There is no single, universally accepted scientific definition of life. Biologists, chemists, physicists, and philosophers have proposed hundreds of definitions over the past century, and none has won consensus. The most widely cited version, often called the NASA definition, describes life as “a self-sustained chemical system capable of Darwinian evolution,” but even its proponents acknowledge it is imprecise and difficult to apply in practice. The difficulty is not just academic hair-splitting. How you define life determines whether viruses count, whether you could recognize alien biology, and whether a synthetic cell built in a lab qualifies as alive.
The NASA Definition and Why It Dominates
In the early 1990s, a working group convened by NASA settled on a compact phrase: life is “a self-sustained chemical system capable of Darwinian evolution.” The definition was shaped heavily by the needs of astrobiology, where scientists needed some operational yardstick for designing instruments that could detect life on other planets. It has since become the single most popular definition in the scientific literature, especially among researchers studying the origin of life and the possibility of extraterrestrial biology.1PubMed. The Origin, Extension, and Future of the “NASA Definition” of Life
The definition packs a lot into a few words. “Self-sustained” is meant to capture metabolism, the idea that a living system harvests energy and raw materials from its surroundings to maintain itself. “Chemical system” anchors life in chemistry rather than, say, software simulations or purely mathematical abstractions. And “capable of Darwinian evolution” points to heredity, variation, and natural selection as the engine that distinguishes living chemistry from mere chemical reactions running in a flask.
The phrase has real strengths. It avoids listing specific molecules like DNA or proteins, which leaves room for life that might use a different chemical toolkit. It also avoids requiring a cell membrane or a particular energy source, which keeps the door open for biochemistries very different from our own. But critics note that it is ambiguous in important ways. What exactly does “self-sustained” mean? A single rabbit is not self-sustained without food, water, and oxygen from outside. And “capable of Darwinian evolution” is tricky: a single organism does not evolve, populations do. A lone bacterium on a petri dish satisfies most intuitions about being alive, yet it is not, by itself, undergoing Darwinian evolution.2PubMed Central. Defining life
The Textbook Checklist Approach
If you took introductory biology, you probably encountered a list of properties that all living things supposedly share. These lists typically include organization (cells), metabolism, homeostasis, growth, reproduction, response to stimuli, and adaptation through evolution. The advantage of this approach is that it is intuitive: you can look at a dog and check off every item. The disadvantage is that almost every item on the list has exceptions or counterexamples that blur the boundary.
Mules are alive but cannot reproduce. Fire consumes fuel, grows, and responds to its environment but is not alive. Crystals grow and maintain an organized structure. Salt solutions can reach a kind of homeostasis. The checklist works well for familiar Earth organisms, but it was never designed to handle entities at the margins, and it offers no principled way to decide which properties are essential and which are just common features of life as we happen to know it.
Autopoiesis and Self-Production
In the 1970s, biologists Humberto Maturana and Francisco Varela introduced the concept of autopoiesis, a Greek-derived term meaning roughly “self-making.” Their idea was that the defining feature of a living system is not any single property but a particular kind of organizational closure: the system produces the very components and boundaries that keep it going. A cell, for instance, manufactures its own membrane, its own enzymes, and its own energy-carrying molecules, and those components in turn sustain the network of reactions that produced them in the first place.3PubMed. Formalizing autopoiesis: Toward a Categorical-Thermodynamic Calculus of Closure
This recursive loop is what autopoiesis tries to capture. A living system is not just a bag of chemicals; it is a network that regenerates itself. Recent experimental work has tried to build laboratory systems that exhibit this kind of closure, testing whether simple chemical networks can sustain themselves in a way that qualifies as autopoietic.4PubMed. Experimental probes of autopoietic self-maintenance The autopoietic framework appeals to researchers who think the NASA definition overemphasizes evolution at the expense of what makes an individual organism alive right now, in this moment, regardless of whether it is part of an evolving population.
The Thermodynamic Perspective
In 1944, physicist Erwin Schrödinger published a slim book called What Is Life? that shaped decades of thinking. His core insight was that living organisms maintain their internal order by feeding on what he called “negative entropy,” essentially importing organized energy from their surroundings and exporting disorder as heat and waste. This lets a living system stay far from thermodynamic equilibrium, the state of maximum disorder that non-living matter tends toward.
