Every Civilization Thinks It's Special

In 410 AD, the Visigoths sacked Rome. The news spread across the Mediterranean like a shockwave. St. Jerome, writing from Bethlehem, captured the disbelief: "The city which had taken the whole world was itself taken."

But here's the thing nobody talks about: Roman senators were still meeting after the sack. The aqueducts still ran. The grain ships still sailed from Egypt. For most people in most of the empire, life continued more or less as it had. The "fall" of Rome wasn't a moment. It was a process so gradual that the people living through it mostly didn't notice.

A generation later, things were noticeably worse. Two generations later, worse still. By the time the Western Empire formally ended in 476 AD, the ending was almost an afterthought—a Germanic general deposing a child emperor nobody cared about. The actual collapse had been happening for decades, invisibly, in a thousand small degradations that only became visible in retrospect.

This is the first thing you need to understand about civilizational collapse: it doesn't feel like collapse from the inside. It feels like things getting slightly worse, year after year, until one day you realize the world your grandparents lived in is gone.

The second thing you need to understand is even more uncomfortable: the historical mortality rate for civilizations is 100%.

The Math Nobody Wants to Hear

Let's be precise about this, because precision matters.

The Sumerians built the first cities. Gone. The Old Kingdom Egyptians built pyramids that have outlasted everything else they made—but the civilization that built them collapsed, reconstituted as the Middle Kingdom, collapsed again, reconstituted as the New Kingdom, and finally collapsed for good. The Assyrians terrorized the ancient Near East for centuries. Gone. The Persians built the largest empire the world had ever seen. Conquered. The Athenians invented democracy, philosophy, and theater. Conquered. Rome itself—the eternal city, the empire that defined "civilization" for half the planet—fragmented into pieces that spent centuries forgetting how to make concrete.

The Han Chinese. The Gupta Indians. The Khmer who built Angkor Wat. The Maya who calculated astronomical tables more accurate than Europe would achieve for another millennium. The Aztecs. The Inca. All of them complex, sophisticated, successful—and all of them either collapsed, were conquered, or transformed so completely that their descendants barely recognize them.

I'm not being nihilistic. I'm being statistical. If you look at every complex society that has ever existed, exactly zero of them persist in their original form today. The pattern isn't "most civilizations collapse." The pattern is "all civilizations collapse." The only question is when and how.

And yet—and this is the part that should keep you up at night—every single one of those civilizations contained people who believed their civilization was different. The Romans didn't think Rome would fall. The Maya lords didn't think the calendar would stop turning. The mandarins of the Qing dynasty didn't think the Celestial Empire was just another political arrangement that would one day end.

Civilizational mortality is matched only by civilizational denial.

What "Collapse" Actually Means

Before we go further, I need to rescue the word "collapse" from Hollywood.

When most people hear "collapse," they picture something like The Road—sudden catastrophe, empty cities, survivors fighting over canned goods. This is not what collapse researchers mean. It's not what the historical record shows. And imagining collapse this way actually makes it harder to see it coming.

The technical definition comes from Joseph Tainter, an anthropologist who spent years studying failed societies: collapse is "a rapid reduction in sociopolitical complexity."

Complexity means: hierarchical government, specialized labor, long-distance trade, information systems, infrastructure, professional institutions. A complex society has tax collectors and judges and engineers and standing armies and administrative records. A simple society has families and villages and maybe a chief.

Collapse means losing those things. Not all at once—gradually, unevenly, and in ways that most people experience as "problems" rather than "the end of civilization."

What does this actually look like?

The emperor stops being able to control the provinces. The roads fall into disrepair because nobody's maintaining them. The tax collectors can't collect, so the soldiers don't get paid, so the soldiers desert or turn to banditry. Long-distance trade becomes risky, then impossible. The specialized craftsmen who made luxury goods—or sophisticated tools—lose their customers and their skills fade. Literacy declines because the schools aren't funded and the libraries aren't maintained. Technical knowledge erodes. Each generation knows a little less than the one before.

This process took centuries in Rome. It took decades in the Maya lowlands. It took a generation or two in some cases of rapid collapse. But the experience from inside is similar: things stop working as well as they used to. Then they stop working at all. Then everyone who remembers when they worked is dead, and the degraded state becomes the new normal.

