The Origin of Life: Evidence and Design

In 1953, a graduate student named Stanley Miller sealed a flask of water, methane, ammonia, and hydrogen, ran electric sparks through it for a week, and found at the bottom a brownish sludge containing several amino acids—the building blocks of proteins.¹ The result electrified the public. It seemed to show that the gap between dead chemistry and living matter might be bridged in a week of lab time. More than seventy years later, the most honest thing that can be said about the origin of life is that we are still, in the deepest sense, staring at that flask. We can make some of the bricks. We cannot account for the building—and above all, we cannot account for the blueprint.

This essay is about that unsolved problem, and about a contested inference some draw from it. It will try to do something harder than cheerleading for either side: state, fairly, what abiogenesis must explain; survey honestly how far the research has and has not gotten; lay out the design argument from biological information at its strongest; and then steelman the mainstream research program just as carefully—because the “god-of-the-gaps” charge is serious and deserves a serious answer. The aim is not a victory but the truth, held with appropriate confidence and appropriate humility.

What Abiogenesis Must Explain

“The origin of life” sounds like one problem. It is at least three, and the third is the deepest.

The first is metabolism: a living cell is a self-sustaining network of chemical reactions that captures energy and uses it to build and repair itself. The second is self-replication: life copies itself, passing on its structure to descendants. But the third problem is the one that has come to dominate the field, and it is the one most easily missed if you imagine life as merely “complicated chemistry.” It is the origin of biological information.

The discovery of the genetic code revealed that the cell is not just a bag of reacting molecules but, in a real sense, an information-processing system. DNA stores instructions in a four-character alphabet (the bases A, T, G, and C); those instructions are transcribed and then translated, by an elaborate molecular machinery, into proteins—long chains of amino acids that must be sequenced just so in order to fold into working three-dimensional shapes. Get the sequence wrong and the protein is, in general, useless. This is not loose analogy. Bill Gates famously observed that “DNA is like a computer program but far, far more advanced than any software ever created.”² Even the simplest free-living organisms, such as the bacterium Mycoplasma genitalium, carry several hundred genes; a 2016 synthetic-biology effort to build a truly minimal cell still required 473 genes whose functions were not all understood.³ The smallest known autonomous life is already staggeringly information-rich.

So the question that drives the design argument is not merely “how did chemistry get complicated?” but “where did the specified sequence—the functional arrangement of parts, the code itself—come from?” Crystals are highly ordered but carry almost no information (they repeat). Random polymers carry no functional message. Life sits in a third category: aperiodic sequences that are also functionally specific, like the letters of a sentence. That combination—what the philosopher of science Stephen Meyer, following others, calls “specified complexity” or “specified information”—is the explanatory target.⁴

The Experimental Program, and Its Limits

It is essential to be fair to the science here, because the popular apologetic habit of waving away decades of careful work is neither honest nor persuasive.

Miller–Urey: a real result that is not a solution

The Miller–Urey experiment was a genuine landmark: it showed that some biological building blocks form spontaneously under plausible energy input. But it does not solve the origin of life, and serious researchers do not claim it does. Two limits are worth naming. First, it produced building blocks, not living systems, and certainly not information—a heap of amino acids is to a cell what a bucket of letter tiles is to the works of Shakespeare. Second, the strongly “reducing” (hydrogen-rich) atmosphere Miller assumed is now widely thought not to match the early Earth, which most geochemists believe was more neutral; in less reducing conditions the yields drop dramatically.⁵ The experiment opened a field; it did not close the question.

The RNA world, and its real difficulties

The deepest conceptual problem in origin-of-life research is a chicken-and-egg loop: in modern cells, DNA stores the information but cannot copy itself without proteins, and proteins cannot be built without the information in DNA. Which came first? The dominant answer since the 1980s is the RNA-world hypothesis, proposed in various forms by Carl Woese, Francis Crick, Leslie Orgel, and named by Walter Gilbert in 1986. RNA can both store information (like DNA) and catalyze reactions (like a protein); the discovery of catalytic RNAs (ribozymes) by Thomas Cech and Sidney Altman, which won the 1989 Nobel Prize, made the idea concrete.⁶ An RNA molecule that could copy itself would, in principle, break the loop.

