Antibiotic resistance and the longest-running evolution experiment in history are the two best observed exhibits Darwinism has. Here is what they demonstrated, and what they did not.
Most arguments about evolution are arguments about the deep past, which means they are arguments about inference. Nobody watched the first eye assemble. That is not a cheap shot; it is the nature of historical science, and the same limit applies to anything a Christian wants to say about Genesis.
So ask a better question. Set the fossils aside and ask what has actually been watched — measured, written up, and open to anyone who wants to check the numbers.
You will get two answers, and they are the right two. Antibiotic resistance, and a flask of bacteria in a Michigan lab that has been running since 1988. Let us take both seriously, which means conceding what they show before pressing on what they do not.
First, the Concession, and It Is Not a Small One
Natural selection is real. It happens. It is observable, quantifiable, and arguing against it is arguing against arithmetic. Populations do change heritably over generations. Mutations do occur. Selection does sort them. A Christian who says otherwise is not defending Scripture; he is embarrassing it.
And the sloppy version of our own side needs retiring. Saying so what, it is still a bacterium, as though nothing happened, is not an argument. Something happened, and some of it is genuinely interesting. If you want to be taken seriously here, say what happened accurately first.
Exhibit One: How Resistance Actually Arises
A hospital strain that shrugs off a drug is held up as evolution you can see, and in a real sense it is. But resistance is not one phenomenon. It arrives by several distinct routes, and the routes decide what the example proves.
- Selection of variation that was already present. In a population of billions, a few cells frequently already carry a variant the drug cannot touch. The antibiotic does not create them; it removes their competition. Resistance determinants have been recovered from bacteria in deep permafrost tens of thousands of years old, and from populations that had never met a modern pharmaceutical.
- Horizontal transfer. A great deal of clinical resistance travels on plasmids, passed between bacteria like a shared tool. A gene that already existed somewhere in the biosphere moved addresses. That is a real and medically serious event. It is not the origin of a gene.
- Breakage that happens to help. This one is the most instructive. Break the porin channel and the drug cannot get in. Break the regulator that keeps an efflux pump switched off and the pump runs constantly, bailing the drug out. Alter the ribosomal protein the drug binds to and you have a degraded binding site that the drug no longer recognizes. In each case the cell got worse at something and, in the presence of the poison, that was an advantage.
The diagnostic tell is what happens when you withdraw the drug. Resistant strains frequently carry a fitness cost and get outcompeted by their ordinary cousins, which is precisely why resistance frequencies in a population can fall after a drug is pulled from rotation. That is the signature of a tradeoff, not an upgrade.
In fairness: not every mechanism is a loss. Some enzymes that chew up antibiotics are efficient, well-built proteins, duplication followed by modification is a documented route to new enzyme specificities, and compensatory mutations can claw back much of the fitness cost. Anyone who tells you resistance is nothing but damage has oversimplified it.
Exhibit Two: The Flask That Has Been Running Since 1988
In February 1988, Richard Lenski started twelve populations of E. coli from a single ancestor and began transferring them to fresh glucose medium every day. He has never stopped. Samples are frozen at intervals, so the experiment carries its own fossil record: any later generation can be thawed and raced against its own great-grandparents.
It is one of the best-designed experiments in modern biology, and Christians arguing about origins ought to say so. Thirty-eight years. More than seventy-five thousand generations. Huge populations, short generations, relentless selection, a sequenced genome throughout. If undirected mutation plus selection is a creative engine, this is the most favorable stage ever built for it. So what came out?
- The cells got substantially better at living on glucose in that flask. Measurably, repeatably, in all twelve lines.
- They got bigger.
- Several lines lost DNA repair function and became mutators, which raised the supply of mutations. Useful in context; a loss nonetheless.
- Many lines independently jettisoned capabilities they no longer needed in a glucose-only world. Streamlining by deletion.
- And around generation thirty-one thousand five hundred, one population out of twelve acquired the ability to metabolize citrate in the presence of oxygen — something ordinary E. coli cannot do. It is such a reliable inability that it is used as a diagnostic test to identify the species.
That last result is the strongest single card in the deck, and it deserves to be played at full strength rather than waved away. It is a new metabolic capability. It required earlier potentiating mutations, so it was contingent on history. It took tens of thousands of generations and it happened in only one line of twelve. And it was not a simple breakage.
What the Citrate Result Turned Out to Be
When the mechanism was worked out, here is what it was. E. coli already owns a citrate transporter; it uses it without oxygen. It already owns the entire downstream machinery for processing citrate — the citric acid cycle is named after the molecule. The missing piece was never the hardware. It was the switch.
The innovation came through tandem duplication of that existing transporter gene, which landed a copy downstream of a different promoter, so the transporter now ran under conditions it had not previously run under. The parts were in the box. A copying error put an old part under new management.
That is a genuine result and I will not shrink it. Duplication and re-regulation is a real mechanism, and this is a clean documented instance of it yielding a function the lineage had never expressed. Anyone citing it as evidence that novelty can arise has a legitimate data point, and I would rather grant it than quibble.
Now the Question That Actually Matters
The issue was never whether populations change. It is whether the kind of change we observe, run out long enough, accounts for the kind of thing biology contains.
So take inventory honestly. Across thirty-eight years, seventy-five thousand generations, and population sizes that dwarf anything in vertebrate history, the catalogue of observed events reads: optimization of existing functions, loss of unneeded functions, broken regulators, broken repair systems, duplicated genes, and one existing transporter placed under a new promoter.
What is not on the list: a new protein fold. A new cell type. A new organ. A new body plan. Not one.
That absence is not an argument from ignorance. The flask is the most closely watched population of organisms in history, and this is what it did.
A second detail is worth sitting with. The fitness gains decelerate hard; each increment costs more than the one before. Lenski’s own group argues that the mathematical form of that curve has no hard ceiling, and that is a fair reading of their data which I will not misreport. But note what the long climb produced in practice: a very good glucose-eating E. coli.
Extrapolation is a legitimate move. But it is a move, and it needs a mechanism with demonstrated capacity, not merely demonstrated direction. What these experiments show clearly is that selection is magnificent at tuning and conspicuously reluctant to invent.
Where Believers Genuinely Disagree
Let me be straight about this argument inside the church, because caricature helps nobody. Christians who hold to an old earth, and some who accept common descent outright, read this same data and draw the line in a different place. They would say that the flask is a tiny window, that a few decades cannot sample what deep time does, and that God can work through processes as freely as He works against them. Those are serious people making a serious argument, and the disagreement is not usually about whether Scripture is true. It is about whether the extrapolation step is warranted.
My own judgment is that it is not warranted on the evidence we have, and that the inventory above is the reason. But a brother who weighs it differently is not therefore a compromiser, and I will not call him one.
The Thing Nobody Watched Happen
Underneath the whole dispute is the question these experiments keep circling without touching: where does the specified arrangement come from in the first place?
Every mechanism on the observed list operates on information that already exists. Duplicate it, break it, move it, re-regulate it, tune it. Those are editorial operations. They presuppose a text. And the flask never once offered an account of the text.
In the beginning was the Word, and the Word was with God, and the Word was God. The same was in the beginning with God. All things were made by him; and without him was not any thing made that was made. (KJV)
John 1:1-3
Scripture is not a laboratory protocol and I will not pretend Genesis is a technical manual. But it does make a claim about the direction of causation: the arrangement came first, and the living things came out of it — rather than the arrangement being the residue of a long enough wait.
Through faith we understand that the worlds were framed by the word of God, so that things which are seen were not made of things which do appear. (KJV)
Hebrews 11:3
So Where Does That Leave You
If you are a skeptic, I am not asking you to pretend the data say something they do not. I am asking you to hold the extrapolation to the standard you would hold any other claim that outruns its evidence. Selection optimizing a bacterium is not the same magnitude of claim as selection authoring a nervous system, and the gap is not covered by saying the word time louder.
And if you are a believer who likes this argument, do the work. Read the papers before you repeat a summary of them. Our credibility on this subject has been spent, repeatedly, by men who were confident about studies they had never opened. The truth does not need our exaggeration, and it is damaged by it.
For the invisible things of him from the creation of the world are clearly seen, being understood by the things that are made, even his eternal power and Godhead; so that they are without excuse. (KJV)
Romans 1:20
One last thing, and it matters most. No argument about a flask has ever saved a man. Grant every extrapolation tomorrow and you would still be standing where you stand, with a conscience that accuses you and a death you cannot negotiate with.
Christianity does not rest on the limits of bacterial metabolism. It rests on a tomb that was found empty by people who had every reason to want the body right where they left it. That is the claim to go and examine.
The bacteria got very good at eating sugar. They never once explained themselves.
Teaching the Word. Watching the Times.
— SmithForChrist