WWII & Convenient Causality

In which I can't help but respond to MartyrMade's thread

Recently Daryl Cooper (MartyrMade) was on Tucker Carlson giving his perspective on WWII, he wrote a long X thread on it as well. His perspective is one that centers Churchill as ‘a chief villain.’ There is no denying that he is well read, but while there is of course a simplified mythology to WWII, I increasingly see this very-online right somehow running with that to “Hitler: Actually kinda good?”

You might think this annoys me because I think Hitler is bad. While I do think Hitler is bad, that’s not why it annoys me. The reason it irritates me is because it’s intellectually lazy. If you like Hitler, just say “I think Churchill was the chief villain because Hitler had legitimate grievances, including the Jewish Question.” At least then you’re being honest.

Richard Hanania

also had a thread on this same point.

Often people rush to debunk these takes by countering them with their own esoteric historical knowledge. That’s fine, but not what I want to do. I instead want to discuss the specific ways in which people take an undisciplined approach to historiography to support their own beliefs. This was an easy outlet for me to do that. The purpose of this post isn’t to convince you Hitler is bad, you already know that.

Compression of history isn’t the same as “them” lying to you

To begin with, most people are incapable of having a heterodox, or even a nuanced perspective, on WWII. This is because they know basically nothing about it other than what they picked up from watching Saving Private Ryan and in their high school history class.

If you were in charge of condensing the lessons of WWII for the masses, how would you do it? You have to take this remarkably complex and catastrophic global conflict, and extract a few global lessons from it simple enough for an average student to internalize. Actually, not even average, ideally the lessons would be simple enough that even someone in the bottom 20th percentile could generally get it. What is the most important thing for citizens in a democracy to take away from it? Probably that it’s bad to round people up and kill them based on their identity, and it’s good to stop people who have done that. Also we were the good guys.

This isn’t wrong, rather it’s been compressed to an incredible degree. We have to take this topic people will spend a lifetime studying, and then teach it to someone in ten hours or less. Still, these lessons need to be taught because they’re not obvious, which is why the history of Europe, of humanity, has been atrocities beyond measure. These world wars no longer happen among developed countries.

The compression of history into a simple story has a lot of the same properties you might expect from a myth: Deep complexity distilled into a simple story and lessons. In this sense “deconstructing the myth of WWII” as they like to say, ends up basically being the same thing as just learning more history.

The tactic they take is to pick any of the untold atrocities from WWII and then telling you that they didn’t tell you the true story. The reality of WWII is that about 80 million people died in total. Since most people know so little, it’s trivial to pick a few atrocities that killed a few hundred thousand people and use them to try and make some sort of point. The reality of the time is that no world leader could sneeze without killing a hundred thousand people.

In fact, the leaders often weren’t even aware of what was going on. Curtis LeMay was a mid-30s general who ordered the initial firebombing of Tokyo when he was riffing on some new strategies. You might think there was a council of elder statesmen and the president in a war room who orchestrated and ordered every event in WWII after debating the morality of such an action, but you’d be wrong. A significant amount of WWII, like WWI, was some guy saying “What if we tried this?” then killing 100,000 people. Only unlike WWI, in WWII the civilians were often the primary targets.

Examples like this demonstrate why simply accumulating more historical facts doesn't necessarily lead to a clearer understanding of causality or moral clarity, in fact often the opposite.

There are diminishing returns to knowing history

When I was younger I had this idea that whoever has read the most books is probably right. Not only are they right, but they’ll win debates by always having a secret reference of esoteric piece of knowledge to drop, and as a result own their opponents.

I eventually found it weird though that as people became more well read, their opinions rarely changed. Sure their ability to win a debate and convince others increased, but they usually believe the same things they always have. This isn’t always true, but it is a common pattern. As David Hume noted, reason is the slave of the passions.

Practically the benefit to being more well read is you can always increment causality back one step further than the other guy. Sure it might seem like Hitler started it, but in fact he was responding to this other guy. The more historically well-read you are, the more you can construct these little vignettes to support your perspective. Another way to think about this, is the more knowledge you have, the more you can overfit your model to the data to support whatever outcome you feel is true.

What this means if you want to be correct is that the more you know, the more you need to regularize your beliefs, which is a term I’m borrowing from statistics where you prevent overfitting a model to data by ‘pulling’ it towards the average. This is a little counter-intuitive at first, because you would think the more you know the stronger your beliefs should be, and of course it should be, just probably not as strong as you think. This is also consistent with the research that generalists tend to produce better predictions than domain experts.

How do you regularize your beliefs without a computer? Unfortunately, it’s a lot harder than doing it statistically. You need to practice a detachment to the past. The more you can calm your passions, the more clearly you see into the past. The way I think of it as a sort of meditative practice: “What has happened is done. These people are not you, you cannot change the past. The past relates to the present less than we might thing amidst the chaos.” One way I also do it is I remind myself that while it may seem like going from having read 1 to 20 books is a large leap in factual gathering, the reality is the amount I’m aware of is still a pathetic fraction of all the information from the event in its totality.

We have to be a little careful here, because we don’t want to run too far with this to suggest we can dismiss everyone who is more well-read than us. Instead you have to try and see what game they are playing, and even if they know more facts than you, try to observe the structure of their argument. Let’s try to pick this one apart. I don’t know all the intricate details of who said what, or who bombed who first from which perspective. I mean I know the general history, but I haven’t read ten or a hundred books on it. However, I do have a keen interest in economic history, and I read the Economic Consequences of the Peace by Keynes a while back.