There has been real progress. In a striking 2009 paper in Nature, John Sutherland’s group (then at the University of Manchester) synthesized activated pyrimidine ribonucleotides under prebiotically plausible conditions, bypassing the long-standing problem of joining sugar and base directly.⁷ Jack Szostak’s laboratory has made advances in protocell membranes and non-enzymatic copying of RNA templates.⁸ This is not a stagnant field, and it would be a mistake to say otherwise.

But the difficulties remain formidable, and—crucially—they are stressed not only by design theorists but by leading origin-of-life chemists themselves. Gerald Joyce and Leslie Orgel, two founders of the field, candidly contrasted the optimistic “molecular biologist’s dream” with the “prebiotic chemist’s nightmare” of actually making RNA from scratch.⁹ The problems include the instability of RNA, the difficulty of obtaining the right-handed sugars and avoiding interfering side-products (the homochirality problem—life uses one “handedness” while undirected chemistry yields both), and, above all, getting from short random strands to a sequence long and specific enough to replicate. Even granting a pool of nucleotides, no one has demonstrated a self-replicating RNA arising and sustaining itself under realistic prebiotic conditions. And note carefully: even a working RNA replicator would not by itself explain the origin of the information—it would presuppose a functional, specified sequence to get started, pushing the central question back one step rather than dissolving it.

Metabolism-first and hydrothermal scenarios

Partly because of these difficulties, other researchers pursue “metabolism-first” models, in which self-sustaining chemical cycles arise on mineral surfaces—for example at alkaline hydrothermal vents, a scenario associated with Michael Russell, Nick Lane, and William Martin—before any genetic molecule appears. These are creative and physically grounded. But they face their own sharp critique, again from inside the field: Leslie Orgel, in a posthumously published 2008 paper, argued that proposed self-organizing metabolic cycles make “unreasonable” assumptions, requiring mineral catalysts so selective that they would, in effect, need “a skilled synthetic chemist” to function.¹⁰ Robert Shapiro, a chemist and longtime skeptic of the RNA world, championed small-molecule metabolic approaches—and was equally scathing about the difficulties on all sides.¹¹ The field is genuinely divided between replicator-first and metabolism-first camps, with no consensus mechanism in either.

The Design Inference from Information

Against this backdrop, the design argument can be stated carefully. It is best understood not as “science can’t explain it, therefore God,” but as an inference to the best explanation about a specific kind of effect.

Premise 1. The origin of the first life required the production of large amounts of functionally specified information—the genetic code and the sequence-specificity of biological macromolecules. Premise 2. Of the available causes—chance, physical-chemical necessity, the two combined, or intelligent agency—only intelligent agency is known, from repeated experience, to produce specified information of this kind. Conclusion. Therefore intelligent design is, at present, the best explanation for the origin of biological information. This is the structure Stephen Meyer develops at book length in Signature in the Cell (2009).⁴

The argument’s engine is Premise 2. Its supporters point to two lines of consideration. The first is positive: in every case where we know the causal story of an information-rich artifact—a paragraph, a software program, a coded message—the cause is a mind. Information of this functional, sequence-specific sort is, in our uniform experience, the signature of intelligence. Here, though, the inference is only as strong as its key term: much depends on whether “specified information” names a genuine natural kind, or quietly smuggles in the intentional, semantic sense that belongs to messages authored by minds—which would make the conclusion circular. The argument should therefore rest on a strictly functional definition: a sequence carries specified information when, of the astronomically many sequences of its length, it falls in the tiny set that actually folds and replicates—a property fixed by chemistry and measurable in the laboratory, not by appeal to any author’s purpose. Defined this way, the claim that such functional specificity is, in our experience, mind-produced is an empirical claim about a measurable feature of nature, and does not lean on the human-artifact analogy to do its work. The second is the apparent inadequacy of the alternatives. Chance faces steep improbabilities: the molecular biologist Douglas Axe, in a 2004 paper in the Journal of Molecular Biology, estimated that functional protein folds might be as rare as one in 10⁷⁷ of possible sequences of that length. A single fold at those odds is not, by itself, beyond the roughly 10⁸⁰ particles and 10¹⁷ seconds the observable universe affords as trials—but a working cell requires many such folds together in one integrated, coded system, and it is that compound improbability, not any single protein, that outstrips the resources of chance.¹² Necessity (law-like chemistry) tends to produce repetitive order, not aperiodic specificity. And selection cannot be invoked at the very first step, because natural selection requires accurate self-replication to already exist—which is precisely what is being explained.

