Definition Of Life

An ongoing discussion on how to define "life".
A system is alive if the whole continuously adapts to the behaviour of its parts.

Define "part" :-)

Define "define" ;-)

{Define LaynesLaw}
A complex system is alive if it uses energy to maintain the order integral to its internal function.

Note this definition was established by fiat, with short or no shrift given to alternative viewpoints. Many here feel that it is not systems, but processes that are alive. Others have suggested that life, rather than being some distinct property of parts of the universe, is itself universal - UniversalLife not in the sense of insurance. And several have questioned whether you can define life in terms of "order" or "function".

Nevertheless we expect the defenders of the faith to descend on this page and wipe out anything that doesn't agree with their preconceptions, ThreadMess endlessly as below, and generally make an unpleasant experience occur. If you don't want to participate in this unpleasant experience, best to leave the definition here alone. Consider instead EvolutionAndEnthalpy, the LeibnizianDefinitionOfConsciousness or some of the other so-called FailedAttemptsAtDefiningLife if you'd like to discuss the fundamentals without getting the cods knocked out of you.

What can be gleaned from this is that:

"Maintain" is an active verb. Is it even possible to "maintain" anything (alive or not-alive) without using energy?

Why is it necessary to specify "the order"? Why not simply, "A sufficiently complex system is called 'alive' if it maintains its internal function''.

Precision. The definition is still far more concise than the Physics-based Definition in FailedAttemptsAtDefiningLife.

It may be preferable to allow "maintains and/or duplicates its internal function", again allowing viruses to be alive throughout their life-cycle under the definition.

This wouldn't allow viruses to remain alive throughout their lifecycle. Nor would I think it correct to say that viruses are alive throughout their lifecycle. This contrasts with the common-sense view of the meaning of "alive" (see Dead -> Live transition below).

Biological viruses generally don't do much activity throughout their "life". They use brute reproduction as their "error correcting" mechanism rather than "self repair". Yet, I would definitely consider them "life", primarily because they are heavily dependent on and respond to natural selection, unlike prions. Prions don't "evolve" to our knowledge.

History of the definition:

A complex thermodynamic system is called alive if it actively maintain its information content.

Isn't entropy information?

A complex thermodynamic system is called alive if it actively maintains its structure.

Doesn't structure exclude logical information in a computer?

A complex thermodynamic system is called alive if it actively maintains its internal order.

What does 'active' mean? What about a structure that keeps it's structure by being composed of materials that can withstand the forces of gravity, wind, etc. Does that count as actively maintaining it's order? Does a clock, with intricate and varied parts that interact to maintain an easily definable order actively maintain it's order?

A complex thermodynamic system is called alive if it uses energy to maintain its internal order.

Doesn't internal mean inside?

Internal doesn't mean "inside". Human skin is part of humans' internal order, the toxic heavy metals in human bodies aren't. But it's worth specifying,

A complex thermodynamic system is called alive if it uses energy to maintain the order integral to its internal function.

What's an internal function?

Internal function means something like a clock's function of keeping even time intervals or a human body's function of maintaining itself. Internal functions can be self-referential or even trivial. What they can't be is external. Like a clock's function of keeping the correct time which is external since it relates to an external standard like a time zone.

Simple thermodynamic systems can't be called alive?

No. Counter-examples are many, including fire. While fire is an essential process in the life cycle of many living creatures, most notoriously Australian eucalypts, which cannot reproduce without burning, it's no more alive than air and water.

Is the word "thermodynamic" crucial to the definition?

You're right, it's redundant with "uses energy". So,

A complex system is called 'alive' if it uses energy to maintain the order integral to its internal function.

What does "complex" mean?

Complex is another thermodynamic concept, dealing with information content. See LifeIsComplex


The definition is correct because it maps well onto the concept of life in all of the following cases, many of which cause FailedAttemptsAtDefiningLife to founder:
  1. cars and refrigerators are not alive since LifeMaintainsItself
  2. laser beams, black holes, ripples in a pond and fire are not alive since LifeIsComplex
  3. sterile humans and non-reproducing mules are still alive, but
  4. frozen, vitrified or otherwise preserved organisms are not alive since LifeUsesEnergy and life maintains ITSELF. They are not alive regardless of whether or not they can be revived (be made alive) under suitable circumstances.
  5. spores and seeds are not alive so long as they are completely dormant
  6. simple self-replicating RNA molecules are only marginally alive,
  7. the genome of a virus is alive when inside a cell when it is maintaining itself by replicating backup copies, and
  8. it is not alive otherwise,
  9. prions are not alive because they do not reproduce their low-level structure and their high-level structure is simply not complex enough
  10. species as individual entities are not alive, maybe,
  11. a biosphere could conceivably be alive,
  12. but ours is not,
  13. the sun is not alive since LifeIsComplex and stars are not
  14. a universe could conceivably be alive (you don't need a separate environment or heat sink to be alive),
  15. but ours is not,
  16. Any self-maintaining computer program is alive since a computer is a thermodynamic system that uses energy for information processing. There is no minimum amount of energy that must be used for information processing, but some energy must always be used.
  17. Any mind, including AI, is alive on a higher level since it must continually maintain and organize its memories in order to prevent its personality from dissolving into chaos.
  18. A perfect simulation of you and your environment running in a computer (which eats simulated food, reproduces simulated babies, etc) would have simulated life.

Interesting cases:
  1. ball lightning
    • Ball lightning disappears in a wink. Really? See What makes you think it's alive? Unlike laser beams and black holes, they might have a complex structure with feedback loops that maintain that structure.
    • {Ball lightning once sent me spam. "Ability to send spam" may be a qualifying factor for "life".}

Life does not always maintain itself. Some people require assistance, or assisted living, such as in an assisted living facility.

Life does not have to be complex if believe bacteria is alive.

The sun may not be alive, but it does have a lifespan.

In other words, your variables are flawed.

Original title: non-living to living transition

Living, dead and dormant

Strange that this definition permits non-alive things (spores, dormant seeds, viruses) to become alive. As commonly understood, this transition is forbidden and only the converse (alive -> not-alive) is allowed.

Metaphors and figures of speech do not prove a thing alive. Just because one can kill the lights doesn't prove they were ever alive. Would it help to say that dormant spores need to be 'deactivated'?

Note also that in medical terminology, organisms make the transition from (dead -> alive) literally all of the time. In those circles, dead just means something like 'irreversibly passed the point where an organism can sustain itself by itself'.

There is no utility in having a definition of life which does not match the common conception of crucial parameters of being alive. As it says on LifeMaintainsItself,We're trying to find a good definition of life that covers the common-sense idea of "alive". This easily encompasses the concept that non-living things do not become alive.

The fact is that seeds and spores are dormant. They aren't alive and they aren't dead. They're dormant, a completely different third state of being. And you know how I know? Because biologists don't call seeds alive. Really? How many biologists have you asked?

Most biologists do call seeds alive, but then, most seeds are alive in the sense given here. Full dormant states, more common in spores (how depressing is it that anthrax, a very rare bacterium [anthrax is not at all rare], is the first example here? [Anthrax was used as an example since the news is likely to have made most people familiar with the concept of spores. Biologically, spores are also not at all rare.]) are less clear. People speak about them as being killed, but they also speak about germination as coming back to life. It's fair to say they're a grey area in common parlance.

Disagree. When people speak of spores being killed they are speaking factually. When they speak of germination as "coming back to life" they are speaking metaphorically.

I don't see how you can come to that conclusion. In cases like killing lights, people are speaking metaphorically, as you can tell by asking them if lights are really alive. For spores, it's not so clear cut. You're only assuming one is metaphorical, and it could just as easily be the other.

Would you call a non-running computer program alive? No, you would not because this would be completely idiotic. The exact same thing with a seed. It's got the potential for life when placed in suitable conditions but it simply isn't alive. Again, disagree.

This is your opinion and you are entitled to it, but it's not an argument.

Oh, but it is. And not only because opinions are arguments when defining concepts, but also because the analogy between a computer program and a seed is exact and I can leverage this analogy to change everyone else's opinions. Since there are no examples of existing computer programs that anyone would call "alive", the analogy is meaningless. Effectively, it is drawing an analogy to something imaginary, which is not at all convincing.

But the killer is the fact that you chose to title this section 'non-living to living transition' instead of 'DEAD to living'. Obviously, you're willing to accept a distinction between the non-living (dormant) and the dead, even if you claim there isn't in order to attack the definition on a perceived fault. IOW, you're fighting against a definition you yourself believe is correct.

What do all the little words mean?

Mention of integral function and integral order aren't begging the question so much as making sure you know which structure you're talking about. The currently suggested definition is in fact behavioral - it says that the state maintains itself against minor disruptions.

States and structures are only your own interpretations of process. They don't exist except in your interpretation of process.

So what? The logistic map is only my interpretation of a process, depending strongly on how I choose to count objects. When I talk about a frog, that's only my interpretation, since there's no hard and fast way of defining which atoms belong and which don't. When we talk about an object being alive, we need a concept of an object first; that's not surprising, it's not a weakness, and it's not something any other definition improves upon. It's also been discussed to death. Why does it need to be hashed out again?

Structural problems of the definition:

Life is a fuzzy concept. A definition of it doesn't NEED to be any more rigid than the concept itself calls for.

No, it's not.

No, it's not. Only external functions are teleological. But those are specifically excluded from the definition.

Proposed further pages:

Removed AutoPoiesis, which is just self-reproduction, and another attempt, which is just natural selection. These things have already been discussed on FailedDefinitionsOfLife?, and unless something new comes up, don't need to be discussed again.

Borderline or Oddball cases to consider

This page is long enough that you should search for answers to your questions in it before you pose the question again. Especially questions which contain obvious keywords like 'prion'.

Prions do not reproduce but propagate. Prions are simply not complex enough to qualify as life.

A prion is only a single level of structure above its fundamental substrate, and its feedstock is at the same level of structure as itself. A prion is a type of protein and proteins are made of amino acids. A prion does not in any way alter the amino acid sequences of other proteins, much like living things do not alter the nuclear structure of matter or the quark structure of nucleons. What prions do is they interact directly with other prion proteins by transmuting them. A group of prions turns another prion into its own kind. This is very similar to what fire does.

Even viruses are three levels of structure above their fundamental substrate, and their feedstock is three levels down. You see, viruses aren't just proteins but highly-organized proteins. And not just any old proteins either but special viral proteins. Viruses alter cells at the level of individual DNA/RNA (genes) and protein sequences (proteins), which build up into chromosomes and complexes, which build up into a virus.

Prions are even simpler than horning genes, which are a single level of structure above their fundamental substrate (nucleic acids) and whose feedstock (other nucleic acids) is one level below them.

And then there's the question of whether a prion is even as complex in its one level of complexity as a virus is at only its topmost level of complexity. Somehow, I doubt that.

What if we try this definition for a while?

1) Evolution is a property of life. 2) Non-perfect reproduction of information is a requirement of evolution.

Which can be written in one line if so preferred. (source: -- ChristianAnder

Evolution is not a necessary property of life. It's a property of biological life, but then so are:


Why don't we ask who buys the current definition and who doesn't?

- Perhaps one can use one definition in science(nature), and one for a more general explanation. People tend to have their own opinion about life, we need more time. The one in the top is a very good textbook description of what we think about life today, but a more strict one is needed for a scientific approach.

Because it doesn't matter, what matters is whether it works. Nobody has come up with a convincing counter-example to the current definition, although some have criticized the form. Most people, though, are just insisting that self-reproduction is necessary for life even though they don't hesitate to call mules alive. I don't understand why this attracts so much hatred - I was happy to find something that works better than the usual textbook version.

It matters, because as near as I can tell, the only people advocating the definition on this page are RichardKulisz and JoshuaGrosse. Is there anyone else?

Other views seem to just get shouted down, ignored or deleted.

The main views being added are that life is about self-reproduction and evolution. These have not been ignored, but they were discredited. Anything new that takes into account the points already raised is welcome. Honestly! What have been deleted are duplicates of material already on these pages, so the views are still presented in the discussion.

Planetary Weather

Discussion moved to LifeMaintainsItself, where it should've been all along.

Is this conversation alive?

"Is this conversation alive?"

The obvious answer is a question: "are you new to wiki?". If the answer is yes, please pay attention to the top of the page where it is stated: "Note this definition was established by vicious warfare in this community". And then you can enjoy reading various pages by or about RichardKulisz like RichardKuliszFanClub, etc.

This wiki community can hardly resist vicious warfare on topics where we are not expected to be taken seriously for obvious lack of competence, so the conversation is dead for all intents and purposes. If you really want to resuscitate it, you're welcome to try :)

Thank you (the both of you). I am new to wiki. I didn't realize it showed. I see now the read awhile before posting suggestion is quite a good one. I have learned a lesson about life (whatever that is). I did enjoy the links, thank you again.

To make what they were saying even more clear: this page largely represents the opinions of RichardKulisz, who has fought extended and fierce EditWars to make sure that opinions he disagreed with did not remain here. The fierceness of his opposition to other views caused essentially everyone but RK to eventually just give up on this page.

