Do miracles happen? Answering Hume

Almost every objection to Christianity eventually arrives here. The virgin birth, the healings, the resurrection — these are the claims that a modern person is trained to set aside before examining.

The training goes back to one eighteenth-century essay.

Hume's argument

David Hume, in An Enquiry Concerning Human Understanding (1748), argued roughly as follows:

1. A miracle is a violation of the laws of nature.
2. The laws of nature are established by firm and unalterable experience.
3. Therefore the evidence against a miracle is as complete as any evidence from experience can be.
4. Therefore no testimony can establish a miracle unless its falsehood would be more miraculous than the event it reports.

It has the feel of a decisive move, and it has functioned as one for two and a half centuries.

Why it fails

It is circular. Premise 2 claims uniform experience against miracles. But how would Hume know experience is uniform? Only by knowing in advance that every reported miracle is false. So the argument assumes its conclusion in order to establish it.

The philosopher John Earman — not a Christian — devoted a book to this, titled with characteristic bluntness Hume's Abject Failure, arguing that the reasoning is confused on probabilistic grounds.

It would rule out any unique event. By Hume's standard, no testimony could establish anything that had not happened before. That would eliminate a great deal of legitimate science, which regularly deals with singular events.

It misunderstands what a law of nature is. Laws are descriptions of regularities under normal conditions, not prescriptions that bind reality. They describe what happens when nothing else interferes.

It rules out evidence a priori. Hume is not weighing testimony and finding it insufficient. He is establishing in advance that no testimony could be sufficient. That is not scepticism; it is a closed system.

What a miracle actually is

Lewis's clarification dissolves the "violation" framing.

If I drop a book, it falls. That is gravity. If I reach out and catch it, I have not violated the law of gravity — I have introduced a new causal factor into the situation. Gravity continues to operate perfectly, now acting on my arm as well.

A miracle, on this account, is God introducing a new factor into the natural order. The laws are not suspended; new initial conditions are supplied, and nature proceeds from there as usual.

If God annihilates or creates or deflects a unit of matter, He has created a new situation at that point. Immediately all Nature domiciles this new situation, makes it at home in her realm, adapts all other events to it.— C. S. Lewis, Miracles

The virgin conception does not suspend embryology. It supplies a fertilised cell by an unusual means, after which nine months of entirely ordinary gestation follow.

The question underneath

Which reveals that miracles are not really the issue. The issue is theism.

If God exists, miracles are possible. A being who created the universe and its laws can obviously act within them. Denying this would be like saying an author cannot write a new sentence into their own novel.

If God does not exist, miracles are impossible. Correctly.

So the argument about miracles is downstream. Someone who says miracles cannot happen is usually not making an argument about evidence — they are stating a prior commitment to naturalism.

The productive move in conversation is therefore to go back one step: "It sounds like the real question isn't whether this particular event happened, but whether there's anyone who could have done it. Can we talk about that instead?"

Methodological versus philosophical naturalism

A distinction worth carrying.

Methodological naturalism is a working rule of science: look for natural causes. This is entirely reasonable. A researcher who concluded "a demon did it" whenever an experiment failed would not be doing science.

Philosophical naturalism is the claim that natural causes are all that exist. That is a metaphysical position, and it is not established by science — it is assumed prior to it.

Confusing the two produces the mistaken belief that science has ruled out the supernatural. Science has ruled it out of its method, which is a very different thing and does not entail the other.

Modern documented cases

The assertion that no miracle has ever occurred under conditions where it could be checked was more plausible before Craig Keener's Miracles (2011) — two volumes, around 1,200 pages, documenting thousands of contemporary claims from around the world.

Keener is careful, and he is honest that many claims do not survive scrutiny. What he establishes is that the blanket dismissal is a claim about the evidence that the evidence does not support. Among the categories he documents:

• Cases with medical documentation before and after.
• Cases witnessed by physicians who were not believers.
• Cases of recovery from conditions with no known spontaneous remission.
• Blindness and deafness reversed with prior clinical records.

His broader sociological point is also worth noting: hundreds of millions of people worldwide report having witnessed or experienced healing. Whatever one concludes about any individual case, treating this as a non-existent phenomenon requires ignoring a very large body of testimony, most of it from the global South.

The Lourdes Medical Bureau is another interesting case: a body including non-Catholic physicians which has assessed thousands of claims and certified around seventy as medically inexplicable — a rejection rate that is itself evidence of seriousness.

What miracles are for in Scripture

They are never random spectacle, and noticing this answers a common objection.

They cluster. They are not evenly distributed across biblical history. They concentrate around Moses and the exodus, Elijah and Elisha, and Christ and the apostles — three periods of major new revelation. In between, centuries pass with very few.

This answers "why don't we see miracles like that today?" The biblical pattern was never continuous. Miracles authenticate a message at the moment it is being delivered.

They authenticate. "God bearing witness by signs and wonders and various miracles" (Hebrews 2:4). Peter at Pentecost: "Jesus of Nazareth, a man attested to you by God with mighty works and wonders and signs" (Acts 2:22).

They signify. John never calls them miracles; he calls them signs (sēmeia). Feeding five thousand signifies the bread of life. Opening blind eyes signifies the light of the world. Raising Lazarus signifies the resurrection and the life. Each one enacts a claim.

Jesus offered them as evidence. "Believe me that I am in the Father and the Father is in me, or else believe on account of the works themselves" (John 14:11). He expected people to reason from evidence to conclusion.

And he refused them as entertainment. He would not perform for Herod (Luke 23:8-9), and he called demanding signs the behaviour of an evil generation (Matthew 12:39). The miracles serve the message; they are not a show.

The one that matters

Christianity does not require you to accept every miracle claim in history, or any modern one.

It requires one.

"If Christ has not been raised, your faith is futile and you are still in your sins" (1 Corinthians 15:17).

Which is where the conversation should end up. Not a general argument about whether the supernatural is possible in the abstract, but a specific historical question about a specific weekend, with specific evidence attached.

Lewis called it the grand miracle: the one that, if it happened, makes all the others unsurprising, and without which none of the rest matter.