There was a definition I encountered in high school, which I never found a source for, but seems to be pretty robust and address many of the non-living things other definitions include.
This definition is a rubric of several qualities:
1. Metabolizes energy
2. Stores Information
3. Self-replicates
#1 rules out crystals; they are formed by outside forces, they don't metabolize themselves. It also rules out viruses and prions; they don't ingest any "food" to perform metabolic activity. #2 rules out fire; it doesn't store information in a DNA-like molecule or anything simliar. It's purely a chemical reaction. #3 is the obvious thing that differentiates non-living things like rocks from plants and animals.
There are still some cool edge-cases. Is a sperm cell alive? An X-chromosome sperm (do X-chromosome sperm have a (probabilistic N)-stage lifecycle?) Are red blood cells alive? Is every molecule of an organism that's alive also considered alive, or just the ensemble? If only the ensemble, are mitochondria alive?
Also, why we consider a single human as living organism, but not a society? Single human without society is a pitiful and helpless chunk of meat, who unable to replicate btw.
Sperm cells can be seen as a haploid phase of many organisms' lifecycle, so are alive by pretty much any definition.
In humans, the haploid phase of the lifecycle is single-celled, while the diploid phase is multicellular. In contrast, in mosses and fungi the haploid phase is multicellular while the diploid phase (sporophytes/zygote) is single-celled.
The unfertilized eggs of bees become drones, so they must be alive. As a side effect, this indirectly make bee-like insects evolve a lifestyle with a big colony with a queen.
In fungus, most of the life is as haploid (i.e. a single copy of the chromosomes, like sperm and eggs) instead of diploids (i.e. two copies of each chromosomes, like most of our cells.)
In some ¿unicellular organism? [I can't find a good link now] the haploid and diploid versions are almost equal.
[And plants are also weird, some have 4 or 6 copies of the chromosomes instead of 2.]
Self-replication doesn't seem necessary. A human who is infertile is clearly still living. Worker ants who can't reproduce would also be well within the living side of the spectrum to me. I give these examples because it seems to me the definition of living should apply at the individual level, not the species. Individuals after all are the ones doing the dying.
replication is still happening in the lower cellular/molecular scales, even when the aggregate organismal level isn't reproducing that pattern. People who take these definitions very seriously aren't implying that elderly people aren't still alive :)
Self replication of information with mutation and evolution would be my definition, which would include viruses, but not crystals, AI could be included if it could evolve on its own.
I think in reality like most things its not black and white, there is a continuum of life and some things are more alive than others.
Edit: perhaps adapting to environment through mutation/evolution or learning or both might be better.
Evolution has led to life but I'm struggling to see why it should be a criterion. What if something happened that stopped humans evolving, and we maintained the same general genetic distribution for the next few millennia? Would we stop living?
Which is why I added the edit. The thing I consider most defining of life is its ability to adapt to its environment both through evolution and later through intelligence and learning.
I’d say life is defined by identity (a bunch of information defining a structure), or a self, and the processes that allow this self to stay identical to itself in time. Something like homeostasis at the organism level. #1 and #2 would be utilities to realize this, and not necessarily integrated. Arguably #3 isn’t necessary as others have said and could be seen as only one manifestation among others of the self-preservation process. A trivial illustration of this definition on which we can all agree is death, where self-preservation processes break and identity disintegrates.
Interestingly this definition would encompass countries and probably any social group as long as they have a name, an identity and processes to maintain it. I think life doesn’t need integrated intelligence to be life, though intelligence could probably be defined as a predictive kind of self-preservation process. Essentially devising a chain of actions for moving from state A to state B with a limited set of possible operations and minimized energy consumption. An efficient way to return to initial state, or to another state that increases likelyhood of identity preservation.
As someone else said in a comment, life may be seen as a continuum where these characteristics are more or less developed and integrated, for instance making coutries or viruses living organisms while still distinguishing the unique character of humans or animals.
I think it's a thin pro-argument. Viruses are completely dependant on the host cell and the host cell's metabolism. The host cell's protein unwraps the jacket, and the host cells' proteins replicate the viral DNA/RNA payload. The virus does not reproduce itself, and it does not metabolize anything. There are no inputs to a virus.
Whereas living cells, give them the proper inputs, and they metabolize energy, catalyze reactions, and create copies of themselves.
I would definitely say biological viruses are a life form. They use as a substrate. However, you need the ability to mutate in order to evolve. Computer viruses can't do that at the moment, they are dumb machines/tools that just keep doing the same thing over and over until we wipe them out.
