Daniel sent us this one — and it comes out of a conversation we were having earlier today about conventional and unconventional weapons, and the international controls around their use. He got curious about something specific: the fact that there are international databases whose whole purpose is to provide a standardized taxonomy of weapon systems and munitions. So his question is basically, what are the main standardizations and taxonomies that have been developed, do they cover things like missiles, ballistic missiles, chemical weapons, and what practical purpose do they actually serve for the people who are concerned with their use? It's a good question, because it touches on something that sits right at the intersection of policy, intelligence, and international law — but almost nobody talks about it.
It's one of those things where, once you know these taxonomies exist, you start seeing them everywhere. They're the invisible scaffolding under almost every arms control agreement, every UN report, every sanction designation. Without them, you literally cannot have a conversation about what weapon was used where, because nobody agrees on what to call it.
Which is the kind of problem that sounds absurd until you remember that even something as seemingly obvious as "cluster munition" — we spent part of this morning debating where the line is.
And that debate is exactly why these systems got built. So let me start with the big one, the one that most people in the field encounter first. It's called the Munitions Reference System, or MRS. It's maintained by the United Nations Office for Disarmament Affairs, and it is essentially the Dewey Decimal System for things that go boom.
The Dewey Decimal System for things that go boom.
I'm not even exaggerating. The MRS assigns a unique alphanumeric code to every category of conventional ammunition — and I mean every category. Small arms ammunition, mortar rounds, artillery shells, rockets, grenades, demolition charges, pyrotechnics. It covers conventional munitions specifically, so it doesn't directly handle chemical or biological or nuclear, but it does cover the delivery systems that could carry those payloads.
You'd have a code for, say, a one hundred twenty millimeter mortar round, and that code is the same whether you're looking at a report from the African Union or NATO or the Russian Ministry of Defense?
That's the goal. And it's more granular than you'd think. The code structure actually captures the caliber, the type of munition, the fuze type, the explosive fill, and even the packing configuration. So you're not just saying "mortar round" — you're specifying whether it's high-explosive, smoke, illumination, practice. Whether it's point-detonating or proximity-fuzed. And critically for the question that prompted this, it has specific codes for submunitions — the individual bomblets inside cluster munitions.
Which means if Iran is documented using cluster munitions, as the prompt mentions, the MRS gives you a standardized way to say exactly which cluster munition. Not "some kind of cluster bomb" but a specific code that links to a specific system.
And that matters enormously for attribution, for sanctions enforcement, and for after-action clearance. Because if you're a demining organization and someone hands you a report that says "unexploded submunitions found in residential area," you need to know which submunition. Different types have different failure rates, different fuzing mechanisms, different render-safe procedures. The MRS code tells you that in a format that's identical whether the report came from the ICRC or the Ukrainian military or a UN investigation team.
That's conventional munitions. What about the bigger stuff — missiles, ballistic missiles, the systems that carry those munitions?
That's a different but overlapping taxonomy, and this is where it gets interesting. For missile systems, the dominant standard is something called the MTCR Equipment and Technology Annex. MTCR stands for Missile Technology Control Regime. It's not a UN body — it's a voluntary association of thirty-five countries that agree to restrict the export of missiles and missile technology capable of delivering weapons of mass destruction.
They needed a taxonomy because you can't restrict export of something if you can't define what that something is.
And the definitions get incredibly specific. The MTCR Annex divides everything into two categories. Category One covers complete rocket and unmanned air vehicle systems capable of delivering a five hundred kilogram payload to a range of at least three hundred kilometers — that's the famous "three hundred kilometer / five hundred kilogram" threshold that you see in every news article about Iranian or North Korean missile programs.
Which is a threshold that was deliberately chosen, right? It's not arbitrary.
Not arbitrary at all. It was calibrated to capture systems that could deliver a first-generation nuclear weapon. A five hundred kilogram payload is roughly the weight of an early implosion-type nuclear device. And three hundred kilometers is enough range to strike strategic targets across borders in most regional conflict scenarios. The MTCR was founded in nineteen eighty-seven, so they were thinking very specifically about the kinds of nuclear threats that existed at the time.
