Adam Frank on Armchair Expert

Armchair Expert with Dax Shepard, December 2023. Guest Adam Frank — astrophysicist, and principal investigator on NASA’s first grant to study technosignatures. Promoting The Little Book of Aliens. Also on Spotify. Transcript supplied by Kyle 2026-09-01; structured capture with proportionate quoting.

Why this belongs to “The Question Marks of Science”

The episode is unusually good on how a question moves from unaskable to answerable, which is the draft’s subject. Frank’s framing:

“We’re the last generation that doesn’t know the answer to the question are we alone — which also means we may be the first generation to have an answer.”

For 2,500 years the question was argued on opinion alone. What changed is not philosophy but instrumentation.

The three revolutions that made astrobiology a science

1. Exoplanets (1995). Before that, the honest answer to “are there other planets” was no data — the prevailing theory was that planets required two stars to nearly collide, so they should be vanishingly rare. Within a decade of the first detection there were thousands; now enough for statistics. Roughly one in five stars has a planet in the habitable zone. Notably the breakthrough wasn’t a bigger telescope but a better instrument — “you got to beat the crap out of the light.”
2. The solar system, visited. Robots have reached every type of body; some we have landed on and scooped. So we have real comparative planetology rather than one example.
3. Earth’s own history, unpacked. The load-bearing one: life doesn’t happen on a planet, it happens to a planet. The Great Oxidation Event is the case — photosynthesis that splits water and vents oxygen rewrote the atmosphere, the chemistry of the rocks, and the trajectory of everything since. It was also an apocalypse for the life already there, oxygen being corrosive.

That third revolution is what makes the first two useful: it tells you what to look for from a distance. If life reorganises a planet, the planet’s atmosphere is the signal.

How you actually detect it

Atmospheric characterisation, explained cleanly: when a planet transits its star, some starlight passes through the planet’s atmosphere on the way to us, and specific wavelengths get absorbed. You see the absences. That fingerprint is a biosignature — oxygen and methane together being the sought-after pair, because oxygen needs continuous replenishment to persist.

Technosignatures are the other half, and Frank’s NASA group builds the library. CFCs are the flagship: either an industrial by-product or deliberately deployed to warm a planet, and hard to produce without chemistry nature doesn’t do. Others discussed: city lights (sodium-lamp spectra), mass solar panels, geosynchronous satellite rings, and interplanetary laser communications sweeping past.

The best correction in the episode: SETI’s “great silence” is an artifact of funding

The popular belief is that scientists have been listening every night for decades and heard nothing. Frank says that is simply false — SETI was marginal, unfunded, and career-damaging (the “giggle factor”), so the search barely happened. His colleague Jason Wright added up every SETI search ever conducted and scaled it against the search space:

If the stars are an ocean and life is fish, we have searched one hot tub — and found no fish. “So are you then going to be like, well there’s no fish in the ocean?”

That is a general lesson about null results, and the most portable thing in the episode: absence of evidence is only evidence of absence if you actually looked.

On UFOs and UAPs — the standard-of-evidence passage

Frank is not dismissive of the phenomena, only of the evidence:

“There is zero evidence of the kind that you need to build a cell phone with… or have hip replacement surgery that doesn’t kill you — that level of science — to connect the uaps to anything related to extraterrestrial or alien life.”

The specifics: the 2017 New York Times story broke the taboo; a NASA panel reviewed roughly 800 cases and found ~94% explainable; one of the famous videos, read off its own instrument numbers, showed an object moving at about 40 mph; the one that appears to rotate impossibly was reproduced by Mick West as the camera rotating. The residue is a few genuinely strange cases worth study — but “they’re like ghost stories… that’s not enough to say aliens.”

His own origin story on this is the good part. As a kid he believed von Däniken’s Chariots of the Gods — until a documentary interviewed the actual archaeologists. On the Easter Island heads, supposedly unmovable because the island has no trees:

“It turns out they chopped down all the trees to move the stone heads. So instead of Ancient Aliens you get a story of human stupidity.”

And the honest inversion, from his own paper with Gavin Schmidt: could you tell if an industrial civilisation had existed on Earth 100 million years ago? No — the surface resurfaces in ~2 million years and almost nothing fossilises. So there is no scientific evidence we have been visited, and none that we haven’t. He is careful not to let that become a licence: “we have no scientific evidence that an elephant hasn’t flown across a canyon.”

Two arguments worth arguing with

The optimist’s case. Reworking the Drake equation with modern exoplanet data, Frank and Woody Sullivan computed the number of habitable-zone planets that have ever existed: 10 billion trillion — a 1 with 22 zeros. His claim is that for us to be the only instance, the experiment must have failed on every one of them.

The pessimist’s reply, which the hosts supply and he concedes, is the timing problem: microbes appeared almost as soon as Earth cooled, but multicellular animals took another ~3.5 billion years, and the Cambrian explosion remains unexplained. So the bottleneck may not be life but complexity — and civilisations occupy a sliver of a planet’s timeline in any case. Frank’s own Fermi-paradox simulations concede the point: if civilisations die, you get pockets of the galaxy empty for 20 million years at a stretch.

The parts most useful outside astronomy

• Against the God’s-eye view. “Science is based on this idea that there’s a third-person view… third person view is a fantasy. It’s a useful fantasy, but there is no such thing — nobody’s ever had the God’s eye view.” He is explicitly anti-reductionist, and ties the “you’re just a bunch of atoms, you’re a meat computer” framing to real harm.
• Redefining intelligence. Early SETI assumed intelligence meant something brain-shaped, because it was designed by “a bunch of white dudes who were reading a lot of science fiction.” The current frontier is liquid brains — ant colonies, fungal networks, octopuses with neurons distributed through their arms. Self-organising complex systems, not centralised skulls.
• Light delay as a way of seeing. “I’m not seeing you as you are, I’m seeing you as you were, like a nanosecond ago… there’s just these concentric shells of pasts.” Which he links to his own Zen practice: the present is “kind of a weird fiction,” something assembled rather than given.
• On climate, the framing worth stealing. “Our job is to not piss it off” — the biosphere “will shake us off like fleas on a dog.” We are not saving the Earth; we are saving the conditions we happen to need. He also rejects the guilt frame as both ineffective and untrue: an individual cannot lay a power grid, so the lever is infrastructure and votes, not shame. ⚠️ He pairs this with a claim that the transition is being held back mainly by incumbents protecting profits — plausible, but stated without the counter-argument getting a fair hearing.
• Why any of it matters. His answer is the Copernican revolution: nothing physically changed in 1543, but the reframing was a building block for the Renaissance, the Enlightenment and the Reformation. Finding life elsewhere would be “the Copernican revolution times a thousand.”

Connections

• The Question Marks of Science — this is the anchor source. The draft’s subject is questions science can and cannot yet touch, and Frank supplies a worked example of one crossing over, plus the “hot tub” correction on how to read a null result.
• The instrumentation-before-answers pattern connects to Forecasting and Predicting The Future: the 2,500-year argument was unresolvable not because people reasoned badly but because they had no instrument. Compare 50 Years Later: Remembering How the Future Looked in 1974 — technologists predict infrastructure well and consequences badly; Frank is describing the same asymmetry from inside.
• The Gavin Schmidt “could you tell?” paper (the Silurian hypothesis) belongs with Historical Futurism — a rigorous argument about the limits of the archaeological record.
• The anti-reductionism and “life is the only thing that innovates” material bears on Motivation and the agency question; his Templeton-funded project asks what makes life different as a physical system.