Daniel's been thinking about a word that gets thrown around constantly in debates about Israel's military operations — proportionality. Everyone uses it, almost nobody defines it, and the gap between what people think it means and what it actually means under international law is... well, it's basically the whole argument. He asked us to unpack what proportionality really is in the narrow legal sense under the law of armed conflict, how that diverges from the colloquial use of the word, and why it keeps getting weaponized specifically against Israel.
And this is one of those terms where the legal definition is so far from the plain-English understanding that you can watch two people argue past each other for an hour and never realize they're using completely different tests. So let's start with what the law actually says, because it's probably not what you think.
It's almost certainly not what you think.
The core text is Additional Protocol One to the Geneva Conventions, Article fifty-one, paragraph five, subparagraph b. It prohibits — and I'm quoting here — an attack which may be 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.
Every word in that sentence is doing work. "May be expected" — that's future-looking. "Excessive in relation to" — that's a balancing test, not a cap. "Concrete and direct military advantage" — that's the thing you're weighing against, not the number of enemy fighters you've killed.
Exactly right. Three elements to pull apart here. First, it's an ex-ante test — judged from the commander's perspective before the strike, based on the information available at the time. Not with hindsight, not after the smoke clears and we know what actually happened. Second, the comparison is between anticipated civilian harm and anticipated military advantage — not between civilian deaths and enemy combatant deaths. That's the most common mistake right there. Third, the word is "excessive," not "any." The law permits civilian casualties. It has to, or you couldn't fight a war in a city at all.
So if you're a commander looking at a target, you're asking: given what I know right now, is the civilian harm I expect to cause out of proportion to the military value of hitting this thing? And if the answer is no — if it's a lot of civilians but the target is critical — the strike can be legal.
The ICRC's own FAQ on the Gaza conflict, which they published in twenty twenty-three, confirms exactly this. Proportionality is not a body-count ratio. It's a balancing test applied to each individual attack. And the ICRC has explicitly stated — I'm quoting again — that proportionality cannot be reduced to a numerical formula. There is no ratio in the treaty text. One-to-one, one-to-two, one-to-ten — none of those numbers appear anywhere in the law.
Which is where the whole thing starts to come apart in public debate. Most people hear "proportional" and think... well, they think if you punch me, I get to punch you back about as hard. Tit for tat. Symmetric retaliation. That's the language of criminal sentencing or self-defense, not the law of armed conflict.
And that frame — the "response must be proportionate to the attack" frame — that's what dominates news coverage and political statements. When someone says Israel's response to October seventh is disproportionate, they're making a claim about the overall scale of the military campaign relative to the initial attack. That's a jus ad bellum question — the law governing whether you go to war in the first place. It has nothing to do with the proportionality rule we just described, which is jus in bello — the law governing how you fight once the war is underway.
Two completely different legal buckets. And the proportionality rule in jus in bello doesn't apply to the war as a whole — it applies to each individual strike. You don't ask "is the Gaza campaign proportionate to October seventh?" You ask "was this specific airstrike on this specific building proportionate to the military advantage of destroying that building?"
The Lawfare analysis of the Israel-Hamas war, which ran through twenty twenty-three and into twenty twenty-four, hammers this point. Commanders assess proportionality based on "the information reasonably available to them at the time." Not perfect information. Not what we learned later. So a strike that looks disproportionate after the fact — because it turns out there were more civilians in the basement than anyone knew — may have been perfectly legal if the commander had reasonable but imperfect intelligence.
And that's a hard thing for people to sit with. The law permits outcomes that feel morally wrong to observers. Civilian casualties are tragic even when they're legally proportionate. The Just Security coverage of the Gaza conflict makes this point repeatedly — critics often conflate "disproportionate" with "regrettable."
They're not the same thing. Regrettable is... a lot of what war produces. Disproportionate is a specific legal finding that the commander got the balance wrong based on what he knew at the time.
Let's make this concrete. The ICRC's own hypothetical: a commander targets a weapons cache in a civilian area and anticipates ten civilian casualties. If the military advantage is destroying a single rocket launcher sitting in storage, that might be excessive — ten dead civilians for one rocket launcher that wasn't about to be used. But if that same launcher is about to fire on a school, the military advantage is enormous, and ten civilian casualties may be proportionate. Same number of dead civilians. Different legal outcome. The variable that changed wasn't the body count — it was the military advantage.
