Gunpowder as a technology was brought with the Mongols to Europe.
In addition, you should read a book by Tonio Andrade called "The Gunpowder Age" which goes into why the Europeans, despite getting gunpowder later, were able to build the best cannons and guns.
His basic thesis is that Europe had terrible walls in comparison to China and this caused China to not develop the gun type weapons earlier.
Chinese walls in the year 200 AD were, as a matter of course, 5 to 10 meters wide, of a similar height, with a trapezoidal cross section made by stacking pounded earth and then cladding the outside in stone.
When gunpowder first appeared in the 800's, it could do nothing to these walls. Short of digging under the walls and blowing them up, gunpowder weapons were too weak to do anything to the walls that they already had had for a thousand years.
So they developed a lot of incendiary and grenade type anti personnel weapons, but not many cannons type weapons. Remember it took the biggest cannon ever produced to knock down the Theodosian walls around Constantinople... And those walls were about as thick as the walls around several dozen Chinese cities.
When gunpowder got to europe, it was a different story.
There were a lot of castles and walls that had walls that were 1 to 3 feet thick and made of stacked stones. Really weak, crude cannons could knock them down easily. And they did. So Europeans started working on the cannons to knock down all the legacy fortifications they had. This caused them to iterate and develop them further. Add in their use on ships, and Europe had an incentive to work on fast, accurate, high velocity guns.
When the Europeans had to develop new walls to resist cannons, the innovation was the star fort. These were carefully designed walls with angles to avoid presenting an enemy with a direct perpendicular shot. In addition, their walls had a squat trapezoidal cross secton and we're built out of packed earth clad with stone. The walls that the Europeans developed to stop gunpowder were similar in construction to the walls the Chinese had been building since before gunpowder was invented.
It was these fortifications and the accompanying cannons that enabled the Europeans to successfully place forts around the world and resist efforts by local kingdoms, that often outnumbered them by orders of magnitude.
The argument is oversimplified here, hut it's a great book.
Very interesting theory. But while the first noteworthy use of cannons in the West--the siege of Constantinople--was not a pitched battle, gunpowder weapons very quickly found their way into the hands of infantry and cavalry. And then field artillery.
When cannons overnight made every castle or city wall in the West obsolete, "Le Trace Italienne" replaced it, or sometimes just retrofitted it. But unlike the Chinese model, these wide walls were also relatively low; and all designed with hidden anti-infantry cannon batteries firing from defilade--something that was never part of Chinese military architecture.
Interestingly, the Romans faced armies of horse-archers for centuries with good results. Using the 'Testudo' or tortoise formation, for infantry to shrug off arrow fire; and using artillery to harass or break up enemy formations forming up for charges (and in sieges), their approach worked adequately for over two millennia.
Constantinople fell in 1453. Cannons had been kicking around Europe for over 100 years by then. They were used by both England and France in the 100 years war in the 1360's.
The Start fort in my post is the Trace Italienne fort. The point was that Chinese walls going back to long before gunpowder times happened to share a lot of the characteristics of the purpose built European anti cannon fortifications... namely the width of the walls and the pounded earth construction. The Chinese did use cannons but as anti personnel weapons which shot arrows or fire because they knew they weren't going to do anything to the walls. As a result, there wasn't a virtuous circle for cannon development in the East where there were lots of small, intermediate sized walls to iterate slightly better cannons on.
The Roman testudo wasn't designed for field deployment against horse archers, it was made for besieging cities (and the arrow shooting people within) and it was rarely used. It just has an outsized role in popular culture because it looks cool. It's like the lorica segmentata (the banded metal armor that Roman soldiers are usually depicted wearing), that was actually not vey common. Most Roman soldiers had the better and more expensive chainmail shirt...., but that turns into a rusty chunk of iron when buried, while the iron band in the segmentata are much more likely to be preserved. As a result, later depictions of Romans would show all of them wearing an armor variant that was not actually very common.
My understanding was that Algecieras (sp?) in 1343 was an outlier, and that Constantinople was the first siege where a battery was used to systematically breach a double set of walls for an assault, where it was arguably the deciding factor in the fall of the city. And got the attention, as it were, of Venetian and Genoese military architects, producing the eponymous "trace Italienne."
"Bastion" is the more accurate, widespread and literal term over "Star", as many nations like the Portuguese were fond of rectangular layouts.
And yes, the testudo was used in campaigns against the Parthians, as mentioned by both Plutarch, Cassius Dio, and Tacitus.
The actual usage of lorica segmentata is unknown, with competing hypotheses about what troop types used it, why, where, when. The only certainty is that it appeared in the late Republic or early Imperium, likely as an economy measure. Also: no archaeological examples of it have ever been found in what were the Eastern portions of the Empire. And if you are referring to actual historical depictions, yes, it was displayed on Trajan's Column.
That's tricky. Nuclear weapons are devastating when used, but it's that very devastation that makes the decision to use them so difficult. It's pretty much a "Go big or go home" type of weapon
Drones don't have that issue (unless they have nuclear payloads, of course).
Why can’t they? Is that obvious? It seems that they can patrol a territory and identify who isn’t supposed to be there - what more is needed to take and hold it?
1. In an environment with EW active, the range of drones is limited. If you want to move your umbrella of drone coverage forward you need to move the operators forward
2. Even with overwhelming drone coverage, you still need boots on the ground to accomplish any real goals for control of an area. Could America control Afghanistan just through use of drones, for instance? I think we found we couldn't, we still needed infantry to man checkpoints, gather intel, search for weapons caches, perform non-lethal area denial, etc.
Yeah, I guess the big question is what the real goals for control of an area are. "Non-lethal area denial" sounds like a useful term for a kind of thing that really is going to be quite hard to do with drones. If the goal is just to ensure that there are no enemy military forces working in the area (including drones) then drone saturation may be enough (if you can rely on electromagnetic communications wavelengths), but if you have some goal of how to relate to the civilian population as well, you're going to need people there.
This is true today. But as sophisticated ML models become capable enough to run on phone-level hardware, will this still be true? Imagine 10 years at the development pace we're seeing today.
Once AI produces autonomous kamikaze short- and long-range drones, everything changes. ECCM and fiber-optics for delivery become unnecessary.
But fully autonomous fighters and other platforms that launch their own munitions--and can survive to return for resupply--are still up to a decade away.
Wire-guided antitank missile operators were super high-risk, as the missile launch blast and the wire trail pointing right at your location made enemy counter-fire very effective.
Ukrainian drone fiber-optic teams probably place their operators at a distance from the wire spool terminus. To minimize casualties.
In a few years, autonomous kamikaze and (possibly) recon drones will make fiber-optics obsolete.
It's very easy to build resiliency into your tech infrastructure; modern data centers by design are built to automatically back up other data centers, and they could be dispersed thousands of miles away behind the front lines.
You can definitely fail to build resilience (as we found out about a month ago), but you can spin up failsafes very rapidly if it's a requirement. It's certainly easier to build backups for your software than for your manufacturing tools.
While I agree with the premise of warfare shaping society--Peter Turchin makes an excellent case for steppe nomads driving large agrarian empire formation--, I don't know if the examples provided make that good of a case for the weapons of war driving the societal changes that precipitated extremely deadly conflicts.
(1) because of the delays from the advent of the military technology to the peak conflict, the military technology just seems to be a background factor, they don't seem to drive much, e.g. if in the 1600s all people had were Roman Empire weapons technology, we would have had extremely deadly conflicts with Roman Empire weapons. If military technologies were driving societal behaviors (e.g. more recruitment for the military), then it doesn't seem as if we'd see conflicts suddenly explode in deadliness and then peter out. There would be continual growths in conflict deadliness over decades/centuries from the 1000s to 2000s. (this is probably a general trend, but the deadliness seems to reflect state size/power and doesn't have an obvious ties to the weapons of war or weapons evolution)
For example, it's true industrialized countries had advantages when fighting non-industrialized countries in the 1800s and this pushed non-European countries (e.g. Japan, China) to industrialize, but it's not clear at all that warfare was a big driver of industrialization in Europe and especially the US. The US' industrialization seemed to have nothing to do with warfare, though it was very helpful for warfare (e.g. the North's victory in the Civil War). My impression is that that's roughly true for Europe in the 1800s. Interstate fighting isn't a big driver of industrialization, though being industrialized helps (e.g. the Franco-Prussian War).
(2) the periods of peak conflict seem a little arbitrary as they ignore the Crusades (~1100s), the Spanish Conquest of the Americas (1500s), and the internal rebellions of the QIng Dynasty (1800s). My impression is the Napoleonic Wars in the 1800s are a better peak for the Age of Gunpowder.
