Big Picture Part Four: Complexity

One of a series of quick posts on the six sections of my book The Big PictureCosmos, Understanding, Essence, Complexity, Thinking, Caring.

Chapters in Part Four, Complexity:

  • 28. The Universe in a Cup of Coffee
  • 29. Light and Life
  • 30. Funneling Energy
  • 31. Spontaneous Organization
  • 32. The Origin and Purpose of Life
  • 33. Evolution’s Bootstraps
  • 34. Searching Through the Landscape
  • 35. Emergent Purpose
  • 36. Are We the Point?

One of the most annoying arguments a scientist can hear is that “evolution (or the origin of life) violates the Second Law of Thermodynamics.” The idea is basically that the Second Law says things become more disorganized over time, but the appearance of life represents increased organization, so what do you have to say about that, Dr. Smarty-Pants?

This is a very bad argument, since the Second Law only says that entropy increases in closed systems, not open ones. (Otherwise refrigerators would be impossible, since the entropy of a can of Diet Coke goes down when you cool it.) The Earth’s biosphere is obviously an open system — we get low-entropy photons from the Sun, and radiate high-entropy photons back to the universe — so there is manifestly no contradiction between the Second Law and the appearance of complex structures.

As right and true as that response is, it doesn’t quite address the question of why complex structures actually do come into being. Sure, they can come into being without violating the Second Law, but that doesn’t quite explain why they actually do. In Complexity, the fourth part of The Big Picture, I talk about why it’s very natural for such a thing to happen. This covers the evolution of complexity in general, as well as specific questions about the origin of life and Darwinian natural selection. When it comes to abiogenesis, there’s a lot we don’t know, but good reason to be optimistic about near-term progress.

In 2000, Gretchen Früh-Green, on a ship in the mid-Atlantic Ocean as part of an expedition led by marine geologist Deborah Kelley, stumbled across a collection of ghostly white towers in the video feed from a robotic camera near the ocean floor deep below. Fortunately they had with them a submersible vessel named Alvin, and Kelley set out to explore the structure up close. Further investigation showed that it was just the kind of alkaline vent formation that Russell had anticipated. Two thousand miles east of South Carolina, not far from the Mid-Atlantic Ridge, the Lost City hydrothermal vent field is at least 30,000 years old, and may be just the first known example of a very common type of geological formation. There’s a lot we don’t know about the ocean floor.

Lost City

The chemistry in vents like those at Lost City is rich, and driven by the sort of gradients that could reasonably prefigure life’s metabolic pathways. Reactions familiar from laboratory experiments have been able to produce a number of amino acids, sugars, and other compounds that are needed to ultimately assemble RNA. In the minds of the metabolism-first contingent, the power source provided by disequilibria must come first; the chemistry leading to life will eventually piggyback upon it.

Albert Szent-Györgyi, a Hungarian physiologist who won the Nobel Prize in 1937 for the discovery of Vitamin C, once offered the opinion that “Life is nothing but an electron looking for a place to rest.” That’s a good summary of the metabolism-first view. There is free energy locked up in certain chemical configurations, and life is one way it can be released. One compelling aspect of the picture is that it’s not simply working backwards from “we know there’s life, how did it start?” Instead, its suggesting that life is the solution to a problem: “we have some free energy, how do we liberate it?”

Planetary scientists have speculated that hydrothermal vents similar to Lost City, might be abundant on Jupiter’s moon Europa or Saturn’s moon Enceladus. Future exploration of the Solar System might be able to put this picture to a different kind of test.

A tricky part of this discussion is figuring out when it’s okay to say that a certain naturally-evolved organism or characteristic has a “purpose.” Evolution itself has no purpose, but according to poetic naturalism it’s perfectly okay to ascribe purposes to specific things or processes, as long as that kind of description actually provides a useful way of talking about the higher-level emergent behavior.

This entry was posted in Big Picture. Bookmark the permalink.

