War in Ukraine V: Alternative Hypotheses

Peter Turchin


Join 36.7K other subscribers

One thing I realized since I posted my blog series on the War in Ukraine (last one here) in July is that previously I explicitly addressed only one hypothesis (which predicts a win for Russia). But what I aim at is a scientific prediction (see Scientific Prediction ≠ Prophecy) and good science requires testing multiple hypotheses against each other. Fortunately, it is not difficult to find a variety of views on the conflict, making opposite predictions. What I am going to do here, then, is focus on two very clear formulations of the rival hypotheses, made by other people than myself. Both predictions were made in January 2023, when it became clear that the conflict had become a war of attrition, as indicated by the flat curve since Month 10 of the war (November 2022):

Percent of Ukrainian territory controlled by Russia. X-axis: months since the beginning of the war (1 = February 2022). Data source.

Both predictions also explicitly invoke the role of mathematics, but come to diametrically opposite conclusions. In other words, this is precisely the situation we want to have in order to do “strong inference” in science.

First, the hypothesis that provided the main basis for the previous series, let’s call it the Casualty Rates hypothesis. This is the view held by group of retired intelligence and military officers, such as Douglas Macgregor, Andrei Martyanov, Ray McGovern, Larry Johnson (and more recently by the IR theorist John Mearsheimer) who tend to be critical of the current US policy. Many of them have aired their views on the podcast Judging Freedom. One particularly clear statement comes from Scott Ritter’s article, “2023 Outlook for Ukraine” published on Jan. 11, 2023. Here is the relevant excerpt:

With the battlelines currently stabilized, the question of where the war goes from here comes down to basic military math—in short, a causal relationship between two basic equations revolving around burn rates (how quickly losses are sustained) versus replenishment rates (how quickly such losses can be replaced.) The calculus bodes ill for Ukraine.

Neither NATO nor the United States appear able to sustain the quantity of weapons that have been delivered to Ukraine, which enabled the successful fall counteroffensives against the Russians.

This equipment has largely been destroyed, and despite Ukraine’s insistence on its need for more tanks, armored fighting vehicles, artillery and air defense, and while new military aid appears to be forthcoming, it will be late to the battle and in insufficient quantities to have a game-winning impact on the battlefield.

Likewise, the casualty rates sustained by Ukraine, which at times reach more than 1,000 men per day, far exceed its ability to mobilize and train replacements.

The second prediction was based on a comparison of the economic fundamentals characterizing the adversaries, in which Russia (with the GDP of $2 trillion) was contending with the combined economic power of the West ($40 trillion, adding together the GDPs of the USA and EU). In his New York Times column, published Jan. 6, 2023, the Nobelist Paul Krugman argued that this 20-fold differential in economic power provided a decisive advantage to Ukraine:

Once Ukraine had beaten off the initial attack and the invasion became a war of attrition, it also ceased to be a simple war between Russia and Ukraine. It’s true that on the Ukrainian side, Ukrainians are doing all the fighting and dying. But they haven’t had to rely on their own military-industrial base.

What this means is that productive capacity — ultimately, economic power — tends to be decisive in a war of attrition. And Russia is just hugely outclassed by that measure

the brutal slogging match may continue for a very long time.

But this is, as I said, largely about math. And the arithmetic, incredibly, seems to favor Ukraine.

What Ukraine Teaches Us About Power

Thus, we have two hypotheses predicting diametrically opposed outcomes, one emphasizing Casualty Rates, the other emphasizing Economic Power.

In a forthcoming post I will show that both hypotheses can be dealt with within a single mathematical framework.

Notify of
Most Voted
Newest Oldest
Inline Feedbacks
View all comments

Now, do the same mathematical analysis of Macedon’s war against Persia.

Peter Mott

Prior to such analysis it must be made clear what victory and defeat consist in for the two sides. Few people can do this!

John Malcolm

Nobelist Paul Krugman argued, after Trump was elected in 2016, that the stock market drop greeting the news would be sustained a very long time. It turned out to be an excellent time to buy stocks!


