Now you see it, now you don’t.

One approach to forming beliefs that’s popular amongst atheists might be summed up by the dictum ‘in the absence of evidence, assume absence.’ It’s an appealingly simple line, made attractive by the way it seems to suit very well a number of instinctive thought processes. In this post I want to look at this idea a little, and see if it’s capable of doing the work often asked of it.

First, the folksy, intuitive version. In any number of situations, we take the absence of evidence as strong support for the proposition of absence. Maybe, while I was writing this and my back was turned, dastardly aliens slipped into my house, abducted my family and replaced them with replicas, identical in every way except that in twelve month’s time they will revert to their true natures and call their friends in to slaughter me. Not impossible, perhaps. And if it did happen, there’d be no way of knowing until it was too late, right? Except, we don’t seriously entertain the possibility (unless we’re unfortunate enough to suffer from Capgras syndrome, wherein the imposter scenario feels terrifyingly compelling).

The human imagination being what it is, it’s possible to invent limitless situations in which a phenomenon for which we have no positive evidence is nevertheless logically possible, and hence impossible to deny. And yet, we instinctively understand that using this as a basis for belief is problematic, to the extent that those who do believe get categorised by a fancy syndrome. Bertrand Russell was famously making a point something like this when he pointed out it was not possible to rule out the existence of a teapot orbiting in space, yet this in itself does not appear to justify the belief. It still feels much more reasonable to say there is no such teapot than to say there is. In the absence of evidence, it would appear that believing in absence makes good sense.

In the same way, we don’t believe in the Loch Ness Monster, or fairies in the garden or any other manner of fancies, and are probably quite justified in resisting their existence. Show me the evidence, we say, and I’m prepared to be swayed. Until then, it’s a better bet to guess/believe they’re not real.

Occam’s razor offers a similar principle. It suggests that the best policy in forming views of reality is to limit ourselves to concepts that are absolutely necessary to explain known phenomena. To sanction invention beyond necessity (so choosing to believe in the teapot when there is no known phenomenon that needs the teapot to exist in order for it to be explained) is to invite unnecessary error.

So, if no to teapots, alien abductions and fairies, why not no to God? Richard Dawkins makes exactly this point when he comes after agnostics like me. You’re agnostic about God, in a misguided bid to be fair minded, he challenges, so why not be agnostic about the teapot too? Come right out and say you do not share the common belief that there is no teapot orbiting in space.

So, does the argument work? As it’s an argument not just against theism, but against agnosticism too, I ought to be able to show it doesn’t. Where might the weaknesses be?

For me there are two major objections to the approach. The first is that, as is so often the case with analogies, the comparison isn’t entirely a fair one. The teapot exists in splendid isolation, invented as a possibility for the sheer fun (or demonstrative power) of it. The God hypothesis isn’t exactly like that. Rather God might be offered as a solution to a conundrum, a way of viewing the world that offers a solution to a particular set of problems (existence, the meaning of morality, consciousness etc). A sort of metaphysical Higgs Boson (prior to the latest results). And, if God stands as one possible solution, no-God stands as the other. We night think then, not of the case of a positive belief being challenged by the sceptic (with the burden of proof lying with the believer) but rather as two competing explanations, each therefore requiring to establish itself, and equally vulnerable to the teapot challenge. A strong atheist isn’t just avoiding belief in God, they’re actively believing in no-God. Atheism is established not just through the negation of the case for God, but in the assertion of a positive alternative, a universe that exists without cause or purpose. At the point we say ‘I don’t know why the universe exists, I just know there was no God involved’ the theist is, I think, entitled to ask how you know that. What we have, therefore, is something of a double-spouted teapot, one that can pour it’s scalding doubt in either direction.

The second counter-argument, somewhat obviously, is that the whole case hinges upon what counts as evidence. What if a particular space mission reported seeing a teapot out their window, but by the time they scrambled to get a photograph, it had disappeared? Later, a second mission reported the same thing, but again failed to get a photograph. Does that make the teapot any more likely? I suspect it does (without making it certain, we might speculate an optical illusion, group delusion, prank…) Regardless of our own beliefs, we need to accept that a great many sincere people report experiences that they consider to be powerful evidence of a God figure. We sceptically disposed types immediately think of the other possibilities, but it perhaps a step too far to say there is no evidence. Rather, as in the case of the astronauts, we have no verifiable evidence. The difference between verifiable and non-verifiable evidence is, to me, crucially important in this debate, and I promise to devote a post to it.

