What is life? Scientists redefine it as information and planetary process

The Short Version

The textbook answer to what is life? is a checklist: metabolism, reproduction, a cell membrane, Darwinian evolution. Three scientists argue that the checklist is the wrong tool. Astrophysicist Adam Frank says life is a planetary process, something that "happens to a planet," co-evolving with its geology and atmosphere, so we should look for whole planets pushed out of balance [1]. Astrobiologist Sara Walker defines life as "the process of how information structures matter over time and space," making lineages, not individuals, the real unit [2]. Chemist Lee Cronin, her collaborator on assembly theory, gives that idea a measurement: complex objects that exist in many identical copies can only come from selection, because "complexity in abundance is evidence of selection" [3]. The shift matters in practice. It moves the search for alien life away from Earth-like chemistry and toward signs that don't assume our biology: planetary disequilibrium, molecular complexity, and structures built by history [1][2][3].

Table of contents

What They Said

Adam Frank: life is something that happens to a planet

Frank, who leads NASA's first grant to search for technosignatures on exoplanets, starts from a correction to how people usually imagine habitability. "You can't just be like, oh, there's a warm pond. Then some interesting chemistry happens in the warm pond," he says. "You actually have to think about the planet as a whole and what it's gone through in order to really understand whether a planet is a good place for life or not" [1].

His evidence is Earth's own history. Plate tectonics, through volcanism and carbon dioxide, likely pulled early Earth out of snowball phases when it was nearly all ice [1]. About a billion years ago, a more vigorous form of plate tectonics began building true Himalayan-style mountains, and their erosion washed nutrients into the oceans [1]. Net primary productivity, how much photosynthesis the bottom of the food chain was doing, "shot up by a factor of almost a thousand," so that plate tectonics "supercharged evolution in some sense" [1]. After the boring billion years of microbes, that turmoil helped set the stage for the Cambrian explosion [1].

The influence runs both ways. Oxygen photosynthesis, invented about two and a half billion years ago, filled the atmosphere with the oxygen that large-brained creatures need [1]. Frank's summary of this, borrowed from colleagues: "Life doesn't happen on a planet. It happens to a planet. This is something that David Grinchpoon and Sarah Walker both say and I agree with this" [1]. He treats James Lovelock and Lynn Margulis's Gaia theory, once mocked as new age, as a scientific idea about feedbacks between life and planet. Whether true Gaian control exists is unclear, but the biosphere does a huge amount of the regulation of the planetary state and over billions of years has strongly modified the evolution of the planet [1]. With Walker and Grinspoon he wrote about planetary intelligence, arguing that a mature biosphere is autopoietic, Francisco Varela's term for self-creating and self-maintaining systems, while today's technosphere is immature and undermines the conditions it depends on [1]. Discussing Kim Stanley Robinson's novel Aurora, in which colonists find a planet habitable but deadly to Earth life, Frank states his view directly: "the true unit of life. It's not DNA, it's not a cell, it's the biosphere" [1].

This planetary view is also why he rejects Brandon Carter's "hard steps" argument that intelligence must be rare. Carter treated the Sun's timescale and evolution's timescale as independent, but "life and the Earth co-evolve," and "the fate of the Earth and the fate of life are inextricably combined" [1]. The most important point, Frank says, "was that the planet and the biosphere have evolved together," something most biogeophysicists accept and the first thing Carter rejected [1]. Windows for evolution open as planet and biosphere change together, so the timescales are naturally linked [1].

From here Frank draws a working definition useful for telescopes. "One way to define life is it is a system that is far from equilibrium," he says; "as soon as it dies, it turns into, it goes back to equilibrium" [1]. That lets astronomers test a planet without knowing its chemistry in advance. Using the Gibbs free energy, you can measure how far an atmosphere is from equilibrium, and "if you see an atmosphere that is wildly out of equilibrium, that indicates that there's something happening on that planet, biosphere or technosphere, that is pumping gases into the atmosphere, that is keeping the whole system from relaxing" [1]. The classic example is oxygen together with methane, which react away quickly unless life keeps replenishing them [1].

