This is an argument I've heard from many New Atheists, including Dawkins and Hitchens.

The conversation typically goes like this. NA stands for New Atheist, P is some other person.

P: What's your problem with my belief in God?

NA: You have no evidence for it. Hence, I believe your position to be irrational. (1)

P: I don't agree with that, but I guess we define the word "irrational" differently. (2)

P: But, anyways, do you believe God doesn't exist?

NA: Yes. God does not exist.

P: Do you have evidence for this non-existence? After all, you are making the active claim that the entity God cannot possibly exist. Hence, the burden of proof lies on you to provide evidence, which, according to your own belief, as stated in (1), is a necessary condition in order for you to classify your own belief as rational.

NA: No, I cannot provide such evidence.

P: Then you cannot believe God doesn't exist. (3)

NA: In that case, maybe you shouldn't believe that the tooth fairy doesn't exist either, or the spaghetti monster! Can't you see how ridiculous such agnosticism gets? (4)

Here's my problem with this argument. When P states in (3) that the New Atheist cannot believe in the non-existence of God, P is doing so based on the statement by the New Atheist in (1). Remember, it is only the New Atheist who ascribes to this paradigm that evidence is necessary for rational belief. P never accepted that paradigm, as they mention in (2). Hence, when P in (3) claims that NA cannot believe in the non-existence of God, P is essentially saying that within NA's paradigm, there seems to be a contradiction.

And so, finally, when NA in (4) attempts to ridicule P's retort by asking whether P also is open to the existence of the tooth fairy, NA seems to be assuming that P also ascribes to NA's evidence-based paradigm. But P never did accept that paradigm, and hence NA's retort (which is an appeal to ridicule either way) doesn't seem to add up.

Essentially, it seems that when NA attempts an appeal to ridicule and tries to mock agnosticism, NA seems to forget that P never ascribed to agnosticism, but merely pointed out that it was NA who originally used an agnostic-like argument in (1), but then refuses to consistently follow their own argument.

  • 2
    Can you give an example of someone saying that they affirmatively believe that God does not exist (not merely that they don't believe God exists), that they don't have evidence that God doesn't exist, and that it is irrational to believe something without evidence? – Acccumulation Apr 9 '18 at 17:55
  • Did you mean "rationality" as in, "I won't assert something that's not the logically defensible implication of evidence"? Or did you mean "rationality" as in, "I won't DO something unless I can assign probabilities for it that makes it desirable"? This is an important distinction! – elliot svensson Apr 9 '18 at 17:59
  • 2
    Putting aside the fact that Dawkins at least does not state that he knows there are no gods, if we amended NA's first statement to the following, would that clear it up for you?: "NA: You have no evidence for it. And I believe one should only believe in the existence of things for which one has evidence. Hence, I believe your position to be irrational." – Chelonian Apr 9 '18 at 18:24
  • 1
    @Chelonian, I don't actually see a problem in NA's reasoning here. The problem is the P statement beginning with "I guess", which I think poorly characterizes a typical theistic stance. Unless "rationality" is in reference to probabilistic arguments against belief in God! – elliot svensson Apr 9 '18 at 18:31
  • 3
    Possible duplicate of Are New Atheists hypocritical? – jobermark Apr 10 '18 at 0:12

Lot: What an amazing day this is! How lucky I am! How jubilant!

Baye: What has your spirits up so high today?

Lot: You would never have guessed! I'm going to win the lottery! I just know it!

Baye: (sighing) You know that's unlikely. What makes your ticket special?

Lot: Oh, I don't have a ticket.

(Baye headdesks)

Baye: Ouch. You're not going to win the lottery. Why would you even think you were going to win the lottery?

Lot: I don't need evidence! What makes you so certain anyhow? How do you know I'm not going to win the lottery? There are thousands of ways it could happen! Someone could walk in right now and hand me a winning ticket!

Baye: The mere possibility of a situation is not reason to believe it!

Lot: Hmph. As far as I'm concerned it will either happen or it won't. That makes it 50-50. Your certitude is no better than mine.

Baye: (screaming inside) That is not how probabilities work!

Lot: Well that's just, like, your opinion, man.

