Life Is Complex

From DefinitionOfLife

Just how complex does something have to be to be alive? Are stars complex enough?

The state that IS maintained in a star isn't sufficiently complex to call it alive.

"isn't sufficiently complex" as measured how?

What structure a star or even large planet possesses occurs is a direct result of gravitational and thermodynamic forces. It's simple enough that they constantly form on their own. Contrast with bacteria, which may have appeared as infrequently as once.

That's a very poor argument since you have no idea what the frequency of bacterial formation might be. Put another way, if you somehow were to find that bacteria formed spontaneously with great frequency throughout the universe, would that make bacteria be not alive?

No. Unless you think the earth is somehow hostile to life, the frequency of spontaneous generation would still be only one or two events per planet per several billion years, in contrast to stars which each form independently and similarly from any suitably dense and disturbed cloud. This is simply a reflection of the enormously different information content in the structure of stars or bacteria (or rather, the parts of the structure that are preserved).

Since you can't quantify the information content of either stars or bacteria or even know which is greater than the other, the argument is meaningless. You might also note that spontaneous generation is not part of the definition. Ultimately the terms "complex" and "order" in the definition are completely subjective and the term "function" has no meaning, such that the definition becomes equivalent to, "Something is alive if I say it is", which at least has the virtue of being honest about it.

Except you can, since information relates to entropy, which incidentally is connected to frequency of spontaneous generation. Even if they were subjective, however, it wouldn't matter, it would just mean life is a subjective concept (like red or hot, still useful concepts). I'm sorry, I answered your question because I thought you wanted an answer, not because you wanted to expose some supposed flaw already discounted in the discussion below. Yeesh. So then spontaneous generation is now part of the definition? Yeesh is right.

No, but as I just said and is explained below, spontaneous generation relates directly to information content. Please make an effort to understand what I've said before you criticise it, ok?

I understand that the identical definition as applied by different people's definitions of "complex" and "order" gives different results to the question of "is this thing alive" for any given thing. This makes the definition worthless. If you can specify the amounts of "complexity" and "order" in the definition, then do so in the definition.

No precise demarcation is given because life is a fuzzy concept, and this is a strength rather than a weakness in the definition, as already discussed below. Why do so many people edit this page with stuff already discussed on it?

By "fuzzy" you evidently mean "subjective".

No. Fuzzy means imprecise.


But not enough structure. All the things listed above have less structure than a single self-replicating molecule of RNA (about 148 bases long).

A fire has structure on a vast fractally interconnected and interdependent set of scales and timescales. So what are you driving at?

Fractals are exceedingly simple structures. Far, far simpler than the simplest self-replicating molecule of RNA.

Most "complex systems" have a structure at many different scales -- in other words, they have fractal structure. Trees, blood vessel systems, human lungs, etc. are all clearly fractal. (But I agree that some fractal structures are generated by very simple non-living processes). If something does *not* have fractal structure, I think it's probably not complex enough to be alive.

The concept of life doesn't have any sharp boundaries, so any faithful definition of that concept won't establish any such boundaries.

The minimal "enough structure" will be set by whatever everyone agrees is alive that has that structure. Since this is not agreed upon, the definition leaves it as an open variable. This is not a weakness of the definition, it's a strength, it generalizes it so everyone can use it regardless of particular opinion.


(Following discussion in LifeMaintainsItself)

(so let's return to laser beams, but inside the light-emitting body itself. They consume energy (is that what you consider "actively"? If not, please define), they maintain their structure, so are they alive? If not, why? -- nb)

Inside of the emitter ... the beam is maintained passively by the emitter. The emitter itself isn't alive because it doesn't maintain its structure. Actually, you have a point. The photons are using energy to reproduce. However, the (single) photons don't have enough structure to qualify as alive. They don't form a 'system' on their own. -- rk, edited by DavidCary

There is a relatively complex electromagnetic field inside the resonator. It can be described as several modes (standing waves) interacting via the electron subsystem (the modes compete for suitable electron-hole pairs). And I cannot say this system doesn't maintain its structure (clearly recognizable by scientists) or does it "passively". But I also wouldn't say it's alive. At least... it's not yet. -- nb

You've made a convincing case that the standing waves are thermodynamic systems which actively maintain their internal order. However, standing waves in a laser beam emitter are not complex enough to be considered alive.

