Iran Today

The Law Is Clear: Human Shields and the Law

The Prison Guard

In November 2019, a sixty-year-old Iranian man named Hamid Nouri stepped off a plane at Stockholm’s Arlanda Airport.1 He had come to visit relatives. Swedish police were waiting.

Nouri had served as an assistant prosecutor at Gohardasht Prison in 1988 — the summer when Iran’s Death Commissions, operating on Khomeini’s fatwa, executed thousands of political prisoners in a matter of weeks. He had helped send men and women down the corridor to the execution hall. And for thirty-one years, he had lived freely — because no court with jurisdiction had reached him.

Sweden’s universal jurisdiction statute changed that. Survivors identified Nouri from events three decades prior. In July 2022, he was convicted of war crimes and murder. The sentence: life imprisonment. The appeals court upheld it. The Supreme Court upheld that.2

Then, on June 15, 2024, Sweden released him.

Iran had detained a Swedish diplomat, Johan Floderus, specifically as leverage during Nouri’s trial. When the prisoner exchange was completed, a life sentence for mass murder was traded for a hostage. Nobel laureate Shirin Ebadi called it “detrimental to justice.” The Atlantic Council warned it had “detonating effects on the universal jurisdiction project.”3

The Nouri case captures the central paradox of international humanitarian law as it applies to Iran: the law is clear. The enforcement is not.


What the Law Says

The legal framework governing human shields is built from four interlocking texts — the Geneva Conventions, their Additional Protocols, the Rome Statute, and customary international humanitarian law. For a topic that generates so much political argument, the actual legal provisions are remarkably direct.4

The prohibition. Geneva Convention IV, Article 28 — written in 1949 — states: “The presence of a protected person may not be used to render certain points or areas immune from military operations.” This is an absolute prohibition with no exceptions. The commentary written by the International Committee of the Red Cross in 1958 explains why it exists: Axis powers during World War II compelled civilians to remain at railway stations, viaducts, and military installations to deter Allied bombing.5

The broadening. Additional Protocol I, adopted in 1977 and ratified by Iran, expanded the prohibition in Article 51(7). Where the original convention covered “protected persons,” AP I covers all civilians. Where it covered “presence,” AP I covers “movements” — addressing both active shielding (directing civilians toward targets) and passive shielding (exploiting existing civilian proximity).6

The defender’s obligation. Article 58 of AP I imposes affirmative duties on the defending party — not the attacker, but the party controlling the territory. To the maximum extent feasible: remove civilians from the vicinity of military objectives, avoid locating military objectives within densely populated areas, and take other necessary precautions to protect civilians under its control.7

These are not aspirational recommendations. They are legal obligations, qualified only by what is “feasible” — and for an IRGC that maintains extensive underground bunker networks, hardened command centers, and dispersed military installations, the range of feasible alternatives to placing assets inside hospitals and schools is very wide.

The medical protection. Article 12(4) states: “Under no circumstances shall medical units be used in an attempt to shield military objectives from attack.” The phrase “under no circumstances” permits no exception. The ICRC’s 2025 analysis confirms that medical protection can only be lost “in the rarest and most exceptional circumstances.”8

The criminalization. The Rome Statute of the International Criminal Court, Article 8(2)(b)(xxiii), codifies human shielding as a standalone war crime: “utilizing the presence of a civilian or other protected person to render certain points, areas or military forces immune from military operations.” This is the first time in history that the act of placing civilians at risk was defined as a crime independent of any resulting harm.9


Who Bears Responsibility

This is the question that matters most — and the one most frequently distorted in public debate. When a military force places its assets next to a school, and an opposing force strikes that target and kills children, who is legally responsible?

The answer, under international law, is both. Independently. Simultaneously.

The ICTY established this in Prosecutor v. Galić, stating that a defending party’s “failure to abide by this obligation does not relieve the attacking side of its duty to abide by the principles of distinction and proportionality.”10 Article 51(8) of AP I — sometimes called “the does not release clause” — makes this explicit: “Any violation of these prohibitions shall not release the Parties to the conflict from their legal obligations with respect to the civilian population.”

What this means in practice: the party using human shields violates Article 51(7), commits a war crime under the Rome Statute, and violates Article 58’s affirmative duty to separate civilians from military targets. The party striking the shielded target retains every obligation of distinction, proportionality, and precaution under Articles 48, 51(5)(b), and 57.11

Neither party’s violation excuses the other. As legal scholar Chris McQuade writes: “Responsibility for civilian casualties is not a zero sum game — such responsibility can be split between both parties.”12

But here is the critical asymmetry that makes human shielding strategically rational: the attacker’s obligations are enforced by international media, human rights organizations, courts, and public opinion. The shielder’s obligations are enforced by almost nothing.