Keynes famously wrote this book in 1919, which warned of future German aggression due to the economic severity of the Treaty of Versaille. I’ve always considered that the earliest reasonable prediction of WWII for two reasons. The first is he called it in real-time without hindsight bias. The second is that he identified early on a system level failure at an abstract economic level, rather than focusing on the precise details of any individual. I personally find these economic and political abstractions more compelling, rather than an obsessively deep historical recounting of “who started it.”

The reason I bring this up is it’s an example of iterating back until you find the type of cause that you find most compelling. Why is this the cause I think is the most convincing? I’m initially trained, and my passions lie, in Political Economy. I could go and make a post, filled with facts, on why a systems level cause of WWII following WW1 is the primarily correct one, and write a 100 long thread on X about it.

See the problem here? Knowing facts, and being smart isn’t sufficient. Causality is mercurial. That’s not to say it doesn’t exist, but you can shape it and frame it in different ways to support what you feel is true, and what you value.

David Hume, an 18th-century Scottish philosopher, famously critiqued the concept of causality. He argued that we never actually observe cause and effect, only the constant conjunction of events. For instance, we see a billiard ball hit another, and the second ball moves. We assume the first caused the second to move, but Hume argues we only observe one event following another, not the 'necessary connection' between them.

Hume got close to the right answer with his critique on causality, where he noted that there is no single ‘necessary connection’ to any event, as each cause as a predecessor cause, so we’re prevented from attributing it to a single cause, and as a result we’ll craft our reason to match our own desires. That’s true, but he took it too far in discarding the idea of causality entirely.

Fast forward to the 20th century, and we find Michel Foucault, a French philosopher, offering another perspective on historical causality. Foucault's approach, which I recently wrote about, suggests that the exact cause we identify comes from the existing power structures within society. In other words, our interpretation of historical causality is shaped by current social and political forces — and we can see some of this in Daryl Cooper’s argument against the ‘myth’ of WWII.

The problem with Foucault, like most scientific critiques, is that it goes too far in abandoning the sanctity of knowledge and causality after identifying what they believe to be a fatal flaw.

I think how we deal with causality is more sanguine. While it’s true there are numerous plausible causes, the cause that’s important depends on what process we are trying to identify in the hierarchy of processes that construct our reality. If we are trying to prevent future conflicts from system and treaty levels, we should pay attention to the failures of the Treaty of Versaille. If we want to avoid despots who are lying between their teeth, we can view Hitler as the proximate cause. If we want to avoid appeasement, we can view Chamberlain as the proximate cause.

Debating which cause was the exact one is a failure of reasoning, because there is always a graph of causality, and while it’s hidden to us, within the infinite counterfactuals of the simulation I do believe it exists. Suppose we could run simulations of our reality using our RealitySim cluster. For example: Out of the 10,000 simulations, when Hitler wasn’t born Germany still went into fascism 84% of the time. Or: Out of the 10,000 simulations, a more generous treaty of Versaille prevented the rise of the Nazi party 54% of the time.

(If you’re wondering where the source of randomization comes from, I’m not sure exactly, but I think you could reasonably approximate it by injecting randomization into phenomena that aren’t correlated with what you’re interested in. A common instrument is something like weather patterns. A change from a sunny to a rainy day in Germany in June 1932 might alter future events, but it would be unrelated to our hypothesis).

To bring this back to the original point of compression and myth, this is a remarkably sophisticated telling of causality. Not only is the framework for counterfactual reasoning non-intuitive, but we don’t have any way to know the true estimates. So how do you teach this to the bottom 20th percentile of the population? Well, it’s probably Hitler’s fault, you know? If you want to go deeper, the books are there for you, it’s not a hidden secret. There isn’t some hidden stash of books the government banned.

It seems pretty clear to me that Daryl has unambiguous Nazi sympathies as

Noah Smith

articulates here. So he’s inclined to find the exact point to begin his analysis that best supports those beliefs. He uses a game theoretical sleight of hand, where the side his sympathies lie with are forced to react to the actions of some external antagonist. Of course, in reality the game of history is real-time, continuous, and extends backwards. All that means is we’re both making and reacting to each other in real time. I make my decisions, thinking about how that will effect your decision, and you make your decision thinking about how it will effect my decisions

If you are motivated you can instead model it as sequential, starting at the exact moment that aligns with the shit you’re on about. We see this all the time: Hamas didn’t start the war with October 7th, they were in fact only responding to something that came before it. Putin didn’t start the war with Ukraine, it was in fact Nato who caused it with their expansion. Hitler didn’t start the war… you get it.

What helps selective conditioning is knowing more about the past than others, since you can always frame the start of the conflict at a cherry-picked point of your choosing, where it begins at the moment the side you prefer was forced to react to some sort of grievance.

When Daryl presents Churchill as 'a chief villain' in WWII, he’s projecting his current political frustrations onto historical events. If you have a lot of problems with America today, and they relate largely to what you see as an excess in areas like sexual transitioning and excess immigration, that’s fine. Lots of people feel that way, and you can advocate and vote on it. The idea that you now need to interpret the causality of the past in such a way to support your current political grievances is only replicating the worst of whig history, and in a particularly boring way.