Collections: Why Don’t We Use Chemical Weapons Anymore?

This week, we’re going to talk briefly about why ‘we’ – and by ‘we’ here, I mean the top-tier of modern militaries – have generally eschewed the systematic or widespread use of chemical weapons after the First World War. And before you begin writing your comment, please note that the mountain of caveats that statement requires are here, just a little bit further down. Bear with me.

Now, when I was in school – this was a topic I was taught about in high school – the narrative I got was fairly clear: we didn’t use chemical weapons because after World War I the nations of the world got together and decided that chemical weapons were just too horrible and banned them, and that this was a sign of something called ‘progress.’ In essence, the narrative I got was, we had become too moral for chemical weapons and so the ‘civilized’ nations (a term sometimes still used unironically in this context) got together and enforced a moral taboo against chemical (and biological) weapons. And, we were told (this was, I should note, the late 90s and early aughts, long before the Syrian Civil War) that this taboo had mostly held.

Which was important, because in this narrative as it was impressed upon that younger version of me, the ban on chemical weapons showed the path towards banning all sorts of other terrible weapons: landmines, cluster-munitions and of course most of all, nuclear weapons. All we would need to do is for the ‘civilized’ nations of the world to summon the moral courage to abandon such brutal weapons of war. Man, the end of history was nice while it lasted! But the example of the ‘successful’ ban on chemical and biological weapons was offered as proof that the dream of a world without nuclear weapons was possible, if only we showed the same will.

When I was a child, I talked like a child, I thought like a child, I reasoned like a child. When I became a man, I put away childish things. But what was my teacher’s excuse? I guess the end of history was a hell of a drug.

More recently, an offhand statement in the Fremen Mirage series prompted a discussion of the same issue both in the comments here and also off of this website in the broader social-media-space. And I saw a few different arguments offered. I want to draw out two, because I think they represent purified versions of many of the others and so are the most useful to examine.

  • The first rejects the moral judgment of the high-school answer: we don’t use chemical weapons, it argues, because we are decadent and lack the correct, morally right Fremen ruthlessness to use this effective weapon-system in order to win.
  • The second accepts the moral judgment of the high school answer, but rejects our claim to moral advancement. It runs we only don’t use chemical weapons because we haven’t had a big war; we only didn’t use them in WWII because of MAD (the term is an anachronism, but it is also the one usually used). We aren’t as moral as we think we are and if you put us in a big, conventional war, we would use them.

And I am going to argue here that all three of these answers (including my high school answer) actually miss the point, because they all assume something fundamental: that chemical weapons are effective weapons, and so the decision not to use them is fundamentally moral, rather than practical.

Quite frankly, we don’t use chemical weapons for the same reason we don’t use war-zeppelin-bombers: they don’t work, at least within our modern tactical systems.

And before we dive in, let me remind you that if you’d like to get email updates when there is new stuff on the blog, you can subscribe using the button here:

And, while we’re here, that if you want to support the blog, I have a Patreon for it, but you can also support the blog by spreading the word, since I rely entirely on word of mouth for readership!


The Exceptions

Let’s start by addressing one of the big caveats to the chemical weapons story, which is that they have never really gone away. Now, I want to leave aside, for the purpose of this essay, the use of lethal chemical agents in genocide, the use of non-lethal chemical agents entirely, as well as the use of things like defoliants that were not intended to cause casualties (even if they did). Those things are all important, but if we get into talking about them, we will never get anywhere. Instead, we’re focusing on the battlefield use of lethal chemical agents against either opposing combatants or civilian populations.

But even within that definition, chemical weapons haven’t gone away. The single biggest burst of chemical weapons usage since WWI was by Saddam Hussein’s Iraq, especially during the Iran-Iraq war (1980-1988), but also against civilian resistance within his own borders. The other large-scale use has been in the Syrian Civil War, by the Assad regime against both military and civilian targets (but mostly civilian targets). That said, there have been a number of other more limited or suspected uses of chemical weapons over the past few decades.

So it is not the case that chemical weapons have gone away completely. Some countries continue to use them when pressed – but only a handful. But you will note that the countries that have fallen back on these weapons have something in common: they don’t really have top-flight militaries. What is going on here?

