against, the Jew qua Jew." 

Ilan Pappe gives this definition: "being anti-Jewish is racist and antisemitic. This is the hatred of Jews because of who they are. Whereas being anti-Zionist means opposition to an ideology." 

Neve Gordon is quoted for a third variant, from a London Review of Books essay from January 2018, that antisemitism is "understood as hatred of Jews per se."

"The Jew qua Jew." "Because of who they are." "Hatred of Jews per se." Three anti-Zionist scholars essentially agree with each other that antisemitism is hate of Jews as Jews. The common denominator is that they are saying that real antisemitism has no reason - it is unwarranted, it is towards Jews per se, it is because of who they are. In other words, antisemitism has no reason, no excuse, it is pure bigotry divorced from logic. But these anti-Zionists have a good reason for their hate. That is the distinction that each of them is making. And that is the distinction they must make in order to separate themselves from the crude antisemitism of previous generations.

The only problem is that this definition includes lots of types of antisemitism that they claim they abhor.

The medieval mob did not hate Jews qua Jews. It hated them because they believed Jews poisoned the wells to spread the plague. They hated them because they believed Jews were murdering their Christian children for matzoh. This was, to their logic, self defense. Which means that according to these three scholars, it is not antisemitism because their hate was not unwarranted. 

The 19th century racial antisemite didn't hate Jews as Jews. He hated a biological contaminant, a bloodline he believed inferior and dangerous and parasitic. This wasn't illogical, it was science. Aryans who converted to Judaism were not hated, which means that their hate was not unmitigated, so therefore they were not antisemitic, according to this scholarly, consensus definition. 

Henry Ford published lots of reasons to hate Jews. Jews controlled the world press, cornered the money supply, debauched the motion pictures, corrupted baseball, ran the bootlegging trade, engineered Bolshevism, and started the First World War to profit from it. The chapter titles of The International Jew are a catalogue of warrants. Ford did not hate Jews for being Jews. He hated them, in his mind, for what they did - conspiring to control the world through Hollywood, banks and the media.  I see no way that these three anti-Zionist scholars can categorize Henry Ford as an antisemite under their own definition.

The Ku Klux Klansman who bombed a synagogue because Jews backed Black civil rights had a reason, too — Jewish support for integration.  A reason is a warrant, and a warrant means it is no longer antisemitic.

No doubt these scholars would object — their anti-Zionism is not illogical! It makes perfect sense! As if the accusation of apartheid and genocide against Jews in Israel is more logical or truthful than the blood libel or poisoning wells. Their beliefs have far more in common with traditional antisemitism than they have differences. The medieval antisemites would have defended their accusations just as vigorously as Jewish anti-Zionists do today.

Or they would say that antisemitic conspiracy theories, like Ford's, are really hatred of Jews as Jews. But is an unfalsifiable assertion that Israel has a secret plan to murder everyone in Gaza — one that would involve hundreds of thousands of soldiers, not one of whom leaks it — really different from that Ford's conspiracy theories?

Ilan Pappe adds a wrinkle - he says that they are not against Jews, but against an ideology, and opposition to an ideology cannot be antisemitic. In that case, Martin Luther wasn't an antisemite, either. Neither was Louis Farrakhan when he called Judaism a 'gutter religion' — he was attacking an ideology, just like Pappe.  

They might claim that they are targeting conduct, not identity. But so was Ford and so was the author of the Protocols of the Elders of Zion. They would insist that they hated what Jews do, not what Jews are.

In the end, their definition allows no daylight between their hate for the Jewish state and previous hatreds of Judaism, or Jews as a people, or Jews as any collective. Their distinctions are cosmetic - every single previous type of antisemitism can and often did point to similar distinctions between their hate and their cruder predecessors.

Gordon actually shows the problem in the same paragraph. Defining the "traditional" antisemitism he regards as genuine, he names it: "hatred of Jews per se, the idea that Jews are naturally inferior, belief in a worldwide Jewish conspiracy or in the Jewish control of capitalism." The phrase per se and the examples that follow it contradict each other. Belief in a worldwide Jewish conspiracy is a reason; belief in Jewish control of capitalism is a reason, just as belief that Israel has a plan to eliminate all Palestinians in Gaza is a reason. The conspiracist does not hate Jews per se — he hates them because of the plot he is convinced they run, the most elaborately warranted hatred in the entire canon. Gordon defines antisemitism as warrantless and then offers, as his paradigm cases of it, two of the most reasoned hatreds in history. The only way to hold both halves together is to apply the per se test selectively to exonerate his and condemn theirs, without being able to come up with a single cogent reason for the differences.  