GDP at purchasing power parity is a better proxy for production capacity than nominal GDP (though neither is perfect).

Roger J Cooper

Wars of attrition are about more than the arithmetic of GDP’s. The will to fight is a key component.

Both Russia and Ukraine are showing signs of stress from manpower losses.

The Ukranians have already had a scandal concerning the sale of draft exemptions resulting in Zelensky’s firing of regional draft directors.

The Russians seem even more stressed. They are using measures such as recruiting from prisons and drafting middle-aged men (normally at last resort, but they started in the first year). There has been a large-scale mutiny. And now protests by the wives and mothers of soldiers.

The wild card here is American politics. The Ukraine war has become caught up in the American right’s idea of being against whatever the other guy is for.

In any case, you can’t just add GDP’s. North Vietnam defeated the US in war of attrition.

Paul McVinney

Despite many differences of opinion on the nature of the Vietnam War, I come down on the side that it was mainly a North Vietnamese insurgency, a war for South Vietnamese ‘hearts and minds’ where the will to fight was paramount. (The US leadership didn’t understand the nature of the war and pursued a WW2-style conventional war strategy.) I view the current Ukraine war as a conventional war for territory where the will to fight is secondary to manpower and industrial factors. Totally agree that the will to fight can be an important factor but I believe the nature of conventional war still lends itself broadly to analysis in the terms presented above.

Last edited 2 months ago by Paul McVinney

The Ukraine war, being a completely textbook example of an anti-colonial war, is a hearts-and-minds war where the will to fight is paramount.

Only with vast technological differentials or monumental population differentials can a colonizing power win a colonial/anti-colonial war against an organized, unified target. (Both happened in the early colonial period, but even so, divide-and-conquer was far more common.)

Ukraine is organized and unified, and has an objectively technologically-superior military to Russia. The population differential is small (only 3:1 in favor of Russia). Russia doesn’t have a *chance* and the main reason is the hearts-and-minds factor. The repeated escalation of crimes-against-humanity and the immiseration of people in occupied Donbass shifted Ukrainian public opinion from having a large minority which thought Russia was OK to having essentially nobody.

In Crimea, two young women poisoned 46 Russian soldiers, and when the Russian secret police came to their door, killed three with a machine gun, severely wounded another, and escaped. This is the *normal attitude* now in Ukraine *including* the occupied territories.

The amount of force necessary to overcome that is well beyond Russia’s capabilities, since Russia has a military where men post complaints on a regular basis that they aren’t getting fed or paid.

The situation is genuinely weird; nobody expected the Russian military to be *this bad*.

steven t johnson

North and South Vietnam were not separate countries. It was one Vietnam divided between a collaborationist regime (much of whose personnel came from Tonkin, by the way, if I remember correctly!) The shocking effect of Tet on US morale came because the army had preached they were winning, winning, winning. And US troop morale was never good because, as usual I believe, the masses were not genuinely supportive of a war of conquest. Or at least one that wasn’t cheap, quick, easy and profitable.

But there seems to me to be an opposite error here in regards to Ukraine. From my US perspective, it seem to me that Russian-language speaker/ethnic Russians and Ukrainians are closely related peoples, to the point it’s not clear it’s always useful to draw the distinction. That’s not for me to decide of course. In 2014, the Maidan coup brought to power fascists who do draw a sharp distinction between Ukrainians (and Rusyn and Tatars and Jews and Roma/Sinti and others to be taken care of later) and Russians. The first act was to attack Russian language rights. The Ukrainian army disintegrated. The Lugansk and Donetsk militias were the Ukrainian army and the emergence of parts of the Donbas into a separate entity was the consequence of this collapse. The Ukrainian fascists have targeted the civilians of the Donbas from the beginning. Thousands have died in the war long before Russia intervened on a large scale. That’s why every effort to resolve the conflict peacefully was rejected by Kyiv. Foreign support for violating Minsk 2 for instance is widely believed, correctly I think, to be key to the war continuing. Zelensky was formally elected on a peace platform but Ukraine is not a democracy in any meaningful sense of the word and he was as much planning war as Poroshenko, the openly pro-war candidate.

