You have heard the claim: "There is no evidence for God." And you have heard the counter: "The universe itself is evidence for God." These two positions cannot both be right, but they can both be making a mistake about what evidence is. Evidence is not a thing that exists independently of a theory. Evidence is data interpreted relative to competing hypotheses. And the interpretation depends on what kind of explanation you are willing to consider. Before you can say "there is no evidence," you must first answer: evidence for what? Under what framework? Against which alternative? The answer to those questions is not scientific. It is philosophical. And it determines everything.

I. Evidence Is Theory-Relative

In formal philosophy of science, evidence E supports hypothesis H over alternative H' if the probability of observing E is higher given H than given H'. This is Bayesian confirmation theory, and it is the most rigorous framework currently available for thinking about what evidence means.

Notice what this requires: you must specify the alternative hypothesis. Evidence does not float free of context. The same observation can simultaneously support one hypothesis and disconfirm another, and a given dataset will point in different directions depending on what theories are in competition. This is not relativism. It is precision.

Applied to the existence of God: the fine-tuned constants of the universe constitute evidence — but evidence for what, against what? Against an eternal, uncaused, self-sufficient physical universe: the fine-tuned constants are extremely improbable. Against a universe created by a rational, information-rich God: the fine-tuned constants are precisely what you would expect. The data is the same. The explanatory power of the competing hypotheses differs. This is a legitimate scientific and philosophical question, not a question that can be dismissed by saying "there is no evidence."

"It is not the case that evidence for theism can be dismissed as non-evidence simply by assuming naturalism is true. That would be question-begging. The question is precisely which framework better accommodates the evidence." — Richard Swinburne, The Existence of God, 2nd ed. 2004

II. The Double Standard Problem

Here is the operative double standard that runs through most popular atheist discourse on evidence: the same type of reasoning that is accepted as legitimate in every other domain of inquiry is declared illegitimate when it points toward design or agency.

Forensic science regularly infers agency from physical evidence. A bullet trajectory, a blood pattern, a fiber transfer: these are physical facts that are interpreted as evidence of human action. No one objects that this inference is "unscientific." SETI research (the Search for Extra-Terrestrial Intelligence) is premised on the assumption that a sufficiently complex, non-random radio signal would be recognized as evidence of intelligent origin. No one objects that this inference is "religious."

But when a molecular biologist notes that the genetic code is a digital information system with properties identical to known designed systems, and infers an intelligent source, the objection is immediate: "You can't invoke design in science." The double standard is not an argument. It is a reflex. And reflexes are not evidence.

III. The Demand for Direct Evidence

A common form of the "no evidence" claim is the demand for direct, unambiguous, repeatable evidence — the kind produced in a controlled laboratory experiment. If God exists and wishes to be known, why doesn't He show up in a reproducible, falsifiable, observable way?

This demand confuses the type of evidence appropriate to the question being asked. No one demands laboratory-reproducible evidence for historical claims. The existence of Julius Caesar is not established by repeatable experiment. It is established by convergent testimony, documentary evidence, archaeological corroboration, and coherent historical narrative. The Resurrection is a historical claim, not a laboratory claim. It requires historical investigation, not experimental protocol.

More fundamentally: if God is the Creator and Sustainer of the physical universe — if He is, in the language of classical theism, the ground of being rather than an object within being — then the demand that He produce laboratory-reproducible evidence of His existence is like demanding that the mathematician appear inside the equation. The Author does not appear within the text as a character. He appears outside it. And the evidence for the Author is in the nature of the text — its structure, its information content, its fine-tuning, its narrative coherence, and the historical record of the moment He broke into it in person.

500+
Witnesses claimed in 1 Corinthians 15:6 to have seen the risen Christ — written within 25 years of the events by a man who explicitly states that most of the witnesses were still living when he wrote, implying they could be questioned. By the standards of ancient historical evidence, this is not weak testimony. It is extraordinarily strong. The question is not whether the evidence exists. It is whether you are willing to evaluate it on its own terms. — Habermas, G.R. & Licona, M.R. (2004). The Case for the Resurrection of Jesus. Kregel.

IV. What Would Falsify Theism?

The challenge "theism is unfalsifiable" is often leveled without considering whether the challenger can specify what would falsify it. Richard Swinburne, John Polkinghorne, and Alvin Plantinga have all argued that theism makes specific empirical predictions — and that those predictions are borne out by the evidence:

A universe created by a rational God should exhibit rational order. Check — and Wigner noticed the astonishment of this. It should have the kind of physical constants that permit complexity and life. Check — and cosmologists have noticed the extraordinary precision of this. It should contain some form of public historical disclosure — a moment where the Creator communicates clearly with His creation. Check — and the historical record of Jesus and the Resurrection is the most documented, most investigated public event in ancient history.

What would falsify theism? A coherent naturalist account of the origin of the universe from literally nothing. A naturalist account of the origin of the genetic code. A naturalist account of the origin of consciousness. A refutation of the Resurrection through a plausible historical alternative. None of these exist. The falsifiers have not been produced. The demand for evidence is legitimate. But the demand must apply symmetrically — and the naturalist position has outstanding debts it has not paid in seventy years of trying.

"Unbelief is not the result of honest investigation. Unbelief is the result of an inconsistent and illogical set of qualities ascribed to the divine — a God who has been defined in advance to be the kind of God who couldn't possibly exist." — Daniel K. Hedrick, GODISNOWHERE.org

The following sources constitute the primary intellectual foundations for reviewing and preparing for this kind of argument.

  • Swinburne, R. (2004). The Existence of God (2nd ed.). Oxford University Press. The most rigorous philosophical argument for theism using Bayesian probability theory. Swinburne applies formal evidence theory to the question and concludes theism is more probable than naturalism given the total evidence. Search this source ↗
  • Habermas, G.R. & Licona, M.R. (2004). The Case for the Resurrection of Jesus. Kregel. Evidence assessment for the Resurrection using the same historical criteria applied to any ancient event. The "Minimal Facts" approach — arguing from facts accepted even by skeptical scholars. Search this source ↗
  • Dembski, W. (1998). The Design Inference: Eliminating Chance through Small Probabilities. Cambridge University Press. Formal framework for inferring design from specified complexity. Published by Cambridge University Press and peer-reviewed — whatever one thinks of Dembski's conclusions, the method deserves engagement, not dismissal. Search this source ↗
  • Hacking, I. (1965). Logic of Statistical Inference. Cambridge University Press. Philosophical foundations of statistical inference — essential background for understanding what it means to say that evidence supports one hypothesis over another. Search this source ↗

Where Does This Argument Lead You?

Select the conclusion that most honestly fits your assessment.