The following is a rough transcript which has not been revised by The Jim Rutt Show or Samo Burja. Please check with us before using any quotations from this transcript. Thank you.
Jim: Today’s guest is Samo Burja. Samo’s a returning guest. He’s been on several times. One of my favorite people to talk to, a real thinker. Samo is the founder of Bismarck Analysis, a consulting and publishing firm that investigates the political and institutional landscape of society. He’s also a research fellow at the Long Now Foundation where he studies how institutions can endure for centuries and even millennia. He’s also a senior research fellow in political science at the Foresight Institute where he advises on how institutions can shape the future of technology. Welcome back, Samo.
Samo: It’s good to be back, Jim.
Jim: Yeah. We had a good conversation very early in the Ukraine-Russia war about what might happen, we both got it horribly wrong, oh, well. Turns out, as you mentioned, in the pregame, it takes two to tango. And so, now, from the other side, two years and a few months later, we’re going to talk today about lessons learned. What have all the military strategists, what lessons should they have taken from the Ukraine-Russia war so far? And as a amateur military historian, I’ve really enjoyed watching this. I also, of course, understand the horrible suffering particularly of the Ukrainian people. Anything bad happens to Russian people, fuck them, they deserve it for starting this goddamn war but I feel terrible about what’s happened to the Ukrainians.
And, in fact, for those who are listening to this, if you want to hear a more blow-by-blow description of the first year of the war, I just recorded in the previous episode, Yaroslav Trofimov’s very interesting book, Our Enemies Will Vanish, where he basically traveled all around Ukraine while the war was happening up close and personal but he also talked to the leaders and things of that sort. So, very interesting, it’s an interesting contrast with today’s conversation. So, let’s start out big picture. From your perspective as a leading strategist, what have you learned from the last two plus years?
Samo: I think that military stockpiles, that is stockpiles of various forms of military equipment, are much less useful than previously assumed and there are two reasons. The first reason is that the rate of attrition in the drone intensified warfare is very high for anything from tanks to trucks, anything to artillery shells, artillery has returned in a positional warfare role which demands huge quantities of shells. It’s impossible to stockpile for the next war, the next war will be waged by an even more transformed technological set of tools. So, you don’t even know what you’re supposed to stockpile in advance.
Jim: Yeah, I think that’s the key point. Having stockpiles of big old tanks, well, that turned that would be an error because what we really needed, and we’ve gone through such rapid evolution of this, is Ukrainians are now producing 80,000 suicide drones a month. What the fuck, right?
Samo: That is a massive scaling up of production that no one would’ve thought possible a few years ago, this would’ve been hailed as a massive revolution in warfare. And if anything, right now, I think decision makers around the world are downplaying it. I think they are underselling what a transformation of warfare this is probably because almost in all countries, be it Russia, be it China, but also be it the United States, no one wants to tell their buddies at big old military corp that is doing business with the government that their whole product line is obsolete and that, probably, if we were thinking truly in national security, we wouldn’t be doing business with you at all, we wouldn’t be buying these obsolete products but we still are.
Jim: We put out functional-based RFPs for 90 days and then you’d have 90 days to deliver and we’d cycle fast because I think that’s one of the … To your point about stockpile, the curve is now so steep on the cutting edge that being able to envision what you need two years from now is getting well-nigh impossible. I’m going to throw out another thought for you, I’d love to get your reaction to it. I think a big eye-opener, for me at least and I presume for the Russians as well, is that most of the fighting since World War II has frankly been one side considerably more sophisticated than the other. And this has really been the first peer or near peer level major war since World War II. I suppose you could say Iraq-Iran but that was mid-tier peer level.
Samo: These were not countries at the cutting edge of either strategy, technology or power. Iran and Iraq were not innovating.
Jim: Yeah, against each other, they were just slugging it out. Of course, Iran had American and Iraq … Anyway, neither here or there. But anyway, this was a near-peer, near-top-of-the-stack war and a lot of the lazy assumptions that the stronger powers had fallen into turned out just not to be true. For instance, I did a little research before this call, a fair bit of us munitions rely upon GPS guidance. Yeah, you want to pound the shit out of Iraq, sure, not a goddamn thing they can do about it but a very technologically advanced country like Russia, relatively quickly, figured out how to jam GPS so the GPS stuff’s now useless. The small HIMARS rockets are almost useless unless you catch them at a fledge area and one of the reasons why we’ve given them the big missiles because they use an inertial system, they don’t use GPS.
I think the Russians also probably took something like that lesson, sophisticated pound on the little guys because, as it came up with Yaroslav the other day, at the actual start of the war, the Russians were amazingly under-resourced. They only threw about 200,000 guys across a seven or 800-mile front and the Ukrainian army was of about similar size. And so, they must have been assuming either that they would do a decapitation that would work, which it didn’t, or there would be being greeted by turncoats which didn’t happen. But when it actually came to fighting, they had no fucking chance, 200,000 against 200,000 across this huge arc of the whole Ukrainian border from north all the way down to almost to Odessa. Of course, that wasn’t going to work unless they were somehow, in their own mind, thinking this was US versus Iraq or something like that. But peer to peer is a very different game than advanced country pounding on not advanced country. What do you think about that?
Samo: I think that there were four big factors that caused a miscalculation on the Russian side. The first one is not even a miscalculation. The gamble was that the president and leadership of Ukraine would simply flee if the country was under serious threat. This was, in the immediate aftermath, people forget of the collapse of the Afghanistan government. So, the Russian perspective was this is an American puppet regime, they have no deep well of support in Ukraine, they know it, as soon as they start losing territory, they’re not going to know how much we’ve committed, they’re not going to know if it’s 200,000 or if we’re going to go in into a total war and the president will just flee the country.
And that is, in a way, a gamble, that’s turned out to not be Zelensky but this thing has happened in the post-Soviet space. Presidents have fled countries with unrest or with war and have usually lost power and, whatever government replaces them after, well, it wouldn’t surrender to the Russians but it might actually become a Russian client state again. So, I think they actually came in with a goal, perhaps similar to that of the United States entering Iraq, I don’t think the … The US didn’t intend to annex Iraq, it just intended to install a friendly government, which Americans would say was a democracy, that failed ultimately even if it was a military success and I think Russia was aiming for something similar.
Now, the next factor that I think is relevant is that they greatly overestimated mobile warfare which I think everyone did. Everyone’s thought that, if you can wage a 20th century war with tanks, surely, you can do it on the planes of Ukraine. That terrain saw some of the biggest tank warfare ever and it might actually be the biggest tank warfare ever. There might never be a tank battle greater than those that occurred in the 20th century, that might just be an obsolete vehicle.
Jim: Yeah, the famous Battle of Kursk.
Samo: Exactly.
Jim: Yeah, Kursk was the most amazing, complicated, gigantic … There were more tanks in that battle than I think exist on earth today. It was 18,000 tanks, something crazy like that, in one battle.
Samo: That kind of fighting is only possible on the wide open planes of Ukraine. Even in World War II, you would never see a battle like that in the hills and mountains of Italy, you wouldn’t see a battle like that in the Netherlands and so on. So, the terrain seemed as favorable as it possibly could be for tank warfare and I think they greatly overestimated how well this would work. And just as a reminder, in the very early stages of the war, the Russians did occupy a lot of territory, territory that the Ukrainians retook.
But for the first few weeks when it was armored warfare, they did surprisingly well even though the forces were matched and that was a big mistake. Historically, the forces have to be much greater on the attacking side even when technology had not changed, when tank warfare was in its prime. We should acknowledge that there are significant parts of Ukraine that remain occupied because of that initial phase of mobile warfare that has given way to this technological transformation that reverts it to something like a positional warfare.
