Alright, today’s prompt from Daniel is looking at the technical and humanitarian history of cluster munitions, which is a heavy one, but incredibly relevant given what we’re seeing in the news right now. Specifically, that shimmering curtain effect over Tel Aviv that everyone’s been posting videos of. It looks like a slow-motion firework display, but the reality on the ground is anything but celebratory.
It’s a terrifying visual, Corn. And as we’re diving into this, it’s worth noting that today’s episode is actually powered by Google Gemini Three Flash. But back to the topic—that shimmer is the literal signature of a cluster munition mid-deployment. When you see those hundreds of tiny sparks drifting down, you’re looking at an area-saturation weapon designed to blanket an entire grid square with explosives.
It’s interesting because we usually think of modern warfare as moving toward extreme precision—you know, the "missile through a window" kind of stuff. But cluster munitions are the exact opposite of that. They’re basically a bucket of grenades delivered by a rocket. Daniel wants us to look at the Convention on Cluster Munitions, why it exists, and why, frankly, it seems to be failing to stop what Iran is doing right now in March twenty-six.
To understand why the treaty exists, you have to understand the engineering. A cluster munition is effectively a "parent" canister. It could be an artillery shell, a gravity bomb, or a missile warhead. Inside that canister are dozens or even hundreds of "children"—submunitions or bomblets. At a pre-set altitude, the parent casing bursts open, and centrifugal force or a small charge scatters these bomblets over a massive area, often the size of several football fields.
Wait, how exactly does the timing work on that? If it opens too high, the wind blows them into the next county, right? But if it opens too low, they don't spread out enough.
It’s a delicate piece of ballistic choreography. Most modern canisters use a barometric fuze or a GPS-timed electronic fuze. At a specific height—say, 500 to 1,000 feet—the canister splits. Some use a "spin-stabilized" method where the canister rotates at high speed, and when the casing shears off, the bomblets are flung outward like water off a spinning tire. That’s what creates that perfect circular or oval "footprint" of destruction on the ground.
Right, so instead of one big hole in the ground, you get a thousand small ones. If you’re a military commander trying to clear a runway or stop a column of tanks, that sounds efficient. But if that rocket is flying over a city like Tel Aviv, those thousand small holes are happening in apartment buildings, parks, and streets.
That’s the core of the humanitarian crisis. There are two main problems. First, the "indiscriminate" nature. Because these things scatter, you physically cannot aim an individual bomblet. If you fire a cluster-tipped missile at a military headquarters in a city, you are guaranteed to hit the civilian bakery three blocks away. It is mathematically impossible to contain the effects.
And the second problem is the one that lasts for decades—the duds.
The unexploded ordnance, or UXO. This is where the engineering tradeoffs become lethal. These bomblets are often cheaply made because you're producing them by the millions. They use simple impact fuzes. If a bomblet lands in soft mud, or hits a tree branch, or just has a manufacturing defect, it doesn't explode. It just sits there. It becomes a de facto landmine.
I was looking at some of the data on failure rates. Depending on the model and the terrain, we’re talking anywhere from five percent to thirty percent. That’s a staggering amount of live explosives just waiting for someone to step on them years after the war ends.
It really is. Think about the math: if a single Iranian missile carries three hundred submunitions and has a twenty percent failure rate, that’s sixty unexploded mines dropped into a neighborhood from a single shot. Now multiply that by the hundreds of missiles we’ve seen in these recent waves. You’re literally seeding a city with thousands of tiny, volatile traps.
Let's talk about that volatility for a second. If a bomblet didn't explode when it hit the ground at terminal velocity, why would it explode later just because a kid picks it up?
That’s the most dangerous misconception. These fuzes are often "armed" by the spin of the fall or the impact itself. Sometimes the firing pin is resting right against the primer, held back by nothing but a rusted spring or a bit of grit. A slight change in temperature, the vibration of a passing truck, or someone tilting it just a few degrees can be enough to complete the sequence. They don't get "safer" with age; they get more unpredictable.
You mentioned the history, and looking back, this isn't a new problem. We saw this in Vietnam, specifically during Operation Commando Hunt in the late sixties and early seventies. The US dropped something like fifteen million cluster bombs on Laos.
Laos is actually the most heavily bombed country in the world per capita because of that. To this day, people are still being killed or maimed there. We’re talking over twenty thousand casualties since the conflict ended. Farmers hitting them with plows, children picking them up because they look like little metal balls or toys. It’s a multi-generational catastrophe.
I read a case study once about the "Balu" or "ball" bombs in Laos. Because they are round and about the size of a tennis ball, children often find them while playing and try to use them in games. It’s heartbreaking because the design—which was intended for aerodynamic efficiency—inadvertently mimics a toy.
It’s a tragic design flaw that persists in almost every iteration of the weapon. Even the modern ones often have bright colors or interesting shapes. And it didn't stop in Southeast Asia. In the ninety-one Gulf War, the coalition used over sixty thousand cluster munitions. Estimates suggest they left behind about one point six million unexploded submunitions in the desert. Even in a relatively "clean" environment like a desert, that makes post-war reconstruction and movement incredibly dangerous.
I remember hearing stories from Gulf War veterans about "the steel rain." That was the nickname for the M26 rockets launched from the MLRS systems. They said it sounded like a constant hail on a tin roof.
The M26 is a perfect example of the scale we're talking about. Each rocket carried 644 submunitions. A full "ripple" of twelve rockets would saturate a huge area with nearly 8,000 bomblets in less than a minute. During the push into Kuwait, the sheer volume of these things meant that even US troops had to be careful moving through areas they had just bombarded. There were reports of US soldiers being killed by their own unexploded duds during the advance.
It’s the persistence that got the international community moving. The real turning point, though, wasn't Vietnam or Iraq—it was the two thousand six war between Israel and Hezbollah. During the final days of that conflict, there was a massive exchange of cluster munitions. The sheer density of UXO left in Southern Lebanon was the final straw for many humanitarian groups. It led directly to the Oslo Process, which gave us the Convention on Cluster Munitions in two thousand eight.
The situation in 2006 was unique because of the timeframe. It’s estimated that Israel dropped the vast majority of its cluster munitions in the final 72 hours of the war, just as a ceasefire was being negotiated. This left Southern Lebanon littered with up to four million submunitions. When civilians tried to return to their homes the following week, they walked into a massive, invisible minefield. The international outcry was so loud that it bypassed the usual UN stalemate and created the independent Oslo Process.
So the CCM, as it’s called, basically says: no using them, no making them, no selling them, and you have to destroy your stockpiles. It sounds like a total success on paper. Over one hundred countries signed on. But there’s a massive asterisk here, isn't there, Herman?
A huge one. If you look at the map of who signed, it’s mostly countries that don't plan on fighting a high-intensity symmetric war anytime soon. The major military powers—the ones with the biggest stockpiles and the most advanced missile programs—stayed out. We’re talking the United States, Russia, China, India, Pakistan, and, notably, Iran and Israel.
So, essentially, the countries that make and use the most cluster bombs are the ones who refused to agree to stop. That feels like a bit of a hollow victory for the treaty.
It depends on how you define victory. The CCM was successful in stigmatizing the weapon. It made it politically expensive to use them. For example, even though the US isn't a signatory, they haven't used them in a major way since the 2003 invasion of Iraq, partly because of the diplomatic headache it causes with their allies who are party to the treaty. But you’re right—in a "gloves-off" conflict like what we're seeing now with Iran, that stigma carries very little weight.
It’s the "Interoperability Loophole" that really gets me. Even for the countries that did sign, like the UK or France, there’s a clause that says they can still engage in military operations with non-signatories who use cluster bombs. So, if the US uses them in a joint operation, a treaty member can basically look the other way while providing logistics or intel. It’s a bit of a "have your cake and eat it too" policy.
It’s a pragmatic compromise that allowed the treaty to exist at all, but it certainly weakens the moral authority of the ban. Now, why did the US or Russia refuse to sign? From their perspective, these aren't just "evil" weapons; they are "force multipliers." If you’re facing a massive armored division, one cluster bomb can do the work of fifty precision missiles. It’s a cost-benefit analysis where military utility outweighs the projected humanitarian cost in their strategic doctrine.
Can you walk me through that math? Why is one cluster bomb better than fifty precision ones? Is it just about the cost of the missile?
It’s partly cost, but mostly physics. If you have a column of fifty tanks moving through a valley, a precision missile can hit one tank. To hit all fifty, you need fifty missiles, fifty target acquisitions, and a lot of time. A cluster munition, however, covers the entire valley floor in armor-piercing "shaped charges." It doesn't need to be precise because it denies the entire area to the enemy. For a military facing a numerically superior foe, that’s an incredibly tempting tool to keep in the shed.
But that calculus changes when you’re talking about a country like Iran using them as a primary terror weapon against civilians. What we’re seeing now in twenty-six isn't "force multiplication" against tanks; it’s area saturation of urban centers.
This is a fundamental shift in doctrine. Historically, Iran focused on precision—the Fateh-one-ten missiles were all about hitting specific hangars or buildings. But as Israeli missile defenses like Iron Dome and David’s Sling got better at intercepting single warheads, Iran pivoted. By using cluster warheads, they’re trying to overwhelm the defense. Even if the interceptor hits the parent missile, if it’s already started the deployment sequence, you still have hundreds of fragments and bomblets falling.
Wait, so even if the Iron Dome "wins" and hits the missile, the city still gets showered?
That’s the nightmare scenario. If the interception happens at a lower altitude, the kinetic energy of the interceptor might actually help disperse the submunitions. Instead of a controlled deployment, you get a chaotic "debris field" of live explosives. The defense system essentially turns one big problem into three hundred small ones. This is why that "shimmering curtain" is so deceptive. It looks like the defenses are working—and they are, in the sense that they prevented a massive explosion at a single point—but they can't stop the gravity-fed descent of the submunitions.
That’s the "shimmering curtain." It’s a way to bypass defense through sheer volume. And because Iran isn't a signatory to the CCM, they don't feel any legal pressure to stop. In fact, reports suggest about half of the submunitions being found in Israel right now are duds—which, as we discussed, is actually worse for the long-term safety of the population.
Well, I mean, that's the tragedy of it. The high failure rate might be intentional. If you’re trying to make a city uninhabitable, leaving thousands of "sleeping" mines in the rubble is a very effective, if horrific, way to do it. The Open Source Munitions Portal has been authenticating photos of these submunitions—many of them look like small cylinders with ribbons on top. The ribbons are meant to stabilize them as they fall, but they also make them look like something a child might want to play with.
I've seen those photos—the "ribbon" is often a bright nylon tail. It looks like a streamer from a bicycle handle or a party decoration.
It’s called a "drag ribbon." Its job is to pull the bomblet upright so the shaped charge points downward when it hits. But after it lands, that ribbon stays attached. If it’s caught in a tree or sticking out of a pile of bricks, it’s an invitation for someone to pull on it. And because these things are often the size of a soda can, they are incredibly easy to overlook until you're standing right on top of them.
It’s interesting to look at the US position here, too. The US hasn't signed the CCM, but they did implement a policy years ago saying they would only use cluster munitions with a failure rate of less than one percent. But then they pushed that deadline back, and eventually, the Trump administration rolled it back further, arguing that "smart" cluster bombs with self-destruct timers were too expensive or not reliable enough yet.
It’s the classic tension between "ideal" tech and "available" tech. The US maintains a stockpile of about six hundred million submunitions. Many of those are decades old. If you use forty-year-old munitions, that one percent failure rate is a fantasy. You’re looking at twenty or thirty percent, just like the Iranian stuff we’re seeing now.
Why is the self-destruct technology so hard to get right? We have computers in our watches now; surely we can put a timer on a grenade?
You’d think so, but it’s a harsh environment. These things are sitting in storage for thirty years in extreme heat and cold. Then they are launched out of a cannon or dropped from a jet, experiencing thousands of G-forces. A tiny battery or a delicate circuit board often fails under those conditions. If the self-destruct mechanism fails, you're right back to having a "dumb" dud. There’s also the cost factor—adding a reliable "brain" to each of the 600 bomblets in a single missile makes the weapon ten times more expensive.
So, if the treaty hasn't stopped the major players, and it hasn't stopped Iran from raining them down on Tel Aviv, what was the point? Did the CCM actually achieve anything?
I think it shifted the "norm." Even if the US or Israel hasn't signed, they are much more cautious about using them because of the global stigma. It created a massive hurdle for the production and sale of these weapons. Most major Western banks and pension funds won't invest in companies that make cluster munitions now. It’s a "financial de-platforming" of a weapon system.
That’s a good point. It’s harder to build a business around something that a hundred-plus countries have labeled a war crime. But Iran, operating largely outside that global financial system, doesn't care. They’ve built their own domestic supply chain for these warheads.
And that’s the scary part of the "shimmering curtain." It represents a failure of international law to keep up with rogue state actors who are willing to ignore the stigma. When you see those sparks over a city, you’re seeing the physical limit of a treaty. Legal papers don't stop physics, and they don't stop a country that has decided that civilian terror is a valid tactical goal.
It makes me wonder about the role of groups like the Mines Advisory Group or the HALO Trust. They’re the ones who have to go in after the "shimmer" fades and actually pick these things up. We talked about the Israeli sappers in a previous episode—the sheer scale of the task they’re facing right now in the center of the country is unprecedented.
It’s a job that never ends. In Laos, they’re still clearing fields fifty years later. In Israel, they’re having to do it in the middle of an active conflict. Every time a new wave comes in, the "minefield" is reset. It’s a constant state of environmental contamination.
Think about the psychological toll on the sappers, too. It’s one thing to clear a minefield where you know the boundaries. It’s another to have to treat an entire metropolitan area as a potential live-fire zone. Every playground, every flat roof, every gutter could be hiding a dud.
And the clearance process for cluster munitions is actually more dangerous than for traditional landmines. With a landmine, you can generally use a metal detector and a probe because they are buried in a predictable way. Cluster bomblets can be anywhere. They can be hanging in a power line, lodged in a balcony railing, or sitting inside a child’s backpack that was left outside. You can't use "flails" or heavy machinery in a city street without destroying the infrastructure you're trying to save. It’s slow, painstaking hand-work with a high risk of "sympathetic detonation"—where one goes off and triggers the others nearby.
So, looking at the practical side for our listeners—what do we do with this info? It’s easy to feel helpless when you see those videos.
One thing is to understand the technical reality so you can spot the misinformation. There’s a lot of talk about "phosphorus" or "incendiaries" whenever people see those sparks. Usually, it’s not that. It’s just the pyrotechnic charges and friction of submunitions. Knowing it’s a cluster event tells you something specific: don’t go near the impact site, even if it looks like nothing exploded.
Right. The "clean" street might be the most dangerous one. And I think supporting the organizations that do UXO clearance is one of the few tangible ways to help. These groups are going to be busy in the Middle East for the next several decades regardless of when the shooting stops.
And keep an eye on the policy side. There’s always a push to tighten the CCM, to close that interoperability loophole, or to pressure non-signatories through trade and diplomacy. It’s a slow burn, but it’s the only path toward actually making these weapons obsolete.
I saw a report recently about "liquidators"—volunteers in conflict zones who try to clear these things with no training. That seems like a recipe for disaster.
It is. We saw this in Ukraine and we're seeing it now in Israel. People think they can just "kick it into a ditch" or pick it up and throw it in the trash. You cannot emphasize this enough: these things are designed to kill. They are more sensitive than a landmine and more powerful than a hand grenade. If you aren't a professional with a blast suit and a remote detonator, you shouldn't be within fifty feet of one.
It’s a grim topic, but a necessary one. The history of cluster munitions is basically a history of us trying, and often failing, to put the genie back in the bottle. Once you invent a way to kill everything in a five-acre radius for pennies on the dollar, it’s very hard to convince everyone to stop using it.
It really is. The engineering is simple, the cost is low, and the effect is devastating. That’s a triple threat that’s hard to beat with just moral persuasion. But the fact that we even have a treaty with over a hundred countries is a testament to how much the world reacted to the horrors of Lebanon and Vietnam. It shows that there is a collective conscience, even if it’s currently being outpaced by the missiles.
Do you think we’ll ever see a "smart" alternative that actually works? Like, a cluster bomb that truly, 100% of the time, turns into a harmless piece of plastic after ten minutes?
The technology exists, but the "will" to implement it is the bottleneck. There are "Sensor Fuzed Weapons" like the CBU-97, which use infrared sensors to find targets and have multiple redundant self-destruct modes. If they don't find a tank, they go inert. But each one of those bombs costs nearly half a million dollars. Compare that to a "dumb" cluster bomb that costs a few thousand. Until the cost of the "clean" version drops, or the legal cost of the "dirty" version becomes too high, the old-school stuff will stay in the silos.
Well, I hope that "shimmering curtain" is something we see less of, though the current trajectory suggests otherwise. We’ll have to see if the technology of defense can eventually outpace the "dumb" volume of cluster saturation.
I'm not optimistic on the short term, but the research into "high-power microwave" defenses might be the answer. Instead of hitting one bomblet with one missile, you fry the electronics of the whole swarm at once. But that’s a deep dive for another day.
I’m sure you’ve already got three papers printed out on that, Herman. I can see your eyes lighting up at the thought of microwave pulses. Anyway, that’s a wrap on the "shimmering curtain" and the complicated world of cluster munitions. Big thanks to our producer, Hilbert Flumingtop, for keeping us on track.
And thanks to Modal for providing the GPU credits that power this show. It’s what keeps our "shimmering" AI brains running.
This has been My Weird Prompts. If you’re enjoying these deep dives, maybe leave us a review on your favorite podcast app. It really does help people find us.
We’ll be back with another prompt soon. Until then, stay curious.
And watch out for anything with a ribbon on it. See ya.