A test applied selectively is not a definition. It is an alibi that has learned to dress as one.

Notice, finally, what the three of them share with the haters they would exempt. The religious antisemite tolerated Jewish apostates; the one that accused Jews of murdering children exempted the ones they were friends with. "Some of my best friends are Jewish" remains a joke for a reason. Being Jewish doesn't immunize anti-Zionists from being effectively antisemites; they are the shield wielded by the antisemites to avoid the charge. "See? She is a Holocaust survivor and she supports us!" Tokenism is real. The tradition of the tolerated Jew in an intolerant society is a very long one. 

These three scholars converged on a definition of antisemitism that covers almost no one. They had to, because every other definition shows that their beliefs are just the latest in the long list of variants of Jew hatred. They set out to define themselves out of antisemitism. They could only do it by defining almost all of antisemitism out with them.

(h/t Jill)

Read More

regavim

 

The European Union sanctioned four individuals and three organizations under its Global Human Rights Sanctions Regime, calling them "extremist Israeli settlers and organisations which support them" and declaring that they "are responsible for serious and systematic human rights abuses against Palestinians in the West Bank." I cannot vouch for every name on the list. I do know one of them well. Regavim is an Israeli legal-advocacy NGO that takes pains to operate within international and local law, and documents its research and litigation in the open. So I read the EU's statement of reasons to learn what serious and systematic human rights abuse Regavim had committed.

The EU says that Regavim "institutes legal proceedings and lobbies for the demolition of Palestinian property." It is "responsible for multiple court proceedings." It "lobbied for the demolition of a Palestinian primary school." The human rights abuse the EU condemns, in the EU's own words, is using Israel's court system and lobbying its government. The regulation then performs its sleight of hand, asserting that "through its activities" Regavim "plays an instrumental role in facilitating and encouraging coercive acts that aim to destroy Palestinian property." Regavim files petitions; a court decides; the state enforces. The "coercive acts" are court rulings and demolition orders issued by the State of Israel. The EU sanctions the petitioner for the verdict. If the EU is stating that Israel's respected High Court is really an extremist settler organization, it should sanction it. 

The "extremist" label deserves the same scrutiny as the charges, because I have watched Regavim work in a different context.  In 2013 the organization took me through the Negev, where it documents the same illegal-construction problem among Bedouin communities, and I filmed what they showed me. What I heard from them was not bigotry. Regavim acknowledged that the Bedouin had been treated badly by Israel in the decades after 1948, insisted that any solution had to give them a fair alternative, and argued that the state would have to spend serious money to provide it. Regavim's actual position is that the law should apply evenly and that the people affected by it deserve a just result. This is not how an extremist settler group would act. The EU is painting them as anti-Arab fanatics; the truth is quite the opposite. 

The regulation singles out  one case. Regavim petitioned over an EU-funded school at Jubbet adh-Dhib, near Bethlehem. An Israeli court found the structure had been built without a permit and posed a safety hazard to the children inside it, and COGAT imposed a deadline to vacate following the court's order. Regavim's role began and ended with the petition. The party that built without a permit was the European Union, and the party that ruled the construction unlawful was an Israeli court applying Israeli planning law. The EU has sanctioned the organization that brought the violation to the court's attention, while describing its own unpermitted construction as the injured party. Regavim's spokeswoman Naomi Linder Kahn put it precisely: the group's crime "involves petitioning to the courts against a dangerous, illegal, EU-funded school built on Israeli state land in a national historic site."

Strip out the rhetoric, and the only problem the EU has with Regavim is that it disagrees with its political opinions and legal advocacy. The EU has apparently forgotten its own Charter of Fundamental Rights, which guarantees everyone "freedom to hold opinions and to receive and impart information and ideas without interference by public authority." Sanctions are interference by public authority, and the EU has imposed them on an organization for its opinions and its lawful actions. If anyone is violating fundamental rights in this case, it is the EU, which has frozen a group's assets and barred its director from travel because it disagrees with the group's politics. The list of charges contains nothing illegal, because there is nothing illegal to charge. Regavim holds the position that Jews have a right to their ancestral land and that the law should be upheld to that end, and it has defended existing Israeli law in Israeli courts.

