Posted by: dacalu | 22 August 2019

The Origins and Probability of Life

The Origins and Probability of Life

My friends Peter Jarrett Schell and John Henry recently posted on Facebook asking me about the origin and probability of life. It’s not a simple question, but I know many people will be interested. I recommend reading up on astrobiology, my book Life in Space, Astrobiology for Everyone (Harvard, 2009) or the more recent Astrobiology: A Very Short Introduction by David Catling (Oxford, 2014). If you’re looking for something more technical, check out the “Astrobiology Primer” (ver. 1, 2006; ver. 2, 2014; ver. 3 in progress). It is a rich and growing field. For those interested in a shorter taste, here are my present thoughts on the earliest life, its timing and chemistry, and what that can tell us about the probability of life. 

Life from Non-Life (Abiogenesis)

Logically, we have two options:

  1. There has always been life in the universe, or
  2. Life arose from non-life

Because I cannot imagine life, at least life remotely similar to Earth life, existing without matter, I think B must be true. There was no life shortly after the Big Bang; there is life now; therefore, life must have arisen from non-life. The simplest explanation available to natural science is that it arose through chemical interactions on Earth. (If you’re interested in the bounds of natural science, see note 1.)

Fossil stromatolites – layered bacterial communities – provide evidence for life at least 3.2 Bya (billion years ago) and probably as early at 3.5 Bya. Chemical fossils – isotope ratios impossible with known abiological processes – have been found at 3.8 Bya and suggested as early as 4.4 Bya. The earliest eon of Earth history is called the Hadean, for Hell-like conditions – high temperature, high radiation from the Sun, heavy meteor bombardment, and volcanoes. No rocks have been found from before 4.0 Bya that have not been melted and reformed. Given the chaos of early Earth, there is a general consensus that life arose on Earth fairly quickly once conditions were good. Most origin of life research is focused on early Earth environments – 4.0-3.6 Bya.

Life from Space (Panspermia)

A few scientists have proposed that life arose on Mars (or somewhere else) and travelled here. We know that meteor impacts can eject material from Mars in a way that eventually brings it to Earth. We know that Earth organisms could survive the ejection process. We do not know of any organisms that could survive the journey, but it is not inconceivable. Having said that, I favor simpler explanations. Given a choice between explaining abiogenesis alone (on Earth) or abiogenesis (on Mars) plus space travel (to Earth), I’ll stick with the former.

What is Life?

Dating the origin of life requires an understanding of what life is. All known life depends on four critical features: replicators (always DNA or RNA genes), catalysts and signal molecules (predominantly amino acid proteins, occasionally nucleotides and other organic molecules), cells (almost always phospholipid bilayers a.k.a. cell membranes), and metabolic networks (a surprisingly small number of organic pathways). We have not observed – and have difficulty imagining – life without all four.

Viruses draw attention because they appear, at first glance, to be lonely replicators. They cannot, however, complete their lifecycle without catalysts, cells, and metabolic networks. Whether or not we consider them alive, they depend on the whole suite of life-features.

In the 1980s, 1990s, and 2000s, many subscribed to the RNA-World hypothesis. RNA molecules can act as both genes and catalysts, making them, potentially, a simpler form of life. Biochemists had high confidence that a ribozyme “autocatalysts” could be created in the lab – a molecule capable of copying itself. If such a molecule did exist, then it would provide modern evidence that the first life could have been RNA-only.

RNA-World research has produced some amazing results. It has revealed fascinating details about metabolism and “chemical evolution.” It has not produced a population of molecules undergoing open-ended Darwinian evolution. Some still have hope that autocatalysts will be found. Many have moved to parallel avenues of research.

Current thinking in astrobiology is that we should pursue multiple research programs on the origin of replicators, the origin of catalysts, the origin of cells, and the origin of metabolic networks. We can learn about all four without committing to which came first, or even if they occurred sequentially. Perhaps they arose in parallel and merged. Similarly, we need not commit to which one, if any “defines” life. They are all interesting features of life as we know it all. They were all involved in the history of Earth life.

There Was No First Organism

In the last ten years, we have started to turn away from positing a first cell or a first organism. All four research programs suggest that life can only be meaningfully understood in the context of population. Replicators like genes can only evolve (undergo evolution by natural selection) through competition and cooperation. Catalysts like proteins can only shape their environment when concentrated and sequestered with reactants. Membrane bubbles must grow, fuse, and divide in order to have life-like properties. Metabolic networks require multiple realizations (perhaps at the level of 10^4) to provide stable complexity akin to even toy models of life. For these reasons, I advocate for speaking of the “first population.” I can’t say it is orthodoxy, but it is coming to dominate in astrobiology and origin-of-life discussions.