The Modern System

Now we need to start by making a crucial distinction between two sorts of armies. I think most people understand this different intuitively, but often express it clumsily. Sometimes you will see it being expressed as ‘Western’ vs. ‘Non-Western’ armies, but many of the armies that operate in ‘Western’ fashion (the JSDF, China’s PLA, the ROKA, the IDF) aren’t ‘Western’ in either a geographic or cultural sense. And, to be honest, talking about ‘Western’ or ‘Non-Western’ armies propagates some bad strategic thinking that obscures more than it clarifies. ‘Western’ is already a fuzzy cultural category (we can have that conversation another day, but I’ll note that I do not think that ‘the West’ is entirely useless as a cultural category, so long as we acknowledge it is a ‘fuzzy set‘), but military systems often cross cultural categories. Assuming that the Western cultural set must come with a Western military package is a dangerous error in strategic judgment.

So instead of that, I am going to borrow an idea from Stephen Biddle’s Military Power (2004). Biddle identifies what he calls the ‘Modern System’ of combat (though I am going to treat it a bit more broadly than he does). In short, it’s a set of tactics and operational art that emerged out of the First World War and were refined in the European theaters (East and West) of the Second, to cope with the tremendous potency of industrialized firepower which had fundamentally reshaped war. Rather than relying on fixed positions for defense and dense shock-formations (‘shock’ here – think ‘bayonets, grenades and trench-knives’), the modern system relies on cover-and-concealment for survivability and maneuver in the offense (go around, not through your opponent’s overwhelming firepower). Adroit use of terrain on the tactical level is a key component of the system, which in turn requires both extensive training of junior officers and NCOs and devolving quite a bit of command agency down to them so that they can make local decisions (compare to, for instance, linear tactics which leave virtually no decision-making to the individual rifleman).

The modern system assumes that any real opponent can develop enough firepower to both obliterate any fixed defense (like a line of trenches) or to make direct approaches futile. So armies have to focus on concealment and cover to avoid overwhelming firepower (you can’t hit what you can’t see!); since concealment only works until you do something detectable (like firing), you need to be able move to new concealed positions rapidly. If you want to attack, you need to use your own firepower to fix the enemy and then maneuver against them, rather than punching straight up the middle (punching straight up the middle, I should note, as a tactic, was actually quite successful pre-1850 or so) or trying to simply annihilate the enemy with massed firepower (like the great barrages of WWI), because your enemy will also be using cover and concealment to limit the effectiveness of your firepower (on this, note Biddle, “Afghanistan and the Future of Warfare” Foreign Affairs 82.2 (2003); Biddle notes that even quantities of firepower that approach nuclear yields delivered via massive quantities of conventional explosives were insufficient to blast entrenched infantry out of position in WWI.)

That means that modern system forces are focused on cover and concealment in defense, but on mobility – often very rapid mobility – in attack. The doctrines that developed to operationalize variations on the modern system (Bewegungskrieg, Deep Battle, AirLand Battle, and so on) all relied (intentionally or not) on pushing the tempo of an attack beyond the ability of a defender to coordinate a defense, on the theory that this would produce tactical and even operational collapse. That theory, it turns out, works very well, but it comes with some costs (I should note here I am stretching the definition of Biddle’s ‘modern system’ a bit and also glossing over a lot of detail in maneuver warfare doctrines).

Via Wikipedia, the modern system being really expensive to implement. Now, to be clear, this isn’t just a question of ‘have tanks’ – a lot of countries have tanks that can’t implement this system, because the system is doctrine + training + hardware. But the hardware sure is pricey (so is the training!).
This is the US Third Armored division in 1991.

The thing is, embracing the Modern System is hard. Actually pulling this off requires a relatively high degree of training. It also requires delegating a lot of authority down to lower officers and NCOs. You need both because individual small units need to keep moving and maneuvering even when they may not have time to get direct orders from above, and they need the freedom to respond to local conditions and utilize local terrain, often down to the squad or fireteam level. In turn, that means it is really hard to do if your common soldiers are undertrained, simply illiterate, or if (as in an authoritarian regime) you can’t trust your officers with any kind of independence. Modern authoritarian ‘coup-proofing’ makes it practically impossible to actually implement the modern system effectively (which is part of why most tin-pot dictators produce such poor military performance; though note that not all authoritarian regimes need to coup-proof in this way).

It’s also expensive. Getting the mobility to pull this off on the operational level means mechanizing almost everything in your army, so that the infantry and artillery can keep up with the advance units and so that the logistics can keep up with them. Doing that requires a lot of command sophistication, but it also just requires a ton of hardware. Everything needs to be motorized, mechanized and portable, and then all of that needs to be tied into radio communications, GPS, and so on. And because you’re delegating authority down to smaller and smaller units, those units need all of that expensive communications, and so on. All of that hardware costs a fortune.