14 Responses to Big Picture Part Four: Complexity

  1. John B says:

    I wouldn’t think that the laws of thermodynamics should hold any sway in quantum mechanics, because it has already been shown to be violated in random particle anti-particle creation and annihilation, which results in a photon from nothing. Then all life is carbon based, and life is created from complex structures of DNA (an emergent property of atoms). If one of these laws can’t hold up, then we shouldn’t expect any of them too. It is really the last piece of classical physics which has stayed in modern physics.

    Also, everything the human race strives for is fighting the second law. People get paid to do it every day. We even build machines to do it for us. Just about every human luxury we have is based on someone acting against the second law in some way, shape, or form. We have adapted to fight against the second law as a major part of our own survival.

  2. Torbjörn Larsson says:

    This is very good! You describe the strongest, if not consensus, theory of life emergence, and the thermodynamic context is not scrambled. The vent theory is strongest because it can be readily tested, and indeed Russell et al describes ~ 20 phylogenetic frozen in traits that we have inherited from Hadean alkaline hydrothermal vents. [“The Drive to Life on Wet and Icy Worlds”: since that review was written it has turned out that the UCA lineage chemiosmosis could only evolve at the ocean/vent interface of alkaline hydrothermal vents, for reasons of ion gradients vs ion transporter and semipermeable membrane evolution.]

    Soup theory on the other hand has a few environmental traits such as vesicles and their ability to divide.

    I prefer soup vs vent theory because neither can be cleanly classified as genes vs metabolism first, even though the general pathway Russell describe is metabolism first.

    “In the minds of the metabolism-first contingent, the power source provided by disequilibria must come first; the chemistry leading to life will eventually piggyback upon it.”

    Of course you must have a disequilibrium of one sort or other, but I note that the metabolic like gluconeogenesis/glycolysis/pentose pathways of Keller et al that can happen in the Hadean ocean/vent interface is more equilibrium reactions as chemists classify them.

  3. Torbjörn Larsson says:

    @John: This nitpick is inconsequential, but the SIWOTI was strong in your comment.

    – Thermodynamics is not violated by virtual particle pairs creation/annihilation since it obeys an uncertainty relation, and no resulting particles are left. (Unless you put in energy.)

    – “Then all life is carbon based”. A leap?

    – “life is created from complex structures of DNA (an emergent property of atoms)”.

    Modern cells are controlled by DNA genomes, but much simpler RNA cells evolved first. (The evidence is at the core of the DNA genetic machinery, which is RNA based. That discovery gave a Nobel Prize.)

    Atoms do not spontaneously arrange themselves in DNA spirals, it is a molecular machinery that evolved.

  4. James Cross says:

    Torbjörn Larsson

    Yeah, I like the vent theory too. I think it is best we have at the moment.

    However, there is still a huge gap from generating amino acids, sugars, and nucleotides to having a viable organism. Even the simplest organism known today has a 120 genes and it is dependent on another organism to survive. If the abiogenesis theory as science currently approaches it is correct, then we should find organisms much simpler than 120 genes or at least be able to artificially create them. I know people are working on that so it might happen. If we go down the metabolism first approach we still need to get reproduction at some point or we do not have life and particularly do not have mechanism to drive evolution.

    Until we fill in a few more gaps I would defer on claims that we understand the origin of life but I don’t think that means we need to invoke God or the supernatural.

  5. John B says:

    Last time I checked, anti-matter/matter collisions produce photons. There is no reason why it should be otherwise in any special circumstance, unless you are just trying to cling to the last bastion of classical mechanics…

  6. Charlie says:

    John B.,

    I’ll let someone else deal with the quantum mechanics, but your statements above about life, human activity, and the second law of thermodynamics are entirely mistaken. There is a huge amount of energy coming from the sun. Low entropy photons hit earth, higher entropy photos radiate away. Even in dark places like thermal vents you have chemical gradients to drive metabolism. Life and human activity are increasing entropy, not reducing it. The chemistry of life requires this, as do things like human power plants, your refrigerator, etc. These processes are driven by the second law, not violations of it.