That's not exactly the ideal for wiki pages, to say the least, but I think that the community gave up on this page in large part because biology issues aren't absolutely dead on topic here (although they're not absolutely off topic, either, so the majority of the community didn't care all that much to start with, unlike other topics that a lot more people cared about and had expertise about.

It also helps that RK had some reasonable technical points on the subject (it's not like he was babbling about protoplasm and mystic life force), even if some people disagreed that he had the absolute final word on the subject that he thought he had. -- DougMerritt

BTW - If I did resuscitate this conversation, would I be creating life?


From a chemical point of view the above cannot be correct. I know many systems (BZ reaction on a thin plate) that shows complex behaviour, and uses energy to maintain its internal function but it is not alive.

That's an interesting thing to note, but I don't think it's actually a counter-example. Some crystals are reasonably intricate, but aside from self-assembly, they don't do anything to preserve their structure. It seems to me like BZ reactions fall in the same category. They show somewhat complex behavior (though not nearly as complex as living things) but in what sense do they maintain their complex behavior against external forces, any more so than ripples do?

Viri are self-maintaining, they are just dormant much of the time.

This topic needs to be reorganized in my opinion, hopefully with a TOC of sorts. I could not find a long discussion on evolution.

Did you try NaturalSelection or SelfReplication? Or perhaps RobHarwood's original NaturalSearch?

FYI, LifeIsComplex, LifeUsesEnergy, and LifeMaintainsItself was meant as a TOC of sorts. But it didn't get used so it didn't exactly prompt any further refactoring.

My proposed definition: "Something that has the ability or recently had the ability to produce copies of itself, and the copies have the potential of being altered somehow to change their characteristics, yet are still able to produce copies. There must be at least 10,000 survivable potential variations of the copies. Products of failed or crippled copies are also considered "life" as long as they have the ability to maintain themselves (consume energy) for a period of time." -- top

Tests against border-line cases:

By this definintion retrotransposons are alive, but most people would not consider them to be so.
A possible new life-form that may be semi-life, somewhat like prions. It's still unknown if it can "evolve".

It is possible that there are certain complex proteins that form cell-like units that reproduce on their own but are otherwise exact clones and not subject to variation. It is kind of "existence selection": they exist and reproduce because nothing stops them.
Prions Evolve :
I suggest much of this may be made mostly moot if we reject the idea that life is "Boolean": all or nothing. Something can be "more alive" than another thing. For example, bacterium may be said to be "more alive" than a virus, or even a dormant spore. See DefinitionOfLifeDiscussion for some possible treatments of this idea. -t

By the way, one of my favorite (but imperfect) short definitions is "Something that produces or induces copies of itself, with variations". It has a high accuracy-divided-by-length rating in my opinion, kind of like 22/7 as an approximation of pi. -t
A definition by using word statistics of a corpus of definitions of life arrived at:

Life is [a] metabolizing material informational system with [the] ability of self-reproduction with changes (evolution), which requires energy and [a] suitable environment.

Or shorter Life is self-reproduction with variations.

See for an explanation of the

"Requires energy" seems superfluous. Reproduction probably cannot happen without it. (Note that parasites steal the host's energy). Same with "suitable environment".
Life is the flow of energy/matter from one form to another. A group of energy/matter that changes form more frequently is considered to be more "alive" then something that changes form very rarely.

Does that make water "alive" when it goes from snow to water to rain, etc?
Another candidate: Something that is the product of evolution that changes behavior in a predictable (patterned) way in response to changes in the environment or external stimulus.

Note that excrement may change its nature based on the environment, but generally not in a repeatable way. Also, the definition intentionally does not exclude an impotent object.

Life is an unstable, yet self-sustained process amidst an chaotic, uncontained environment. (credit to:

...defend or destroy!!!

{But one has to define "sustain", "uncontained", "chaotic", etc. to be rigorous enough for our needs.}

Is this page supposed to be a parody of some sort?

It's not supposed to be, but of course that has little to do with whether it is or not.

See also UniversalLife, DefinitionOfDeath, EndOfLife, NaturalSelection, WhatIsInformation, SelfReplication, HumanUniversals, DetectingLife, DefinitionOfLifeDiscussion.

Should fit within the DefinitionOfUniverse and the DefinitionOrdering (?)

Contributors: RichardKulisz, JoshuaGrosse, RobHarwood, DavidCary and AnonymousDonor

CategoryDefinition CategoryBiology

View edit of December 1, 2014 or FindPage with title or text search