The Proportionality Debate

Scholars disagree about one specific question: when shields are present, should the proportionality calculation — the legal requirement to ensure civilian harm is not “excessive in relation to the concrete and direct military advantage anticipated” — be adjusted?13

Three camps exist.

Those who argue proportionality should be relaxed. International law scholar Yoram Dinstein argues that “the actual test of excessive injury to civilians must be relaxed” when shields are present, and that human shields should “carry less weight than other civilians in the proportionality calculation.”14 The U.S. Department of Defense Law of War Manual and the UK Joint Service Manual both treat the defending party’s responsibility as a relevant “factor” in proportionality assessment.

Those who argue full attacker liability regardless. Beth Van Schaack — Stanford professor and former U.S. Ambassador-at-Large for Global Criminal Justice — argues that efforts to “relax” proportionality should be “rejected,” documenting how Sri Lanka invoked LTTE shield use to justify “deliberate, indiscriminate, and disproportionate attacks.”15 The ICRC maintains that all shields retain full civilian status and full weight. Amnesty International and Human Rights Watch hold the same position.

The middle ground. Michael Schmitt of the Lieber Institute at West Point — lead author of the Tallinn Manual and one of the most cited scholars in the field — opposes relaxation for involuntary shields: “It allows a malevolent enemy engaged in a war crime to deprive the very individuals IHL is meant to protect of some of the protection they enjoy as civilians.” He objects on practical grounds — it “begs the question, ‘relaxed how and how much?’” — and principled ones — it “relies upon an approach to IHL that has long been rejected.”16

What all three camps agree on: the party using human shields commits a violation of international law regardless of whether the shields are attacked. The Blaškić judgment holds that the crime is complete upon the act of placing civilians at risk — “regardless of whether those human shields were actually attacked or harmed.”17


Iran’s Obligations

Iran is not a bystander to this legal framework. It is a party to it.

Iran ratified all four Geneva Conventions. Iran ratified Additional Protocol I. Every provision discussed above — Article 51(7)‘s prohibition on shields, Article 58’s obligation to separate civilians from military targets, Article 12(4)‘s absolute protection of medical facilities — is directly binding on the Islamic Republic and its military forces, including the IRGC.18

Iran signed but never ratified the Rome Statute, which means the ICC cannot exercise jurisdiction over Iranian territory directly. But customary international humanitarian law — particularly ICRC Customary IHL Rule 97 — extends the prohibition on human shields to all conflicts and all parties, regardless of treaty status.19

The IRGC has no publicly available military manual addressing international humanitarian law. It has no known proportionality guidance for commanders. Its internal doctrine, as characterized in a Foundation for Defense of Democracies assessment of IRGC force structure, centers on “the concealment of strategic assets in places the enemy cannot target” — language that describes the precise conduct international law prohibits.20


Accountability Pathways

If the law is clear, why does enforcement fail? The Nouri case provides the template — and the warning.

Universal jurisdiction. Germany’s Code of Crimes Against International Law, Sweden’s legal framework, and similar statutes in roughly a dozen countries allow prosecution of international crimes regardless of where they were committed. Nouri’s conviction proved the concept: survivor testimony, even decades old, can sustain convictions for crimes committed in Iran. But Nouri’s release proved the vulnerability: hostage diplomacy can nullify what courts establish.21

The ICC pathway. The ICC’s investigation into the Palestine situation — which produced arrest warrants in November 2024 for both Israeli and Hamas leaders — offers a theoretical pathway from Gaza to Tehran. If Hamas commanders face charges for crimes directed by IRGC officers, command responsibility doctrines could extend to Tehran. But reaching up the command chain from Gaza to the IRGC faces jurisdictional barriers that have never been tested.22

UNSC referral. The Security Council can refer any situation to the ICC regardless of treaty membership. Russia and China have reliably vetoed such referrals for every case where their allies might be implicated.23

Targeted sanctions. IRGC commanders are already designated under U.S. terrorism authorities. But no Magnitsky-style sanctions have specifically targeted individuals for ordering human shield deployment — a gap that could be addressed without requiring UN consensus.24