You see the problem. If collapse is gradual and unevenly distributed, how would you even know if you were living through one?

The Science of Collapse

Here's where it gets interesting.

For most of human history, we explained collapse the same way we explained everything else: morally. Rome fell because of decadence. The Maya collapsed because they angered the gods. Civilizations died because they deserved to.

This is psychologically satisfying and intellectually useless. It explains nothing, predicts nothing, and mostly serves to reassure us that we're not decadent, so it won't happen to us.

But starting in the 1980s, something changed. Researchers from completely different fields—archaeology, ecology, complexity science, economic history—started converging on the same questions: Are there structural patterns in collapse? Do civilizations fail in predictable ways? Can we identify warning signs?

The answers turned out to be yes, yes, and maybe.

Joseph Tainter examined dozens of collapse cases and found a common dynamic: societies add complexity to solve problems, and eventually the complexity itself becomes the problem. Early complexity additions pay for themselves. Later ones deliver diminishing returns. Eventually you're spending more to maintain the complexity than you're getting back from it.

Jared Diamond analyzed environmental collapses—Easter Island, the Norse in Greenland, the Maya—and found recurring patterns of resource depletion. Societies that destroy their own foundations, sometimes watching the destruction happen, unable to stop.

Peter Turchin built databases quantifying 10,000 years of historical societies and found cyclical patterns of instability. Empires rise and fall in waves that last two to three centuries. The waves have measurable precursors—elite overproduction, popular immiseration, state fiscal crisis. In 2010, based on these patterns, Turchin predicted that America would experience major political instability around 2020.

The prediction aged well.

C.S. Holling, an ecologist studying forest fires, realized that the same dynamics apply to societies. Systems that suppress small disturbances eventually get catastrophic ones. The longer you prevent the correction, the harder it hits when it finally comes.

These researchers didn't coordinate. They came from different disciplines, studied different cases, used different methods. And they converged on remarkably similar conclusions about how complex systems fail.

The Patterns

So what are the actual patterns? What do collapsed civilizations have in common?

Complexity overshoot. This is Tainter's core insight. Societies become complex because complexity solves problems. More specialization, more hierarchy, more coordination mechanisms—at first, these deliver huge returns. Your first standing army, your first writing system, your first professional bureaucracy—game-changing.

But the curve bends. Each additional layer of complexity delivers less benefit than the one before. The tenth layer of administration helps less than the first. The hundredth regulation adds less than the tenth. Eventually you're adding complexity just to manage the complexity you already have.

And then—and this is the part that gets people—the curve can go negative. Past a certain point, more complexity makes things worse. The bureaucracies conflict with each other. The regulations create more problems than they solve. The system becomes so elaborate that nobody understands how it works, including the people running it.

Societies don't intentionally go past this point. They end up there because every problem looks like it needs a complex solution, and nobody ever simplifies. Complexity accumulates. Maintenance costs compound. One day you realize the empire costs more to run than it produces.

Energy constraints. Every civilization runs on energy—food, fuel, labor power. The complexity you can sustain is limited by the energy available to sustain it.

This seems abstract until you realize it's not. Rome ran on agricultural surplus and conquest booty. When the frontiers stabilized and conquest stopped, one energy source disappeared. When agricultural land degraded and populations declined, the other shrank. The complexity that had built the empire was now being starved.

The modern term for this is EROI—Energy Return on Investment. In 1930, oil had an EROI of 100:1. You invested one barrel's worth of energy to get a hundred barrels out. Pure surplus, raining from the sky. Today it's more like 15:1 for conventional oil, 5:1 for tar sands. The net energy available to run civilization is shrinking even as we pump more oil than ever.

Elite overproduction. This is Turchin's signature finding. Complex societies produce elites—trained specialists competing for high-status positions. When the society is expanding, there are positions for everyone. When expansion stops, there are more elites than positions.

Frustrated elites become dangerous. They have the education and connections to organize. They have the ambition and entitlement to believe they deserve power. When they can't get it through normal channels, they start looking for other ways.

