In the summer of 1950, the physicist Enrico Fermi sat down to lunch at Fuller Lodge in Los Alamos, New Mexico, with Edward Teller, Herbert York, and Emil Konopinski. The conversation turned to a New Yorker cartoon by Alan Dunn depicting aliens stealing New York City's trash cans, then to the probability of faster-than-light travel, then to the general question of extraterrestrial civilizations. The discussion moved on. Some minutes later, in the middle of lunch and without apparent context, Fermi looked up and asked his companions: "But where is everybody?" All three companions independently recalled, decades later, that everyone at the table instantly understood what he meant.
The question, as reconstructed by Eric Jones of Los Alamos in his 1985 report "Where is Everybody?": if the galaxy is filled with civilizations even marginally older than ours, they should have colonized it by now. A single technological civilization — one, anywhere — expanding at reasonable speeds with self-replicating probes would saturate the Milky Way in a span of cosmic time that is, by the galaxy's age, instantaneous. So where is everybody?
The silence that followed is not a puzzle a casual answer resolves. It is, when taken seriously, one of the most disturbing questions in all of science.
The universe is approximately 13.8 billion years old. The Milky Way alone contains between 100 and 400 billion stars. Kepler mission data, published in the early 2010s, established that most stars host planets, with Erik Petigura's widely cited 2013 PNAS paper estimating that roughly 22% of Sun-like stars in the Milky Way have an Earth-sized planet in the habitable zone. That is something between 8 and 40 billion Earth-analogue planets in our galaxy alone. Across the roughly two trillion galaxies in the observable universe, the numbers become computationally obscene.
Even with extremely conservative assumptions — life is rare, intelligence is rarer, technology is rarer still — the cumulative arithmetic produces a single conclusion: the galaxy should be obviously inhabited, many times over, by civilizations far older than ours.
Michael Hart made this argument in mathematical form in a 1975 paper in the Quarterly Journal of the Royal Astronomical Society, "An Explanation for the Absence of Extraterrestrials on Earth." Hart's observation — which has come to be called Fact A in the subsequent literature — is simply that there are no extraterrestrials on Earth. No artifacts, no probes, no radio signals, no confirmed anomalous presences. Hart argued that this empirical absence, given the cosmic parameters, is itself extraordinary, and that any resolution of the Fermi Paradox must explain not just the silence at interstellar distance but the absence here, in our own solar system, of signs of prior visitation. Frank Tipler extended the argument in 1980 in QJRAS, demonstrating that self-replicating Von Neumann probes — machines that land on a planetary body, construct copies of themselves from local materials, and launch each copy toward the next star — could saturate the galaxy in between five and fifty million years of steady expansion, a timescale that is one ten-thousandth of the galaxy's age. The conclusion Hart and Tipler drew: we are alone.
That conclusion is the one that most people refuse. But every alternative to it is worse.
Frank Drake wrote down, in 1961, the equation that has organized the field since. The Drake equation calculates N — the number of detectable civilizations currently in the Milky Way — as the product of seven factors: the rate of star formation, the fraction of stars with planets, the number of habitable planets per such star, the fraction of those on which life arises, the fraction of those on which life becomes intelligent, the fraction of those that develop detectable technology, and the fraction of the galaxy's age during which such civilizations are detectable. The equation is not empirical. It is a decomposition — a way of breaking an impossibly broad question into categories that future observation might constrain one at a time.
Sixty years later, the first three terms are roughly known. Star formation rates are measured. Kepler has pinned down the prevalence of habitable-zone planets. The remaining four are essentially free parameters. Our best estimate of the probability that abiogenesis occurs on a habitable world ranges across many orders of magnitude, depending on whether one believes (as Francis Crick proposed in 1973) that life on Earth was seeded from elsewhere (directed panspermia) or arose independently. Our best estimate of the fraction of life that becomes intelligent depends on whether one treats the evolution of complex brains as a generic convergent outcome or as a one-off catastrophe — the Cambrian explosion, the fortuitous retention of a particular jaw structure, the chance survival of primates through the K–Pg extinction.
Drake himself, in his later years, estimated N at around ten thousand detectable civilizations in the Milky Way. Adam Frank and Woodruff Sullivan, in a 2016 Astrobiology paper, argued that Drake-equation pessimism had collapsed enough under Kepler data that the probability of humanity being the only technological civilization to ever have arisen in the observable universe can now be bounded at less than one part in 10^22. The argument is not that other civilizations currently exist. It is that other civilizations almost certainly have existed at some point. Which reproduces the puzzle, not dissolves it: if so many civilizations have existed, why have none of them left traces we can see?
Robin Hanson proposed in 1998 the framework that has come to dominate the serious literature: the Great Filter. Somewhere between dead matter and galaxy-spanning civilization, there is at least one step — and possibly several — so difficult or catastrophic that virtually no species gets past it. The question is not whether a filter exists. The arithmetic requires one. The question is where in the sequence the filter is.
The optimistic interpretation is that the filter is behind us. Perhaps the emergence of self-replicating chemistry from non-replicating chemistry is astronomically unlikely — the so-called "abiogenesis bottleneck." Perhaps the transition from prokaryotic to eukaryotic cells (which happened only once in Earth's history) is a vanishingly rare endosymbiotic accident. Perhaps the evolution of multicellular life, language, or civilization itself involves bottlenecks we have already passed through by luck. If so, we are rare — possibly, as Hart concluded, unique — and the silence is the sound of an empty galaxy.
The terrifying interpretation is that the filter is ahead of us. Hanson's paper laid out the logic: the more rapidly we find evidence of past life elsewhere in the solar system (fossil microbes on Mars, living chemistry under Europa's ice), the more bad news it is for our prospects. Every instance of independent abiogenesis outside Earth reduces the probability that the Great Filter is behind us — and therefore increases the probability that it is ahead. Civilizations, in this reading, reach the industrial threshold routinely. They also destroy themselves routinely, at some point between industrial capability and interstellar colonization, for reasons we have not yet identified but may be about to. Nuclear war, engineered pandemics, AI catastrophe, ecological collapse, something we have not imagined. The silence is not the sound of emptiness. It is the sound of graveyards.
Nick Bostrom, in a 2008 essay in Technology Review, put the conclusion as bluntly as possible: "I hope that our Mars probes will discover nothing. It would be good news if we find Mars to be completely sterile. Dead rocks and lifeless sands would lift my spirit."
Beyond the Great Filter, the Fermi question has generated a decades-long taxonomy of possible resolutions. The literature — summarized most exhaustively in Stephen Webb's If the Universe Is Teeming with Aliens... Where Is Everybody? (2002, expanded 2015) — catalogues seventy-five. A few of the most serious:
The Rare Earth hypothesis (Ward and Brownlee, 2000). Simple life is common, but the specific conditions that allowed complex multicellular life on Earth — a stable G-type star, a large Moon stabilizing the planet's axial tilt, plate tectonics, a Jupiter-class gas giant sweeping out most inbound comets, a position in the galactic habitable zone far from the densely populated and radiation-rich galactic center, the evolution of oxygenic photosynthesis through the Great Oxidation Event — are rare enough in combination that complex life is astronomically uncommon. Microbes yes; engineers no.
The Zoo Hypothesis (John Ball, 1973). Advanced civilizations are aware of us and have deliberately chosen not to make contact, either as a matter of ethics analogous to not interfering with endangered species or as a matter of collective policy enforced among themselves. The galaxy is not empty. It is supervised.
The Dark Forest (Liu Cixin, 2008). The galaxy is full of civilizations, but revealing your location is an existential risk: any civilization that can detect you can in principle destroy you, and at interstellar distances you cannot verify the intentions of a detected neighbor, so the game-theoretically stable strategy is silence and, if you must respond to an outside detection, pre-emptive destruction. The galaxy is a dark forest in which every hunter moves in silence, killing anything that reveals itself. Everyone remains silent because everyone who does not is dead.
The Transcension Hypothesis (John Smart, 2012). Civilizations that survive long enough develop the capacity to manipulate the structure of spacetime at small scales, and their energy-optimal path is inward — into computational substrates operating at the Planck scale, black-hole-adjacent environments, or dimensions we cannot observe. They do not expand outward across the galaxy because the physics of information rewards contraction. The civilizations we cannot see are not dead. They have left the observable universe.
The Aestivation Hypothesis (Sandberg, Armstrong, and Ćirković, 2017). The thermodynamic cost of computation falls asymptotically as the temperature of the cosmic microwave background cools. A civilization that has optimized for long-term computational output therefore has a strong reason to hibernate — to dormantly wait out the heat death of the universe's current thermal regime so that its future computations can be performed at exponentially greater efficiency. The silence is not emptiness or death. It is sleep.
The Simulation resolution. If we are in a simulation, the Fermi Paradox dissolves: aliens are not present in the program because they were not added to the program. The silence is not a physical fact to explain but a rendering decision at the simulation's root. This is the resolution explored directly by The Simulation Hypothesis.
The Berserker hypothesis (Fred Saberhagen / Frank Tipler, 1980). Self-replicating probes exist, but they were designed as weapons — dispersed by one ancient civilization to destroy any emerging intelligence. The galaxy is silent because every emerging civilization has been sterilized by machines left behind by a paranoid predecessor. The reason we have not seen the Berserkers is that we have not yet tripped the tripwire.
In 1964, the Soviet astronomer Nikolai Kardashev proposed a classification of civilizations by the energy they command. A Type I civilization uses all the energy available to a planet (humanity is roughly Type 0.7). Type II uses the full energy output of a star — potentially captured via a Dyson sphere, a shell of energy-collecting structures enclosing a star proposed by Freeman Dyson in a 1960 Science paper. Type III uses the full energy of a galaxy.
If even a handful of Type II civilizations existed in the Milky Way, we would expect to see infrared excesses from their Dyson spheres — the waste heat of a fully enclosed star. A systematic search for such signatures, using WISE infrared data across 100,000 galaxies, was conducted by Jason Wright and colleagues at Penn State and published in 2015 in The Astrophysical Journal. The result: no unambiguous infrared signatures consistent with Kardashev Type III civilizations in any surveyed galaxy. Upper limits suggest that if such civilizations exist, they are using less than 85% of the typical galaxy's stellar output — which is a strangely specific constraint for a K3 civilization to observe.
The tentatively anomalous case of KIC 8462852 ("Tabby's Star"), identified in 2015 by Tabetha Boyajian and colleagues, showed irregular luminosity dips of 20% — far too deep for any known transit — that briefly raised the possibility of a Dyson-sphere-like megastructure. Subsequent infrared observations ruled out the megastructure hypothesis; the dips are now widely attributed to dust clouds of uncertain origin. But the fact that the case briefly made scientific news is a measure of how much weight even a marginal candidate carries.
SETI's systematic radio searches have now operated for over sixty years. The Breakthrough Listen initiative, launched in 2015 with $100 million in funding from Yuri Milner, represents the most comprehensive survey to date: dedicated time on the Green Bank Telescope, Parkes Observatory, and the Automated Planet Finder, covering the nearest million stars and the plane of the Milky Way across a wide frequency range. The 2020–2023 data releases have produced no unambiguous extraterrestrial signals. One candidate, BLC1, a narrowband signal detected in 2019 during observations of Proxima Centauri, was determined in 2021 to be a terrestrial radio interference artifact.
This is a not-yet-negative result, not a confirmed absence. Breakthrough Listen's sensitivity has limits; frequency ranges remain unexplored; cultural assumptions about what a transmitting civilization would choose to broadcast are almost certainly wrong. But the cumulative weight of sixty years of non-detection across the full spectrum of what we have been able to examine is not zero. Something about the expectation generated by the Drake equation is in tension with the observed silence. Either the arithmetic is wrong, the detection methods are inadequate, or the civilizations are there but not signaling in ways we can hear.
The mainstream discussion of the Fermi Paradox has always been conducted under a specific assumption: that no credible evidence of extraterrestrial contact has been observed. That assumption has become contested. The December 2017 New York Times article on the Pentagon's Advanced Aerospace Threat Identification Program, the subsequent AATIP & The Pentagon UAP Disclosure of Navy Tic Tac videos, the 2023 Congressional testimony of former intelligence officer David Grusch, and the ongoing work of the All-Domain Anomaly Resolution Office represent a slow institutional concession that something has been observed, over multiple decades, that cannot be categorized as known-origin human technology or natural phenomena.
This repositioning does not answer the Fermi Paradox. But it changes which version of the paradox is being asked. The classical formulation assumes Fact A — that no extraterrestrials have visited Earth. If the Tic Tac UAPs, the 2021 USS Omaha video, and the AARO-documented transmedium objects are non-human, then Fact A is false, and the paradox becomes: not where is everybody, but why have they not introduced themselves. The serious answers to that narrower question converge on Zoo Hypothesis variants, Dark Forest variants, and (speculatively) the possibility that the observed UAPs are either ancient probes left by a long-ago visit (the Ancient Astronauts framing at maximal strength) or long-resident non-human intelligences that are neither interstellar colonizers nor ancient visitors but something for which our ontology has no current word.
What makes the Fermi Paradox so powerful is not any individual resolution but that every serious candidate answer fundamentally reshapes our view of our own situation. If we are alone, the value of human consciousness becomes something close to cosmic: in all this universe, only this spark. If civilizations die at our stage, the existential stakes of the twenty-first century are not exaggerated. If they are hiding, the galaxy is political in a way we have not yet learned to inhabit. If they transcend into substrates we cannot observe, the question of what civilization is for reopens at the deepest level. If they are already here, the institutional arrangements of the societies we inhabit are built on a lie that will eventually be forced into daylight. And if we are simulated, the silence was never a physical fact at all.
Fermi's question is a question in disguise. It asks where everybody is, but it answers, in whatever form the answer eventually takes, what kind of universe we live in. The arithmetic of cosmic absence is not a technicality. It is the largest piece of evidence we have about the nature of reality, and we do not yet know how to read it.