That hypothetical does more work than any ten op-eds. It shows you that proportionality isn't a counting exercise — it's a weighing exercise. And the weight on the military advantage side changes depending on context. How imminent is the threat? How important is the target? What happens if you don't strike?
So that's what the law actually says. Now let's talk about why almost nobody uses it that way.
The gap between the legal test and the colloquial understanding isn't just an academic curiosity — it's the engine that drives a huge amount of the public debate about Israel's military operations. And Israel is the most consistent target of this confusion for a few reasons.
One of them is that Israel's own military doctrine makes the confusion worse in a weird way. Roof-knocking — the practice of dropping a small munition on a roof as a warning before the real strike — and the use of precision-guided munitions mean Israel actually produces lower civilian casualty ratios than many other conflicts. But the legal test doesn't care about ratios. It cares about excessiveness relative to military advantage. So critics point to absolute civilian death tolls without ever assessing the military advantage of each strike.
And the military advantage side of the equation is where the public debate just... stops. Nobody wants to do it. Partly because the information is often classified — you can't assess the military advantage of striking a particular building if you don't know what was inside it. And partly because it feels callous to say "well, that hospital was also a command center, so the military advantage was high."
But that's exactly what the law requires you to weigh. And Hamas's tactic of embedding military assets in civilian infrastructure — hospitals, schools, residential buildings — deliberately complicates these assessments. The military advantage of striking a command center in a hospital may be very high. The anticipated civilian harm is also very high. The law says the commander has to weigh both. The public debate skips the first half entirely.
Lawfare's analysis notes this explicitly. When a military target is placed in or near a civilian object, the civilian object doesn't automatically become a legitimate target — but the proportionality calculus shifts because the military advantage of striking it may be very high. The law doesn't let you ignore the civilians. It also doesn't let you ignore the military objective.
And this is where the weaponization really kicks in. Activists and states — Ireland comes to mind, and we've talked about their legal activism before — use the language of proportionality to argue that Israel's campaign is disproportionate in the colloquial sense. Too many civilians relative to the initial attack. That's a legally meaningless frame, but it's politically devastating.
Because it sounds legal. It uses the right word. And the person hearing it thinks "oh, they're applying international law" when what they're actually doing is applying a completely different standard and borrowing the law's vocabulary to give it weight.
The conflation works in two directions. You've got jus ad bellum proportionality — is the overall war justified relative to the provocation? And you've got jus in bello proportionality — was this specific strike legal? Critics slide between them depending on which one makes Israel look worse in the moment.
And even within jus in bello, the test is inherently subjective. The Just Security archive shows that reasonable legal experts disagree on specific applications all the time. Is twenty civilian casualties excessive for destroying a tunnel network? Depends on the tunnel network. Depends on what's moving through it. Depends on what you knew when you struck. The law provides a framework, not a calculator.
Which makes it very easy to weaponize. If there's no bright line, you can always find a legal expert who'll say a given strike was disproportionate. And the fact that another legal expert disagrees doesn't make headlines.
There's a knock-on effect here that I think is actually the most damaging part of all this. When the legal term is consistently misused to mean something else, people draw one of two conclusions. Either they conclude the law itself is meaningless — because it permits outcomes that look like disproportionality to them. Or they conclude that Israel is uniquely lawless — because it's accused of disproportionality far more often than other states fighting similar wars.
Both conclusions are wrong. But they're politically potent. And they feed each other. The more people hear "Israel's campaign is disproportionate," the more they assume Israel must be violating international law. The fact that the accusation is using the wrong legal test never enters the conversation.
And this isn't just sloppy journalism — though there's plenty of that. Some of it is deliberate. The legal definition is inconvenient for the argument people want to make. If you say "Israel's strikes must comply with the proportionality rule, and here's what that actually means," you have to then do the work of assessing individual strikes based on the information available to commanders at the time. That's hard. It requires information nobody has. Saying "look at all the dead civilians, clearly disproportionate" is easy.