It is worth registering how seriously secular thinkers have felt the force of the problem. Francis Crick, co-discoverer of DNA’s structure, wrote that “the origin of life appears at the moment to be almost a miracle, so many are the conditions which would have had to have been satisfied to get it going”—though, in fairness, Crick added at once that this “should not be taken to imply” that life could not have begun by ordinary chemistry, and he himself floated directed panspermia (life seeded from elsewhere) rather than design.¹³ The synthetic chemist James Tour, who actually builds molecules for a living, argues forcefully that the field has vastly underestimated how far it is from synthesizing even the components of a cell, let alone assembling them.¹⁴ Crick’s hedge matters: the difficulty is widely conceded; the inference to a designer is not.

Steelmanning the Mainstream Program

Here honesty requires real work, because the strongest objections to the design inference are good ones, and a Christian who cares about truth should be able to state them as well as their proponents do.

“This is a god-of-the-gaps argument”

The most important objection—pressed by philosophers such as Graham Oppy and Sahotra Sarkar, and by working scientists generally—is that the design inference simply baptizes ignorance. The history of science is a history of gaps closing: lightning, disease, planetary motion, and the diversity of species were all once attributed to direct divine action and later given natural explanations. To infer a designer from today’s missing mechanism, the critic says, is to repeat a long-discredited move and to bet against the very process that has reliably delivered. Worse, it can be a science-stopper: if “a mind did it” is on the table, why keep working at the chemistry?

This deserves a genuine concession and a careful reply. The concession: absence of a current explanation is not proof of impossibility, and anyone who says “abiogenesis is impossible” has overreached the evidence. The honest claim is comparative and provisional, not a demonstration. We do not know that no undirected pathway exists; we know that none has been demonstrated.

The reply: the design inference, properly stated, is not an argument from ignorance but an argument from knowledge—from a known causal power. We are not saying “we can’t imagine how nature did it.” We are saying “we positively know one kind of cause that produces specified information, and we know of no other.” That is the same abductive logic a forensic scientist or an archaeologist uses, and it is in principle defeasible: demonstrate an undirected chemical route from simple precursors to specified information, and the inference is overturned. A genuine god-of-the-gaps claim cannot be falsified that way; this one can. The science-stopper worry also cuts the other way—many design theorists actively want the chemistry pursued, precisely because a clean negative result strengthens the inference and a positive result would settle it. The right posture is to keep doing the chemistry while being candid that, as of now, it has not delivered.

“The improbability numbers are wrong”

The second serious objection targets Premise 2’s probabilistic backbone. Douglas Axe’s 1-in-10⁷⁷ figure has been sharply contested. Critics—including biologists such as Arthur Hunt and the authors at venues like the Panda’s Thumb—argue that Axe’s experimental method measured something narrower than the true density of all functional folds, that functional sequences are far more common than he estimated, and that proteins can arise and improve gradually rather than by one-shot assembly.¹⁵ This is a real scientific dispute, and the design argument should not lean on a single contested number as though it were settled. The mastery summary for this argument states the point plainly: Axe’s estimate is contested by other biochemists, and the case must rest on more than one line.¹⁶

The measured reply is that the comparative inference does not require Axe’s exact figure. Even on far more generous estimates of functional density, no demonstrated undirected pathway yet bridges the gap from chemistry to a self-replicating, information-bearing system—and the origin-of-life problem is not only about one protein fold but about an integrated, coded system. The probability argument is one strand of a convergent case, not its load-bearing pillar. Where the numbers are contested, say so; lean on the convergence.