Jim: Yeah. Before you go on to your other two points, let me make a very key point here and this is a big wave of history that has happened again and again is the switch between offensive dominance and defensive dominance and it’s often driven by technology and sometimes by surprisingly modest technology. For instance, the classic example was the period between the Napoleonic War and the US Civil War. In the Napoleonic War, basically, smoothbore muskets that were only effective to about 75 yards, it turns out, when you do the math, a frontal assault will work more often than not if you have berserker soldiers like Napoleon had. And so, he won a lot of battles on frontal assaults or oblique assaults that were frontal.
Between the Napoleon and the Civil War, and it’s worth noting that all the generals on both sides, most of them were educated at West Point and they were educated on the works of Jomini, he a Swiss strategist, not Clausewitz, Clausewitz wasn’t translated into English until later as it turns out. But Jomini, like Clausewitz, based his strategies and his calculations on Napoleon and so these guys were all taught that, under this situation, a frontal attack would more or less work. Well, between the Napoleonic era and the Civil War, the Minie ball was invented which is a new kind of bullet for muskets that has a little wedge in the back that expands the flange that allows it to dig into rifling. And a large percentage of the muzzleloaders by the Civil War were rifled, not all of them, but a large percentage. And instead of 75 yards effective range, you’re now talking 200 yards, three X approximately.
So, think of the Battle of Pickett’s Charge, every few years, every 10 years, go to Gettysburg and walk the battlefield and you go, “What the fuck were they thinking? These guys charging across this mile of open space with cannons which will take a certain number out but the thing that really wiped them out was the guys with the rifled muskets on the stone wall there, just slaughtering them from a couple of hundred yards away. By the time they got to the wall, there’s almost none of them left, a few of them jumped over the wall and the Union guys just clubbed them, basically.” And so, that little technological change meant that, in the Civil War, essentially, 100% of the time, whoever was on the tactical defensive won because the attackers would break their force on the defense which now had switched because of this one technical innovation from offense usually winning to defense, almost always winning, and then they’d counterattack and fuck them up big time.
Almost no examples of classic maneuver warfare where the person that struck first have won, which is quite interesting. There’s that example. And then, of course, that trend of defense having the advantage continued to grow until the famous World War I on the eastern front where the machine gun and rapid fire light artillery, again, made it exceedingly difficult to do any breakthrough, at least on a compact front like we had on the East, until, late in the war, the addition of tanks and maybe poison gas provided a little bit of friction. But the first three years, it was about a static a warfare as you’re likely to see. And then of course, World War II, we saw quite the opposite, we saw a return to mobile warfare with air, land, radio, tanks, all this stuff, lines moved quickly. In fact, I did some math, in World War I, the lines moved, on average, 0.3 kilometers a day, in World War II, 1.9 kilometers per day. So, basically six times as much.
Even on the Italian front, I went and did the numbers, 1.2 kilometers a day so four times as much as the western front World War I. So, we had moved back to a world of mobile warfare and, to your point, the Russians probably had learned that their military academies were in an epoch of mobile warfare and guess what? They were wrong. Drones and inexpensive anti-tank weapons said, “Ooh, that game doesn’t work so well anymore.”
Samo: Mm-hmm. And perhaps they had learned incorrectly from the US example because the US had a very successful military intervention in the Gulf War and then another very successful intervention in 2004. Successful on purely military terms, not talking here about strategic success. And both of those actually made heavy use of air power and heavy use still of tank warfare. We all remember those iconic CNN images of Iraqi tanks burning but American tanks driving into Baghdad, that happened, technically, in the 21st century but I think that was a 20th century war. Including its computer component, including the reliance on GPS, including all the information warfare stuff, people don’t realize that, in the 1980s, warfare would already have been computerized had it been a Soviet versus US war in the late 1980s.
So, whenever you heard the Pentagon talking about information warfare, they just meant adding an information layer to 20th century warfare, that’s all they talked about. However, as we talked earlier, Iraq was not at the cutting edge of anything. They had failed, really, to even defeat Iran so the US could wage an obsolete form of war on an opponent that was not a peer or not close to a peer. And here’s perhaps the third reason that … I initially said there were four big reasons for the Russian miscalculation, the third reason is, had the Russians attempted this scale of invasion in 2014, there would have been enough sympathizers and traders in the Ukrainian government that there would have been a surrender and the Ukrainian military would not have been ready.
But the Ukrainians had since 2014, to a great extent, pursued training, acquisition of modern equipment and also did a full security sweep of the Ukrainian government to find people who might be Russian agents or Russian sympathizers.
Jim: Yaroslav made that point very strongly and gave the other example which is apparently the actual nature of the Russian regime in Donbas was totally horrible and that there were lots of refugees that left Donbas and came to the Ukraine. And even though they were ethnic Russians, Russian speaking and one might’ve thought they would be sympathetic to the Russian cause, apparently uniformly described the Russian Donbas regime as hellish and horrible. And even Kharkiv, which was, again, very strongly Russian ethnic and Russian speaking, became a fortress of anti-Russia significantly due to reports on what a Russian puppet state actually looked like.
Samo: Yeah, those rump states were essentially very exploitative, in some ways, borderline totalitarian. The best analog would be you are living in a warlord era state like some of the crazy states that popped up during the Chinese Civil War at the start of the 20th century, had some truly unusual, strange but fragile regimes show up there.
Jim: Okay. So, let’s go on to your fourth point on how Ukraine did better than you and I both thought back in 2022.
Samo: Mm-hmm. Well, the fourth point is, and this was, I think, something that both Western analysts got wrong and the Russians themselves got wrong, it was the assumption that the level of corruption in the Russian military was much lower than it actually was. That the reforms that had been initiated in the early 2000s to try to streamline it, modernize it, remove, basically, people who had been corrupt and were selling off pieces of equipment, that just turned out to be [inaudible 00:18:58]. There was just so much rot in that institution that it was actually borderline embarrassing in certain ways in the first year of the war. Various supply shortages for this very 20th century style military and this very 20th century style of war, they were just missing so you couldn’t even wage that war very well for very long.
At first, it was not the Ukrainian counteroffensive that stopped them, they exhausted their own supply. They extended beyond their own supply lines after a relatively modest incursion into what is Europe’s second-largest country. If you count the European part of Russia as a European country, then Ukraine is a huge country too, it is actually a vast country. So, assuming you’re going to have much more logistical capability than you actually do because you have paper supplies, well, paper supplies make for a paper tiger.
Jim: Yeah, I remember the analysis somebody did of looking at the tires on the armored personnel carriers and the heavy trucks and such and, apparently, various corrupt supply sergeants had taken the pretty good quality Russian made tires and sold them and had bought really cheap Chinese tires to replace them. And there was apparently an absolute … Because the Russian military tires were relatively puncture proof and the cheap-ass Chinese knockoffs that these corrupt supply sergeants bought to replace them did not have that armor in the tread and there was an amazing breakdown of just flat tires. And the Russian army, a lot of its APCs are still tire driven and their trucks as well so that turned out to be … And the fuel trucks, particularly, the armor unit consumes a bodacious amount of fuel and, if you can’t get your fuel trucks right behind the lines, you’re going to be in a world of hurt.
Another one and, again, this came out of Yaroslav’s book that the Russians just did not calculate correctly is that, while the Ukraine only had a active military of 200,000, they had a massive reserve force and they were able to ramp up their guys with guns and girls with guns from 200,000 to 700,000 by July of 2022. So, after July of 2022 until quite a while after Putin did his mobilization and his draft and started shaking people out and … Prigozhin started shaking people out of the prisons, the Ukrainians actually had a significant numerical advantage on the ground. So, no surprise that the Russians just ran out of gas and were actually hunkered down after, basically, April 1st because the Ukrainians were very rapidly bringing people into the battle.