Are these two objections sufficient to dismiss the ‘no-evidence equals no justifiable belief’ case against God? I’d say yes, but with an important caveat. Burk noted in his comments on the last post that it’s interesting how there are so very few deists about, those who believe there’s something out there beyond our understanding, responsible for the way things are, but who don’t take the next step of naming and personalising it. It’s not just that people believe in God, they often believe in a very specific and detailed version of that God. My hunch is that, with regard to objection one, the teapot case could still be levelled against specific versions of God belief, in that these beliefs contain teapot like features, details that are not strictly necessary to answer the big ‘why are we here?’ questions (in that a stripped down version would still provide an answer.) A bit like claiming, prior to the confirmation, that not only does the Higgs Boson exist, but its real name is Bob.

A similar case might be constructed against the second objection. If we begin to think of personal revelations in terms of their details, we find that in many cases these revelations are in conflict, perhaps even contradictory. It’s like the astronaut looks out the window and says, look, a teapot, and then the second astronaut says, no, it’s a cow. It’s fair to consider this diminishes the credibility of the teapot sighting, and send us in search of alternative explanations.

So, my take on this, not surprisingly, is that the no-evidence principle doesn’t demand atheism. There’s still room for those like me who deny the teapot but remain open to the possibility of a God. As always, I’m open to refutations.

13 thoughts on “Now you see it, now you don’t.

  1. The problem with accepting the evidence of people who claim to have seen god is that it often seems a bit too convenient, a case of wish-fulfilment, like seeing a mirage in the desert. They are seeing what they want to see. When it is something arbitrary, like a kettle outside an air craft, we are more likely to take it seriously, because there is no reason why they would want to see that.

    On the whole an interesting argument. And I agree with your main point, the atheist is lacking evidence of his case for no-God, just as much as the theist is in his case for God.

  2. Burk says:


    Excellent discussion.

    “Rather God might be offered as a solution to a conundrum, a way of viewing the world that offers a solution to a particular set of problems (existence, the meaning of morality, consciousness etc). “

    I think we have to be very careful about the conundrums being addressed. The reason for morality is not such a big conundrum- we all benefit from moral conventions. Clothing all this with a god has its conveniences, but hardly makes for a “solution”, given the social nature (i.e. written by humans) of everything god tells us to do.

    The reason for the origin of the universe is, conversely, a real conundrum, but solved by a god of very simple characteristics- no need to extend the concept to morality, prayers, beards, and the like.

    And the reason for existence? Another false conundrum, like morality. A pet rock can give us a reason for existence, or a dog, or a child, or .. the possibilities are endless. They tend to be social relationships, and so an imaginary relationship, as we typically think a child may have, can be very powerful as well. Whether all this speaks to actual purpose pre-existing in the universe is quite another question, and one that we have received a uniformly negative answer to, in any aspect about reality that we actually understand.


    I think that a dedicated theist’s approach to the question of your post is to claim that the proposition of god is so popular, so natural, so deeply held and so satisfying in its explantory powers, that it comes under an entirely different class of hypotheses than the tea pot hypothesis. Thus the claim that comparing god to pink ponies and Santa Claus is not only rude in the extreme, but a “category error”. It demands some presumptive respect due to these special characteristics.

    Obviously, I differ. The formal category is highly similar, because the positive evidence is equally absent, while the psychological trappings have strong resemblences. Its special characteristics, in the end, have nothing whatsoever to do with philosophic plausibility.

    Popularity does not make ideas correct. Assuming god due to a god-shaped hole in our mystical/emotional apparatus doesn’t cut it either, especially given all the other known weaknesses that render the likely reason one of hope (or fear) over one of cognition. Indeed, if our cognition were variable in this regard, like we are with color-blindness, it would be a simple matter for those with the “gift” to show others by prophecy and special knowledge the error of their doubt. Such things have been claimed with shameless abandon, only to be brought into total disrepute, to the point that serious theists don’t even attempt this kind of defense, retreating to the position that theism equates with all the finest feelings of art and sociability, thus atheists are closet psychopaths. It is, at this point, a paltry argument.

    The consensus that leads to correct models of reality is of a special kind that arises among those with expertise in the evidence being proffered, dedicated to rigorous, critical examination of all hypotheses. It is not a matter of democracy, or of the most shrill or of the “deepest” believers carrying the field. Strength of belief may only make such belief more dangerous if it is wrong.

    This is far from saying the the case demands atheism- that is a conclusion about the psychological landscape that I would leave to each individual. Cosmically speaking, we have no idea.

  3. Hi Jonathan

    A good example, I think. I want to look at the whole nature of evidence, and how we respond to it, in a further post, but you’re right to note that we do need to be careful how we interpret evidence.


  4. Thanks Burk

    I want to address both the morality argument, and the ‘why something not nothing?’ case over the next two posts, when I look at why the theism case is not a rationally compelling one either, and agree that in the morality case in particular, other solutions present themselves.


  5. Edwin says:

    You claim that God is a problem to a specific set of conundrums, but as I see it, all of your presented examples have already been solved by modern science – morality and consciousness in social sciences, neurology, and evolution, and existence in physics. The conundrum therefore only exists for those choosing to disbelieve scientific facts, for others, there is no problem needing to be solved.

    There is of course the argument that God is not contradictory with science, but rather he “started the Big Bang and then left us”. To this, I can only say that atheists are not claiming that we know the universe as we see it is not the creation of a sentient being or beings, because we don’t. We are however claiming that the universe follows a set of rules that we call the laws of physics, and there exists no supernatural deity which transcends these. This may not be provable, but it is entirely logical to make this assumption in the fact that there exists no reason to disbelieve this and it has held true for everything in the past.

    For your other solution, that of non-verifiable evidence, there exists countless sightings of the Loch Ness Monster and other cryptids, of magic and of dragons, of every being of fantasy and imagination. There are people who believe in such things, but the generally accepted belief is that these do not exist, in the lack of any real evidence. So it is also with God.

    On another point, since most of your readers here are atheists, we’re seeing a lot of argument from one side and none from the other. Any theists care to step up and argue their point of view in the comment section?

  6. Hi Edwin

    I think you’re right to suggest alternative solutions exist for most of these problems, and I want to dwell upon the morality one in particular in a further post. Why anything exists at all, I would suggest, doesn’t yet fall into that category. My reasoning is, while physics might show how, given physical parameters, universes might come into existence, it doesn’t answer the, why these parameters and not others? question. There’s a distance to go with consciousness, too, I suspect. Will we one day have an entirely satisfactory model of these things? I don’t know. Can such a model be built with reference to an outside power, I don’t know that either, so I’m agnostic with regard to such questions. I have no urge to pre-judge.

    There are of course many stripes of atheist, and many ways of defining it. For me, the atheist who says, ‘I don’t know if there was some sentient being who created the universe’ sounds very much like what I’d call an agnostic, but as I said a while back, these are fluid terms.

    With regard to non-verifiable evidence, I find it hard to see how one can rule on this one without pre-judging the issue.

    My next two posts will cover why I think two very common cases in support of theism don’t work, which may well change the make-up of the comments. A litlte polite to and fro could be illuminating. We’ll see.


  7. Edwin says:

    My comment on a sentient being was a reference to science fiction, but it’s based upon the fact that we know humanity could theoretically progress to the point of being able to create an entirely contained virtual reality with actual sentient beings within their own world that is as complex as our own, and therefore there exists the possibility – unlikely, but theoretically possible – that our own so-called reality is an equivalent. Would you call them a God? I would not, because a deity as I see it needs to have some supernatural power, some mystical force that humanity can never attain. Again, it’s a matter of definition.