IMO, self-replication is the most fundamental characteristic, and the ability to mutate/evolve is key as well. The rest is all details.
The "metabolism" requirement is an important one, though. If you remove it, and you allow for "self-replication" that's entirely dependent on an external agent, then it seems like you would have to consider all kinds of things as "alive" that would fall outside any common-sense definition of the term. For example, works of literature.
It seems to me literature doesn't self-replicate, we replicate it... And apart from the bible and such, literature usually dies out fairly quickly in the grand scheme of things.
Viruses rely entirely on the host cell's proteins for all of their replication. They don't metabolize anything themselves, nor do they create their own copies of themselves. The host cell creates copies of the virus.
I agree with the other responder. However, humans stopping evolving seems unlikely to happen. For instance, right now, we have birth control. Birth rates are declining rapidly. I don't think that's going to continue. We are in the process of selecting for people who have more children. Nature will find a way to evolve around birth control.
Both you and the other responder just completely ignored my question though. I'm not asking if humans will stop evolving or if other species have gone extinct. I'm saying that if we somehow used CRISPR to stop genetic mutation (let's assume it's possible), but in every other way remained the same, are you saying we would no longer be life?
I'm asking this question because the aforementioned definition of "life" is flawed in my opinion. Evolution has nothing to do with it.
1. Unless you have some way to stop time, it is impossible to stop humans from mutating/evolving.
2. I would argue it is part of being human, and being alive, to evolve and adapt to your environment. Humans have multiple ways of doing this. Beyond our ability to genetically evolve, our intelligence is a mechanism we use to adapt to your environment much faster than genetics permit.
Life is self-replicating energy states. As thus, i would include stable energy phenomena.A https://en.wikipedia.org/wiki/Extraterrestrial_vortex could fulfill the requirements for life if it was able to manipulate its environment into producing a re-occurrence aka offspring. So self-replication, aka storing the instructions for re-occurrence into the environment system, is kind of in a flux in extreme cases.
#3 removes the whole class of AI based life forms that would not need to replicate. Our future AI citizens may not appreciate not being thought of as living.
That seems to me a form of begging the question: you assume "AI based life forms" will indeed be life forms, then argue the definition from the parent post fails to account for this new life form.
But that such a thing as AI life forms will exist and be accepted as such by scientific consensus is not a given, at all.
I'm not sure it is begging the question, at least not if framed properly.
The question seems to be to be "what exact criteria categorises what is commonly meant when people talk about what is alive". I think you could interpret OP as claiming that there is an idea of artificial or alien life as presented in science fiction which people would commonly describe as life but may not replicate.
That's not what the OP claimed, and in a second comment he/she implies AI life forms are inevitable. This is... debatable, to put it mildly.
You cannot say "this definition of life fails to account for an uncertain future phenomenon not everyone is convinced will count as life, therefore your definition of life is incomplete".
Fictional life is not real life. Authors, especially fiction authors, are not bound by the rules of nature. Sometimes they guess right, sometimes they don't, sometimes they simply provide thought experiments or parables.
Not really. There is no objectively true unarguable definition of life. When in silico life becomes reality and humans do their human thing of devaluing anything that's not like them, it won't matter what human consensus deems as life. I'm not worried about determining what "indeed will be life forms" no more than I worry about objectively defining what is a good meal or which is the most beautiful color. Humans have no standing it determining what true life is. What's important is developing an intentional stance towards whatever is not us.
The human mind invented the concept of life. There’s no objective definition other than what people decide that it means, because no other form of intelligence tries to make the distinction.
I’m pretty sure siliconers do pasteurize their silk and disilfect medical equipment as well. Life is too intrusive and too eager for food to not care about it.
Bees and Humans fit the definition. Therefor, aggregations of bees and humans, like hives and countries, also fit the definition.
"Countries" seem more of a stretch, as replicating would technically mean creating other countries like itself. Not just sustaining itself into the future.
Do hives seed other hives? I suppose they would, so they better fit the definition.
Individual humans don't quite fit the definition, many don't reproduce at all, the ones that do only do in groups of two. Strictly speaking you'd have to wait and see whether a given human child eventually reproduces to decide it's alive...
But with bees it's much worse -- in each hive, only the queen and some drones reproduce, most bees are worker bees that can't. So this definition fits bee hives better than it does individual bees.
Seems the definition would need to say something about the species as a whole reproducing. But that leads to the species definition problem. And what about that tortoise that was the last of its species?
> "Countries" seem more of a stretch, as replicating would technically mean creating other countries like itself. Not just sustaining itself into the future.