And I then went and did some research, I said, “How big is the US’s Reserve?” and I was actually surprised how small it is, according to Perplexity at least which is … I cross-checked it a little bit. But the US Army reserve is about 180,000 people, the Marine Corps reserve is bigger proportionally, it’s 100K, and the Air Force reserve is only 70K. So, across the whole bunch of them, the US Reserve force is smaller than the Ukrainian Reserve force, not in proportionate, but in absolute numbers even though the US is what, seven, eight times the population of Ukraine. And so, that suggests to me that one of the lessons that I learned from this is that having ready reserves that you can deploy within a couple of months may turn out to be really important in peer-to-peer warfare in the future and the US may be well under reserved.
Samo: Well, part of the aspect of the Ukraine War is that, for Ukraine to succeed at defense, it had to engage in a total war, in a war of mobilization and the Russians expected them to not be capable of it. It is an interesting question whether the US could mobilize and prepare a larger reserve if necessary. Could you prepare a larger reserve and could you activate it? There’s an argument that there are moves in the United States today to increase the size of the reserve and to prepare for the possibility of carrying out a draft. I don’t know if you’ve heard of the recent legislation.
Jim: Yup, just read it a couple of days ago and this time they’re going to include the ladies and they’re also proposing to make no draft registration required, make it automatic which they could easily do today when we should be thinking about that if we have to fight a peer-to-peer, extended multi-year war. Of course, there’s the other change. Most of the wars, again, the Iran-Iraq thing was a bit of an anomaly, but most of the wars since World War II have been pretty short in time duration. Even Gulf War I where Iraq still, at least at the material level, had a fairly substantial military, we whacked their ass in four days, Jesus Christ. The various Middle Eastern Israel versus the Arab Wars went a few days, couple of weeks max until this most recent snafu.
So, we may also have … Everybody may have lost focus on the fact that peer-to-peer wars could go on for years as they did in World War I, World War II, Korea, having reserves and stacks of reserves, having a ready reserve and then having an inactive reserve that maybe it might take you six months to get those guys up but having that deeper stack of reserves may actually be indispensable in fighting a years-long peer-to-peer war.
Samo: Part of the change that I have observed in strategic thinking around the world is precisely around the role of conscription. There was actually an argument in the 1990s and early 2000s that because of the profoundly professionalized nature of war, because of how complex some of this equipment is, a draft is not actually very useful. But of course, Ukraine could not have sustained its defense without having the capability to institute a wide draft. So, there are a bunch of places that had mandatory military service that abolished it in the ’90s and early 2000s. Usually men but perhaps, in the US example, in some countries such as Israel, women too would have to have some military training and go through it when they reached age because the assumption was that they might be drafted.
And because people thought that this would be small professional militaries, this wide military training for your population was seen as not necessary. So, the result of this is I think that there’ll be more mandatory service around the world and, especially in Europe, we might see many of the countries that abolished mandatory military service in the early 2000s bring it back. But also in places like Asia, I think decision makers in small states such as Singapore, Taiwan, maybe South Korea might extend the services that they require of their citizens and new countries in Asia might pick it up as well.
Jim: Yeah, of course, the US abolished its draft in the ’70s and has managed to run a relatively substantial pure professional military since and we’ve had amazingly good quality but that then comes to the question, in long term peer-to-peer, what’s the relationship between quality and quantity. If you’re kicking ass on Iraq, quality wins every time. If you’re involved in a slugfest with the Russians or, even worse, the Chinese, probably both are important and, of course, the western Europeans have taken this to an extreme. People like the UK and France have only a few hundred, 300 tanks, I think, each something, some remarkably small number. They’re good tanks, they’re as good or better than our tanks but they don’t have very many of them and, in a protracted peer-to-peer slugfest, some hybrid between World War I and World War II, they’re going to be in a world of hurt real quick.
Samo: Part of the interesting calculus here is that, if you were to fight China, you are fighting a country with a vastly larger population than that of the United States, they would just have more manpower despite their older age structure. The older age of both Ukraine and Russia is also a change, the higher average age of both Ukraine and Russia than any historical war that I know of, I don’t think countries this old have ever gone to war. I think that’s worth considering.
Jim: That’s very interesting. Both are extremely old countries and you see it on the news. You see these 55-year-old fat guys sitting there running the artillery for the Ukrainians and we know the math on the other side is almost identical so they probably have the same.
Samo: These are both countries that have an average age of 40.5 for Ukraine and, in Russia, 40.7 years. Consider that, that’s the average age and the mean, the mean. That might add interesting nuance to the data to think about but, on the most fundamental level, this might be the first war where more 40-year-olds are dying than 18-year-olds, where 50-year-olds might be dying. And you see the truck drivers, they have deep lines on their faces, and life can be hard and especially was hard in the 1990s, but these are clearly 50-year-old men driving supply trucks and 40-year-old men and 30-year-old men trying to run away from drones as drones chase them on the battlefield. And this is both sides because they’re both countries in deep demographic decline, both Russia and Ukraine, and they’re not unique in this regard in the modern world.
They’re not unique in Europe, they’re not unique in Asia and the United States will likely have a much higher age, average age, mean age and median age for every single decade for the next 50 or 60 years because US fertility has now converged to, let’s say, a good western European fertility of 1.6 kids per woman, a TFR of 1.6. A good European fertility is not a good absolute fertility, that leads to a very old society in one or two generations.
Jim: And then, of course, we have truly disastrous examples like South Korea where it’s 0.7, talk about a going out of business curve. I don’t know how the hell they’re going to be able to maintain their military but they’re going to have to be thinking-
Samo: If South Korea has a fertility of 0.7 and North Korea has a fertility, let’s say they’re lying about their official numbers of 1.8, let’s say it’s just 1.3, that actually still gives North Korea an advantage.
Jim: Over the long haul, for sure.
Samo: In the long haul advantage. Both of them go to zero eventually but one would go to zero decades or centuries earlier.
Jim: Yeah, I had a very interesting podcast with Robin Hanson about this not too long ago and we both were aware of the interesting anomaly that, if you project current curves, which of course is a horrible error, but let’s say you project current curves out 300 years, you end up with a world where a quarter of the population is Hasidic Jews, a quarter of the population is Amish and the other half of the population is everybody else.
Samo: More profound demographic transformations have happened in the history of the world. Consider farmers versus hunter-gatherers, that was a massive inversion.
Jim: Or the Mongols, at least in a big region, it wasn’t the whole world but a big, big area. Thirty, 40% of people in Central Asia are directly descended from Genghis Khan himself which is quite an impressive statistic. Again, if we think peer-to-peer, long-term slugfest, numbers are going to matter and we think, against China, you say, “Okay, where can we get men?” India, India’s bigger than China now and they sit all along the-
Samo: Has a still young population.
Jim: Very young population. Now, they don’t have a very good military tradition there but they have a pretty decent military and, if we thought of a multi-year slugfest, getting the Indians to attack from the south might be a really clever move to suck up and attrit the Chinese.
Samo: I think that any war the US might have with China would necessarily be a coalition, it would have to be the US plus several countries just from a purely logistical standpoint. You need the Japanese islands, you need the Philippines to even start thinking of waging war on the Chinese coast. The complication with India though, of course, is that the terrain and the Tibetan Plateau is a natural boundary. It’s likely that Indian troops would, if the Indians were to fight, they would probably be fighting, not actually directly on their border, but in Southeast Asia or perhaps in Central Asia if a country such as Turkmenistan or Kazakhstan sided against China.
Again, these are contested regions of the world. There’s an argument to be made that, in Central Asia, actually, it’s Chinese, Russian and American influence that contest. There are many US military bases in Central Asia, a legacy of the war on terror period, and many of these countries try to look to the United States for security because they do not want to be client states of either Russia or China. At the same time, they appreciate Chinese economic opportunities so they are becoming economically more dependent on China. And this is the tricky thing that is true of all of these countries, not just of these poor, low population density countries in Central Asia, it is true of large populous East Asian nations such as Japan.
