What it actually is
In 2018 the astrophysicists Adam Frank and Gavin Schmidt published “The Silurian Hypothesis: Would it be possible to detect an industrial civilization in the geological record?” in the International Journal of Astrobiology. The name is a joke borrowed from Doctor Who, whose Silurians are a reptilian civilisation that predates humanity.
The paper does not claim such a civilisation existed. It asks a narrower and more interesting question: what would the geological record have to contain for us to detect one? And it answers by looking at what our own industrial civilisation is leaving behind.
This is the rarest thing in this archive: a fringe-sounding question, asked rigorously, by working scientists, with a published answer.
The signatures they considered
- Carbon and oxygen isotopes. Burning fossil fuel shifts the ratio of carbon isotopes in the atmosphere and oceans, and leaves a mark in sediments. Our own signal is already in the rock record.
- Radioactive fallout. Nuclear testing produced isotopes — plutonium, caesium, iodine — that will remain detectable for millions of years.
- Plastics and synthetic molecules. Novel organic compounds that do not occur naturally, and would persist as chemical traces even if the objects themselves were destroyed.
- Fossil fuel depletion. A civilisation that burned the easy coal and oil leaves a geological record of something that was there and is no longer.
- Technofossils and altered sedimentation. Cities, roads and mines change erosion patterns; the “urban stratum” is already being described.
- Nitrogen and phosphorus cycles. Fertiliser production and agriculture leave global chemical fingerprints.
For the deep past: a pre-human industrial civilisation is very unlikely — there is no such signature in the record — but the question of detectability is not trivial. Surface structures are erased by erosion and subduction within a few million years; a civilisation 100 million years old might leave only a thin chemical line.
For the present: our own signal is legible and will remain legible. The paper's real subject is the Anthropocene — what we are writing into the strata, and whether future geologists (or future civilisations) would read it.
Why the answer is not "yes, aliens"
The hypothesis is sometimes retold as evidence for a hidden predecessor civilisation, which inverts its logic entirely. Frank and Schmidt assume the null hypothesis — none existed — and test whether we could detect one if it had. Absence of a signal is evidence of absence, in a way that is unusually strong here, because we know exactly what to look for: we are producing the signature ourselves.
The comparison to make is with ancient astronauts, which starts from monuments and infers builders; the Silurian hypothesis starts from the rock and asks what a builder would leave. One predicts nothing. The other predicts a specific, searchable chemistry.
The uncomfortable reading
If industrial civilisations are short-lived — a few centuries of fossil-fuel use, then either collapse or transition — then the galaxy could be full of planets that hosted one briefly and left a chemical line nobody was around to read. That is a version of the Fermi problem that requires no hunters: not silence, but brevity.
Sources for this page
- Frank & Schmidt (2018), “The Silurian Hypothesis”, International Journal of Astrobiologycambridge.org
- The full paper on arXivarxiv.org
- NASA GISS, Gavin Schmidt — Anthropocene and climate modellinggiss.nasa.gov
- Wikipedia, “Silurian hypothesis” — summary and receptionen.wikipedia.org