(3) While the Mongol conquests are extremely deadly, I don't know if Mongol social reforms are what enabled their large conquest. Steppe nomads before them were able to conquer Northern China (e.g. the Khitan and the Jurchens did so right before them) and my impression is that these are just the tip of the iceberg historically. The Mongols then use the resources (and siegecraft) of Northern China to take on other agrarian empires. It's possible the big difference between the Mongols and the Jurchens who occupied Northern China before them is an expansionist drive, i.e. a willingness to direct Chinese agrarian resources for further conquest more than anything.
Taking the modern example at face value, did Russia invade Ukraine because of the emergence of drones? Or is a drone advantage/disadvantage going to be the reason for China to invade Taiwan? It seems doubtful.
The thing that gunpowder enabled was peasant armies that could stand up to professional soldiers - it took a lifetime to become a skiled archer or cavalryman, but anyone could learn to fight with muskets or rifles quickly enough to be useful before the war ended. I seem to recall that Napoleon was the first to take this to its logical extreme and became the inventor of modern conscription, using the manpower of an entire nation to fuel its army rather than being stuck with however many people it could support as a professional warrior class in peacetime. (I've also heard that a lot of Napoleon's enemies in continental Europe were reluctant to do the same thing because they were hereditary nobility that were afraid that an armed citizenry would overthrow them, but they ended up with no choice and did it anyway to fight him. I don't know if any of this is accurate, though.)
You keep leaving a great deal out. And you are taking credit for points of view that have been thought out decades ahead of you.
1) The first role of defense is to prevent the conflict. This is the bulk of expenditure, thinking and effort
2) The UKR conflict would have been over in the first month with US current technology. To your point, howitzers don't matter in the modern conflict; we determined that in 1998.
3) At nation state level, conflict, electronic warfare, social engineering, etc matter a lot more than battlefield conflict.
4) Battlefield conflict doctrine is constantly updated, hence the army's procurement of UAVs. However, the military has no interest in a UKR like conflict. It is awful game theory. Their goal is an asymmetric battle. UAVs are symmetric battle. Neither side will have a huge advantage. It is a poor doctrine. The military has been thinking about the points you raise, but 30 years ago. And made other blocking maneuvers.
5) US military is thinking about macro issues like how food supply fits into military power. Or how Ozempic going off patent in China helps China redirect from HC cost to military.
6) I doubt China thinks it can militarily dominate the US with UAV technology as a silver bullet. Since military doctrine is mostly about creating a threat to avoid war, in what world would US capitulate out of UAV fear?
7) UAVs are more likely to end proxy conflicts. Much like 9/11 ended airplane hijackings. We were working with UKR years ago on network-centric warfare, which is one reason they have advantages in this conflict.
All of these are good points that I would summarize as “don’t over-learn lessons from Ukraine”. The Ukraine conflict has demonstrated that drones are powerful weapons in a conflict between two decrepit Soviet militaries with very outdated hardware and barely any supply of modern stealth technologies for either missiles or aircraft. This conflict should be extremely instructive for developing nations that might go to war.
But that doesn’t really describe the kind of wars the US might fight in the future. Quadcopter drones will do nothing against conventional aircraft that aren’t parked on a tarmac; they can’t move fast enough, fly high enough or travel far enough. And NATO powers in general, and the US in particular, aren’t going to be zoned out of a battlespace by air defenses of the kind that are shutting down both Ukraine and Russia from air superiority. This war would have been over in a couple of weeks had the US Air Force been sent in to support Ukraine from the beginning; Russia had (and still has) no way of dealing with that degree of conventional air power. (Whether it would have been over because Russia retreated or because they initiated the nuclear apocalypse, though, is an open question, and the fact that nobody but Putin knows the answer for sure is probably the main reason we didn’t do this.) Even if you accept their own claims about their most advanced modern air defense systems (and their claims are historically very exaggerated), they still can’t target an aircraft like an F-35 unless it’s practically right on top of them, by which time the aircraft has been in weapons range to destroy the defensive radar for over a hundred miles.
The longer-ranged attack drones that Russia and Ukraine are using are also not a new technology; we just usually call them “cruise missiles”. They were first developed in the 1940s and have been continually refined since then, but the low speed flight that makes them relatively affordable by comparison to high-speed ballistic missiles also renders them highly vulnerable to air defenses. Modern electronics have somewhat tilted the cost equation in favor of the attacker when using these kind of basic cruise missiles, but solutions to that have already been found, such as the already-deployed Advanced Precision Kill Weapon System that retrofits old 70mm unguided rockets with a basic guidance package that’s more than sufficient to seek and destroy subsonic electric-powered drones, at a cost of around $30k per shot, comparable to the price of even the cheapest long-range strike drones. The US has been using this to great effect against the Houthis, flipping the cost equation on them and forcing them to use much-more-expensive and limited-supply anti-ship missiles instead of cheap drones, which puts much more significant strain on their resources and those of their backer, Iran.
I think by far the scariest thing we’ve seen from Ukraine has been the shipping container attacks that they successfully executed against Russia. This is something genuinely new and dangerous as most military facilities are not currently equipped with the kind of point defenses needed to respond to an attack of this kind. And the ubiquity of cargo container transport in all developed nations — and the degree of blind trust afforded to globalized trade to keep the system moving — render us especially vulnerable to this kind of strike. That said, this is more of a capital investment question than a technological one. Building automated point defenses to shoot down autonomous quadcopter drone swarms is just an application of technologies we already have, not something that will require breaching new frontiers of scientific research.
Honestly, my main reason to despair for the US is simply that we’ve already succumbed to the most devastating weapon of the modern age: the twisted, destructive version of ourselves forged in the demonic hellfire of social media. The only wars the public cares about fighting are culture wars with one another, and they much prefer to keep getting cheap consumer goods from China, paying low taxes and getting fat entitlement checks to spending money on war preparations because they think no one would be crazy enough to attack us. And I think that’s going to remain the case until we suffer a catastrophic defeat, probably against China: it’ll take another Pearl Harbor to wake people up to the threat. Except this time, we won’t have the industrial base to do anything about it, and we’ll be up against an opponent that’s not going to be inclined to be a charitable victor and is more likely to want to knock us flat and keep us that way so we can’t challenge them in the future.
While I agree with many of your points, this statement stood out:
"This war would have been over in a couple of weeks had the US Air Force been sent in to support Ukraine from the beginning; "
That has been the supposition for many wars, including US wars, and hasn't borne out often, and not between major powers since the Napoleonic Wars.
In 1998, it took NATO air forces 78 days to get Serbia to agree to end hostilities. And Serbia didn't have a nuclear deterrent that would reduce US scope of action, the way it would be with Russia.
Would the war be different with US involvement? That is a certainty, but the only one. Would it be over by now? Possibly, but there are reasons to doubt. It's possible it ends because we're all dead. It's possible that Putin is prepared to let a much larger share of men of military age expire on the battlefield, which might take more than a decade. Another possibility is the war evolves into asymmetry. Another is that the US decapitates Russia, and then we have endless internecine war across much of the Russian space, including Ukraine.
Over in a couple weeks? That's the most common delusion in history, and the costliest.
These are all good points, but I would note that the situation in the first few weeks of the war bore a great deal more resemblance to Kuwait in 1991 than to Bosnia. The Russians had brought in poorly supported columns of armored units without significant logistics, anticipating an easy blitz to Kyiv and the end of the conflict within days. After the Ukrainians successfully repelled the initial wave, you had disorganized and stranded Russian troops, columns of armored units without food or fuel, and no ability to entrench or create defenses, a radically different situation from the Bosnian campaign where the enemy was defending territory they had controlled for over five years before NATO got involved. The majority of the Russian ground forces probably could have been annihilated in very short order, just as Saddam’s forces were when they were caught unprepared in Kuwait and forced to retreat.
The situation today is, of course, radically different. The Russians have had years to entrench and build the necessary logistics for a war of attrition. But defending against aircraft from an entrenched position is much, much easier than trying to take and entrench your position when the opposition has overwhelming air superiority. If the initial Russian forces had been destroyed and pushed back to the pre-2022 lines of conflict, they would have had a much, much harder time trying to advance further and the cost would have been astronomically higher because they wouldn’t have the use of the artillery units that were the central pillar of their offensive strategy for the first two years of the war. Obviously, no one can prove the counterfactual, but my intuition is that at the very least it would have forced a chaotic retreat from Ukraine similar to what we saw in the Kharkiv offensive, but much more sweeping, and Russia would have been very quickly pushed back to the pre-invasion lines and needed a significant amount of time to regroup and rebuild before attempting to restart the conflict.
