Deep dive · Biology · Origin of life
Ten trillion years: the full case
The complete argument, with the numbers, the sources, and the strongest objections answered rather than avoided.
Life, set against the scale of the universe it appeared in
Section 01
The claim, and what it really says
In February 2017, NASA announced that the TRAPPIST-1 system, 39 light-years away, holds seven roughly Earth-sized planets. Commentators added a detail that sounded decisive: TRAPPIST-1 is an ultracool red dwarf, and stars that small burn so slowly that their main-sequence lifetime is measured in trillions of years. Ten trillion years, some said. Surely that is enough time for life to appear somewhere.
It is worth separating three claims that get folded into one:
- The star will last a very long time. This is well-supported stellar physics.
- Its planets will remain habitable for that long. This does not follow, and the evidence runs the other way. TRAPPIST-1 is a flare star, and the planets orbit close enough that repeated stellar outbursts are expected to strip atmospheres. Direct JWST observations of the inner planets have so far failed to detect any thick atmosphere at all.
- Enough time makes the origin of life likely. This is the claim this article is about — and it is not an astronomical claim. It is a probability claim, and it can be checked.
Deep time is a powerful intuition. We are used to long spans doing work that short spans cannot: canyons carved, mountains worn down, species spread across continents. So “ten trillion years” sounds like a solvent that dissolves any improbability. The question is whether the improbabilities involved in the origin of life are of a size that time can touch.
Section 02
What the universe can actually compute
Before running any calculation about life, it helps to know the ceiling. How many events of any kind can the observable universe have hosted?
In 2002, MIT physicist Seth Lloyd worked out the answer from first principles. Using the total energy available and the fundamental limits that quantum mechanics places on how fast a physical system can change state, he calculated the maximum number of elementary logical operations the observable universe could have performed since the Big Bang. The figure is about 10120.
That is not a number about biology. It is a hard physical bound on everything — every particle interaction, every collision, every state change, of any kind, anywhere, for the whole history of the cosmos.
| Quantity | Approximate size |
|---|---|
| Seconds since the Big Bang | 1017 |
| Seconds in ten trillion years | 3 × 1020 |
| Atoms in the observable universe | 1080 |
| Elementary particles in the observable universe | 1080 |
| Maximum logical operations since the Big Bang (Lloyd) | 10120 |
| Read the bottom row as the hard ceiling: nothing in this universe has had more chances than that. | |
Keep 10120 in mind. Any event whose probability is far smaller than 1 in 10120 has, on the resources of this universe, no realistic opportunity to occur even once — no matter how many years you add. Time is only one of the three inputs, and it is the smallest of them. Adding zeros to the clock does not help when the deficit is a hundred orders of magnitude wide.
Why this matters
Notice what has just happened to the “ten trillion years” argument. Ten trillion years is 3 × 1020 seconds. Even if every atom in the universe attempted a new chemical arrangement every Planck time for that entire span, you would still be nowhere near the numbers that follow. The debate is not about whether the time is long. It is about whether the target is small.
Section 03
One protein: the arithmetic in full
Every living cell runs on proteins. A protein is a chain of amino acids that folds into a specific three-dimensional shape, and the shape is the function. Get the sequence wrong and the chain either fails to fold or folds into something inert. An average protein runs 300 to 400 amino acids long; the calculation below uses a conservative 150, which is short.
Step one: the sequence
In 2004, molecular biologist Douglas Axe published a study in the Journal of Molecular Biology estimating how rare functional sequences are within the space of all possible sequences. Working with a 150-residue section of a bacterial enzyme, he estimated the ratio of functional to possible sequences at roughly 1 in 1077.
Step two: the bonds
Amino acids in a protein are joined by peptide bonds. In an unguided chemical environment, other linkages form just as readily. The standard estimate is that roughly half of the bonds formed would be peptide bonds. For a 150-unit chain you need about 149 correct bonds in a row: 0.5149, or about 1 in 1045.
Step three: the handedness
Amino acids come in two mirror-image forms, left-handed (L) and right-handed (D). Laboratory synthesis produces both in equal measure. Life uses only the left-handed form — a single right-handed residue in the wrong place can prevent folding. For 150 residues: 0.5150, again about 1 in 1045.
