Posted by Some time in the summer of 1950, four physicists walked to lunch at Los Alamos joking about flying saucers. A New Yorker cartoon had blamed a rash of missing city trash cans on aliens. Enrico Fermi, Edward Teller, Emil Konopinski and Herbert York kicked the idea around, the conversation drifted elsewhere, and then, partway through the meal, Fermi interrupted himself with a question nobody at the table needed explained. Where is everybody? Three quarters of a century later that remark has a name, the Fermi paradox, and it is still doing quiet damage to our assumptions about the galaxy.
The bones of it go like this. The Milky Way holds a few hundred billion stars and has been assembling them for close to 13 billion years. Our Sun is a latecomer at 4.6 billion. If a small fraction of those older systems produced a technological species, and if one of them ever sent out ships or self-replicating probes at a decent fraction of light speed, the galaxy should have been crossed and re-crossed many times over. Michael Hart ran that calculation in 1975 and got a colonization time well under a million years.
Note what the argument does not require. It does not require aliens to be friendly, curious, or even alive. Only that somebody, once, left something behind.
The Part of the Argument That Actually Got Stronger
When Fermi asked his question, not one planet outside the Solar System had been confirmed. The first rung of the ladder was assumption dressed as reasoning. That ended in October 1995, when Michel Mayor and Didier Queloz announced 51 Pegasi b: a Jupiter-mass world circling a Sun-like star every 4.2 days, so close in that nobody had a category for it. They took a share of the 2019 Nobel Prize for it.
Then Kepler, which did one thing with enormous patience. From 2009 it stared at one patch of sky between Cygnus and Lyra, watching roughly 150,000 stars for four years for dips of a few hundredths of a percent. It found thousands of planets. Kepler-186f, in 2014, was the first Earth-sized world in the habitable zone of its star. February 2017 brought TRAPPIST-1, seven Earth-sized planets around an ultracool dwarf 40 light years out. And in 2016, Proxima b, orbiting the nearest star to the Sun.
So one term in the Fermi paradox argument is no longer speculation. Planets are common, and small rocky ones in temperate orbits appear to be common too. Everything downstream of that, whether chemistry becomes biology and biology becomes something that builds telescopes, is still guesswork.
Seven Terms, Three of Which We Can Fill In
Frank Drake wrote his famous equation on a blackboard in 1961, not as a prediction but as an agenda for a two-day meeting at Green Bank. It has been misread as prophecy ever since. The first three factors are pure astronomy, and sixty-five years of observation have given us real ranges for all of them.
The other four are the fraction of habitable worlds where life starts, the fraction where it turns intelligent, the fraction that builds detectable technology, and L, how long a civilization stays detectable. For those we have no measurement. Not a rough one. None. Feed the equation optimistic values and the galaxy holds millions of civilizations; feed it pessimistic ones and the answer comes back as one, and the formula shrugs.
The Filter Nobody Can Locate
Robin Hanson gave the problem its bleakest framing in 1996 with the Great Filter. Somewhere on the road from dead chemistry to a civilization that leaves marks visible across light years, there may be one step so improbable that almost nothing gets through.
Where that step sits matters more than anything else. Put it early, at the origin of life, or at the messy cellular merger that produced complex eukaryotic cells roughly two billion years ago and appears to have happened exactly once here, and the silence is what you would expect. We would be the freak result of a lottery nobody else won. Lonely, but not threatening. Put the filter after our current stage and the picture changes.
Which is why an independent second origin of life in our own Solar System, under the ice of Europa or Enceladus or in Martian rock, would be unsettling news wearing the costume of good news. It would mean the early steps are easy, which pushes the filter forward, toward us.
How Long Does a Civilization Stay Loud?
Humans have been radio-bright for less than a hundred years, and the bright phase may already be closing. High-power analog television and early warning radar leaked energy in every direction, including up. Digital broadcasting is far more efficient, traffic has moved to fibre and tight satellite beams, and Earth is going quiet in exactly the frequencies an alien astronomer would monitor.
If that pattern is typical, L is small. A society might go conspicuous for two or three centuries, then fade into efficiency: not dead, just no longer wasteful. Two species can share a galaxy and miss each other by ten million years. There is no way to test any of it. It is a hypothesis built on a sample of one.
Sixty-Five Years of Listening, and What That Covered
Drake pointed an 85-foot dish at Tau Ceti and Epsilon Eridani in April 1960 and listened near the 1420 MHz hydrogen line. Project Ozma heard nothing. Project Phoenix worked through around 800 nearby Sun-like stars between 1995 and 2004. Also nothing.
The famous near-miss is still the Wow! signal. On 15 August 1977 the Big Ear telescope at Ohio State recorded a narrowband burst near the hydrogen line, clean and strong, lasting the full 72 seconds the beam spent on that patch of Sagittarius. Jerry Ehman circled the printout and wrote “Wow!” in the margin. It has never been seen again, and a signal that never repeats confirms nothing.
A recent case shows how the field polices itself. Breakthrough Listen picked up a narrowband tone near 982 MHz at Parkes in 2019 while observing Proxima Centauri, drifting in frequency the way a transmitter on an orbiting planet should. Called BLC1, it made headlines in December 2020, and within a year the team had traced it to human electronics and published the debunk themselves. That same month Arecibo lost its cables and its platform crashed into the dish, so Green Bank, the Allen Telescope Array and the VLA carry the load now.
And here is the number that rarely makes the coverage. In 2018 Jason Wright and colleagues quantified how much of the total search space every SETI programme in history has actually covered, counting sky position, frequency, sensitivity and observing time. Their comparison was a hot tub of water drawn out of Earth’s oceans. Scoop up that much seawater, find no fish, and you would not declare the ocean sterile.
Looking for Exhaust Instead of Messages
Radio was always a choice of convenience, not a prediction about how an older species would behave. Freeman Dyson made the alternative case in 1960 in a two-page paper in Science. A civilization that keeps growing its energy use will eventually want a serious fraction of its star’s output, and whatever it does with that energy, thermodynamics insists it comes back out as waste heat. So look for stars dimmer than they should be in visible light and brighter in the infrared. Nobody has to be sending anything. Decades of searching have found no convincing candidate.
K2-18b is the cautionary tale of the moment. Teams reported hints of dimethyl sulfide in JWST spectra of that sub-Neptune, a molecule produced on Earth almost entirely by marine plankton, and independent reanalyses of the same data found it far weaker than claimed, or absent. JWST did not fail. A spectral bump at the edge of the noise is not a discovery.
Where an Answer Might Actually Come From
Astrobiology works the question from underneath, and its record checks enthusiasm. In 1996 NASA held a press conference about ALH 84001, a Martian meteorite containing structures some researchers read as microfossils. The consensus moved away from biology, but that fight launched modern astrobiology more effectively than any grant. Curiosity has measured methane in Gale Crater that rises and falls with the seasons, while the Trace Gas Orbiter, looking down from above, has struggled to see it at all. Still unresolved, like the 2020 report of phosphine at Venus.
OSIRIS-REx dropped its capsule into the Utah desert in September 2023 with material from the asteroid Bennu that turned out to contain organic compounds and amino acids. Not life, but it narrows the question of where the ingredients come from. The instruments arriving this decade change what is answerable: the Extremely Large Telescope in Chile, with its 39-metre mirror, should characterize atmospheres of rocky planets around nearby stars, and the Square Kilometre Array will make wide technosignature surveys routine.
None of it guarantees a resolution. Maybe technological civilizations are vanishingly rare. Maybe they are everywhere and quiet. Maybe we started looking eighty years ago and that is simply not long enough. The Fermi paradox is less a puzzle waiting for a clever answer than a standing reminder of how thin the evidence is and how much of the sky nobody has checked. If you want to follow that work as it happens, the surveys, the false alarms, the retractions, the occasional genuinely strange result, SETIworld is where we keep track of it. Come and argue with us about where the filter sits.