But – and this is the core of Biddle’s argument – when modern system armies encounter armies that have not implemented the modern system in conventional pitched battle, the result is generally a crushingly one-sided affair. That’s going to matter a lot for the analysis going forward: so far, it doesn’t seem possible to hold territory (meaning not fighting as guerillas, but actually engaging in positional warfare in the Maoist sense) with a non-modern system army against a modern system army. Non-modern system armies that try get pretty badly wrecked. The 1991 Iraq war is the traditional case study in just how badly wrecked: the coalition (using the modern system) took 292 KIA; 776 WIA compared to the Iraqi Army (without the modern system) suffering 25,000+ KIA and 75,000+ WIA. That stunning lopsidedness is going to matter a lot for the argument going forward.

Via Wikipedia, what happens when a non-modern system army attempts to fight a conventional battle against a modern-system army. Here, a number of destroyed Iraqi armored vehicles, including a T-72 tank (center of frame), in March, 1991. Heavy Iraqi losses occurred during the chaotic retreat out of Kuwait as the Iraqi frontline rapidly collapsed.

Doctrine and Weapons

It’s going to be incredibly cumbersome to keep having to say ‘Non-modern system’ armies, so I’m going to start calling them Static-System armies, since – as we’ll see – in modern warfare, they tend to be a fair bit more fixed and static than the modern system armies (note: I’m going to keep calling them ‘armies’ for simplicity, but the modern system combined land and air assets), preferring to dig in for sieges and trench warfare. So again: static system (old, cheap) vs. modern system (new, expensive). And remember: this is a difference in doctrine not equipment, in how an army expects to fight their battles and how they actually do – a difference in how, not in what. It is possible to have all of the tools of the modern system, and still not have the training or will to do the modern system (indeed, Iraq did just this in 1991 and got torn apart for it). You can buy tanks and planes, but you cannot buy the modern system, you must train it.

So why contrast these two systems? Because the value of chemical weapons varies wildly between these two systems. Put bluntly: the modern system has few, if any, uses for chemical weapons. Chemical weapons still work against a static-system, but the modern system is already more effective against a static-system in ways that, as we’ll discuss in a moment, chemical weapons cannot enhance. That restricts the usefulness of chemical weapons to static-systems fighting each other or as weapons of terror, which, as it turns out, accounts for the great majority of their use (and in turn means that chemical weapons see most of their use in battles of weak armies against other weak armies).


The first reason is that it is far easier to protect against chemical munitions than against an equivalent amount of high explosives, a point made by Matthew Meselson. Let’s unpack that, because I think folks generally have an unrealistic assessment of the power of a chemical weapon attack, imagining tiny amounts to be capable of producing mass casualties. Now chemical munition agents have a wide range of lethalities and concentrations, but lets use Sarin – one of the more lethal common agents, as an example. Sarin gas is an extremely lethal agent, evaporating rapidly into the air from a liquid form. It has an LD50 (the dose at which half of humans in contact will be killed) of less than 40mg per cubic meter (over 2 minutes of exposure) for a human. Dangerous stuff – as a nerve agent, one of the more lethal chemical munitions; for comparison it is something like 30 times more lethal than mustard gas.

But let’s put that in a real-world context. Five Japanese doomsday cultists used about five liters of sarin in a terror attack on a Tokyo Subway in 1995, deployed, in this case, in a contained area, packed full to the brim with people – a potential worst-case (from our point of view; ‘best’ case from the attackers point of view) situation. But the attack killed only 12 people and injured about a thousand. Those are tragic, horrible numbers to be sure – but statistically insignificant in a battlefield situation. And no army could count on ever being given the kind of high-vulnerability environment like a subway station in an actual war.

In order to produce mass casualties in battlefield conditions, a chemical attacker has to deploy tons – and I mean that word literally – of this stuff. Chemical weapons barrages in the first World War involved thousands and tens of thousands of shells – and still didn’t produce a high fatality rate (though the deaths that did occur were terrible). But once you are talking about producing tens of thousands of tons of this stuff and distributing it to front-line combat units in the event of a war, you have introduces all sorts of other problems. One of the biggest is shelf-life: most nerve gasses (which tend to have very high lethality) are not only very expensive to produce in quantity, they have very short shelf-lives. The other option is mustard gas – cheaper, with a long shelf-life, but required in vast quantities (during WWII, when just about every power stockpiled the stuff, the stockpiles were typically in the many tens of thousands of tons range, to give a sense of how much it was thought would be required – and then think about delivering those munitions).