Multiply the three together
1 in 10167
The odds of a single functional 150-residue protein assembling by chance
Set that against Lloyd’s ceiling. The universe’s entire computational history is 10120 operations. The requirement here is 10167 — forty-seven orders of magnitude beyond everything the cosmos has ever done.
Put in terms of the ten-trillion-year claim: to get a single protein inside that window you would need roughly 10147 independent attempts every second, for ten trillion years, without pause.
And a protein is not a cell. It cannot replicate, cannot metabolise, cannot repair itself, and in a warm aqueous environment it will hydrolyse back into its parts long before anything else arrives to work with it.
Section 04
The problem isn’t odds. It’s information
Improbability alone proves nothing. Every hand of cards you are dealt is astronomically improbable, and yet you were dealt one. The philosopher’s objection is fair, and it has to be answered properly.
The answer is that the origin of life does not present an improbable outcome. It presents an improbable outcome that also matches an independently specified pattern — namely, functional protein folds and a working genetic code. A random hand of cards is improbable but unspecified. A royal flush dealt to the dealer five times running is improbable and specified, and no one attributes it to chance.
Stephen Meyer’s argument in Signature in the Cell turns on precisely this. DNA does not merely have structure; it carries sequence-specific instructions that are read, translated, and executed. In every other case where we know the origin of information of this kind — a book, a program, an alphabet — the cause was a mind. Meyer’s claim is not that we cannot yet explain it. It is that we do know of a cause sufficient to produce it, and it is not chance.
A measurable version of the same idea
This is not only an apologetic argument. Jack Szostak — a Nobel laureate and one of the leading origin-of-life researchers in the world — proposed in Nature in 2003 that biological molecules be measured by their functional information: the number of bits required to specify a sequence that performs a given function at a given level. Robert Hazen and colleagues developed the measure formally in 2007.
The point is that mainstream science accepts the quantity. What it does not have is a mechanism that generates large amounts of it from undirected chemistry. The disagreement is about the source, not the existence, of the information.
Section 05
The chemistry nobody can do
The probability arguments above are theoretical. There is a separate, blunter line of evidence that comes from the laboratory bench.
James Tour is a synthetic organic chemist at Rice University, one of the most cited chemists working today, with a career built on constructing complex molecules deliberately. In 2019 he published an open letter to his colleagues about the state of origin-of-life research. His complaint was not philosophical. It was technical.
His argument, in outline:
- Prebiotic syntheses do not produce clean products. They produce tars — complex mixtures of thousands of compounds. Getting a usable yield requires chemists to isolate, purify, and protect intermediates at every stage. There is nothing on the early Earth to do that job.
- The steps interfere with each other. Conditions that favour one reaction destroy the products of another. Sugars degrade in the presence of amino acids. Nucleotides hydrolyse in water — the very solvent the story requires.
- The published successes are heavily engineered. When a paper reports the synthesis of a biological building block, the reagents were purified, added in a chosen order, at chosen concentrations, at chosen times, by a chemist. Tour’s point is that the intelligence required to run the experiment is not acknowledged in the conclusion drawn from it.
- We cannot build a cell. Not from purchased, purified components, with unlimited funding and every tool modern chemistry possesses. Tour’s position is that chemists who work with molecules daily know how far this is from possible, and that popular accounts systematically obscure the gap.
Tour is not making a probability claim. He is making an availability claim: the pathway does not exist, not even on paper, not even with a chemist running it.
Section 06
The simplest cell we have ever built
In March 2016, J. Craig Venter’s team published JCVI-syn3.0 in Science — a bacterium with a designed, minimised genome, the smallest of any self-replicating organism known. The goal was to strip life to its irreducible parts and find out what those parts are.
The final count was 473 genes across roughly 531,000 base pairs. Earlier comparative-genomics work had suggested the floor might sit somewhere near 250 to 300 genes. Two decades of work could not get below 473.
The more arresting result was what those genes are doing. Of the 473, the function of 149 — nearly a third — could not be determined. About 70 of them can be broadly classified by structure. For 79, the function is a complete blank. They are essential: remove them and the cell dies. Nobody knows why.