He calls these "agnostic signatures," alongside unusual networks of chemical reactions and excess information in atmospheric compounds [1]. Spotting them is hard: "You really have to use information theory to say, like, what's the background?" [1]. And he goes deeper still. For Frank, the biggest question in physics is what life is, and specifically agency, how an autonomous agent arises [1]. Life's defining feature is that it innovates: physics can predict a star's whole life from its mass and composition, but no one could predict from a single cell that billions of years later there would be a kangaroo that can punch you in the face [1]. He is interested in informational organization, which Kant recognized when he coined the word organism, and in Robert Rosen's contested argument that living systems cannot be fully captured by algorithms [1].

He is also candid about how little we can yet generalize: "We have n equals one example of life. So that means we could be an accident" [1]. One more example would change everything [1].

Sara Walker: life is information structuring matter across time

Walker, an astrobiologist and theoretical physicist, argues we keep failing to define life because we look at individual organisms in a thin slice of the present [2]. Every standard definition, whether self-reproduction, metabolism or compartments, has counterexamples, which is why viruses and fire are so troublesome [2]. She recalls Carl Sagan's essay in which aliens watching Earth might conclude cars are the dominant life form [2].

She takes apart the popular astrobiology definition, life as a self-sustaining chemical system capable of Darwinian evolution, and says she was offended to be quoted agreeing with it, because she thinks every word is wrong [2]:

• Not self-sustaining. Nothing is. Humans depend on societies, so by that logic societies might be more alive than individuals [2].
• Not chemical. Life emerges in chemistry because chemistry is the first combinatorial space the universe can't exhaust. Borrowing Cronin's example, making every three-dimensional arrangement of the molecular formula of taxol would fill about 1.5 universes [2].
• Not individual evolution. Darwinian evolution happens to populations, not individuals [2].

Her alternative: "life is the process of how information structures matter over time and space," with any individual just an instance of lineages structured across time [2]. She even suggests language, memes and mathematics may be alive in this sense, as open-ended combinatorial spaces with room for creativity [2].

The key move is to see life as big in time. We evolved perception that makes space feel important and time fleeting, she says, so we miss most of what we are [2]. "The universe is far larger in time than it is in space," and "the modern technosphere is the largest object in time in the universe that we know about" [2]. Lineages are interwoven and constantly exchange parts, so the individual can't be the unit of life, and life is high-dimensional: something can be alive in some dimensions and not others [2]. She sees life at every scale, from cells to societies to the planet, and says the origin of life can only be solved by integrating all of them [2].

On materialism, she takes a middle path. Our concept of matter is itself tied to particular technologies and eras; just as Aristotle's earth-like and air-like properties gave way to mass and acceleration once we could measure time, we may find life is a material property once we learn to think about information in a material way [2]. And she parts ways with a long-standing hope among physicists, going back to Schrödinger, that non-equilibrium physics and entropy will explain life; she thinks that is the wrong approach, because the real question is how life generates so much order and complexity from randomness [2].

Her account of the origin of life follows. It is a boundary the universe can cross only when a structure emerges that reinforces its own existence against the randomness of chemistry [2]. Once a loop of mutually catalyzing molecules closes, those objects can be made in high abundance, and everything after that point is holding on for dear life to the causal structure that put it there [2]. Cronin's lab has an example: a large molybdenum ring of perhaps around 150 atoms that templates its own production, the first non-organic autocatalytic set [2]. For the first time in her career, she says, she now thinks the origin of life is very abrupt rather than gradual [2]. Her work on chirality supports this: among small molecules with fewer than about seven to 11 heavy atoms almost nothing is chiral, while above that threshold almost everything is, and her lab's work suggests the origin-of-life transition happens around the complexity where chiral molecules become very likely [2].