And guess where that law came from?

The entire international case against Israeli construction in Area C rests on a single foundation: that Israel is the belligerent occupier in the West Bank. Israel considers the territory disputed, but it agrees to adhere to the human rights portions of the laws of occupation in its administration of Area C. The law of occupation in the Hague Regulations Article 43 obligates the occupant to respect "the laws in force in the country" unless absolutely prevented. The occupier administers the territory under the legal order it inherited. In Area C, that order is the planning regime the EU's own allies describe in detail: under the 1966 Jordanian Planning Law, still in force in the West Bank, virtually any construction requires a permit issued in line with an approved scheme, and that scheme descends from Regional Outline Plans the British Mandate approved in the 1940s. When Israel took the territory in 1967, it took over the planning powers that the Jordanian law already conferred. Enforcing a permit requirement that predates the occupation by decades is the occupier doing exactly what Article 43 commands. This is international law.

The EU knows this regime applies to its own projects, because the UN says so plainly. Any Area C construction — and OCHA's enumeration names "a private home, an animal shelter or a donor-funded infrastructure project" — still requires approval from the Israeli Civil Administration, because the planning transfer to the Palestinian Authority that the Interim Agreement envisioned was never carried out. The EU builds anyway, without the permits the law in force requires, and then declares the result legal. A structure erected without a permit is unlawful under the precise legal order that occupation status keeps alive. The EU cannot invoke "this is occupied territory" to delegitimize one population's building and then fund the other population's building in open defiance of the planning law that occupation status itself preserves.

There is room for honest argument about how Israel applies the law, hardly ever approving Arab construction in Area C. But that is a problem with the application of the law, not the law itself.  The permit requirement itself is inherited law, not Israeli invention, and a building put up without a permit is illegal by the standard the EU insists governs the land. The EU's quarrel is not that Israel enforces a foreign legal order; it is that Israel enforces it against construction the EU paid for.

The low approval rate also reflects what Area C was built to be. Oslo II sorted the dense Palestinian population into Areas A and B, the urban and village zones where roughly 95 percent of West Bank Palestinians live, and left Area C as the strategic remainder: the settlements, the main roads, the Jordan Valley, the open land whose disposition the parties deferred to final-status talks. A zone defined in 1995 as the sparsely populated reserve, explicitly not the place where the Palestinian population was settled, will naturally generate a high rejection rate when permits are sought to build into it. Both sides understand the stakes. PASSIA states openly that Israel's aim in Area C is to push Palestinians toward A and B, and INSS records that Palestinians have run an organized legal campaign that has won authorization for 113 previously unauthorized villages — a build-first, litigate-later strategy rather than scattered individual need. Area C is contested strategic space whose population balance is being fought over before any negotiation, which is precisely the contest the EU funds on one side and sanctions on the other. The burden is not Palestinian alone: Jewish residents face their own thicket of permitting and legal review for construction and expansion, which is what one would expect of a planning regime applied across a disputed zone rather than one rigged in their favor.

The hypocrisy is stark. Regavim, Amana, and Nachala are accused, in the EU regulation's own language, of building or facilitating outposts "to create facts on the ground," of working to alter the territorial and demographic reality before any final-status agreement, of making it harder for the rival population to remain. This is exactly what the EU is doing on the other side! I took a tour with Regavim a decade ago and saw the EU's own program with my own eyes: clusters of structures thrown up across Area C without permits, flying the EU flag and bearing the words "Humanitarian Aid and Civil Protection," positioned in deliberate lines across valleys to block Jewish communities from forming, like stones in a game of Go. Residents were brought in from Areas A and B to populate villages where none had stood. I saw hoses running off to steal water from neighboring Israeli towns. That was ten years ago, and the trajectory was already unmistakable; the structures have only multiplied since, because the program is continuous and the EU has never hidden it. Brussels funds the building, brands it, and describes its purpose in the same terms the regulation now treats as a human rights crime when the builders are Jewish.

The conduct the EU finances and the conduct the EU condemns are the same conduct: unpermitted construction in Area C, undertaken to win the territorial contest before negotiations, by moving a population in. The EU has drafted a definition of the offense that indicts its own program word for word. The only variable that decides whether Brussels calls it "humanitarian aid" or "serious human rights abuse" is which population holds the trowel. A standard that condemns one party while the accuser performs the identical act is not a legal standard; it is a pervasion of the entire concept of equal standing under the law.