We have shown how complex even the simplest life forms are. And these discoveries have humbled us.J. Craig Venter
If a single 150-residue protein sits at 1 in 10167, then 473 of them arriving together — on the same assumption of independence, and treating each as a comparable target — gives roughly 1 in 1078,991. A number of that size is not improbable in any ordinary sense. It is a number for which the universe contains no resources at all.
Stating the assumptions plainly
That figure is 473 × 167, and it assumes each protein is an independent 150-residue target. It is offered as an indication of scale, not as a precise measurement — and it should be presented that way. But the honest version does not weaken the case: even reducing the exponent by a factor of a hundred leaves a number beyond 10780, against a universal ceiling of 10120. The conclusion is not sensitive to the assumptions. It is only sensitive to whether the target is specified at all — and it is.
Section 07
The multiverse escape hatch
There is one move that neutralises every probability argument at once: make the number of trials infinite. If reality contains unlimited universes with unlimited chemistry running in each, then any event with a non-zero probability occurs somewhere, and we necessarily find ourselves in the branch where it did.
This is not a fringe suggestion. The most striking version of it comes from Eugene Koonin, an evolutionary biologist at the US National Center for Biotechnology Information and no friend of creationism.
In a 2007 paper in Biology Direct, Koonin set out to calculate the probability of the origin of the “breakthrough stage” — a coupled replication–translation system, the minimum machinery for Darwinian evolution to begin operating at all. His conclusion was that within the observable universe, the probability is on the order of 10-1018.
A replication–translation system in one observable universe
1 in 101018
Published in Biology Direct, 2007, by an evolutionary biologist arguing against design
Koonin’s own response was to invoke eternal inflation: an infinite number of causally disconnected regions, in which even this probability is realised somewhere. He was explicit that the model is required in order to keep the origin of life within the reach of chance.
Three things follow.
First, the number is conceded from the other side. When apologists produce figures like 10-167, the standard reply is that the calculation is naive. Koonin’s figure is worse by eight hundred orders of magnitude, and it comes from a working evolutionary biologist publishing in a peer-reviewed journal.
Second, the multiverse is not observational. Regions beyond our horizon are, by construction, causally disconnected from us. No experiment can detect them. This is not a temporary limitation of instruments; it is a structural feature of the model.
Third, the reasoning explains too much. If any finite improbability can be absorbed by postulating unlimited unobservable trials, then no observation can ever count as evidence for design — or against any hypothesis whatever. A move that immunises a position against all possible evidence has stopped functioning as science and started functioning as a commitment.
Section 08
The objections, answered
These are the strongest replies, stated as their defenders would state them.
“Nobody claims life assembled at random. Chemistry is not a lottery.”
This is the most important objection, and it is fair against the crudest form of the argument. Real proposals involve autocatalytic cycles, mineral surfaces, self-assembling lipid membranes, and chemical selection — processes that are not random draws.
The reply is that none of these mechanisms address sequence specificity, which is where the improbability lives. Chemical affinities determine which reactions occur; they do not determine which sequence results. If chemistry did specify the sequence, the sequence could not carry information — for the same reason that magnetic letters that snap into one fixed order cannot spell a message. The freedom of the sequence is what makes the genetic code possible, and it is precisely that freedom the probability calculation is measuring. Selection cannot help before replication exists, and replication is what needs explaining.
“The RNA world solves this. RNA is both information and catalyst.”
The RNA world is the leading hypothesis, and it has real experimental support: ribozymes exist, and RNA can catalyse some reactions on itself. But four problems remain unsolved. Ribose is difficult to produce prebiotically and unstable once produced. Attaching bases to sugars and phosphates without a chemist’s intervention is unsolved. RNA hydrolyses rapidly in water. And no ribozyme has ever been produced — even by directed laboratory evolution — that can copy a strand as long as itself with the fidelity required to sustain a lineage. Leslie Orgel, who co-founded the hypothesis, spent his later career detailing why the chemistry does not work as advertised.
“Hydrothermal vents provide free energy and a natural gradient.”