But of course, the most unpredictable element here is the fact that Russia is ruled by one man who likes to rattle the nuclear saber when threatened, and a 1991-style wipeout of his invasion force, representing the majority of his available ground forces, would be a threat unlike anything Russia has faced since WW2. This would likely push him into a desperation mindset rather than the attritional mindset he’s employing now, confident that his individual will to win will significantly outlast the fickle, self-absorbed voting publics that ultimately rule the NATO states. Desperate people and nuclear weapons are a very dangerous combination, and the “escalate to de-escalate” strategy of employing tactical nukes against Ukrainian targets as a signal of willingness to go all the way to strategic weapons may well have been his choice of response.
I appreciate your thoughts here. I don't doubt that Ukraine would be in much better shape today if the US had decided upon providing air superiority from the outset.
Mine are that Russia made the classic mistake of thinking the war would be over fast, because it expected things to go as it planned. You proposed the same mistake for the US (don't worry - you have plenty of historical company). Events almost never go to plan, and never when one or more belligerents have unlimited objectives, as Russia does.
In play is misunderstanding or at least ignoring Russian goals and rationales. Destroying Russia's assets in Ukraine and pushing them out of Ukraine is likely not enough to end the war. Ending the war would likely either involve an invasion NATO is not capable of currently, or decapitation, both of which are spectacularly unwise.
In short, the war won't be over until Russia's goals change, and changing them is a long term effort.
Appreciate the thoughtful response. I think the core disagreement came down to the way we were each using the word “war”. When I said the war would be over in a few weeks, I was speaking of the immediate hot conflict, not the lower-intensity hybrid warfare Russia has been waging for a long time. I completely agree with you that Russia will continue pursuing hostilities through whatever avenues it deems feasible for as long as the current leadership and their ideological fellow travelers — believers that sovereignty over the maximalist Russian Empire is their divine birthright — remain in power. (Which, given history, will likely be for as long as distinguishably Russian culture continues to exist. Fortunately for the world, alcohol and low birth rates are slowly but surely fixing that problem without any assistance from the rest of us. Putin’s even kindly accelerating the process with his military misadventures.)
I don’t think, though, that Russia would have had a ton of options for continuing high-intensity conflict after a chaotic unplanned retreat under heavy fire from US airpower. Moving more ground troops into Ukraine would have been infeasible with the enemy holding air superiority. Sitting on the Russian side of the border and lobbing long-range ordnance into Ukraine would probably have been infeasible as well, working under the assumption that a US that had the political will to directly intervene in the conflict to protect Ukraine would probably not draw an artificial line at striking targets on Russian soil, even if it perhaps drew an arbitrary line at sending manned aircraft into internationally-acknowledged Russian airspace. (Even under our current, incredibly timid and cowardly approach, we’re still allowing the Ukrainians to use NATO-provided weaponry to strike targets on Russian soil, after all.)
The main things that initially allowed Russia asymmetrical strike capability against Ukraine were their larger supply of longer-range guided ordnance and their fleet of bombers capable of taking off from airbases Ukraine couldn’t strike and launching from high altitudes far behind the line of battle where Ukrainian ground-based air defenses weren’t a threat. Neither of these factors would be a significant impediment to the US, though: we have plenty of accurate long-range guided ordnance too, and plenty of capability to deliver it to anywhere in Russia, meaning that no staging point would be safe from retaliation. (Any Russian aircraft getting too close to the border would be highly vulnerable to combat air patrols armed with long-range air-to-air missiles, too, severely limiting the ordnance they could use for strikes on Ukrainian territory: no cheap glide bombs, only expensive ballistic missiles.) We’re already seeing significant shifts in Russian strategy forced by Ukraine’s development of the no-really-we-swear-it’s-not-British FP-5 Flamingo long-range cruise missile; imagine if they’d had similar capability from day one.
My general assessment is that in this scenario, assuming Putin wasn’t willing to uncork the nuclear genie — and I’m quite sure his good buddy Xi would be pulling out every bit of leverage he had to ensure he didn’t, given that China is by far the country that has the most to lose if the nuclear taboo is broken — his only real option would have been to retreat, call a cease-fire and then let things settle down while he rebuilt his forces and prepared for round two. But the task in front of him would be very daunting: he couldn’t just rebuild forces sufficient to fight the Ukrainian military; he’d need forces sufficient to fight NATO as a whole. This is likely out of reach of Russia’s economy even on a total war footing unless the NATO states believed the threat was completely gone and returned to a 1990s peacetime disarmament mindset. And while I would never underestimate democratic governments’ ability to ignore the obvious and bury their heads in the sand, the European response to the Ukraine invasion suggests that the invasion woke them up fairly quickly and they’d be unlikely to go back to sleep if intelligence indicated Putin was stockpiling weapons to try again in a few years.
I disagree that US ordinance stockpiles are great. One of the war's surprises is that excepting reserves, there are some shortages, with production ramps lagging far behind what was previously assumed possible. Usage would have soared with actual US engagement.
I think your assessment that the Europeans were woken up quickly is not correct. Of course, again definitions matter. But for comparison, it didn't take years to mobilize for WW1 or WW2, yet today Western Europe remains still far from awake nearly 4 years into the wide-scale invasion. Some like Spain determinedly so. I have few doubts that they wouldn't fall entirely asleep if Putin desists to stockpile weapons. (Are Eastern Europeans more awake - oh yes!)
Europe hasn't even attempted to onshore all drone materials fabrication, or opened a single mine for this purpose.
I think what's interesting about these takes is that we may get a chance to actually test these theories, if the "Russia is going to start a war with NATO" predictions are actually true.
That theory can't be tested in the future, because Russia has already been prosecuting a low-intensity hybrid war against NATO since at least 2008.
So the question isn't whether Russia is going to start a war, but whether NATO is going to fight one. The answer appears to be 'no'.
Russia is currently increasing the intensity of conflict, but so gradually or incrementally it is (intentionally) difficult to draw a line. That line so far isn't testing defences, such as drone incursions in Poland or infantry incursions into the Baltics.
How long did it take Israel to force Iran to capitulate? And they have approximately the same level of military technology as Russia; Russia is using Iranian drone specs, not vice-versa, for instance.
This is year 46 and counting. You're referring to one of dozens of confrontations in a conflict that has raged since 1980. Iran lost control of its skies, but it didn't capitulate, cede anything, or abandon its goals. Iran has never looked weaker, but the conflict continues.
You should ask yourself: "What are Russia's goals, and why?" The answer will tell you why the US cannot fight and win a war with them "in a couple weeks".
FWIW, I'm all for bleeding the bear, except that it serves to strengthen China.
I mean, I like to think I'm clear-eyed about the limits of air superiority. Can it stop an enemy advance? Certainly. Can it dislodge troops? Unless they've had significant time to dig in, yes. Can it dislodge troops in population centers, or pacify civilian populations? Most likely not. So I guess it comes down to your win conditions in Ukraine; whether the war would be over without Ukraine retaking Donbas & Crimea or whether it would be over if Russia could not support any new advances.
PGMs would have a field day on the way Russia positioned. UKR was killing Generals with our coordination. Russia has not been running active operations like Iran. They are rusty. The command structure is rusty.
It is correct Russia had other options, but even going nuclear is questionable with their fleet. And the calculus of "How reliable is this" fits into their thinking of response. It would have triggered many other issues, but from a doctrine standpoint, UAVs would have done little in the geopolitical context. And that is my point. A JDAM matched with our C4ISR would wipe out their command and control, and many other capabilities quickly.
Fair enough. The thing is, Russia would likely just sit outside Ukraine and lob munitions, unless the US is going to go deep into Russia. And then Russia will wait for the US to lose interest. For the Ukrainians, that's obviously preferable, but it doesn't solve anything for anyone else.
China could easily make us capitulate in a wartime environment, no?
1. Immediately ban all exports and watch our war production (and everything else, eventually) seize up. Shut down pharma exports.
2. Flip all the "destroy critical infrastructure" switches they've set up. Make power plants blow up, take out telecommunications and power grids across the country, etc.
3. Do 100 distributed drone attacks out of shipping containers and trucks - same thing Ukraine did to blow up Russian bombers within their borders.
I don't get the sense we have any similar tools to fight them with. We'd have to use actually, long-range weapons, which seems like something we wouldn't do even in the scenarios above, for MAD reasons.
Modern geopolitical conflict is multi-domain. UAVs are a tiny part of the story. Economic forces are a powerful component of the calculus.
This reminds me that most people seem to have little comprehension of how forward-thinking the military is on these conflict scenarios. Not sure how to educate people, but if you look at some of the non-classified work on war games, you get a sense of how much they think about these scenarios.
I would add that many policymakers are fairly ignorant of the matter. That recent film on Netflix about nuclear war seemed pretty weak, even though it claimed to have policy makers involved.