The Last Universal Common Ancestor (LUCA)

All is not lost. We know almost nothing about the first organism or first species. We can speak of the most recent ancestor of all known life. By reasoning backward from present organisms, we can imagine a common ancestor of life or, more realistically, a single population that gave rise to modern species. That population would have had all (or at least most) features common to extant life: including a DNA, RNA, proteins, cells, and common metabolic pathways (such as the TCA cycle).

I think common descent is usually a better explanation than convergence. I suspect that global trends toward cooler temperatures, lower radiation, and more available oxygen did produce some convergence (e.g., symbiosis). There are also some interesting possibilities for convergent adaptations to denser population (e.g., increase in maximal body size at the order level, increase in genome complexity). It is, of course, a historical question, so it’s hard to say how much proof or what kind of proof should convince us of what actually happened (Note 2).

Messy Details

Two factors complicate matters: horizontal transfer and coalescence. Discussion of LUCA gets complicated by the messy, multi-level, recombining networks of descent. There is likely a species tree that connects all current species trees through lines of decent. At the level of species, branches rarely recombine. Species can fuse; individuals can cross fertilize; and genes can jump from one to another; but not that often. I think of the species tree as made up of hollow pipes. Within those pipes, individuals reproduce to form their own organism level trees. The organism trees branch and flourish like vines within the population tree pipes. One family might flourish for thousands of year only to fail while a slender lineage that lived on the margins suddenly takes over the population. To make things even worse, the individual trees are also pipes, with gene trees growing inside them. Genes can replicate, diversify, and compete within individual organisms. Horizontal transfer describes times when trees don’t look like trees, when individuals break out of their species level pipes, or genes break out of their organism level pipes and break back in somewhere else. “Coalescent theory” provides math for tracking the last common ancestor at any given level.

Unfortunately, the LCA may differ between different genes, different organisms, and different species.  For example, “mitochondrial Eve” describes the LCA of all human mitochondria, transmitted from mothers to offspring. A single mitochondrion, roughly 150,000 years ago, gave rise to all modern mitochondria (in humans). The descendants of other mitochondria around at that time have all died off (though they could have hung around until very recently). So, one woman can be said to be the mother of all modern humans, at least as far as mitochondria are concerned. Meanwhile “Y-chromosomal Adam” describes the LCA of all human Y-chromosomes, transmitted from fathers to sons. A single chromosome, more than 200,000 years ago, gave rise to all modern Y-chromosomes. So, one man can be said to be the father of all modern humans, as far as Y-chromosomes are concerned. But this Eve and this Adam were 50,000 years apart.

In the end, the LUCA population is really just a narrow ring drawn around a mess of pipes and vines in the distant past. We know many things about the genes, organisms, and populations within the ring, but we don’t know how they related to one another. We believe that there were genes, organisms, and populations outside the ring as well. They may have been very successful at the time. None of their descendants remain today.

Shadow Life

In this way, LUCA is much like the observable universe. There may be stuff outside, but the expansion of stuff inside prevents us from knowing. Life as we know it has outcompeted other forms of life. My personal belief – and I think it is common among astrobiologists and origins researchers – is that proto-life arises regularly on Earth, but known life has grown so good at survival that it eats alternative forms of life as soon as they arise. They never get a chance to get started.

Some biologists have proposed a “shadow biosphere” made up of organisms using alternate chemistries, unavailable or unappetizing to life as we know it. It doesn’t show up in our familiar environment, but may persist underwater or underground. Personally, I’m skeptical. It’s hard to get the energetics right without carbon-carbon bonds (“organic chemistry”) and known life finds carbon-carbon bonds tasty.  A shadow biosphere would need a highly evolved defense mechanism and would be in a constant arms race with known life for survival.

The Probability of Life Arising

This is one of my favorite questions and therefore one I like to see handled with rigor. Philosophy of probability can be contentious. We want probability to estimate the frequency of future events, but we don’t have access to future events. So, we can restrict it to the frequency of past events (frequentism), or call it subjective, or attempt a compromise.

The frequentist probability of life arising in the universe is 100%. It did. The frequentist probability of life arising on a planet that humans have visited is 100%. The presence of humans is the presence of life. The frequentist probability of life arising on a planet humans have visited independent of human presence is either 100% or 50% if you think we have studied Mars sufficiently. None of these “probabilities” really satisfies our curiosity. Astrobiologists call this the N=1 problem.

Some subjectivity will be involved. We should ask how much. We have information about the history of life on Earth and good evidence that Earth-like life is not abundant on any other body in the Solar System. Earth life adapts and spreads quickly, suggesting that, if it arises and has a congenial environment, it will take over. Earth is awash with life: a mile below the surface and a mile above, in the driest deserts (Antarctic and Atacama), in cold and heat and radiation. People disagree about how to quantify this subjectivity. For this reason, I prefer to talk about the plausibility or likelihood (probability given specific hypotheses) of life arising.

