Posted by On 15 August 1977 a telescope in Ohio built out of aluminum panels and a great deal of stubbornness recorded a burst of noise that lasted 72 seconds and never came back. Jerry Ehman went through the printout days later, circled six characters and wrote one word beside them in red pen: Wow. Half a century on, that scribble is still the most famous thing the search for alien signals has produced, and it is still unexplained. It also says something honest about what radio telescopes are. They do not find anything. They turn faint energy into numbers, and the hard part falls to people.
The Big Ear observatory was demolished in 1998 to make room for a golf course. Nobody has heard that signal again.
Chicken Wire, Steel and a Lot of Patience
The field started by accident. In 1932 Karl Jansky, an engineer at Bell Telephone Laboratories, was hunting for the static that spoiled transatlantic phone calls. He built a rotating antenna from brass pipe and lumber on wheels off a Model T, and found a hiss that rose and fell every 23 hours and 56 minutes. That is the sidereal day, not the solar one, so the source was neither on Earth nor on the Sun. It came from Sagittarius, the center of our galaxy.
Bell Labs moved him on to other work. An amateur in Wheaton, Illinois named Grote Reber read the papers, built a nine-meter dish in his back yard in 1937, and mapped the radio sky more or less alone.
The basic idea has not changed. A curved surface gathers radio waves over a large area and focuses them onto a receiver, which amplifies a signal so weak that everything collected by every radio observatory in history would not warm a cup of coffee. The numbers really are that small. Bigger surface, fainter sources within reach. Amplifiers chilled to a few degrees above absolute zero keep the instrument’s own noise from drowning the sky.
Radio telescopes do not care about sunset. They work through daylight, cloud and rain, which is why the Green Bank Telescope in West Virginia, a hundred meters across and fully steerable, observes around the clock. Human noise is the real problem. Green Bank sits inside the National Radio Quiet Zone, where transmitters are restricted by law and staff drive a truck around the valley hunting down leaky microwave ovens.
Why Radio Won the Argument
Two physicists made the case in 1959. Giuseppe Cocconi and Philip Morrison argued in Nature that if anyone wanted to signal across interstellar distances, radio was the cheap way to do it: it crosses gas and dust that scatter other wavelengths, it moves at light speed, and the sky is quiet in the microwave band. They even proposed a frequency. 1420 MHz, the emission line of neutral hydrogen, a landmark any technological species knows.
A young astronomer at Green Bank had been thinking the same thing. In the spring of 1960 Frank Drake pointed an 85-foot dish at Tau Ceti and Epsilon Eridani and listened near the hydrogen line for a few weeks. Project Ozma heard nothing. The next year Drake chalked his famous equation on a blackboard as an agenda for a small meeting there, and people have argued over it since, because most of its terms are still guesses dressed as variables.
The band between the hydrogen line and the hydroxyl line at 1662 MHz picked up a nickname: the water hole. H plus OH. The pun is dreadful, the logic under it less so.
What Counts as a Suspicious Signal
Nature is loud, but nature is broad. Pulsars, quasars, collapsing gas clouds and the hot edges of black holes shout across wide swathes of frequency. Very little in astrophysics produces a tone squeezed into a few hertz. Transmitters do it constantly, because packing power into a narrow band is how you get heard. So narrowband is the first filter.
The second is motion. A transmitter on a rotating, orbiting planet is never still relative to us, so its frequency should slide in a smooth, predictable way over minutes, and the software looks for that drift. Signals that refuse to move are usually sitting next to the telescope, riding around with it.
Unusual is not the same as alien. Every candidate has to be argued down before anyone is allowed to argue it up.
The Microwave Oven in the Kitchen
For seventeen years the Parkes dish in New South Wales picked up strange millisecond bursts that looked tantalizingly like the fast radio bursts turning up in distant galaxies. They were nicknamed perytons. They clustered around lunchtime. In 2015 a team led by Emily Petroff traced them to the observatory’s own microwave ovens: open the door before the timer finishes and the magnetron fires a dying pulse into a very sensitive receiver.
