The frontier
What is still argued over.
- First look
- How it works
- The physics, still to be written
- The frontier
The frontier
What is still argued over
Three piles. No life has been found anywhere but the Earth. Everything written on the subject falls into one of three piles: things that have been measured, claims that were made and then disputed, and guesses. This level keeps the three apart, and each heading says which pile it is.
Measured: planets are common. One survey used gravitational lensing, in which a star and its planets briefly magnify the light of a more distant star behind them. It concluded that stars have planets as a rule. It found that 17 per cent of stars have a planet of about Jupiter’s mass between 0.5 and 10 times the Earth’s distance from the Sun, and that smaller planets are commoner still.
Measured, with wide error bars: rocky planets at the right distance. The Kepler telescope’s survey gives the number of planets between 0.5 and 1.5 times the Earth’s radius, in the habitable zone, for each star of roughly the Sun’s temperature (4,800 to 6,300 K):
| Definition of the habitable zone | Planets per star |
|---|---|
| Conservative | Between 0.37 (+0.48, −0.21) and 0.60 (+0.90, −0.36) |
| Optimistic | Between 0.58 (+0.73, −0.33) and 0.88 (+1.28, −0.51) |
The two ends of each range come from two extreme ways of allowing for planets Kepler would have missed, and the errors are large because few such planets were found. The same study estimates that the nearest one, round a star of the Sun’s kind or the kind a little cooler, is about 6 parsecs away, and that about four lie within 10 parsecs. By our own sum those distances are 20 and 33 light years.
What this figure does not say matters as much. A planet in the habitable zone gets the right amount of starlight for liquid water. Whether it has water, air or life has not been measured for any of them.
Measured: some gases in one distant atmosphere. K2-18 b is a temperate planet smaller than Neptune. The Webb telescope found methane in its air at 5 sigma and carbon dioxide at 3 sigma. Neither gas needs life to explain it, and the team that found them did not claim it did. The dispute described below is about a different gas.
Claimed and disputed. Each of the following was reported in a serious journal, and each ran into trouble when others checked it.
| Claim | What was measured | What happened next |
|---|---|---|
| The Martian meteorite ALH84001 | Carbonate globules, magnetite grains and carbon compounds, argued to be the work of microbes | Others built a fuller case that all of it could form without life. The consensus now is that the features are probably not signs of life |
| Phosphine on Venus, 2020 | One absorption line at radio wavelengths, read as about 20 parts in a billion of a gas with no known way of forming there. Its authors offered unknown chemistry, or life, as the source | An independent team found that the line cannot be told apart from one of sulphur dioxide beside it, and set upper limits that disagree with the claim |
| Dimethyl sulphide on K2-18 b, 2023 and 2025 | A hint in 2023 that its authors called of low significance. In 2025, 3 sigma evidence for that gas or a close relative, from a different instrument | A joint analysis of all the data found insufficient evidence for either gas. Other molecules, such as ethane, fit equally well. It estimated that about 25 more passes of the planet in front of its star would be needed to tell the claimed feature from a flat line at 3 sigma |
| Spots on a Martian rock, 2024 and 2025 | Two iron minerals, vivianite and greigite, in a pattern that on Earth microbes can produce, in rock made of clay and silt and rich in organic carbon | NASA calls it a potential biosignature. The minerals can also form without life, and the team says that cannot be ruled out. The sample is one of 27 rock cores the rover has collected, and it is still on Mars |
| A radio signal from the direction of Proxima Centauri, 2019 | A narrow signal near 982 MHz | Traced to interference from electronics near the telescope |
A second study of the K2-18 b claim made a wider point. Whether a faint gas appears to be detected depends on which other gases the computer model is allowed to try. When more were allowed, the claim vanished.
