The red cross on a white background. You see it on ambulances, on hospitals, on aid trucks in every war zone on earth. But what the organization behind that emblem actually does, and what it's legally allowed to do, is something most people have never had to think about until a conflict lands on their doorstep. Daniel's been looking at this and he's got a whole set of questions. He wants to know about the ICRC's founding mission as a neutral humanitarian actor, its legal mandate under the Geneva Conventions to visit prisoners of war, and then he zeroes in on a specific controversy. After October seventh, Hamas took around two hundred and fifty hostages into Gaza. Israel spent months publicly criticizing the ICRC for failing to demand visiting rights. Daniel's asking whether those hostages qualified as prisoners of war under Article Four of the Third Geneva Convention, what the actual mechanics of Article One Twenty-Six visiting rights are, and whether the criticism of the ICRC was legally and practically justified. That's four questions wrapped in one, and they all turn on a legal definition most people get wrong.
To answer that, we need to start with what the ICRC was actually created to do, and what it was not created to do. The organization was founded in eighteen sixty-three by a Swiss businessman named Henry Dunant. He'd wandered onto the aftermath of the Battle of Solferino in northern Italy in eighteen fifty-nine, something like forty thousand dead and wounded left on the field with no medical care, no organized relief, just local villagers doing what they could. And here's the detail that always gets me. Dunant wasn't a doctor, wasn't a soldier, wasn't a diplomat. He was there on a business trip, trying to get land rights from Napoleon the Third. Instead he spent days organizing local women to bring water and bandages to the wounded, regardless of which side they'd fought on. He wrote a book about it called "A Memory of Solferino," and out of that came two things. The ICRC itself, and the first Geneva Convention in eighteen sixty-four. The original mission was narrow. Neutral humanitarian relief for wounded soldiers. Not political advocacy, not enforcement, not public condemnation. Just get the bandages where they're needed and don't get shot while doing it.
Neutrality as the price of entry.
And that's still the core of how the organization works. Today the ICRC's mandate is rooted in the four Geneva Conventions of nineteen forty-nine and their Additional Protocols. It's often described as the guardian of international humanitarian law. But guardian doesn't mean enforcer. The ICRC has no police, no army, no power to compel anyone to do anything. Its access depends entirely on being trusted by all parties to a conflict. If one side decides the ICRC is hostile or biased, the doors close and stay closed. Think of it like a doctor in a prison. The doctor gets to see the inmates because everyone agrees the doctor is just there to provide medical care, not to judge, not to report to the warden, not to take sides in whatever gang conflict is happening inside. The moment the doctor starts picking favorites, they lose access to the other side's inmates. And then those inmates get no medical care at all.
So the thing people miss is that the ICRC's effectiveness and its structural weakness are the same thing. Neutrality buys access, but it also means you can't publicly shame anyone without losing the access that neutrality bought you.
That's the tension. And it's the key to understanding the Israel-Hamas controversy. The ICRC negotiates privately. Its reports on what it finds in prisons and detention centers go to the detaining power, not to the public. The theory is that quiet diplomacy gets results that public grandstanding doesn't. Sometimes that's true. Sometimes it isn't. But that's the model, and it's baked into the organization's DNA. There's actually a famous example from World War Two that illustrates this perfectly. The ICRC visited Allied POW camps in Germany and Axis POW camps in Allied territory throughout the war. They saw terrible conditions on both sides. They documented them. They reported them confidentially to the detaining powers. And they said almost nothing publicly. After the war, they were criticized heavily for not exposing what they'd seen. Their response was, if we'd gone public, we would have been expelled, and then we couldn't have delivered millions of Red Cross parcels to POWs who would have starved without them. That's the trade-off, right there.
So with that structural tension in mind, let's get into the legal weeds. Because the definition of who counts as a prisoner of war is where this whole controversy turns.
Article Four of the Third Geneva Convention. This is the one that defines who gets POW status. The main categories are members of the armed forces of a party to the conflict, and members of militias or volunteer corps that belong to a party. Then there's a subcategory in Article Four A Two. Members of other militias, volunteer corps, and organized resistance movements, even if they're not part of the regular armed forces, provided they meet four conditions.
Four conditions. Walk me through them.
One, they're commanded by a person responsible for their subordinates. Two, they have a fixed distinctive sign recognizable at a distance. Three, they carry arms openly. Four, they conduct their operations in accordance with the laws and customs of war. If a group meets all four, its fighters get POW status if captured. If they don't, they're unlawful combatants and the protections are much thinner.
Okay, so let me push on this a little. Those four conditions, they were written in nineteen forty-nine. The model they had in mind was basically the French Resistance, right? Partisan fighters operating behind enemy lines in occupied Europe. And the logic was, if you're going to fight outside the regular army, you have to at least look like a soldier and follow the rules, or you don't get the protections soldiers get.
The drafters were thinking of the Maquis in France, the partisans in Yugoslavia, groups that fought the Axis occupation. And the four conditions were a compromise. States wanted to limit who got POW status. The resistance movements wanted recognition. The four conditions were the middle ground. Wear something that marks you as a fighter, carry your weapon where people can see it, have a command structure, and follow the laws of war. Do those four things, and even if you're not in a regular uniform, you get treated as a POW if captured.
And if you don't do those four things, you're just an armed civilian, which in the logic of the Conventions means you're an unlawful combatant.
Right. And unlawful combatants don't get POW status. They can be prosecuted under domestic law for things like murder and kidnapping, even for acts that would be lawful for a soldier in uniform. The distinction matters enormously. A lawful combatant who shoots at enemy soldiers is engaged in a legitimate act of war. An unlawful combatant who does the same thing is committing a crime under the domestic law of the capturing state.
And Hamas?
Hamas is a non-state armed group. Under Article Four A Two, its fighters could theoretically qualify as POWs if they meet those four conditions. But there's a real question about whether they do. Hamas fighters don't wear a fixed distinctive sign recognizable at a distance. They don't carry arms openly in the sense the Convention means. And the operations on October seventh, the deliberate targeting of civilians, that's about as clear a violation of the laws and customs of war as you can get. So even for the fighters, POW status is debatable at best.
But that's almost beside the point for the question Daniel's asking.
It is. Because the hostages taken on October seventh were not fighters. They were civilians. Israeli civilians, foreign nationals, some dual citizens. Seized from their homes in kibbutzim, from a music festival. A grandmother. A baby. Civilians are not combatants and they cannot be prisoners of war under Article Four. The category simply doesn't apply. You can't be a prisoner of war if you were never a combatant to begin with. That's not a loophole or a technicality. That's the fundamental architecture of the Geneva Conventions. The whole system is built on the distinction between combatants and civilians. Blur that line and the entire legal framework collapses.
So the legal answer to Daniel's first question is no. The hostages were not POWs.
Not even close. And this is where the Fourth Geneva Convention comes in. GC Four protects civilians in wartime. Article Thirty-Four of GC Four explicitly prohibits hostage-taking. Flat ban. No exceptions. Taking civilians hostage is a war crime. The people Hamas seized are victims of a war crime. They are not prisoners of war. That distinction matters enormously because it determines which legal framework applies. If they were POWs, the Third Convention would govern their treatment, and that Convention actually allows quite a bit. You can interrogate POWs. You can hold them until the end of hostilities. You can put them to work in certain circumstances. But for civilian hostages under the Fourth Convention, none of that applies. Their detention is illegal from the start. Every hour they're held is a continuing war crime.
So calling them prisoners of war isn't just legally wrong, it actually gives the taking of civilian hostages a veneer of legitimacy it absolutely doesn't deserve under international law.
And that's why the legal terminology matters so much. If you call them POWs, you're implicitly accepting that their capture was a legitimate act of war. It wasn't. It was a war crime. The distinction isn't academic. It's the difference between lawful detention and ongoing atrocity.
And the visiting rights framework?
This is where it gets specific. Article One Twenty-Six of the Third Geneva Convention gives the ICRC the right to visit prisoners of war. And it's the strongest access right in international humanitarian law. The ICRC can visit POWs at any time, as often as they wish, without witnesses. The only restriction is imperative military necessity. The detaining power can't say no, can't delay, can't set conditions. It's a near-absolute right. I want to emphasize how unusual this is in international law. There are very few rights this absolute. The ICRC can literally show up at a POW camp unannounced and demand to walk in. The detaining power has to let them. If they don't, they're in clear violation of the Convention.
That's the gold standard.
The gold standard. But it only applies to POWs as defined by Article Four. For civilian detainees, the right is weaker. GC Four Article One Forty-Three grants the ICRC access to places where protected persons are held, but it's subject to imperative military necessity and comes with more restrictions. And critically, it assumes the detaining power is a state that has agreed to be bound by the Conventions. The whole machinery of the Conventions, the protecting powers, the ICRC's formal role, the reporting requirements, all of it is built on a system of states. The drafters in nineteen forty-nine were not thinking about non-state armed groups holding civilian hostages in tunnels.
So here's where I want to pause and make sure we're being precise. When people say the ICRC has a right to visit detainees, they're usually thinking of Article One Twenty-Six. That's the famous one. But that right is specific to POWs in international armed conflicts. If you're not a POW and your captor isn't a state, Article One Twenty-Six simply doesn't apply. Is that right?
And it's the single most misunderstood fact in this entire controversy. Article One Twenty-Six is a powerful tool, but it's a specialized tool. It works on one specific type of person held by one specific type of actor. POWs held by states. Outside that box, the tool doesn't fit.
Hamas isn't a state.
Hamas is not a state party to the Geneva Conventions. It is bound by Common Article Three, which applies to all parties to a non-international armed conflict, and that article prohibits hostage-taking, torture, and inhuman treatment. But Common Article Three doesn't contain the detailed visiting-rights machinery of Article One Twenty-Six. It's about a hundred and fifty words long. It says you can't murder people, you can't torture them, you can't take hostages, you have to treat the wounded and sick. That's basically it. No visiting rights. No ICRC access mandate. No inspection regime. So the ICRC's strongest legal tool, the one that lets it walk into a POW camp anywhere in the world and demand to see the prisoners, doesn't apply here. Legally, the ICRC has no Article One Twenty-Six right to visit the hostages because the hostages are not POWs and Hamas is not a state.
So legally, the criticism of the ICRC for failing to demand visiting rights is... what, misdirected?
Legally, the demand was for something the ICRC doesn't have the legal authority to compel. The organization can and did request access as a humanitarian matter. It has been pressing Hamas privately since day one. But it can't invoke Article One Twenty-Six. There's no legal lever to pull. It's like demanding that a building inspector enforce the fire code on a building that isn't in his jurisdiction. He can ask nicely. He can offer advice. He can stand outside and hope they let him in. But he can't write a citation. He has no authority to enforce anything.
But that doesn't end the story. Because the criticism of the ICRC wasn't really about legal categories. It was about what people expect the Red Cross to do.
Right. Throughout late twenty twenty-three and into twenty twenty-four, Israeli officials publicly accused the ICRC of failing to demand visiting rights, of not using its unique access to Hamas, of being too cautious. The ICRC's response was that it had been pressing Hamas privately since the day of the attack, but Hamas refused, and public demands would destroy the ICRC's ability to operate in Gaza at all. And this is where the public debate completely diverged from the legal reality. The Israeli government was making a political argument dressed up in legal language. They were saying, the ICRC should publicly demand access. The ICRC was saying, if we do that, Hamas expels us from Gaza tomorrow, and then we can't help anyone, not the hostages, not the civilians, not the wounded in the hospitals we're supporting. That's not a legal argument. That's an operational judgment call. And reasonable people can disagree about whether the ICRC made the right call. But it's not a question of law. It's a question of strategy.
And that's where the structural tension you described earlier becomes the whole story.
It is. The ICRC has maintained a continuous presence in Gaza since nineteen sixty-seven. During this war, it facilitated some hostage releases during the November twenty twenty-four ceasefire deal. It has delivered medical supplies, helped evacuate the wounded, supported hospitals. It has never been granted access to the remaining hostages. As of today, August thirty-first twenty twenty-six, Israeli estimates put around sixty hostages still alive in captivity. The ICRC still hasn't seen them.
Over two and a half years.
Over two and a half years. And the question is whether quiet diplomacy was ever going to work with a group like Hamas. The ICRC's model assumes that the party holding detainees wants to be seen as complying with international law, or at least cares about its reputation enough to make concessions. But Hamas openly violates IHL. It doesn't recognize the Geneva Conventions. It has no reputation to protect with the international community in the way a state does. The ICRC's toolkit, which is built entirely on persuasion and moral pressure, has almost nothing to grip. It's like trying to use a negotiation manual written for diplomats on someone who isn't playing the same game. The manual assumes your counterpart wants a seat at the table. What do you do when they don't even acknowledge the table exists?
So the ICRC did what it could with the tools it has, and the tools were the wrong tools for the job.
That's the structural argument. And I think it's largely right. But I also understand the frustration. From an Israeli perspective, the organization whose entire reason for existing is to visit people held in wartime couldn't get in the door for over a year. That looks like failure. The families of the hostages don't care about the legal distinction between Article One Twenty-Six and Common Article Three. They see the red cross emblem and they think, that's the organization that's supposed to see my son, my mother. And they didn't. That's not a legal failure. It's an emotional and symbolic one. But it's real, and it's devastating, and it's not something you can fix by explaining the finer points of treaty interpretation.
The gap between legal reality and public expectation is basically the whole story.
It is. And there's a historical parallel that helps here. The Iran hostage crisis, nineteen seventy-nine to nineteen eighty-one. Fifty-two American diplomats and civilians were held in the US embassy in Tehran. They were never classified as POWs. The ICRC's role was limited to humanitarian mediation. It couldn't invoke Article One Twenty-Six because the hostages weren't POWs and Iran wasn't playing by the rules. The ICRC did what it could, which wasn't much, and the hostages were held for four hundred and forty-four days.
Same dynamic.
Same dynamic. And there's another one. The ICRC's access to detainees in the US war on terror. The ICRC visited Guantanamo detainees. It had access. It produced confidential reports to the US government documenting concerns about treatment. Those reports were ignored for years. The ICRC couldn't compel changes. It could only report privately and hope. Same structural limitation, different conflict. In both cases, the ICRC had access but no enforcement power. In the Hamas case, it doesn't even have access.
So the pattern is, when a detaining power doesn't care what the ICRC thinks, the ICRC's leverage evaporates.
Completely. And Hamas is the extreme case. It's not just that they don't care. Granting the ICRC access would actively harm their interests. The hostages are Hamas's most valuable bargaining chip. Letting the ICRC see them would provide evidence of their condition, their location, who's alive and who isn't. That's information Hamas has every incentive to control. The ICRC's request for access isn't a neutral humanitarian ask from Hamas's perspective. It's an intelligence risk. And this is the part that I think the public criticism of the ICRC fundamentally misses. People imagine the ICRC knocking on a door and Hamas just being stubborn. But from Hamas's operational perspective, opening that door isn't a humanitarian gesture. It's a strategic concession. They're not being stubborn. They're being rational, in the coldest possible sense.
Even the quietest of quiet diplomacy was never going to work. The ICRC could have whispered its request through every back channel on earth and Hamas still would have said no, because saying yes would cost them their only real leverage.
That's the grim logic of it. And it raises a deeper question. Does the ICRC's neutrality model still work in asymmetric conflicts with non-state armed groups that don't recognize IHL? The ICRC's own publications acknowledge that its access model is strained in these situations. The Geneva Conventions were designed for interstate wars. The assumption was that states want to be seen as lawful. Even when they violate the rules, they usually deny it rather than embrace the violation. That gives the ICRC something to work with. You can go to a state and say, your reputation is at risk, your standing in the international community is at risk, let us in and we can help you demonstrate compliance. That argument works, imperfectly, but it works, because states care about those things.
But a group that openly violates IHL and doesn't bother denying it, there's no entry point.
None. The ICRC's founding mission assumed a world of state actors who cared about their standing. When a group like Hamas has no interest in IHL compliance and no reputation to protect, the ICRC's toolkit is nearly empty. This isn't a failure of the ICRC. It's a limit of the Geneva Conventions themselves. The Conventions are a contract among states. They bind states. They create obligations for states. They were never designed to regulate the behavior of non-state armed groups that reject the entire premise of the international legal order.
That's the thesis. The criticism of the ICRC is understandable but misdirected. The organization did what it could with the tools it has. The real question is whether those tools are adequate for twenty-first-century warfare.
That gap between legal mandate and practical expectation is something our producer Hilbert Flumingtop knows firsthand.
Hilbert: South Sudan, twenty fourteen to twenty fifteen. I was a logistics coordinator for the ICRC during the civil war. My job was arranging convoy access to opposition-held areas. Medical supplies, surgical kits, the basics. Spent months negotiating with local commanders who had no idea what the Geneva Conventions were. They'd never heard of it. They didn't care.
What was the actual negotiation like? I mean, walk us through the mechanics. You show up at a checkpoint. What happens?
Hilbert: You don't just show up. That's the first thing people get wrong. You spend weeks building relationships first. You find out who the commander is, who his brother is, who he owes money to, who he's afraid of. You drink tea. A lot of tea. You don't mention the Conventions. These guys don't care about Geneva. They care about whether their wounded fighters are going to die of infected gunshot wounds. That's your leverage. Not the law. The surgical kits.
The law is almost irrelevant to the actual negotiation.
Hilbert: The law is why we're there at all. It's the framework that says we have a mandate. But it's not what gets you through the checkpoint. What gets you through is the commander deciding it's in his interest to let you through. Six weeks. That's how long it took to get one truck of surgical kits through a checkpoint held by a commander who wanted us to pay him in cash. He wasn't being ideological. He just had a checkpoint and he wanted to get paid. The ICRC doesn't pay. So we waited. Six weeks.
And the leverage was?
Hilbert: We might leave. That's it. If we left, he'd have no one to treat his wounded fighters. That was the entire negotiation. Not the law. The law is just the reason we're there. The access is the only thing that matters. Everyone talks about Article One Twenty-Six like it's a magic key. It's not. It's a piece of paper. The paper gets you in the door if the person on the other side wants to let you in. If they don't, the paper does nothing. I've had commanders literally laugh when I mentioned the Geneva Conventions. Because they don't know what it is, and even if they did, there's no enforcement mechanism that reaches them. Who's going to arrest them? The International Criminal Court? Good luck serving that warrant in the middle of a swamp in South Sudan.
Did you ever actually leave?
Hilbert: Twice. The ICRC has an internal rule. Never threaten to leave unless you mean it. We meant it. Packed up the convoys, pulled back to Juba. Both times the fighting parties eventually called us back. Took weeks. But they called. Because they needed the surgical kits more than they needed the checkpoint revenue. That's the calculation. And it only works if the other side actually needs what you're offering.
The difference with Hamas?
Hilbert: Hamas doesn't need the ICRC. The ICRC needs access. Hamas doesn't need anything the ICRC has. They've got their own medical supplies, their own networks. And letting the ICRC see the hostages gives up information. Location, condition, who's alive. That's not leverage the ICRC can offer. That's leverage Hamas keeps by keeping the door shut. In South Sudan, both sides needed us. Here, only one side does. That's not a negotiation. That's a locked door and you don't have a key.
The model works when the detaining party has something to gain from cooperation. When they don't, the model has nothing to push with.
Hilbert: That's the job. You show up. You ask. You wait. Sometimes they say yes. Sometimes they don't. The law doesn't say yes for you. The law just tells you which door to knock on.
If the door never opens?
Hilbert: Then you stand there. That's what the ICRC has been doing in Gaza. Standing there. Asking. For two and a half years. I don't know what else people expected them to do. Kick the door down? They don't have a door-kicking department. That's not a joke. People seem to think the ICRC has some kind of enforcement arm. They don't. They have Land Cruisers and surgical kits and a mandate. That's it.
I want to sit with that image for a second. The most powerful humanitarian organization on earth, and its actual operational capacity is standing at a checkpoint and hoping the person with the gun decides to let you through. That's not a bug in the system. That is the system.
Hilbert: That is the system. And most of the time, it works. The ICRC visits hundreds of thousands of detainees every year in dozens of countries. Most of those visits happen because the detaining power sees some benefit in cooperating. But when you hit a case where there's no benefit to cooperation, the system has no answer. And that's Gaza.
If the ICRC's toolkit is this limited against non-state groups, what does that mean for the future of international humanitarian law? If the Geneva Conventions were built for a world of state armies and POW camps, and today's conflicts look more like October seventh than the Battle of Solferino, is the whole framework due for an overhaul?
The problem is, any protocol or mechanism designed to cover non-state armed groups runs into the same wall. Getting such groups to sign on is the same problem all over again. They have to want to be bound. And if they don't want to be bound, no treaty language in the world changes that. You can draft the most beautiful, comprehensive, morally unimpeachable protocol in the history of international law, and it will have exactly zero effect on a group that doesn't sign it and doesn't care that it exists.
Right. And that's the fundamental limit of law as a tool. Law works on people and entities that have opted into the system, either explicitly by signing treaties or implicitly by being part of a state that has signed them. It doesn't work on those who reject the system entirely. And we don't have a good answer for that. We never have.
The ICRC isn't broken. The system it operates within has a gap where non-state actors sit. And filling that gap isn't a legal drafting problem. It's a political reality problem. The law can't force a group like Hamas to open a door it has every incentive to keep shut. No amount of clever treaty language changes the fact that the person with the key doesn't want to use it.
That brings us back to Daniel's original question. The criticism of the ICRC is understandable but it's aimed at the wrong target. The organization did what it was built to do. Quiet diplomacy, private requests, sustained presence. The fact that those tools produced zero access for over two years isn't evidence that the ICRC failed. It's evidence that the tools themselves have limits that no amount of good faith negotiation can overcome. The ICRC didn't fail the hostages. The legal and political framework within which the ICRC operates failed to provide any mechanism that could reach them.
That's the open question Daniel's prompt leaves us with. If the ICRC's model of neutral access is structurally incapable of securing visits to hostages held by non-state groups that reject IHL, what alternative mechanisms exist? The UN? Third-party states? The answer is, none that work reliably. The ICRC is the best option we have. It's just that in some cases, the best option still doesn't get you through the door. And when that happens, we're left with the uncomfortable reality that international humanitarian law, for all its moral weight and historical significance, is ultimately a voluntary system. It works when people want it to work. When they don't, the gap between what the law promises and what it can deliver is measured in human lives.
In this case, in the lives of roughly sixty people who've been held in captivity for over a thousand days, without a single visit from the organization the world created to visit people exactly like them. That's not a failure of will. It's a failure of design. And fixing that design is a problem no one has solved.
Thanks to our producer Hilbert Flumingtop for keeping this show running. This has been My Weird Prompts. If you want to send us your own questions the way Daniel does, you can reach us at show at my weird prompts dot com. We'll be back soon.