Alkaline vents give a genuine proton gradient and mineral compartments, and the proposal is serious. But an energy source is not an explanation of specified sequence. A vent supplies power; it does not supply instructions. A generator connected to a printing press does not produce a book. The energy problem and the information problem are separate, and only the first has a candidate answer.
“Miller’s atmosphere was wrong, but the experiment still works.”
This is correct and should be granted. The current consensus is that the early atmosphere was largely neutral rather than strongly reducing, which weakens the original setup. But locally reducing environments — volcanic plumes, for example — are plausible, and when Jeffrey Bada’s team reanalysed Miller’s own archived samples in 2008 with modern instruments, they recovered a far richer set of amino acids than Miller had been able to detect.
None of which touches the argument. Amino acids are not the difficulty. They occur on meteorites. The difficulty is arranging them in a specified order, in a single chain, with uniform handedness, in an environment that also destroys them.
“Axe’s 10⁻⁷⁷ figure has been criticised.”
It has, and the criticism deserves stating. Axe measured functional sequence density within one particular protein fold, and critics argue this cannot be generalised to all functional sequences — there may be many more functional folds than his method samples. Estimates from other groups have come out considerably less extreme.
Two things in response. First, even the most generous published estimates leave the target far below what 10120 operations can search. Second, the argument does not depend on this single number. Koonin’s figure was derived independently, by an opponent, and is vastly larger. Remove Axe entirely and the case stands.
“This is a God-of-the-gaps argument.”
The objection has force against any argument of the form “we don’t know, therefore God.” But that is not the structure here. The inference is not from ignorance; it is from a positive and repeated observation: in every case where we know the causal history of specified information, the cause was intelligent. That is an argument from what we do know, and it is the same inferential form archaeology and forensic science use without controversy.
It is also worth noting that the gap has not been closing. Seventy years after Miller, the numbers have grown worse, not better, and the leading naturalistic answer now requires an infinite set of universes nobody can observe.
Section 09
The men who changed their minds
Arguments are not settled by who holds them. But when people who spent their careers defending a position abandon it under the weight of their own calculations, that is worth recording.
Astronomer
Sir Fred Hoyle
Cambridge · coined the term “Big Bang” as mockery
One of the leading atheists of his generation. After calculating the probability of life emerging by chance at 1 in 1040,000, he became a theist.
Philosopher
Antony Flew
The defining atheist philosopher of the twentieth century
An idol to Richard Dawkins. In 2004, citing the complexity of DNA and the failure of every naturalistic account of its origin, he announced his belief in a Creator.
Biochemist
Dean Kenyon
Co-author of Biochemical Predestination
He wrote the standard defence of chemical evolution. As he came to grasp the complexity of life, he abandoned his own earlier work and became a creationist.
Hoyle’s figure is worth pausing on. He arrived at 1 in 1040,000 by estimating the odds of the roughly two thousand enzymes a cell requires all arising by chance. His comparison has become famous: the likelihood is that of a tornado sweeping through a junkyard and assembling a Boeing 747 from the material inside.
Hoyle also gave the argument its most quoted line, drawn from the fine-tuning of carbon production in stars — a discovery that was his own:
A superintellect has monkeyed with physics.Fred Hoyle, Engineering and Science, 1981
He added that the numbers seemed to him overwhelming enough to put the conclusion almost beyond question. This was the man who had spent thirty years resisting a universe with a beginning because he disliked its theological flavour.
Section 10
The motive beneath the method
The honest ones admit they do not know. Richard Dawkins has said in interview, “Nobody knows how the first living matter came into existence.” That admission is not a scandal — it is the correct scientific position, and it is shared across the field.
What is worth examining is the rule that governs what may be concluded from the admission. That rule is not itself a discovery. It is a prior commitment, and some of its defenders have described it with unusual candour.
The Harvard geneticist Richard Lewontin, reviewing Carl Sagan in The New York Review of Books in 1997, explained that the commitment to materialism is not a conclusion drawn from method but an assumption brought to it, because — in his phrase — “we cannot allow a Divine Foot in the door.”
The philosopher Thomas Nagel, an atheist, wrote in The Last Word that his own resistance is not merely intellectual: it is not just that he does not believe in God, but that he does not want there to be one. He named the condition the fear of religion, and noted that it affects a great many people, including himself.
The evolutionary biologist Michael Ruse put it more bluntly than most. After conceding that the origin of life is “desperately difficult,” he closed a debate with these words:
I would rather be a fool than surrender to the Bible.Michael Ruse
There it is. The real motive on the table. Not a shortage of evidence — a decision made in advance about what the evidence will be permitted to say.
Paul’s claim in that verse is not that the evidence is hidden and must be uncovered by specialists. It is that the evidence is plain, and that the failure is not a failure of information. The next two verses name what it is instead: a refusal to honour what is already known. “Although they claimed to be wise, they became fools.”
The complexity that science keeps unveiling, the more it advances, points clearly to one conclusion. Seventy years of increasingly sophisticated work has made the origin of life harder to explain by chance, not easier. Every gain in resolution has revealed more machinery, more information, more specificity — and no mechanism capable of producing it.
The God who made this world is alive. He made everything for us — and He made us.
Questions?
An objection, a source you want checked, a point you think is wrong — we would rather hear it than not. Send it over; it will be read and answered.
Write to us co**********@***il.comSources
- Axe, D. D. (2004). “Estimating the prevalence of protein sequences adopting functional enzyme folds.” Journal of Molecular Biology, 341(5), 1295–1315.
- Lloyd, S. (2002). “Computational Capacity of the Universe.” Physical Review Letters, 88(23), 237901.
- Hutchison, C. A. III, et al. (2016). “Design and synthesis of a minimal bacterial genome.” Science, 351(6280), aad6253.
- Venter, J. C., quoted in Singer, E. (2016). “In Newly Created Life-Form, a Major Mystery.” Quanta Magazine, 24 March 2016.
- Koonin, E. V. (2007). “The cosmological model of eternal inflation and the transition from chance to biological evolution in the history of life.” Biology Direct, 2:15.
- Meyer, S. C. (2009). Signature in the Cell: DNA and the Evidence for Intelligent Design. HarperOne.
- Szostak, J. W. (2003). “Functional information: Molecular messages.” Nature, 423, 689.
- Hazen, R. M., Griffin, P. L., Carothers, J. M., & Szostak, J. W. (2007). “Functional information and the emergence of biocomplexity.” PNAS, 104 (suppl. 1), 8574–8581.
- Tour, J. M. (2019). “An Open Letter to My Colleagues.” Inference: International Review of Science, 3(2).
- Miller, S. L. (1953). “A Production of Amino Acids Under Possible Primitive Earth Conditions.” Science, 117(3046), 528–529.
- Johnson, A. P., Cleaves, H. J., Dworkin, J. P., Glavin, D. P., Lazcano, A., & Bada, J. L. (2008). “The Miller Volcanic Spark Discharge Experiment.” Science, 322(5900), 404.
- Orgel, L. E. (2004). “Prebiotic Chemistry and the Origin of the RNA World.” Critical Reviews in Biochemistry and Molecular Biology, 39(2), 99–123.
- Hoyle, F. (1981). “The Universe: Past and Present Reflections.” Engineering and Science, November 1981, 8–12.
- Hoyle, F., & Wickramasinghe, C. (1981). Evolution from Space. J. M. Dent.
- Flew, A., & Varghese, R. A. (2007). There Is a God: How the World’s Most Notorious Atheist Changed His Mind. HarperOne.
- Kenyon, D. H., & Steinman, G. (1969). Biochemical Predestination. McGraw-Hill.
- Lewontin, R. (1997). “Billions and Billions of Demons.” The New York Review of Books, 9 January 1997.
- Nagel, T. (1997). The Last Word. Oxford University Press, ch. 7.
- Greene, T. P., et al. (2023). “Thermal emission from the Earth-sized exoplanet TRAPPIST-1 b using JWST.” Nature, 618, 39–42.
- Ruse, M., closing remarks in public debate. [Full citation to be supplied.]
- Dawkins, R., interview remarks on the origin of the first living matter.
- Scripture quotations from the Holy Bible, New International Version (NIV).
Leave a Reply
Your email is safe with us.