Via Wikipedia, a chlorine gas attack during WWI. Note how much of it is required to blanket the area. Against targets equipped with gas masks, chlorine gas is much less effective and was mostly replaced with the more lethal mustard gas, which could also blister the skin.

But then comes the other ‘problem’: protection. If you want to protect something against high explosives, your only option is armor and lots of it. Heavy, expensive armor. But if you want to protect a soldier against gas? Well, an state-of-the-art M50 gas mask costs about $270 with a set of filters. A complete NBC (Nuclear-Biological-Chemical) suit – for contact-chemicals that the mask doesn’t full protect against – costs a bit less than that (I’ve seen around $200 – and something tells me that researching this has put me on every watchlist). The soldier’s rifle probably costs around $700, for comparison. And remember: these modern system armies have all sorts of hideously expensive hardware. A unit price of c. $500-600 per soldier is cheap. And it makes your entire, very expensive multi-thousand ton arsenal of chemical munitions almost useless at a stroke.

Via Wikipedia, a pair of Canadian soldiers in NBC suits during an exercise in 1987. They’re preparing to use a Blowpipe anti-air missile to apologize some aircraft straight out of the sky.

And that’s Meselson’s point: compared to the protection available, chemical weapons don’t have the punch of traditional high explosives. Rich nations can even afford – if they expect chemical attack – to equip their entire civilian populace with these devices. And in fact, during WWII, the British did exactly that, distributing gas masks generally among the civilian populace, fearing German chemical attacks against their cities. But there’s no mask – or indeed, any sort of man-portable armor of any kind – which can stop a 1-ton high explosive bomb. So the question becomes: what is more effective for cost, a ton of mustard gas, or a ton of high explosives (speaking from artillery casualty figures in WWI, the answer is very clearly, ‘a ton of high explosives’)?

Via Wikipedia, Air Raid Warden in kit during WWII, including gas mask. This one is from Brisbane, Australia (1942).

But that’s not the only problem – the other problem is doctrine. Remember that the modern system is all about fast movement. I don’t want to get too deep into maneuver-warfare doctrine (one of these days!) but in most of its modern forms (e.g. AirLand Battle, Deep Battle, etc) it aims to avoid the stalemate of static warfare by accelerating the tempo of the battle beyond the defender’s ability to cope with, eventually (it is hoped) leading the front to decompose as command and control breaks down.

And chemical weapons are just not great for this. Active use of chemical weapons – even by your own side – poses all sorts of issues to an army that is trying to move fast and break things. This problem actually emerged back in WWI: even if your chemical attack breaks the enemy front lines, the residue of the attack is now an obstruction for you. (Compare the War on the Rocks podcast on landmines which deals with artillery-delivered-landmines on the same terms – sure I’ve prohibited the enemy, but now I have fewer maneuver options, and remember: I want go fast and break things). A modern system army, even if it is on the defensive operationally, is going to want to make a lot of tactical offensives (counterattacks, spoiling attacks). Turning the battle into a slow-moving mush of long-lasting chemical munitions (like mustard gas!) is counterproductive.

But that leaves the fast-dispersing nerve agents, like sarin. Which are very expensive, hard to store, hard to provision in quantity and – oh yes – still less effective than high explosives when facing another expensive, modern system army, which is likely to be very well protected against such munitions (for instance, most modern armored vehicles are designed to be functionally immune to chemical munitions assuming they are buttoned up).

This impression is borne out by the history of chemical weapons; for top-tier armies, just over a century of being a solution in search of a problem . The stalemate of WWI produced a frantic search for solutions – far from being stupidly complacent (as is often the pop-history version of WWI), many commanders were desperately searching for something, anything to break the bloody stalemate and restore mobility. We tend to remember the successful innovations – armor, infiltration tactics, airpower – because they shape subsequent warfare. But at the time, there were a host of efforts: highly planned bite-and-hold assaults, drawn out brutal et continu efforts, dirigibles, mining and sapping, ultra-massive artillery barrages (trying a wide variety of shell-types and weights). And, of course, gas. Gas sits in the second category: one more innovation which failed to break the trench stalemate. In the end, even in WWI, it wasn’t any more effective than an equivalent amount of high explosives (as the relative casualty figures attest). Tanks and infiltration tactics – that is to say, the modern systemsucceeded where gas failed,in breaking the trench stalemate, with its superiority at the role demonstrated vividly in WWII.

