Below is a lightly edited transcript. For the article that inspired this one, see Hydraulic Despotism.
Welcome to the btrmt. Lectures. My name is Dr Dorian Minors, and if there is one thing I’ve learned as a brain scientist, it’s that there is no instruction manual for this device in our head. But there are patterns. Patterns of thought, patterns of feeling, patterns of action. That’s what brains do. So let me teach you about them. One pattern, one podcast, and you see if it works for you.
So today I want to talk about something a little bit different from the usual brain science fare or the other stuff that I teach here at the academy. I want to talk about infrastructure, and specifically, I want to talk about a dead theory from political science that I think describes modern life a little bit better than any of the live ones.
Now, I should say, as I always do, that this is my own perspective. Nothing to do with my day job at Sandhurst, just Dorian doing his little podcast. But today I want to talk about a theory called hydraulic despotism, proposed by a guy called Karl Wittfogel in the 1950s. And I’ll be up front — it was absolutely savaged by his peers, and rightly so, because it has some real problems. But not only does it have an excellent name, there’s a pattern inside it that I think is worth extracting, because once you see this, I think you’ll start seeing it everywhere. And it has some interesting implications for all of us.
So let’s get into it.
Control the Water, Control the People
Here’s the basic idea. People need water. Not just as a basic need, but also because it allows us to lean into agriculture. With a rich flow of water, you can irrigate your crops. And so where water flows, you’re going to find more people congregating around that water. And as populations grow, you’re going to find that infrastructure grows to support those populations. And where infrastructure grows, mobility is going to drop because people become more reliant on these conveniences — the local goods and services to support the size of that community.
And so at some point — and this is the key move that Wittfogel makes — the organisation, the government that controls the water at the point that mobility starts dropping, has the capacity to become a despot. Because if they were to take the water away, people wouldn’t just be able to move. They would face death. So whoever controls the water has total control of the people.
This is a thesis that he wrote about in his book Oriental Despotism. And he argued that this explained the historical empires of China, of Egypt, of Mesopotamia and pre-Columbian Peru. All these ancient empires arose because somebody figured out irrigation and then used it as a lever to become despots, to become a tyrannical force. In contrast, he liked to point out, you had these enlightened Europeans elsewhere who had this beautiful system of feudalism where people were free. Incidentally they were starving, but at least they were free.
And here come the problems with Wittfogel’s thesis. Not just that the comparison to European feudalism doesn’t seem like quite the comparison he thought it was, but also that the examples he picked weren’t really demonstrating what he thought they were demonstrating. The causality often ran backwards — a lot of successful states built irrigation, it wasn’t that irrigation led them to take control of a group of people. And plenty of actually despotic states didn’t bother with irrigation at all. If you look at the history of the man, the whole thing was fairly transparently motivated by a shift from being a committed communist to a committed anti-communist. He was really trying to explain why the Soviet Union and China were the way they were, and he worked backwards from the conclusion.
So his theory of hydraulic despotism died in the water. Nobody serious is citing Wittfogel approvingly. It’s a dead theory and that’s where most people leave it.
Modern Hydraulic Resources
But if you’re a cynic like me, you hear something like this — controlling the water means controlling the people — and you start looking around at the way that we live our lives with a bit of a raised eyebrow.
So let’s do an autopsy on Wittfogel’s theory. Put aside the specifics about ancient irrigation. The insight that I think is interesting is that any resource that flows through increasingly centralised infrastructure, when it becomes essential and non-substitutable — you can’t just go and get it somewhere else — then whoever controls it controls you. And once you frame Wittfogel’s theory like that, you start seeing hydraulic resources everywhere.
Start with the obvious one. Actual water. Here in the UK, Thames Water has something like a 20 billion pound debt pile. They can’t even manage the literal water resource. So that’s a bit embarrassing for the metaphor, but it does rather prove the point.
Or you could think about energy. Hydrocarbon fuels, electricity, gas — these things are controlled by relatively small and relatively centralised forces. And it’s really evident in the resistance we face trying to move from these products for energy to more modern products for energy. The infrastructure makes it very, very difficult to change. Or Russia’s activity in Ukraine demonstrated this almost overnight — you had global food insecurity because grain, another flowing resource through centralised infrastructure, was suddenly being controlled by a conflict.
You can also think about communities. I’ve written about this a couple of times — about rising rates of loneliness, about the whittling away of familial connections. But the short version is that we increasingly outsource all the things we used to do for each other to organisations, both governmental and non-governmental. And a lot of the problems we have in terms of mental health and well-being are because the infrastructure we outsourced it to is becoming increasingly fragile, particularly in a changing political environment.
Or you could think of social media. Your links to your friends and your family flow through just a handful of companies. You have Meta, you have Google, you have ByteDance. And so this aspect of your relationships is subject to what I’d call hydraulic conditions. If Meta decides to change how its algorithm works tomorrow, your ability to maintain your connections to your friends changes. And I think that’s one of the things that’s so problematic about social media — not just the content it feeds us, but how little control we have over how that content is fed to us.