Turchin found this pattern preceding virtually every major period of instability in recorded history. Late Republican Rome. Pre-revolutionary France. Antebellum America. The decade before the Cultural Revolution in China. When you produce more credentialed aspirants than your system can absorb, something breaks.

Environmental degradation. Diamond's territory. Societies that destroy their resource base are destroying themselves. Deforestation. Soil exhaustion. Aquifer depletion. Climate disruption. These recur across collapse cases, and they share a common feature: the costs are delayed and distributed while the benefits are immediate and concentrated.

Resilience loss. Holling's insight from ecology. You can optimize for efficiency or resilience, but not both. Systems that remove all redundancy, that tighten every connection, that eliminate all slack—they're incredibly productive right up until they shatter.

A forest that experiences regular small fires stays healthy. A forest that suppresses all fires accumulates fuel until one spark creates an inferno. Economies that allow regular recessions clear out bad investments. Economies that prevent all recessions accumulate fragility until the crash takes down everything.

Why Nobody Sees It Coming

If the patterns are real and the research is solid, why didn't the Romans see it coming? Why didn't the Maya?

Gradual onset. Each year is only slightly worse than the last. The degradation accumulates below the threshold of perception. By the time it's obvious, it's been happening for generations.

Baseline shift. Older people who remember how things were die. Younger people grow up with the degraded state as normal. The knowledge that things were different evaporates.

Sunk cost commitment. The more you've invested in the current system, the harder it is to imagine abandoning it. Roman senators had spent their whole lives climbing the hierarchy. They weren't going to accept that the hierarchy was collapsing.

Complexity obscures causation. In a complex system, effects are distant from causes. The decision that doomed you might have been made fifty years ago by people who are now dead. The feedback is too slow for learning.

The system selects for denial. People who raise alarms are punished. Optimists get promoted; pessimists get marginalized. The system filters for exactly the kind of thinking that will prevent it from seeing its own failure.

Roman senators weren't stupid. Maya lords weren't blind. They were embedded in systems that made collapse difficult to perceive, difficult to discuss, and difficult to address even when perceived.

Sound familiar?

The Contemporary Question

I've been careful so far to talk about the past. Historical case studies. Academic research. Patterns in data.

But you're not reading this for history. You're reading this because of a question you're probably afraid to ask directly: Is it happening to us?

The honest answer: we don't know. But we can be more precise about what we don't know.

The patterns identified in historical collapses aren't just historical. They're structural. And the question of whether those structures exist today has empirical answers.

Is complexity increasing faster than returns on complexity? Is the administrative state growing faster than what it administers? Are regulations accumulating faster than problems are getting solved? Look at the data and draw your own conclusions.

Are we producing more credentialed elites than positions for elites? Is there credential inflation, frustrated ambition, political polarization driven by educated people who feel cheated by the system? Look around.

Are we degrading the environmental systems we depend on? Is climate changing? Are aquifers depleting? Are fisheries collapsing? The data exists.

Are our systems optimized for efficiency at the cost of resilience? Did supply chains break during a pandemic? Did a single ship in a canal cause global disruption? Are we running leaner and more fragile than ever before?

This series will walk through the frameworks in detail. Tainter on complexity. Diamond on environment. EROI on energy. Turchin on cycles. Holling on resilience. The researchers who built the science of collapse, and what they found.

I'm not going to tell you we're doomed. The researchers themselves are careful to distinguish analysis from prophecy. We have capabilities previous civilizations lacked. Global communication. Scientific method. Historical awareness.

But I'm also not going to tell you we're safe. The patterns are real. The warning signs are visible. And believing we're special—believing the math doesn't apply to us—is exactly what every previous civilization believed.

They were wrong.

Further Reading

- Tainter, J.A. (1988). The Collapse of Complex Societies. Cambridge University Press. - Diamond, J. (2005). Collapse: How Societies Choose to Fail or Succeed. Viking. - Turchin, P. (2023). End Times: Elites, Counter-Elites, and the Path of Political Disintegration. Penguin.

This is Part 1 of the Collapse Science series. Next: "Joseph Tainter: Diminishing Returns on Complexity."