If you think the US economy is co-dependent with China’s economy, look at Japan, look at Taiwan. Taiwan and China fighting would be disastrous for the Taiwanese economy even if no factory is blown up because all of the assets on the books of Taiwanese corporations that are in mainland China, poof, they go and smoke, they’ve been nationalized or redistributed to Chinese companies.
Jim: Let’s now switch. So, I think we had a very interesting conversation about a lot of cool things but one of the ones that’s been the biggest wick up and change is drones and how we think about drones of all sorts from small to big. When the war started, it was the … The Ukrainians had the big Turkish drones, which were pretty cool, they fired missiles, et cetera, but they were expensive, they were slow and the Russians eventually figured out how to counter them. What has happened since then? What lessons do you take about this amazingly rapid evolution in this whole new category?
Samo: I think part of the observation is that you can currently build a militarily useful drone from off the shelf electronics components. You can just take, basically, civilian cameras, civilian engines and slap an explosive on them. And if you have a secure line of communication, which is somewhat tricky, so that does require specialization, you can pilot them fairly effectively and you’re starting to see that there is a vast difference in skill between drone pilots. So, it’s starting to become apparent that drone pilot’s skill matters and they are getting to experience combat over and over and over again. There’s an interesting way in which … Of course, troops historically have always learned from war but you suffered a significant risk of death every time you entered combat. With virtual reality goggles and drones as your piloted vehicles, you can have thousands of hours of combat flight experience and almost no chance that you yourself will die meaning that you can have some very, very skilled people.
I think we’re going to see a little bit, this is always very tentative prediction wise because the technology might change and AI might just very quickly obsolete human pilots of these drones but let’s assume they don’t, I think the dynamic will be very similar to snipers where, if snipers survive, they can become extremely skilled at their trade. And you have these truly astounding kill ratios where you have a sniper that, during World War II, might kill a hundred people or 200 or 500 people, 500 soldiers, I would not be surprised at all if we start hearing about crack Ukrainian drone pilots or eventually Russian ones who keep getting better and better. And the interesting thing is there’s reasons to expect that young people are better at this, faster response times. Faster response times, higher neuroplasticity.
So, ironically, to tie it a little bit back to the age discussion, it’s an interesting question, under conditions of this technology that needs fast response times, maybe the optimal thing is to have your infantry be guys in their 30s and 40s and your drone pilots be all the guys in their 20s, so that’s funny.
Jim: Or 17.
Samo: Yeah, or 17.
Jim: 17-year-olds, right?
Samo: Yeah, exactly.
Jim: They could still be in their mother’s basements, what the fuck, right?
Samo: It’s not their mom’s basement, it’s the military’s basement, it’s the bunker.
Jim: Yeah, and they’re easy to keep happy, just give them a pizza now and then and they can go for it. That’s actually a very interesting insight that the age profile may actually become relevant at the cutting edge of drone pilotry, assuming there still is drone pilotry, at the moment, there still is. When I step back a little bit and look at this amazing change in drone warfare in one war in two years, what it reminds me of is the evolution of aviation during World War I.
In the beginning, basically, it was just for reconnaissance and then guys started dropping bombs, mortar bombs over the side, basically, and then they got to be effective enough that the other side said, “Well, we need something to chase those guys away,” so the rudimentary Sopwith Camels and stuff, anti-bomber planes came out and then the SPADs and the Fokkers came out to be anti anti-bomber planes and then you got the very rapid evolution of fighter aircraft and it went “ju-ju-ju-ju-ju” and then you got the synchronized machine gun that could shoot through the propeller and, basically, aviation was still not much more advanced than the Wright brothers in 1914.
By 1918, zoom, huge, huge change in the pressurization of that industry and the same going on with drones, obviously, with a different trajectory and one of the things that, again, maybe some people thought about but I don’t think it was in anybody’s plans. Everyone was still building these big drones, the Reapers and all that stuff but the real cutting edge is now the suicide drones, as I mentioned earlier. Ukraine is aiming to, I don’t think they’re quite there yet, but they’re getting close, to make 80,000 of those suckers a month, that’s just amazing. If they could actually trade one per Russian soldier, the Russian army would be out of business in about two months. But the war is so complicated, you can’t even get a one drone to one soldier kill ratio but you can certainly …
I’ve been reading more and more reports on how both the people and the material, the instant they start to move, the surveillance drones see them and then the suicide drones swarm them. This is really a qualitative change in how flight infantry and armor combat is likely to play out here in this next round.
Samo: Partially, the transformation of this war is a military transformation but it is also perhaps becoming an industrial transformation. Countries now know that, if you want to participate in a modern war, you just have to build your own drones. Going around your allies even, if they’re reliable allies, asking them for drones is not going to adapt the technology rapidly enough for combat conditions. The Russians themselves are now also building drones, previously they were buying Iranian ones. So, I think we’re going to see, basically, the creation of lots of lots of drone factories in both countries and this will result in rapid prototyping but it will also result in something of an industrialization.
After this war, I would not be surprised at all if, say, an independent Ukraine was actually a top world drone producer and actually sold civilian drones of a quality comparable to DJI drones from China.
Jim: Yeah, it makes sense and it may be better because they’re under very adverse situations and they have to be able to survive in much more intense environments than … I’ve got a fleet of eight DJI drones from the very beginning to the current close to state of the art, and they’re good but they’re very, very, very fragile and so I would expect that what the Ukrainians are going to learn to do is to build really good rugged stuff and, as you say, they may well be quite the industrial power.
Samo: Well, one of the interesting things, if I can go a little bit on another consequence of drone warfare.
Jim: Go ahead, yeah.
Samo: I note that the dog that didn’t bark is the very important thing here. The relevance of the Air Force, both the Ukrainian and the Russian Air Force, has actually been very muted during this war. Why? Because drones are the new Air Force. So, I think drones have, with the reconnaissance and munition delivery, essentially obsoleted older type, classical manned Air Force and that is a big deal for the United States. If you talk to people who estimate US military power, sometimes they point out that, out of the world’s top five Air Forces, actually, the US has three, three of the greatest Air Forces, it has the US Air Force and then the Marine Corps has its own Air Force and the Navy has its own Air Force and each of those is a top five in the world Air Force.
However, these are 20th century Air Forces with jet fighters that are manned, with aircraft carriers and so on. So, I’m not sure the US is looking at producing drones at the scale and cost efficiency that the Ukrainians and the Russians are producing right now.
Jim: Yeah. And I’ve read a little bit about this, apparently, as you would expect, there’s gigantic institutional/bureaucratic pressure from the Pilot Corps. “My whole career, I’m Colonel Blunderbuss, that I’m the squadron leader of a group of F-22s,” which what, they cost $300 million each, something ridiculous. “And me and my boys, we’ll defend the homeland.”
In reality, a $300 million fighter plane may just be an absurd thing in the next turn of the screw and we better get our shit together on that. Certainly the Chinese are also watching all this stuff as well because they can just tell people what to do a little bit better than we can. They may well be able to make this transition faster than we can. It’s very interesting you say the dog that did not bark, because the thing that gave the U.S. its amazing supremacy over these lesser non-peer foes was the AirLand Battle and then the JATO. The JATO, which was the thing that came after it.
A little bit surprising, the Russians haven’t been able to do anything at all like AirLand. They’re just utterly inept at it appears or they don’t have the right equipment, even though drones are more cost-effective, U.S. has shown that if you can pull off AirLand, you can actually do some pretty amazing breakthroughs against the narrow front and then fan out and then very much like the World War II style Blitzkrieg. But that’s not happening at all in the Ukraine-Russia war.
Samo: And of course, part of it might be the deficiency of the Russian Air Force, but part of it just is that I think airplanes are too easy to shoot down with modern anti-air equipment. And if you have state-of-the-art anti-air equipment, I’m not sure the U.S. planes will have better survivability than, say, Russian planes. Let’s remember in the 1990s, stealth bombers were shot down successfully by Serbian forces using 1970 Soviet technology. So 1970 Soviet technology could, in the 1990s, shoot down the stealth bombers that the U.S. was sending.
You notice that after that period, they quietly de-emphasized stealth in designs. It was a failed hypothesis. It didn’t work.
Jim: It works a bit.
Samo: Or didn’t work well enough.
Jim: At least not against peer level opponents, right?
Samo: Of course. Of course. And Serbia wasn’t a peer level opponent. Again, they were using 1970 Soviet equipment, which was considered obsolete by the Russians already then. So, I think, that Chinese anti-air defenses would be probably very lethal and the Russian ones are probably as well.
Jim: As you know, as a very serious strategist, near the heart of strategy, particularly for protracted warfare, are cost ratios. And one of the things that certainly seems to be clear is smart weapons are changing the math of asymmetric exchanges quite rapidly. Russian tank, couple of million dollars. A Ukrainian drone, thousand dollars, maybe something like that, maybe less. You have to trade 10 of those for one tank. You’re trading $10,000 for 2 million. That’s a hell of an economic exchange ratio and in a long war, that just can become decisive and then it’s going to apply in every category.
Talk about, say, your defense contractors, I hate to be in the business of building aircraft carrier combat groups right now. What do those things cost, 25 billion each or something nuts like that. And they’re just big, old fat sitting ducks against, even if you take top of the line U.S. equipment called million dollar rockets versus at the low end $500 million ships. You don’t have to have a very high hit rate to make that economic exchange ratio look pretty dreadful in a hurry. How do you see that playing out?
Samo: I think that for prestige reasons, people still, countries around the world, are still building aircraft carriers. Let’s say, China has decided to build a number of aircraft carriers. The correct historical analog is, I think, the World War II era battleship, where lots of steel, lots of equipment, lots of fuel is put into battleships that become too risky to use. You can’t sail the Yamato. You can’t sail the Bismarck into anywhere where it might be militarily useful because the risk of it sinking is too great.
And that just makes them these empty symbols of military strength. Battleships that would have been decisive in naval engagements in 1917 by 1942, ’43 are just sunk costs. They’re just sunk costs waiting to sink. And I think the aircraft carriers of the world today, it’s important for a country to say, we have the most aircraft carriers just as at the start of World War II, it was still important to say we have the most battleships because that sounds powerful. That seems powerful to people whose intuitions were built up by the previous wars.
So they have maybe a small deterrence value, but as soon as you think through something in terms of an actual war, I think, these are just pointless, pointless exercises.
Jim: Yeah. At least for peer level combat.
Samo: For peer level combat, yeah.
Jim: Yeah. Want to go kick the out of the hooties or something. Yeah.
Samo: Without aircraft carriers, say, the operations in Iraq wouldn’t have been possible. But even then, it did require the cooperation of Saudi Arabia. I think airfields would be plentiful in many of these situations and the ranges of modern aircraft, they’re just longer than the ranges of World War II or early Cold War era aircraft and have options such as air refueling and so on.
Jim: Yep, that’s absolutely right. And, again, I think this is the mindset that you got to think about peer warfare as very different than pounding on the periphery. A country like U.S. who seems to have taken over the job of the world’s policeman, they still need its aircraft battle groups to go pound the hooties or something. But you try to fight the Chinese toe to toe, those are just big fat targets that could be overwhelmed even by not a very high kill ratio rockets and still be losing 20 to 1 on the economics, maybe 100 to 1 and that’s not sustainable.
Let’s go back to drones a little bit. We’re seeing the proliferation of drones at every scale for every mission imaginable. What we haven’t yet seen is very much autonomous dronery. I know the Russians have been experimenting with it. I assume the Ukrainians have been as well. Where do you see that going and what’s the curve going to be on that?
Samo: I think that the progress of truly autonomous drones is going to be more incremental than technologists might expect. For the analog here, I would consider truly autonomous cars where arguably much of the technology had been there and most of the problems were solved in 2014. Yet, it still took until 2024, not just because of regulation, okay, it’s been legal to drive autonomous cars here in San Francisco for a while now. And since starting in March, it’s legal to charge money to passengers for driving those. But because there is a long tail of problems and situations that are not covered, when you solve 98% of the problems related to driving, there are still 2% that are stubborn and take a lot of work to get down to 1%.
It takes even more work to get down to half a percent of situations that need to be accounted for where human intervention is still necessary. So because of that, the progress has been measured in years or maybe even in one decade. I suspect, given the complexity of the battlefield, even with massive progress in other domains of artificial intelligence, I think, it’s going to be 20 years or so before we really start seeing autonomous vehicles thrive. Now, I could be wrong, but I think you’re going to see incremental automation of more and more steps of it.
The flying is easy. Computer pilots have been a staple of the industry. Autopilot for aircraft, that’s been solved for a long time. What’s tricky is understanding the battlefield. That’s difficult for human general intelligence. To be effective, you have to respond to these combat circumstances. And I think that this will be constrained by data as well. Look, traffic has rules. Sure, there are some rules in war. There are such things as war crimes. But really, most of the reality of how war is waged changes war to war to war, which makes … You can’t really train on a huge data set.
There’s not going to be a billion samples of the same drone versus drone flight. There might be a billion data points for a pedestrian crosses the sidewalk, okay, than pedestrian goes off the sidewalk in a place where there’s no crossing crosses it. You can train a machine on that. You can train a machine learning algorithm on that much better.
So, because of these reasons, the first one being that this was incremental even in a much more structured environment. And the second one, the data sets are actually going to be tiny for actual combat encounters to train actual combat situations and the data sets will keep changing as the warfare changes. Because of that, I think, human pilots will stay with us for a longer time to come. But it’s going to be remotely piloted they’re not going to be in the vehicle itself. They might not even be in the tank itself very soon. Wheeled drones work just fine.
I think we might actually see a return in mobile warfare when we see the first autonomous tanks roll on the battlefield. That might be a plausible future for the tank.
Jim: I wonder if the other forcing function though might actually force autonomous quicker than you think, which is there’s also been a back and forth very rapid evolution of electronic countermeasures attacking the communications links. We talked about earlier, the U.S. when it’s pounding on little guys can use GPS weapons. The Ruskies have now pretty much taken GPS out of the equation as useful for most situations.
And this ECM, electronic countermeasures, arms race continues, I would argue that the premium attached with truly autonomous goes up very rapidly. And so, if I were both the Russians and the Ukrainians, and by the way, the Americans and the Chinese, I would be spending like crazy to get good enough full autonomous for suicide drones. I’d also point out, as you did, you did make the model, and I have had at least three podcasts about self-driving car technology most recently with George Hots, an amazingly interesting guy who has an open source self-driving car software company, which is an amazing idea.
A thousand dollars piece of hardware, you can plug into any of 250 models of cars, run his software on it, and it does a level two autonomous drive thing. Probably do level three, but he won’t tell you that. And the other thing about it is, I guess my point is, and as he says, it’s okay, we’re down to the 99 point something level. And for killing civilians, 99.9 isn’t good enough. You need 99.999 to be good enough. Warfare, maybe you don’t need anywhere near that good. We do know that friendly fire deaths occur in every war at a higher number than people like to admit.
They don’t have to be perfect, they have to be better or as good as the alternative and more effective in some other way that a thousand dollars a shot. Even if you’re having a problem identifying decoys reliably. Let’s say, you can identify 9 out of 10 decoys appropriately, the math may still work for a thousand dollars drones to take out $2 million tanks. So, if I had to put a bet down, I’d put a small bet down, which is that autonomous hunter killer air drones will be a thing within a year because of the forcing function of electronic countermeasures making remote piloting more and more difficult.