    • Ned Grenfell says:

      Dear Bernard
      Yes the God entity may be a plausible answer to existence but that is what the God entity must remain in order for the belief in God and the belief in No God to be equally likely and therefor justified by agnosticism. However once Religion extrapolates a plausible answer for existence into something personal and attached to humanity the God entity becomes something not rational but ridiculous. Seeing as the God entity serves only as a possible answer to a few questions (that grow smaller in number as science advances) then the God entity must remain ex hypothesis. One cannot claim as Religion does that God is beyond Human comprehension and then make hypothesis about the nature of God. In situations where the God entity has been tailored to suit the needs of those involved in Religion then the God entity must be refuted. The argument between a divine creator and the non existence of a divine creator is one that is entirely rational to be agnostic. However the argument between the existence of a vengeful, attention seeking, demanding yet loving, caring God and the non existence of such a thing should is not a disagreement in which Agnosticism is appropriate. If one feels the need to view God as the most plausible reason for existence then so be it, their stance is a defendable and semi-rational one. However if one then molds the God entity into something far more complex and personal, then they need to be reigned in and held accountable for their same misguided ideals, in the same way one would reign in any true teapot believer.
      Thanks, Ned

      • Hi Ned

        Yes, it’s important to be clear about what precisely one is agnostic. So I’m agnostic about the purpose, if any, of creation, but not agnostic about fairies in my garden, or the restorative powers of the market place, for reasons much as those you outline. An interesting question is what happens if you turn your critique on its head and ask what warrant the atheist then has to personalise their own response to a Godless universe? Are descriptions like purposeless, or morally neutral or whatever, also projections of human needs rather than valid extrapolations from the initial commitment? There then becomes a further question, I think, about what warrant we may give ourselves to believe, once we have jumped off the agnostic fence and committed to any particular existence or non-existence.


  8. Hi Ned

    Now that I have more than three minutes to shape a reply, let me expand slightly. Take as the starting point the question ‘is existence purposeful?’ I’m agnostic on this, but understand full well that many, indeed most, choose to commit to an answer, one way or another, and I don’t think one can be called irrational.

    Now, imagine you have as a foundational belief, the belief that you have been created as part of some purposeful plan. Is it perhaps, at this point, a reasonable leap to consider one’s own deepest intuitions are indeed a guide to this purposeful reality? I’m not sure it’s an off the chart nuts response, if purpose is your first premise.

    If one takes as their opening premise the belief that existence is purposeless, then it’s a very simple leap to interpret religious intuition as an evolved response, as a suspiciously human construction designed to help us cope with the essential fear of being a self aware mortal.

    What I’m not sure you can do is critique the theistic extrapolation from the atheistic premise, any more than the theist critiquing the atheist because their conclusions are not in line with theistic premises, makes any sense. Thus, it might be that a test to one’s commitment to agnosticism is the ability to accept not just alternative foundational beliefs as plausible, but also to accept the conclusions that might flow naturally from them.


    • Ned Grenfell says:


      It is an interesting point about the leap of intuition and one that I had never fully considered so thank you for that. What I have trouble with is where does the leap end, at which stage does one consider the infringement that belief may have on other areas of Humanity such as science and even laws and then stop. By saying that just because one is justified to make certain intuitions about a their theory of existence do you then say that any creation myths and stories (I am speaking when they are taken literally rather then metaphorically) are something to remain agnostic about. In my opinion The God entity is something to remain agnostic about when it remains an entity to whom one gains moral values and purpose and that is all. But when one claims that the God entity performs miracles that defy physics or when the God entity supposedly stands in opposition with evaluation then do you still remain agnostic about that specific version of that religions God. By no means do I wish to make attacks on the metaphorical believers of the Bible but the ones who believe in the Bible in a literal sense, is their idea of God still something to remain agnostic about ?

      • Hey Ned

        For me, the point where intuitions lead one to make predictions that are at odds with our most reliable physical models, is the point where agnosticism ought to end. If somebody believes that praying for their ailing child is a better bet than chemotherapy, for example, then I’m fairly comfortable with the authorities intervening and usurping the parent’s rights. It is one thing to back ones intuition when all we have to put it up against is the intuition of others, but quite another to put it up against models whose accuracy is well established. Clearly, a great deal of that which masquerades as religion is plainly nuts. Then again, a great deal of what masquerades as rationalism is pretty dodgy too, so the goal is to set up lines of delineation that don’t lead to us throwing the baby out with the bath water.

  9. Ned Grenfell says:

    Thanks a lot for clearing that up, up till now that was possibly my biggest problem with committing fully to agnosticism. Many Atheists out their need to realize that when religion is stripped back of all its balls and whistles of irrationality it is actually not too much of an obscene idea but it saying that I still believe it remains an idea without enough evidence to decline or commitment.
    Many thanks for the time spent explaining

Leave a Reply

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

%d bloggers like this: