•  Is Atheism the Null Hypothesis?
  •  Is Atheism Falsifiable?
  •  Does Atheism Carry the Burden of Proof?

Atheism has distinct definitions which can be categorized as follows:

• Weak/Soft Atheism: I do not believe "god exists". Weak atheism is the rejection of the positive claim that "god exists".

• Strong/Hard Atheism: I believe "god does not exist". Strong atheism is the acceptance of the negative claim that "god does not exist".

Strong atheists are a subset of weak atheists: those who believe god does not exist form a subset of those who do not believe god exists.

Theism/atheism address belied/disbelief in whether god exists.

Gnosticism/agnosticism address knowing/not knowing that god exists.

  • Agnostic Theist: someone who believes god exists but does not know or claim to know this.
  • Agnostic Atheist: someone who does not believe god exists, but does not know or claim to know whether god exists.
  • Gnostic Theist: someone who believes god exists and knows or claims to know this.
  • Gnostic Atheist: someone who believes god does not exist and knows or claims to know that god does not exist.

I, Karlen K., am an agnostic hard atheist, depending on the definitions of knowledge, and god: I believe there is no god but do not claim to know that god does not exist. I do not believe there is a god, and in fact believe there is no god.

The burden of proof is on the proposition, not on the opposition!

The burden of proof is on the one who makes a claim, regardless of the positive or negative content of the claim. The burden of proof is on the claimant, not the respondent.


  • Null Hypothesis: H{0}:= "God does not exist" = "There is no god" = "No god exists"
  • Test hypothesis: H{T}:= "God exists" = There is a god" = "Some god exists"

There are four possible believe positions here: Let: B[]:= believe [], ~B[]:= do not believe:

  • B[H{T}]:= I believe "god exists". ------------------ acceptance of a positive claim.
  • ~B[H{T}]:= I do not believe "god exists".------------ rejection of a positive claim.
  • B[H{0}]:= I believe "god does not exist".----------- acceptance of a negative claim.
  • ~B[H{0}]:= I do not believe "god does not exist.----- rejection of a negative claim.

Note that: ~B[H{T}] is the rejection of the positive test hypothesis H{T} (where "rejecting" = "not accepting"). On the other hand, B[H{0}] is the denial of the positive test hypothesis H{T} (where "denying" = accepting that H{T} is not true (i.e. false)).

One need not satisfy any burden of proof for rejecting a claim, whether positive or negative in content, but for accepting (or asserting) that the claim in question is not true (i.e. false).

~B [H {T}] =/= B [H {0}]:

I do NOT believe "god exists" =/= I believe "god does NOT exist"

Definitions Let: b{X}:= "I believe {X}" = I accept that X is true. Then: ~b{X}:= "I do not believe {X}" = I do not accept that X is true (i.e., I reject that X is true)

Let: X:= a proposition, ~X = the negation of X (i.e., not X)

  • b{X}:= I believe{X};
  • ~b{X}:= I do not believe{X};
  • b{~X}:= I believe {~X};
  • ~b{~X}: = I do not believe {~X}.

Consider: X = "god exists", then ~X = "god does not exist".

  • H{0} = Null Hypothesis = ~X = "God does not exist."
  • H{+} = Positive Hypothesis = X = "God exists."

An hypothesis that can be expressed in terms of an equality relation ("="), zero ("0"), or a negation ("not") is to be chosen as the null hypothesis. In our case, one (H{0}) of the two mutually opposing, exclusive, and exhaustive hypotheses carries the negation operator of formal logic ("not"); therefore, its alternative hypothesis expresses a positive proposition, hence H{+}. Therefore, in this case the negative claim "god does not exist" is to be chosen as the null, and the positive claim "god exists" is to be assigned the (alternative) positive hypothesis.

  • b{X}: I believe god exists. ------------------- Theism
  • ~b{X}: I do not believe god exists. ------------ Atheism (weak)
  • b{~X}: I believe god does not exist.------------ Atheism (strong)
  • ~b{~X}: I do not believe god does not exist.----- Rejection of Strong Atheism

There are only two possibilities:

H{0}: God does not exist. H{+}: God exists.

H{0} can be falsified, but cannot be accepted; it can only fail to be rejected. In statistics, it is incorrect to accept the null hypothesis because of failing to be able to accept the (alternative) positive hypothesis.

