People often feel that they can intuitively recognize whether something is alive, but nature is filled with entities that flout easy categorization as life or non-life — and the challenge may intensify as other planets and moons open up to exploration. In this excerpt from his new book, Life’s Edge: The Search for What It Means to Be Alive, published today, the science writer Carl Zimmer discusses scientists’ frustrated efforts to develop a universal definition of life.
“It is commonly said,” the scientists Frances Westall and André Brack wrote in 2018, “that there are as many definitions of life as there are people trying to define it.”
As an observer of science and of scientists, I find this behavior strange. It is as if astronomers kept coming up with new ways to define stars. I once asked Radu Popa, a microbiologist who started collecting definitions of life in the early 2000s, what he thought of this state of affairs.
“This is intolerable for any science,” he replied. “You can take a science in which there are two or three definitions for one thing. But a science in which the most important object has no definition? That’s absolutely unacceptable. How are we going to discuss it if you believe that the definition of life has something to do with DNA, and I think it has something to do with dynamic systems? We cannot make artificial life because we cannot agree on what life is. We cannot find life on Mars because we cannot agree what life represents.”
With scientists adrift in an ocean of definitions, philosophers rowed out to offer lifelines.
Some tried to soothe the debate, assuring the scientists they could learn to live with the abundance. We have no need to zero in on the One True Definition of Life, they argued, because working definitions are good enough. NASA can come up with whatever definition helps them build the best machine for searching for life on other planets and moons. Physicians can use a different one to map the blurry boundary that sets life apart from death. “Their value does not depend on consensus, but rather on their impact on research,” the philosophers Leonardo Bich and Sara Green argued.
Other philosophers found this way of thinking — known as operationalism — an intellectual cop‐out. Defining life was hard, yes, but that was no excuse not to try. “Operationalism may sometimes be unavoidable in practice,” the philosopher Kelly Smith countered, “but it simply cannot substitute for a proper definition of life.”
Smith and other foes of operationalism complain that such definitions rely on what a group of people generally agree on. But the most important research on life is at its frontier, where it will be hardest to come to an easy agreement. “Any experiment conducted without a clear idea of what it is looking for ultimately settles nothing,” Smith declared.
Smith argued that the best thing to do is to keep searching for a definition of life that everyone can get behind, one that succeeds where others have failed. But Edward Trifonov, a Russian‐born geneticist, wondered if a successful definition already exists but is lying hidden amidst all the past attempts.
In 2011, Trifonov reviewed 123 definitions of life. Each was different, but the same words showed up again and again in many of them. Trifonov analyzed the linguistic structure of the definitions and sorted them into categories. Beneath their variations, Trifonov found an underlying core. He concluded that all the definitions agreed on one thing: life is self‐reproduction with variations. What NASA’s scientists had done in eleven words (“Life is a self‐sustained chemical system capable of undergoing Darwinian evolution”), Trifonov now did with three.
His efforts did not settle matters. All of us — scientists included — keep a personal list of things that we consider to be alive and not alive. If someone puts forward a definition, we check our list to see where it draws that line. A number of scientists looked at Trifonov’s distilled definition and did not like the line’s location. “A computer virus performs self‐reproduction with variations. It is not alive,” declared the biochemist Uwe Meierhenrich.
Some philosophers have suggested that we need to think more carefully about how we give a word like life its meaning. Instead of building definitions first, we should start by thinking about the things we’re trying to define. We can let them speak for themselves.
These philosophers are following in the tradition of Ludwig Wittgenstein. In the 1940s, Wittgenstein argued that everyday conversations are rife with concepts that are very hard to define. How, for example, would you answer the question, “What are games?”
If you tried to answer with a list of necessary and sufficient requirements for a game, you’d fail. Some games have winners and losers, but others are open‐ended. Some games use tokens, others cards, others bowling balls. In some games, players get paid to play. In other games, they pay to play, even going into debt in some cases.