Evidence preservation. What may matter most in the long term. The Iranian Archive has preserved over two million digital artifacts. The UN Fact-Finding Mission has collected over 38,000 evidence items. Mnemonic’s forensic verification methodology produces evidence that meets courtroom standards. Leaked documents reference specific IRGC mission orders from the Mohammad Rasulullah Corps of Greater Tehran and authorization from the Imam Ali Central Security Headquarters — providing chain-of-command evidence linking individual commanders to specific hospital occupations.25

The information half-life of atrocities is compressing — from decades (the 1988 massacres) to days (Bucha 2022) to hours (the January 2026 crackdown, where footage from the Kahrizak forensic center showing over 250 body bags leaked within hours of brief satellite connections).26 What this means: the evidence is being preserved faster than it can be destroyed. The accountability question is when, not whether.


The Paradox

The Nouri case distills the paradox that governs all of this.

The law against human shields is clear — codified in treaties Iran ratified, established in customary law binding on all states, upheld by international tribunals, and defined as a standalone war crime. The evidence of the IRGC’s systematic violation is extensive — twelve documented incidents across nine cities, verified by multiple independent organizations, corroborated by the regime’s own officials.

The enforcement is broken. The IRGC’s commanders rarely travel to countries with universal jurisdiction statutes. Nouri’s arrest was opportunistic — a former prison guard visiting relatives who stumbled into Swedish jurisdiction. The senior commanders who ordered the hospital sieges and school occupations of 2025-2026 are not likely to make the same mistake.

Consider the logic from a commander’s perspective: if placing a weapons depot next to a school means your enemy either cannot strike it or will be condemned internationally for doing so, the incentive is clear. The children pay for it either way.

This is the perverse incentive that makes human shielding rational: the party using shields faces near-zero accountability while the party striking shields faces intense legal, political, and media scrutiny. Every commander watching this asymmetry draws the same conclusion.27

The counterfactual discipline is instructive. Any interpretation that reduces attacker liability for strikes on shielded targets incentivizes more shield use. Any interpretation that ignores the shielder’s responsibility concentrates enforcement on the more legally accessible party. The only framework that serves civilian protection is one that holds both parties simultaneously accountable — and that requires something that does not yet exist: effective, symmetric enforcement.28

Marco Sassòli of the Geneva Academy captures the core principle: international humanitarian law’s “fundamental approach is not to evacuate civilians so that belligerents can conduct hostilities with minimal restraint, but rather to oblige the military to conduct hostilities taking the presence of civilians into account.”29

A military force that hides behind the civilians it claims to protect has forfeited any claim to legitimate authority — and violated the Islamic jurisprudence on warfare that it invokes to justify its existence.

The law was written to protect the girls at Kosar Elementary. The law was written to protect the patients at Imam Khomeini Hospital. The law was written to protect the students in Minab.

The evidence is being preserved. The legal framework exists. The precedent is set. The question is whether anyone will enforce it against the people who put those children in harm’s way.30


This article is part of The Shield and the Schoolyard. For the proxy pipeline, see From Beirut to Minab. For the forensic evidence, see Twelve Incidents, Nine Cities.

Footnotes

  1. Hamid Nouri arrested at Stockholm’s Arlanda Airport, November 2019, on a tip from a former son-in-law. He had served as assistant prosecutor at Gohardasht Prison during the 1988 mass executions.

  2. Nouri trial: ninety-two sessions, thirty-four plaintiffs, twenty-six witnesses. Convicted of war crimes and murder (July 2022). Life imprisonment upheld on appeal (December 2023) and by the Swedish Supreme Court.

  3. Nouri released in prisoner exchange on June 15, 2024, for Swedish diplomat Johan Floderus and dual-national Saeed Azizi. Iran had detained Floderus specifically as leverage during the trial.

  4. The four-text framework: Geneva Conventions (1949), Additional Protocols (1977), Rome Statute (1998), and customary international humanitarian law as documented in the ICRC Customary IHL Study.

  5. Geneva Convention IV, Article 28 (1949). Pictet Commentary (1958) identifies the provision as a direct response to Axis civilian shielding practices during World War II.

  6. Additional Protocol I, Article 51(7), broadening coverage from “protected persons” to all civilians, and from “presence” to “movements.” Iran ratified AP I.

  7. AP I, Article 58, affirmative obligations on the defending party. Qualified by the “feasible” standard — not absolute perfection but reasonable effort.