Baye: No, it's not. The way we reason about probabilities is not an arbitrary choice out of all the ways you can assign numbers to events. Bayesian statistics is not just just a postulate, it is a mathematically ideal solution given natural assumptions. Competitive markets repeatedly demonstrate that these formal models of the world are better predictors. A 50% probability should be right 50% of the time. Some ways of looking at the world are just right, and some are just wrong.

If the story wasn't clear enough, I am saying that NA is right to treat this scenario asymmetrically. One's tools of rationality are designed precisely so that you are able to talk about areas of uncertainty. These are not a mere matter of convention, they are facts about the way the world works, and the further you stray from them the more wrong you will be.

  • 1
    While the story is mildly amusing — iff you know Bayesian probability — it does not clarify the answer, because trying to understand the final claim — and why it is not valid — is harder than trying to understand it in the original post. -1 for favouring being funny over clarity and pedagogy. – MichaelK Apr 10 '18 at 9:16
  • Discussion in chat. – MichaelK Apr 10 '18 at 9:26
  • I had to delete some comments as they did not contribute to an improvement of the answer but offered a differing perspective that should, if anything, have been part of an alternative answer. Please head for chat for material discussions and dissent. – Philip Klöcking Apr 10 '18 at 9:42
  • 1
    @Veedrac Well my point is that you are not doing any disabusing with that story. If the reader cannot — or has difficulty to — associate the story you weaved with the subject at hand, it is pointless. – MichaelK Apr 10 '18 at 11:42
  • 2
    @MichaelK Well I hesitate to just drop them because that's the part that shows that Baye really is coming from a uniquely justified position, but I'll reword it when I get time. – Veedrac Apr 10 '18 at 12:52

The atheist makes no claim that they can be agnostic about

First a note: the label "New Atheist" serves no purpose here, because it does not refine or clarify anyone's arguments or stance.

The vocabulary for the context for this post:

Atheism means rejection of faith-based doctrine, arguments and claims

Faith means belief without supporting evidence

Hence there is no such thing as New or Old or Somewhat Out-Of-Fashion Atheist. There is only one manner in which you can reject religiously inspired claims and/or arguments, and that is state "No, I reject that claim". There is no "New" or "Old" way of doing it, except possibly to bring new arguments into the discussion.

Second note, the Flying Spaghetti Monster is not an attempt at ridicule. It points out that whoever wants to remain an agnostic — in the complete absence of evidence — is entertaining the notion of a deity by Special Pleading. Special Pleading is not a logical fallacy per se, but it does point to an argumentative inconsistency (colloquially called "hypocrisy" or "double standard").

The main point... you want to make the argument that:

NA seems to forget that P never ascribed to agnosticism, but merely pointed out that it was NA who originally used an agnostic-like argument in (1), but then refuses to consistently follow their own argument.

You are trying to make a Tu Quoque towards the atheist. But it fails. Let me clean the fluff out of your hypothetical dialogue.

Theist: I claim to know the divine will.

Atheist: I reject that claim.

T: Why do you reject my claim?

A: Because you have not presented convincing arguments for it.

T: You cannot prove that the claim is not true. It could be true.

A: It could indeed be true. But we — you included — have a habit of not taking unsubstantiated claims for fact, even when they could possibly be true. This goes especially for claims about the supernatural. Instead we relegate such claims to the category of "fantasy".

In brief: the theist is making a claim; the atheist rejects it. The atheist is not making an assertive claim, so there is no such claim for them to be agnostic about.

Hence your argument — that the atheist is being agnostic about their own claims in a hypocritical manner — falls.


If your conversation had played out a bit differently, your argument would have held.

Theist: I claim to know the divine will

Atheist: I reject that claim

T: Why do you reject my claim?

A: Because there are no gods at all, therefore there is no divine will to be known, and therefore you cannot know of any such will.

T: You cannot prove there are no gods.

A: It could be true there are no gods.

...then your argument would have held true. If the atheist brings forth such an assertion — "there are no gods" — then that is a claim that they can be agnostic about and then it would be hypocritical to strike down on agnostic arguments from the theist.


...atheists very rarely make that claim. Not even Dawkins — that you referred to — goes to a seven-point-oh on his seven-graded scale, and stops at "six point nine".