Roger Austin

I look forward to seeing how both hypotheses are incorporated into one mathematical framework. It would be especially interesting if the model provided some sort of estimate on how long it took to produce a “winner” and the losses both sides would suffer during that time. That might help sides decide whether to seek a peace agreement. 

Another thing perhaps worth considering in another future framework is the potential use of tactical nuclear weapons in wars of attrition.

Would artificial intelligence paired with human expertise be any help with building frameworks/being the frameworks to address the likely outcomes of conflicts? Is that even something humans want an AI to “know”?

If this and other “conflict frameworks” proved accurate, how would the frameworks themselves shape conflict, war planning, peace planning, and public opinion?

Paul McVinney

At this time, the Ukraine war is a war of attrition, making the Casualty Rates hypothesis more plausible. However, a wider variety of weapons with the necessary training can change the nature of this war to a more effective combined arms war on Ukraine’s side, leaving Russia still focused on infantry attrition. An expert opinion on this comes from this profile of US Army Brig Gen (ret) Mark Arnold who is in Ukraine advising their forces: https://www.washingtonpost.com/opinions/2023/09/04/arnold-ukraine-counteroffensive-united-states-weapons/
Arnold’s main point is that Western donations of tanks, aircraft, and other equipment, and the training to go with them have not had time to take effect. He sees Ukraine being in a better position to make significant gains in 2024. At that time, if he’s right, the Economic Power hypothesis will be more plausible.


None of these analyses are worth a damn without analyzing morale. Most wars end when one side is *unwilling* to keep fighting.

Russia’s ability to recruit manpower, and to produce materiel, are morale-limited. So are Ukraine’s. Except Ukraine has the fanatical morale of people defending their homes, while Russia has the appallingly bad morale of people being sent to die in a foreign war for no reason which makes sense.

It’s important to recall how WWI ended for Russia: the Tsar was overthrown. How WWI ended for Germany: the naval mutiny. Etc. There’s no chance of a Ukrainian morale break. And there’s a *guaranteed* Russian morale break.

Smart people said that even if Russia occupied the whole of Ukraine, they wouldn’t be able to hold onto it, with non-stop partisan warfare, sabotage, poisoning of Russian officials, etc. The Russian government, with sky-high levels of inequality, is unable to keep the heating systems working in dozens of cities, and is on the verge of social collapse. It could never hold onto Ukraine, with >95% of the population fighting it at every turn.

When Nazi Germany gave up, it was long past the point when most Germans actually believed in the Nazi cause. What would it take for Ukrainians to stop believing in the cause? At this point Russia would have to kill over 90% of Ukrainians to make that happen. Russia can’t do it — they don’t have the manpower or materiel.

By contrast, what it would it take for enough Russians to stop believing in the cause to make it impossible to recruit manpower for the Russian Army? Frankly not much. They’re practically there already.

So if you want to do an attritional analysis which is actually going to work, you have to do a *morale attrition* analysis. This is tricky but there *are* polls (despite the fact that in Russia, people will lie to the pollsters because they’re afraid the secret police will show up if they don’t say what Putin wants to hear), and you can graph them over time.

steven t johnson

It is highly doubtful that this commenter has any use for facts. There is no reason at all to think 95% of the population of Donbas wants to slaughter Moskals, given that local militias (ultimately derived from the old, pre-fascist Ukrainian army) are a large part of the fighting forces. Their morale? It’s hard to tell, the local people have been bombed by Kyiv since 2014. I believe that takes a toll. The real question in this fantasy world is, why hasn’t Russia collapsed yet?