The part I really don't like in that show was the thought we would do a full response without confirmation of the initial bad actor. After one hit of a nuke, we would not have a need to respond in a use-it-or-lose-it scenario. Frankly, we would be better positioned to respond because NCA would be bunkered.
Yeah, it seemed insane to me that they would seriously consider launching "counter-attacks" against _everyone_, because they couldn't tell who the responsible party was, based on a single strike, with no evidence yet of any more incoming weapons.
The other things the enemy knows, but movie producers do not, is the quality of their equipment. Russian tanks have poor metallurgy. UAVs fail. Bombs do not blow up. Nuclear launch systems fail.
The US has been operating at a very high tempo, working through defects. Others have not. The Russian nuke fleet probably has 25% chance of getting to the target. If you are planning an attack and only 25% are even capable of getting to the target, do you really think your strategy wins?
Also the notion of a system-wide successful cyber attack on military infrastructure is a stretch. These are moated systems with limited ability to test an attack.
It is obviously more likely that China would go after commercial infrastructure.
"Leave the World Behind" is a more interesting first strike. It has its own problems, but it is no surprise that Obama produced it. He would know the threat scenarios much better than many.
Ha ok, so you’re just saying UAVs alone won’t make the US capitulate to China, but agree that China has a suite of tools that could make us capitulate.
China needs us as much as we need them - so neither is eager for a way. China does want to scare our population so we spend money we may not have. China would rather spend money on growth than defense.
China and we are running wargames on economic warfare, and other domains. Noad is on a side issue that could become bigger, but people who have fought wars for the last 30 years understand the balancing nature. The US has controlled battle domains for 30+ years and has tremendous practical experience that others lack. Others, including China.
This is true, but keep in mind that the Chinese will be checked by the same fears that prevented the US from going all-out against Russia: it is extremely dangerous to land a knockout punch to a heavily-armed nuclear power, because they may decide that if they’re going down, they’re taking you with them. That will remain the case and force some degree of caution with massive conventional overmatch until someone develops a perfectly reliable countermeasure to modern nuclear missiles, one reliable enough that national leaders are willing to gamble the entire existence of their nation on it.
If you read military doctrine, US and China have been thinking about economic warfare, cyber, and non-nuclear foring functions like food, as baseline warfare planning for 30 years. In 1998 when I was involved, we were more concerned about the flow of Soy than nuclear for balancing superpower status.
China has some very real vulnerabilities as well. We rely on them for technology and pharma inputs, sure, but they're a net food importer, which is rather more important. I doubt the Chinese have a "destroy critical infrastructure" switch and we don't, and the only country that's successfully deployed a distributed drone attack is a US ally. And their population is dense enough that one JDAM on the Three Gorges Dam would probably be the deadliest strike in military history.
Underneath drone and AI technology are huge amounts of metals and minerals, which China largely controls now. That’s a choke point the Western allies are finally waking up to. One tiny example: Alcoa is testing ways to extract gallium from bauxite at a facility in Western Australia. The U.S., Japan, and Australia governments are funding this project, with capital and off-take agreements. A good sign, IMO. But we are way behind China in terms of controlling critical mineral resources. Rio Tinto is testing the same thing in Canada with support from Western allies.
Rare earth minerals and critical minerals aren't rare. They are just expensive to extract and process in a way that isn't ridiculously toxic.
Myanmar is the world's third largest producer of rare earths. Not because it has the third largest reserves, but because the area along the Chinese border is governed by a patchwork of revolutionary organizations, militias, and narco groups who all need money. And Chinese companies are only too happy to give it to them in exchange for mining rights.
The way they do it there is two ways.
One, open mining where they haul off the rock for processing. This is pretty toxic and expensive.
Two, they find a deposit of so called ionic clays, where the rare earth elements have saturated clay in the ground for a few thousand years, and drill several hundred holes vertically to and through the clay. Then, they drill a fee holes horizontally lower down the hill to connect with the vertical holes. Finally, they pump a washing fluid into the vertical holes. It goes through, dissolves the clays, and washes out through the horizontal holes into a collection pond. The pond is then dried and the sediment shipped to China for processing...
We are in a mineral arms race now. Mining will continue with or without the Western allies. The iPhone I’m using to type this message is filled with critical minerals from mines I will never see. Tim Cook understands this. He knows China’s supply chains better than most experts.
Tim Cook's Apple sent money and industrial equipment to China worth more than twice the Marshal plan in inflation adjusted terms. He's kind of a negligent traitor in the sense that he did more to indusrialize the United states's greatest rival than any other American.
Iran has good drone tech and lots of drones. Didn’t help at all against Israeli and American attacks from the air. Eventually, of course, drones will be fighting drones in the air, but conventional air superiority over the battlefront is still highly effective now. Ukraine is really a demo of what happens when neither side has an airforce capable enough to suppress air defenses (in part because American AWACS are able to operate safely from behind NATO lines). With your opponent having air superiority, dug in front lines would be insanity, and the infrastructure necessary to transport and supply drones and armor and artillery would be at risk of destruction.
Dispersed guerilla forces with small drones able to launch terrorist type attacks from a distance or launch a massed attack against a military target will still be impossible to bomb into submission, of course, but this would be very different than what we see in Ukraine (more like the USMC’s new strategy)
The connections you draw are interesting. Perhaps mass-training of AIs by observing people playing video games would be an adaptation to move society toward readiness for drone warfare.
You were prescient to call quadcopter drones as a military technology of the future. However, I'd advise you to not be caught fighting the last war. In the 2010s, for instance, the cutting edge of military technology was the fixed-wing loitering drone; after the USAF Predator drone was used to great effect in the GWOT, the Bayraktar was the decisive technology in regional conflicts in Kurdistan and Nagorno-Karabakh. Who else remembers the "People's Bayraktar" campaigns to buy Ukraine these drones? Of course, when used against a modern military they proved worthless; slow, high visible and easy to shoot down by basically any anti-air platform.
This is how I feel about quadcopter drones today. Obviously they're being used to great effect in Ukraine, where the frontline is fixed, approximate enemy positions are known and relatively close, civilian casualties are acceptable, and supply lines are secure. Remove any one of those, though, and their usefulness plummets. Plus, as you note we are one technological innovation away from making them a sidenote in the next war - anti-drone technology to shoot down any fast-moving flying object seems relatively straightforward, especially compared to handwavy AI-piloted drones that will only kill the enemy. And as other commenters note I don't see any use for them in a potential naval conflict for Taiwan.
How much of the global success of democracy and participatory government was contingent on the dominance of the infantryman, and what does his eclipse mean for a country like the US founded by and for the firearm-wielding citizen soldier?
Thanks for the essay. As noted in the comments below unclear if drones are the be-all, end-all or whether the three eras you noted perfectly fit...nevertheless, your point that AI, electric stacking and other technological developments are upending and will continue to upend how nations battle, seems to me, totally on point. Whether they are the cause of effect of societal change, however, is unclear to me. But they certainly go hand in glove.
Very depressing read for me to start my Thursday (but a worthwhile read nevertheless.)
The domestic use of drones for population control scares me, or the use by terrorist groups. One of my dystopian predictions is that within our lifetimes, a drone bunker, anti-drone drones, or point lasers may be things we have installed in our communities and/or our homes.
As you noted, the broader issue is the acceleration of technological change. Won't that become highly destabilizing as perceptions of unknowable adversary capabilities increases? I'm reminded of Nick Bostrom's Vulnerable World Hypothesis. Not only are we pulling balls from his metaphoric urn faster, but the destabilizing FOMO, potentially seen as 'strike 1st or lose/die', becomes a major geopol risk itself.
Gunpowder as a technology was brought with the Mongols to Europe.
In addition, you should read a book by Tonio Andrade called "The Gunpowder Age" which goes into why the Europeans, despite getting gunpowder later, were able to build the best cannons and guns.
His basic thesis is that Europe had terrible walls in comparison to China and this caused China to not develop the gun type weapons earlier.
Chinese walls in the year 200 AD were, as a matter of course, 5 to 10 meters wide, of a similar height, with a trapezoidal cross section made by stacking pounded earth and then cladding the outside in stone.
When gunpowder first appeared in the 800's, it could do nothing to these walls. Short of digging under the walls and blowing them up, gunpowder weapons were too weak to do anything to the walls that they already had had for a thousand years.
So they developed a lot of incendiary and grenade type anti personnel weapons, but not many cannons type weapons. Remember it took the biggest cannon ever produced to knock down the Theodosian walls around Constantinople... And those walls were about as thick as the walls around several dozen Chinese cities.
When gunpowder got to europe, it was a different story.
There were a lot of castles and walls that had walls that were 1 to 3 feet thick and made of stacked stones. Really weak, crude cannons could knock them down easily. And they did. So Europeans started working on the cannons to knock down all the legacy fortifications they had. This caused them to iterate and develop them further. Add in their use on ships, and Europe had an incentive to work on fast, accurate, high velocity guns.