During WWII, everyone seems to have expected the use of chemical weapons, but never actually found a situation where doing so was advantageous. This is often phrased in terms of fears of escalation (this usually comes packaged with the idea of MAD (Mutually Assured Destruction), but that’s an anachronism – while Bernard Brodie is sniffing around the ideas of what would become MAD as early as ’46, MAD itself only emerges after ’62). Retaliation was certainly a concern, but I think it is hard to argue that the combatants in WWII hadn’t already been pushed to the limits of their escalation capability, in a war where the first terror bombing happened on the first day. German death-squads were in the initial invasion-waves in both Poland, as were Soviet death squads in their invasion of Poland in concert with the Germans and also later in the war. WWII was an existential war, all of the states involved knew it by 1941 (if not earlier), and they all escalated to the peak of their ability from the start; I find it hard to believe that, had they thought it was really a war winner, any of the powers in the war would have refrained from using chemical weapons. The British feared escalation to a degree (but also thought that chemical weapons use would squander valuable support in occupied France), but I struggle to imagine that, with the Nazis at the very gates of Moscow, Stalin was moved either by escalation concerns or the moral compass he so clearly lacked at every other moment of his life.

Both Cold War superpowers stockpiled chemical weapons, but seem to have retained considerable ambivalence about their use. In the United States, chemical weapons seem to have been primarily viewed not as part of tactical doctrine, but as a smaller step on a nuclear deterrence ladder (the idea being that the ability to retaliate in smaller but still dramatic steps to deter more dramatic escalations; the idea of an ‘escalation ladder’ belongs to Herman Kahn); chemical weapons weren’t a tactical option but baby-steps on the road to tactical and then strategic nuclear devices (as an aside, I find the idea that ‘tactical’ WMDs – nuclear or chemical – could somehow be used without triggering escalation to strategic use deeply misguided). At the same time, there was quite a bit of active research for a weapon-system that had an uncertain place in the doctrine – an effort to find a use for a weapon-system the United States already had, which never quite seems to have succeeded. The ambivalence seems to have been resolved decisively in 1969 when Nixon simply took chemical weapons off of the table with an open ‘no first use’ policy.

Looking at Soviet doctrine is harder (both because I don’t read Russian and also, quite frankly because the current epidemic makes it hard for me to get German and English language resources on the topic) The USSR was more strongly interested in chemical weapons throughout the Cold War than the United States (note that while the linked article presents US intelligence on Soviet doctrine as uncomplicated, the actual intelligence was ambivalent – with the CIA and Army intelligence generally downgrading expectations of chemical use by the USSR, especially by the 1980s). The USSR does seem to have doctrine imagine their use at the tactical and operational level (specifically as stop-gap measures for when tactical nuclear weapons weren’t available – you’d use chemical weapons on targets when you ran out of tactical nuclear weapons), but then, that had been true in WWII but when push came to shove, the chemical munitions weren’t used. The Soviets appear to have used chemical weapons as a terror weapon in Afghanistan, but that was hardly a use against a peer modern system force. But it seems that, as the Cold War wound down, planners in the USSR came around to the same basic idea as American thinkers, with the role of chemical weapons – even as more and more effective chemicals were developed – being progressively downgraded before the program was abandoned altogether.

This certainly wasn’t because the USSR of the 1980s thought that a confrontation with NATO was less likely – the Able Archer exercise in 1983 could be argued to represent the absolute peak of Cold War tensions, rivaled only by the Cuban Missile Crisis. So this steady move away from chemical warfare wasn’t out of pacifism or utopianism; it stands to reason that it was instead motivated by a calculation as to the (limited) effectiveness of such weapons.

And I think it is worth noting that this sort of cycle – an effort to find a use for an existing weapon – is fairly common in modern military development. You can see similar efforts in the development of tactical nuclear weapons: developmental dead-ends like Davy Crockett or nuclear artillery. But the conclusion that was reached was not ‘chemical weapons are morally terrible’ but rather ‘chemical weapons offer no real advantage.’ In essence, the two big powers of the Cold War (and, as a side note, also the lesser components of the Warsaw Pact and NATO) spent the whole Cold War looking for an effective way to use chemical weapons against each other, and seem to have – by the end – concluded on the balance that there wasn’t one. Either conventional weapons get the job done, or you escalate to nuclear systems.

(Israel, as an aside, seems to have gone through this process in microcosm. Threatened by neighbors with active chemical weapons programs, the Israelis seem to have developed their own, but have never found a battlefield use for them, despite having been in no less than three conventional, existential wars (meaning the very existence of the state was threatened – the sort of war where moral qualms mean relatively little) since 1948.)