Christopher Hitchens — that you also referred to — made this distinction very clear, and argued that atheism is no belief in the existence of gods, as separate from belief in the non-existence of gods, in a debate with his brother Peter Hitchens (32 minutes, 10 seconds into the video).

[The] atheist proposition is the following — most of the time: It may not be said that there is no God; it may be said that there is no reason to think that there is one.

And no atheist needs to go as far as to claim the non-existence of gods in order to achieve the rejection of theistic claims. Admittedly, it is true that when the claim that there are no gods is accepted into the discourse, then all theistic claims fall consequently, and it would therefore be a convenient shortcut for the atheist to reject all theistic claims by using the claim of non-existence. But — barring that — the atheist can simply go the slightly longer route and show that the theistic claims are unproven.

The only difference this makes is in regards to agnosticism. The claim that there are no gods kills agnosticism outright, while the position that claims about gods are unproven allows for it.

That is where the counter-argument you listed above shows up: sure... it could, possibly, be true that there are gods. But why would we — collectively — bother about that possibility, when we never bother about any other supernatural claims other than to use them as fanciful flights of fantasy that tickle our imagination?

  • Seriously? Not a new way to reject a claim? – user32096 Apr 10 '18 at 11:50
  • @nocomprende Yes, seriously. The atheistic stance is the same as it has always been: "You — fellow human — have not convinced me". Hitchens, Dawkins, Dennet, Harris, Dillahunty, Krauss, Hirshi Ali carry the same claim that all unbelievers have carried throughout the ages: your argument is not convincing. The only rejuvenation that happens is that new answers are being put forth to theistic claims. But most of these are in response to new theistic claims that the theists have to resort to when their old claims are finally widely and commonly rejected, and not just by atheists. – MichaelK Apr 10 '18 at 11:59
  • The Flying Spaghetti Monster is a new argument, yes... but the only reason it exists is because of the new theistic argument that is "We have to consider the possibility of an Intelligent Designer that created the species [cough]AndThatDesignerIsGod[cough]" . Previously the theists did not need it because then they just claimed — outright — that "God created the species, period" and simply steamrolled anyone that claimed differently... par example. – MichaelK Apr 10 '18 at 12:03
  • You said that there is not a new or old way to reject a claim, and so I mentioned a recent one: you look surprised and say 'Seriously?" It seems rather effective in bridging the Sar chasm. Should see someone about that cough. – user32096 Apr 10 '18 at 23:04
  • @nocomprende Was there a point - a profound or at least relevant point - you wanted to make? – MichaelK Apr 11 '18 at 3:53

I don't think a typical theist says that they define rationality differently than a New Atheist. Instead, I see that P does indeed have evidence, although it is either evidence not acknowledged by NA or unavailable to NA.

Defining rationality differently is asking for trouble. I would call that a bad plan by P.

The form of P's evidence could be any of the following:

A) Personal witness of supernatural events, such as a divine messenger or answer to prayer.

B) Acknowledgement of the intended outcome to the moral argument, "watchmaker argument", ontological argument, Kalam cosmological argument, etc. ...for the existence of God. None of these arguments are irrational on the face of it, although possibly NA could demonstrate that with effort.

C) Reception of historical accounts and letters which testify to supernatural events attributable to God. On the face of it, it is not more irrational to believe the Bible than it is to believe the sayings and writings of Richard Dawkins or Chris Hitchens.

