A long-lasting civilization might be invisible. A visible one might not last.
Long before human memory or record, the Earth cycled through billions of years of deep time — and in 2018, NASA researcher Gavin Schmidt and astrophysicist Adam Frank posed a quietly unsettling question: if an intelligent civilization had risen and fallen in that vast, unwitnessed span, would we possess any means of knowing? Their peer-reviewed thought experiment, the Silurian Hypothesis, does not claim such a civilization existed, but rather illuminates the limits of geological memory itself. It is a meditation on absence — on how completely time erases, and how the very marks of a civilization's virtue or destruction determine whether it can ever be found.
- The geological record preserves only fragments of deep time, meaning entire civilizations could have lived and vanished without leaving a single legible trace.
- Industrial signatures — carbon spikes, temperature surges, chemical shifts — are indistinguishable from natural events like volcanism, making it theoretically impossible to confirm an ancient civilization's existence.
- A real event, the Paleocene-Eocene Thermal Maximum 56 million years ago, mirrors today's climate disruption so closely that scientists cannot rule out an intelligent cause — though no evidence supports one.
- The hypothesis carries a cruel paradox: sustainable civilizations leave almost no mark, while destructive ones leave more evidence but are unlikely to survive long enough to matter.
- Popular imagination has run far ahead of the science, weaving the hypothesis into myths of sea peoples and hidden beings — interpretations the evidence does not support.
- What the hypothesis ultimately reveals is not a hidden history, but the humbling depth of what we cannot know across 4.5 billion years of planetary time.
Imagine a skyscraper built 100 million years ago. By now, nothing would remain — no steel, no concrete, no whisper of its existence. This image anchors a question that has begun to trouble even skeptical scientists: if an intelligent civilization rose and fell long before humans, would we have any way of knowing?
In 2018, NASA researcher Gavin Schmidt and astrophysicist Adam Frank published a peer-reviewed paper in the International Journal of Astrobiology exploring exactly this. They called it the Silurian Hypothesis — not a claim that such a civilization existed, but a thought experiment probing the limits of geological detection. Over deep time, everything — buildings, machines, bones — eventually becomes indistinguishable from the planet itself. And the signals a collapsed civilization might leave behind would look identical to natural events: a carbon spike from industry or from a volcano, rising temperatures from human activity or from tectonic methane release.
Schmidt and Frank point to the Paleocene-Eocene Thermal Maximum, roughly 56 million years ago, as an example. Global temperatures spiked, carbon flooded the atmosphere, ocean chemistry shifted. It resembles the climate disruption humans are causing today — yet natural processes alone could produce the same signature. There is no way to prove an intelligent species caused it.
The hypothesis carries a deeper paradox. A long-lasting civilization would likely become sustainable, living in harmony with its environment — and therefore leaving almost no detectable mark. A destructive civilization might leave more visible evidence, but such practices would probably not sustain a society across millions of years. The most durable civilizations may be the most invisible.
The idea has captured popular imagination, inspiring viral videos and speculation about ancient sea peoples or hidden beings — interpretations that stretch far beyond what the evidence allows. For any such scenario to hold, a civilization would need to have persisted for tens of millions of years and then vanished without a trace. There is no evidence this occurred.
Yet the core insight endures. We cannot be certain intelligent life did not exist in Earth's distant past. We cannot be certain we would recognize it if it had. The geological record is not a complete archive — it is a fragmentary document. Our species has existed for roughly 300,000 years; the Earth itself is 4.5 billion years old. In that immense span, we can account for only the tiniest fraction of what may have lived and died. The Silurian Hypothesis does not answer whether another civilization existed. It simply reveals how little we can actually know.
Imagine a skyscraper built 100 million years ago. By now, there would be almost nothing left of it—no steel, no concrete, no trace that it ever stood. This thought experiment sits at the heart of a deceptively simple question that has begun to trouble even skeptical scientists: if an intelligent civilization rose and fell long before humans walked the Earth, would we have any way of knowing?
The question is not new, but it gained serious scientific weight in 2018 when NASA researcher Gavin Schmidt and astrophysicist Adam Frank published a peer-reviewed paper in the International Journal of Astrobiology exploring exactly this possibility. They called it the Silurian Hypothesis—a thought experiment designed to probe the limits of what we can actually detect in Earth's deep past. The paper does not claim that such a civilization existed. Rather, it asks: if one did, what would we be looking for, and would we even recognize it?
The challenge is profound. The fossil record preserves only a tiny fraction of the life that has ever existed on Earth. Over geological timescales, everything—buildings, machines, bones, waste—eventually becomes indistinguishable from the planet itself. If a civilization industrialized, consumed resources, and then collapsed, it might leave behind only the faintest of signals, if any at all. And here is where the problem deepens: many of those signals would look identical to natural events. A spike in atmospheric carbon dioxide could come from a factory or from a volcano. Rising global temperatures could result from human activity or from tectonic shifts releasing methane from the ocean floor. How would we tell the difference?
Schmidt and Frank point to a real geological event as an example: the Paleocene Eocene Thermal Maximum, which occurred roughly 56 million years ago. During this period, global temperatures spiked, carbon flooded the atmosphere, ocean chemistry shifted, and some species went extinct while others adapted. On the surface, it resembles the climate disruption humans are causing today. But the paper itself acknowledges that natural processes—volcanism, tectonic activity—can produce identical signatures. There is simply no way to prove that an intelligent species caused it rather than nature itself.
Then there is the paradox that makes the hypothesis even more unsettling. A civilization that lasted a very long time would likely need to become sustainable, living in relative harmony with its environment. But the more sustainable it became, the less damage it would inflict on the planet—and the less visible it would be in the geological record. Conversely, a civilization that burned fossil fuels and ravaged its world might leave detectable marks, but such destructive practices would probably not sustain a society for millions of years. A long-lasting civilization might be invisible. A visible one might not last.
The hypothesis has captured popular imagination in ways the original paper perhaps did not anticipate. A TikTok video exploring the idea accumulated over 88,000 views, and some have attempted to connect it to ancient myths—stories of sea people, underground beings, creatures that influenced early human societies. These interpretations stretch credibility. For such a scenario to work, a civilization would need to have retreated into the oceans around 56 million years ago, then somehow persisted for tens of millions of years until encountering early humans only a few hundred thousand years ago, then vanished without a trace. There is no evidence whatsoever that this occurred.
Yet the core insight of the Silurian Hypothesis remains genuinely unsettling, even to skeptics. We cannot be certain that intelligent life did not exist in Earth's distant past. We cannot be certain we would recognize it if it had. The geological record is not a complete archive—it is a fragmentary document, and vast stretches of deep time remain opaque to us. We know our own species has been here for roughly 300,000 years, our ancestors for about 6 million. The Earth itself is 4.5 billion years old. In that immense span, we can account for only the tiniest fraction of what may have lived and died and left no mark. The hypothesis does not answer the question of whether another civilization existed. It simply reveals how little we can actually know.
Citações Notáveis
The fossil record only preserves a tiny fraction of life that ever existed. If a civilization rose, industrialized, and collapsed, it might not even leave a trace behind.— Concept from Schmidt and Frank's paper, popularized by Jeff Foran
A Conversa do Hearth Outra perspectiva sobre a história
So the paper isn't actually claiming that some ancient civilization was here before us?
No, it's a thought experiment. Schmidt and Frank are asking: if one existed, would we have the tools to detect it? That's the real question.
But why would we not be able to detect it? Wouldn't there be fossils, artifacts, something?
Over 100 million years, a skyscraper becomes dust. Fossils are rare—the record preserves maybe a fraction of a percent of what lived. And here's the trap: if a civilization burned fossil fuels, it would leave a chemical signature. But so does a volcano. We couldn't tell them apart.
So a destructive civilization would be more visible?
In theory, yes. But that's the paradox. A civilization that destroys its environment probably doesn't last very long. A sustainable one leaves almost no mark at all.
That's genuinely unsettling.
It is. We're looking at 4.5 billion years of Earth's history, and we can really only account for the last few million. The rest is mostly dark to us.