• Strange Notions Strange Notions Strange Notions

4 Errors About the Burden of Proof for God


I used to be a lawyer before entering seminary to prepare for the Catholic priesthood. It’s perhaps unsurprising, then, that I’m fascinated by questions about the “burden of proof” in religious questions. For example, does the burden of proof fall on the believer or the atheist? What sort of evidence is permissible to meet this burden of proof? Do “extraordinary” claims require extraordinary evidence? Should they meet an extraordinary burden of proof, above the burden required for other sorts of claims? Here are four ways that those questions are answered incorrectly:

Error #1: The burden of proof falls with theists, and not with atheists.

Frequently, atheists will claim that “atheism isn’t a belief,” and therefore doesn’t require evidence, and that the burden of proof falls solely with the believer. This is false. As Luke Muehlhauser at Common Sense Atheism explains, “I think the burden of proof falls on whoever makes a positive claim.” It’s true, this means that believers should be able to provide support for this, but it also means that if you disbelieve in God, you should also be able to support this belief:

“If you claim that Yahweh exists, it’s not my duty to disprove Yahweh. [….] But most intellectually-inclined atheists I know do not merely “lack” a belief in God – as, say, my dog lacks a belief in God. Atheists like to avoid the burden of proof during debates, so they say they merely “lack” a belief in God. But this is not what their writings usually suggest. No, most intellectual atheists positively believe that God does not exist. In fact, most of them will say – at least to other atheists – that it’s “obvious” there is no God, or that they “know” – as well as we can “know” anything – that God does not exist.
Thus, if the atheist wants to defend what he really believes, then he, too, has a burden of proof. He should give reasons for why he thinks that God almost certainly doesn’t exist.”

This is the critical distinction. To go from “I’m not convinced from the evidence that Christianity/theism is true” to “therefore, Christianity/theism is false” is a logical leap not supported by the evidence. Alvin Plantinga has a helpful illustration:

“[L]ack of evidence, if indeed evidence is lacking, is no grounds for atheism. No one thinks there is good evidence for the proposition that there are an even number of stars; but also, no one thinks the right conclusion to draw is that there are an uneven number of stars. The right conclusion would instead be agnosticism.”

I don’t believe that there are an even number of stars. But I also don’t doubt that there are an even number of stars. Lack of evidence for X isn’t evidence of its opposite, and in this case, the weight of the evidence is perfectly 50-50.

That doesn’t mean that there aren’t ways in which lack of evidence can be probative: if I claim that it’s been raining all afternoon, the lack of water on the ground would be evidence against my claim. So there’s no reason atheists couldn’t argue that, if God existed, we’d see X and Y, but don’t see those things, and therefore He doesn’t exist. That would be a logical proof, but would take actual intellectual legwork. The alternative of pretending to be agnostic (a phenomenon Muehlhauser rightly treats as widespread) is much easier. It just happens to be intellectually dishonest.

Error #2: Christian Beliefs are either scientifically-evaluable or non-provable / non-falsifiable.

Given that the party making a positive claim (either that there is a God, or that there isn’t) has the burden of proof, what counts as proof? Oftentimes, there’s a false dichotomy that truth-claims (like religious claims) are analyzable in the way that scientific questions are, or else they’re nonsense. Here’s Muehlhauser apparently falling into that trap:

“Christians have done a good job of making it impossible to disprove their God. Yahweh used to be hiding just above the clouds, from where he would throw rocks at the Amorites and do other fun stuff. Now he’s some kind of invisible, transcendent being we couldn’t possibly disprove. But we don’t have to. It’s the duty of Christians to show us some reason to think Yahweh exists. Christians have the burden of proof, because they are making a positive claim. The atheist merely says, “I see no reason to accept your claim, just like I see no reason to accept the claims of Scientology.”

If this is any indication, Muehlhauser’s understanding of Christianity and history is a big part of the problem. He assumes that we used to think that God was “hiding just above the clouds,” because he takes Joshua 10:10-11 embarrassingly literally to mean that God was on a cloud throwing rocks. Further, he claims that Christians did “a good job of making it impossible to disprove their God,” as if the transcendence of God was something we invented as an evasion from these brilliant atheist rebuttals (where does God sit on cloudless days? Shucks!).

In reality, Christian theology has been clear about God’s transcendence for the entirety of Christian history. God’s transcendence can also be shown to be metaphysically necessary from the work of pre-Christian philosophers like Aristotle. Further, you can trace God’s transcendence all the way back to Genesis 1:1, which says, “In the beginning, God created the heavens and the earth,” not “in the beginning, God sat on a cloud in the heavens and created the earth.” It’s true that, out of necessity, the Bible frequently uses anthropomorphic language to describe God and His actions, but what other language could we use? It’s also clear, from the very start, that much of this language is understood by author and reasonably-smart readers alike to be metaphorical and analogical. When God says in Exodus 19:4, “You have seen what I did to the Egyptians, and how I bore you on eagles’ wings and brought you to myself,” no reasonable person takes that to mean that Landroval swooped in and brought the Jews out of Egypt. After all, the prior 18 chapters just finished explaining how the Israelites escaped Egypt.

If you read the Bible by assuming that it is written by and for idiots, don’t be surprised if your Biblical exegesis is idiotic. This might seem like a side point (and admittedly is, somewhat), but Muehlhauser goes on from here to conclude that belief in the Christian God is like believing in a being like Odin, a categorical error only made possible by treating God like an artifact of this universe rather than the universe’s Creator.

So that’s part of the problem. The more important point here is Muehlhauser’s implicit admission that he doesn’t even know how to evaluate the Christian claim of a transcendent God. He needs to imagine that God is a silly rock-throwing cloud monster, because that’s the kind of being he understands how to analyze. Elsewhere, he writes that:

“Skepticism and critical thinking teach us important lessons: Extraordinary claims require extraordinary evidence. Correlation does not imply causation. Don’t take authority too seriously. Claims should be specific and falsifiable."

Underlying this appears to be an attempt to analyze God the way that one would approach the question of whether or not quasars exists. Even the categories of “falsification” assume a particular approach to rational inquiry, an approach well-suited for the natural sciences, but often ill-suited outside of the realm for which it was invented. Take the principle of non-contradiction, for example: it’s a non-falsifiable, untestable logical axiom, but is true nevertheless. This is true of literally all logical axioms. (By the way, without these logical axioms, science is impossible, so this idea that all truth must be falsifiable can be shown to be false). The Wiley Blackwell Handbook of Humanism, while in no way favorable towards religion, acknowledges the incompleteness of this worldview:

“12 times 12 is 144 is something you can establish from the comfort of your armchair by reason alone. You can do this with other conceptual truths. It’s possible, for example, to figure out whether my great-grandmother’s uncle’s grandson must be my second cousin once removed by just unpacking these concepts and examining the logical relations that hold between them. Again this can be done from the comfort of an armchair. No empirical investigation is required. Or suppose an explorer claims to have discovered a four-sided triangle in some remote rainforest. Do we need to mount an expensive expedition to check whether this claim is true? No, again we can establish its falsity by conceptual, armchair methods.
So, even while acknowledging that science, as characterized here, is an extraordinarily powerful tool, let’s also acknowledge that other non-scientific but nevertheless rational methods also have their place when it comes to arriving at reasonable belief – including armchair methods. Science is merely one way – albeit a very important way – of arriving at reasonable beliefs.”

Given this, consider the kinds of claims that Christians make about God. Unlike, for examples, we’re not claiming that God is a creature that originated from this universe, came into power, and reshaped the universe. We’re saying that God is an uncreated Being (indeed, Being itself) and is the origin of all created reality. By definition, such a God isn’t going to be confined to the law of nature… laws He created. We’re making metaphysical claims, and Muehlhauser, like many atheists, is trying to evaluate them like physical claims. It’s true that we also believe that this God became man (without ceasing to be God), but this is a historical claim, and history doesn’t permit of scientific laboratory testing particularly well, either.

I’m not here attempting to prove either God’s existence or the truth of the Incarnation, only to say that those propositions aren’t claims that the natural sciences is equipped to handle, just as it’s not equipped to handle claims like “John Quincy Adams was a member of the Anti-Masonic Party” or “When an equal amount is taken from equals, an equal amount results” or “beauty is a transcendental.”

Error #3: Extraordinary claims logically require extraordinary evidence.

Carl Sagan was fond of quoting Marcello Truzzi’s saying (alluded to above, by Muehlhauser) that “extraordinary claims require extraordinary evidence.” If this is meant as a description of the way we normally approach truth-claims, it’s true enough. We tend to hold things we find credible to a much lower burden of proof than things we find incredible. But trying to turn it into a logical rule is a disaster.

For starters, it renders incoherent results. Imagine a murder trial in which three people were in the room with the victim when he was shot, and forensics proves that there were two shooters. All three are brought up on trial. Using Truzzi’s standard, does this mean that the elevated burden of proof is on each of the three defendants (since there’s only a one in three chance that he’s guilty, making innocence more extraordinary in this case) or on the prosecution (because murder is an extraordinary sort of event)?

Worse, almost everything turns on what you consider “extraordinary,” a term that only appears objective (for example, a person who believed that all weather events were caused by the actions of the gods wouldn’t view such divine intervention as “extraordinary). In practice, this is an example of confirmation bias, which refers to “a person’s tendency to favor information that confirms their assumptions, preconceptions or hypotheses whether these are actually and independently true or not.” If something agrees with an atheist’s assumptions, it’s ‘ordinary,’ and held to one standard. If it disagrees, it’s ‘extraordinary,’ and held to a much higher standard.  All of us are prone to confirmation bias, but the “extraordinary claims / extraordinary evidence” mantra only serves to entrench it.

Error #4: Religious claims should be held to a higher burden of proof than other claims.

The final error I want to address is a permutation of the third one: it’s the idea that, as “extraordinary claims,” religious claims should be held to a a higher standard of proof than ordinary claims.

The normal standard for believing in something is what’s called a “50+1” standard. If you think of assent as balance between “belief” and “disbelief,” any tilting of the scales, however slight, points to the proper outcome. And this is how we normally use “belief,” to the point that it appears illogical and incoherent to do otherwise. G.E. Moore’s famous paradox is that statements like “It is raining and I don’t believe that it is raining” don’t mean anything. You’re affirming two contradictory statements. So, too, to say that “God probably exists, but I don’t believe He does” doesn’t appear to mean anything. And if the likelihood of God’s existence is above 50% (however slightly), then He probably exists.

Although apparently incoherent, this error actually points to an important feature of religious belief. Faith isn’t just an intellectual assent to the historical and metaphysical data. It’s also an act of trust, requiring an act of the will. No matter how clear the historical evidence of Jesus’ Resurrection, you can always choose to ignore or deny it. Pope St. Gregory the Great points out that this was even true of the Apostles who encountered the Resurrected Christ, which is why Jesus can still refer to Thomas’ faithful response as “belief” (John 20:29): “Thomas saw a human being, whom he acknowledged to be God, and said: My Lord and my God. Seeing, he believed; looking at one who was true man, he cried out that this was God, the God he could not see.”

This is an important dimension, because it’s easy to pretend that this is all exclusively on the level of the intellect, that belief and disbelief are motivated solely by the weight of the evidence (and that therefore, all wrong opinions in matters of faith are a matter of ignorance or simple mistake). When a person announces that they will choose not to believe in God even if the weight of evidence tips in His favor, they’re announcing something else is at hand.


So there you have it: (1) the burden of proof falls to the party making a claim (whether that claim is that God does or does not exist); (2) this burden should be met in a manner appropriate to the type of claims (so don’t expect scientific claims to be proven in the same way that historical ones are, for example); (3) requiring special evidence for claims you deem “extraordinary” opens the door for confirmation bias [and so you should be extremely cautious about doing so]; and (4) there’s no rational, disinterested reason to hold religious claims to a higher burden of proof than any other kinds of claims.
(Image credit: Valdosta Today)

Joe Heschmeyer

Written by

Until May 2012, Joe Heschmeyer was an attorney in Washington, D.C., specializing in litigation. These days, he is a seminarian for the Archdiocese of Kansas City, Kansas, and can use all the prayers he can get. Follow Joe through his blog, Shameless Popery or contact him at joseph.heschmeyer@gmail.com.

Note: Our goal is to cultivate serious and respectful dialogue. While it's OK to disagree—even encouraged!—any snarky, offensive, or off-topic comments will be deleted. Before commenting please read the Commenting Rules and Tips. If you're having trouble commenting, read the Commenting Instructions.