“You’re demanding the present mechanism instead of the historical one”

A subtler objection, urged by philosophers of biology, is that we do not know early-Earth conditions well enough to declare any gap real. Prebiotic chemistry may have run along routes that left no trace and that we have not thought to test. To infer design from our ignorance of four-billion-year-old conditions is, on this view, premature. This is a fair caution, and the right response is not to deny it but to locate the inference correctly: it rests on the positive track record of intelligence plus the persistent failure of undirected proposals so far, and it remains revisable the moment prebiotic chemistry delivers a real mechanism. That is how inferences to the best explanation are supposed to work—held as the current best, not the final word.

The State of the Scholarship

What can be said without controversy is this: there is no scientific consensus mechanism for the origin of life. The leading textbooks and review articles describe the problem as open. The evolutionary biologist Eugene Koonin, no friend of design, has written candidly about how intractable the origin of the translation system and the genetic code remains, even invoking the multiverse to supply enough probabilistic resources.¹⁷ The field is vigorous, well-funded, and genuinely advancing on sub-problems—nucleotide synthesis, protocell membranes, mineral catalysis—but it has not produced, and does not claim to have produced, a demonstrated route from chemistry to a coded, self-replicating cell.

It is equally important to say what the scholarship does not show. It does not show that such a route is impossible. The majority of working scientists expect one to be found, and that expectation is not irrational given the history of the discipline. The design inference is a minority position in the academy, defended by a serious but small community (the Discovery Institute’s circle, and sympathetic philosophers), and it is widely—though not universally—judged to fall outside methodological naturalism. A reader should weigh that sociological fact honestly: most experts disagree with the design conclusion, even while agreeing that the problem is unsolved.

One further point of intellectual honesty. None of this essay’s argument depends on rejecting common descent or biological evolution more broadly. The origin of life (abiogenesis) and the origin of species (evolution) are distinct questions; Darwin himself bracketed the former. Apologia Daily takes no position binding Christians to a particular view of evolutionary biology—faithful believers hold a range of views—and the information argument here is about the first cell, not about the mechanisms of descent thereafter.¹⁸

What the Argument Does and Doesn’t Establish

The design inference from biological information does not prove that life was designed, and it certainly does not prove the God of the Nicene Creed. At most it argues that, among the causes currently on offer, intelligent agency is the one we know can produce the relevant effect—and that this makes design the best present explanation of life’s informational signature. It is a probabilistic, defeasible, comparative claim. It could be overturned tomorrow by a laboratory result, and the honest defender should say so cheerfully.

What it does do is real and worth holding. It shows that the confident assurance—sometimes heard—that science has essentially solved the origin of life, or soon will, outruns the evidence. It locates a genuine and deep mystery at the foundation of biology: the origin of code-like, functionally specified information, for which the only uncontested known source is a mind. And it shows that the inference to design, properly stated, is not anti-science but a particular reading of what the science has and has not delivered. The paleobiologist Simon Conway Morris, himself a Christian and no design theorist, has argued that life’s deep order and convergence are at least suggestive of a cosmos that is ripe for life—a reminder that wonder at these questions is not the property of one party.¹⁹

For the Christian, the cell’s astonishing informational depth is not a club to wield but an invitation to attentiveness—“in him all things hold together” (Colossians 1:17). The right posture before the origin of life is the one Scripture commends for every contested thing: to give a reason for our hope “with gentleness and respect” (1 Peter 3:15), to concede what should be conceded, to overstate nothing, and to follow the evidence wherever it leads—confident that truth, in the end, has nothing to fear from inquiry.