And I want to stress this point: it isn’t that chemical munitions do nothing, but rather they are less effective than an equivalent amount of conventional, high explosive munitions (or, at levels of extreme escalation, tactical and strategic nuclear weapons). This isn’t a value question, but a value-against-replacement question – why maintain, issue, store, and shoot expensive chemical munitions if cheap, easier to store, easier to manufacture high explosive munitions are both more obtainable and also better? When you add the geopolitical and morale impact on top of that – you sacrifice diplomatic capital using such weapons and potentially demoralize your own soldiers, who don’t want to see themselves as delivering inhumane weapons – it’s pretty clear why would wouldn’t bother. Nevertheless, the moral calculus isn’t the dominant factor: battlefield efficacy – or the relative lack thereof – is.

Ok, – you say – so throwing chemical weapons at a fancy modern-system army is going to be less effective than just using high explosives. But what if we introduce static-system armies?

System Static

The situation changes significantly when older-style static-system armies fight each other.

Let’s start by remembering that the static system isn’t some coequal way of doing modern ground warfare, so much as the inevitable fallback when, for some reason, you can’t do modern-system warfare. There are, in practice, really just about three reasons why that might be: first, the army might not have access to the fancy, expensive equipment to supply the mobility the modern system demands; second, the army might not have the training to execute the modern system; third, the front-line officers in the army might not have the command independence to execute the modern system (if this last one seems nonsensical – many authoritarian regimes are hesitant to give junior officers a lot of independence, since junior officers tend to be festering-grounds for coups in such systems).

And those factors feed directly into vulnerability to chemical munitions. While modern system armies are so damned expensive that effective chemical protection is a rounding error on the budget, for armies struggling to get enough rifles and bullets to provide even basic firepower, chemical protection (or modern body armor, for that matter) is an unaffordable luxury. Likewise, armies with weak organization, training and discipline will find chemical preparedness – which involves a lot of training on how to get those gas masks and NBC suits on fast – very difficult; actually getting all of the fidgety equipment to the right spots will also prove hard (but is second-nature to a modern system military which has nothing but fidgety equipment).

And, so, where do we still see chemical weapons used? In static-system vs. static-system warfare. Thus, in Syria – where the Syrian Civil War has been waged as a series of starve-or-surrender urban sieges, a hallmark of static vs. static fighting – you see significant use of chemical weapons, especially as a terror tactic against besieged civilians. The limited manpower and capabilities of regime forces have caused the war to deteriorate into a series of sieges, sometimes stretching out years (fighting in Aleppo lasted for four years, for instance; the final siege itself ran from February 2014 to its conclusion in December 2016). Anti-regime forces are often poorly equipped (often completely unable, for instance, to engage regime air-assets) and the civilian populace was completely unprotected against chemical munitions, making them far more vulnerable targets.

But a major factor here is actually weakness, in the Syrian regime forces. Assad simply didn’t have a lot of modern air-to-ground munitions; chemical munitions weren’t being compared for cost- and mission-effectiveness against such modern weapons, but against barrels loaded with explosives, nails and scrap – weapons which would have been primitive by the standards of the 1940s, much less now. And – let’s be honest here – his ground forces lack manpower, but also perform quite poorly. Remember: the question for the effectiveness of chemical weapons is value-over-replacement – while the vulnerability of anti-regime forces increased the value, we also must note that Assad’s heavily weakened, static system forces also substantially reduced the value of the replacement. In a fight between what are, in the last analysis, two weak forces, the calculation on the effectiveness of chemical weapons changes.

(I want to pause here. I am being a bit bloodless talking about this, because I want my point to be clear. But just so we understand each other: using barrel bombs against civilians is monstrous. Using poison gas at all is monstrous. Assad is a monster. Also, I don’t want to make it out that the international response doesn’t matter – one assumes that Assad gambled that while using chemical weapons would prompt international outrage, it wouldn’t isolate him diplomatically anymore than he already was (a gamble that, unfortunately, seems to have paid off). A state in the same position but more reliant on international support which might be jeopardized by such a breach of international norms might make a very different calculation).

Likewise in the Iran-Iraq war (1980-1988), when Iraqi efforts to execute a maneuver war fell apart, the Iraqi army broke out the gas to try to stop Iranian human-wave assaults and later to open up areas to attack. Once again, the use of chemical weapons here was a consequence of weakness. The Iranian military was woefully underprepared for hostilities, with Iranian units reportedly sometimes not even having enough rifles for all of the soldiers (much less artillery, anti-tank or anti-helicopter support – to try to fight without that in a modern conventional battle is to go to the fight naked), a significant number of whom were children. And while Iraq could buy modern weapons, they could not buy modern military competence (a point Biddle makes in the above-cited book quite well). That the Iraqi army was not up to the task of waging modern war was made stunningly apparent in 1991.