When the Europeans had to develop new walls to resist cannons, the innovation was the star fort. These were carefully designed walls with angles to avoid presenting an enemy with a direct perpendicular shot. In addition, their walls had a squat trapezoidal cross secton and we're built out of packed earth clad with stone. The walls that the Europeans developed to stop gunpowder were similar in construction to the walls the Chinese had been building since before gunpowder was invented.
It was these fortifications and the accompanying cannons that enabled the Europeans to successfully place forts around the world and resist efforts by local kingdoms, that often outnumbered them by orders of magnitude.
The argument is oversimplified here, hut it's a great book.
Very interesting theory. But while the first noteworthy use of cannons in the West--the siege of Constantinople--was not a pitched battle, gunpowder weapons very quickly found their way into the hands of infantry and cavalry. And then field artillery.
When cannons overnight made every castle or city wall in the West obsolete, "Le Trace Italienne" replaced it, or sometimes just retrofitted it. But unlike the Chinese model, these wide walls were also relatively low; and all designed with hidden anti-infantry cannon batteries firing from defilade--something that was never part of Chinese military architecture.
Interestingly, the Romans faced armies of horse-archers for centuries with good results. Using the 'Testudo' or tortoise formation, for infantry to shrug off arrow fire; and using artillery to harass or break up enemy formations forming up for charges (and in sieges), their approach worked adequately for over two millennia.
Constantinople fell in 1453. Cannons had been kicking around Europe for over 100 years by then. They were used by both England and France in the 100 years war in the 1360's.
The Start fort in my post is the Trace Italienne fort. The point was that Chinese walls going back to long before gunpowder times happened to share a lot of the characteristics of the purpose built European anti cannon fortifications... namely the width of the walls and the pounded earth construction. The Chinese did use cannons but as anti personnel weapons which shot arrows or fire because they knew they weren't going to do anything to the walls. As a result, there wasn't a virtuous circle for cannon development in the East where there were lots of small, intermediate sized walls to iterate slightly better cannons on.
The Roman testudo wasn't designed for field deployment against horse archers, it was made for besieging cities (and the arrow shooting people within) and it was rarely used. It just has an outsized role in popular culture because it looks cool. It's like the lorica segmentata (the banded metal armor that Roman soldiers are usually depicted wearing), that was actually not vey common. Most Roman soldiers had the better and more expensive chainmail shirt...., but that turns into a rusty chunk of iron when buried, while the iron band in the segmentata are much more likely to be preserved. As a result, later depictions of Romans would show all of them wearing an armor variant that was not actually very common.
My understanding was that Algecieras (sp?) in 1343 was an outlier, and that Constantinople was the first siege where a battery was used to systematically breach a double set of walls for an assault, where it was arguably the deciding factor in the fall of the city. And got the attention, as it were, of Venetian and Genoese military architects, producing the eponymous "trace Italienne."
"Bastion" is the more accurate, widespread and literal term over "Star", as many nations like the Portuguese were fond of rectangular layouts.
And yes, the testudo was used in campaigns against the Parthians, as mentioned by both Plutarch, Cassius Dio, and Tacitus.
The actual usage of lorica segmentata is unknown, with competing hypotheses about what troop types used it, why, where, when. The only certainty is that it appeared in the late Republic or early Imperium, likely as an economy measure. Also: no archaeological examples of it have ever been found in what were the Eastern portions of the Empire. And if you are referring to actual historical depictions, yes, it was displayed on Trajan's Column.
When talking about drones, these two facts should not go unmentioned:
- Ukraine is currently producing more combat drones than both Russia and all NATO countries combined (about 4 million, growing fast).
- And the drone supply chain in Ukraine is already highly local and becoming more so by the day; China plays an ever-smaller role.
If free countries want to learn, Ukraine is who they can learn from.
Read more:
- https://jamestown.org/ukraine-leads-world-in-drone-innovation-and-production/
- https://www.forbes.com/sites/davidhambling/2025/04/08/ukraine-is-making-fpv-drones-without-chinese-parts-and-at-lower-cost/
Is THIS is a big part of why Russia doesn't want Ukraine to ever join NATO?
Nuclear weapons still beat drones.
That's tricky. Nuclear weapons are devastating when used, but it's that very devastation that makes the decision to use them so difficult. It's pretty much a "Go big or go home" type of weapon
Drones don't have that issue (unless they have nuclear payloads, of course).
Unless you want to conquer not just destroy
They're pretty good at scaring off would-be conquerers, though.
Drones are a useful complement to traditional firepower but not a replacement.
Recently Russia have made gains, with Ukraine citing the lack of legacy systems.
Ukraine continuously call for more tanks, artillery, and long-range missiles - drones were born from necessity because of lack of alternatives.
Geographically the future battlefield of Taiwan with its vast beaches and forests does not suit the type of drone warfare seen in Ukraine.
https://challengerresearch.substack.com/p/the-drone-frenzy-why-preparing-for?r=5jzlze
Agreed, and not to state the obvious but drones can't take and hold territory.
Why can’t they? Is that obvious? It seems that they can patrol a territory and identify who isn’t supposed to be there - what more is needed to take and hold it?
1. In an environment with EW active, the range of drones is limited. If you want to move your umbrella of drone coverage forward you need to move the operators forward
2. Even with overwhelming drone coverage, you still need boots on the ground to accomplish any real goals for control of an area. Could America control Afghanistan just through use of drones, for instance? I think we found we couldn't, we still needed infantry to man checkpoints, gather intel, search for weapons caches, perform non-lethal area denial, etc.
Yeah, I guess the big question is what the real goals for control of an area are. "Non-lethal area denial" sounds like a useful term for a kind of thing that really is going to be quite hard to do with drones. If the goal is just to ensure that there are no enemy military forces working in the area (including drones) then drone saturation may be enough (if you can rely on electromagnetic communications wavelengths), but if you have some goal of how to relate to the civilian population as well, you're going to need people there.
This is true today. But as sophisticated ML models become capable enough to run on phone-level hardware, will this still be true? Imagine 10 years at the development pace we're seeing today.
This.
Once AI produces autonomous kamikaze short- and long-range drones, everything changes. ECCM and fiber-optics for delivery become unnecessary.
But fully autonomous fighters and other platforms that launch their own munitions--and can survive to return for resupply--are still up to a decade away.
Noah, incomplete sentence in footnote 2?
It takes a huge amount of energy...
Aren’t servers the main point of failure then? The answer would be to identify server farms and bomb them.
Drones linked by fiber optic to the operator’s terminal reduce that risk, but for yes, for networked drones.
Wire-guided antitank missile operators were super high-risk, as the missile launch blast and the wire trail pointing right at your location made enemy counter-fire very effective.
Ukrainian drone fiber-optic teams probably place their operators at a distance from the wire spool terminus. To minimize casualties.
In a few years, autonomous kamikaze and (possibly) recon drones will make fiber-optics obsolete.
It's very easy to build resiliency into your tech infrastructure; modern data centers by design are built to automatically back up other data centers, and they could be dispersed thousands of miles away behind the front lines.
When AWS us-east1 went down, a lot of the internet went down with it. My assumption is that govcloud is more resilient, but…
You can definitely fail to build resilience (as we found out about a month ago), but you can spin up failsafes very rapidly if it's a requirement. It's certainly easier to build backups for your software than for your manufacturing tools.
An autonomous kamikaze drone packed with high explosive looking for targets of opportunity behind enemy lines doesn't need backup data centers...
That’s a good point, easier to back up data than have redundant factories
While I agree with the premise of warfare shaping society--Peter Turchin makes an excellent case for steppe nomads driving large agrarian empire formation--, I don't know if the examples provided make that good of a case for the weapons of war driving the societal changes that precipitated extremely deadly conflicts.
(1) because of the delays from the advent of the military technology to the peak conflict, the military technology just seems to be a background factor, they don't seem to drive much, e.g. if in the 1600s all people had were Roman Empire weapons technology, we would have had extremely deadly conflicts with Roman Empire weapons. If military technologies were driving societal behaviors (e.g. more recruitment for the military), then it doesn't seem as if we'd see conflicts suddenly explode in deadliness and then peter out. There would be continual growths in conflict deadliness over decades/centuries from the 1000s to 2000s. (this is probably a general trend, but the deadliness seems to reflect state size/power and doesn't have an obvious ties to the weapons of war or weapons evolution)
For example, it's true industrialized countries had advantages when fighting non-industrialized countries in the 1800s and this pushed non-European countries (e.g. Japan, China) to industrialize, but it's not clear at all that warfare was a big driver of industrialization in Europe and especially the US. The US' industrialization seemed to have nothing to do with warfare, though it was very helpful for warfare (e.g. the North's victory in the Civil War). My impression is that that's roughly true for Europe in the 1800s. Interstate fighting isn't a big driver of industrialization, though being industrialized helps (e.g. the Franco-Prussian War).
