The Burden of Proof

The phrase “burden of proof” has to do with who it is who needs to back up his claims with evidence; the one who needs to prove what he is saying. The person who has the burden of proof is the one making a controversial claim. If the claim is noncontroversial, no argument, evidence, reasons, or proof, is necessary to have the claim accepted. Where no proof is necessary, there is no burden of proof. The person questioning what is normally considered a noncontroversial claim then has the burden of proof because he is making the controversial claim that the accepted wisdom is wrong. 1Fluoride is added to almost all toothpaste because it has been accepted as a scientific fact that fluoride is good for teeth. If someone disputes this claim, he has the burden of proof to counter a noncontroversial claim, whereas someone asserting that fluoride is good for teeth has no need to marshal evidence for his assertion. When Copernicus claimed that the earth orbits the sun, the burden of proof was on him, because that was not accepted wisdom at the time. It certainly looked as though the sun orbits the earth. Having the burden of proof in no way suggests that a person is wrong. Who has the burden of proof is just a matter of logic; not of who is right and who is wrong.The reason the person making a controversial claim has the burden of proof is that evidence and reasons will need to be presented before anyone has a rational reason for believing him. It is not rational to believe unsubstantiated claims with no evidence. If someone claims your car will not start today, but gives you no reason to believe him, then you have good reason to doubt his claim.

It is important to note that if the burden of proof has been met; the person making the controversial claim has presented strong supporting evidence, then if someone still disagrees with the conclusion being argued for it is not sufficient to just say “I disagree.” Once someone has discharged the burden of proof, anyone disputing the well-supported claim now has the burden of proof on him. The person disputing the conclusion of anyone who has met the burden of proof must point out what was wrong with the evidence presented, or point out some piece of opposing evidence or consideration that was not mentioned. For instance, introduction to philosophy students in my classes will read a long article arguing that there is a rather high degree of agreement between people as to what counts as beautiful, especially when it comes to human beauty. Lots and lots of studies are cited and ordinary observations that everyone is familiar with are appealed to. For instance, people tend to date people who resemble them, including in terms of physical beauty. This implies that most people know approximately how attractive they are, which they do.  Very unattractive people do not usually plan to become a model.  One particular student in an online discussion comment wrote: “I disagree because assessments of beauty are entirely subjective and everyone differs.” That opinion is absolutely fine, but only if the student shows what was wrong with pages and pages of evidence and reasons provided to the contrary. The burden of proof will be very high indeed on the student. In such circumstances, the student might be better off taking a single scientific study or commonplace observation, or a very modest selection of the studies and observations, and showing why they are incorrect or inconclusive, or what things had been overlooked and were missing from the analysis.

Complicating matters is the fact that the burden of proof is context dependent. If a mathematician at a conference refers to an equation famous to other professional mathematicians and relies on this equation, he has no burden of proof because, in that context, the truth of the equation is noncontroversial. If, however, that same mathematician asserts that an equation is valid to people unfamiliar with the equation he will have to do a proof, proving the two sides are in fact equal to each other. An equation is2 saying A = B; that, for instance, 2 x 3 equals 6. Giving me 6 balls is the same as giving me 2 x 3 balls. The left side of an equation is equivalent to the right side of an equation. E = MC2 is saying that energy is mass times the speed of light squared. Energy (E) and mass (M) are interchangeable. Likewise, most people who attend church agree that God exists. For them, the existence of God is a more or less settled matter and uncontroversial. That means that the priest can treat the existence of God as a given; as a fact, and does not need to provide evidence for God’s existence. If, however, a philosophy professor, for instance, makes the claim that “God exists” then, since some of his students probably do not agree, he is rationally required to provide evidence for God’s existence. Note that providing evidence and discharging the burden of proof does not need to be definitive. The professor makes the claim of God’s existence, and provides evidence. For the moment, his job is done. If a student wishes to object, he or she must show that the evidence is either incorrect, or that it omits some other important consideration. It is the student who has the burden of proof. Once the student has made his case, it is now up to the professor to show that there is some deficiency in the evidence presented or some important consideration has been omitted. So, the burden of proof passes back and forth like a ball in a tennis match.

3It should be mentioned that there is an argumentative technique that everyone needs to use at times that bypasses the burden of proof for one stage of the argument, and that is when we agree to a questionable, unproven assertion “for argument’s sake.” Here we are agreeing to a counterfactual, or at least something that has not yet been shown to be factual. So, for instance, a philosopher might assert that the existence of eternal hell is inconsistent with the notion of a forgiving, loving, God. In other words, if God exists, which we will accept for argument’s sake, i.e., for the moment, then eternal hell as a place of damnation by God, does not exist. Someone might agree that this is true as a counterfactual even if he does not believe in God. The burden of proof is still being discharged, just later in the argument. The arguer is asking for hypothetical, not actual, acceptance of the first premise, to try to prove a point. A more prosaic example would be a situation where a roommate says she wants you to leave because you left the top off the toothpaste once. You deny that it was you who left the top off and then say “even if I did leave the top off the toothpaste, asking me to leave for that reason is an overreaction.” Here you accept, for the moment, for argument’s sake, that you did something, in order to argue that even if it were true, the suggested course of action would be unwarranted.

Returning to the burden of proof, in a criminal trial, there is a logical requirement that there be an “assumption of innocence.”  The prosecutor is making the controversial claim that the defendant is guilty. This is not common knowledge known to all. Thus, the prosecutor must prove his claim by offering good evidence for it. Lots of rules of evidence have been developed over the last two thousand years to try to make sure that the evidence is legitimate and veridical (true). At one point in time mere rumors were accepted as evidence. Since anyone at all can start a rumor and rumors need have no connection with reality, accepting them as “evidence” was a travesty of justice. Due process also involves knowing exactly what the charges are against you, who is making the charges, and being given an opportunity to cross-examine the person making the claims to assess his credibility. It will not be possible to respond to charges unless the defendant knows what they are. “Hearsay” is not permitted. If someone testifies “I heard Sue say that John is a murderer” the court needs to hear from Sue. Sue may have been joking, exaggerating; she may have based her claim on flimsy evidence, or no evidence, or the person reporting this might have misheard. The court needs to find out from Sue her evidence for thinking John is a murderer in order to evaluate this evidence.

