The Fermi Paradox
A single civilization, launching self-replicating probes at one percent of light speed, saturates the Milky Way in 10 million years. The galaxy is 13.6 billion years old. The thing needs to happen once. It has not happened — or we cannot see that it has. That gap is the paradox, and every improvement in our instruments makes it sharper.
Fermi posed the question over lunch at Los Alamos in 1950: where is everybody? Seventy-six years later, the answer is still nothing, and the silence is no longer cheap. Breakthrough Listen has spent $100 million scanning a million stars. JWST is now reading exoplanet atmospheres for industrial pollutants. The null result is data.
At a glance
Half a dozen serious answers, no resolution after 75 years.
The argument that resists dismissal
The von Neumann probe version is the form that hurts. One civilization, anywhere in galactic history, only has to launch a self-replicating probe once. No faster-than-light travel. No civilization-spanning patience. The probe travels to the nearest star, mines material, builds copies, sends them onward. Even at 0.01c with long pauses for replication, the galaxy fills in 1–10 million years — under 0.1% of its age.
The argument doesn't need civilizations to survive. It doesn't need them to want to colonize. It needs one of them, ever, to have wanted it for the duration of a single launch.
Solutions, ranked by how much they cost to believe
| Solution | What it assumes | What it costs |
|---|---|---|
| Great Filter behind us (Hanson, 1996) | Abiogenesis or eukaryotic complexity is astronomically rare | Earth's life is a near-miracle |
| Great Filter ahead | Civilizations routinely self-destruct | We probably will too |
| Rare Earth (Ward & Brownlee, 2000) | Plate tectonics, large moon, Jupiter shield, stable star — all required | Microbes common, minds nearly unique |
| Dark Forest (Liu, 2008) | Revealing position is suicidal; rational play is silence | Universal game-theoretic compliance |
| Grabby Aliens (Hanson, 2021) | Expanders exist but haven't reached us yet | Testable; predicts we are early |
| Transcension / aestivation | Advanced minds go inward or wait for the cold | No expanding civilizations needed |
| Zoo hypothesis (Ball, 1973) | They know, they hide, they enforce non-contact | Universal compliance across species |
Hanson's "grabby aliens" model is the most quantitative recent contribution — it predicts that if expanding civilizations are out there, humanity should find itself unusually early in cosmic history. Which we do.
What the silence has bought us
- BLC1 (2020): narrowband signal from the direction of dest proxima centauri. Resolved as terrestrial radio interference by 2022.
- MeerKAT survey (2023–2024): 60+ nearby stars. Strongest-ever constraint on transmitters within 100 light-years. No technosignatures.
- FAST candidate signals (China, 2022): all RFI on follow-up.
- Dyson sphere search (Suazo et al., 2024): seven candidates with mid-infrared excess in Gaia data. Natural explanations remain more likely. Follow-up ongoing.
- JWST: TRAPPIST-1b and c appear to lack thick atmospheres. The biosignature search continues at other targets; industrial pollutants (NO₂, CFCs) are in principle detectable to ~30 ly.
- PANOSETI: wide-field optical/IR detector for nanosecond laser pulses, prototype operational 2024.
Each null trims the parameter space. The hypothesis "loud civilizations within 100 ly" is now essentially dead. "Loud civilizations within 1,000 ly" is bleeding.
What's contested
Whether the paradox is even a paradox. The empirically conservative read: we have searched a tiny fraction of the sky, in a tiny slice of the electromagnetic spectrum, for a tiny window of time, assuming civilizations broadcast at frequencies and powers we'd recognize. The "Great Silence" may be a measurement problem, not a metaphysical one. David Kipping and others argue the priors on the Drake equation are so wide that almost any answer fits.
The harder version — von Neumann probes — resists this defense, because probes persist and replicate without continuous broadcasting. The standard counter is replication-error accumulation: across thousands of generations, probes degrade catastrophically. This was modeled computationally in 2022–2023 with mixed results. The "lurker" hypothesis — that probes are here, nano-scale, parked in asteroids or at Lagrange points — is untestable in the way most interesting hypotheses are not.
Why this has to do with other realms
The paradox is, in disguise, a question about concept existential risk: if the Great Filter is ahead of us, what is it, and how close are we to it? Nuclear weapons, engineered pandemics, runaway AI, climate collapse — each is a candidate. The "ahead of us" interpretation of the filter turns a SETI puzzle into a question about whether our own century is the last bottleneck humanity passes through. Robin Hanson's economic models, Nick Bostrom's existential risk framework, and the SETI null result are the same conversation under different lighting.
An open question
If a Bracewell probe were already in our solar system — small, cold, listening — what observation would distinguish its presence from its absence?
Key sources
- Where Is Everybody? by Stephen Webb (2015, 2nd ed.) — the canonical survey of 75 proposed solutions.
- Rare Earth by Peter Ward and Donald Brownlee (2000) — the strongest case for the contingency view.
- Hanson, R. "The Great Filter — Are We Almost Past It?" (1998 essay, hanson.gmu.edu) — the original framing.
- Hanson, Martin, McCarter, Paulson, "If Loud Aliens Explain Human Earliness, Quiet Aliens Are Also Rare" (2021, arxiv:2102.01522) — the grabby aliens model.
- The Three-Body Problem trilogy by Liu Cixin (2008–2010) — the Dark Forest articulated as fiction; the cleanest statement of the game-theoretic version.
- Breakthrough Listen public data releases (breakthroughinitiatives.org) — the underlying observational record.
Further reading
- The Eerie Silence by Paul Davies (2010) — argues SETI is searching the wrong way, with an honest accounting of post-radio approaches.
- Kipping, D., "An Objective Bayesian Analysis of Life's Early Start and Our Late Arrival" (PNAS, 2020) — the conservative statistical read.
- Suazo et al., "A Search for Dyson Spheres Around Late-Type Stars" (MNRAS, 2024) — the Gaia infrared-excess paper.
- Liu Cixin's The Dark Forest — the most disturbing fictional solution, and the most influential outside academic SETI.
- The 99% Invisible episode on the Golden Record — what we chose to say if anyone is listening.
See Also
- mission voyager 1 — the only message we've physically sent, now 24 billion km out and still inside our heliosphere
- mission breakthrough starshot — the first serious engineering case for reaching another star within a human lifetime
- dest proxima centauri — the nearest star, the BLC1 false alarm, and the actual distance the probe argument has to cross
- concept existential risk — if the Great Filter is ahead of us, this is what it looks like
- concept habitable zone — the prior that's been narrowing fastest under JWST data
- overview milky way neighbors — the volume of sky the null result now covers