  • 1
    1) Personal witnessing is not evidence, at least not in a public discourse, because those claims cannot be reviewed by your peers. 2) Such arguments demands omniscience on the level of a god, because otherwise those arguments do not defeat the null hypothesis, that there could be another explanation than a personal god involved 3) that would — at best — be witnessing. And saying that the bible is as credible as Dawkins and Hitchens is nonsensical, because those people can at least be attested to with certainty and confronted about their arguments. No holy scripture offers that. – MichaelK Apr 10 '18 at 9:24
  • 1
    @MichaelK, since the judge and the witness are the same person, it is evidence if a person experiences an answer to prayer or a divine message, even though the evidence may not be transferable. – elliot svensson Apr 10 '18 at 13:25
  • The question regards a conversation... an exchange between (at least) two people. Your latest comment only argues that personal witness can be evidence — or at least a source of conviction — to only one person; in a conversation it is useless as evidence. – MichaelK Apr 10 '18 at 13:28
  • 1
    It is no more irrational to believe the Bible than to believe Dawkins and Hitchens on the face of it. Each of these sources provides words that are accepted by some group of people, and if a person has no previous knowledge of such things, the justification for listening to each is pretty much the same. With further investigation the wisdom of believing any of these voices may be questioned and known more fully. – elliot svensson Apr 10 '18 at 13:29
  • 2
    I agree: first impressions may be deceiving. On first impression the bible may seem convincing. Although I would opine the opposite: if the bible is dropped into the lap of an adult informed person that has no previous knowledge of it, nor has been raised in a culture that reveres belief in revelation, nor takes it as a virtue to have faith in religious belief, then I do not see the Bible is convincing of anything — neither on the face of it, nor when examined in depth — considering its advocacy of things we today consider wholly unethical, and its inconsistencies and contradictions. – MichaelK Apr 10 '18 at 13:33

I would say there is a miscommunication occurring here, rather than faulty ideas.

As Chelonian stated in their comment, NA's 1st statement should be amended to "NA: You have no evidence for it. And I believe one should only believe in the existence of things for which one has evidence. Hence, I believe your position to be irrational."

If NA did not clarify that, then one could argue that it's either NA's fault for not stating that, or it's P's fault for not asking for clarification about the differences in their idea of rationality. Regardless, the statement about the tooth fairy is a poor show of character from NA, and I would say that if a hypothetical argument were to end that way, it would have ended due to emotions clouding the effective delivery of argument. However, I do not think P is entirely without blame, as (3) is a forcing statement even when (2) might have suggested that the two were not aware of each others' full ideas.

  • I think it's a mix. One thing the internet taught me is that often, when someone seems to advocate an unreasonable position, it's not a miscommunication: they really believe it. – Hurkyl Apr 10 '18 at 2:06
  • I try to believe six impossible things before breakfast. – user32096 Apr 10 '18 at 11:48

What is going on here is the implicit use of a criterion like Popper's falsifiability to assign the burden of proof; and a helping of Laplace' "I have no need of such assumption".

If a sufficiently specified God is discussed, then it becomes reasonable to ask for evidence for either existence or non-existence. We have climbed Mount Olympos, and found no sign of Zeus, etc.

However, typically such debates are about very unspecific notions of God. The believer could still offer evidence, by specifying God however he wants, and then suggesting matching evidence. The New Atheist argues that until that happens, the default position should be to assume non-existence - just as we do with many other hypothetical constructs.

  • This argument, of a preferred default position, implies a contradiction to the New Atheist position that for a thing to be known it must be demonstrated through evidence. How does the New Atheist know what the default position should be, and is this applicable to P? – elliot svensson Apr 9 '18 at 17:49
  • 2
    As I said, essentially its about falsifiability. Assuming non-existence of X is generally preferred as the baseline, because it can be falsified by showing an X. Claiming that a specific X exists is fine, because we can examine the candidate. Claiming that some unspecific X exists is rejected as a base assumption. – Arno Apr 9 '18 at 17:54
  • Oh, did you mean rationality not as it pertains to logic but to risk-evaluation and probability-decisions? – elliot svensson Apr 9 '18 at 17:58
  • @elliotsvensson: Statistics and science, in general, are about sentences like "With 99.95% probability we can not say that XYZ is not the case." This is different from being able to say that something is actually positively the case. Especially Dawkins is quite explicit in referring to this nuance which does not make any difference for most scientific knowledge but sometimes is important. – Philip Klöcking Apr 10 '18 at 9:49
  • But for different people, the 0.05% possibility that actually what just happened was a "medical miracle" (for example) is much more plausible because of their priors. To say such belief is irrational is incorrect- it's the perfectly rational response to the evidence set before them. – elliot svensson Apr 10 '18 at 13:43

The argument, as you present, certainly does not add up. NA is contradicting his own epistemological stance. NA's acceptance of (1) makes him accept criticism of P (as in 3) by default. NA also makes an appeal to ridicule, conflating tooth-fairy and spaghetti monster to God, and making an implicit assumption that all have equal amount of evidence.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.