Whether life is carbon based or something else, we are presuming that the fundamental units are extremely tiny. We need metabolism where parts are made or modified (by existing parts), which (if naturally occurring) probably gets booted up through a synergy of auto catalysis and self assembly.

In terrestrial life as we know it, it's implemented on the smallest possible scale involving chemistry, and re-arranging atoms (and electrons) is the smallest moving parts available at our energy scale.

So I would like to challenge this assumption. Can life emerge that's not carbon-based, and isn't simply using different atoms instead in an analogous way, but is based on units that are larger than single atoms and basic molecules? Does it have to be based on the smallest available things?

I'll entertain ideas of alternatives as well as arguments as to why it indeed must be atom-scale.

Note: the relevant features of life would be self-repair and (as applicable) homeostasis, being able to reproduce, and subject to Darwinian evolution.

I'm specifically asking about the possibility abiogenesis, not whether something could be constructed to have these features.

  • $\begingroup$ I am assuming that your statement "Can life emerge that's not carbon-based" is using "emerge" to imply that you are restricting any outside help in the genesis of your form of life. In other words, no creator, ether theological or scientific. $\endgroup$ – Mark Ripley Jun 5 '16 at 10:22
  • 1
    $\begingroup$ Yes. In Code of the Lifemaker an ecosystem was based on automated manufacturing gone feral, and had machine parts and building materials as lowest level. But it had to be made by another intelligence first and did not boot itself up from chaos. $\endgroup$ – JDługosz Jun 5 '16 at 10:26
  • 3
    $\begingroup$ I find difficult to understand what is being asked, since everything is formed by atoms. If you wanted an example of "life" formed by, say, screws and bolts, those screws and bolts would be formed of atoms too. In the same way, you could say that life "is based" on quarks, since there are no atoms without quarks. $\endgroup$ – SJuan76 Jun 6 '16 at 8:34
  • 1
    $\begingroup$ When I use a screw, I don't care about the individual atoms, only about the emergent properties of the material and its shape. Chemists don't solve functions modeling quarks; they just use (approximate) emergent rukes of chemistry. $\endgroup$ – JDługosz Jun 6 '16 at 8:41
  • 1
    $\begingroup$ Oh, then human beings is a solution, because organic chemists do not care about atoms, either (organic chemistry is based on molecules containing carbon). DNA, glucids, lipids, aminoacids and so on... And of course, many scientist are not worried about those but by upper functions (transcrition of DNA, how an organ or system works) and to those molecules are usually no more important than atoms. $\endgroup$ – SJuan76 Jun 6 '16 at 9:59

For life to be macroscopic and not also microscopic, there needs to be a reason why microscopic life is not effective. "Nature finds a way," so if there's value to be had at the microscopic level, it tends to find ways to miniaturize.

A macroscopic form of life runs into trouble with "attention." It needs to protect its resources, which means it needs to pay attention to them. If a parasite is small enough that it can sneak through the sensor grid and start stealing resources, that parasite is going to flourish. This means the macroscopic life needs more powerful ways of dealing with these smaller parasites. The solution which we see most in life is smaller sensors. We humans cannot see bacteria, which would eat us alive in days if we weren't stopping them. However, we have our own microscopic immune system which is always paying attention to everything at the microscopic level, protecting resources.

This is an arms race, encouraging creatures to get smaller and smaller if they can. Development of a molecular level way of thinking could eventually become a necessity due to this arms race.

The alternative to an arms race is to rely on macroscopic techniques, like santization. If you can make an environment inhospitable for life, you can protect your resources. We have no antibodies or white blood cells protecting a scalpel that's about to be used for surgery. However, we have an autoclave! We can sanitize things before putting them in our bodies!

However, sanitization takes energy. Lots of energy. You don't actually know how much energy is needed, because you don't exactly know what bacteria or virii are on the scalpel. You have to saturate the area, making sure that, if there were bacteria, they will die.

This is a very wasteful use of energy. For surgical purposes, it is a necessity. However, for day to day use, it would doom us to consume more energy than we do already! Eventually, some creature may figure out how to do a molecular scale trick, and become more efficient. Over time, this trick may dissiminate out into the rest of life.

Thus we find that, while it is not a necessity that molecular scale life occur, it is the natural direction taken by many different paths evolution might take. If macroscopic life existed, it is very likely that it would start learning how to operate microscopically at the speed of evolution.


must be atom-scale are hard words. Like there is no possibility than atom-scale live. In the end it comes all to chance.

In the end for atom based life, you need to get the right atoms that interact together in an way so self reproducing process starts. (and of course the other traits that are needed for life) Eventually creating life (you have some probability x).

If you want life based on bigger parts, that parts need to interact with other bigger parts in a similar way like those atoms in the previous paragraph. Those parts themselves however are based on atoms. So not only you need to gather the right parts that eventually creating life (x) but also those parts must previously generated by putting together atoms. So additionally to your probability x for your life to be created form your building blocks, you also have the probability y that those blocks are created. Depending on the complexity those blocks need to archive the goal y could be much smaller then x.

This results in an even overall lower probability (x * y < x).

So while I wouldn't say it is impossible, it is so unlikely that saying impossible wouldn't be wrong.

  • $\begingroup$ Minerals exist. Things like grains of sand exist. $\endgroup$ – JDługosz Jun 5 '16 at 10:18
  • $\begingroup$ Well yes. But as I understood your questions those Minerals that makes the life may not interact on molecular level, but something larger. And those minerals will need special traits to interact on such an level with other minerals. Maybe magnetism? What I meant was the chance, that minerals that have such specific traits emerge and in addition form life, are far lower then molecules to form something similar. $\endgroup$ – lokimidgard Jun 5 '16 at 11:23

Yes, Such life does exist Computer Viruses. They self replicate and their fundamental units are logic gates and processor cycles. Some research equivalents even evolve through selection and mutation. http://link.springer.com/chapter/10.1007%2F978-3-540-85984-0_113

  • 1
    $\begingroup$ But would they have emerged by themselves? My question is specific to natural abiogenesis, not just systems that continue to maintain themselves and replicate once built. $\endgroup$ – JDługosz Jun 5 '16 at 15:04

It seems to me, that chances are heavily against the emergence of macro-based life. The old Miller-Urey experiment shows that at least some of the most important amino acids can 'spontaneously' come to existence from non-organic compounds in a 'soup'. But there is no way, that molecules in such a soup could form a macroscopic part (like a gear) from themselves.

Therefore I suggest to use micro-based life as 'catalyst' or 'precursor'. The simplest way to do this would be intelligent design by the micro-based beings, like in my previous, quite inadequate answer. But there is a more evolutionary-style possibility:

Micro based life emerges, and flourishes on the planet in great variety. But then something happens, which exposes the planet to steady radiation (the soil is not contaminated by radioisotopes, the radiation probably comes from cosmic sources). Micro-based life sucks, since even a few lingering high-energy electrons or gamma-photons are capable to break its fine molecular structures.(for example mess up their DNA-equivalent, killing the cells or causing cancer.)

As the life resettles in the irradiated areas, the beings start to develop thick, metallic shells. These grow from inside, and protect the tissues from the radiation. But it renders the first Shellers almost immobile. As evolution goes on, the mechanichs of the shell becomes more complex. It allows the muscles to move the rigid external legs from the inside.

The classic photosynthesis requires the pigments to be exposed to direct sunlight. Under the new circumstances, this equals destroying them. More robust, macroscopic systems develop to harness energy from sunlight and radiation: the 'plants' grow solar panels, direct converters and electrothermal generators, while their whole metabolism slowly changes, to make use of their electricity.

After some millennia, the first macro-nerves appear: inorganic conductive lines in the legs, capable of transmitting basic information to the vulnerable brain from the outside. They are crude when compared to subtle micro-nerves, but instead of killing them, the radiation only causes some noise in their signals.

As the biomass of their food is in high percent made up from macrostructure shells, the digestion systems of the heterotrophic beings is in trouble. It is optimized for processing organic molecules, not for large pieces of shell. Crushing-jaws and smelter-stomaches emerge as they develop organs optimized for reshaping the ingested macroparts into useful shell-pieces.

At last, electromechanical systems start to substitute muscles. Initially, they are pretty clumsy, but there are plenty of electricity, and their usage allows significant reduction of micro-biomass, further reducing cancer risk.

It is hard to imagine that macroisation slowly takes over reproduction, but the evolution is capable of interesting tricks. Maybe, for example, the eggshell modifies into the macro-body of the offspring, premade by the mothers macro-manufacturer organs, and later continued by the hatchling.


I would say, "yes and no".

Life as you describe would be much more complex than our kind of life. As such, I think it could only emerge randomically on enormously longer time scales, so long actually that the entropy would make short work of the individual, not-yet-"living" components.

Before you can build life on simple nanomechanisms, there must be a way for lots and lots of those nanomechanisms to form and be available.

Hence, my "No".

But your kind of life - DNA replicators - has arisen from more rough and unstable RNA replicators, and there's evidence that those might have arisen from simple repeating patterns in clays or other suitable materials. Yet, many do not consider simple replicators as being life.

So even if biological life has arisen from chemical replicators, it is said that life arose from the absence of life, and by "life" you mean "Earth-type life".

If we defined chemical replicators as life, we would need to say that: - Earth life did not arise from the absence of life, but from earlier life. - it was that earlier, simpler life that did arise from the absence of life.

So my "yes" stems from the possibility that we consider biological life as a necessary stage for the rise of nonbiological life:

  • simple random electrochemical processes
  • electrochemical self-repeating patterns
  • simple chemical replicators
  • complex chemical replicators
  • nonbiological life

Each stage (except for the first that's ubiquitous given the appropriate environment) cannot arise by itself, but is brought forward by the previous stage.


In a sense, we are life that is based on single units that are larger than atoms and molecules. The units are called cells. This sort of life emerged very suddenly around 600M years ago after two billion years of green and grey slimes.

