The burden of proof and God

Because I recently discussed a blog post from March 2013, I was wondering what *I* was writing around the time.  So here’s a blog post from that period.  Please note that my opinions from six years ago do not necessarily reflect my current opinions.

One of the more tedious arguments concerning gods is the argument over who has the burden of proof.  Whereas many atheists argue that the theist must first make the argument for the existence of gods, their opponents argue that this is a cop out.  For example, on NY Times:

Contemporary atheists often assert that there is no need for them to provide arguments showing that religious claims are false. Rather, they say, the very lack of good arguments for religious claims provides a solid basis for rejecting them. The case against God is, as they frequently put it, the same as the case against Santa Claus, the Easter Bunny or the Tooth Fairy. This is what we might call the “no-arguments” argument for atheism.

I take the side of atheists; I think theists have the burden of proof.  This is not about giving atheists an unfair advantage in the debate, nor is it about making a “no-arguments” argument.  In fact, I do not believe it is an advantage, fair or otherwise, at all.  It’s simply about who takes which role.

I’m somewhat influenced by the role of burden of proof in law, though law does not necessarily provide a paragon of debate format.  In a court of law, attorneys set out to prove certain facts.  But producing the arguments and evidence costs money, so the law must specify whose role it is to prove the facts, and to what extent they must be proven.  In most criminal cases, the plaintiff must prove “beyond reasonable doubt”, and in most civil cases, the burden of proof is a “preponderance of the evidence”.  If you’re wondering how to interpret that, it depends whether you ask the defendant’s or plaintiff’s attorney.

(Please correct me if I made any error with regard to law.)

In the case of proving God, it is only sensible that the theist has the burden of proof.  The atheist doesn’t know beforehand what particular god or gods the theist believes in, and doesn’t know what arguments to use.  Theists keep on telling us that not all religious people believe the same things, and I wholeheartedly agree.  That’s why it’s the theist’s role to explain what they believe and why.

But this does not confer an advantage to atheists, any more than it confers an advantage to the defendant in a court of law.  It just means that the theist makes the first argument.  If the argument, taken at face value, meets the burden of proof, then it is up to the atheist to counter it.  In other words, the burden of proof shifts to the atheists.  The burden of proof shifts back and forth indefinitely, and does not give an advantage to either side.

By saying theists have the burden of proof, I only mean that they have the initial burden of proof.  It is not meant as a “get out of an argument free” card.


  1. says

    In most criminal cases, the plaintiff must prove “beyond reasonable doubt”,
    Quibble: In criminal cases, the party bringing the action (and bearing the initial burden of proof) is usually termed the prosecutor. Typically, only governments can initiate criminal proceedings. Plaintiff usually refers to the complainant in a civil action.

  2. EnlightenmentLiberal says

    To Larry

    Typically, only governments can initiate criminal proceedings.

    Note that this has not always been this way in America and England.

    To OP
    I think that the phrase “burden of proof” is a misunderstanding as commonly used. When used correctly, “burden of proof” is a fairness doctrine in conversation, and doesn’t have much to do with “what should I believe”.

    The proper place to invoke “the burden of proof” is when a person comes up with a farfetched idea and says “the evidence is easy to find; do the research yourself”. If this was socially permitted, then someone could easily waste another person’s time. The “burden of proof” is a social fairness convention, designed to prevent someone from unfairly placing typically time-wasting demands on other people.

    However, if someone comes to you with a seemingly outlandish claim, and you already know some things relevant to the claim, you shouldn’t ignore that background knowledge just because the other person, the proponent, didn’t bring it to the conversation. The point here is that the “burden of proof” as commonly invoked has very little to do with epistemology itself. Rather, when “deciding” what you should believe, you should look at all of your available evidence and knowledge, and use that and only that to figure out what beliefs are warranted to the appropriate level of confidences. In other words, you should take a Bayesian approach. In light of your current evidence, sometimes the correct personal position will be “I don’t know”, and sometimes it will be “that claim is true”, and sometimes it will be “that claim is false”.

    In other words, saying that one side has this special burden of proof is wrong-headed. In other words, the whole notion of the “null hypothesis” is wrong-headed, or at least overly simplistic. However, if you have a boatload of background evidence for a proposition, then yes, that should be the “default” position, but it’s the default position precisely because of all of that evidence that you already have for the proposition. It’s only this sense that the “burden of proof” makes any sense, but it would be better phrased as “you should take into account all of your available evidence every time, and no proposition enjoys a special privileged position except as indicated because of already known overwhelming evidence”.

    In other words, “the burden of proof” is there to say to someone “I’m pretty sure that your claim is false, and I’m not going to spend my time researching your claim because that would probably be a waste of my time, and I’m not going to let you do that. If you want to bring up specific evidence and citations, then I’m more likely to listen, because that requires spending less of my precious time for your cockamanie idea”.

    The burden of proof in criminal court is a different beast entirely. There is the old legal maxim that it is better to let 10 guilty persons go free than put 1 innocent person in jail. This itself is another fairness doctrine, and we could debate its merits on its own. However, assuming we grant this fairness principle for when we should put people in jail, then the criminal court burden of proof necessarily follows. However, trying to draw parallels to epistemology and an epistemological “burden of proof” ala “null hypothesis” is incredibly wrong-headed. There is no such thing as the null hypothesis per se for whether someone is guilty or not. Rather, we as society have determined that in order to put someone in prison, one must have very strongly compelling evidence. Outside of court, we should evaluate claims of innocence and guilt like any other propositions – we should take into account all of our available general background knowledge, like the frequency of such crimes, etc., and also take into account all of our available evidence that is specific to this instance, i.e. do we have a bloody murder knife found in his hands. Then, based on those evidence, we should choose a position like “I don’t know / I am undecided”, or “it’s slightly more likely that they’re guilty”, or “it’s very likely that they’re guilty”, or “I’m very certain that they’re guilty”, etc. Beliefs and knowledge come in degrees of confidence, and every proposition must be evaluated in the context of competing propositions, e.g. the Bayesian approach.

  3. Matthew Herron says

    To me, it’s largely about who is making a claim. The various religions make claims that a particular god exists, that the god has certain characteristics, that its thoughts can be known, and so on. The burden of proof is appropriately on them to support those claims. Atheism doesn’t necessarily make any claims but rather asks ‘what is the evidence for your god’ and concludes that the evidence is too weak to justify a belief.

  4. bmiller says

    I take the “angry atheist” view: Even if YOUR God is “true,” given the character of said God as outlined in the Bible, prove to me such a deity is worthy of craven worship.


    I know this does not really answer your points, but…

Leave a Reply

Your email address will not be published. Required fields are marked *