Via Wikipedia, an Iranian soldier wearing a gas mask during the Iran-Iraq War.

(While those two are the largest incidents of chemical warfare on the battlefield after WWI, they’re not the only ones – poison gas was used in the North Yemen Civil War (1962-1970), probably by Cuban troops during the Cuban intervention in Angola, and possibly by Soviet forces in Afghanistan as a terror weapon against civilians. These weapons never really went away, they just dropped out of use by most modern system armies.)

But if chemical weapons can still be effective against static system armies, why don’t modern system armies (generally) use chemical weapons against them? Because they don’t need to. Experience has tended to show that static system armies are already so vulnerable to the conventional capability of top-flight modern system armies that chemical munitions offer no benefits beyond what precision-guided munitions (PGMs), rapid maneuver (something the Iraqi army showed a profound inability to cope with in both 1991 and 2003), and the tactics (down to the small unit) of the modern system do. Modern system armies don’t need to use gas to stall out human wave attacks, because they can simply deliver overwhelming conventional firepower. Likewise, they don’t need gas as a terror weapon to win long sieges – contrast the Second Battle of Fallujah (admittedly, a smaller urban center) with the long urban sieges of the Syrian Civil War; American, Iraqi and British forces cleared Fallujah in less than two months, while inflicting casualties at a rate of 10-to-1 on a defending force (of course, inflicting casualties alone doesn’t win wars, but then chemical weapons can’t secure gains, win hearts and minds, etc. either).


Throughout all of this, the contrast between the effort to ban chemical munitions and similar efforts to ban other ‘morally objectionable’ weapon-systems is instructive. Take cluster munitions, which are opposed generally because the bomblets they release can remain in the ground, unexploded for long periods and thus injure or kill civilians years after the end of a conflict. The Convention on Cluster Munitions is a limited effort to ban these weapons, but 12 years on from its implementation, most of the world’s largest militaries remain aloof (It’s actually worse than the wikipedia map makes it look – keep in mind that all of the ratifying countries that are in NATO or otherwise have a US-backed security guarantee can simply count on American cluster-munitions to fill battlefield roles in the event of a conflict, which is a great way to be able to do some moral grandstanding while still not actually standing on any actual principles, per se).

Via Wikipedia, a map of the countries in the CCM, banning cluster munitions.

The effort to ban landmines (the Ottawa Treaty) is slightly better, but it has the same key gaps: the United States, Russia, China, India, Iran, Egypt, Israel, Pakistan, North and South Korea and so on (Turkey, interestingly is in Ottawa, but not the CCM; I wonder why). For countries whose conventional military capabilities are, in the age of the Pax Americana, largely ornamental (or, again, are part of a larger military alliance which can cover for their capabilities), signing these treaties is easy. But the moment you look at states that expect to actually fight someone in a conventional war in the near future, signature rates collapse.

And, Via Wikipedia, a map of all of the countries in the Ottawa Treaty, banning many types of landmines. Note that it does not cover anti-tank mines, remote-controlled mines, or even anti-personnel anti-handling devices on your anti-tank mine (so an anti-tank mine that is designed to blow up a person trying to pick it up is still not banned). It is, as such, a pretty limited treaty. And yet – look at all those grey non-signatories.

And, of course you have the nuclear weapons states – the USA, France, Britain, Russia, India, Pakistan, China, North Korea and Israel (sorry Israelis, this blog is a no-ambiguity zone). You may notice some overlap between these groups, with the exception of the two smaller NATO-nuclear powers (Britain and France).

But those same countries that won’t sign paper on these issues not only aren’t using chemical weapons on the battlefield (and many of them are involved in at least one active conflict or have been in just the lack decade, so it’s not for lack of opportunity), they are actively dismantling production and storage. The USA has destroyed about 90% of its stockpile and is moving towards complete removal (expected by 2023). Russia claims to have completed its destruction, and although it still sometimes uses chemical agents in espionage activities, there’s no sign of anything like preparedness for battlefield use. That includes dismantling or converting production facilities. Now – are some of these countries probably holding back small, hidden stockpiles? Sure. But that’s not enough for battlefield use. And I really want to stress that – these are militaries who will not make even token commitments to getting rid of other ‘immoral’ weapons, but will actively remove their ability to rapidly deploy these weapons in the future.

