The Elusive Definition Of ‘Life’

Ever since we humans developed language, certain words have had exact and universal definitions. But other words… not so much. Perhaps strangely, some of the most important words/concepts have been the most difficult to define.

Dead As Dirt
This patch of dried mud looks entirely dead. However, a microscope would reveal that it’s teeming with life. And a rainstorm would quickly reveal that other life had only been dormant within it.
(Copyright © 2011 by Wil C. Fry.)

For example, when early humans came up with the first word for “sun”, nobody ever questioned its meaning: “that big hot thing in the sky; the biggest, brightest one”. In every language, through all of history, the meaning of sun only really changed once — when we discovered that “star” and “sun” were actually synonyms instead of two separate things; it’s just that one of the stars is relatively close to us while the others are very far away.

But at the same time, words like “life” and “alive” have elusive meanings that we have yet to precisely define — though many have tried. Physicist Fred Adams (in Origins Of Existence) noted: “Achieving a universal definition of life is unquestionably of fundamental importance, but no such definition has yet been forthcoming.” Adams and others have described characteristics of life, including reproduction, metabolism, and homeostasis, but there is still not a universally agreed-upon definition of life — or what makes something alive and something else not alive.

“Whenever biologists try to formulate definitions of life, they are troubled by the following: a virus; a growing crystal; Penrose’s tiles; a mule; a dead body of something that was indisputably alive; an extraterrestrial creature whose biochemistry is not based on carbon; an intelligent computer or robot.”

— William Poundstone, The Recursive Universe

Merriam-Webster takes the cop-out route for alive: “having life; not dead” — and defines life as: “the quality that distinguishes a vital and functional being from a dead body”, which doesn’t really say anything. It doesn’t say what that distinguishing “quality” is.

Legal definitions of life are, of course, something separate from scientific definitions, but have great import over our lives. One I found is: “a state in which energy of function is ever resisting decay and dissolution” (a 1924 federal district court case). There are many others; each of them seems to have the same problem as most scientific definitions: you can always think of exceptions or problems with each definition. For example, the one I quoted uses the phrase “ever resisting decay”, and we know that’s not true of life. Every living thing that we know of eventually decays and dies — and not always in that order.

“I know it when I see it.”

Why is this so difficult? Remember when a U.S. Supreme Court Justice said “I know it when I see it”? The court and everyone involved had been struggling to define obscenity and failing. The same seems to be true of life: “I know it when I see it.” I can point to something that is alive (grass, for example) and tell my child: “That is alive”, and point to something else that is not alive (a sidewalk) and say “that is not alive”. Eventually, my child will learn from these examples how to categorize things into one category or the other. But, like me, my child will still not have a working definition that includes every instance of life.

At least one scientist, James Lovelock has posited that the Earth itself is a single living organism. The rest of us readily accept that our entire living bodies — and the bodies of all living things are made up entirely of materials that were not alive at some point, and that eventually all of those materials will be considered not alive once more. From dust to dust.

We search for life on other planets. We find new life forms in the depths of the sea. We have huge social arguments over when a person should or should not be removed from life support technology, but we still can’t define life itself.

Living Fire
Many definitions of “life” end up including flames inadvertently, as we’ll see below. This “fire tornado” certainly seemed alive at the time. It was about 20 feet tall and twisting violently when I saw it leap over a four-lane highway.
(Copyright © 2005 by Wil C. Fry. Some rights reserved.)

• The Ability To Reproduce

This is seemingly the most common thread in all definitions of life. However, there are problems with it. First, I cannot reproduce. Neither can you. Are we not alive? Many of my individual cells can reproduce, and billions of them do so on a daily basis, but I — the adult male human known as “Wil C. Fry” — cannot reproduce myself. At best, I can supply a gamete — half the genetic code necessary to form a new human. Without another gamete from someone else, I’m stuck. Do I only qualify as “alive” if I’m paired with another gamete-producing human? If you dropped me off alone on Mars, with no way back, you would still consider me alive, despite my utter lack of ability to reproduce. Women on the other hand are much closer to being able to reproduce without help, but still only produce gametes. Any fertile woman set alone on another planet is unable to reproduce but still alive.