Schrödinger’s framing has held up remarkably well. Careful thermodynamic analysis of microbial growth confirms that cells do rid themselves of internally generated entropy through both heat production and the chemical transformation of substrates into higher-entropy products.5PubMed. Does microbial life always feed on negative entropy? Thermodynamic analysis of microbial growth Living organisms require continuous energy input to exist. All known life is built around metabolic processes that extract energy from the environment and channel it into maintaining the organism’s internal order.6PubMed Central. Metabolic Homeostasis in Life as We Know It: Its Origin and Thermodynamic Basis
This thermodynamic view does not, by itself, distinguish life from other dissipative systems like hurricanes or convection cells, which also maintain organized structures by processing energy flows. But it does capture something essential: life is a process, not a substance. Cut off the energy supply and the organization collapses. A dead organism has the same atoms as a living one; what it lacks is the sustained, far-from-equilibrium chemistry.
Information as the Thread
Some researchers have tried to define life through the lens of information theory. The idea is that what makes living matter special is not its chemistry per se but the way it stores, copies, and acts on information. DNA is, at bottom, an information-storage molecule. The genetic code is a mapping between sequences and functions. Evolution is, in one framing, an algorithm that edits and refines information over time.
One proposal frames life as a “continuum of self-maintainable information,” treating the living entity as a distinct element within that continuum.7PubMed. Shannon’s information, Bernal’s biopoiesis and Bernoulli distribution as pillars for building a definition of life This approach has the advantage of being agnostic about specific chemistry. In principle, any system that stores heritable information and uses it to maintain and reproduce itself could qualify as alive, whether it runs on DNA, some alien polymer, or something else entirely. The challenge is making these ideas precise enough to be testable. Lots of things store information, from hard drives to snowflake patterns, without being alive.
Where Definitions Break Down
The real stress test for any definition of life is the boundary cases, entities that satisfy some criteria for life but not others. Viruses are the most famous example. A virus has a genome, evolves by natural selection, and can be exquisitely adapted to its host. But outside a host cell, a virus particle is inert. It has no metabolism, no internal energy production, and no ability to replicate on its own. The majority of virologists consider viruses to be subcellular genetic parasites that do not self-replicate but are replicated passively by the cellular machinery of their hosts.8PubMed Central. Are viruses alive? The replicator paradigm sheds decisive light on an old but misguided question.
Whether viruses are alive turns entirely on which definition you adopt. Under the NASA definition, they are arguably capable of Darwinian evolution but are not self-sustained. Under autopoiesis, they clearly fail, since they do not produce their own components. Under a checklist approach, they hit some marks and miss others. One useful reframing places all biological replicators on a continuum from completely selfish (lytic viruses that destroy their host) to fully cooperative (genes that benefit the organism), which sidesteps the alive-or-not binary altogether.8PubMed Central. Are viruses alive? The replicator paradigm sheds decisive light on an old but misguided question.
Viroids push the question even further. These are tiny circular RNA molecules, much simpler than viruses, with no protein coat and no genes of their own. Yet viroids can cleave and join RNA, replicate inside host cells, and undergo Darwinian evolution.9PubMed Central. Viroids and the Origin of Life If an entity that small and simple can evolve, where exactly do you draw the line?
What Minimal Cells Reveal
Synthetic biology has taken a different approach to the question by trying to build life from scratch, or at least strip it down to its barest essentials. In 2016, researchers synthesized a near-minimal bacterial cell by keeping only the essential and quasi-essential genes of Mycoplasma mycoides. The resulting organism, called JCVI-Syn3.0, could replicate DNA, transcribe RNA, translate proteins, and divide, and not much else.10PubMed Central. Minimal Cells-Real and Imagined
What surprised researchers was what the stripped-down genome still kept. Even after removing every gene that was not strictly necessary, the minimal cell retained systems for repairing and preventing metabolite damage, the chemical wear and tear that accumulates as a side effect of metabolism. This suggests that damage control is not a luxury but a fundamental requirement for any living system.11PubMed Central. Metabolite Damage and Damage Control in a Minimal Genome Life, even at its simplest, has to constantly maintain itself against its own chemistry going wrong.