    This is a common misunderstanding. (And sadly, an intentional misunderstanding promoted by some to mislead others for reasons that are hard to fathom. I cannot tell from your post if you are a victim of this misinformation or one who knowingly perpetuates it.)

  7. Torbjörn Larsson says:


    I fail to see the relevance to what I said. Matter/anti-matter collisions obey thermodynamics, ask Einstein. (E = mc^2 is the tip off.)

  8. Torbjörn Larsson says:

    James Cross

    Seems all the great minds think alike! [/shameless self promotion]

    I am going to fudge over the number of genes question in the UCA lineage, but I note that simplified parasites like viroids and viruses get away with 0 (viroids) to 3 (viruses) genes in some cases.

    “generating amino acids, sugars, and nucleotides”.

    Vents wouldn’t be producing amino acids, but they are efficient to adsorb and concentrate organic molecules, circumventing the dilution problem of soups.

    Keller et al show that they should have been producing pentose in their outer pores. I am miffed that they haven’t tried to see if purines – who are metabolic products of pentose and amino acids – also would be produced non-enzymatically in Hadean ocean vents.

    But as soon as we have purine base pairs, we are done. There has been recent work that shows RNA strands are produced to a length that is set by individual pore geometry when vents grow. (Earlier work used huge temperature differences to show strand elongation, but now there is no need.) They are the unique systems that do so, soups tend to shrink strand length.

    The resulting strands can replicate non-enzymatically under thermal cycling, and will sort into ‘stores’ of gels of ‘like’ types, as ordered by their functional purine sequence.

    An important advance is that this will happen in racemic conditions, and that resulting chiral sorted ribozyme replicases are more efficient, shorter and faster, when they replicate the opposite chirality. That could have started the transformation of the inorganic vent PCR reactor into individual pore ‘cells’ with increasingly non-thermal genetic machinery, i.e. transformed the developing reactor into darwinian small survivable step evolutionary organisms.

    Ribosome and protein fold evolution show the remaining UCA lineage evolution. The ribosome started out as a simple ‘knicked’ ribozyme strand that could put amino acid monomers on other ribozymes, That would have increased their functionality. Since that bond is temporary, aa monomers are recycled and by chance sometimes a dipeptide would be formed. A product funnel evolved as the random polypeptides grow into 5-6 mer long strands, in which case that product became more important for fitness. They form ‘nests’ that can bind catalytic metal atoms, a start for the proteins.

    Eventually a supporting proto-mRNA strand would support and eventually chose what has now become tRNA aa monomer ‘guide’. When the number of guide sites hit first 2, then 3, the genetic code can evolve. It can only do so if there is a chiral choice, and there are still two frozen in such in today’s genetic machinery that needs none since it is now a metabolic choice. The tRNA and the rRNA binding sites are chiral selective.

    This is when the RNA/RNA non-chiral, still environmentally metabolic reliant cell lineage transform into a RNA/protein chiral, “in house” metabolic cell lineage. Early protein folds were multifunctional and not very faithful, but combined in a haphazard manner that leapfrogged evolutionary bottlenecks, especially as the ancestral ribosome became quickly elaborated.

    Remaining identifiable steps is DNA evolution, from which 3 major viral lineages stem as seen by their replicases. The DNA strand ‘loop’ was first used as a temporary genetic memory, but would have consolidated the RNA strands that in some cases were still used as ribozymes. Here is when the number of genes increase, both because they can and because there is a demand.

    Also remaining is evolution of chemiosmosis and a modern type organic semipermeable membrane with transporter proteins in pores. That could only happen on the alkaline hydrothermal vent surface for reasons of ion flows and pH differentials, and was essential for purely organic cell growth outside the vent. Chemiosmosis is also what enables cells to adapt peculiar metabolisms to the environment, that doesn’t waste too much energy while growing et cetera.