For all this confusion, however, we never get tripped up talking about games. Toy stores are full of games for sale, and yet you never see children staring at them in bafflement. Games are not a mystery, Wittgenstein argued, because they share a kind of family resemblance. “If you look at them you will not see something that is common to all,” he said, “but similarities, relationships, and a whole series of them at that.”
A group of philosophers and scientists at Lund University in Sweden wondered if the question “What is life?” might better be answered the way Wittgenstein answered the question “What are games?” Rather than come up with a rigid list of required traits, they might be able to find family resemblances that could naturally join things together in a category we could call Life.
In 2019 they set out to find it by carrying out a survey of scientists and other scholars. They put together a list of things including people, chickens, Amazon mollies, bacteria, viruses, snowflakes, and the like. Next to each entry the Lund team provided a set of terms commonly used to talk about living things, such as order, DNA, and metabolism.
The participants in the study checked off all the terms that they believed to apply to each thing. Snowflakes have order, for example, but they don’t have a metabolism. A human red blood cell has a metabolism but it contains no DNA.
The Lund researchers used a statistical technique called cluster analysis to look at the results and group the things together based on family resemblances. We humans fell into a group with chickens, mice, and frogs — in other words, animals with brains. Amazon mollies have brains, too, but the cluster analysis put them in a separate group close to our own. Because they don’t reproduce by themselves, they’re set a little apart from us. Further away, the scientists found a cluster made up of brainless things, such as plants and free‐living bacteria. In a third group was a cluster of red blood cells and other cell‐like things that can’t live on their own.
Furthest away from us were things that are commonly not considered alive. One cluster included viruses and prions, which are deformed proteins that can force other proteins to take their shape. Another included snowflakes, clay crystals, and other things that don’t replicate in a lifelike way.
The Lund researchers found that they could sort things pretty well into the living and the nonliving without getting tied up in an argument over the perfect definition of life. They propose that we can call something alive if it has a number of properties that are associated with being alive. It doesn’t have to have all those properties, nor does it even need exactly the same set found in any other living thing. Family resemblances are enough.
One philosopher has taken a far more radical stand. Carol Cleland argues that there’s no point in searching for a definition of life or even just a convenient stand‐in for one. It’s actually bad for science, she maintains, because it keeps us from reaching a deeper understanding about what it means to be alive. Cleland’s contempt for definitions is so profound that some of her fellow philosophers have taken issue with her. Kelly Smith has called Cleland’s ideas “dangerous.”
Cleland had a slow evolution into a firebrand. When she enrolled in the University of California, Santa Barbara, she started off studying physics. “I was a klutz in the lab, and my experiments never turned out right,” she later told an interviewer. From physics she turned to geology, and while she liked the wild places that the research took her to, she didn’t like feeling isolated as a woman in the male‐dominated field. She discovered philosophy in her junior year and was soon grappling with deep questions about logic. After graduating college and spending a year working as a software engineer, she went to Brown University to earn a Ph.D. in philosophy.
In graduate school Cleland mulled space and time, cause and effect.
When Cleland finished grad school, she moved on to subjects that were easier to talk about at dinner parties. She worked at Stanford University for a time, contemplating the logic of computer programs. She then became an assistant professor at the University of Colorado, where she remained for the rest of her career.
In Boulder, Cleland turned her attention to the nature of science itself. She examined how some scientists, like physicists, could run experiments over and over again, while others, like geologists, couldn’t replay millions of years of history. It was while she was reflecting about these differences that she learned about a Martian rock in Antarctica that was posing a philosophical conundrum of its own.
[The Martian rock, a meteorite designated Allan Hills 84001, was examined in 1996 by a NASA team led by David McKay. They reported seeing signs of ancient life in it, including microbial fossils, but most scientists dismissed the evidence as too ambiguous to be credible.]
A lot of the arguments over Allan Hills 84001 had less to do with the rock itself than with the right way to do science. Some researchers thought the NASA team had done an admirable job of studying it, but others thought it was ridiculous to conclude from their findings that the meteorite might contain fossils. The planetary scientist Bruce Jakosky, one of Cleland’s colleagues at the University of Colorado, decided to organize a public discussion where the two sides could air their views. But he realized that judging Allan Hills 84001 required more than running some experiments to measure magnetic minerals. It demanded thinking through how we make scientific judgments. He asked Cleland to join the event, to talk about Allan Hills 84001 as a philosopher.