FrederickPohl?'s Gateway (SciFi) describes a species of ArtificialIntelligence made up of standing waves inside of a matter-less black hole. If they were possible then they would be alive.

On a fundamental level, the standing waves in the emitter aren't much more complex than the simple flip-flop memory units made up of logic gates inside of a computer. They consume energy to maintain one of two states. The high-level information contained in one of these units is very low (a single bit) and so it's not alive. A laser beam might contain several bits but this would still be much too low. -- RichardKulisz

How many bits does life require then? Is there some particular number you have in mind?


Discussion on the importance of the principle of biogenesis... and viruses and prions

I think the easiest way to handle things like the crystal example is probably by biogenesis. Crystals can come from other crystals, but they can also arise all by themselves. Not so with fruit flies. At some point life appeared spontaneously, but it doesn't happen every few days. I would suggest that this is the essence of life - being a very high entropy state that is common because it copies itself, rather than arises spontaneously, though I wouldn't be surprised if this is too broad. It's pretty vague, but you are deluding yourself if you think there is a sharp dichotomy between life and non-life. -- JoshuaGrosse

There is a clean separation between living and non-living. The variety comes in the various forms of life and non-life.

This is simply not true. Biologists had one heck of a time deciding whether or not viruses are alive, and although the general similarity of our biochems has more or less left the consensus that they are, there are other intermediate forms. In fact you can find a near-continuous spectrum all the way down. Good examples near the bottom are prions, which are slight variations of natural animal proteins that happen to catalyze the very change in DNA that creates them.

In short, prions don't contain their own gene, just on how to assimilate something to their gene. Was it reproduction when the borg assimilated Picard (and does anyone have a decent analogy)? This is once again the problem of where the info is coming from, which is discussed below.

Biogenesis does not add any power or usefulness to the definition of life, it's just a curious side-effect of natural life's affect on its own environment. The same is true of entropy; it just so happens that most living things display negative entropy, but this is just an interesting side-effect of the more fundamental requirement of replication.

Biogenesis is hardly a curious side-effect. A good example here is mutating computer viruses, which I think you would admit as alive, and at the very least make a good sim. These do not usually just appear from the random bit-patterns on your computer, and it's not because of something they did. In fact, if they spontaneously appeared frequently enough (more often then they reproduced, say), you would always have a random mix; natural selection would not occur, and they would be treated pretty much the same as fire. The infrequency of spontaneous generation is crucial for life and evolution to work; the fact that living things have low entropy is obviously tied up directly with that. You shouldn't dismiss these so lightly. In fact, I would bet this is the crucial feature of life - see RK's definition above.


Moved from DefinitionOfLife

It doesn't matter. Ripples are simply not complex enough to qualify as life. LifeIsComplex.

How complex is complex enough?

this question is never answered

Then the distinction is useless.

[It hasn't been answered on this page, but some things have been noticed about the subject in recent decades: the greatest complexity is found neither in complete order, nor in complete randomness, but roughly in between, at the onset of chaos. This is not just metaphorical, it can potentially be quantified for a system in terms of the behavior of the system's state space (the statistical mechanics == information theoretical approach to thermodynamics).]

I still think some kinds of life might be very simple (compared to our kind of life).

You know, if anyone had bothered to read this page they'd know that How complex is complex enough IS answered on it. Something can only be complex enough if its abiogenesis is exceedingly improbable.

Not a good argument because you don't specify the environmental conditions under which the probability of spontaneous generation can occur. Trivial example: in a large expanse of space seeded with sufficient quantities of hydrogen and helium gas stars are more likely to form than bacteria. In aqueous conditions with sufficient quantities of appropriate organic molecules, bacterial are more likely to form than stars. This argument is considered in more detail in cosmologist Fred Hoyle's novel "The Black Cloud" (see http://www.daviddarling.info/encyclopedia/B/BlackCloud.html)

Oh, of course, which is why scientists have such a swell time doing abiogenesis. In fact, abiogenesis is a first year undergrad lab in any half-decent biology 101 class.

The facts are that we know of at least a dozen spontaneous LARGE planetoid formation events in our solar system alone. Outside of our solar system, we know of several billion stellar formation events and we can extrapolate many times more planetoid formation ones. How many spontaneous bacterial formation events do we know of in the entire universe? ZERO. How many spontaneous self-replicating RNA formation events do we know of in our solar system? We speculate there were no more than a few, maybe as low as one.