The ICRC's rocket launcher hypothetical makes this painfully clear. You can't look at a body count and know whether a strike was proportionate. You have to know what the target was, what the commander knew, and what the military advantage was. None of that is visible in a casualty figure.
And this dynamic isn't unique to Israel — the same thing played out in Kosovo, in Libya, in Ukraine. But Israel is the most consistent target because the conflict is the longest-running, the most legally scrutinized, and the one where the political incentives to use the colloquial meaning are strongest.
There's also a weird asymmetry in how the test gets applied in public debate. When the US strikes an ISIS target and civilians die, the conversation — when it happens at all — tends to accept that the military advantage was real and the question is whether the civilian harm was excessive. When Israel strikes a Hamas target, the military advantage half of the equation often just... vanishes. As if the target wasn't there.
Part of that is Hamas's information strategy. They don't acknowledge military assets in civilian areas. So the public sees a destroyed residential building and assumes it was just a residential building. The IDF might say it was a command center, but by then the image of the rubble is already the story.
And the proportionality test doesn't care about the image. It cares about what the commander knew when he authorized the strike. If he had reasonable intelligence that the building contained a command center, the military advantage side of the scale has real weight — even if Hamas denies it, even if journalists can't verify it afterward.
This is what makes retrospective proportionality judgments so unreliable. We're applying the test with information the commander didn't have — and without information the commander did have. It's not just hindsight bias. It's fundamentally the wrong inquiry.
So where does this leave us? The legal test is narrow, context-dependent, and applied ex-ante to individual strikes. The public debate uses the word to mean something completely different. And the gap between them is not an accident — it's useful.
Pulling on that thread a bit more: the Just Security coverage notes that even within legal circles, there's a debate about whether the proportionality rule is too permissive. Some scholars argue that "excessive" should be interpreted more strictly in urban warfare because the civilian consequences are so severe. Others argue that the test already accounts for that — the anticipated civilian harm in a dense urban area is higher, so the military advantage needs to be correspondingly higher to justify the strike.
Which is exactly the kind of debate the law is designed to accommodate. Reasonable people weighing the same factors and reaching different conclusions. That's not a bug — it's the whole point of a balancing test. But it also means the law can't settle the public argument in the way people want it to.
People want the law to say "this many dead civilians is too many." The law refuses to say that. It says "it depends." And "it depends" is a terrible soundbite.
It's a worse slogan.
Hilbert: Forty-seven times.
...Go on.
Hilbert: I spent three months in two thousand seven cataloging every instance of the word proportionality in UN Security Council resolutions about the Middle East. Two thousand to two thousand ten. Forty-seven resolutions used the term. Every single one used it in the colloquial sense — response proportionate to provocation. Not one used the IHL definition. Not one.
You were working for...
Hilbert: Small NGO. We submitted amicus briefs to the ICC. My job was to build the evidentiary record showing that even the Security Council's own lawyers were using the wrong definition in formal resolutions.
Forty-seven out of forty-seven.
Hilbert: The thing that drove me crazy — and I mean actually crazy, I had to take a week off — was that the lawyers drafting those resolutions knew the legal definition. They'd been to law school. They'd read the protocols. They just knew the colloquial version was more politically useful. It's not that people don't understand the law. It's that the law is inconvenient, so everyone pretends the colloquial meaning is the legal one.
And the Security Council resolutions are where the rest of the international system takes its cues.
Hilbert: That was our argument. If the Council itself can't use the term correctly, the entire downstream debate is poisoned. You get human rights reports citing Council resolutions as authority, and the wrong definition just... propagates.
Did the brief go anywhere?
Hilbert: Never cited by any ICC prosecutor. One of my colleagues later found a footnote in a twenty fifteen UN Human Rights Council report that paraphrased our argument without attribution. So the idea got through — just not in a way that changed anything.
That's... grim.
Hilbert: The report's conclusion was that Israel should be investigated for disproportionate attacks. Used the colloquial definition. The footnote basically said "some have argued that the legal definition is narrower than the one employed here" and then moved on.
So the institution that's supposed to uphold the law acknowledged that it was using the wrong definition and kept going.
Hilbert: I had the footnote framed. It's in a box somewhere. My point is: this isn't ignorance. It's not even really confusion, most of the time. It's that the legal definition produces outcomes that are politically inconvenient for one side, and the colloquial definition produces outcomes that are politically inconvenient for the other side, and the side that benefits from the colloquial definition has more votes in the General Assembly.
And it's not just Israel. You said this played out elsewhere.
Hilbert: Kosovo, Libya, Ukraine. Any conflict where one side can claim to be the underdog, the proportionality argument gets deployed the same way. Israel just draws it more consistently because the conflict never ends and the legal scrutiny never stops. There are entire careers built on litigating Israel's proportionality. You don't see that for, say, Turkey's operations in northern Syria.
Because the political incentive structure is different.
Hilbert: The resolutions pass. That's what matters. You get more of what you reward.
The institutional machinery has a stake in maintaining the confusion.
Hilbert: I'm not saying it's a conspiracy. I'm saying it's a habit that nobody wants to break because breaking it would make their jobs harder. If you're a human rights lawyer and you have to start every report by explaining that proportionality doesn't mean what everyone thinks it means, you lose half your audience before you get to the actual argument.
If you don't explain it, you're using the word in a way you know is legally inaccurate.
Hilbert: Correct. And most of them choose the second option. I would too, probably. It's easier.
That's... I was going to say depressing, but it's more like structural. The gap between legal meaning and political utility is built into the system.
Hilbert: The gap is the system. The law says one thing, the politics needs it to say another, and the institutions that are supposed to bridge that gap are the ones doing the most to widen it. I don't have a solution. I just had to count them.
Forty-seven.
Hilbert: I still remember the spreadsheet. Column A was the resolution number, column B was the paragraph, column C was the quote, column D was which definition they were using. Column D was the same word in every row.
What word?
Hilbert: Wrong.
If you take one thing from this whole discussion, it's that the next time someone accuses Israel — or any military — of a disproportionate attack, you now know what to listen for. Are they talking about the legal test or the colloquial one? They're almost never the same thing.
The giveaway is usually whether they mention military advantage at all. If the argument is just "look at the civilian death toll" with no assessment of what was being targeted and why, they're not applying the proportionality rule. They're applying a different standard and borrowing the law's vocabulary.
Which leaves me with a question I don't know the answer to. If the legal definition is this narrow and this context-dependent, is it even possible to have a productive public debate about civilian casualties in war without everyone learning IHL first? Or is the gap just too wide to bridge?
I think the gap is structural, and I think Hilbert's spreadsheet proves it. When the Security Council can't use the term correctly across forty-seven resolutions, the problem isn't that people haven't read the treaty. The problem is that the correct usage is politically useless for the arguments people want to make.
There's a future dimension to this that the law hasn't caught up with yet. The IDF has acknowledged using AI-assisted targeting systems. If a commander has near-perfect intelligence — if the system is telling them exactly how many civilians are in the building and exactly what's in the basement — does the proportionality test become stricter? Because the "information reasonably available" standard shifts when the available information is vastly better.
That's a open question. The law was written for a world where commanders had fog of war, incomplete intelligence, best guesses. If AI reduces the fog, the "reasonably available" standard might demand more of commanders. Or it might not — because the standard has always been about what's reasonable to expect a human to process in real time, not what's technically knowable.
Either way, the gap between the legal test and the public debate isn't closing. If anything, better intelligence makes the gap wider — because the information commanders have gets better while the information the public has stays the same.
That's probably where we should leave it. The law has a definition. The definition is narrow and specific and doesn't do what most people think it does. Knowing that doesn't settle any arguments, but it does tell you which arguments are serious and which ones are just using legal-sounding words.
This has been My Weird Prompts. Thanks to our producer Hilbert Flumintop for... the spreadsheet, and the footnote, and the framed reminder that institutions can know they're wrong and keep going anyway.
If this episode gave you a lens for evaluating the next proportionality argument you hear — and you will hear one — tell someone about the show. Or don't, but at least you'll know what to listen for. We're at myweirdprompts dot com.
We'll be back soon.