What started as a quick prep for a talk turned into a dive into the philosophy of extraterrestrial life. Cleland concluded that the fight over Allan Hills 84001 sprang from the divide between experimental and historical sciences. The critics made the mistake of treating the meteorite study as experimental science. It was absurd to expect McKay’s team to replay history. They couldn’t fossilize microbes on Mars for 4 billion years and see if they matched Allan Hills 84001. They couldn’t hurl a thousand asteroids at a thousand copies of Mars and see what came our way.
Cleland concluded that the NASA team had carried out good historical science, comparing explanations for the ones that explained their evidence best. “The martian‐life hypothesis is a very good candidate for being the best explanation of the structural and chemical features of the martian meteorite,” she wrote in 1997 in the Planetary Report.
Cleland’s work on the meteorite impressed Jakosky so much that he invited her in 1998 to join one of the teams at NASA’s newly created Astrobiology Institute. In the years that followed, Cleland developed a philosophical argument for what the science of astrobiology should look like. She informed her ideas by spending time with scientists doing different kinds of research that fit under the umbrella of astrobiology. She traveled around the Australian outback with a paleontologist searching for clues to how giant mammals went extinct 40,000 years ago. She went to Spain to learn how geneticists sequence DNA. And she spent a lot of time at scientific meetings, roaming from talk to talk. “I felt like a kid in a candy store,” she once told me.
But sometimes the scientists Cleland spent time with set off her philosophical alarms. “Everybody was working with a definition of life,” she recalled. NASA’s definition, only a few years old at that point, was especially popular.
As a philosopher, Cleland recognized that the scientists were making a mistake. Their error didn’t have to do with determinate attributes or some other fine philosophical point understood only by a few logicians. It was a fundamental blunder that got in the way of the science itself. Cleland laid out the nature of this mistake in a paper, and in 2001 she traveled to Washington, D.C., to deliver it at a meeting of the American Association for the Advancement of Science. She stood up before an audience made up mostly of scientists, and told them it was pointless to try to find a definition of life.
“There was an explosion,” Cleland recalled. “Everyone was yelling at me. It was really amazing. Everyone had their pet definitions and wanted to air them. And here I told them the whole definition project was worthless.”
Fortunately, some people who heard Cleland talk thought she was onto something. She began collaborating with astrobiologists to explore the implications of her ideas. Over the course of two decades she published a series of papers, culminating in a book, The Quest for a Universal Theory of Life.
The trouble that scientists had with defining life had nothing to do with the particulars of life’s hallmarks such as homeostasis or evolution. It had to do with the nature of definitions themselves — something that scientists rarely stopped to consider. “Definitions,” Cleland wrote, “are not the proper tools for answering the scientific question ‘what is life?’”
Definitions serve to organize our concepts. The definition of, say, a bachelor is straightforward: an unmarried man. If you’re a man and you’re unmarried, you are — by definition — a bachelor. Being a man is not enough to make you a bachelor, nor is being unmarried. As for what it means to be a man, well, that can get complicated. And marriage has its own complexity. But we can define “bachelor” without getting bogged down in those messy matters. The word simply links these concepts in a precise way. And because definitions have such a narrow job to do, we can’t revise them through scientific investigation. There is simply no way that we could ever discover that we were wrong about the definition of a bachelor as being an unmarried man.
Life is different. It is not the sort of thing that can be defined simply by linking together concepts. As a result, it’s futile to search for a laundry list of features that will turn out to be the real definition of life. “We don’t want to know what the word life means to us,” Cleland said. “We want to know what life is.” And if we want to satisfy our desire, Cleland argues, we need to give up our search for a definition.
From the book Life’s Edge: The Search for What It Means to Be Alive by Carl Zimmer, published by Dutton, an imprint of Penguin Publishing Group, a division of Penguin Random House LLC. Copyright © 2021 by Carl Zimmer.