    Oh, and subsequent possible evolution of the push-pull filament extrusion system that would be enabling cell movement outside the vent. It roots both bacteria flagella and archaea archaella before they split and ended the UCA lineage. (In my mind I see cells that go that-way and those that go this-way, and as they can’t agree they evolve very different genetic and membrane systems. =D)

    Now, nobody has shown the individual steps in an integrated fashion, and of course such a lengthy theory confabulated by my opinion alone without detailed testing may at best be morally correct. But testing should be doable, and individual shorter evolutionary pathways should be essentially correct if the underlying work is.

    [References omitted for the time being, but can be hunted up if asked. Yikes, that would be a lot of references! ]

  9. John B says:


    Einstein’s opinions about quantum mechanics where his biggest failure. He fought it most of the way through its development, because it did not agree with classical mechanics. God plays dice with the universe, and the moon isn’t there when you are not looking at it.

    The problem here is that quantum mechanics breaks a lot of classical laws, but science doesn’t change those laws to fit what quantum mechanics says. We still have Newton’s Laws even though quantum mechanics may violate them, so we still have the Laws of Thermodynamics even though quantum mechanics violates the first law. The big difference here would be that we do not still use Newton’s Laws in quantum mechanics…

  10. John B says:

    What has thermodynamics gotten us here lately anyways? I will tell you. The holographic principal that says an object falling into a black hole will stop falling from one frame of reference and continue to fall in another frame of reference at the same time (creating a relative velocity paradox). That is not consistent with the theory of relativity. An object cannot have two different relative velocities from each frame in the theory of relativity. It would create a circular argument which says that they would then have different amounts of time dilation and length contraction based on their differing relative velocities from each frame. Then the theory of the holographic universe in black hole physics has to be wrong, which uses the laws of thermodynamics to come to this conclusion.

  11. Torbjörn Larsson says:

    John, you persist in pushing erroneous claims that a little checking would perhaps make you abstain from.

    – “Einstein’s opinions about quantum mechanics where his biggest failure. He fought it”.

    Einstein considered his cosmology/cosmological constant prevarications his greatest failure. And he was a pivotal figure in discovering (photelectric effect) and developing quantum physics and its usefulness (laser model, et cetera).

    – “the moon isn’t there when you are not looking at it.”

    Of course it is, you are misunderstanding the role of observers.

    – “quantum mechanics breaks a lot of classical laws,”

    Such as? Thermodynamics is more fundamental and remains unbroken, say.

    – “quantum mechanics violates the first law.”

    The 1st LOT states that when energy flows into or out from a system, its internal energy changes in accord with the law of conservation of energy.

    How would QP violate that? According to the correspondence principle it can’t, and the Schroedinger equation is explicit in that it describes wave functions belonging to systems of fixed energy (energy conservation).

    … and I could go on.

    Rather I will end with the observation that life, which relies mostly on classical behavior anyway, obeys thermodynamics to the point that you can’t understand its emergence or its metabolic opportunism (‘life is a way to find an electron sink’) without it.

    I think you somehow wanted to ‘magify’ life, but got lost in your script somewhere. Well, you can’t describe life in such terms anyway.

  12. darrelle says:

    There was that interesting math paper a couple of years ago or so that suggested that given an open system, i.e. an outside source of energy, that chemical systems will naturally tend to evolve towards living systems because living systems are more efficient at increasing entropy than non living systems.