  8. AP I, Article 12(4): “Under no circumstances shall medical units be used in an attempt to shield military objectives from attack.” ICRC 2025 analysis on the rarity of legitimate circumstances for loss of protection.

  9. Rome Statute, Article 8(2)(b)(xxiii). The Elements of Crimes require proof the perpetrator moved or took advantage of civilian location with specific intent to shield military objectives.

  10. ICTY, Prosecutor v. Galić (IT-98-29), Trial Chamber, paragraph 61. Appeals Chamber upheld and increased sentence to life imprisonment.

  11. AP I Articles 48 (distinction), 51(5)(b) (proportionality), and 57 (precaution). Article 51(8) establishes non-reciprocal obligation — one party’s breach cannot justify another’s.

  12. Chris McQuade, Opinio Juris, 2023, on the non-zero-sum nature of civilian casualty responsibility.

  13. AP I, Article 51(5)(b): prohibition on attacks “expected to cause incidental loss of civilian life, injury to civilians, damage to civilian objects, or a combination thereof, which would be excessive in relation to the concrete and direct military advantage anticipated.”

  14. Yoram Dinstein, The Conduct of Hostilities Under the Law of International Armed Conflict, 4th edition (2022). See also Rubinstein and Roznai, Stanford Law and Policy Review (2011), proposing “proportionate proportionality.”

  15. Beth Van Schaack, Stanford, former U.S. Ambassador-at-Large for Global Criminal Justice. Documents Sri Lanka’s invocation of LTTE shields to justify attacks on civilian hospitals and safe zones.

  16. Michael Schmitt, Lieber Institute at West Point, lead author of the Tallinn Manual. Schmitt distinguishes voluntary from involuntary shields — accepting that voluntary shields may constitute direct participation in hostilities while maintaining full protection for involuntary shields.

  17. ICTY, Prosecutor v. Blaškić (IT-95-14), Trial Chamber, paragraph 716. Using civilians as shields constitutes “inhuman treatment” and “cruel treatment” — the “considerable mental suffering” alone suffices even without physical harm.

  18. Iran ratified all four Geneva Conventions and Additional Protocol I. These treaties are directly binding on all Iranian military forces including the IRGC.

  19. ICRC Customary IHL Rule 97: the prohibition on human shields applies in all armed conflicts, international and non-international, regardless of treaty ratification. Iran signed the Rome Statute on December 31, 2000 but never ratified it.

  20. The IRGC has no publicly available military manual addressing IHL compliance. Internal doctrine described as focused on “concealment of strategic assets in places the enemy cannot target.”

  21. Universal jurisdiction statutes exist in Germany, Sweden, France, Belgium, the Netherlands, and approximately a dozen other countries. The Nouri case demonstrated both the potential and fragility of this pathway.

  22. ICC Palestine situation: arrest warrants for Netanyahu, Gallant, and Hamas commander Deif (November 2024). Charges focused on starvation, murder, and crimes against humanity — not specifically human shields. Command responsibility pathway from Gaza to Tehran has never been tested.

  23. Russia and China have vetoed UNSC referrals of Syria and other situations to the ICC. The same dynamic applies to any potential Iran referral.

  24. No Magnitsky-style sanctions have specifically targeted IRGC commanders for ordering human shield deployment. Existing IRGC designations are under terrorism authorities rather than human rights frameworks.

  25. The Iranian Archive (Atlantic Council DFRLab + Mnemonic): 2+ million digital artifacts. UN FFMI: 38,000+ evidence items, ~300 interviews. Leaked IRGC documents reference mission orders from the Mohammad Rasulullah Corps and Imam Ali Central Security Headquarters.

  26. Information half-life compression: 1988 massacres (decades to document), MH17 2014 (months), Bucha 2022 (days), Iran January 2026 (hours). Kahrizak forensic center footage leaked via Starlink connections.

  27. The enforcement asymmetry: attackers face media, legal, and political scrutiny; shielders face near-zero accountability. This makes human shielding strategically rational under current enforcement conditions.

  28. The “complementary duties framework” (Stanford) — holding both parties simultaneously accountable — produces the best civilian protection outcomes in theory but requires symmetric enforcement that does not yet exist.

  29. Marco Sassòli, Geneva Academy, on IHL’s fundamental approach to civilian presence and military operations.

  30. Evidence preservation mechanisms (the Iranian Archive, UN FFMI, Mnemonic forensics) create permanent records. The trajectory from decades to hours in exposure speed is transforming the calculus — but the enforcement gap remains.