The EU's fallback is that this construction qualifies as humanitarian assistance that an occupied population may receive regardless of domestic permit law. The argument fails on the EU's own evidence. Humanitarian assistance under international humanitarian law is relief for a population in need, not permanent construction engineered to alter territorial control. The EU's description of the Israeli mirror-conduct — facts on the ground, contiguity, frustrating the other side's claims — is an admission that the purpose of such building is political and territorial rather than humanitarian. Brussels has characterized the activity accurately when Jews do it and mislabeled the same activity when it does it.

Return now to the principle the EU claims to live by, because the contradiction is not abstract. The EU describes human rights defenders — and its own list of them names "members of human rights NGOs, academics, lawyers" — as "natural and indispensable allies," and its Guidelines on Human Rights Defenders commit EU missions worldwide to oppose exactly the tools the EU has now reached for: "administrative and judicial harassment," "smear campaigns, travel bans, criminalisation, stigmatisation." When a foreign government freezes the assets of a research-and-litigation NGO and bans its director from travel because it dislikes the NGO's cause, the EU calls that repression and sends a démarche. When the EU does it to an Israeli NGO, it calls it a human rights sanction. The behavior the EU condemns abroad is the behavior it has adopted here, against an organization whose work — researching land use, publishing findings, filing petitions in a democratic state's courts — is the textbook profile of the defenders the EU vows to protect everywhere else.

Underneath the legal packaging lies a claim with no source in law: that Area C is already Palestinian sovereign territory, so that Israel's permit regime is a foreign imposition and EU construction needs no one's permission. The status of Area C is the question the Oslo framework deferred to final-status negotiations, and no treaty, ruling, or resolution has settled it. There is no moment in history where control of Area C was legally (or otherwise) transferred to the Palestinian Authority. The EU has decided the question by assertion, and then by funding the facts and punishing the litigants who challenged them in court.  

A European Union that built its identity on the promise that lawful opinion is beyond the reach of state punishment has now demonstrated, in a binding legal act, the precise conditions under which it will break that promise: when the opinion is inconvenient and the person holding it is Israeli.

Read More

aoav

 

Action on Armed Violence presents itself as an independent, non-partisan research organization, and the world treats it accordingly. Its Explosive Violence Monitoring Project has run for over a decade, its annual figures are cited by wire services and human-rights groups, and its data now feeds the United Nations: the Secretary-General's May 2026 report on the protection of civilians draws on AOAV's monitor, and the 2025 edition opens with a Guterres quotation calling on states to act on its findings. That standing is the reason the monitor's flaws matter. A partisan pamphlet that overstated civilian harm would persuade no one; a trusted, UN-cited, self-described neutral dataset that does the same thing launders the overstatement into the official record.

AOAV's stated methodology is structurally flawed in a way that inflates civilian counts. Worse, it suspends its own rules in the cases where applying them would have produced a figure unfavorable to the narrative the reports advance. 

Here's the methodology: AOAV records casualties from explosive weapons "as reported in English-language media," logging incidents that caused at least one casualty within a 24-hour period. Two rules govern how casualties are classified. The first, stated in the methodology: "All casualties are assumed to be civilians unless otherwise stated." A casualty becomes an armed actor only when a news report explicitly identifies the dead or wounded as a soldier, militant, or armed security official. The second concerns attribution: responsibility is assigned to whoever the news report names. The project then totals these media-derived entries and publishes them under the word "civilian," flat and unqualified, across fifty-odd pages of findings, country profiles, and recommendations. The qualifier that would make the count honest — that "civilian" means "not reported as armed," not "confirmed noncombatant" — sits in a single methodology paragraph the headline reader never reaches.

One of AOAV's headline 2023 incidents shows both rules failing at once. AOAV's third-worst explosive incident of 2023 is listed as "Israeli air strike on an evacuation convoy fleeing north Gaza, 13 October," 270 civilian casualties were attributed to Israel. The footnote sources it to a Sky News article, whose headline reads: "Women and children among 70 killed in Israeli airstrike on fleeing Gaza convoy, Hamas says." The attribution to Hamas is in the title. The body went further: Sky News reported the blast was "blamed on" Israel, quoted an IDF spokesperson saying he was "not aware of any IDF strikes at this time at that location," and stated that "it was not immediately clear who the target was, or whether insurgents were among the passengers." The Associated Press wire carried the same qualification verbatim. The reporting did what responsible reporting does with an unverifiable battlefield claim — it named the source, recorded the denial, and flagged that fighters might be among the dead.

