Concept

Argument from Ignorance

Intro

"You cannot prove God exists, so God does not exist." That is the argument from ignorance in its most common form. The move treats we have no proof as if it were the same thing as the thing is false. They are not the same.

A simple example shows the gap. Imagine someone tells you they have a coin in their pocket. You cannot see it, weigh it, or test it from where you stand. Does that mean the coin is not there? Of course not. Failing to find evidence is different from finding evidence of failure. You only get to conclude the coin is not there when you have looked carefully in places where the coin should have shown up and it did not.

That last part matters. Sometimes the absence of evidence really is evidence of absence. If you searched your house thoroughly for a missing dog and found nothing, the dog is probably not in the house. Carl Sagan's line catches it: absence of evidence is not evidence of absence, unless you would have found the evidence if it were there.

Christian apologists can fall into this fallacy too, with moves like science cannot explain consciousness, so it must be God. That is the God-of-the-gaps version. The careful work is sorting the genuine fallacy from real burden-of-proof inferences.

The page walks the structure, shows the symmetric atheist and theist versions, lists diagnostic signs, and surveys the legitimate cousins (legal presumption of innocence, scientific null hypotheses) that look like the fallacy but are not.

In full

The informal fallacy of inferring that a proposition is true because it has not been proven false, or, symmetrically, that it is false because it has not been proven true. Latin argumentum ad ignorantiam: "argument from ignorance." Canonical form: "P has not been proven true. Therefore P is false." (Or the symmetric "P has not been proven false. Therefore P is true.") The fallacy lies in treating the absence of proof as if it were proof of the contrary, when in fact the two are distinct epistemic states.

The fallacy was named in the modern logical tradition by John Locke in An Essay Concerning Human Understanding (1690, Book IV, ch. 17, §22), the same chapter that names ad hominem and ad verecundiam. Locke described argumentum ad ignorantiam as a tactic that "drives an opponent to confess he has no argument" without thereby establishing one's own conclusion. Modern formal treatment in Walton Informal Logic (Cambridge 2008); Hurley A Concise Introduction to Logic; Copi-Cohen-McMahon Introduction to Logic. The popular-cultural anchor for the false-fallacy distinction is Carl Sagan's dictum from The Demon-Haunted World (Random House, 1995): "absence of evidence is not evidence of absence", except, importantly, when evidence WOULD have been expected to be found if the claim were true.

In apologetic discourse argument from ignorance is one of the most-deployed atheist conversation-stoppers, "you can't prove God exists, so God doesn't"; "there's no scientific evidence for the soul, therefore no soul"; "miracles can't be replicated under controlled conditions, therefore miracles don't happen". The fallacy can also appear in Christian apologetic when poorly-formed, "science can't explain consciousness, therefore it must be God" (this is also a God of the Gaps form). The careful work is distinguishing legitimate burden-of-proof and absence-as-evidence inferences (the false-fallacy examples below) from genuinely fallacious appeals to ignorance (the actually-fallacious cases).

Canonical structure

Two symmetric forms:

Negative form (atheist deployment most common)

P1: P has not been proven true (no positive evidence found / cannot be demonstrated)
C: Therefore P is false (or unwarranted)

Positive form (theist deployment when poorly formed)

P1: P has not been proven false (no positive disproof found)
C: Therefore P is true

Both fallacious moves treat absence of one kind of proof as if it constituted affirmative proof of the contrary. The fallacy: lack of proof for X is not the same as proof of not-X. Many propositions are not yet proven in either direction; not-yet-proven is not the same as proven-false.

How to spot it (diagnostic)

The argument appeals to lack of proof rather than to positive evidence. Listen for "you can't prove" / "there's no evidence for" / "it has never been demonstrated", these signal an absence-based inference.
The conclusion treats absence as proof. The leap from "P is unproven" to "P is false" (or vice versa) is the fallacy; positive evidence for not-P would be required to establish P is false.
The argument doesn't engage why the absence of evidence might exist. Genuine inferences from absence require asking whether evidence WOULD have been found if the claim were true. If yes, absence is meaningful; if no, absence is not.
The burden-of-proof framing is asymmetric. The arguer typically requires opponent to prove their case while exempting their own from the same standard.
Conflation of "no evidence found" with "evidence of absence." Sagan's dictum: these are different. Found-no-evidence is consistent with both "doesn't exist" and "exists but undetected"; only when evidence WOULD have been expected does absence become evidence-of-absence.

Common apologetic deployment

Atheist deployment against Christianity

"You can't prove God exists, therefore God doesn't exist." Standard New-Atheist trope. The fallacy: lack of proof for God's existence (granted for the sake of argument) doesn't establish God's non-existence. Many propositions are unproven without thereby being false. Treated in God of the Gaps (which engages the related "no scientific gap requiring God, therefore no God" form) + Faith and Reason (engaging the Plantingian warrant framework).
"There's no scientific evidence for the soul, therefore no soul." Conflates "no scientific evidence" with "no evidence at all", phenomenological + introspective + philosophical-of-mind evidence is not scientific in the empirical-laboratory sense but is evidence nonetheless. The argument also assumes the soul WOULD be detectable by empirical-scientific methods if it existed (the question-begging assumption, see Begging the Question).
"Miracles can't be replicated under controlled conditions, therefore miracles don't happen." The argument-from-non-replicability: assumes that if miracles happened, they would be replicable on demand. But miracles by definition are not natural-law-bound events; the absence of laboratory replication is what we'd EXPECT for the miraculous, not evidence against. (See Argument from Miracles.)
"Science has explained X / Y / Z without invoking God; therefore God isn't needed for the remaining unexplained things." Compound: argument-from-ignorance + induction-from-specific-cases-to-general-claim + the God of the Gaps structural pattern. Even if science explains many things naturalistically, that doesn't entail it explains all things naturalistically; and explanation-by-naturalism doesn't entail no-need-for-theism (theistic-evolution, methodological-naturalism, classical-theism's compatibility with all naturalistic causation).
"There's no evidence for an afterlife, therefore no afterlife." Often combined with materialism: assumes that if afterlife existed, it would be empirically detectable from this side of death. The premise is itself non-empirical (it's a metaphysical claim about what kind of evidence the afterlife would produce).
"You can't prove the resurrection happened, therefore it didn't." The atheist position requires affirmatively explaining the evidence (empty tomb + appearances + transformation of disciples + early creeds) without the resurrection; "we don't know what happened, but it wasn't the resurrection" is argument-from-ignorance about the alternative + assumption-of-naturalism rather than positive engagement.
"You haven't shown me a Christian who has lived perfectly, therefore Christianity isn't transformative." Conflates "Christians can be inconsistent / sinful" with "Christianity has no transforming effect", the standard isn't perfection but the directionality of moral / spiritual growth. Also engages No True Scotsman Fallacy dynamics.

Christian counter-deployment

The Christian apologist needs to check their own argument-from-ignorance deployment:

"Science can't explain consciousness, therefore consciousness must be from God." This is the God-of-the-gaps form of argument-from-ignorance, fallacious as deployed. The substantive Christian engagement is via positive arguments (the Hard Problem of Consciousness as positive challenge to reductive materialism + the Phenomenal Concept Strategy + property-dualism) rather than gap-reasoning. (See God of the Gaps / Methodological Naturalism.)
"You can't prove naturalism is true, therefore Christianity is true." Doesn't follow logically. The Christian apologetic must be on positive evidence for Christianity, not on negative evidence against naturalism alone.
"Atheism can't ground morality, therefore Christianity must be true." Better form: "Atheism can't ground objective morality. Some grounding of objective morality is required. Christianity provides such grounding. Therefore Christianity is the better explanation." This is positive Moral Argument (see Moral Argument) rather than argument-from-ignorance.

How to rebut it

1. Distinguish absence-of-evidence from evidence-of-absence (Sagan's distinction)

Carl Sagan's The Demon-Haunted World (Random House, 1995) gave the popular-cultural articulation: "absence of evidence is not evidence of absence", except when evidence WOULD have been expected if the claim were true. The diagnostic test:

If the claim, were it true, would produce easily-detectable evidence, AND careful search has produced no evidence → absence IS evidence of absence.
If the claim, were it true, would NOT necessarily produce easily-detectable evidence → absence is not evidence of absence; it's just absence.

Apply this to "no evidence for God" / "no evidence for the soul" / "no evidence for miracles", would these claims, were they true, produce the kind of evidence the atheist demands? Often the answer is no (God isn't a physical object whose existence would be detected by physical instruments; the soul isn't a measurable substance; miracles are by definition not natural-law-bound replicable events). The "no evidence found" is consistent with "exists but undetectable by the demanded methodology" rather than "doesn't exist."

2. Distribute the burden of proof correctly

The standard epistemic principle: the burden of proof falls on the affirmative claimant. Saying "I cannot prove not-P" is not "P is true." Saying "I cannot prove P" is not "P is false." Both atheist ("there is no God") and theist ("God exists") positions are claims that bear burden of proof. The proper response to the atheist's "you can't prove God" is: "I'm offering positive evidence, the cosmological argument, fine-tuning, moral argument, historical resurrection. Engage these. And what's YOUR positive evidence for atheism, the metaphysical claim that there is no God?"

3. Inference-to-best-explanation is not argument-from-ignorance

The major theistic arguments take IBE form, not gap-reasoning form. The Kalam, fine-tuning, moral, ontological, transcendental, resurrection-historical arguments are positive inferences from what we DO know (the universe began to exist; constants are precision-tuned; moral truths exist; etc.). These are not arguments from ignorance. The atheist who charges "argument from ignorance" against these arguments is misidentifying their inferential structure (see God of the Gaps for the full IBE-vs-gap-reasoning distinction).

False-fallacy examples

Cases where what looks like argument from ignorance is NOT actually fallacious, the inference from absence is legitimate because evidence WOULD have been expected.

Absence-of-evidence when evidence would have been expected. "We've searched the room thoroughly and found no elephant. Therefore there's no elephant in the room." The room was thoroughly searched; an elephant would be detectable; absence is genuinely meaningful. Why this isn't argument from ignorance: the conditional "if there were an elephant, we'd see it" is independently established. Absence here IS evidence of absence.
Presumption-of-innocence in legal contexts. "Not proven guilty beyond reasonable doubt" → "treat as not guilty for legal purposes." Why this isn't argument from ignorance: this is a procedural-legal-rule about how the state should act, not an inferential claim about whether the defendant actually committed the crime. The legal presumption is a decision rule, not a metaphysical claim.
Default-positions in inquiry when one position has positive evidence and the other doesn't. When evidence for P is robust and evidence against P is absent, the rational default IS to accept P. Why this isn't argument from ignorance: the inference rests on positive evidence for P, not merely on absence of evidence against P. The "absence of disproof" is a supplementary consideration, not the load-bearing element.
Russell's teapot, when used correctly. Bertrand Russell's analogy: an undetectable teapot orbiting between Earth and Mars cannot be disproven, but the burden of proof falls on the claimant. Why this isn't argument from ignorance (when properly deployed): the analogy is about burden-of-proof distribution + the asymmetry between positive evidence and unfalsifiable claims. The claim "I cannot disprove the teapot, therefore the teapot doesn't exist" would be argument from ignorance; the claim "the burden of proof rests on the teapot-affirmer to provide positive evidence" is correct burden-of-proof distribution.
Scientific reasoning from negative results. "We searched for X using methodology Y; we did not find X. Methodology Y is sensitive enough to detect X if X exists. Therefore X probably doesn't exist (within the limits of Y's sensitivity)." Why this isn't argument from ignorance: the inference rests on the established sensitivity of the detection methodology; the absence is meaningful given the methodology can detect what's being searched for.
Plantinga's Reformed Epistemology + properly basic belief. Properly basic beliefs don't require inferential support, and the absence of inferential support is not an argument against them. Why this isn't argument from ignorance: the position is about warrant-conditions for properly basic beliefs (the sensus divinitatis, perceptual beliefs, memory beliefs), not about absence-as-evidence. (See Reformed Epistemology.)
Inference-to-best-explanation when absence-of-evidence-for-alternatives is part of the inference. When the inference is "the best explanation of E is H1, given that no good evidence supports H2 / H3 / H4," the absence of evidence for alternatives is a legitimate part of the IBE. Why this isn't argument from ignorance: the inference rests on positive evidence for H1 + evaluation of alternative hypotheses, not on absence alone.

The diagnostic test: does the claim, were it true, produce easily-detectable evidence that careful search has not produced? If yes, absence IS evidence. If no, absence is just absence.

When it's actually fallacious

Clear cases where the argument-from-ignorance charge sticks:

"You can't prove God exists, therefore God doesn't exist." Lack of proof for God's existence (even granting for argument's sake that there's no proof, a contested premise) doesn't establish God's non-existence. The Christian apologetic has positive evidence (cosmological / fine-tuning / moral / ontological / resurrection). The atheist requires positive evidence for atheism, not just absence-of-proof-for-theism.
"There's no scientific evidence for the soul, therefore no soul." Conflates "no scientific evidence" with "no evidence at all" + assumes the soul WOULD be empirically-scientifically detectable if it existed (the question-begging assumption).
"Miracles can't be replicated under controlled conditions, therefore miracles don't happen." Assumes miracles WOULD be replicable on demand if they happened, but miracles by definition aren't natural-law-bound replicable events. The non-replicability is what we'd EXPECT, not evidence against.
"You can't prove the resurrection happened, therefore it didn't." Argument from ignorance about the resurrection's evidential status + assumption-of-naturalism rather than positive engagement with the historical-evidence. The historical-Jesus + minimal-facts + Bauckham eyewitness-testimony evidence is positive evidence for the resurrection that needs engagement.
"Christian-counter-instance: science can't explain consciousness, therefore God." Argument from ignorance + God of the Gaps form. Substantive Christian engagement is via positive arguments (Hard Problem of Consciousness as positive challenge to reductive materialism), not via gap-reasoning.
"Atheism is true because no theism has been proven true." Argument from ignorance applied to theism in general. Even if no specific theism could be conclusively proven, that doesn't establish atheism; agnosticism would be the appropriate position if argument from ignorance were the only consideration.
"There's no evidence for an afterlife, therefore no afterlife." Assumes afterlife WOULD be empirically detectable from this side of death. The premise is itself non-empirical.

Christian scholarly resources

John Locke, An Essay Concerning Human Understanding (1690), Book IV, ch. 17, §22. Original modern naming of argumentum ad ignorantiam alongside ad hominem and ad verecundiam.
Douglas Walton, Informal Logic 2nd ed. (Cambridge, 2008). Standard taxonomic treatment.
Patrick Hurley, A Concise Introduction to Logic (Cengage, multiple eds.). Textbook treatment.
Irving Copi, Carl Cohen, & Kenneth McMahon, Introduction to Logic (Routledge, 14th ed.). Alternate canonical textbook.
Carl Sagan, The Demon-Haunted World: Science as a Candle in the Dark (Random House, 1995). The popular-cultural anchor for "absence of evidence is not evidence of absence", though Sagan also engages cases where absence IS evidence (the Russell's teapot / dragon-in-my-garage examples).
Bertrand Russell, "Is There a God?" (1952, commissioned by Illustrated Magazine but unpublished until 1997 in Last Philosophical Testament). The teapot analogy: about burden-of-proof distribution, not argument from ignorance per se.
Norman Geisler & Ronald Brooks, Come, Let Us Reason: An Introduction to Logical Thinking (Baker, 1990). Christian-apologetic logic primer.
Alvin Plantinga, Warranted Christian Belief (Oxford, 2000). Engages the de jure objection's burden-of-proof structure and defends properly basic belief against argument-from-ignorance charges.
William Lane Craig, Reasonable Faith 3rd ed. (Crossway, 2008). Engages the burden-of-proof in the existence-of-God debate.
John Lennox, Gunning for God: Why the New Atheists Are Missing the Target (Lion, 2011). Engages New-Atheist argument-from-ignorance moves.
Peter Lipton, Inference to the Best Explanation 2nd ed. (Routledge, 2004). The IBE methodology that distinguishes legitimate inference from gap-reasoning.