Category Two covers everything else — components, propulsion systems, guidance sets, re-entry vehicles, and also complete missiles that don't cross the Category One threshold. The export controls are less stringent for Category Two, but they're still there. And the taxonomy is detailed enough that if you're exporting a gyroscope that could be used in a missile guidance system, the Annex tells you exactly which subcategory it falls under and what the licensing requirements are.
The MTCR taxonomy is built for export control. It's a regulatory tool, not primarily a monitoring or reporting tool.
And that's an important distinction. Different taxonomies serve different purposes, and they slice the world differently as a result. The MRS is built for ammunition reporting and stockpile management. The MTCR Annex is built for technology denial. And then there's a third one that's built for arms trade transparency, which is the UN Register of Conventional Arms.
Which is the one where countries voluntarily report their arms imports and exports.
The UN Register has its own set of seven categories of major conventional weapons. Battle tanks, armored combat vehicles, large-caliber artillery systems, combat aircraft, attack helicopters, warships, and missiles and missile launchers. And within each category, there are sub-definitions. For example, a battle tank is defined as a tracked or wheeled self-propelled armored fighting vehicle with high cross-country mobility and a high level of self-protection, weighing at least sixteen and a half metric tons unladen, with a main gun of at least seventy-five millimeters.
Which is a definition that probably excludes some things that a layperson would call a tank, and includes some things that might surprise them.
It absolutely does. And those boundary disputes are where the taxonomy does its real work, because once you have a definition, you can have a debate about whether something meets it. Without the definition, you're just arguing past each other.
We've got MRS for ammunition, MTCR for missile technology, UN Register for major conventional arms. What about chemical weapons? The prompt specifically asked about those.
Chemical weapons have their own dedicated taxonomy under the Chemical Weapons Convention, which is administered by the Organisation for the Prohibition of Chemical Weapons — the OPCW. And this one is fascinating because it's structured around what are called "Schedules.
Schedules of chemicals.
Schedule One chemicals are things that have been developed or used as chemical weapons and have little or no legitimate industrial use. Sarin, VX, mustard gas, ricin. These are subject to the most stringent verification measures — you basically can't produce them outside of very small quantities for research or protective purposes, and every gram is accounted for.
Schedule Two covers chemicals that are precursors to chemical weapons or that have some limited legitimate use. Things like thiodiglycol, which is a precursor to mustard gas but is also used in inks and dyes. These are subject to export controls and declaration requirements but aren't as tightly restricted.
Schedule Three is the interesting one. These are dual-use industrial chemicals that can be used to make chemical weapons but are also produced in massive quantities for legitimate purposes. Phosgene, hydrogen cyanide, chlorine. The verification regime for Schedule Three is lighter, but production facilities still have to be declared and are subject to inspection.
The CWC Schedules are essentially a risk-tiering system.
And it's a taxonomy that has to balance two competing demands. On one hand, you want to catch everything that could be weaponized. On the other hand, you don't want to strangle legitimate chemical industries that employ millions of people and produce everything from pharmaceuticals to plastics. The Schedule system is the compromise.
Which brings us to the practical question — the prompt asked what purpose these taxonomies actually serve for people concerned with weapons use. And I think the answer has layers.
It absolutely does. Let me go layer by layer. The first and most obvious is monitoring and verification. If you're a UN panel of experts investigating whether an arms embargo has been violated, you need to be able to look at a photograph of a weapons cache and identify exactly what you're seeing. The standardized taxonomies mean that a French investigator and a Kenyan investigator and a Russian investigator are all using the same language.
If they're not, you get exactly the kind of disputes we've seen in actual conflict investigations — where one party says "that's not a ballistic missile, it's a guided rocket" and the distinction matters for whether an agreement was violated.
And the taxonomy doesn't eliminate those disputes, but it provides a framework for adjudicating them. You can point to the definition and say, "Does this system meet these specific technical criteria or doesn't it?" That's a lot more productive than arguing about what the word "missile" means.