Do I only qualify as “alive” if I’m paired with another gamete-producing human?

(Weirdly, I haven’t found a single scientific criticism of this “requirement” for life that includes my complaint above. It’s universally overlooked that most sexually-reproducing species require at least two members of the species in order to reproduce. Yet this is why I would dismiss this requirement immediately and out of hand as a way of determining whether a single organism is alive.)

What about people who are infertile? They cannot reproduce, even with the help of another person.

A mule cannot reproduce, even with the help of a mate, yet is still considered alive.

Second, it isn’t difficult to imagine that someday we will create a machine smart enough to make copies of itself. Robots already perform many complex tasks — in auto factories, for example. What if, instead of being programmed to attach bumpers to automobiles, these robots were programmed to assemble the ingredients for more identical robots?

Third: Is Henrietta Lacks still alive? Cells that were scraped from her in 1951 are still alive today, though Lacks herself died a few months later. Those cells — called “HeLa” — continue to reproduce.

• Metabolism

Another very common characteristic listed for life is the “capability of metabolism”. Interestingly, when I looked up the definition of metabolism, what I found was: “the sum of the physical and chemical processes in a living organism”. So we’re going in circles now. Life is at least partially defined as something having the ability to metabolize, but metabolism is defined as “stuff living things do”. If you remove the “living organism” part of the definition, then what you have left is a list of chemical reactions that can conceivably occur outside living organisms. Back to my hypothetical robot above, it is easy to imagine a machine being built that can perform these same chemical reactions.

This characteristic too, then, has problems.

• Growth

“The capacity for growth” is also often listed as a characteristic of life. Of course, as we all know, some living things don’t grow. Again using myself as an example; I* haven’t grown in a decade; my height and weight are today approximately the same as they were 10 years ago. It is also true that our hypothetical robot can be programmed to add parts to itself — growth.

(* It is true that my hair and fingernails/toenails continue to grow, and are regularly trimmed. I mention this only in case someone smart aleck thinks of it and believes they’ve derailed the argument.)

• Reaction To Stimuli

This characteristic has the same problems as the others: (1) we each can think of living things that don’t react to stimuli — at least not obviously or immediately — and (2) we all know that non-living machines can react to stimuli. We don’t even need the hypothetical robot; I’m currently using a machine that reacts to stimuli.

• Organization

I found a source that listed organization as the primary characteristic of life, in the context that living cells are highly organized, and that all life is composed of living cells. Interestingly, quite a few scientists now consider viruses to be living organisms, and they are not composed of cells. They are, however, still somewhat well organized.

Of course, many non-living things are well organized too. Consider again our hypothetical robot. So far, it meets all the qualifications of life, whereas certain living things do not.

(Many sources insist that life must be composed of cells, and therefore exclude viruses and other non-cellular organisms. My robot can still fit this definition, because it is hypothetical, and could be built in a cellular fashion if so desired.)

• Homeostasis

The current definition of life on Wikipedia includes homeostasis, which (like metabolism) is usually defined as something living things do. Homeostasis is “the property of a system within the body of an organism in which a variable… is actively regulated to remain very nearly constant.” The definition notes that mechanical systems do this too (for example, thermostats), but explains that homeostasis “is an almost exclusively biological term”.

So I will register the same complaint that I did with metabolism: you can’t define life as “having homeostasis” and then define homeostasis as “something living things do”. Our hypothetical robot can easily be designed to actively regulate variables to remain very nearly constant. Further, homeostats within living organisms are known to break down (examples are listed on the Wikipedia page), yet the organisms are still considered “alive”.

• Organic

Some have insisted that to be truly alive, an organism must be organic — that is, composed of carbon-based compounds. This one would finally leave out our hypothetical robot, unless we managed to build it using these same compounds (but we probably wouldn’t). However, as with the word life, the word “organic compound” is one for which “a consensus definition remains elusive and likely arbitrary”. And it is often defined as “the carbon compounds necessary for life” — another circular definition.

The problem some scientists have with this definition is the possibility that we’ll someday encounter life that isn’t carbon-based.

• Resistance To Entropy

Physicist Erwin Schrödinger (yes, that Schrödinger) used an entirely different definition in his 1944 book What Is Life?: “When is a piece of matter said to be alive? When it goes on ‘doing something’, moving, exchanging material with its environment, and so forth, and that for a much longer period than we would expect an inanimate piece of matter to ‘keep going’ under similar circumstances.” He said that something was “dead” when: “A permanent state is reached, in which no observable events occur. The physicist calls this the state of thermodynamical equilibrium, or of ‘maximum entropy’.”

Critics said his definition failed to account for a very simple example of non-life: flames. (Of course, some non-modern societies did consider flame to be a kind of living thing.)

Does this mean tardigrades die and come back to life? Are tardigrades like the mythical Christ? Some would argue yes.

It also fails to account for tardigrades — my new favorite super-animal. Tardigrades, when exposed to extreme conditions like outer space, extreme cold or heat, or absolute absence of water, cease to be “doing something” and completely desiccate themselves. When conditions are right, they reanimate. Does this mean they “die” and come back to life? Are tardigrades like the mythical Christ? Some would argue that yes, tardigrades must be considered “dead” during such a time, and then considered “alive” when they get the conditions they need.

What about seeds? It is an interesting proposition. As we know from research, certain seeds can be stored for many thousands of years, doing nothing, and then grow a plant under the proper conditions (32,000 years later, in at least one case). Are seeds alive or not alive? By Schrödinger’s definition, they are not alive while they are “doing nothing”, and only produce a life when they start to sprout a plant.

Of course, seeds and dehydrated tardigrades aren’t “not alive” the same way a sidewalk or spoon is “not alive”. Because a spoon or sidewalk won’t wake up someday under the right conditions and suddenly begin to grow or reproduce, while we know that tardigrades and seeds will, or can.

(Going back to the very idea of “resisting entropy”, it is well known that some living things cause increased entropy around them, in order to survive. So, while an individual living organism might be considered as staving off entropy for some specified time period, the overall entropy in the environment continues unabated.)

And still, I think, we are capable of producing machines that “resist entropy”. I’m not sure that we’d want to, but I think it’s within the realm of possibility, given our current technological knowledge. Already our automobiles contain sensors throughout their various systems, notifying us (via our mechanics) when something goes wrong. Not long ago, I lacerated my shin while edging my lawn. I did not buy a replacement part; my shin’s skin simply repaired itself over time. Imagine a future automobile that notified itself of a failed tail light and acquired — or manufactured — a replacement. Low oil? A process within the car itself creates more oil. And so on.

If we did create such a machine, would we consider it “alive”? We would have to, by this definition.

• Autopoiesis

This is very similar to the metabolism requirement, and also related to the “resisting entropy” argument. Autopoiesis “refers to a system capable of reproducing and maintaining itself”. The “reproducing” here doesn’t refer to reproduction of another living organism, but replacing all the pieces and parts within an organism. As an example, authors Lynn Margulis and Dorion Sagan (wife and son, respectively, of the famous Carl Sagan) wrote in their 1995 book “What Is Life?” (they were aware of Schrödinger’s title): “Every five days you get a new stomach lining. You get a new liver every two months. Your skin replaces itself every six weeks. Every year, ninety-eight percent of the atoms of your body are replaced. This nonstop chemical replacement, metabolism, is a sure sign of life.”

In other words, “you” aren’t a collection of specific atoms or cells; those specific atoms and cells keep leaving your body at a regular rate, and your body continues to replace them with new ones, via ingesting food, water, and air. “You” are the organism that conducts this process, and you’re only alive while you’re doing it.

This definition works for some of the problematic examples like mules — and even for all sexually reproducing organisms, who can’t manufacture offspring alone — and excludes viruses. But it too fails with flames.

Clearly, autopoiesis alone isn’t enough to define life.