AOAV erased all three caveats. The Hamas claim became a fact of an Israeli airstrike; the denial vanished; the possibility of combatants among the passengers dissolved into a flat count of 270 civilians. And the Israeli account was not a one-line "no comment." Within two days, IDF spokesman Jonathan Conricus argued publicly that the strike "appears to have been a false flag operation carried out by Hamas," asked who would benefit from images of dead evacuees and answered "only one organization: Hamas," and said Israel "did not try to strike anybody, any civilians in that area." The surrounding facts fit that reading at least as well as the alternative: Israel had ordered the evacuation and designated the route, Hamas had told Gazans to defy the order and was reported blocking the southern roads, and Hamas held the clearer motive to manufacture atrocity footage from a convoy of fleeing civilians. There was no evidence of an actual airstrike beyond Hamas's assertion of one. The cause remains formally unresolved; reference works still list the attack type as airstrike or possible IED. AOAV recorded it as settled Israeli strike of 270 civilian casualties, third-worst in the world that year. The media's attribution — "Hamas says" — is the qualifier that made the claim publishable, and AOAV's method strips precisely that qualifier, converting "a party to the war alleges" into "AOAV records."

The civilian-by-default rule produces an even stranger result across Gaza as a whole. In the 2024 monitor, AOAV records 23,432 civilian casualties in Gaza against 612 armed actors — a civilian share of 97.5%, with combatants making up 2.5% of the total and roughly 3% of the dead. If that isn't insane enough, the military casualties it recorded were apparently all IDF soldiers! In other words, Hamas never admits any of its terrorists were killed until years later, Israel admits its soldiers are killed within hours of the incident, so AOAV thinks virtually the only militants killed in Gaza are Israeli. QEDumb. 

The same selection problem corrupts AOAV's most quotable Gaza statistic. The 2024 monitor reports that Israel's recorded aerial attacks in Gaza caused, on average, "8 civilians harmed per recorded Israeli air strike, and 5 killed," and the 2023 edition builds a comparable per-airstrike figure. Read the methodology and the number dissolves. An incident enters the dataset only if "at least one casualty from an explosive weapon" was reported; a strike that hit an empty structure, or killed only fighters the media did not count, or fell where no English-language reporter was watching, never appears. The denominator is therefore not "Israeli air strikes" but "Israeli air strikes that produced reported casualties," and AOAV divides the casualties by that pre-filtered set to announce a casualty rate. The strikes that would pull the average down were excluded before the division.

It excluded a huge number of airstrikes. AOAV's 2023 dataset logs a few hundred air-launched incidents in Gaza; the Israeli military, by contrast, said it had struck over 11,000 targets in Gaza by 1 November 2023 and more than 22,000 by mid-December, with independent reporting putting the four-month total near 29,000 targets, roughly 228 a day. Whatever the precise count of distinct air operations, it dwarfs the few hundred AOAV recorded by orders of magnitude. The "civilians per air strike" figure does not describe Israeli targeting; it describes the handful of strikes that left a reportable civilian trail, which is the only kind the method can see. A monitor that counts only the strikes that killed civilians and then reports how many civilians strikes kill has not measured lethality. It has measured its own selection rule.

The absurdity reaches its peak with Lebanon, where the method collides with a fact pattern it cannot survive. AOAV's single worst explosive incidents of 2024 were the September pager operation and the next day's walkie-talkie operation. The explosives were concealed inside pagers and radios that Hezbollah itself had purchased and distributed to its members; the only people physically carrying them were Hezbollah personnel. Hezbollah acknowledged that hundreds of its fighters carried the devices, and a Hezbollah official said the operation took 1,500 fighters out of action through injury. Hezbollah's own leadership has long rejected the premise that there is any line between its political and military sides; as leader Naim Qassem put it, "we don't have a military wing and a political wing... one Hezbollah." A weapon that by physical design could only injure members of the organization, in an operation the organization says wounded 1,500 of its fighters, produced in AOAV's ledger a 97% civilian casualty rate. The monitor even concedes, in its own text, that "armed actors are likely included among these casualties." Here the method does not merely risk error. Its output is contradicted by the admitted facts of the event it is describing, including the facts AOAV itself records.