Second layer is stockpile management and security. Governments use these taxonomies internally to track their own ammunition inventories. NATO has an entire system called the NATO Codification System — the NCS — which assigns a unique NATO Stock Number to every item in the supply chain. If you're a logistics officer trying to make sure your ammunition depots are properly accounted for, you're using these codes constantly. And that has a direct connection to proliferation concerns, because poorly managed stockpiles are a major source of weapons for non-state actors.
The taxonomy is a tool for preventing diversion before it happens, not just for investigating it afterward.
And the third layer is clearance and risk education. When a conflict ends and you have unexploded ordnance contaminating farmland and villages, the humanitarian demining organizations need to know exactly what they're dealing with. The MRS codes feed directly into something called the Collaborative ORDnance Data Repository — CORD — which is maintained by the Geneva International Centre for Humanitarian Demining.
And it's essentially a technical encyclopedia of every munition that might be encountered in a post-conflict environment. Photographs, diagrams, fuze details, explosive contents, recommended render-safe procedures. All indexed by the standard taxonomies. So a deminer in Laos who finds an unfamiliar submunition can look it up in CORD and know exactly how to handle it.
Which literally saves lives. That's not a rhetorical point — improper handling of unexploded ordnance kills deminers every year.
And the taxonomy is what makes that knowledge transferable across borders and across organizations. Without it, every demining group would have to build its own reference library from scratch, and they wouldn't be able to share information efficiently.
There's a fourth layer I want to pull out, which is legal accountability. When you're building a war crimes case or a sanctions designation, the precision of the taxonomy matters enormously. You can't just say "they used cluster munitions." You need to say exactly which cluster munition, when, where, and ideally trace it back to a specific manufacturer or supplier. The taxonomy gives you the vocabulary for that.
It connects to the evidentiary standards that courts require. The International Criminal Court isn't going to accept "some kind of rocket" as evidence. They want a specific designation that can be cross-referenced against export records, end-user certificates, and chain-of-custody documentation. The taxonomy is what makes that cross-referencing possible.
We've got MRS, MTCR Annex, UN Register, CWC Schedules, NATO NCS, CORD. Are there others worth mentioning?
There's the Wassenaar Arrangement Munitions List, which is used by forty-two participating states to control exports of conventional arms and dual-use goods. It has its own categorization system that covers everything from smooth-bore weapons to electronic warfare systems. There's the International Ammunition Technical Guidelines — the IATG — which is a UN framework for safe ammunition management and uses the MRS taxonomy as its backbone.
Then there's the one that intelligence agencies use.
Which we know exists but the details are largely classified. The US Department of Defense maintains something called the DoD Ammunition and Explosives Hazard Classification Procedures, which includes a detailed taxonomy for hazard classification and compatibility grouping. And the intelligence community has its own internal designators for foreign weapons systems — you'll see things like "SA-21" or "SS-N-27" in open-source reporting, which are NATO reporting names that map onto a much more detailed classified taxonomy.
The NATO reporting names are their own interesting subsystem. They're not just random — the initial letter tells you the type. S for surface-to-air, A for air-to-air, K for air-to-surface, and so on.
They're deliberately designed to be easy to remember and hard to confuse. A "SA-10" and an "SA-11" are different systems, but the naming convention makes it immediately clear they're both surface-to-air missiles. That's a taxonomy designed for operational use — you need pilots and intelligence officers to communicate quickly and unambiguously under stress.
Which is a different design goal than the CWC Schedules, which are designed for legal precision and diplomatic negotiation.
That's the thing I find genuinely fascinating about this whole space. All of these taxonomies are doing the same basic thing — naming and categorizing weapons — but they're optimized for completely different use cases, and those optimizations shape what they capture and what they miss.
Give me an example of something that falls through the cracks between taxonomies.