  13. James Rose says:

    Hello, Sean, sorry for coming ‘late to this party’, but I just found your blog, via your Science Friday interview May 20, 2016 where you talked about “The Big Picture”. I’m delighted you’ve tackled the ‘broad umbrella of the universe’ notion Sean, but am a bit bemused (albeit I’m behind the curve on getting a copy of the book to read – will rectify that soon). I didn’t catch the whole interview, but I would have liked to have heard mention of George Sarton, Ludwig von Bertalaffy, the General System Theory movement (circa 1950 to present) and similar writers/theory groups. Some of us in General System have identified that there might be a next-Copernican revolution needed in order to accomplish a GUT – grand unification theory – that goes beyond the physics notion, to indeed capture~explain the smooth tiering evolving of complexity levels (against the gradient of entropy) – that is more subtle and complete than the Howard Odum ’emergy model’ of ‘solar energy source’ drivers. Do you have any thoughts or notions about these topics? And, some corollary questions, if you might comment on them: ‘entropy’ is so important and ubiquitous . . . Is it a derived phenomena coming “out of” the fundamental forces? Or, is it derivable from the geometric topology of dimensions generally? (meaning . . . could it be a relations-set that precedes the fundamental forces? And, could there be an improved understanding of it that goes beyond Boltzmann, to a more generalized model~definition (eg, Shannon~Weaver model ‘information entropy’, which is a related but different phenomena altogether.) ? There are flocks of diverse uses~redefinings of ‘entropy’. * * * with apologies, I’m not terribly practiced in ‘blogging’. Look forward to thoughts/remarks. Thanks, James

  14. Bernd Knipperdolling says:

    It requires more effort than scientists normally can be bothered to invest in order refute creationist’s claims, as the thermodynamic argument is actually better than it looks.

    When the entropy in an open system decreases, it gathers some of the information from outside of the system. If for example a vessel full of lukewarm water is heated on one end (say the upper one) and cooled down on the other end (the lower end), the entropy becomes lower, although the average temperature might stay the same. But the water in the upper bit will be hotter while in lower bit will be cooler. What the water will not do is to become hotter on left side and cooler on the right side. Or take some spread out metal cuttings. If a strong magnet is placed close to that system, the cuttings will adjust themselves in line with the magnetic field. They reproduce the information given by the magnet. But the cuttings will (most likely) not form the word “Inshallah” for example. The information to form this word is not given by the outside environment of our system.

    Other examples like in the first linked article ( ) refer to complexity created by humans. Take a person building a house from nails and a pile of logs of wood. Surely the entropy of the pile of wood decreases while it forms a house. But again, all the information required is already present in the human species and its knowledge and civilization. The low entropy of the environment (civilization as a consequence of evolution) passes on information to the system in question (pile of logs) and decreases entropy in that system in accordance with the second law.

    What has to be explained though is how all the information required to enable evolution, humanity, civilization and knowledge was gathered in the first place. This is how the main argument from the linked article ( also is flawed, as it simply compares the absolute values of increase or decrease of entropy while missing out what sort of information it is all about in either case.

    Our solar system in a whole is an open system only in a strictly formal sense. Practically it is almost closed, except for some radiation escaping the system and some light of distant stars entering it. If we want to explain evolution in accordance with the second law we would have to assume at least all the information from the human genome (actually a tiny part of the whole complexity involved, since a single string of human DNA can’t survive on its own) is somehow encoded in the flickering of the stars in the sky. Well, as long as this information has not been encoded within the system in the first place.

    So Creationists may very well be right about evolution breaching the second law. However they are wrong in the conclusions they draw. Thermodynamics is “only” statistics. Events breaking the second law are not illegal but simply very rare. Like winning a lottery is a rare event. It is extremely unlikely that I will win a lottery next week. But still someone somewhere wins a lottery every day. In a similar way the a priori chance of a successful evolution on earth may have been extremely low. But that’s not the point. Had evolution been successful on alpha centaury then we would just sit over there wondering about why we are here.

    The only point creationist could make is to claim that the chances for evolution are so vastly low compared with the size of the universe that it would be unlikely to be successful even once at all. But the size of the universe is unknown. It must at least be much bigger than the visible part. It may well be infinitely big. There are also good reasons to assume that our universe is only one part of a vastly bigger multiverse. The burden of proof is on the creationists to show evidence for a sufficiently small universe if they want to prove that evolution is unlikely. This evidence does not exist.