Samo: And that we will just accept a higher rate of civilian deaths and friendly fire. I think that’s a plausible hypothesis because we tend to think about our acceptance of civilian casualties decreasing over time as a sign of moral progress, but possibly this is just a convenient story. And what has always happened is that when a new technology is effective in war and, say, increases civilian casualties, when you don’t have precision bombing, when you just have to carpet bomb a city, lo and behold, we accept much higher civilian casualties than we did even a decade or two earlier.
So I think the strategic bombing of World War II is a point in favor of the theory that we will, just in these wars, come to expect that maybe there’s a 90% chance that I as a soldier am killed by enemy drones and maybe there’s a 10% chance I’m killed by friendly drones, and that’s just how we wage war now. And then, you hide from both the enemy drones and the ally drones.
Jim: Yep, exactly. Exactly. To your point about World War II, the change of expectations, as I mentioned earlier, I studied the U.S. Civil War fairly carefully. The biggest battle ever fought on the American continent was the Battle of Gettysburg. A huge, amazing battle. How many civilians were killed in the whole Gettysburg campaign?
Samo: No, I’m curious to learn.
Jim: Zero.
Samo: Fascinating.
Jim: Zero. And the estimate of civilians caught in the whole U.S. Civil War is like a thousand. And most of that was in the burning of the Shenandoah Valley and partisan battles and things where the civilians were mixed up. But in terms of the actual military on military, essentially, no civilians were killed at all. And that was part of it was the very professional chivalrous ideas about warfare and the nature of the weapons. They weren’t long range. They weren’t huge explosives.
But by the time they got to World War II, you think about Hiroshima and Nagasaki, in reality, the Americans killed 2.5 million Japanese civilians with firebombing. The stick and paper Japanese cities of the ’30s and ’40s, they go up like that. And Curtis LeMay ran this amazing campaign where they burned the shit out of basically the top 100 Japanese cities. And they actually kept Hiroshima and Nagasaki and one other city, I forgot the third one, off the target list so they’d have something to demo the nukes with.
Samo: Right, right. And it’s fascinating that the perspective, I mean, Curtis LeMay’s own perspective was that, had he lost the war, he’d probably be tried for war crimes. That’s a quote by him.
Jim: Yeah, he should have. I mean, by the traditional measure and Dresden, that’s certainly a war crime. But on the other hand, fighting Nazis fuck up, we’ll do what we got to do and that’s just the way things go. So I think to your point that we may just reevaluate as things work. So now, let’s switch to the ground. As I understand it, the Russians have done a small amount of drone, not quite tanks yet, but like-
Samo: Scout vehicles, delivery vehicles.
Jim: ATVs with machine guns on them, things like that. And they really haven’t worked out so well yet. But what are your thoughts on that? Are there some fundamentally more difficult things about ground warfare? Or are we just on an exponential curve but we’re further behind on it?
Samo: I think there are some additional difficulties in ground warfare, but I have to note that many of the flying drones are essentially participating in something as complex as ground warfare. When you have a soldier running behind cover to hide from a drone that’s chasing it and the drones try to chase it and avoid getting shot by the soldiers and then collide with the soldiers killing them. That is as information complex as ground warfare. So there are at least some drones that are encountering flying drones are encountering comparable complexity.
Now, with regard to ground warfare, I think it’s partially also the case that the capability is currently provided by human soldiers, so it just takes longer for bureaucracies to decide to use an autonomous ground vehicle than just put a guy in a Jeep, there’s no guy you can put into a drone. So it’s the question of do we develop a totally new capability? Or do we leave that capability only to the enemy? Most militaries will decide to develop that new capability.
But then, when you compare the pluses and minuses of a human in a piloted vehicle versus an autonomous vehicle, then you’re like, oh, well maybe this is not worth the cost. Actually, our survival rates are fine. And from Brown’s words, the human brain is a supercomputer that you can produce with unskilled labor.
Jim: Exactly.
Samo: But we are now an older world, so perhaps, actually, it doesn’t even make sense to have fifty-year-old truck drivers be blown up. Why? Here in S.F., I can hop into a Waymo and it drives me to the other side of town. Surely that Waymo could deliver ammunition to the other side of town just fine. If San Francisco was in some sort of civil war, right, there’d probably be a street you’d have to manually mark on its map as, okay, this is clear. This is our green zone. The military ammo delivery Waymo can drive in the green zone, but not the enemy combat zone. Maybe you do something like that, but the technology is there.
And note, Russia does have companies like Yandex that just make solid test vehicles that have fulfilled the same autonomous driving function. So I think it’s just a matter of time before many of the deliveries in the combat zone with the vehicles have shown low survivability, those little vehicles running ammo to the troops. You might’ve seen, they even picked up this obsolete, I don’t know how to describe, it looks like a van from the 1960s, you probably know which vehicles I’m referring to, that can just race fast enough so that it doesn’t get taken out by drones. Why does that have to have a human in it? I don’t think it does.
And I think if you eliminate the human there, even if it’s remotely driven, I think that more supplies make it. So, the electronic countermeasures, like you said, remain a forcing function, but I just think that a lot of these vehicles, if they have onboard autonomous driving, would do just fine to deliver supplies to allies. That is perhaps the least complicated role, but a vital one since it is the logistics and the supplies that limit how far in advance can go or where in advance where an offensive can even begin.
Jim: Yep. That’s always critical. The logistics are, in many ways, the base upon which everything else is built. Yeah, that makes actually a fair bit of sense. I’m just curious, have you driven in a fully autonomous Waymo yet?
Samo: Yes. Yeah. I’ve taken it many times.
Jim: Oh wow. It’s almost worth flying out to California as much as I hate flying just to do that.
Samo: Well, but the fact that you’re probably going to have to fly to California is also interesting. They’re not going to start driving in your town a year from now or two years from now, maybe five years. But that suggests that they are very incremental in which territories they’re mapping.
Jim: Yeah. I’ve noticed they’ve avoided snow also. It’s interesting, we go up to Pittsburgh quite a bit. Our granddaughter lives up there and a daughter we visit there a little bit. And there were a bunch of full autonomous driving companies in Pittsburgh that were spun out of Carnegie Mellon, the famous robotics lab there. And one of their claims was, yeah, we’re going to build more robust because Pittsburgh, it’s this horribly laid out city on a pointy little triangle. It’s in Appalachia. In fact, they laughingly call Pittsburgh, the Paris of Appalachia. And so, lots of steep hills, non-square grids and snow and ice and bad weather and fog and all that.
Well, guess what? They all went out of business. There ain’t no autonomous driving cars in Pittsburgh anymore. Too hard of a problem to solve as it turned out.
Samo: California and Texas, they essentially don’t have weather. They just have temperature. There’s no-
Jim: Yeah. And of course, where Waymo first did its thing was in Mesa, Arizona, which is flat grid like and perfect weather, so it’s quite interesting how that will move along. All right, so let’s move on to the next piece, which is … We’ve talked about how the ECM, the electronic countermeasure, is such a forcing function. This was in your paper from February 23rd, 2022, which is perhaps surprisingly artillery may still seem to be the queen of battle for a while yet. What do you think about that? Because you can still throw a shitload of mass with artillery.
Samo: I think that is one of the parts of the analysis and forecast that we got right. It’s been born out as we talked about 80,000 drones, but we should also talk about tens of thousands of artillery shells that have been expanded.
Jim: A day, a day, a day tens of thousands, millions of artillery shells have been expanded.
Samo: Of course. Of course. And that rate of production is not something that the United States has prepared currently. It was not even something the Russians had prepared. The Russians thought, oh, our huge stockpiles mean we don’t have to build artillery shells. And of course, they had to then buy them from places like North Korea and so on. And the same is true of Ukraine. I think Ukraine gets South Korean artillery shells so-
Jim: And Czech-
Samo: … you could jokingly maybe the Russian Ukraine war is, in a way, Korean War by proxy.
Jim: I love it.
Samo: Both countries are deploying artillery shells and using production that would otherwise be used against each other. The City of Seoul famously, of course, as you know, in the sites of North Korean artillery and would be-
Jim: 40,000 artillery pieces sitting up on the mountains just on the other side of the border. And it’d be a hell of a shit show should that one ever go up.
Samo: Well, the thing is, the artillery shells are themselves basically dumb drones that you can’t use electronic countermeasures against. They fly. They’re hard to intercept. I think that with the end of positional warfare, the key countermeasure against artillery was always you just move fast enough that the artillery has trouble hitting you in time from a distance. That’s the key because once the artillery is in the air, its final position is fixed unless it’s a missile. So you could just move out of that area and that was the advantage often of rapid warfare.
So, as soon as you cancel mobile warfare, the advantage of artillery returns and then you can’t move into zones that are heavily shelled. And if you keep on shelling a zone, guess what, no one can go there. And that’s why for a little bit we saw even the return of trenches. There was some literal and is still some literal trench warfare happening in Ukraine today.
Jim: Yeah, think about the combination. I always like to think about layers of weapons. If you have the thousand dollar suicide drone attacking anything that moves, it means it’s difficult to use maneuver. And now everybody’s pinned down, that makes artillery more effective again. That’s quite interesting, the two together.
Samo: Also, there is a direct synergy sometimes where a surveillance drone can update live the targeting of artillery pieces and that’s a pairing both the Russians and Ukrainians use.
Jim: Yeah, hugely important. And in fact, that was one of the big advantages the Ukrainians got to first in the very early days of the war. They were able to call in their artillery. Well, I guess, one last thing on artillery, and again this gets to logistics. The Ukrainian Army, as we know, was all Soviet caliber stuff. So all their artillery was Soviet artillery. They have Soviet caliber, Soviet shells and all. And obviously, the Russians aren’t going to be nice and give the Ukrainians anymore artillery shells.
And so, the Ukrainians burned through their supply of Russian block munitions pretty quickly. And we were able to round some up from the Czechs, I think, and a few other folks that had stockpiles of them. But one of the key things, as it turned out, was just barely in time, probably the U.S. started supplying a fairly large amount and the French and the Brits and the Poles, I think, NATO caliber artillery. So there are now thousands of NATO calorie artillery pieces. But we’re draining the hell out of our stockpiles very, very rapidly because again, this peer level extended warfare was not something that was in our planning horizon.
And if I had to say, one of the stupidest things I’ve seen so far in the U.S. is why didn’t we appointed George Marshall to ramp up our artillery shell production and other basic supplies that we know we’re going to need. Now, this is a little known fact, but to at least kids today is Roosevelt appointed George Marshall in 1939, ’38 I think actually, to begin getting America ready for war. And he was the smartest general we had. And they did not use him in the battlefield. They used him to run everything else behind the scenes and it was a hugely important, brilliant move by Roosevelt to take his smartest, most capable general and basically put him in charge of grand strategy and logistics.
And that made a huge difference for our readiness once the war actually started. As far as I know, Biden has not done that, has not found a George Marshall and said, you’re in charge of accelerating our readiness here over the next couple of years.
Samo: Well, I feel that a telling thing to look at here is the trillion dollar infrastructure bill. I’ve yet to meet someone who can point to a concrete bridge or doc then we could at least call that bridge or doc that was built a trillion dollar bridge. So, I feel like the modern U.S. government would perhaps have a war readiness bill where 1 or $2 trillion are spent and somehow it just vanishes all up in the air with very few tangible results. The core problem has been that it is politically always more useful to spend money under pretext and basically hand it to your political allies than it is to achieve a deliverable.
And with regard to peer conflict and long war, the war between Ukraine and Russia is already a war of production. It’s a question of who builds more artillery shells? Who builds more artillery pieces? Who builds more drones? And sure, there are coalitions at play, but this is really no longer a globalized economy. Putin can only go to so many countries asking them for their artillery shells and drones before he’s run out of diplomatic options or given up too much.
Some have to be produced in Russia. And Ukraine can only rely on its allies to some extent because the allies will always give their own demands for how and under which conditions will they give weaponry. Sure, the allies could be more generous, but at the end of the day, Ukraine has to assemble its own drones. And I think that the United States will have to, in any prolonged war, assemble most of its own weapons. The problem is the U.S. has given away its industrial base.
Jim: Yeah. That’s a very scary thing that we have done that.
Samo: I mean, consider that essentially the U.S. is trying to be a global naval power while having very few dry docks that could build modern ships.
Jim: I think we’re like 23rd or something in ship tonnage building and we’re beaten out by amazingly small countries. It’s-
Samo: Well, but the most concerning one is that China is number one.
Jim: Yup. China.
Samo: China is building truly a vast number of civilian ships. And they’re actually building a decent number of military ships as well. So, any civilian production capability in conditions of war will be turned to military production. And perhaps the quality will be lower, but quantity has a quality of its own.
Jim: Yeah. That’s what we talked about earlier that especially if you talk about peer-to-peer war over a period of time, quantity is going to matter again. And it’s not going to be a small, very elite professional force kicking ass on Third World countries. That’s a very different thing. And if we don’t get that in our head, we’re going to be fucked. All right, so let’s hear in our last 10 minutes or so, take what we’ve learned and think about the other potential hotspot, which is the Straits of Taiwan. What do you think what’s been learned in this war and the thinking around this war that’s going to inform everything about the Straits of Taiwan situation?
Samo: Well, four factors. One is countries that might side with China in such a conflict now know that sanctions only matter if both China and the U.S. sanction you. If you are only sanctioned by the U.S. and its allies but are not sanctioned by China, your economy won’t be happy, but it will survive just fine. That was demonstrated by this war and applies directly to the Taiwan Strait crisis because there are some governments that try to balance the U.S. and China. And there are governments that might consider, for example, staying out of the conflict or even joining the Chinese side.
The second factor that I think is important is that it’s been demonstrated that an untested military cannot be relied on. So, the Chinese will probably get themselves into some smaller conflict where they can deploy an expeditionary force probably towards some kind of naval objective before they go for Taiwan. Or at least, if they’re smart, they would do that because, again, a paper tiger is not valuable in a war. Do we expect that the corruption problems in the PLA are going to be easier or less severe than they are in the Russian military? No, not at all.
The Russian military actually had stress tests for small military expeditions to places like Syria and things like the Georgian war. They couldn’t fight a scale of the Ukraine war. They didn’t have enough functionality there. But China hasn’t even done that. They’re spending a lot on their military. They’re building some good hardware. But do any of these systems actually work? By which I don’t mean the physical system, I mean the bureaucratic system. Every military is a bureaucracy. It’s a struggle to make a bureaucracy functional. Unless you test your bureaucracy, all your bureaucrats will always say that we are competent and we’re doing our job and we’re not stealing any of your money.
In reality, though, it’s not tested, they’re doing all the above. They’re not prepared. They’re pursuing careerism. They are giving contracts to their cousin who gives them a kickback. It’s a mess. And I think China’s military has good hardware. I think the U.S. underestimates the quality of Chinese military technology, however, its bureaucracy is probably a mess. Its logistical capabilities are probably a mess. They have spent a lot of time crafting military theory and very little time implementing military practice. They have no stress tests for this bureaucratic machine.
The last big war they had, it was the Korean War. What a different world, 60 years ago. The listeners might not know that the Chinese were absolutely a combatant, however, unofficially. The Korean War was partially a China versus U.S. proxy war. Something the Chinese remember, Chinese cinema actually had a very popular movie or several very popular movies in recent years valorizing Chinese troops in Korea fighting Americans. So, there is jingoistic nationalistic sentiment.