Armed drones are a perfect example. When the MTCR was written in nineteen eighty-seven, armed drones as we know them today didn't exist. So there's been a years-long debate about whether and how the MTCR Annex applies to systems like the Bayraktar TB2 or the Shahed-136. The Annex covers "unmanned air vehicle systems" in Category One, but the definition was written with cruise missiles in mind, not loitering munitions that can orbit a target area for hours and then dive into it.
You have a weapon system that's clearly militarily significant, but the taxonomy was built for a different era and doesn't quite know what to do with it.
And different countries have different interpretations, which creates exactly the kind of ambiguity that the taxonomy was supposed to eliminate. The United States takes the position that the MTCR does cover armed drones. Some other countries disagree. And in the gap between those interpretations, you get export decisions that might not be what the framers of the MTCR would have intended.
Another gap — what about cyber weapons? None of these taxonomies cover them.
No, and that's a huge open question. The Tallinn Manual process — which is an academic effort to apply international law to cyber operations — has tried to develop some taxonomies for cyber capabilities, but there's nothing with the institutional weight of the MRS or the CWC Schedules. And the fundamental challenge is that a cyber weapon isn't a physical object you can photograph and measure. It's code, it's ephemeral, it can be deleted or modified, and its effects depend on the target system in ways that a bullet's effects don't.
The taxonomy problem is even harder for cyber, because the thing you're trying to classify doesn't have stable physical properties.
And that's one of the reasons cyber arms control is so much harder than nuclear or chemical arms control. With a nuclear warhead, you can count them, you can measure the yield, you can verify the fissile material. With a cyber capability, what are you even counting? Lines of code? None of those map cleanly onto the kind of taxonomies we've built for physical weapons.
Let me pull on a thread you mentioned earlier — the NATO Stock Number system. How granular does that actually get?
An NSN is a thirteen-digit code that identifies a specific item of supply. And I mean specific. There isn't just an NSN for "one hundred twenty millimeter mortar round." There are different NSNs for the high-explosive round, the smoke round, the illumination round, the practice round. And within high-explosive, there are different NSNs for different fuzing configurations. And then there are separate NSNs for the fuze itself, the propellant charge, the packaging container, and the technical manual.
If you're a logistics officer and you need to order replacement fuzes for a specific type of mortar round, you're not writing a prose description. You're giving a thirteen-digit number.
That number is the same across all thirty NATO member states. A Canadian supply sergeant and a Turkish supply sergeant are using the exact same code. That interoperability is one of the quiet enablers of coalition warfare. Without it, you'd have ammunition incompatibilities, mis-shipments, and in the worst case, troops at the front receiving the wrong munitions.
Which has happened historically.
There's a famous example from the First World War where French and British artillery units couldn't share ammunition even when they were fighting side by side, because their shell dimensions were different by a few millimeters. The NATO Codification System was built specifically to prevent that from ever happening again.
The taxonomy isn't just a bureaucratic exercise — it has real operational consequences.
Real economic consequences. The NCS is also the backbone of NATO's logistics marketplace. When a manufacturer wants to sell something to a NATO military, they have to get an NSN assigned. That process involves submitting detailed technical specifications, which get reviewed and codified. It's a barrier to entry, but it also creates a transparent market where buyers can compare exactly what they're getting.
Let's go back to the prompt's specific scenario for a moment — the documentation of cluster munition use. If you're an investigator on the ground and you find remnants of a cluster munition, what's the actual workflow for identifying it through these taxonomies?
The first step is physical documentation — photographs with scale markers, GPS coordinates, serial numbers if visible, any markings on the munition body or fuze. Those get sent to a technical analysis team, which compares the physical characteristics against reference databases. The MRS code is usually the first level of identification — it tells you the category and type. Then you cross-reference against manufacturer databases to identify the specific model and likely country of origin.
There are organizations that specialize in this.
Conflict Armament Research — CAR — is one of the best known. They do field documentation in active conflict zones and trace weapons back through supply chains. They've documented Iranian-made munitions in Yemen, Syrian weapons in Iraq, North Korean ammunition in Africa. Their reports are incredibly detailed, and they rely heavily on the standardized taxonomies to make their findings legible to international audiences.