Minimal cells also clarify the role of compartmentalization. Life-like systems need to import building blocks, export waste, recycle internal molecules, and maintain stable internal conditions, and all of that requires a boundary that separates inside from outside.12PubMed Central. Minimal Out-of-Equilibrium Metabolism for Synthetic Cells: A Membrane Perspective Research on the origin of life suggests that the earliest protocells formed when simple fat-like molecules spontaneously assembled into membrane structures, concentrating and organizing biologically relevant chemicals in their confined spaces.13PubMed. Organization and Compartmentalization by Lipid Membranes Promote Reactions Related to the Origin of Cellular Life A primitive protocell consisting of a fatty-acid compartment and an information-carrying polymer would already have the potential for Darwinian evolution.14PubMed Central. The origins of cellular life
Why It Matters for Finding Alien Life
The definition of life is not just a philosophical puzzle. It has practical consequences for how space agencies design missions to search for biology on other worlds. If your instruments are calibrated to detect DNA, amino acids, or other molecules specific to Earth biochemistry, you might miss life that uses a completely different chemical toolkit. The concept of contingency suggests that non-biomolecules, chemicals that play no role in Earth life, could have been essential at the origin of life elsewhere, meaning extraterrestrial organisms on the moons of Jupiter or Saturn might look nothing like anything in our biology textbooks.15PubMed Central. Alternative Pathways in Astrobiology: Reviewing and Synthesizing Contingency and Non-Biomolecular Origins of Terrestrial and Extraterrestrial Life
This has led to growing interest in what researchers call agnostic biosignatures, indicators of biology that do not depend on the specific molecules or metabolic pathways used by Earth life. Instead of looking for DNA or oxygen, these approaches look for general patterns: unusual chemical complexity, molecular structures that are far from thermodynamic equilibrium, or distributions of compounds that are hard to explain without invoking biology.16Annual Review of Earth and Planetary Sciences. Agnostic Biosignatures: Expanding the Search for Life in the Solar System The hope is that even life built on entirely foreign chemistry would leave detectable fingerprints in its environment, as long as the search is not too narrowly tuned to Earth-specific signatures.
Meanwhile, xenobiology, the effort to build life using artificial genetic polymers, is actively testing the boundaries of what counts as biological. Researchers have designed molecules called xeno nucleic acids (XNAs) that store genetic information like DNA but have different chemical structures, making them essentially invisible to natural biological systems.17PubMed Central. Xenobiology: a new form of life as the ultimate biosafety tool If a self-replicating system running on XNA were ever achieved, it would satisfy most definitions of life while sharing almost nothing chemically with existing biology. That would be a powerful argument that life is defined by its organization, not its ingredients.
The Continuum Problem
One of the deeper reasons no single definition has won is that the boundary between living and nonliving may not be a sharp line at all. Some researchers have argued that the origin of life is better understood as a seamlessly continuous process rather than a discrete event at a specific point in time.18Perspectives on Science and Christian Faith. Rethinking Abiogenesis: Part II, Life as a Simplification of the Nonliving Universe Under this view, there was never a moment when non-life suddenly became life. Instead, chemical systems gradually acquired more life-like properties, metabolism, information storage, compartmentalization, heritability, until at some fuzzy point they crossed a threshold that we retrospectively label “alive.”
This is frustrating if you want a clean yes-or-no answer, but it reflects what the chemistry actually looks like. Simple self-assembling vesicles are not alive. A modern bacterium is. But between those two endpoints lies a long gradient of increasingly complex, increasingly life-like chemical systems, and no obvious place to draw a line. The question “what is life?” may ultimately be less like “what is water?” (which has a precise chemical answer) and more like “what is a mountain?” (where the answer depends on how tall you insist a hill has to be before it counts).
Definitions as Tools, Not Truths
Given the lack of consensus, many working scientists have started treating definitions of life less as eternal truths and more as practical tools that serve different purposes. The NASA definition is useful for astrobiology mission planning. Autopoiesis is useful for thinking about what makes an individual organism distinct from its environment. The thermodynamic perspective is useful for understanding why life requires energy. The checklist approach is useful for teaching students. None is universally correct, and none is useless.
The honest state of affairs is that biology has no equivalent of water’s molecular formula. There is no Hâ‚‚O of life, no single criterion you can point to and say “that, and only that, is what makes something alive.” What exists instead is a cluster of deeply related properties, self-maintenance, information storage, evolution, compartmentalization, metabolism, that tend to travel together in the systems we recognize as living. Entities at the boundary, such as viruses or self-replicating RNA molecules, satisfy some of these properties and not others, and reasonable scientists disagree about where they fall. The search for a universal definition continues, driven in large part by the practical need to know what to look for on other worlds, but there is a growing recognition that the question itself may not have a single clean answer.