And you can keep going. Payment processing — there are just a handful of processors that control almost all of the flow of money, and you can’t really run a business without them. Visa, Mastercard, Stripe. AI compute — the absurd Nvidia valuation tells you everything you need to know about who controls access to the resources that are going to shape whatever comes next. Medicine and medical aid, data storage, cloud infrastructure. All of these things are flowing resources that have become centralised and increasingly non-substitutable.
Convenience as Control
It’s certainly not what Karl Wittfogel had in mind. And to just call it despotism off the bat is certainly unfair. But I think it’s an interesting pattern, because one of the things that I’m very interested in is mental well-being. And there is an enormous correlation between mental well-being and a sense of controllability. This is an old theory now — Selye was an endocrinologist, but he had this idea of the difference between distress, bad stress, and eustress, good stress. I have an article that talks about this in more detail. But fundamentally the difference is partly biological, partly psychological — and it comes down to perceived controllability. If you don’t feel in control, you’re going to interpret the same stress that could be excitement as something threatening.
And I don’t think that in the main it’s truly despotism. People aren’t doing this on purpose. In the main, it’s just industrial society doing what it does — centralising life in urban areas, delegating our responsibilities to institutions, making things more convenient, making things more productive, and doing so at the cost of our communal ties. I don’t think many people would be keen to go back to a time where hospitals weren’t a thing because people weren’t centralised enough for that to make sense.
But I think there is something we can do about it from a mental health perspective, because we often actively participate in relinquishing this control. The delegation is convenient. When it comes to technology, because it’s so difficult to understand, we happily hand it over. We don’t have to use Meta for our social connections, but it’s easier than the alternatives. There’s a reason not many people ended up migrating to Mastodon when Twitter started getting wild. And we don’t have to use a particular cloud provider, but migrating once you’re in the Google or the Apple ecosystem is very difficult.
And it’s not just convenience. There is also an element of coercion here. And it’s not the coercion that Wittfogel’s ancient subjects suffered from — coercion by threat of death. It’s coercion by marginalisation. Try and migrate your friends from WhatsApp to Signal and see how many people join you there. If you don’t buy into the infrastructure, you’re going to miss out. It’s a softer coercion, but the endpoint looks awfully similar.
By this point, you can’t leave. Your relationships are there on whatever service you’re on. Your data’s there, your workflows are built around it. Your flowing resource has been captured and now the terms can change.
And I think that’s why I like hydraulic despotism for this. There are other theories that point directly at what I’m talking about — Michael Mann has something called infrastructural power; there’s platform capitalism; Callon has something on obligatory passage points from actor-network theory; even the essential facilities doctrine from antitrust law. All of these things describe what I’m talking about. But none of those things have that same kind of motivating force as hydraulic despotism. Hydraulic despotism sounds worrying. Good. I think we should be worried.
What Can Be Done
But I don’t just want to leave you with the worry, because unlike Wittfogel’s ancient imagined subjects who really couldn’t leave, we do have alternatives now.
Cory Doctorow works at the Electronic Frontier Foundation and gave a talk at the 39th Chaos Communication Congress in early 2026, where he describes what I’m about to talk about in great detail. I think that article is well worth reading, and I’ll leave it in the show notes. But essentially you have a lot of forces aligning against this kind of hydraulic control now in a way that hasn’t really been the case in previous years. You have people in the digital rights space who’ve been harping on about this for long enough that governments are starting to do something about it. You have tech entrepreneurs who are trying to raid big tech’s margins, particularly with AI. And then you’ve got national security hawks increasingly worried about dependency on concentrated infrastructure because of the way the global stage is changing. So this concern isn’t fringe anymore. It’s starting to become mainstream.
And Doctorow is also excited about some of the changes in the legal landscape. Anti-circumvention laws exist in almost all jurisdictions that criminalise modifying devices without the approval of the manufacturer. The US often makes these a condition of a trade agreement. These are the reason why it’s so hard to jailbreak an iPhone — you own the technology, but you’re not allowed to jailbreak it because it’s illegal. Doctorow thinks that now with trade agreements being weaponised, these kinds of pressures might go away and people might start building software that allows you to have more independence.
But I don’t think we have to wait for these changes. All the alternatives already exist. Federated social media, self-hosting your own services, community-owned infrastructure, open-source software — probably the single most successful anti-hydraulic movement in history. All of these things have been possible for a long time and none of them have been as convenient as the alternatives. And in fact, that’s the point. The convenience is the mechanism of control. Every time you choose the convenient option, you are strengthening the hydraulic infrastructure.
But now is the time of AI. And I think the landscape looks very different, because where once you had to figure out these complicated things yourself, now you have the opportunity to have AI walk you through this alternative landscape — to help you take back control of these hydraulic resources and move you away from the infrastructure that controls them. And in doing so, I think we have a chance of gaining back a sense of control and hopefully alleviating some of the pressure that is contributing to the lack of mental well-being that’s increasingly prevalent in our society.
So I’ll leave you with this. The defining feature of human beings, I reckon, is our capacity for nurture — to share ideas and bring them into the world. To live in a modern society is to relinquish control of our water. That’s unavoidable at this point. But we don’t need to lean so hard into that fact and hand over everything else as well. I write more about this — about recreating systems from zero, about digital literacy as a civic responsibility, and about what happens to our relationships when they’re mediated by hydraulic infrastructure. Now I think that we have the ability to do something a little bit different.