She also frames evolution by what doesn't exist: "the most significant struggle for existence in the evolutionary process is not among the objects that do exist but between the ones that do and those that never have the chance to" [2]. And she sees death as part of the definition, not its opposite: "death is a way to make room in the universe for other things to exist" [2].

For the search for aliens, this view cuts two ways. It broadens what could count as life, and it warns we might not see it. As a biosphere packs more time into the same space, it may become so virtualized and abstract that it can't perceive other such structures, a "great perceptual filter" [2]. She isn't sure she agrees with the idea, but notes that humans didn't recognize microbes until we invented microscopes [2]. Her grounded work is on life detection in the solar system and biosignature planning for the Habitable Worlds Observatory, and she predicts that "the first contact with alien life will actually be in an origin life experiment" [2].

Lee Cronin: measure the history inside an object

Cronin, a chemist at the University of Glasgow, co-wrote with Walker the Nature paper "assembly theory explains and quantifies selection and evolution," which drew fire from biologists, physicists, complexity theorists and chemists [3]. His core idea: the history is in the objects [3].

Assembly theory uses two measurements [3]:

• Assembly index: the minimum number of steps needed to rebuild an object from its basic parts, where anything already built can be reused for one step. For molecules, the parts are atoms and the steps are bonds [3]. His lab found physical shortcuts: mass spectrometry, infrared spectroscopy and NMR each correlate with the calculated index [3].
• Copy number: how many identical copies exist [3].

Copy number is what turns complexity into a signal of life. "The ultimate randomness and ultimate complexity are indistinguishable until you can see a structure in the randomness," Cronin says [3]. One complex molecule could be a fluke, but a million identical copies of a molecule with a hundred distinct parts could never arise by chance, even across an infinite set of universes [3]. Something must have built them, a cell or a factory; "memory" in the universe is a real physical process that directs construction [3].

He places this in four nested spaces: the assembly universe, where anything goes; the assembly possible, where physical laws like chemical bonding apply; the assembly contingent, where complex objects can't be reached until their precursors exist; and the assembly observed, what we actually see [3]. The key shift is from possible to contingent [3]. That's also why he says, provocatively, "I don't think chemical reactions exist": named reactions are shorthand for constraints like Earth's gravity, temperature, pressure and pure reagents in glass flasks, and a reaction that works on Earth may not be accessible on Venus [3].

Selection, in his view, happens when objects persist in environments that would destroy others, like a carbon chain surviving an acid pond, but it becomes interesting when objects are continually made and destroyed [3]. The environment is the first factory, but it is fragile; biology's leap is robustness, error correction that lets life survive almost anywhere [3]. Cronin thinks selection itself isn't special; what's special is Earth's particular history of environments [3]. As he puts it, "selection is the force in the universe it creates novelty" [3].

Crucially, the theory produces a test. In his lab's validation, samples including peaty Islay whisky, a lowland whisky, vodka, E. coli, beer yeast, taxol and five blind samples from NASA (fossils, Antarctic sediment and a meteorite among them) were measured; biological samples produced molecules with an assembly index above 15, abiotic ones stayed below, and NASA confirmed the living and dead calls were right [3]. On Earth, he says, molecules above about 350 molecular weight or with more than 15 fragments are only produced by life, and skeptics can simply raise the threshold [3]. His lab has even recreated the tree of life with mass spectrometry and no gene sequencing, by fingerprinting samples' most complex molecules [3].

For other worlds, he would bring a high-resolution mass spectrometer, drilling into billions-year-old soil because surfaces are harsh [3]. He and Fridman playfully design a life meter that sniffs volatiles, separates them and measures complexity from dead, through Earth life, up to alien intelligence [3]. The big advantage, he says, is that it looks for complexity without an Earth-chemistry or Earth-biology bias, because "complexity in abundance is evidence of selection" [3].