The world is full of countries, nations and so forth. At some time in the distant past there was no nations. After that there was one. Now there is over a hundred, with the corpses of many more lying in our past.
All usable land on Earth has been claimed now so they have no space to grow in number without cannabilising one another. Their walls have grown hard and inflexible with the passage of time and laws and agreements, and the evolutionary pressures of conflict have subsided of late, as they cannot move against one another without incurring the wrath of alpha predators, and the alpha predators cannot attack one another without the assurity of mutual destruction.
But countries live a long time, and their frame of reference is different from our own. Their moods encompass entire generations of the lives of mankind. They will eventually again fight among themselves unless ordered into cells within a higher organism. They will fight and devour one another, birthing new border states or vassal states. To any greater being it would appear like an almost peaceful dance, the ebb and flow of lines on a map, the respiration of civilisation.
To us it would be every bit as savage and chaotic as it must be for the cells and bacteria which comprise us.
Well EO Wilson posited certain eusocial insect colonies, including bees, as a super-organism. I think the idea has some currency in biology. The idea is that Darwinian selection is acting upon the colony, so it is the unit of evolution. Remember that eukaryotic cells are a symbosis of two prokaryotic cells.
Not sure how often or even if countries self-reproduce.
I'm mulling this over... It's an interesting idea for sure. I'm not sure that it can't be reduced to human populations in general, which makes it more like the super-organism Wilson describes.
OTOH, not all human populations are states, as in countries, and arguable that is the "organism" said to be reproducing itself....
I've always found it amusing how in our day-to-day lives we speak with such completely assured confidence of highly complex interactions, problems, phenomena, etc. that we know something is or isn't, yet we still struggle with far more basic/fundamental questions like "what is life?"
The thing I've always loved about pure science is that the best work is approached with much humility. We do not know what something is or why something is. We have a lot of data, may be able to predict a secondary behavior or predict a primary behavior fairly consistently, but that's a bandaid at best--we know it (hehe) and try to dig deeper. Outside of science, we tend to just "know" and brush off deviations from what we know as mysticism or something else.
Sometimes I wish I wasn't aware of how little I really know compared to what I'd like to know. When you peel away the facade of certainty in a lot of aspects of life, it's a bit disheartening to see how people treat each other based on the assumptions they often make of such falsely placed certainty.
I don't think struggling to define the definition of life is a matter of not knowing or knowing. It's simply us attempting to impose our invented taxonomy on nature and reality.
Indeed. I really like the way Feynman explained how nature doesn’t care about what models or descriptions we come up with. Nature just is. This made me look at physics and science in a whole new perspective.
There are so many phenomena that are beyond human intellect, intuition, and language that the best we can do is to observe and marvel. It doesn’t mean that we should give up trying to seek explanation or modeling of observed reality, just that don’t be dogmatic about it.
I think it's more an issue of how language and human brains work, than features of the world around us.
The concepts behind words in human languages are extremely fuzzy and ambiguous. That is why progress on computational linguistics was stuck for a long time, until researchers started using statistical models.
This fuzzy and ambiguous way of modeling the world seems to better fit whatever it is our brains do. The definition of what is or isn't "life" is fuzzy and ambiguous, because however the concept is encoded in our brains is fuzzy and ambiguous.
The problem isn't "we can't come up with a definition of life", it's "every time we define life, we don't like the results".
We could make up any definition. The problem is that we instantly think "well, that's stupid, let's define it again". But I'll say it again, that wasn't ever the problem. The problem was when we thought we were talking about one thing, but we were really talking about something else - our own personal heuristic.
And nature doesn't care about your heuristic. Hueristics make terrible definitions.
Virus's are alive by a lot of heuristics. They act alive-ish. They're not alive by a lot of definitions. Bacteria are one species by definition. But since that's not useful, we make up heuristics that separate them into all sorts of species. Go through every situation, every time we are disappointed with the definition of life, and we find that the reason is that we just don't like it. We WANT something else to be called life, because it feels that way.
Interestingly, most definitions of life that would characterize a virus as alive, would also characterize ideas and culture as living (unless you add some explicit requirements like "must have DNA" or "transmission between hosts must happen by transfer of matter"). It's easy to reject that out of hand, but if you think a bit about the Ship of Theseus, then it might not be so crazy to think that what is important is the replication and mutation of patterns rather than the specific matter carrying that pattern. Cf. also The Selfish Gene, where Dawkins coined the term "meme" as a mental analogue to a gene.
Humans need to categorize things; it is how we are able to navigate our complex world, by recognizing patterns and acting on them. We get in trouble when we think our categorizations are objective fact, or that we can ever cleanly categorize things in a way that will satisfy all our categorizing needs.