If the day comes when one of these countries suddenly is in a war and realizes they need chemical weapons, they won’t be available. Production would have to start fresh (look into the 1915 ‘shell crisis‘ to get a sense of what a disaster it is in industrial warfare to only realize you need large amounts of munitions after hostilities have begun). They are dismantling these systems because they do not expect they will ever have a use for these weapons.

Not because chemical weapons are immoral. Because they are ineffective, whereas landmines, cluster munitions and nuclear weapons are assumed to have future uses. Morality was never at issue, effectiveness was.

That has some big consequences, most of which are actually unpleasant. It means that efforts to ban ‘immoral’ weapons that are effective will likely never succeed. The model that activists thought they were building on is a foundation of sand – the only way you ban weapons is after they are basically useless anyway (to cut that point finer: you get to ban weapons after they stop being useful to the great powers, who may then enforce the ban on smaller states). The idea that we might summon up enough moral indignation to set these weapons aside seems unfounded. We are not nearly so moral as my high school teachers imagined, nor are we so cowardly as a ‘Fremen apologist’ might think. We’ve simply made an amoral calculation about effectiveness.

It also means that, should some development make chemical weapons effective compared to conventional high explosives or nuclear weapons – some agent that can defeat NBC protection, for instance – then chemical warfare will come screaming back. Now, the reassuring thing is that it looks like none of the premier military research systems in the world think that is likely (or they’d be investing more money into getting there first – look at the tremendous resources thrown into getting railguns, lasers, nuclear-powered missiles and the like on to the battlefield, even when it is not clear that they will be effective). But these terrible weapons haven’t been put to bed forever, only for as long as they remain relatively ineffective.

As someone who thinks humans must learn to peace as well as we have learned to war if we wish to survive in the long run, it is a humbling and concerning thought that we have not come so far as we might like.

136 thoughts on “Collections: Why Don’t We Use Chemical Weapons Anymore?

  1. “Adroit use of terrain on the tactical level is a key component of the system, which in turn requires both extensive training of junior officers and NCOs and devolving quite a bit of command agency down to them so that they can make local decisions […].”

    I believe you just described why the Russian military is failing in Ukraine. It would seem that no amount of hardware + doctrine can overcome poor training and total lack of trust in junior officers, ot to mention a startling lack of NCOs.

  2. I’m curious if you have any response to the idea that chemical weapons would have been effective against Germany in WWII because Germany didn’t have enough gas masks for their horses and were dependent upon horse-drawn logistics. I don’t recall where I encountered this idea or if it’s accurate but I’ve seen it cited as a reason Germany didn’t initiate the use of chemical weapons.

    1. That only matters if you can kill horses better with gas warfare than with high explosives. They presumably also didn’t have enough armour to protect horses on the march against bombing or even strafing runs either.
      A big point of this essay is that usually, you can kill more with HE.

  3. I think the either/or distinction, morality or utility, isn’t quite appropriate. Both can contribute to the eventual result.

    If I walk give my representative a treaty to ban nerf guns in armed conflict, that treaty is never going to get signed, and not because anybody has an elite nerf team. By the same token, without the popular will behind it, the US never would have signed the chemical weapons treaty, even if they had no plans to use cw themselves.

    Throughout this, you speak of the political costs involved in the use of cw. What are those costs? They are the costs of being populalrly considered less moral than the other guy.

    If will is sufficient against certain useful weapons, then treaties will get signed– just as attrition was not a politically viable strategy in WWI, landmines may not be a politically viable strategy in WWIII. It is not utility or popularity alone that determine this; it is the combination of the two. (Not really the two, I’d suspect some other factors play into it as well.)

    1. While your point is good, it seems apparent that landmines are absolutely still considered valid in modern warfare from a political standpoint. Ukraine has been using landmines extensively (most recently and notably at Vuhledar), but there has been no outcry that I’ve heard, just resigned comments about how difficult and dangerous the clean-up will eventually be.

  4. This means a good way to get rid of particular inhumane weapons — or tactics — is to design a technological counter which makes them ineffective. This is often far more possible than people realize…

  5. A deep irony I’ve sometimes wondered if is even possible is that we might see a scenario where a ‘counterforce’ nuclear exchange happens, and the end result is a nuclear war that while it hurts all belligerents extremely badly, also leaves both of them intact enough to continue a conventional war for at least awhile, because the primary targets of both sides’ nuclear stockpiles was the other sides’ nuclear stockpile.

Leave a Reply