(2) the periods of peak conflict seem a little arbitrary as they ignore the Crusades (~1100s), the Spanish Conquest of the Americas (1500s), and the internal rebellions of the QIng Dynasty (1800s). My impression is the Napoleonic Wars in the 1800s are a better peak for the Age of Gunpowder.
(3) While the Mongol conquests are extremely deadly, I don't know if Mongol social reforms are what enabled their large conquest. Steppe nomads before them were able to conquer Northern China (e.g. the Khitan and the Jurchens did so right before them) and my impression is that these are just the tip of the iceberg historically. The Mongols then use the resources (and siegecraft) of Northern China to take on other agrarian empires. It's possible the big difference between the Mongols and the Jurchens who occupied Northern China before them is an expansionist drive, i.e. a willingness to direct Chinese agrarian resources for further conquest more than anything.
Taking the modern example at face value, did Russia invade Ukraine because of the emergence of drones? Or is a drone advantage/disadvantage going to be the reason for China to invade Taiwan? It seems doubtful.
The thing that gunpowder enabled was peasant armies that could stand up to professional soldiers - it took a lifetime to become a skiled archer or cavalryman, but anyone could learn to fight with muskets or rifles quickly enough to be useful before the war ended. I seem to recall that Napoleon was the first to take this to its logical extreme and became the inventor of modern conscription, using the manpower of an entire nation to fuel its army rather than being stuck with however many people it could support as a professional warrior class in peacetime. (I've also heard that a lot of Napoleon's enemies in continental Europe were reluctant to do the same thing because they were hereditary nobility that were afraid that an armed citizenry would overthrow them, but they ended up with no choice and did it anyway to fight him. I don't know if any of this is accurate, though.)
You keep leaving a great deal out. And you are taking credit for points of view that have been thought out decades ahead of you.
1) The first role of defense is to prevent the conflict. This is the bulk of expenditure, thinking and effort
2) The UKR conflict would have been over in the first month with US current technology. To your point, howitzers don't matter in the modern conflict; we determined that in 1998.
3) At nation state level, conflict, electronic warfare, social engineering, etc matter a lot more than battlefield conflict.
4) Battlefield conflict doctrine is constantly updated, hence the army's procurement of UAVs. However, the military has no interest in a UKR like conflict. It is awful game theory. Their goal is an asymmetric battle. UAVs are symmetric battle. Neither side will have a huge advantage. It is a poor doctrine. The military has been thinking about the points you raise, but 30 years ago. And made other blocking maneuvers.
5) US military is thinking about macro issues like how food supply fits into military power. Or how Ozempic going off patent in China helps China redirect from HC cost to military.
6) I doubt China thinks it can militarily dominate the US with UAV technology as a silver bullet. Since military doctrine is mostly about creating a threat to avoid war, in what world would US capitulate out of UAV fear?
7) UAVs are more likely to end proxy conflicts. Much like 9/11 ended airplane hijackings. We were working with UKR years ago on network-centric warfare, which is one reason they have advantages in this conflict.
All of these are good points that I would summarize as “don’t over-learn lessons from Ukraine”. The Ukraine conflict has demonstrated that drones are powerful weapons in a conflict between two decrepit Soviet militaries with very outdated hardware and barely any supply of modern stealth technologies for either missiles or aircraft. This conflict should be extremely instructive for developing nations that might go to war.
But that doesn’t really describe the kind of wars the US might fight in the future. Quadcopter drones will do nothing against conventional aircraft that aren’t parked on a tarmac; they can’t move fast enough, fly high enough or travel far enough. And NATO powers in general, and the US in particular, aren’t going to be zoned out of a battlespace by air defenses of the kind that are shutting down both Ukraine and Russia from air superiority. This war would have been over in a couple of weeks had the US Air Force been sent in to support Ukraine from the beginning; Russia had (and still has) no way of dealing with that degree of conventional air power. (Whether it would have been over because Russia retreated or because they initiated the nuclear apocalypse, though, is an open question, and the fact that nobody but Putin knows the answer for sure is probably the main reason we didn’t do this.) Even if you accept their own claims about their most advanced modern air defense systems (and their claims are historically very exaggerated), they still can’t target an aircraft like an F-35 unless it’s practically right on top of them, by which time the aircraft has been in weapons range to destroy the defensive radar for over a hundred miles.
The longer-ranged attack drones that Russia and Ukraine are using are also not a new technology; we just usually call them “cruise missiles”. They were first developed in the 1940s and have been continually refined since then, but the low speed flight that makes them relatively affordable by comparison to high-speed ballistic missiles also renders them highly vulnerable to air defenses. Modern electronics have somewhat tilted the cost equation in favor of the attacker when using these kind of basic cruise missiles, but solutions to that have already been found, such as the already-deployed Advanced Precision Kill Weapon System that retrofits old 70mm unguided rockets with a basic guidance package that’s more than sufficient to seek and destroy subsonic electric-powered drones, at a cost of around $30k per shot, comparable to the price of even the cheapest long-range strike drones. The US has been using this to great effect against the Houthis, flipping the cost equation on them and forcing them to use much-more-expensive and limited-supply anti-ship missiles instead of cheap drones, which puts much more significant strain on their resources and those of their backer, Iran.
I think by far the scariest thing we’ve seen from Ukraine has been the shipping container attacks that they successfully executed against Russia. This is something genuinely new and dangerous as most military facilities are not currently equipped with the kind of point defenses needed to respond to an attack of this kind. And the ubiquity of cargo container transport in all developed nations — and the degree of blind trust afforded to globalized trade to keep the system moving — render us especially vulnerable to this kind of strike. That said, this is more of a capital investment question than a technological one. Building automated point defenses to shoot down autonomous quadcopter drone swarms is just an application of technologies we already have, not something that will require breaching new frontiers of scientific research.
Honestly, my main reason to despair for the US is simply that we’ve already succumbed to the most devastating weapon of the modern age: the twisted, destructive version of ourselves forged in the demonic hellfire of social media. The only wars the public cares about fighting are culture wars with one another, and they much prefer to keep getting cheap consumer goods from China, paying low taxes and getting fat entitlement checks to spending money on war preparations because they think no one would be crazy enough to attack us. And I think that’s going to remain the case until we suffer a catastrophic defeat, probably against China: it’ll take another Pearl Harbor to wake people up to the threat. Except this time, we won’t have the industrial base to do anything about it, and we’ll be up against an opponent that’s not going to be inclined to be a charitable victor and is more likely to want to knock us flat and keep us that way so we can’t challenge them in the future.
While I agree with many of your points, this statement stood out:
"This war would have been over in a couple of weeks had the US Air Force been sent in to support Ukraine from the beginning; "
That has been the supposition for many wars, including US wars, and hasn't borne out often, and not between major powers since the Napoleonic Wars.
In 1998, it took NATO air forces 78 days to get Serbia to agree to end hostilities. And Serbia didn't have a nuclear deterrent that would reduce US scope of action, the way it would be with Russia.
Would the war be different with US involvement? That is a certainty, but the only one. Would it be over by now? Possibly, but there are reasons to doubt. It's possible it ends because we're all dead. It's possible that Putin is prepared to let a much larger share of men of military age expire on the battlefield, which might take more than a decade. Another possibility is the war evolves into asymmetry. Another is that the US decapitates Russia, and then we have endless internecine war across much of the Russian space, including Ukraine.
Over in a couple weeks? That's the most common delusion in history, and the costliest.
These are all good points, but I would note that the situation in the first few weeks of the war bore a great deal more resemblance to Kuwait in 1991 than to Bosnia. The Russians had brought in poorly supported columns of armored units without significant logistics, anticipating an easy blitz to Kyiv and the end of the conflict within days. After the Ukrainians successfully repelled the initial wave, you had disorganized and stranded Russian troops, columns of armored units without food or fuel, and no ability to entrench or create defenses, a radically different situation from the Bosnian campaign where the enemy was defending territory they had controlled for over five years before NATO got involved. The majority of the Russian ground forces probably could have been annihilated in very short order, just as Saddam’s forces were when they were caught unprepared in Kuwait and forced to retreat.
