The Signal Worth Chasing

That is the news, and it sounds like failure. It is not. It is one of the more quietly magnificent things we did this month, and it happens to arrive at the exact moment a good-faith skeptic is asking, out loud, whether we should be doing it at all.

The point of light is K2-18b, a planet orbiting a red dwarf in the constellation Leo — a "Hycean" world candidate, possibly draped in a hydrogen atmosphere over a global ocean, its air laced with carbon dioxide and methane. A couple of years ago it briefly detonated the internet when the James Webb Space Telescope caught a faint chemical whisper of dimethyl sulfide, a molecule that, here on Earth, is made almost exclusively by living things. It was a three-sigma hint, not a five-sigma confirmation — the statistical equivalent of hearing your name in a crowded room and not being sure. Other teams promptly showed that dead chemistry could produce the same whisper. The signal was real; the certainty was not. And so K2-18b became the planet we cannot stop thinking about.

This summer, two of the world's great radio instruments — the Very Large Array in New Mexico and the MeerKAT array in South Africa — turned toward that planet and asked a different, blunter question. Not is there life? but is anyone broadcasting? They swept the narrowband radio frequencies, the kind of tight, artificial signal our own civilization leaks and beams. Software named COSMIC and BLUSE strained out the roar of Earth's own transmitters. Millions of candidate blips were run through five separate screens. And at the end of it: no convincing technosignatures. Nothing. The search set an upper limit — if something over there is transmitting, it is doing so more quietly than the old Arecibo dish could have shouted.

So we spent enormous effort, coordinated across two hemispheres, to establish that a specific corner of the sky is, as far as we can tell, radio-silent. A skeptic might reasonably ask what, exactly, we bought.

i · the economics of listening to nothing

The skeptic is not a strawman. In mid-July 2026, the neurologist and science writer Steven Novella fielded exactly this challenge on his NeuroLogica blog: a reader argued that finding an alien species we could actually talk to is so improbable — requiring some cosmic parallel evolution to hand us a conversation partner — that SETI is not worth pursuing on economic grounds. Why fund a lottery ticket whose jackpot may not exist?

It is a fair question, and it deserves better than a starry-eyed dodge. Novella's answer is worth sitting with, because it is a case study in how to think clearly under maximum uncertainty.

First: we are working from a sample size of one. We have exactly one example of a life-bearing planet, one example of a technological civilization, and that civilization is us, reading this sentence. From an N of 1 you cannot compute a probability. You cannot say the odds are low; you cannot say they are high; you can only say you do not know, and that anyone claiming a firm number is smuggling in intuitions dressed as math. What little data we have cuts against the skeptic's confidence: intelligence, in the loose sense of flexible problem-solving, evolved more than once on our own planet — in primates, in cetaceans, in the octopus, in crows. Nature does not seem to consider intelligence a freak accident. It keeps reaching for it. That is not proof, and it is important to say where it stops: flexible minds arising repeatedly inside a single biosphere, from shared chemistry and shared ancestry, tell us nothing directly about the odds of life igniting elsewhere — and even less about the far longer-odds jump from a clever animal to one that builds radios. It is not evidence the jackpot exists. It is only a thumb laid gently against the skeptic's side of the scale, where he had assumed the pan was empty.

Second: two species need not be evolutionary cousins to communicate. They need a shared language, and the universe already ships one. Mathematics and physics are the same everywhere; a prime number is a prime number in Leo as surely as in Los Angeles. A civilization that wanted to be found could build a signal out of those universals, addressed to anyone with the wit to factor it. The conversation would be stilted, sure. But "stilted" and "impossible" are not the same word.

Third, and this is the part the accountants miss: a negative result is not a wasted result. SETI does not become worthless if it fails to find aliens, because along the way it is radio astronomy — real instruments pointed at real objects, generating real data about the electromagnetic sky. The search for a signal and the study of the cosmos are the same activity wearing two hats.

And fourth, the principle that ties it together: a mature scientific portfolio does not bet everything on sure things. It balances a mountain of high-probability, incremental research against a handful of low-probability, civilization-altering long shots. You fund the safe work because it reliably pays. You fund a little of the wild work because the one time it hits, it rewrites the human story. Refusing the long shot entirely is not prudence. It is a failure of nerve disguised as budgeting.

ii · a negative result is still a result

Here is where the K2-18b silence stops looking like failure and starts looking like exactly the kind of dividend Novella is describing.