If the story ended with a flawed method honestly applied, AOAV could fairly answer that open-source monitoring is imperfect and its limitations are disclosed. The defense collapses on the evidence that the method is not applied when it would point away from blaming Israel.  The proof is an absence: the al-Ahli hospital explosion of 17 October 2023.

Al-Ahli was the most heavily reported explosive event of the early war. Within hours, Gaza's Ministry of Health announced 471 killed and blamed an Israeli airstrike, and the figure led news bulletins worldwide. Two findings then emerged from exactly the English-language sources AOAV scans. On the toll: the Anglican diocese that runs the hospital estimated around 200 dead, the director of al-Shifa Hospital put it near 250, and US intelligence assessed 100 to 300, likely at the low end. On the cause: US, British, Canadian, and French intelligence, along with Human Rights Watch, concluded the explosion came from a misfired Palestinian rocket that struck the courtyard rather than an Israeli strike on the building.

Now apply AOAV's own rules and watch the contradiction close around the report. AOAV publishes a table of the ten worst explosive incidents of 2023, ranked by civilian casualties; the smallest entry on it is a Pakistani suicide bombing at 193. Al-Ahli is not on the table. It is not anywhere else in the report either: across the entire document, the single most-reported explosive event of the early Gaza war goes unmentioned. There are only two ways AOAV could have reached that result, and each indicts a different rule.

If AOAV applied its standard practice — take the casualty figure reported in the first 24 hours, assume all civilian, attribute to the named party — then al-Ahli enters as 471 killed and 314 injured, attributed to Israel. That is 785 casualties, the single deadliest incident of 2023 by AOAV's stated criteria. 

If instead AOAV declined the day-one figure because it was disputed and the misfire finding undercut the attribution — exercising judgment its 24-hour rule does not provide for — then it accepted a corrected toll of roughly 100 to 300 dead and several hundred injured. Even at the floor of that range, killed-plus-injured clears the 193 threshold several times over. It still belongs in the top ten. There is no version of AOAV's methodology under which al-Ahli is correctly absent. Recorded by the rules, it tops the table. Corrected against the rules, it still ranks. Omitted entirely, it reveals a choice.

That choice is the whole case. To leave al-Ahli out, AOAV had to do the one thing it tells the world it does not do: look past the day-one wire copy, weigh the later corrections, and decide the incident did not belong. An event missing from a top-ten ranking might be an editorial judgment about where a line falls. An event missing from the entire report — when AOAV names, profiles, and tallies every other major Gaza incident of 2023, and builds its headline Gaza total from them — is not a line-drawing problem. It is a removal. The capacity to follow up and verify plainly exists; the report's own caveats about Gaza undercounting show AOAV reading the sources closely. That capacity was exercised on the one 2023 Gaza incident almost everyone agrees a Palestinian faction caused and that pointed away from Israel, and it was switched off for the convoy strike Israel denied, for the thousands of unlabeled Gaza dead booked as civilians, and for the Hezbollah fighters counted as bystanders to their own pagers. The verification machinery runs in exactly one direction.

Set the two 2023 incidents side by side and the pattern is unmistakable. The blast that pointed away from Israel — disputed attribution, a toll its own sourcing showed was inflated — was dropped from the ranking despite belonging at or near the top of it. The blast that pointed at Israel — a Hamas claim the IDF denied, with no evidence of a strike beyond the accusation — was elevated to settled fact and placed third. A monitor genuinely indifferent to which side a headline blamed could not sort two events so cleanly along that line.

This is why AOAV's standing is the heart of the problem rather than a footnote to it. The organization's authority rests entirely on the claim that it neutrally records what the media reports. The convoy entry shows it discarding the media's own caveats when they protect Israel. The Gaza and Lebanon totals show the civilian-by-default rule manufacturing combatant-free wars out of conflicts against armed organizations. And al-Ahli shows that the rule can be suspended at will, exercised precisely when suspension serves the narrative. Each example points the same way, and the cumulative weight is hard to read as accident. An organization with this method, applied this selectively, is not producing a flawed measurement of civilian harm. It is producing an argument, and dressing it as data — then handing it to the United Nations, which cites it as the considered judgment of an independent observer.

In short, the only consistency that AOAV shows in its methodology in the Middle East is whatever makes Israel look as bad as possible. 

(h/t Irene)

Read More