H{+} cannot be falsified, but can be accepted if there were sufficient evidence constituting proof for a given standard of proof (degree of certainty), such as a "clear and convincing evidence", "beyond a reasonable doubt", "beyond a shadow of a doubt", "95% confidence interval", etc. If sufficient evidence is gathered, then the null can be rejected. If the evidence is insufficient to reject the null, one does not accept the null, but merely fails to reject the null.

  • 2
    Talking about a null hypothesis only makes sense when we have an empirical test to run. There is none for God, so the debates take place on a different plane. And the only side without a burden of proof is agnosticism that suspends judgment either way. Both atheism and theism have their burdens of proof.
    – Conifold
    Jul 12, 2020 at 3:27
  • 1
    Does this answer your question? Asserting that Atheism and Theism are both faith based positions
    – Conifold
    Jul 12, 2020 at 3:30
  • Theism / Atheism address belief / disbelief respectively regarding god's existence. However, Gnosticism / Agnosticism address knowledge respectively regarding god's existence. There are 4 positions here: Agnostic Theist, Gnostic Theist, Agnostic Atheist, Gnostic Atheist. Knowledge is a subset of belief: those beliefs that are true and justified. Therefore one can be an agnostic theist: someone who believes god exists but does not claim to know it. A gnostic atheist is an atheist who asserts there is no god and knows or claims to know it. Jul 12, 2020 at 3:39
  • 3
    The arguments for/against God's existence have become perpetuum mobile patent applications of philosophical forums. Jul 12, 2020 at 11:07
  • 1
    Too many concepts mashed together in this question. Null hypothesis is only relevant in statistical testing with the context of an actual statistical experiment, and must be about statistical observations. Which of the positions should be default and who should have the burden of proof is not a useful philosophical question, as it has to many existing answers already.
    – tkruse
    Jul 13, 2020 at 5:07

6 Answers 6


All these different categories seem to be a crude attempt to reproduce the notion of probability. If we assume that the proposition "god exists" is a well defined boolean, then we must assign some probability to it. This probability should be strictly between 0 and 1. (If you assign probability 0 or 1, by the formal math of baysian updating, you can't ever change your mind, however good evidence you get.)

You don't have a burden of proof, you have a prior, which assigns probabilities between 0 and 1 to statements, probabilities that are then updated on evidence.

Some popular priors are versions of Occams razor, that assign smaller probability to more complicated hypothesis. Technically, this spreads probability out over complete hypothesis. So imagine a giant list of equations, all possible equations, as possible physical laws. Some are the equations of string theory or quantum field theory ect. Each equation has a prior probability attached to it, which is smaller the longer the equation is. All the probabilities add up to 1. THe equations are split into those that describe a universe with a god, and those that describe a universe without.

If your definition of "god" is very complicated and specific, something you would need many long equations to describe, then very few of the equations will contain a god, and your prior on "god exists" is very low. If your definition of god is wide enough, you might count alien pranksters and timetravelers and the abstract force of evolution as close enough, and your probability on "god exists" will be high.

(Don't mistakenly jump from the broad definition to the narrow one. If you have convinced yourself that the universe must have a first cause, and labelled that cause "god" it is a mistake to assume the cause is sentient and cares about you in particular, and also gave stone tablets to desert nomads.) Once you have your definition, think of several possibilities, as different as possible but still fitting the definition.


The title of your question is based on multiple subjective assumptions, which are classical of people who wants to proceed to assess the existence of God using scientific tools without fulfilling the minimum requisites to make science. Reading Mario Bunge (Science, Method and Philosophy) could help. Your questions have only speculative answers, so, this is intended to clarify the problem.

Atheism is essentially a lack of belief, which is not an hypothesis, but moreover a subjective preference. Starting from there, your question is ill-formed.

Your question refers probably to the proposition "God does not exist", which belongs not in the domain of science but to the domain of philosophy (that's why you're posting on the philosophy forum). Scientific knowledge is essentially empirical, and science seeks for empirical truth. Philosophy deals with the final truths.