The scan did not find ET. What it did was carve away a piece of ignorance. Before this summer, "is anyone broadcasting from K2-18b?" was a total unknown — a shrug. Now it is a bounded unknown: whatever might be there is quieter than Arecibo, within the frequencies we searched, during the window we listened. That is not nothing. That is the difference between a blank map and a map with one region marked surveyed, no dragons found at this resolution. Science advances at least as often by ruling things out as by ruling them in. Every "no" tightens the shape of the eventual "maybe."

And the search proved something about the searcher. Millions of candidate signals, filtered by automated systems robust enough to separate a genuine anomaly from the endless self-noise of a planet drowning in its own radio chatter — that is a rehearsal. The instruments and the software just demonstrated they can handle the data firehose that the next generation of surveys will unleash. We didn't only listen to K2-18b. We tuned the ear.

But honesty about the dividend requires honesty about the target, and here the scan is on shakier ground than the wonder wants to admit. Aiming these dishes at K2-18b specifically was not cold optimization. A biosignature and a technosignature are unrelated questions — a hydrogen-and-methane ocean world has no special reason to host broadcasters any more than a tide pool has reason to host a telegraph. The choice of K2-18b rode the Webb dimethyl-sulfide publicity wave at least as much as it followed any ranked list of promising targets. That does not make the search worthless; it relocates the worth. The dividend — that a null buys knowledge, that the instruments got their rehearsal — is true of SETI as a discipline, and mostly independent of whether this particular pointing was wise. The listening earns its keep as method. The address it happened to be listening to earns rather less.

There is a doubling here worth naming precisely, because it is easy to over-read. The same planet gave us, within a couple of years, a tentative chemical yes from Webb and a firm radio no from the VLA and MeerKAT. Both are true — but they are not in tension, and it would be a cheat to make the silence feel profound by pretending they were. A world can be, at once, chemically intriguing and technologically silent, and that is not a paradox; it is the single most likely description of a living planet. Earth itself was chemically screaming with biology for billions of years before it emitted a single deliberate radio watt. A planet can be drenched in life and utterly quiet. Life and broadcasting life are separated by four billion years of getting around to it — which means the radio "no" was never really a comment on the chemical "yes" at all. The two instruments answered two different questions, and only one of them was ever going to answer quickly.

iii · the coherence question

Strip away the hardware and SETI is not really a question about them. It is a question about us.

To fund a multi-decade search whose payoff might land next year or might land never — or might, most likely, land on a generation not yet born — a civilization has to be capable of an unusual act of self-location. It has to be able to hold the thought: we are one instance of a phenomenon we do not understand, embedded in a context we cannot see the edges of, and finding out where we sit is worth some of our treasure. That is not a scientific capacity. It is a temperamental one. It asks whether a species can care about a question that is bigger than its own attention span.

This is the coherenceism angle, and it is sharper than it first appears. A coherence is not measured by how tidy it is on the inside; it is measured by how much of what is real it manages to include. A civilization that refuses to look outward because the answer might not arrive on time has drawn its circle of concern tightly around the present quarter. A civilization that keeps a few dishes pointed at the dark — not from certainty, but from the discipline of admitting it does not know its own cosmic neighborhood — has widened that circle to include the parts of reality it cannot yet touch. The listening is the humility made operational. It is what intellectual honesty looks like when you build it out of steel and put it in the desert.

But turn that same standard back on the listening, and it bites. To listen is cheap and safe; to broadcast is costly and exposing. A civilization that only ever cups its ear to the dark — that wants to include the reality of others while offering none of its own to be included — has built itself a half-coherence: it seeks to see without consenting to be seen. There is a name for the instinct underneath that posture, the dark-forest suspicion that the safest move in an unknown universe is silence. And if that is the posture, then the humility is not yet complete — it is humility that costs us nothing. The fuller version would eventually have to answer the harder question the listening lets us defer: not just whether we will keep asking, but whether we will ever consent to be found. For now we only listen, which is an honest enough place to begin. But knowing that our humility is, so far, the free and one-directional kind is part of the price of listening clearly.

The skeptic's instinct is not wrong, exactly. It is just small. It optimizes for a payoff we can see and price, and the whole point of the exercise is that the most important thing about our situation might be something we currently cannot see or price at all. You do not find the edge of the map by only walking the parts already drawn.

So we listened to K2-18b, and it said nothing back, and that silence cost real money and returned no aliens. And it was worth it — not because this particular planet was a clever bet, but because the alternative is a species clever enough to build the ear and too frightened of an empty answer to ever cup it to the sky. The void, as it happens, is mostly silence. The interesting question was never whether it would answer. It was whether we would keep asking — and, someday, whether we will dare to speak. This month, gorgeously, expensively, against the odds, we at least kept asking.