COLLECTIVE PUNISHMENT

In antizionist use, “collective punishment” leads two lives that have to be told apart. The first is legal. Article 33 of the Fourth Geneva Convention holds that no protected person may be punished for an offense he or she did not personally commit. The second is rhetorical. On campuses, in NGO reports, and at protests, the phrase often shrinks into shorthand for any Israeli action that harms civilians in Gaza or the West Bank, with the legal theory left unstated. The two meanings deserve to be kept separate, because only one of them is law.

Article 33 is a serious rule, not a synonym for every civilian hardship in war. It does not decide, on its own, whether a blockade or a closure, an evacuation order or a strike, a checkpoint or a sanction or a demolition meets the legal test. That depends on facts and intent, on authority and necessity and proportionality, and on the status of the people affected. Lawyers and governments genuinely dispute how Article 33 applies to particular Israeli measures in Gaza and the West Bank. The phrase ought to name a legal argument. It should not be allowed to stand in for one.

The social use runs in a different direction. There, “collective punishment” slides from a claim about Israeli conduct to a demand placed on Jews elsewhere, and the logic is brutally simple: if Israel is guilty, then Jews who do not condemn Israel loudly enough are complicit. Once tagged as complicit, they can be shamed or boycotted, excluded or made to answer for policies they never enacted. A Jewish student in Boston does not become a cabinet minister in Jerusalem because someone wants a target closer to home. So the term lives a strange double life. It condemns punishing people for acts they did not personally commit, yet it circulates in spaces that hold Jews collectively answerable for Israel. Its legal meaning is precise; its social use is often anything but.

The phrase was standard NGO language for Gaza well before the current war. In July 2008, Amnesty International published “Gaza Blockade: Collective Punishment,” describing the tightened blockade after June 2007 as a man-made humanitarian crisis. UN officials and NGOs, student groups and boycott campaigns have leaned on it ever since, especially in arguments over the blockade, movement restrictions, punitive demolitions, and closures. After October 7, 2023, it became ordinary vocabulary in campus protests and faculty letters, divestment resolutions and ceasefire campaigns. In many of those settings it no longer arrived as a legal claim to be proved. It arrived as a conclusion already reached.

A related logic drives the academic boycott campaigns. In March 2022, the Middle East Studies Association passed a BDS resolution endorsing the 2005 Palestinian civil-society call. In July 2023, the American Anthropological Association voted 71 percent to endorse an academic boycott of Israeli institutions. The AAA said the boycott reached institutions and not individual scholars, yet the policy still treats institutional affiliation as enough to trigger exclusion from association benefits. That is not Article 33, and the entry does not pretend it is. It belongs, though, to the same broader habit of assigning liability through association, which is the thread worth following.

The move from institutions to Jews is visible in the campus demands that followed October 7. The ADL’s May 2024 survey of encampment demands reported calls to cut ties with Hillel and Chabad, Jewish charities, “Zionist” speakers and organizations, and Israeli faculty. Some of those specific claims need case-by-case review, and the pattern still matters: a demand aimed at Israeli state policy keeps turning into a demand to exclude Jewish communal life. The same logic surfaced off campus too. In Philadelphia in December 2023, protesters targeted Goldie, a falafel restaurant tied to the Israeli-born Jewish chef Michael Solomonov, accusing him of complicity in attacks on Gaza. At UC Berkeley on February 26, 2024, a scheduled talk by the Israeli lawyer Ran Bar-Yoshafat was shut down after about 200 protesters descended on Zellerbach Playhouse; the university reported broken windows and a broken door, and attendees described assaults and antisemitic slurs. None of these actions touched a decision-maker in the Israeli cabinet. They fell on local Jews, Israelis, and Jewish student spaces serving as stand-ins.

The implied message is that Jewish identity carries state liability. The person being confronted may not live in Israel or vote there, may hold no government post and play no part in any military decision—and none of that matters once he is treated as the local face of the Jewish state. That standard is rarely applied to anyone else. Arab Americans are not asked to answer for every Arab government, Iranian Americans are not held personally responsible for the Islamic Republic, and Chinese Americans are not made to account for Beijing. Jews, however, are told directly and indirectly that their standing in public life depends on where they come down on Israel. That selective demand is exactly what makes the term dangerous once it leaves legal argument and enters social sorting, because it borrows the language of anti-collective liability while imposing collective liability on Jews. The contradiction is the whole point.

The notion that Jews share group guilt is very old. One of its earliest and most consequential forms was the deicide charge, the claim that Jews as a people bore permanent responsibility for the crucifixion of Jesus, which over centuries helped mark them as a suspect community inside Christian Europe. The Fourth Lateran Council of 1215 required Jews and Muslims in Christian lands to be set apart by dress. Enforcement varied, and later eras built their own systems of marking. The claim here is not that medieval Europe and the modern campus are the same place. It is that a community could be marked in public because of an inherited group status it did nothing individually to earn.

The ritual-murder myth carried that pattern into a new shape. Beginning with William of Norwich in 1144, Jews were said, as a people, to murder Christian children, and the charge never depended on evidence against any particular Jew. It worked because the group had already been rendered suspect. The expulsions from England in 1290, France in 1306, and Spain in 1492 had many causes, yet each drew in part on the premise that a whole Jewish community could be treated as a danger. Modern antisemitism then changed the vocabulary while keeping the structure. In the Dreyfus Affair, one Jewish officer’s false conviction for treason became a national drama because many assumed Jews held divided loyalties. Under the Nuremberg Laws of 1935, German Jews were stripped of citizenship and defined by ancestry instead of conduct. In Stalin’s Soviet bloc, Jewish guilt was recoded as “Zionist” conspiracy, and the 1952 Slánský trial in Czechoslovakia prosecuted 14 Communist officials, most of them Jews—all convicted, 11 executed—in one of the clearest cases of antisemitism dressed in anti-Zionist language.

In its legal arena the term brings Israel within the rules of international humanitarian law, which can be entirely legitimate where a speaker identifies a specific state act, a legal rule, a factual record, and a theory of liability. Article 33 is real law. The trouble starts when the legal term becomes a shortcut, because then every hardship reads as proof and every reply sounds like evasion. Used that way, the phrase delivers a finding before the case is built, flattening the blockade, a strike, an evacuation order, a checkpoint, and a border closure into one identical conclusion. The harder questions vanish in the process: What was the target? What was the stated purpose? What alternatives existed? What did Hamas do? What did Palestinian authorities do? What did Israeli officials know? Who actually caused the harm?

Socially, the phrase becomes a gate. Jewish students and scholars, artists and business owners, synagogues and campus groups can be treated as suspect until they first perform the required denunciation. The question stops being “What did you do?” and becomes “What kind of Jew are you?”—which is how a courtroom term gets dragged outside and used at the door. The April 1933 Nazi boycott of Jewish businesses is not an exact comparison, since it was state-organized and enforced by intimidation and violence. Even so, the underlying move is worth marking, because it punished Jewish businesses as Jewish businesses and not because any owner had personally done wrong. Whenever a campaign shifts from opposing a state to targeting Jews as Jews, that same warning light comes on.

For the record, Article 33 reads: “No protected person may be punished for an offence he or she has not personally committed. Collective penalties and likewise all measures of intimidation or of terrorism are prohibited.” The ICRC’s casebook explains that collective punishment covers not only criminal penalties but also sanctions, harassment, and administrative action taken against a group in retaliation for individual acts. A person can argue in good faith that a specific Israeli policy violates that article; that is a legal argument and a fair one. It turns into something else when Jews in London or Boston, Los Angeles or Toronto or Melbourne are made to answer for the policy because they are Jews or Zionists or Israelis or members of a Jewish institution.

Incident data tracks the shift, with the usual caution. The ADL’s final 2023 audit reported 8,873 antisemitic incidents in the United States for the year, 5,204 of them between October 7 and year’s end, and said 52 percent of the post-October 7 incidents referenced Israel, Zionism, or Palestine. The ADL’s methodology and classification choices are debated, so these belong in the record as tracking data and not as a court’s findings. They still show how fast Israel-related anger entered antisemitic incident reporting. None of this places Israeli policy above criticism, which should be as open as criticism of any state. The line is crossed when criticism of Israel hardens into a demand that Jews prove their innocence—when the message stops being “state your view” and becomes “prove you are not one of the bad Jews.” A government can be criticized without turning a people into defendants, and a serious account of civilian suffering in Gaza has to include Hamas governance, the use of civilian sites and tunnels, hostage-taking and rocket fire and aid diversion, along with the Palestinian Authority’s role in the West Bank. That is accountability. The other thing is exclusion.