Admissible in legal proceedings.
CAR's documentation has been used in UN panels of experts reports, in sanctions designation packages, and in some cases as evidence in national courts prosecuting arms trafficking. The taxonomy is what gives it evidentiary weight — you're not saying "we think this is an Iranian mortar round," you're saying "this item matches the technical specifications of the HM-16 mortar round as catalogued in the MRS under code such-and-such, and the lot numbers are consistent with production at the such-and-such facility.
That shift from "we think" to "this matches" is the whole game.
And it's worth noting that the taxonomies themselves are constantly being updated. The MRS is revised regularly to add new munition types and retire obsolete ones. The MTCR Annex has been amended multiple times to keep pace with technological developments — though, as we discussed, it's often playing catch-up. The CWC Schedules have a formal amendment process that requires consensus among states parties.
Which means the taxonomies are living documents, not static reference works.
They're political documents, whether we like it or not. Every decision about what to include, what to exclude, where to draw a boundary, is a political decision with real-world consequences. When the CWC negotiators decided to put a particular chemical on Schedule Two rather than Schedule One, they were making a judgment about its legitimate industrial uses versus its weaponization potential. That judgment affects which facilities get inspected, which exports get blocked, which countries face scrutiny.
Those judgments can be contested. I'm thinking of the debates around riot control agents — tear gas — and whether they should be considered chemical weapons for the purposes of the CWC.
The CWC explicitly allows the use of riot control agents for law enforcement purposes, but prohibits their use as a method of warfare. But what counts as "law enforcement" versus "warfare"? If you use tear gas in an occupied territory, is that law enforcement or is it warfare? Different countries have different interpretations, and the taxonomy doesn't fully resolve that.
Because the taxonomy can tell you what a substance is chemically, but it can't tell you what the context of its use means legally.
And that's a limitation that applies across all of these systems. The taxonomy is a tool for clarity, not a substitute for judgment. It can tell you that a particular munition is a cluster munition as defined by the Convention on Cluster Munitions. It can't tell you whether its use in a specific circumstance constitutes a violation of international humanitarian law. That's a separate analysis that requires legal expertise and contextual knowledge.
To answer the prompt's question directly — yes, there are robust international taxonomies covering conventional munitions, missiles, ballistic missiles, and chemical weapons. They serve monitoring, verification, stockpile management, export control, humanitarian clearance, and legal accountability functions. They're detailed, they're continuously updated, and they're politically contested at the margins.
The prompt mentioned non-proliferation and monitoring as the driving concern. I'd add that the taxonomies are most effective when they're embedded in a broader institutional framework. The MRS on its own is just a reference book. It becomes powerful when it's used by UN investigator teams, by demining organizations, by sanctions committees, by national export control agencies. The taxonomy enables coordination across all of those actors.
One thing we haven't touched on — the prompt mentions conventional and unconventional weapons. The taxonomies we've discussed cover conventional and chemical. What about biological weapons?
The Biological Weapons Convention — the BWC — is the odd one out here, because it doesn't have a detailed taxonomy in the way the CWC does. The BWC prohibits the development, production, and stockpiling of biological weapons, but it lacks a verification mechanism and it doesn't have a schedule of controlled agents. There have been efforts to develop lists — the Australia Group maintains a list of biological agents and toxins for export control purposes — but there's no universal, treaty-based taxonomy for biological weapons the way there is for chemical weapons.
Partly because the BWC was negotiated in nineteen seventy-two, before the CWC, and the verification provisions were weaker from the start. And partly because biological agents are inherently harder to classify than chemicals. A chemical has a defined molecular structure. A biological agent is a living organism that can be modified, engineered, or even synthesized from scratch. The boundary between a pathogen and a research tool is blurry in ways that the boundary between a chemical weapon precursor and an industrial chemical is not.
The taxonomy gap for biological weapons is a real vulnerability in the non-proliferation architecture.