A scale for claims. In 2021 a group of NASA scientists proposed that evidence for life should be reported on a ladder of confidence, not as yes or no. Their example has seven steps. The wording below is ours, following their paper.
| Step | What has been shown |
|---|---|
| 1 | A signal of a kind that life is known to produce |
| 2 | Contamination ruled out |
| 3 | The setting is one where a biological explanation makes sense |
| 4 | Every source that is not life ruled out |
| 5 | A second, independent signal, or checking by independent investigators |
| 6 | Several independent lines of evidence, with rival explanations dismissed |
| 7 | Follow-up observations, made because of the earlier steps, that find what life would predict |
By the authors’ own reckoning the meteorite reached step 3 and no higher. They judged in advance that a rover on Mars could reach step 3, and that only samples studied on Earth could reach step 6. They single out step 4 as especially hard for planets that can only be studied from afar. NASA has not placed the Martian rock of 2024 on the scale. But its team says explanations without life cannot be ruled out, and ruling them out is step 4, so by the scale’s own terms it has not passed that step.
Why one gas is never enough. Oxygen was once thought a sure sign of life. A review of the question found it fails in both directions. Life on Earth may have gone several billion years without filling the air with oxygen, so a living planet can show none. And there are ways for a lifeless planet to build up a good deal of it. Oxygen counts as evidence only when the rest of the planet’s circumstances are known well enough to rule those out.
Guesswork: the number of civilisations. In 1961 Frank Drake wrote the question as a product of seven factors:
is the number of civilisations in the Milky Way giving out signals that could be detected.
| Factor | Meaning | State of knowledge |
|---|---|---|
| Suitable stars formed each year | Measured. It was the only factor known in 1961 | |
| Fraction of those stars with planets | Measured. The 1961 guess, close to all of them, agrees with modern surveys | |
| Planets in each system fit for life | Partly measured. The Kepler figure above counts planets of the right size at the right distance, not planets known to be fit for life | |
| Fraction of those where life appears | Unknown. One example, the Earth | |
| Fraction of those where intelligence arises | Unknown | |
| Fraction of those that give out detectable signals | Unknown | |
| Years for which they go on doing so | Unknown. We have been doing it for less than a century |
Four of the seven have never been measured, and the answer is only as good as the worst of them. Published values of run from 1, meaning only us, to several million. Drake’s own figure of 10,000 comes from supposing one new civilisation a year and a lifetime of 10,000 years. The SETI Institute says plainly that the equation cannot be solved or even accurately calculated. It is a list of what would need to be known. Any number put on today is speculation.
Measured: how much silence. Between 2016 and 2019 one search listened to 1,327 nearby stars at radio frequencies from 1.10 to 3.45 GHz, the most thorough search of that band up to then. After setting aside interference from Earth it was left with no candidates at all. That is a real measurement with a real limit. For those stars, in that band, while the telescopes were pointed at them, there was no transmitter stronger than about watts, or for the stars watched with the smaller of the two telescopes. It says nothing about weaker transmitters, other frequencies, other times or the rest of the galaxy’s stars.
Where it stands.
- Measured: planets are common, and rocky ones at the right distance for liquid water are probably not rare.
- Measured: no confirmed sign of life, past or present, on any other world, and no confirmed signal.
- Disputed: a potential sign in one Martian rock, which needs a laboratory on Earth, and a claimed gas on K2-18 b, which the fuller analysis does not support.
- Speculation: how often life starts, how often it becomes intelligent, and how long civilisations last. Nothing measured so far bears on these.
Where this comes from
- Bryson and others 2021, rocky habitable-zone planets round Sun-like stars from Kepler data
- Cassan and others 2012, one or more bound planets per Milky Way star
- Madhusudhan and others 2023, carbon-bearing molecules in the atmosphere of K2-18 b
- Madhusudhan and others 2025, new constraints on DMS and DMDS on K2-18 b
- Luque and others 2025, insufficient evidence for DMS and DMDS on K2-18 b
- Welbanks and others 2025, challenges in the detection of gases in exoplanet atmospheres
- Greaves and others 2020, phosphine gas in the cloud decks of Venus
- Villanueva and others 2021, no evidence of phosphine in the atmosphere of Venus
- NASA 2025, Mars rover discovered a potential biosignature
- Green and others 2021, call for a framework for reporting evidence for life beyond Earth
- Meadows and others 2018, oxygen as a biosignature in the context of its environment
- SETI Institute, the Drake equation
- Price and others 2020, the Breakthrough Listen search of 1,327 nearby stars
- Sheikh and others 2021, analysis of the Breakthrough Listen signal blc1
The words are written for this site. Sums in the text are worked from the figures in these sources.