Yes it would be perverse to deny that monocellular organisms are also alive. But therein lies the difficulty. The fundamental building blocks are atoms and simple molecules. That's where evolution has to start.

However, there is also the possibility of "scaffolding": an intermediate step which can go extinct after some more evolutionary fit form has evolved from it. It's an open question as to whether there was some form of non-cellular life before cells emerged and wiped it out. There are interesting parallels between clay minerals and RNA. But there is no longer any clay-based life to be found on Earth.

So similarly, imagine some intelligent beings evolving from "animals" and discovering themselves and all other "life" to be made out of what we might call small voraciously self-reproducing nanotechnology. Will they ever work out that a good few hundred million years previous, the human race wiped itself out by developing self-reproducing nanotechnology with insufficient restraint and control mechanisms? (An SF grey goo catastrophe). Or, that their once lifeless planet was seeded with nanotechnology by an interstellar probe launched by far-sighted aliens or post-humans?


Question is: what do you consider a living object? There was a conversation about that in Monday begins on Saturday by Arkadi and Boris Strugatsky:

On the table next to the tub appeared a revolting creature resembling both a hedgehog and a spider. Eddie raised himself up and looked at the table.

"Ah," he said, and lay down again. "That's not life. That's un-life. Isn't Koschei the Undead nonprotein life?"

"What more do you want?" asked Korneev. "Does it move? It moves. Does it eat? It eats. It can reproduce, too. Would you like it to reproduce right now?"

If these conditions are all you want then big enough colony of robots with decent AI, where each unit is part of bigger "body", would be a "living" thing.

  • $\begingroup$ Certainly, as noted in comments to the OP. But I don't see how a robot factory could come into existence de novo. I don't understand your quotation. $\endgroup$ – JDługosz Jun 6 '16 at 15:21
  • $\begingroup$ @JDługosz Well, you may consider terrestrial life as a pretty complex robots as well. As for quotation I ask again: what qualities should object have to be a living thing? An ability to consume & reproduce? Intelligence? Adaptivity? Something else? $\endgroup$ – Mr Scapegrace Jun 6 '16 at 15:29

It depends on how you define "life" and "organism". You are a collection of cells, each of them are alive. They have basically the same DNA, but even within your body they've differentiated a little bit. Each of your cells contains a collection of organisms, mitochondria, which have their own completely different DNA. A good video on this subject is Kurzgesagt's What Are You?

You're alive. You're also a cooperative collection of things which are alive. Those things contain other things which are alive. What if you expand this up instead of down?

Is a city alive? It respirates. It grows. It regulates its internal environment. It changes with the external environment (evolves). It can even send out people to found a new city (reproduction). It's made up of other things which are also alive, but working cooperatively for their collective benefit (whether they know it or not).

Does the life on a planet constitute an organism? The Gaia hypothesis. Like a plant in a too small pot, it grows to fill every niche. It respirates. It changes with the environment, but it also alters that environment to better suit itself. And, if panspermia is to be believed, it can reproduce.

This probably wasn't what you had in mind. You asked about abiogenesis, and what I've described all requires existing life. I bring this up to suggest a new place to look for macroscopic life. Many of the building blocks of life, including amino acids, form spontaneously. Point is, what if life, macroscopic life in the expanded definition, could also appear from spontaneously forming macroscopic structures dominated by macroscopic forces like gravity and electromagnetism?

  • $\begingroup$ I don't follow what you're referring to in the last sentence. $\endgroup$ – JDługosz Jun 9 '16 at 1:02

Macroscopic life can exist.

Consider a robotized city. In the core there is a big factory: smelters and chemical plants, machining and circuit-printing..., it can produce any parts needed for the city. A lot of workbee robots are lingering around. They mine resources, carry it to the main factory, and use the parts fabricated in the main factory to assemble new workbees and repair the factory. The city can grow and heal.

If enough resources were stockpiled, the factory shifts to production of its own parts. As they are produced, some workbees take these parts, and carry them away. They assemble a new factory, and start to service it. reproduction is achieved.

This 'lifeform' is analogous to monocellular life, since it has no subcomponents capable for reproduction. But if many robotic cities forge an alliance, and start to specialize for different tasks, something like multicellular life is possible. (a robotic empire from human perspective.)

But there is a problem. There is no way for such system to emerge naturally. How could this all come to existence? You would probably need some microscopic-manipulation-based classical organisms to design and build the first factory.

  • $\begingroup$ From the Question: “I'm specifically asking about the possibility of abiogenesis, not whether something could be constructed to have these features.” I'm asking about the emergence of life, not the definition. $\endgroup$ – JDługosz Mar 26 '17 at 8:32
  • $\begingroup$ @JDługosz OOps. I have overlooked this sentence. But you can view thing as multistage abiogenesis: first microlife emerges, reaches sentience, and creates macrolife. After that, it goes extinct. $\endgroup$ – b.Lorenz Mar 26 '17 at 10:22

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.