> The problem isn't "we can't come up with a definition of life", it's "every time we define life, we don't like the results".
Well said. E.g., for almost any definition of life, we end up with superorganisms like ant colonies and bee hives being live things--not to mention cities and human societies being superorganisms too.
Yeah, trying to fit the complexity of reality into a taxonomy is always going to run into problems. That doesn't mean it's not a useful exercise (sometimes), but we need to keep in mind that it's just us trying to mentally organize reality.
After being introduced to Christopher Alexander's conception of life* through his The Nature of Order textbooks (so verbose...), it's brought a clarity to existence I've found otherwise lacking from other quarters.
Simply put, every thing (and grouping of things up and down the scale of size) is more or less alive depending on the order / configuration within itself and in relation to the things surrounding it. So a door can be more or less alive, taken by itself or taken in context of the wall, floor, and ceiling around it (and so on). Or compare a city with no sidewalks, completely paved with concrete, all buildings identical, square grid streets, no plant life whatsoever to some of the great cities in Europe which organically grew according to the needs of the populace over time. Can't you say one is literally more alive than the other, both the physical city itself as well as its inhabitants?
Since adopting this perspective, I've been able to make great strides in improving the order of my home which has resulted in an increase in well-being and mental health. The house is more alive and so I'm more alive for it, and vice versa, and so the positive feedback cycle continues.
He goes into great depth identifying principles we see occurring in nature to help define this gradient of liveliness and analyzes countless examples of great works of art from human history. He also shows how to bring such life to the world around you and in the things you make. Check it out. :)
I think Schrödinger in his book of the same name as the title of the article gave by far the clearest definition of life. Life is a local state of negative entropy. Which is to say that life orders matter and exports heat. There are other definitions mentioned here that are kind of equivalent, i.e life stores information, life as an out-of-equilibium state etc, but just expressing it in terms of entropy I think draws a clear line between living and non-living matter.
I think the consequence of this is that quite a few systems can be considered alive that traditionally wouldn't, like crystals or DNA itself but I don't think that's that big of a deal.
And maybe viruses, books... languages? Books seems to me a little like tardigrades sometimes, if you squint. Information stored indefinitely (hibernating?), which can be rehydrated and spring back to life within human minds. We talk about tardigrades like they're really weird, but maybe we just can't see how truly weird a form of life WE are :)
The definition of life is trivial: it is what biologists are interested in. If you want to know if something is alive, just count the biologists following it. Rabbits? Yes. Bugs? Yes. Viruses? Yes. Protons? No. Memes? Not really. Computer viruses? No. Prions? Borderline.
Might the definition change, as biologists encounter new stuff and develop an interest? Sure. Does that make the old definition wrong? No.
Dividing the world into life / non-life boxes is a purely human distinction. The universe totally doesn't care. Who does? People. Who cares most? Biologists.
I would caution that we biologists study more than life, but also its environment! So, the abiotic conditions of soil, the function of enzymes (including proton pumps :D), and even the culture of humans (and dolphins and orangutans, etc.).
It's true that life can't really be defined by some essentialist statement (i.e. it has properties that neatly divide X from not-X), and the family resemblance approach works better (things like rabbits and viruses and oaks and cyanobacteria), even though it's not as satisfying. Mind, it's OK to just tell kids in biology classes a more convenient definition.
These are good questions. I care most about these sorts of definitions when teaching, since they can orient students and describe what it is you are talking about. Some criteria I was taught about life:
> 1. Reproduces itself/making things similar to itself (sort of; viruses are on the periphery, prions...).
> 2. Metabolizes chemicals for energy
> 3. to grow, and as a result,
> 4. produces waste.
> 5. Reacts to stimuli in the environment.
> 6. Maintains homeostasis (reacts to stimuli in itself and keeps 1-5 going).
I really like these, because while they are wrong on some level, they are pedagogically useful: they orient students to what life does, what we can learn about it, and what sorts of data we can collect. Plus, it points towards both ecology and evolution.
Your definition is more encompassing in a way, so it better describes any life-form, but so encompassing that it also describes crystalization! (The parallels between life and crystals are admittedly interesting, especially when thinking about abiogenesis.)
Just from a practical perspective, I would rather start from those criteria than from physical chemistry and physics. Then, come exceptions and complications, which are great! Even when learning physics, I recall every other lecture beginning with, "Those assumptions we made were wrong. Instead, ..."
The definition I remember being taught early on includes that, so that's on the right track.
