Burden of Proof: Who Has to Prove What?
"Prove to me that God doesn't exist." "You can't disprove that crystals have healing energy." "Show me evidence that my homeopathic remedy doesn't work." In each of these challenges, something has been flipped. The person making a positive claim — that God exists, that crystals heal, that homeopathy works — is demanding that sceptics disprove it rather than offering evidence in support of it. This inversion is the burden of proof fallacy, and it is one of the most effective rhetorical moves in the arsenal of motivated reasoning.
What Is the Burden of Proof?
The burden of proof (from the Latin onus probandi) is the obligation to provide evidence or reasoning sufficient to justify a claim. The fundamental rule is simple: whoever makes a claim bears the burden of demonstrating that it is true — or at least plausible. This rule underlies scientific method, legal standards of evidence, and everyday rational discourse.
The burden of proof fallacy occurs when someone making a claim (the positive claim-maker) successfully shifts the obligation of proof to their interlocutor, demanding that the other party disprove the claim rather than offering evidence for it themselves. The lack of disproof is then treated as confirmation.
The Latin Principle
The classical formulation is: "Onus probandi incumbit ei qui dicit, non ei qui negat" — the burden of proof rests on the one who asserts, not the one who denies. This principle is ancient, appearing in Roman law and classical rhetoric. Its modern applications span courtrooms, scientific journals, philosophical debate, and media fact-checking.
In law, the principle has been codified into specific standards: criminal courts generally require proof "beyond reasonable doubt"; civil courts often use "balance of probabilities." Both standards place the burden on the party making the positive claim — typically the prosecution or plaintiff.
Russell's Teapot: The Classic Illustration
Philosopher Bertrand Russell formulated the most memorable illustration in 1952. He proposed imagining that he claimed a small teapot was orbiting the Sun somewhere between Earth and Mars — too small to be detected by any telescope. If Russell asserted this and demanded that others disprove it, their inability to do so would not, Russell argued, entitle him to expect human belief. The burden of proof for the teapot claim rests with Russell, not with those who are sceptical of it.
The parable was aimed at religious apologetics: the claim that God exists is a positive claim, and the appropriate response to it is to examine the evidence offered for it — not to demand that atheists prove a negative. Russell's teapot has since been extended to countless analogous cases: invisible dragons in garages, orbiting spaghetti monsters, and other unfalsifiable entities specifically designed to illustrate the structure.
Hitchens's Razor
The journalist Christopher Hitchens formulated what has become known as Hitchens's Razor: "What can be asserted without evidence can be dismissed without evidence." This is a compressed statement of the burden of proof principle. If no evidence is offered for a claim, no disproof is required to dismiss it — the claim simply doesn't rise to the level of requiring a response.
This principle is particularly useful in debates with a high density of unsupported assertions. Each assertion does not create an obligation to disprove it; the obligation runs in the other direction.
Common Forms of the Fallacy
The Unfalsifiable Claim
"There is an invisible presence in this room that affects your emotions." When asked for evidence, the response is: "Prove there isn't." The claim has been structured so that no evidence could possibly refute it, and the burden has been placed on sceptics to disprove what cannot by definition be disproved.
Conspiracy Theories
Conspiracy theories routinely deploy burden-shifting. "The moon landing was faked." "Show me evidence it wasn't." The conspiracy claim is made without evidence, but any evidence produced against it (footage, technical documentation, independent confirmations) can be dismissed as "part of the cover-up" — which means the theory is structured to be unfalsifiable, and sceptics are perennially on the defensive.
Pseudoscience
"This supplement improves cognitive function." "Prove it doesn't." The supplement manufacturer has made a specific claim that a substance has an effect. The burden rests with them to demonstrate that effect — through controlled trials, replication, peer review. Demanding that sceptics prove a negative is an attempt to bypass this entire evidential structure.
Shifting the Default
A subtler form involves establishing an undefended default assumption and treating deviation from it as what needs justification. "Science has been wrong before — why do you trust it?" The implied burden is on the person who trusts scientific consensus to justify that trust at every juncture. But this ignores that the alternative — rejecting scientific consensus — is also a positive position requiring justification.
Negative Claims and the Asymmetry Myth
A common confusion involves the idea that "you can't prove a negative" — which is taken to mean that negative claims automatically lack a burden of proof. This is incorrect on several grounds.
First, many negative claims are perfectly provable: "There are no elephants in this room" can be verified by inspection. "No human has run a marathon in under one minute" is supported by extensive evidence.
Second, most apparently negative claims can be restated as positive ones. "There is no God" is logically equivalent to "the universe is godless," which is as much a positive claim as its negation — and carries the same burden of justification.
Third, the asymmetry between positive and negative claims is a matter of degree, not kind. Extraordinary claims require extraordinary evidence; modest claims require modest evidence. The question is always proportionality, not a blanket exemption for one polarity.
In Scientific Method
The burden of proof principle is institutionalised in scientific practice. The null hypothesis — the assumption that a proposed effect does not exist — functions as the default position that must be overcome with evidence. A new drug must demonstrate efficacy in controlled trials; the burden does not rest with sceptics to prove it doesn't work. This is why double-blind controlled studies were developed: to create a rigorous evidential standard that appropriately places the burden on those claiming an effect.
When scientific consensus is challenged, the challenger bears the burden of producing evidence that meets the same standards. This is not intellectual conservatism — it's the mechanism that has allowed science to progressively improve its models of reality by holding new claims to high standards before accepting them.
Relationship to Other Fallacies
The burden of proof fallacy is closely related to Ad Hominem in its effect: it shifts the conversation away from the merits of a claim by making sceptics responsible for disproving it. It also connects to Tu Quoque as a diversionary tactic — both deflect rather than address. And when a claim is structured specifically to be unfalsifiable, it often involves the Circular Reasoning pattern of protecting a conclusion from any possible disconfirmation.
How to Respond
When you encounter a burden-shifting argument:
- Identify the positive claim. What has actually been asserted? Who made the assertion?
- Return the burden. "You've made a claim. What evidence do you have for it?"
- Apply Hitchens's Razor when appropriate. "You've made this claim without evidence. I'm comfortable dismissing it without evidence until you provide some."
- Avoid defending a negative indefinitely. You don't have a general obligation to disprove every unsubstantiated claim. Acknowledging that you can't currently disprove X is not the same as acknowledging that X is likely true.
- Ask about falsifiability. "Under what conditions would you change your mind?" If no conditions exist, the claim is unfalsifiable and the burden of proof structure becomes irrelevant.
Summary
The burden of proof is not an arbitrary procedural rule — it is the foundation of rational inquiry. Without it, any assertion becomes defensible simply by challenging others to disprove it, and the production of evidence becomes unnecessary. Keeping track of who made what claim, and ensuring that claims are accompanied by evidence proportional to their strength, is one of the most basic and most important habits of rigorous thinking.
See also: Ad Hominem, Tu Quoque, Circular Reasoning
Sources & Further Reading
- Russell, Bertrand. "Is There a God?" (1952). Reprinted in The Collected Papers of Bertrand Russell, vol. 11.
- Hitchens, Christopher. God Is Not Great. Twelve Books, 2007.
- Wikipedia: Burden of proof (philosophy)
- Effectiviology: The Burden of Proof
- Logical Fallacies: Burden Of Proof