Evidence of life 3.9-4.0Bya suggests a high likelihood of life arising, if conditions are right. As I mentioned above, I do think shadow life, or precursors of shadow life, have arisen repeatedly in Earth’s history only to be consumed by life as we know it. On the other hand, we’ve had no luck making life in the lab, so it can’t be too likely. The barrenness of Mars and, requiring far more assumptions, the silence of interstellar space suggest a low likelihood. Given the ridiculous number of stars and planets, it seems plausible we are not alone. I can’t really say more than that.

Characteristics of Alien Life (Should it Exist)

I love this question and speculate some at the end of my book (Life in Space: Astrobiology for Everyone). We have learned a great deal in the last 10 years, but some things are basic chemistry; they remain the same. Molecules with carbon-carbon bonds (organic chemistry, not necessarily biochemistry) have really nice properties for life. This is basically the only way to have robust chain molecules with 4 strong bonds. Anything else will not have the same flexibility. Add to that the abundance of carbon in the universe, and it seems highly likely that any life-form will be carbon based (Note 3).

Similar arguments can be made for the environment of any form of life. Water has amazing and unique properties. It is liquid over an unusually broad range of temperatures. It is slightly polar, making it a good medium for many types of chemistry. Ice floats, creating a layer of insulation over lakes in cold temperatures. And, once, again, the atoms (hydrogen and oxygen) are very abundant.

A weaker case can be made for energy capture and storage. Visible light (380-740nm) turns out to be close to peak Solar output (Note 4). It carries almost enough energy (162-315 kJ/mol) for basic organic reactions (300-450 kJ/mol). If it were stronger it would dissolve organic molecules; if it were weaker, it could not power organic systems (Note 5). Sun-like stars, and Earth-life-like energy capture work well together.

I would also note that Earth life, at the most basic level operates by pumping protons (H+) across membranes. [This process creates osmotic potential, which drives ATP synthesis as protons pass back through the membrane. ATP acts as storable redox potential, a battery for life.] Given the abundance of hydrogen, I would be surprised if this were not a universal strategy for energy use.


Astrobiology can tell us a great deal about the characteristics of Earth life and make decent predictions about alien life, should it exist. Small sample size (n=1) means that our inferences are likely to be heavily biased when trying to make statements about life at a larger scale. We cannot know the extent of that bias until we find (or make) a second instance of life. For the moment, alien life seems highly likely, as long as we remember that this plausibility is largely driven by a philosophical belief that we are, in some meaningful way, “normal.” I look forward to finding out more and am excited about the growth of knowledge in coming decades.


Note 1: Caveat on natural science. My understanding of natural science is that it deals with natural explanations. Appeals to the supernatural, unnatural, or anything outside the universe – including God, Spirit, etc. – do not meet this requirement. Natural science often fails to provide answers we want. Currently it does not answer the question of how life arose. This has led some to look for answers beyond natural science. As a natural scientist, I cannot assess whether those answers are satisfactory. For a great defense of “methodological naturalism” see Robert Pennock’s article on “Naturalism, Evidence and Creationism” (1996, Biology and Philosophy 11(4):543-549).

As a whole person, I suspect I will still want the natural science explanation, in any case. I want to know how events proceeded within the bounds of nature and think it will be worth our time to pursue those types of explanations. To wit, whether or not God was involved, an interesting natural science question remains. When, where, and in what manner did life appear within our universe?

The metaphysical question of whether it could have occurred without unnatural intervention strikes me as poorly framed. It hangs on an equivocation between two uses of the word “natural” described by J.S. Mill in an 1885 essay on “Nature.”

Note 2: Caveat on epistemology. I want to be clear that sometimes we can say. Excellent work has been done on the evolution of the ribosome which provides good evidence for historical structures and common descent. Nonetheless, it is difficult to say a priori what will be convincing in the future.

Note 3: Note on “artificial life.” Many of my colleagues believe we will contact alien robots before we encounter organic alien intelligence. I have mixed feelings about this. Intelligence is tricky for a number of reasons. In any case, I do not think that silicon-based life could arise without the aid of organic life. Whether it may, one day, supplant or dominate organic life, inorganic abiogenesis strikes me as implausible. Silicon-silicon bonds are too weak.

Note 4: Note on Solar output. I have not looked at data in detail recently. A quick scan suggests that Solar radiation entering the lower atmosphere has a peak around 500 nm with roughly half the total radiation falling in the visible range. The spectrum falls steeply in the UV, but has a broad infrared tail.

Note 5: Note on Stellar frequency. Recently, astronomers have become excited about the prevalence of Red Dwarfs in the galaxy. They are far more common than Sun-like stars and many have rocky planets orbiting in the habitable zone, a region where water on the surface of planets would be liquid. They output sufficient heat to warm their planets, but I have not seen commentary on whether their weaker photons (peaking ~1000nm, 120 kJ/mol) would be effective for supporting organic chemistry.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s


%d bloggers like this: