Can Science Prove God?

The Question and Its Ambiguity

The question "Can science prove God?" is ambiguous, and the ambiguity needs to be cleared up before anything useful can be said. There are at least three things it could mean.

First, it could mean: can the scientific method — hypothesis, prediction, experiment, replication — establish the existence of God the way it establishes the existence of electrons or genes? The answer to this is almost certainly no. God, as classically conceived, is not an object within the natural world, and scientific method is designed to investigate objects within the natural world. You cannot put God in a test tube.

Second, it could mean: can scientific discoveries — the Big Bang, fine-tuning, the genetic code, the structure of the brain — provide evidence that is more probable on the hypothesis that God exists than on the hypothesis that God does not? The answer here is more contested. Richard Swinburne and the proponents of the fine-tuning argument say yes: the specific values of physical constants are extremely unlikely on naturalism and quite likely on theism. Critics say the evidence is ambiguous or that naturalistic explanations are available.

Third, it could mean: should the question of God be treated as a scientific question at all? This is a question about the scope of science — about whether scientific method is the only route to knowledge, or whether there are questions science cannot answer but philosophy and theology can. Dawkins argued that the God hypothesis is a scientific hypothesis and that science has refuted it. Plantinga argued that the God hypothesis is not a scientific hypothesis and that treating it as one is a category mistake.

These three readings lead to different answers, and the debate is largely about which reading is the right one.

Four Models of Science and Religion

Ian Barbour, a physicist and theologian, proposed a fourfold classification of the possible relationships between science and religion that has become standard in the field. It helps organize the options.

Conflict. On this model, science and religion are fundamentally at war. The history of the relationship is a history of opposition — from Galileo to Darwin to the modern creationism debates — and the conflict is real and irresolvable. This is the model of the new atheism: Dawkins treats the God hypothesis as a scientific claim that has been tested and found wanting. It is also the model of the young-earth creationist, who treats the biblical account as a scientific description and concludes that science is wrong where it disagrees. Conflict models, from opposite directions, agree that science and religion make competing claims about the same things.

Independence. On this model, science and religion are about different things and do not overlap. Stephen Jay Gould called this "non-overlapping magisteria" (NOMA): science covers the empirical world, religion covers values and meaning, and the two have nothing to say to each other. Science answers "how" questions; religion answers "why" questions. The advantage is peace: no conflict, because no contact. The disadvantage is that both sides tend to violate the boundary. Religion makes empirical claims (miracles, resurrection, creation), and science makes philosophical claims (materialism, reductionism) that go beyond its empirical warrant.

Dialogue. On this model, science and religion are distinct but in conversation. They do not make competing claims about the same things, but they do inform each other. Scientific discoveries can raise theological questions — the Big Bang raises the question of cosmic origins, fine-tuning raises the question of design — and theological commitments can shape the interpretation of scientific results. The dialogue model is the most popular among philosopher-theologians, because it takes both science and religion seriously without forcing them into conflict.

Integration. On this model, science and religion are ultimately aspects of a single inquiry. Theological claims and scientific claims are integrated into a unified worldview. Aquinas is an integrationist: nature and grace, reason and revelation, are both gifts of the same God and cannot ultimately conflict. The integration model requires that both science and theology be open to correction by the other, which is a high standard.

The Fine-Tuning Argument

The most prominent current argument that draws on science for the God question is the fine-tuning argument. The argument goes like this: the physical constants of the universe — the strength of gravity, the mass of the electron, the cosmological constant, the ratio of electromagnetic to gravitational force — are set within extremely narrow ranges. If any of them were slightly different, stars would not form, carbon would not be synthesized, the universe would be a soup of hydrogen or a black hole. The probability of these constants taking life-permitting values by chance is astronomically low. But on the hypothesis that a rational designer set them, the probability is much higher. Therefore, the fine-tuning is evidence for God.

Swinburne has made the Bayesian version of this argument: P(fine-tuning | theism) > P(fine-tuning | naturalism), and by Bayes's theorem, the fine-tuning raises the probability of theism relative to naturalism. The strength of the argument depends on the probability assignments, and these are contested. But the argument is a serious one, and it is the closest contemporary philosophy of religion comes to a scientific argument for God.

The main objections are three. First, the many-worlds objection: if there are infinitely many universes, each with randomly set constants, then the probability that at least one is life-permitting approaches 1, and the fine-tuning is no longer surprising. This is a genuine scientific hypothesis (some versions of inflationary cosmology generate it), and if it is true, it undercuts the fine-tuning argument. Second, the necessity objection: maybe the constants are not really free to vary. Maybe there is a deeper physical theory that fixes them, and the apparent contingency is an artifact of our ignorance. Third, the who-designed-the-designer objection: if the designer is complex enough to set the constants, the designer is even more improbable than the fine-tuning it is meant to explain. Dawkins pressed this in The God Delusion, though philosophers generally regard it as weak, since theism does not claim God is complex — classical theism claims God is simple.

The God of the Gaps

A persistent worry about arguments from science to God is the God-of-the-gaps problem. The pattern is familiar: science cannot explain X, so we invoke God as the explanation. Then science explains X, and God is squeezed out. Lightning was attributed to Zeus; now we know it is static electricity. Disease was attributed to demons; now we know it is pathogens. Consciousness is not yet explained; some invoke God. The pattern suggests that using God to fill gaps in scientific knowledge is a strategy that loses over time, as science explains more and more.

The God-of-the-gaps problem is a real concern, and thoughtful theists have responded in two ways. The first is to deny that they are making a gaps argument. The fine-tuning argument, they say, is not based on what science has not yet explained but on what science has explained: the narrowness of the life-permitting range is a scientific discovery, not a gap. The argument uses science, not ignorance, as its basis. The second is to distinguish between God as an explanation within nature and God as an explanation of nature. The God-of-the-gaps worry arises when God is used as a competing scientific explanation — as an alternative to a naturalistic mechanism. But if God is an explanation of why there is nature at all, or why nature has the order it does, then God is not filling a gap in science but answering a question that science does not address.

This is the independence or dialogue model at work. Science explains how the universe works. Theism explains why there is a universe and why it is the kind it is. The two explanations operate at different levels and do not compete. Whether this response succeeds depends on whether you think the question "why is there something rather than nothing?" is a real question or a pseudo-question — and that is itself a philosophical disagreement.

Methodological Naturalism and the Limits of Science

The deepest issue is about the scope of science. Science operates on the principle of methodological naturalism: it explains natural phenomena in terms of natural causes. This is a methodological commitment, not necessarily a metaphysical one. You can practice science as a naturalist without being a metaphysical naturalist — you can believe that God exists and also believe that God is not a hypothesis that science should invoke, because invoking God would short-circuit the scientific process.

Methodological naturalism means that science, by its nature, will never prove God. It cannot, because it has ruled out supernatural explanations by methodological fiat. This is not a weakness of science — it is a strength. The commitment to naturalistic explanation is what has made science so successful. But it does mean that the God question is not a scientific question. It is a philosophical and theological question, and the evidence that bears on it is not the kind of evidence that science, as science, can produce.

This is why the answer to "Can science prove God?" is no, but with an important qualification. Science cannot prove God, but it can provide evidence that a philosopher can use in a philosophical argument for God. The fine-tuning data is scientific data. The argument that uses it is a philosophical argument. Science does not prove God, but it contributes to the conversation, and the conversation — about whether the universe looks more like a designed system or a brute fact — is one that science has made richer, not poorer.

Sources

• Stanford Encyclopedia of Philosophy, "Philosophy of Religion" — https://plato.stanford.edu/entries/philosophy-religion/
• Stanford Encyclopedia of Philosophy, "Science and Religion" — https://plato.stanford.edu/entries/science-religion/
• Stanford Encyclopedia of Philosophy, "Fine-Tuning" — https://plato.stanford.edu/entries/fine-tuning/

Related Topics

• Theism vs Atheism — the broader debate science feeds into.
• Faith vs Reason — whether arguments are needed at all.
• Theism vs Deism — deism's reliance on natural theology.
• Why Is There Suffering? — the scientific and theistic challenge.
• Philosophy of Miracles — where science and divine action meet.