His view of what life produces is expansive. Life mines novelty from the future: taxol, one of biology's most complex natural products, was not written into the universe's initial conditions but is an accident unique to Earth [3]. Because of the combinatorial scaffolding built since the last universal common ancestor, "the largest place in universe is earth" [3]. Life is probably common around other stars, he thinks, but each planet's history may be so different that our causal cones never overlap: "my biggest fear in a way is that life is everywhere but we become infinitely more lonely" [3]. His answer is a planet simulator to create alternative life in the lab, learn which kinds of worlds would overlap with ours, and point telescopes there, searching signals for copy-number structure [3]. And he rejects Sagan's pale-blue-dot lesson: "I think Carl Sagan did the wrong thing," because the point is how big Earth is for having produced Voyager [3].

Where They Agree, and Where They Don't

All three reject the checklist. Frank says the warm-pond picture misses the planet [1]; Walker says every word of the standard definition is wrong [2]; Cronin replaces properties with a measurable history [3].

All three move the unit of life up and out. Frank's unit is the biosphere [1]. Walker's is the lineage stretched across time, with life at every scale up to the planet and technosphere [2]. Cronin's is the factory: whatever system makes many copies of complex objects [3]. They also overlap personally: Frank quotes Walker's line that life happens to a planet and wrote on planetary intelligence with her [1], and Walker and Cronin built assembly theory together [2][3].

All three emphasize history and contingency. Frank's Earth needed snowball exits, mountains and oxygen [1]; Walker's objects carry billions of years of causal history [2]; Cronin's assembly contingent space means complex things can't exist until their precursors do, and taxol is unique to Earth's history [3]. Frank's point that life innovates and can't be predicted from a cell [1] echoes Cronin's life that mines novelty [3].

They disagree on equilibrium. Frank's practical definition is a system far from equilibrium, and disequilibrium measured with Gibbs free energy is one of his agnostic signatures [1]. Walker says the physicists' hope that non-equilibrium physics and entropy will explain life is the wrong approach, since the question is how life makes order and complexity from randomness [2]. These don't fully conflict: Frank uses disequilibrium as a detection sign, while Walker is rejecting it as an explanation of what life is. But they point searches toward different evidence: atmospheres out of balance [1] versus complex objects in abundance [3].

They differ on where to look. Frank favors telescopes reading whole-planet atmospheres [1]. Cronin favors instruments in the soil measuring molecules, plus lab-built alternative life [3]. Walker spans both: solar system life detection, the Habitable Worlds Observatory, and origin-of-life experiments [2].

They differ on how recognizable alien life will be. Frank expects Darwinian evolution with convergent features like limbs [1]. Walker raises the possibility that advanced life becomes too abstract for us to perceive [2]. Cronin thinks life may be everywhere yet so contingent that we never overlap [3].

On the computational question. Frank cites Rosen's contested argument that living systems can't be fully captured by algorithms [1]. Walker argues computation sits at the top of reality, not its base, and is our best current language rather than what the world is [2]. Cronin says assembly theory is about causation, not compression, and differs from Kolmogorov complexity, which assumes a computer [3]. All three resist reducing life to an algorithm, though for different reasons.

Controversy. Cronin's paper offended evolutionary biologists, physicists, prebiotic chemists and even creationists [3], and he admits the title might have said non-biological selection and evolution [3]. Neither Frank nor Walker is shown in these sources endorsing every part of it, and the thresholds are provisional: Cronin says the 15-step number differs for other spaces and skeptics can raise it [3][2].

Why It Works

Selection leaves a signature randomness can't. A random process can make one complex object, but not many identical ones [3]. Copy number plus assembly index separates the two, and it can be measured with standard instruments [3]. Blind tests with NASA samples got the living and non-living calls right [3].

Life pushes planets out of balance. Living systems keep reactive gases in the air, like oxygen and methane together, that would otherwise react away [1]. A dead planet relaxes toward equilibrium; a living one doesn't [1].

Life and planet co-evolve. Because geology feeds life (nutrients from eroding mountains, carbon dioxide from volcanoes) and life reshapes geology and air (oxygen), you can't understand either in isolation [1]. This is also why Carter's hard-steps logic fails, since it assumed the two were independent [1].

Histories collapse onto self-reinforcing structures. Out of an exponentially large space of possibilities, autocatalytic loops lock in and get made again and again, as in Cronin's molybdenum ring [2]. Because the space grows exponentially with assembly index, objects come into existence along near-shortest paths close to what already exists [2]. That explains both why life's products are complex and why they come in copies [2][3].

Instruments define what we can see. Walker's microbes-before-microscopes example [2], Cronin's point that identical depends on detector resolution (mass spectrometry can't tell left- and right-handed molecules apart) [3], and Frank's information-theoretic problem of defining the background [1] all show that the definition of life is tied to how we measure.

Where It Applies, and Where It Doesn't

These definitions apply most directly to the search for life beyond Earth, where the checklist assumes too much. Frank notes the habitable zone concept is already too strict: Europa, far outside it, has a hundred-mile-deep ocean under about ten miles of ice [1]. A definition based on complexity or disequilibrium doesn't need Earth-like water on a surface [1][3].

They apply to the origin-of-life question, where Walker and Cronin both think lab experiments can watch complexity rise: Cronin's Petri dish of E. coli showing total assembly rising in an S-curve, and his lab trying to see molecular networks do the same from raw chemicals [3], and Walker's prediction that first contact may be in such an experiment [2].

They apply less cleanly at the edges. Frank acknowledges that detecting anomalies against an unknown background is really difficult [1]. Cronin's thresholds are calibrated on Earth chemistry and may need adjusting [3]. Walker's perceptual filter is explicitly speculative [2]. And Walker's broad view, in which language or technology may count as alive in some dimensions [2], stretches the word further than a practical biosignature search requires.

They also don't settle the everyday question of whether a virus or a fire is alive. Walker's answer is that life is high-dimensional, so something can be alive in some dimensions and not others [2], which dissolves the question rather than answering yes or no.

Finally, with only one example of life, every definition is still being tested against a sample of one [1].

What To Do With It

1. Drop the checklist when thinking about alien life. Ask instead about process: is there a planet out of balance [1], or complex objects made in many copies [3]?

2. Think of habitability as planetary history. Plate tectonics, snowball escapes, mountain erosion and oxygen all shaped when complex life could arise on Earth [1]. A single warm pond is not enough information.

3. Look for disequilibrium. Atmospheres "wildly out of equilibrium," like oxygen with methane, are signs of something pumping gases in, whether a biosphere or a technosphere [1].

4. Look for complexity in abundance. On Earth, molecules above an assembly index of about 15, or above about 350 molecular weight, found in quantity, are only made by life [3]. Take the thresholds as provisional and adjustable [3][2].

5. Measure below the surface. Cronin would drill into ancient soil rather than sample harsh surfaces [3].

6. Watch origin-of-life labs as closely as telescopes. Both Walker and Cronin think life's origin and alien life are linked, and that experiments producing self-reinforcing chemistry may be where "first contact" happens [2][3].

7. Think of yourself as part of something big in time. Walker's lineage view and Cronin's "largest place in universe is earth" both treat living things as the current tip of billions of years of history [2][3]; Frank's view makes the biosphere, not the individual, the unit [1].

8. Stay alert to what our instruments can't see. The microbes-before-microscopes lesson [2] and the problem of defining the background [1] mean a negative result may say more about our tools than about the planet.

Sources

1. Alien Civilizations and the Search for Extraterrestrial Life · Lex Fridman Podcast
2. Sara Walker: Physics of Life, Time, Complexity, and Aliens · Lex Fridman Podcast
3. Lee Cronin: Controversial Nature Paper on Evolution of Life and Universe · Lex Fridman Podcast