The situation today is, of course, radically different. The Russians have had years to entrench and build the necessary logistics for a war of attrition. But defending against aircraft from an entrenched position is much, much easier than trying to take and entrench your position when the opposition has overwhelming air superiority. If the initial Russian forces had been destroyed and pushed back to the pre-2022 lines of conflict, they would have had a much, much harder time trying to advance further and the cost would have been astronomically higher because they wouldn’t have the use of the artillery units that were the central pillar of their offensive strategy for the first two years of the war. Obviously, no one can prove the counterfactual, but my intuition is that at the very least it would have forced a chaotic retreat from Ukraine similar to what we saw in the Kharkiv offensive, but much more sweeping, and Russia would have been very quickly pushed back to the pre-invasion lines and needed a significant amount of time to regroup and rebuild before attempting to restart the conflict.
But of course, the most unpredictable element here is the fact that Russia is ruled by one man who likes to rattle the nuclear saber when threatened, and a 1991-style wipeout of his invasion force, representing the majority of his available ground forces, would be a threat unlike anything Russia has faced since WW2. This would likely push him into a desperation mindset rather than the attritional mindset he’s employing now, confident that his individual will to win will significantly outlast the fickle, self-absorbed voting publics that ultimately rule the NATO states. Desperate people and nuclear weapons are a very dangerous combination, and the “escalate to de-escalate” strategy of employing tactical nukes against Ukrainian targets as a signal of willingness to go all the way to strategic weapons may well have been his choice of response.
I appreciate your thoughts here. I don't doubt that Ukraine would be in much better shape today if the US had decided upon providing air superiority from the outset.
Mine are that Russia made the classic mistake of thinking the war would be over fast, because it expected things to go as it planned. You proposed the same mistake for the US (don't worry - you have plenty of historical company). Events almost never go to plan, and never when one or more belligerents have unlimited objectives, as Russia does.
In play is misunderstanding or at least ignoring Russian goals and rationales. Destroying Russia's assets in Ukraine and pushing them out of Ukraine is likely not enough to end the war. Ending the war would likely either involve an invasion NATO is not capable of currently, or decapitation, both of which are spectacularly unwise.
In short, the war won't be over until Russia's goals change, and changing them is a long term effort.
Appreciate the thoughtful response. I think the core disagreement came down to the way we were each using the word “war”. When I said the war would be over in a few weeks, I was speaking of the immediate hot conflict, not the lower-intensity hybrid warfare Russia has been waging for a long time. I completely agree with you that Russia will continue pursuing hostilities through whatever avenues it deems feasible for as long as the current leadership and their ideological fellow travelers — believers that sovereignty over the maximalist Russian Empire is their divine birthright — remain in power. (Which, given history, will likely be for as long as distinguishably Russian culture continues to exist. Fortunately for the world, alcohol and low birth rates are slowly but surely fixing that problem without any assistance from the rest of us. Putin’s even kindly accelerating the process with his military misadventures.)
I don’t think, though, that Russia would have had a ton of options for continuing high-intensity conflict after a chaotic unplanned retreat under heavy fire from US airpower. Moving more ground troops into Ukraine would have been infeasible with the enemy holding air superiority. Sitting on the Russian side of the border and lobbing long-range ordnance into Ukraine would probably have been infeasible as well, working under the assumption that a US that had the political will to directly intervene in the conflict to protect Ukraine would probably not draw an artificial line at striking targets on Russian soil, even if it perhaps drew an arbitrary line at sending manned aircraft into internationally-acknowledged Russian airspace. (Even under our current, incredibly timid and cowardly approach, we’re still allowing the Ukrainians to use NATO-provided weaponry to strike targets on Russian soil, after all.)
The main things that initially allowed Russia asymmetrical strike capability against Ukraine were their larger supply of longer-range guided ordnance and their fleet of bombers capable of taking off from airbases Ukraine couldn’t strike and launching from high altitudes far behind the line of battle where Ukrainian ground-based air defenses weren’t a threat. Neither of these factors would be a significant impediment to the US, though: we have plenty of accurate long-range guided ordnance too, and plenty of capability to deliver it to anywhere in Russia, meaning that no staging point would be safe from retaliation. (Any Russian aircraft getting too close to the border would be highly vulnerable to combat air patrols armed with long-range air-to-air missiles, too, severely limiting the ordnance they could use for strikes on Ukrainian territory: no cheap glide bombs, only expensive ballistic missiles.) We’re already seeing significant shifts in Russian strategy forced by Ukraine’s development of the no-really-we-swear-it’s-not-British FP-5 Flamingo long-range cruise missile; imagine if they’d had similar capability from day one.
My general assessment is that in this scenario, assuming Putin wasn’t willing to uncork the nuclear genie — and I’m quite sure his good buddy Xi would be pulling out every bit of leverage he had to ensure he didn’t, given that China is by far the country that has the most to lose if the nuclear taboo is broken — his only real option would have been to retreat, call a cease-fire and then let things settle down while he rebuilt his forces and prepared for round two. But the task in front of him would be very daunting: he couldn’t just rebuild forces sufficient to fight the Ukrainian military; he’d need forces sufficient to fight NATO as a whole. This is likely out of reach of Russia’s economy even on a total war footing unless the NATO states believed the threat was completely gone and returned to a 1990s peacetime disarmament mindset. And while I would never underestimate democratic governments’ ability to ignore the obvious and bury their heads in the sand, the European response to the Ukraine invasion suggests that the invasion woke them up fairly quickly and they’d be unlikely to go back to sleep if intelligence indicated Putin was stockpiling weapons to try again in a few years.
I disagree that US ordinance stockpiles are great. One of the war's surprises is that excepting reserves, there are some shortages, with production ramps lagging far behind what was previously assumed possible. Usage would have soared with actual US engagement.
I think your assessment that the Europeans were woken up quickly is not correct. Of course, again definitions matter. But for comparison, it didn't take years to mobilize for WW1 or WW2, yet today Western Europe remains still far from awake nearly 4 years into the wide-scale invasion. Some like Spain determinedly so. I have few doubts that they wouldn't fall entirely asleep if Putin desists to stockpile weapons. (Are Eastern Europeans more awake - oh yes!)
Europe hasn't even attempted to onshore all drone materials fabrication, or opened a single mine for this purpose.
I think what's interesting about these takes is that we may get a chance to actually test these theories, if the "Russia is going to start a war with NATO" predictions are actually true.
That theory can't be tested in the future, because Russia has already been prosecuting a low-intensity hybrid war against NATO since at least 2008.
So the question isn't whether Russia is going to start a war, but whether NATO is going to fight one. The answer appears to be 'no'.
Russia is currently increasing the intensity of conflict, but so gradually or incrementally it is (intentionally) difficult to draw a line. That line so far isn't testing defences, such as drone incursions in Poland or infantry incursions into the Baltics.
How long did it take Israel to force Iran to capitulate? And they have approximately the same level of military technology as Russia; Russia is using Iranian drone specs, not vice-versa, for instance.
This is year 46 and counting. You're referring to one of dozens of confrontations in a conflict that has raged since 1980. Iran lost control of its skies, but it didn't capitulate, cede anything, or abandon its goals. Iran has never looked weaker, but the conflict continues.
You should ask yourself: "What are Russia's goals, and why?" The answer will tell you why the US cannot fight and win a war with them "in a couple weeks".
FWIW, I'm all for bleeding the bear, except that it serves to strengthen China.
I mean, I like to think I'm clear-eyed about the limits of air superiority. Can it stop an enemy advance? Certainly. Can it dislodge troops? Unless they've had significant time to dig in, yes. Can it dislodge troops in population centers, or pacify civilian populations? Most likely not. So I guess it comes down to your win conditions in Ukraine; whether the war would be over without Ukraine retaking Donbas & Crimea or whether it would be over if Russia could not support any new advances.
PGMs would have a field day on the way Russia positioned. UKR was killing Generals with our coordination. Russia has not been running active operations like Iran. They are rusty. The command structure is rusty.
It is correct Russia had other options, but even going nuclear is questionable with their fleet. And the calculus of "How reliable is this" fits into their thinking of response. It would have triggered many other issues, but from a doctrine standpoint, UAVs would have done little in the geopolitical context. And that is my point. A JDAM matched with our C4ISR would wipe out their command and control, and many other capabilities quickly.
Fair enough. The thing is, Russia would likely just sit outside Ukraine and lob munitions, unless the US is going to go deep into Russia. And then Russia will wait for the US to lose interest. For the Ukrainians, that's obviously preferable, but it doesn't solve anything for anyone else.
China could easily make us capitulate in a wartime environment, no?
1. Immediately ban all exports and watch our war production (and everything else, eventually) seize up. Shut down pharma exports.
2. Flip all the "destroy critical infrastructure" switches they've set up. Make power plants blow up, take out telecommunications and power grids across the country, etc.
3. Do 100 distributed drone attacks out of shipping containers and trucks - same thing Ukraine did to blow up Russian bombers within their borders.