You're assuming that all life is like our life.

Of course, life is defined by being in some ways like our life. But I think you're missing the point. The full range of living things on earth originated from one, possibly two or three abiogeneses. Thus, unless the earth is unusually hostile, you can expect only a few biogeneses per planet's worth of appropriate materials. A jar of water and organic nutrients won't produce new bacteria unless some were added. In direct contrast, stars form on their own, constantly. In short, even given appropriate start materials, bacteria are many orders of magnitude less likely to appear than stars. Here appropriate means high entropy, i.e. you can't start with spores; exactly how high entropy doesn't matter, so long as you compare things according to the same standard.

Entropy provides the standard measuring stick between aqueous nutrient baths and interstellar gas clouds.

None of that implies that all life is as complex as our life. The universe is a big place. There could be much simpler replicators out there that have evolved to meet any reasonable definition of life. There could be forms of life that started out complex and became vastly simpler.

Of course there could be simpler replicators than bacteria. Viruses and prions give some idea of what the lower limit of complexity to be considered alive is. It may be noted that if something regularly appears through spontaneous generation, its self-replication is not especially distinct. It might also be noted that self-replication is not necessary for life, as sterile organisms prove. But all these things were already discussed above.

Replicators are necessary for evolution. I'm not sure what "spontaneous generation" has to do with this discussion.

Spontaneous generation, i.e biogenesis, relates to complexity - it is the discussion. Replication is necessary for evolution, which allows more complex forms of life to develop. However, some living things (such as mules) do not replicate.

Replication may be a prerequisite for life. It's a prerequisite for evolution, and evolution may be the only way that life comes into being. I'm not convinced life on earth was generated "spontaneously". My personal suspicion is that there were other replicators that evolved into RNA, or somehow enabled RNA to form. I'm unaware of any known living thing that isn't the product of a replicator.

I don't disagree with any of that, though I would note that living things don't have to be replicators, as sterile organisms show and strong AI potentially would. And if life on earth didn't appear spontaneously (I suppose it depends on how many steps you allow), that only reinforces how unlikely it is.

Inside those sterile organisms are replicators. I'm not saying that each organism has to replicate to be alive. I'm saying all known living things are produced by replicators (RNA and/or DNA). And I doubt strong AI will be free of replicators.

Fair enough on the first, much less likely on the second unless you count weak things like ideas. But where are you going with that?

The replicators that have spread over the surface of the earth (RNA and DNA) may be far more complex than replicators that arise on other planets, or replicators that predate RNA on this planet. There could be forms of life that are vastly simpler than ours. Complexity isn't a precondition for replication and may be a localized consequence of it.

I agree entirely. But replication isn't a criterion for life. Counter-examples are many, including most notably fire. Early RNA can be compared with viruses (variously considered alive and not), virioids (usually not), and prions (almost always not). So all this stuff about simple replicators doesn't reflect on life. This is already discussed to a painful degree on FailedAttemptsAtDefiningLife, and doesn't need to be repeated here.

Replication may be a criterion for life. It's a criterion for any evolutionary process (not just biological evolution) and it's hard for me to imagine how life can arise except as a product of evolution. By simple replicators I don't mean viruses, virioids or prions. I mean chemical replicators we've never observed.

I understood that. I'm simply pointing out that simple replicators don't count as alive, with these and fire as examples. Further, although all known living systems involved replicators, they don't have to be replicators. Considering living things as by-products of evolution has the fatal flaw of not distinguishing between sterile offspring and excreta. All in all, it strongly looks like replication is not a sufficient or necessary criterion for defining life. It comes into things because it's the only way of construction objects too complicated form spontaneously, besides artificial design. Once again, this is already discussed.

By the way, without a minimal complexity requirement, how do you hope to exclude fire?

I don't. Excluding fire is less important to me than including all life. Heraclitus believed that everything was fire.

If you don't exclude fire, your definition doesn't match the one people use. In that case, there's the very valid objection to be raised that what you're calling simple life might be nothing more than alien fire, which would not be interesting for the same reasons. And a definition that includes all things we consider alive defines what other things would be, so can't exclude any. Expecting biologists to study minerals is a waste. And Heraclitus doesn't come into things, because he would still distinguish between a naked flame and a horse.


See also: ComplexityMetrics

EditText of this page (last edited September 14, 2009) or FindPage with title or text search