And it's been flagged repeatedly at BWC review conferences, but the political will to develop a more robust taxonomy and verification system hasn't materialized. Some states actively resist it, arguing that it would impose burdens on their biotechnology industries without providing meaningful security benefits.
Which is the same tension we see in every one of these systems — the balance between control and legitimate use.
That balance is never static. Technology changes, threats evolve, and the taxonomies have to evolve with them. The challenge is that the evolution is slow — treaty amendments take years, sometimes decades — while the threat landscape can change in months.
Where does that leave us practically? If someone is concerned about weapons use and wants to use these taxonomies effectively, what should they know?
I'd say three things. First, know which taxonomy applies to your concern. If you're tracking cluster munitions, it's the MRS and the Convention on Cluster Munitions definitions. If you're tracking missile proliferation, it's the MTCR Annex. If you're tracking chemical weapons, it's the CWC Schedules. Second, understand what the taxonomy can and can't tell you. It can identify a weapon, but it can't interpret the legality of its use. Third, follow the organizations that operationalize these taxonomies — CAR, the OPCW, the UN panels of experts, the humanitarian demining community. They're the ones turning the reference books into actionable intelligence.
If you're a citizen trying to make sense of news reports about weapons use in a conflict, the taxonomies give you a vocabulary for asking better questions. Instead of "were cluster munitions used," you can ask "which specific munition, what's its failure rate, who manufactured it, and how did it get there.
Which are much harder questions for governments and armed groups to dodge.
There's one more taxonomy I want to put on the table that we haven't mentioned — the International Traffic in Arms Regulations, ITAR, which is the US system. It has its own categorization, the United States Munitions List.
And ITAR is interesting because it's a national system that has global reach. The US Munitions List covers everything from firearms to spacecraft, and because the US is such a dominant arms exporter, the ITAR categories effectively shape global arms trade patterns. If something is on the US Munitions List, it's subject to export controls regardless of whether the Wassenaar Arrangement or the MTCR would cover it.
The US Munitions List has its own quirks. It covers items that other taxonomies don't — things like cryptographic systems, certain types of software, even some training services.
Which reflects the American view that arms control isn't just about the hardware, it's about the knowledge and the enabling technologies. That's a view that's increasingly shared by other countries, but the US was early in codifying it into a formal taxonomy with legal force.
We've got a layered system — international taxonomies for coordination and treaty compliance, national taxonomies for export control, and operational taxonomies for battlefield logistics and humanitarian clearance. They overlap but they don't perfectly align, and the gaps between them are where a lot of the interesting problems live.
Where a lot of the proliferation happens. If you're a state or a non-state actor trying to acquire weapons while evading controls, you look for those gaps. You find the component that's not quite covered by the MTCR Annex, the chemical that's on Schedule Three instead of Schedule Two, the ammunition type that hasn't been added to the MRS yet. The taxonomies are a fence, and like any fence, they're only as strong as their weakest post.
Which is a sobering thought to end the substantive discussion on. But it's also an argument for paying attention to these systems, for keeping them updated, for closing the gaps when we find them. The taxonomies aren't glamorous, but they're load-bearing.
They're the load-bearing infrastructure of arms control. Nobody gets excited about the Dewey Decimal System, but try running a library without one.
Covering the covers.
Now: Hilbert's daily fun fact.
Hilbert: In nineteen sixty-three, a Soviet biological research station in Suriname secretly documented a three-year cooperative relationship between a wild capybara and a spectacled caiman, in which the capybara would lead the caiman to fish traps set by local villagers, the caiman would break the traps open, and the two would split the catch. The Soviet researchers reportedly destroyed their field notes when the station was abruptly shuttered in nineteen sixty-four, and no corroborating account has ever surfaced.
I have so many questions about why the Soviets had a biological research station in Suriname.
I have so many questions about the capybara-caiman profit-sharing arrangement.
Thank you to Hilbert Flumingtop for producing. This has been My Weird Prompts. You can find every episode at myweirdprompts.We're back next week.