But even skipping over this commenters uncritical acceptance of every atrocity story produced by the Telemarathon of Ukraine while twisting every story from the much more open Russian media space into proof of catastrophe, the notion that “smart people” are objectively analyzing the inability of Russia to conquer all Ukraine is grotesque. There is no reason to think Russia intends to even attempt such a thing. This so-called analysis, plus equally malicious nonsense about Russian plans to invade and conquer Europe (yes, this is exactly what is being claimed, though. The official aims of the “Special Military Operation” say nothing about annexing Ukraine. The annexation even of the Russian lands of the Donbas and Crimea was a consequence of Kyiv refusing peace, not the cause of the war which began in 2014. The standard of discourse is so low in regards to Ukraine, that this elementary fact is “forgotten,” not just inconvenient.

steven t johnson

Krugman may have a Nobel* but he’s wrong. You cannot simply add GDP of different countries, not even PPP estimates. As Prof. Turchin noted in the comment, what really matters is military production. The only way GDP would matter in the way Krugman implies is if there was a mobilization of the economy, what’s called a war economy. Then the GDP could be a proxy measure of what will ultimately be the materiel available as the economy is re-oriented to military production. Further, though Prof. Turchin hasn’t (yet?) emphasized the point, the diversity of interests prevents simply addition of EU and US GDP. Imposing higher energy costs on the EU benefits the US at the EU’s expense, a notable instance some believe is already trashing the German economy in particular. Of course, the decided imperfections in the NIPA statistics like GDP tend to treat many financial transactions as actual production.

Another aspect to the GDP argument is the fetishization of expensive weapon systems. The talk about combined arms warfare, so far as I can tell, is largely drawn from brochures from weapons manufacturers and the retired generals they hired. It is entirely unclear why anyone assumes that expensive technology guarantees victory. Unless the true notion of victory is terrorizing the world by ruining dissident countries, the US has not won its professed aims in decades. Surely this should account for something in the assessment of the GDP argument.

The conditionality of the Osipov-Lanchester/economic combined model in projecting the objective limits of attritional war assumes parity in morale. But morale is historically non-linear. Defeat is morale destroying. Loss of territory hampers recovery, compounding the lowering of morale. Victory elevates morale. Gaining territory is “merely” a negative benefit that strikes at enemy replacements of men and materiel, thus attritional gains can be popularly perceived as not-victory, so that attritional “victory” can still be a political problem. As a US citizen, the devastating effects on morale of Grant’s successful attrition of Lee’s army in the Overland campaign convinced Lincoln that he would fail re-election. Fortunately, a popularly perceivable victory in the capture of Atlanta suddenly boosted morale.

But as this historical example also implies, the question is, whose morale? The morale of the northern people mattered because there was an election. The morale of the Ukrainian people doesn’t matter because the core of the Ukrainian state, the armed forces, includes a fascist component, complete with extensive international connections, openly with other foreign militaries and illicitly too I have no doubt, with other fascist sympathizers in US and EU military officer corps. In the political/economic arena, it is entirely unclear that “Ukraine” is calling the shots, rather than the paymasters. The morale of those paymasters is what matters. On the Russian side, the problem with the US/EU goal of overthrowing Putin and installing another shamelessly Yeltsinite criminal assault on the people of Russia is that it is highly unlikely that the masses of the people want another Yeltsin. Personally I think the popular image of Putin as the reverse of Yeltsin, rather than the sober and cautious and supposedly competent version of Yeltsin, is a carefully cultivated myth. But I’m pretty sure they don’t want another Yeltsin. Any likely oligarch friendly replacement will probably be aggressively nationalist and even more jingoistic than Putin. There are some segments of Russian society that see a bright future from copying and serving the US and the EU. But I suspect they are largely counter-elites, in Prof. Turchin’s terminology, seeking support from outside the country, not from the population at large.

*Technically, the Swedish central bank prize. So far as I understand it, Krugman won for mathematical theorems designed to prove international division of labor based on comparative advantage was Good. This is more or less regarded as proving that the current state of affairs in world economy is Good. And there is no such thing as imperialism, at least not since decolonization. I am not certain Krugman should be deemed a sound thinker rather than a useful one.


Are you sure the GDP data are correct? There might be a possibility that US and EU data are as precise as GDP data published by the UDSSR?

  1. Home
  2. /
  3. Cliodynamica
  4. /
  5. Regular Posts
  6. /
  7. War in Ukraine V:...

© Peter Turchin 2023 All rights reserved

Privacy Policy