The story is funny and completely typical. Satellites, radar, cell networks and unshielded laptops in the control room all get heard by sensitive radio telescopes, and much of it looks artificial for the excellent reason that it is.
The most interesting candidate of the modern era failed exactly this test. In 2019 Breakthrough Listen, observing with Parkes, recorded a narrowband tone near 982 MHz while pointed at Proxima Centauri, host of Proxima b, a roughly Earth-mass planet found in 2016. It drifted the way a planet-bound transmitter should, and seemed to appear only on target. The team labeled it BLC1, spent over a year picking it apart, and published in 2021 the conclusion that it was almost certainly human interference.
Slow, deflationary, published anyway. That is what the process is supposed to look like.
One Dish, or Many
Radio telescopes come in two broad shapes, and the choice is a trade. Arecibo’s 305-meter bowl, slung across a sinkhole in Puerto Rico, was the sensitivity champion for half a century and the workhorse of Project Phoenix, the privately funded survey that examined roughly 800 nearby stars between 1995 and 2004. Its cables began failing in 2020 and the platform came down that December. The loss still stings.
The alternative is many smaller dishes wired together. Interferometry combines signals from separated antennas so they behave like one instrument as wide as the gap between them, buying sharpness rather than collecting area. The Very Large Array spreads 27 dishes along three arms of New Mexico desert. The Event Horizon Telescope took the trick to its limit, linking observatories on several continents into an Earth-sized aperture to image a black hole’s shadow in 2019.
The Allen Telescope Array at Hat Creek in northern California is the only major facility ever built with technosignature searches as a founding purpose. It has 42 small dishes rather than the 350 originally planned, but its time belongs to the search instead of being borrowed in the gaps. Elsewhere the strategy has shifted toward riding along. Systems now tap the data flowing off the Very Large Array while it does ordinary astronomy, and Breakthrough Listen has a similar deal with MeerKAT in the Karoo, precursor to the Square Kilometre Array now rising in South Africa and Australia.
China’s 500-meter FAST dish in Guizhou inherited Arecibo’s title, though not its transmitter.
Somewhere Specific to Point
Drake had to guess which stars deserved his few weeks of telescope time. That changed in 1995, when Michel Mayor and Didier Queloz found 51 Pegasi b tearing around a Sun-like star every four days, and again as Kepler and TESS turned exoplanets into a census of thousands of worlds.
Target lists now write themselves. Kepler-186f and Kepler-452b, rocky planets in or near their stars’ habitable zones. The seven worlds of TRAPPIST-1, confirmed in 2017 around a dim red dwarf some forty light-years off. K2-18b, where JWST spectra have been read by one team as a hint of dimethyl sulfide and by others as showing nothing of the kind, a dispute that shows how thin the evidence gets at these distances.
Nobody Has Heard Anything Yet
Breakthrough Listen has worked through more than a thousand nearby stars since 2015 and reported no technosignatures each time. Every dedicated search before it ended the same way. The silence is a real result, if a modest one, since it rules out only loud, persistent beacons aimed our way from a tiny sample of a galaxy holding hundreds of billions of stars. Jill Tarter, who ran Project Phoenix, likens the effort to dipping one glass from an ocean and declaring it fishless.
Turn the question around and it gets uncomfortable. Earth has leaked radio for a century, but broadcasts are weak, diffuse and increasingly replaced by cable and beamed satellite links, so a civilization with our equipment would struggle to pick our chatter out from a few light-years away. Planetary radar is another matter. Arecibo’s transmitter threw about a megawatt into a narrow beam, and in 1974 it sent a coded message toward the cluster M13.
What improves from here is mostly arithmetic. More antennas, wider bandwidth, faster processing, machine learning trained to cut millions of hits per observation down to a pile a human can inspect.
There is no promised ending here. Nobody knows whether the sky holds a signal at all, and radio telescopes may go on hearing nothing for another sixty years. They will keep listening, because the alternative is leaving the question open on purpose. If you want to follow what the dishes pick up, the candidates and the retractions, the fights over K2-18b, the first data out of the Karoo, that conversation is happening at SETIworld, and there is room for one more pair of ears.