The other reason for the assumption of innocence and the requirement that prosecutors present their evidence, rather than assuming defendants are guilty and having them prove their innocence is summed up by the phrase “you can’t prove a negative.” What this means is that it is usually impossible to prove that you did not do something. I cannot prove that I have never murdered anyone or never had a secret affair. If the burden of proof is put on the wrong person, then all people will be found guilty of crimes whether they are innocent or not.  If I can be found guilty of murder if I cannot prove that I have never murdered anyone, then I will be illegitimately convicted for sure. This is why we must start with the assumption of innocence and then try to prove guilt. This is to ensure the correct person has the burden of proof. It is crucial to get this right. The assumption of innocence is an absolute logical necessity if the justice system is to work at all in terms of justly convicting or acquitting defendants.


Technically, it is usually never possible to prove that someone is innocent; only that the person is not guilty.  A finding of “not guilty” in a court of law does not mean the defendant is for sure innocent of the crime he is accused of – it only means insufficient evidence has been provided for thinking he did do it. This does not mean we should continue to harbor suspicions about the defendant just because he cannot prove he is innocent because that is typically just not possible.

The prosecutor presents his case – the evidence – and the defendant can proceed to poke holes in the case to demonstrate that it is not sufficient or that, for example, he was out of the country at the time of the murder, etc. If the prosecutor were simply to make an accusation without evidence such as asserting “You are a murderer!” all defendants would be unable to defend themselves because to murder someone can take just a minute or two depending on method and none of us can account for our actions and whereabouts every minute of every day. 

Sometimes a defendant will refuse to “take the stand,” testify on his own behalf, and “answer the charges.” This is sometimes treated as suspicious in itself. However, the defendant does not need to “defend himself” or to vouch for his good character. It is instead up to the prosecutor to prove his case since it is he who has the burden of proof. Besides, years of not killing anyone, or being generally friendly and well-respected, is not evidence that someone did not kill someone and then went back to his usual routine.  It is a nasty fact that defendants who are good looking, male and female, appear to be nicer and more trustworthy than ugly people. Jurors are unaware that looks are motivating their assessment but it affects their assessment anyway and some people are going to seem more likable than others on the stand by quirk of personality. Some people have a cold manner and are otherwise off-putting. At the extreme end, there are people so charismatic and “fun,” being in their company is like being at the most enjoyable party you have ever attended. I have known two people like this. Students flocked to the classes of one of them. But there is no point telling anyone – be like that guy! Neither for purposes of teaching, nor in being a defendant in court.

5 thoughts on “The Burden of Proof

  1. The burden of proof is totally on the one making a claim without evidence that a certain claim of someone else is a controversial claim. All claims are presumed innocent and non-controversial until proven to be controversial by the one claiming without evidence that the other man’s claim is too controversial and requires evidence. That statement itself requires evidence and cannot be accepted face-value without any proof/evidence.

    • scottrobertharrington – You write: “All claims are presumed innocent and non-controversial until proven to be controversial.” I’m afraid that is not true. I have given the example of Copernicus. He really did need to provide evidence for his views. Science can’t progress without reasons being given for new contentions. If you have in mind politically motivated assertions of controversiality, then you are dealing with an evilly motivated liar and fraud who is not in the least interested in rational discussion and with whom further interaction is pointless. E.g., anyone who claims that the existence of biological sex is controversial is a liar or mentally challenged – or possibly both.

      • How can we be sure that a proposition is true?

        Take a simple example. Sometimes, when they can’t find something, one hears people say, “It can’t have simply vanished.” It is something they take for granted.

        But how do they know this? It is not a logical impossibility that things should simply vanish; we can quite easily imagine something just disappearing and re-appearing, like the Cheshire Cat in Alice in Wonderland.

        This suggests “things don’t simply vanish” is an empirical or scientific proposition, an hypothesis that needs to be tested. But there is an obvious difficulty here; accepting even the possibility of things “simply vanishing” has all sorts of implications for what would count as testing, proving, how we interpret evidence and our whole system of verification.

        The solution is simple enough: “things don’t simply vanish” is a rule, a practice, a regulative principle for the way we judge and act, from the most rigorous scientific enquires to the most ordinary everyday activities. Regarding it as absolutely solid is part of our method of doubt and enquiry.

        To treat it as an empirical proposition would be a grammatical error; it would show we do not understand the kind of proposition we are dealing with.

  2. Pingback: The Burden of Proof | Reaction Times

  3. “A finding of “not guilty” in a court of law does not mean the defendant is for sure innocent of the crime he is accused of – it only means insufficient evidence has been provided for thinking he did do it.”
    An excellent illustration of this is the case of Creasey v Creasey 1931 S.C. 9. It was a divorce action, at a time where the criminal standard of proof applied in consistorial cases: proof beyond reasonable doubt and on corroborated evidence.
    HELD “The confessions of the wife, defender, may warrant the Court in finding that adultery is proved against her, while, not being evidence against the co-defender, he escapes; and thus divorce may be granted against the wife for adultery committed by her with him, while he himself is assoilzied from the action.”
    In Creasy her (extra-judicial) admissions were corroborated by evidence of clandestine association; however, as against the co-defender, this was uncorroborated, as it came from a single witness, a waitress at the tea-rooms they frequented.


Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.