• A Useful Definition

Writer Joseph Morales attempted here to come up with a useful definition of life. He gave five criteria for a good definition, which I like: (1) sufficiency, (2) common usage, (3) extensibility, (4) simplicity, and (5) objectivity. It should sufficiently enable us to say whether something is alive or not; it should apply to all common examples; it should be extensible to difficult examples, including viruses, mules, fire, Gaia, ETs, robots, etc.; it should be as simple as possible; and it should rely on measurable and objective properties so that anyone can independently use the same definition to arrive at the same conclusion about whether something is alive or not.

Morales begins with: “Living things are systems that tend to respond to changes in their environment in such a way as to promote their own continuation.” He immediately notes exceptions, and adds an “important addendum”: “Living things are not perfect. Through internal breakdowns, mistakes, insufficient skill or strength, or sheer bad luck, all living processes fail eventually and die.”

By using the word “systems” (implying organization), he avoids inadvertently including flames. And by not including reproduction, he avoids excluding mules, infertile organisms, or organisms like humans that cannot reproduce alone.

But within a few paragraphs, Morales admits that it fails: “…seemingly it must collapse under the weight of a significant counterexample…” — he’s referring to altruism, whereby one organism will sacrifice itself for another, which is decidedly neither promoting its own continuation nor an example of breakdown, mistake, or bad luck. He notes that evolutionists tend to think of self-sacrificing altruism as a genetic instinct that arose to further the species — I’ll die sooner than my children in the natural course of events; by sacrificing myself for their survival, I’m ensuring that something very much like me continues a little longer than I could.

Morales thinks this is enough to fight off the counterexample, but I don’t agree. It still directly contradicts his definition, as written. While my children (or siblings, or nephews, etc.) are very much like me, genetically (more like me than other organisms), they are still not me.

Then he adds a phrase into his initial definition: “Living things are systems that tend to respond to changes in their environment, and inside themselves, in such a way as to promote their own continuation.” The three new words are meant to apply to feedback systems inside living things.

But in the end, after all this (and much more I haven’t mentioned here), he reaches no conclusion, leaving us with a definition that not only doesn’t accurately describe living things (doesn’t explain self-sacrifice), but also does not exclude our hypothetical complex machine built by humans. It is reasonable to believe we can create machines capable of responding “to changes in their environment in such a way as to promote their own continuation”.

Further, his definition doesn’t exclude corporations. A corporation is a complex, organized system that responds to changes in its environment (and inside itself) in such a way as to promote its own continuation. And his “important addendum” applies to corporations too.

We could probably add a single phrase into his definition that would exclude both robots and corporations/clubs, something along the lines of: “something not designed or developed by humans”, but we have to be careful here, since we’re very close to being able to clone humans (some say we’re already there), and the definition should be clear on whether a clone is alive or not.

• So What Is It?

Are we stuck with “I know it when I see it?” On its face, “life” seems to be a fairly basic concept, but as you can see, there isn’t a definition of it that includes all living things while leaving out all non-living things.

Before concluding, Morales includes a quotation from Aristotle — “Nature proceeds little by little from things lifeless to animal life in such a way that it is impossible to determine the exact line of demarcation” — and asks: “Could it be that the property of life is something that can be possessed in varying degrees?”

Is there a spectrum, a continuum of that elusive life quality that is possessed in very small degrees by self-propagating systems like crystals or flames, slightly greater degrees by viruses, and ever-increasing degrees by more complex systems like bacteria and multi-celled life forms? (And even more so by corporations, systems like the internet, etc.) Where would a very complex, self-replicating automaton fit into this spectrum?

• Conclusion

I am not here to formulate my own definition, but only to explore the work of those more educated than I am.

I don’t remember now why I began to think along these lines, but several subjects I’ve considered recently are related to the definition of life.

One is a curiosity I’ve always thought about — the possibility extra-terrestrial life. Would we even recognize it if we can’t define it here on our own planet? What if it was composed of pure energy — some evolutionary descendant of fire or lightning or even an energy we really don’t understand yet, such as gravity? What if it was a form of advanced AI, some combination of robots and an internet that became conscious and outlived its creators? Would we count it as “alive” in the same way we are alive — even if it could communicate with and understand us? (Conversely, would it consider us to be alive?)

Another subject is the abortion “debate” (debate in quotations because it’s really just two sides talking past each other). One side uses the slogan “life begins at conception”, yet we as a civilization still haven’t even defined life in any satisfactory way. By most definitions, the gametes (sperm and ovum) are each living things in their own right, separate not only from their organisms of origin, but also from each other. When they do meet to form a new cellular form, their previous lives didn’t end; they merely continued as a new kind of organism, one that (like a seed) has the potential (but far from the guarantee) of eventually becoming an organism recognizable as a human.

I mentioned Lovelock’s “Gaia” theory near the top, and it’s worth thinking about. Few few single organisms on Earth could survive without many of the others. Humans, for example, contain more bacteria cells than human ones, and those companions are of “vital importance in human health”. They’re in our mouths, on our skin, in our intestinal tract, and elsewhere. Without them, we can’t survive. Also, the bulk of what we eat is recently-deceased life of many kinds. Other mammals are carnivores, herbivores, or omnivores. None of them survive by eating inanimate matter. Almost all life on Earth depends on other types of life, not only for immediate survival, but for reproduction and continuance. (There known exceptions — mealworms that eat Styrofoam, bacteria that consume petroleum products, etc. — but they are rare.)

Perhaps all of us together — from microbes to blue whales, from tardigrades to humans — really are one massive organism, a vast sea of moving, interlocking and interdependent parts. All of us reside in a very thin spherical layer just above and just beneath the surface of the Earth. Everywhere else we know of is completely lifeless.

  1. Dana says:

    Very impressive. I haven’t thought this much about existential ideas since my college days in the 80s (I was a philosophy major). If I had more (free) time, I’d love to engage in a discussion/debate about the definition of life, but right now I only have time for a knee-jerk, off the cuff reply, which wouldn’t do this justice.

    • Wil C. Fry says:

      Unfortunately, my college years were spent learning that “God did it”, so I missed those discussions and lessons. Trying hard to catch up later in life. :-)

      If we ever do get to sit down and talk some day, we’ll just pretend we’re bright-eyed college students and discuss away…

      • Dana says:

        Although, as a lawyer, the “I’ll know it when I see it” methodology works just fine for me. (Every first year law student is taught the case of Jacobellis v. Ohio.)

        • Wil C. Fry says:

          It seems to work for most people, and seems to be the best we have so far. At least with “life”, most of us agree where the line is drawn (unlike “obscenity”, on which very few people agree where to draw the line).

  2. I was a math major in college. The definition of life is as irrational as pi. In pi’s case, that means that it can’t be simplified to a fraction. You can use a fraction, 22/7, for all practical purposes. But it won’t be accurate. In fact, since pi is also a non-repeating decimal, you can’t be entirely accurate anyway. You can only be accurate to the extent of the number of decimal places you’re willing to use for the value of pi. Life is like that. It exists somewhere between not-living(unborn) and not-living(dead). Sperm is alive. And an egg is alive. Yet when they combine, we say that they create life. (That’s a bit hypocritical, isn’t it.) Tardigrades can encyst themselves and stop all their metabolic functions without dying. (Okay. Some die and some don’t. But the only way to tell is to try and re-animate them. If they do; they weren’t and if they don’t; they are.) And I only remember being a bleary eyed college student. I’m not sure that ever went away.

    • Wil C. Fry says:

      Thanks for the contribution, Michael.

      Indeed, just as a few decimals of pi are “good enough” for most uses (3.14159 is how far I memorized it), a few words in the definition of “life” are good enough for most uses.

      But unlike the value of pi, the *definition* of it is very precisely defined, if I’m not mistaken: “the ratio of a circle’s circumference to its diameter.” If you manage to make a circle (which is also precisely defined), you can measure its diameter and circumference, and come up with a ratio.

      With life, however, both the value and the definition are ambiguous and disagreed-upon.

      I think.

      (I majored in neither in college, nor do I have a degree.)

Write a comment...

Welcome , today is Tuesday, 2018.01.16