I don't get the sense we have any similar tools to fight them with. We'd have to use actually, long-range weapons, which seems like something we wouldn't do even in the scenarios above, for MAD reasons.
Modern geopolitical conflict is multi-domain. UAVs are a tiny part of the story. Economic forces are a powerful component of the calculus.
This reminds me that most people seem to have little comprehension of how forward-thinking the military is on these conflict scenarios. Not sure how to educate people, but if you look at some of the non-classified work on war games, you get a sense of how much they think about these scenarios.
I would add that many policymakers are fairly ignorant of the matter. That recent film on Netflix about nuclear war seemed pretty weak, even though it claimed to have policy makers involved.
Yeah, considering that was by Kathryn Bigelow, I was pretty disappointed. Some fairly good performances, but the writing was weak.
The part I really don't like in that show was the thought we would do a full response without confirmation of the initial bad actor. After one hit of a nuke, we would not have a need to respond in a use-it-or-lose-it scenario. Frankly, we would be better positioned to respond because NCA would be bunkered.
Also, the exfil of POTUS was simply stupid.
Yeah, it seemed insane to me that they would seriously consider launching "counter-attacks" against _everyone_, because they couldn't tell who the responsible party was, based on a single strike, with no evidence yet of any more incoming weapons.
The other things the enemy knows, but movie producers do not, is the quality of their equipment. Russian tanks have poor metallurgy. UAVs fail. Bombs do not blow up. Nuclear launch systems fail.
The US has been operating at a very high tempo, working through defects. Others have not. The Russian nuke fleet probably has 25% chance of getting to the target. If you are planning an attack and only 25% are even capable of getting to the target, do you really think your strategy wins?
Also the notion of a system-wide successful cyber attack on military infrastructure is a stretch. These are moated systems with limited ability to test an attack.
It is obviously more likely that China would go after commercial infrastructure.
"Leave the World Behind" is a more interesting first strike. It has its own problems, but it is no surprise that Obama produced it. He would know the threat scenarios much better than many.
Ha ok, so you’re just saying UAVs alone won’t make the US capitulate to China, but agree that China has a suite of tools that could make us capitulate.
Not reassuring!
UAVs are a tiny part of this.
China needs us as much as we need them - so neither is eager for a way. China does want to scare our population so we spend money we may not have. China would rather spend money on growth than defense.
China and we are running wargames on economic warfare, and other domains. Noad is on a side issue that could become bigger, but people who have fought wars for the last 30 years understand the balancing nature. The US has controlled battle domains for 30+ years and has tremendous practical experience that others lack. Others, including China.
This is true, but keep in mind that the Chinese will be checked by the same fears that prevented the US from going all-out against Russia: it is extremely dangerous to land a knockout punch to a heavily-armed nuclear power, because they may decide that if they’re going down, they’re taking you with them. That will remain the case and force some degree of caution with massive conventional overmatch until someone develops a perfectly reliable countermeasure to modern nuclear missiles, one reliable enough that national leaders are willing to gamble the entire existence of their nation on it.
Yes, but that’s true of Russia and other nuclear powers too.
In a non-MAD domain, standard conflict domain, it means China is the primary superpower, not the US.
They could hurt us a lot more than we could hurt them in ways that don’t necessarily trigger global annihiliation.
If you read military doctrine, US and China have been thinking about economic warfare, cyber, and non-nuclear foring functions like food, as baseline warfare planning for 30 years. In 1998 when I was involved, we were more concerned about the flow of Soy than nuclear for balancing superpower status.
China has some very real vulnerabilities as well. We rely on them for technology and pharma inputs, sure, but they're a net food importer, which is rather more important. I doubt the Chinese have a "destroy critical infrastructure" switch and we don't, and the only country that's successfully deployed a distributed drone attack is a US ally. And their population is dense enough that one JDAM on the Three Gorges Dam would probably be the deadliest strike in military history.
Underneath drone and AI technology are huge amounts of metals and minerals, which China largely controls now. That’s a choke point the Western allies are finally waking up to. One tiny example: Alcoa is testing ways to extract gallium from bauxite at a facility in Western Australia. The U.S., Japan, and Australia governments are funding this project, with capital and off-take agreements. A good sign, IMO. But we are way behind China in terms of controlling critical mineral resources. Rio Tinto is testing the same thing in Canada with support from Western allies.
Rare earth minerals and critical minerals aren't rare. They are just expensive to extract and process in a way that isn't ridiculously toxic.
Myanmar is the world's third largest producer of rare earths. Not because it has the third largest reserves, but because the area along the Chinese border is governed by a patchwork of revolutionary organizations, militias, and narco groups who all need money. And Chinese companies are only too happy to give it to them in exchange for mining rights.
The way they do it there is two ways.
One, open mining where they haul off the rock for processing. This is pretty toxic and expensive.
Two, they find a deposit of so called ionic clays, where the rare earth elements have saturated clay in the ground for a few thousand years, and drill several hundred holes vertically to and through the clay. Then, they drill a fee holes horizontally lower down the hill to connect with the vertical holes. Finally, they pump a washing fluid into the vertical holes. It goes through, dissolves the clays, and washes out through the horizontal holes into a collection pond. The pond is then dried and the sediment shipped to China for processing...
Method two is "How to poison groundwater 101"
But it is cheap.
We are in a mineral arms race now. Mining will continue with or without the Western allies. The iPhone I’m using to type this message is filled with critical minerals from mines I will never see. Tim Cook understands this. He knows China’s supply chains better than most experts.
Tim Cook's Apple sent money and industrial equipment to China worth more than twice the Marshal plan in inflation adjusted terms. He's kind of a negligent traitor in the sense that he did more to indusrialize the United states's greatest rival than any other American.
Iran has good drone tech and lots of drones. Didn’t help at all against Israeli and American attacks from the air. Eventually, of course, drones will be fighting drones in the air, but conventional air superiority over the battlefront is still highly effective now. Ukraine is really a demo of what happens when neither side has an airforce capable enough to suppress air defenses (in part because American AWACS are able to operate safely from behind NATO lines). With your opponent having air superiority, dug in front lines would be insanity, and the infrastructure necessary to transport and supply drones and armor and artillery would be at risk of destruction.
Dispersed guerilla forces with small drones able to launch terrorist type attacks from a distance or launch a massed attack against a military target will still be impossible to bomb into submission, of course, but this would be very different than what we see in Ukraine (more like the USMC’s new strategy)
The connections you draw are interesting. Perhaps mass-training of AIs by observing people playing video games would be an adaptation to move society toward readiness for drone warfare.
You were prescient to call quadcopter drones as a military technology of the future. However, I'd advise you to not be caught fighting the last war. In the 2010s, for instance, the cutting edge of military technology was the fixed-wing loitering drone; after the USAF Predator drone was used to great effect in the GWOT, the Bayraktar was the decisive technology in regional conflicts in Kurdistan and Nagorno-Karabakh. Who else remembers the "People's Bayraktar" campaigns to buy Ukraine these drones? Of course, when used against a modern military they proved worthless; slow, high visible and easy to shoot down by basically any anti-air platform.
This is how I feel about quadcopter drones today. Obviously they're being used to great effect in Ukraine, where the frontline is fixed, approximate enemy positions are known and relatively close, civilian casualties are acceptable, and supply lines are secure. Remove any one of those, though, and their usefulness plummets. Plus, as you note we are one technological innovation away from making them a sidenote in the next war - anti-drone technology to shoot down any fast-moving flying object seems relatively straightforward, especially compared to handwavy AI-piloted drones that will only kill the enemy. And as other commenters note I don't see any use for them in a potential naval conflict for Taiwan.
How much of the global success of democracy and participatory government was contingent on the dominance of the infantryman, and what does his eclipse mean for a country like the US founded by and for the firearm-wielding citizen soldier?
Thanks for the essay. As noted in the comments below unclear if drones are the be-all, end-all or whether the three eras you noted perfectly fit...nevertheless, your point that AI, electric stacking and other technological developments are upending and will continue to upend how nations battle, seems to me, totally on point. Whether they are the cause of effect of societal change, however, is unclear to me. But they certainly go hand in glove.
Very depressing read for me to start my Thursday (but a worthwhile read nevertheless.)
The domestic use of drones for population control scares me, or the use by terrorist groups. One of my dystopian predictions is that within our lifetimes, a drone bunker, anti-drone drones, or point lasers may be things we have installed in our communities and/or our homes.
As you noted, the broader issue is the acceleration of technological change. Won't that become highly destabilizing as perceptions of unknowable adversary capabilities increases? I'm reminded of Nick Bostrom's Vulnerable World Hypothesis. Not only are we pulling balls from his metaphoric urn faster, but the destabilizing FOMO, potentially seen as 'strike 1st or lose/die', becomes a major geopol risk itself.