And I think a lesson learned also is that autocratic countries and countries with autocratic tendencies do not have a hard time getting their population to be on board enough with aggressive war to not be a problem. Remember all the theories that like Putin’s going to be overthrown in a wave of protests and anti-war stuff. None of that happened. And over time, Russian public opinion went from uncertain to hardening and to being more committed to the war.
And I think any situation in China would be similar. So, don’t believe anyone who says that if China wages an aggressive war, the CCP will just be overthrown from the inside. That’s not going to happen. I actually can’t think of a historical example where that happened even in the last hundred years.
Jim: Yup, even the most rotten regimes.
Samo: It wasn’t even true in the 20th century and it seems to not be true now.
Jim: Yup. Indeed. What else do we have about the Straits of Taiwan?
Samo: The recent war has reminded everyone of this. I think that the massive number of drones used, those straits are short enough that drones can fly over. So, I think, we might see a truly huge, non-amphibious invasion, but electric motor, air flight invasion. Imagine something like 300,000 drones just flying over the strait one day.
How can the Taiwanese shoot up all of them? They can’t. And then going through the streets, and at the end of the day, what prevents you from putting out a loudspeaker in Chinese saying that a curfew is now in effect and anyone leaving their home is violating martial law. And that the mainland government will be deploying human police forces within 48 hours. Sounds a bit like science fiction, but it’s really not impossible. You could do that in a city or a lighting site.
Jim: That’s a really interesting idea that could happen.
Samo: Well, if the soldier is stuck at home and can’t leave his home to go get his gear because there’s a drone-
Jim: Over his house, yeah.
Samo: … his street. Yeah. And of course that would be very aggressive. That would require a Chinese government that is willing to inflict huge civilian casualties on the Taiwanese, which I don’t think they are. But still, the military capability exists and that’s a technological capability that changes how an amphibious invasion might go. And I’ve not seen many people grapple with that, though some have begun to realize it with various alternative concepts being floated for how do you deploy troops onto an island?
Now, again, I think the Chinese will first test any such thing in an island context. I do think China will try to keep the war low on civilian casualties because the ideological premise is that this is China and that this is going to be a policing action against a rebelling province, if say, Taiwan declares independence or something like that. They will frame it as, oh, this local band of political extremists is moving away from China. But at least half of the population support reunification with China and the ones that are confused by American propaganda, well, we’ll sort them out soon enough. That’s going to be the story, but they will believe the story.
Jim: Yeah. Even though the reality is only 4% want to reunite. It’s 96 to 4 apparently is the actual ratio.
Samo: They’ll take any sort of opposition to escalation with China and equivocate that with reunification desires with China. About 50% are in favor of the status quo and not to pursue moves such as declaring independence, et cetera, et cetera.
Jim: Correct. That’s about correct. Yep. That is an interesting idea, this drone thing. I’m going to kick back on that, which is there’s an interesting exchange ratio thing here. If Chinese drones have to operate across the Straits of Taiwan at a range of 80 or 100 miles, they need to be much more powerful and much more expensive than a point interdiction drone that just needs to fly up over Taipei and shoot down everything it sees.
And so, I suspect if someone actually envisions this in time, the Taiwanese could build a counterforce at a 20th the cost per unit, and may well be able to counter it if they can envision it in time, that would be quite interesting.
Samo: Well, perhaps we will then have automated drone carriers rather than aircraft carriers.
Jim: Yeah. Yeah. Yeah. Now, let’s think about the more canonical story that gets told about Taiwan, which I have serious reasons to doubt that Chinese will actually try, which is across the straits amphibious assault. What do you think we’ve learned about from Ukraine that puts light on that scenario?
Samo: I think there’s been a relative lack of amphibious warfare in the Ukraine war, even though on paper there is a case for it. Crimea is essentially an island. And there was discussion that the Ukrainians might attempt operations to take Crimea back even before all of the other territory is liberated. That ended up not happening. And the Russians did not end up using that much amphibious operations either nothing near Odessa, et cetera, et cetera.
It seems to me that the defensive being favored in this war at a macro level goes against the Chinese attempting an invasion. I think they’ve learned that the technology favors the defensive, and that’s just another multiplier on an already difficult operation. It’s much easier to drive tanks across a border than it is to land them on the shore of an island. And all of the problems encountered in Ukraine are multiplied if attempted over the strait.
Jim: Yeah. And I think the other point that there hasn’t been any amphibious operations of consequence beyond Snake Island, a few little things like that, is the fact that smart weapons, precision, inexpensive, asymmetric exchange, has allowed Ukraine to drive the Russian Navy out of the eastern half of the Black Sea. They don’t even have a Navy or they got a couple of rowboats or something. A homemade knockoffs of the harpoon called the Neptune were enough to sink the biggest ship in the Russian Navy, at least in the Black Sea fleet.
And now they’ve had to redeploy out of Sevastopol over farther to the east because even Ukraine’s relatively small stock of anti-ship missiles make it untenable to have anything that floats anywhere near. And if I were China, I’d say, fuck the idea of trying to come across the straits. I ran the numbers, let me see if I can find them here. For less than 1% of Taiwan’s GDP, they could build 2000 harpoon equivalents per year. And you run that for a couple of years, and I even came up with a homemade strategy.
Let’s deploy them out to people’s garages and backyards and we’ll use audio signaling with big sirens to say when you should fire, which salvo of these missiles. And I looked into the harpoon and it does not use GPS. It uses radar and inertial guidance and some other things. And so, it should not be easily jammed. And if Taiwanese fired 10 salvos, 500 harpoons each at an attacking fleet, there’d be nothing left. They’d all go to the bottom probably before they got halfway across the strait.
Samo: Well, interesting thing here, however, is autonomous submersible vehicles.
Jim: That’s the other thing. The Taiwanese could do that too, very inexpensive.
Samo: But on the other hand, the Chinese could also deploy many autonomous submersible vehicles to surround, interdict and possibly blockade the island.
Jim: I do, because that’s the next scenario, which is a lot of people now have looked at the cross Taiwan and what they’ve learned from Ukraine. Remember, the people were thinking that was the scenario three years ago. But now, after watching the situation in Ukraine, I think, big learning that would be really, really, really, really dangerous for China to do. And now, it’s the interdiction strategies. But then you go, okay, the initial flavor of that was the Chinese Navy would do it. Well, wait a minute. The same analysis about aircraft carriers applies to the Chinese Navy trying to interdict Taiwan.
We just fall back and lob missiles at them, sink their fleet. And of course, the other one, it’s a traditional benefit of the U.S. is we’ve got our 45 attack subs. Once we’ve got the attack subs to Taiwan game over for the Chinese surface navy that’s a long distance away from their shore where they can use mines and support aircraft.
Samo: The strategy of using submarines to say choke off an island supplies did not work during World War II because convoys and because there were not enough submarines to do it, however, I think, autonomous submersibles might just be able to do it. Declare a red zone where every single boat that goes a certain direction from the shore of Taiwan is fair game. And then, deploy a few thousands submersible drones that have a simple AI that just tells them, hey, collide with anything that moves and keep on replenishing those. It’s self-deploying mines. It’s basically self-deploying mines.
Jim: Yeah. Attack mines that could attack, right?
Samo: Yeah. Attack mines. Yeah.
Jim: But then of course there’ll be an arms race, not that hard to build something to counter those. And you see anything under the surface that’ll water attack it. And so, we’d have another arms race, so that’s interesting. All right. This has been a very interesting conversation as I figured it would. Samo Burja from Bismarck Analysis, is that what you call your thing?
Samo: Yes.
Jim: We’ve had a very wonderful conversation, stimulating. It’s going to cause me to have even more thoughts. And I look forward to having you back on the Jim Rutt Show in the future. Thanks for coming on board.
Samo: Thanks for having me and looking forward to next time.
Jim: Yeah. This was great.