From memory:
1. Reproduces itself/making things similar to itself (sort of; viruses are on the periphery, prions...).
2. Metabolizes chemicals for energy
3. to grow, and as a result,
4. produces waste.
5. Reacts to stimuli in the environment.
6. Maintains homeostasis (reacts to stimuli in itself and keeps 1-5 going).
I'm probably forgetting some and there are exceptions to all of these. Notably, in interesting cases like at the inception of life, or at the periphery of life: viruses, parasites, potential extraterrestrial chemical systems, and arguably cultural systems. But those criteria are still convenient.
Wait till you try to find the boundary between species!
Many plants, and some animals such as birds, can produce fertile hybrids after being separated for millions of years. Especially true for organisms that have reproductive barriers besides genetic and physical incompatibility (different pollinators, geographic separation, birdsong differences, etc).
Crazy idea but what if plants are actually higher lvl beings? They are more self-sustainable and helpful to ecosystem than humans.
Actually they probably aren't because I guess otherwise all different planets would have them (but hey we are also only here so far).
There's interesting question what would god-like(unlimited power over mater and limitless knowledge) entity do? I guess killing oneself and recreating with sprinkle of rng makes some sense after such event as you probably get bored thus I'm speculating that universe could be god itself (if it's bound by same laws of physics - if not then IDK). Don't ask me what I'm smoking :p
In trying to imagine ways to unify knowledge of the worlds phenotypes (not DNA/RNA) we ran various exercises to think of just one "standard" measurement that could be measured across all life. All failed at one scale or another. An individual clonal tree (poplar), good luck measuring its mass. A blob like radially symetrical single celled organism, "length" is meaningless. You get the idea ...
This is similar to a previous HN thread [1] about a post about paper called the "information theory of individuality" [2] that talked about life as the ability to "cohere information forward in time."
There is another related idea I stumbled on called the "constructal law," [3] which seems related.
Where I think they are related to the question of what life can be defined as is that you could express this constructal force in terms of that information theory idea by summarizing it with a slight extension - as a bounded system is living based on its ability to dynamically cohere information forward in time. The limits of this constructal theory will be defined in using information theory terms.
When we discover stuff in the universe and ask, "is/was this life?" there might be some criteria in these ideas in terms of asking, does this thing take feedback from its environment and produce coherent information? Or, is this thing a source of neg-entropy in its environment? If we look at life as a temporary, dynamic whorl of neg-entropy, it might yield some useful criteria. Half-baked, but measuring information-in, information-out could tell us whether something other than classical physics is processing it. It suggests we could figure out whether something was created by life as well, since the internet is not "alive," but it is evidence of life. Sorting things based on whether it is evidence of life, seems testable.
> if you want to understand the laws of gravitation, then you go and study black holes because you have a very intense gravitational field there. And you can understand and probe gravity very well there.
> If you want to understand the laws of information and how information operates in the physical world, then you should go and study a living thing. Because that's where information actually is at its most dense or most intense.
I always wondered how does life pass information between generations. The many Behaviours that seem to be built into animals. For example like a dog barking at strangers. Or recent generations of Kangaroos that know to avoid crossing roads.
Life should be defined extremely narrowly, as in requires DNA or RNA. This frees us to consider equally complex non-living things (like AIs) on their own terms, without being hung up on whether they are alive or not.
Doesn't that make us willfully blind to when they might arrive on our terms. This sort of "let's put it aside" thinking led to some dark places in the past, and we'll eventually come to them again with non-biological complex systems of emergence.
I'd personally define it more widely, as I think that's more aligned with reality. It's much more of a gradient than we usually take it for, imho
Very interesting article, but if you are short on time I suggest stopping once it mentions Carol Cleland - it spends too much time talking about her, and not enough about the topic after that.
I would say any process with differential replication and descent with modification is life. That is, any process that exhibits evolution through natural selection.
"Your honor, Science shows us that life is vastly more complicated than antiquated ideas of 'alive' and 'dead'. I submit that my client cannot be guilty of murder as the very concept relies on binary categories that Science has discredited. Who are we to say if a murder has occurred?"
This definition is a rubric of several qualities:
1. Metabolizes energy
2. Stores Information
3. Self-replicates
#1 rules out crystals; they are formed by outside forces, they don't metabolize themselves. It also rules out viruses and prions; they don't ingest any "food" to perform metabolic activity. #2 rules out fire; it doesn't store information in a DNA-like molecule or anything simliar. It's purely a chemical reaction. #3 is the obvious thing that differentiates non-living things like rocks from plants and animals.