In 1913, a Polish-German socialist named Rosa Luxemburg published a book1 that made a strange argument. She said capitalism had a math problem. Not a moral problem — a structural one. And that problem would eventually destroy it.
The argument went like this. Capitalism works by selling things for more than they cost to make. The profit is real. But who buys everything? Workers only earn enough to buy part of what they produce. Capitalists spend some of their profits but reinvest the rest. So there's always more stuff being made than workers and capitalists combined can purchase. The system needs extra buyers. It needs an outside.
For two hundred years, that outside existed. Colonial territories. Peasant economies. Subsistence communities that hadn't yet been pulled into the market. Capital could expand into these spaces — sell goods, extract resources, put people to work — and keep the circuit running.
But the outside is finite. And as capitalism spreads globally, it keeps incorporating more of that outside into itself. When there's nothing left outside, the circuit breaks.
Why this matters beyond the math
Luxemburg wasn't just describing a technical problem in economics. She was identifying something that every powerful system does: it depends on resources it cannot produce itself. Capital needs markets, labor, and raw materials that come from communities, ecologies, and relationships it didn't create and can't replicate. It can enclose them, extract from them, and destroy them. It cannot build them from scratch.
This is what we mean when we talk about the commons — the shared productive capacity that communities build together. The common language and knowledge people use without paying for it. The informal networks of care and support. The natural systems that provide clean water and breathable air. The social trust that makes institutions function at all. None of this is produced by the market. All of it is required by the market to function.
Capital can capture the commons. It can privatize water, commodify attention, patent seeds. But once captured and privatized, those things stop being what they were. A commons that's been enclosed isn't a commons anymore — it's an asset. And assets don't spontaneously regenerate. They depreciate.
It cannot replace it.
Reform versus revolution
Luxemburg was also one of the sharpest critics of a certain kind of progressive politics. In Reform or Revolution2, she drew a line between two different kinds of change.
One kind operates within the system's rules. Win elections. Pass laws. Negotiate better wages. These are real gains — she wasn't dismissing them. But they work within the existing structure. They can improve conditions under capture. They cannot change the structure of capture itself. If the routing architecture stays the same, the system eventually routes around any reform.
The other kind of change is structural. It doesn't ask for better terms within the existing arrangement — it asks why this arrangement exists and whether a different one is possible. This is harder, riskier, and slower. But it's the only kind of change that can address the underlying problem rather than managing its symptoms.
She wasn't saying reform is useless. She was saying you need to know which kind of work you're doing, and not mistake one for the other.
How change actually happens
Her most underread work is The Mass Strike,3 published in 1906 after the Russian revolution of 1905. In it, she describes something she observed happening in the streets: change that coordinated without a central command. Strikes spreading from factory to factory, city to city, without a party organization issuing instructions. Workers reading each other and responding. Distributed intelligence, operating faster than any hierarchy could manage.
This wasn't chaos. It was a different kind of order. And it was more effective than the centralized models that claimed to speak for the movement.
She was describing what we now understand as a property of complex systems: distributed coordination often outperforms centralized planning when the environment is changing fast and the knowledge is local. City hall doesn't know your street better than you do. A party headquarters doesn't know your factory floor. The mass strike worked because it processed local knowledge locally, in real time.
Luxemburg mapped the "outside" that capital requires as geographic — colonial territories, pre-capitalist economies. What she didn't fully ask was: who is the outside at home? What labor, within the capitalist center itself, is being used without being counted?
The next article in this series begins to answer that question.
Rosa Luxemburg was arrested in January 1919 and executed without trial by a right-wing paramilitary unit acting with the knowledge of the Social Democratic government she had spent years opposing.4 She was forty-seven years old. Her body was thrown in a canal.
The ideas survived because they were right. The system does depend on what it cannot produce. The outside is real. And the question of what that outside is made of — who lives there, whose labor it runs on, whose knowledge it extracts — is the question the next generation of thinkers picked up and refused to let go.