In order to prove the existence of God as an empirical truth (in order to make science about God), you need at least to have an objective definition of God, and a formalization of the empirical knowledge of God (which don't exist at all, because there is no proof of any experience of God); that is, you need to have a solid empirical base. The required foundation is not related to the types of theisms or the discussion of beliefs that you are making. Far from it. That is completely irrelevant without a solid empiric base (and if you want to discuss your statements, make separate questions of each one).

Perhaps you are not aware about the fallacious start of thermodynamics, but your problem is identical, you need to read this. Temperature being essentially a subjective feeling, it was used as an objective foundational element for the development of the laws of thermodynamics (in the same way, you are trying to address the problem of God based on a subjective understanding of God). Scientists noticed such problem after the laws were fully developed, so they developed a concept of temperature which would make it an objective and measurable property, coherent with the subjective feeling. Since it is foundational to thermodynamics, they called it the "zeroth" law, because without it, the rest have no sense (likewise, you are developing the rest, without the required foundation). In the same way, you need a zeroth-law of theologics before proceeding to further hypotheses. That meaning, you need at least to address the problem of God by being objective about the meaning of God, and what we can experience about it. Again, science is about experience, not about belief and deep truths.

Having said that, the rest of your question is largely speculative. You are developing concepts based on the previously described issue: the fact that your understanding of God and the logic analysis based on it is far from being objective.


Is atheism the null hypothesis?

That entirely depends on the experiment being proposed. This can’t be answered in an absolute sense. I’d argue that theism couldn’t be the null hypothesis though. You’d need to choose which god, or which attributes of god, you were testing for.

Is atheism falsifiable?

Absolutely. The good old flaming letters, half way between Earth and the sun, that read “I’m here” or equivalent should do the trick.

Does atheism carry the burden of proof?

Possibly. If the atheist is using their belief to extract money from you or make you live by their moral code then yes. They should justify their actions.


Your "question" is not actually a question, but instead an assertion of a particular, -- and false -- set of claims.

The evaluation of a proposition is more complex than your two false dichotomies. For a given proposition, one may have sufficient justification to hold that it is 1) True, 2) false, 3) invalidly articulated, such that it is unevaluatable, or 4) one does not have sufficient information to justify any of the three prior options. Note -- this is four options, not your false dichotomy of two.

One can hold by each of these four options with varying degrees of certainty. Certainty is also not the gnostic/agnostic false dichotomy you attempt to force it into.

When applied to the proposition of a God's existence, these four options USED to map into 1) == theism, 2) == atheism, 3) == agnoticism, and 4) == still searching/evaluating.

The "null hypothesis", of course, would be 4). Not 2).

However, for purposes of debating tactical advantage, one atheist spokesperson argued for the false dichotomies you now embrace. That was Anthony Flew, who made this argument to change the definition of atheism in his 1976 book The Presumption of Atheism. For a quarter century, his proposal was basically ignored, but in this century, advocates of atheism have increasingly embraced the two false dichotomies, and one special pleading fallacy that constitute Flew's proposal. Wikipedia's summary is:

The Presumption of Atheism

One of Antony Flew's most influential professional works was his 1976 The Presumption of Atheism[26] in which Flew forwarded the proposition that the question of God's existence should begin with the presumption of atheism:

"What I want to examine is the contention that the debate about the existence of God should properly begin from the presumption of atheism, that the onus of proof must lie upon the theist. The word 'atheism', however, has in this contention to be construed unusually. Whereas nowadays the usual meaning of 'atheist' in English is 'someone who asserts that there is no such being as God, I want the word to be understood not positively but negatively... in this interpretation an atheist becomes: not someone who positively asserts the non-existence of God; but someone who is simply not a theist.

The introduction of this new interpretation of the word 'atheism' may appear to be a piece of perverse Humpty-Dumptyism, going arbitrarily against established common usage. 'Whyever', it could be asked, don't you make it not the presumption of atheism but the presumption of agnosticism?[27]

Flew's proposal to change his profession's use of the term atheism saw limited acceptance in the 20th century, but in the early 21st century Flew's negative sense of 'atheism' came to be forwarded more commonly.[28][29] The impact of Flew's proposed negative atheism, which is often referred to today as 'weak atheism' or 'soft atheism', is illustrated by analytic Philosopher William Lane Craig's 2007 assessment that the presumption of atheism had become "one of the most commonly proffered justifications of atheism."[30] And BBC journalist William Crawley 2010 analysis: "The Presumption of Atheism (1976) made the case, now followed by today's new atheism, that atheism should be the ... default position".[31][32] In recent debates, atheists often forward the Presumption of Atheism referring to atheism as the "default position"[33][34][35][36] or has "no burden of proof"[37][38] or asserting that the burden of proof rests solely on the theist.[27][39][40]

Note that Flew, in 1976, believed that God claims were incoherent and unevaluatable in principle. Agnosticism such as Flew's has a burden of proof, just as claims for both truth or falsity of God claims. And there were efforts put forward by theists to rebut Flew's assertion of the incoherence of God claims. Flew himself found one of those, The Coherence of God, by Richard Swinburne, to be sufficiently powerfully argued that he himself eventually conceded his own agnostic presumption was false.

That Flew himself found his own reasoning invalid, at roughly the same time that other atheists were finally taking up his definition (which relies upon, for its tactical effectiveness, Flew's presumption of the inability of theists to even articulate a God claim, much less support it), is a historic peculiarity in current atheist "reasoning".

For specific answers:

  • Atheism is not the null hypothesis -- that would be uncertainty
  • Yes, atheism is as "falsifiable" as basically any metaphysical claim. One can identify multiple circumstances that, if they took place, would make accepting it less and less plausible.
  • Yes, all metaphysical claims carry a burden of proof. Theism, atheism, and agnosticism all carry a burden of proof. As does any claim that there is no support for any of those other three claims.

Null hypothesises and default positions are scientific concepts about reproducible observations of nature. Gods however conceptually are agents which observably act out of their own volition, not deterministically and not reproducibly, while remaining unobservable outside such actions.

Thus it is not valid to apply null hypothesis or default positions on the future observability of gods and miracles.

Instead, for any individual gods are hypothesis to explain past observed or reported events (miracles) which can not likely be reproduced, proven or disproven.

Thus any atheist/theist position is not based on it being a philosophical default position in itself, but rather a default position based on prior subjective assumptions of truth of holy scriptures and/or other miraculous experiences or reports.

In the theoretical absence of any so-called "holy" scripture or reports of miracles, atheism could be a default position. But since humanity already has "holy" scriptures and diverse reports of miracles, individuals are forced to choose what events to believe in first, and then pick a logical position on gods as a second step based on what they chose first.

Note that people may also chose to consider reproducible observations as miracles, such as the miracle of life, the beauty of nature, the vastness of the universe, the power of love, etc.


As a matter of fact, I would argue that atheism is the alternative hypothesis. On the basis that theism and faith predated atheism. Most of the world had some form of religious faith. Within any society that is predominantly, say for example Christian, then this is the status quo, this is the norm. It is in fact atheism that challenges this preconception. Indeed, atheism wouldnt exist without theism first against which atheism is deliberately contrasted. Atheism is constructed as the antithesis of theism.

You might argue that before religion when man was primitive they were atheist, so atheism actually predates theism. I would contest that for two reasons. The first is that it presupposes the falsity of biblical creationism, so its circular reasoning. The second is that atheism, as I previously stated, exists only to oppose theism; one cannot be an atheist if one has no conception of god in the first place. The very definition of atheism testifies to the conception of a deity.

Now, both positions make their own claims. One says god does exist, one says it doesnt. Why is the burden to prove immediately on the side that held the status quo for far longer? That makes no logical sense. Atheism is the challenger, the accuser if you will, claiming that the status quo has been wrong. The burden of proof is clearly placed on the atheist.

Depending on religion - Ill use Christianity as an example - its a well established assertion that faith is a key component to the religion. "God wants you to have faith", "faith begets faith", etc. Okay, but that implies that there wont ever exist conclusive evidence because it explicitly undermines faith and replaces it with certain knowledge. A lack of evidence is therefore self-consistent with Christianity.

But the atheist, for example, claims science and logic can be utilitized to prove their assertions. They have no such hindrances from pursuing truth, yet they frequently deny the duty to do so. Which I find not only odd and anti-intellectual, but ideologically inconsistent and self-contradictory within the framework of atheism.

If Christianity is inherently incapable of proving their claims, as an inherent quality of the religion, why should they be obliged to do so? Especially when the dominant, preexisting status-quo belief system. You see its a catch 22, a rock and a hard place. If they cant prove then they cant persuade through logic, but if they can prove then in a major way they undermine faith and the entire religion collapses anyway. Intellectually speaking, its an unfair burden to place on faith-based religions - an intellectually dishonest and ideologically cowardice game to play. If atheism is so hung up on logic and empiricism, then utilize such tools before making an assertion.

Atheism isnt atheism at all... it is by and large anti-theism. If youre simply "devoid of faith" then youre atheist (what you call soft atheism), as long as youre open minded to either side. But very few self-described "atheists" are open minded. If however you are positively asserting the absurdity of religion and denying gods existence then youre an anti-theist (what you call hard atheism). Im describing the meanings of the prefixes a- and anti- here, the true meanings of the words and not how theyve been falsely interpreted by todays world.

Anti-theism is a religion in itself. It makes an assertion about transcendent matters... be it deities, origin of life or of creation, origin of morality, spiritualism or after lives... it doesnt matter what the claim is, they are asserting one of this nature and they do so without any evidence of their own to back it up. If after all the "atheist", i.e. anti-theist, did have substantive evidence then the world wouldnt have religion anymore.

The anti-theist believes what they believe in faith, and nothing more. But what they believe is of a spiritual, transcendent matter. Put the two together and you have the definition of religion.

I have no doubt that I will get down voted for this response. Even though I explicitly answered each and every one of your questions. Thats what "open minded people" do, after all.

Take me for example. Im a true atheist, not an anti-theist. I am open minded and willing to be persuaded by either side. Through facts or through divine inspiration (the only way I recognize theism to be persuasive). Such evidence hasnt crossed my path yet. I simply "dont know".

Wait, I hear you saying "Isnt that agnosticism?" Well, maybe Im agnostic too, Im not sure. But agnostic simply means you dont believe evidence exists, it doesnt mean youre undecided. There are agnostic theists and gnostic theists. Within the Christian framework there should only be agnostic theists. And I would hope all anti-theists, the ones who pretend logic and science are on their side, would be gnostic.

I will take your lead on this and use "soft" and "hard". I would argue that soft anti-theism is a religion unto itself, for reasons already explained, what Ive previously described as anti-theism. I would thus call hard anti-theism a form of faith bigotry. Its usually a visceral over-reactionary response to other peoples beliefs, rooted in resentment over principles and of other peoples successes.

Then there is the concept of "burden to prove" itself. This does not stem from logic or science. I would make the claim that there is no basis in empiricism for it. Although it potentially can be adopted into it. The notion of a burden to prove is actually a moralistic philosophy stemming from Abrahamic religions, and has more to do with criminal court than it does scientific or philosophical inquiry. The idea is, of course, that it is a greater sin (moral wrong) to punish the innocent than it is to fail to punish the guilty. We get "proven guilty beyond a shadow of a doubt" from the same source and for the same reasons. The burden to prove is always on the accuser, not on the accused. This is why in my first paragraph it was relevant to point out that atheism is challenging the status quo, not the one being challenged. Historically this is so, and so the burden should have always been on atheism.

On the surface, the idea that the burden to prove rests with the side proposing a claim is not unreasonable even in a scientific framework. But then again it kind of is unreasonable if you think about it in terms of intellectual honesty. If science is about the pursuit of truths then shouldnt all sides of any claim be pursued? If one side of the debate is constitutionally incapable of providing proof, then as previously argued why should they be compelled to? Why should the alternative who is not hindered in the same way deny the responsibility?

A lack of evidence is not evidence of lacking, as Im sure youre aware. So any atheists who takes the theists inability (or refusal) to try cannot claim scientific victory. The atheist must still prove his own claim if he is to believe it is truth.

I assert now that there is no burden to prove in science, and no shifting of that burden either. The burden rests squarely and equally on both sides of any claim. Prove a proposition or else prove its opposite. Or admit there is no evidence either way; or else admit that you simply lack the will or desire to pursue truths in that subject, and graciously bow out of the conversation. A scientist - any rational person - would not deny the duty to pursue the truth of any claim if at all possible.

And no rational person would expect rationality from faith; if it isnt logically inconsistent with the faith then it cannot be utilized to disprove it. Religion only needs to be self-consistent, not consistent with its alternative viewpoints.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .