The Problem of Divine Hiddenness

In this blog I have surveyed what I believe to be seven (or so, depending on how you count them) fairly strong arguments for the existence of God. But it is nonetheless true that God’s existence is not completely obvious to most people in the way that, say, the existence of the physical world is obvious. So why is the evidence for God’s existence not more direct? Why doesn’t God make his presence as clear as day to everyone – especially when he supposedly wants people to know him?

This is the problem of divine hiddenness, also called the argument from non-belief, and to me it is one of the two strongest arguments for atheism (together with the problem of evil, which I will discuss in the next post). It can be formulated simply as follows:

  • If God exists, then he is perfectly loving.
  • If God is perfectly loving, he would make it so that every person believes that he exists.
  • Some persons do not believe that God exists.
  • Therefore, God does not exist.

The gist of the argument is that, if God existed, he would want to be in a loving relationship with every person he created. But a precondition for being in a loving relationship with God is believing that God exists. So, God would ensure that everyone is aware of his existence in order for it to be at least possible for them to enter into a relationship with him.

Since in fact it does not appear to be the case that everyone is aware of God’s existence, this line of reasoning lends support to the belief that God does not exist.

Evaluating the Argument

As a theist who finds the axiological and ontological arguments for God compelling, I am in complete agreement with the first premise: if God exists, then he is perfectly loving. So we can take that as a given.

The third premise, I believe, is also fairly obviously true. There are Christians who claim (on the basis of a couple verses in the Bible) that deep down, everyone really believes that God exists: no one is truly an atheist, and anyone who claims they are is lying, even to themselves. I have come to think that this is a very mistaken response. It is extremely uncharitable, it actually isn’t well supported by Scripture, and it isn’t well supported by the testimony of many current and former non-theists. See this article by Randal Rauser for more on that. So I also accept the premise that there are real atheists and agnostics out there.

The second premise is more questionable. Here is how it might be justified. Since God is perfectly loving, he desires the best for everyone. Since God is the locus of all value, being in a personal, loving relationship with God is the greatest good that anyone can experience. So God desires for everyone to be in a right relationship with him. And more simply, since he loves everyone, he desires relationship with them for its own sake: and he would certainly reach out to them rather than abandoning them to an existence devoid of the goodness of his presence.

Now, if belief in God is indeed a precondition for right relationship with him, then we can make the inference from “God desires for everyone to be in a right relationship with him” to “God desires for everyone to believe that he exists”. And I think it is certainly the case that, for the kind of deep and reciprocal relationship that God ultimately desires us to have, belief is required. However:

  • Not all persons may be capable of belief in God. (For example, infants or cognitively impaired persons.) Nevertheless, it may be that such persons can still have a kind of relationship with God, the way that even an unborn child has a kind of relationship with her mother.
  • Since God is not just one personal being among others, but is also the ground and locus of all value, it may be possible to have a positive relationship with God even without believing in him, by relating to Goodness, Truth, and Beauty.
  • If God has foreknowledge of our free choices, he knows whether producing belief would lead to the kind of positive relationship that he is interested in. There may be those who, upon coming to believe that God exists, would still reject relationship with him, enter into an improper relationship with him, or enter into a relationship with him which they would ultimately choose to forsake. (Such persons may or may not be culpable for being in these contrary states; there can be a variety of reasons for such dispositions.) God’s reasons for producing belief in these cases are much reduced.
  • Similar to the above point, there may be people to whom God has given sufficient rational grounds for belief in his existence, who have freely and culpably rejected those grounds, or who have shut themselves off from God in some other way (for example, by refusing to seek God because of a desire to be in control of their own life). Not desiring a coerced relationship, God’s reasons for forcing belief on such persons are also reduced.

This means there is some reason to doubt that God’s desire for relationship automatically leads to God’s desire to produce belief in his existence. Furthermore, even if God desires for someone to believe that he exists, that does not imply that he has an all-things-considered desire for that person to believe that he exists. God may have other considerations, some in favour of allowing created persons to remain unaware of his existence, at least for a time. Here are some reasons for divine hiddenness:

  • Delaying in making evidence available for a person to believe in God may alter their circumstances in such a way that they enter into a higher quality relationship with him later than they would have if he had made such evidence available earlier. (Daniel Howard-Snyder argues along these lines in this paper.)
  • Divine hiddenness may allow for greater independence and interdependence of creatures, benefiting our moral development. (Dustin Crummett makes this suggestion here.)
  • Divine hiddenness may reduce coercion in some situations and allow for more opportunity to freely choose what is good for the right reasons (i.e. because it is good, not just for fear of punishment, for example).
  • Perhaps it is appropriate to God’s holiness to maintain a certain degree of distance from his creatures.
  • Because of the butterfly effect, divine hiddenness could lead to better possible outcomes in ways that are totally unpredictable from a human perspective, but that can be foreseen by an all-knowing God.

(Take a look here for a resource that explores some of these possibilities further, with references to other philosophical works.)

In evaluating why God may remain hidden from some people, we also have to consider that if God exists, then human persons may have an eternity before them in which to relate to God: our mortal lives may just be an infinitesimal sliver of the whole of our existence. Which means that for all we know, God’s sacrificing some depth of relationship for a short time in order to obtain other goods may very well be justified.


Given these considerations, I am far from certain that the second premise of the problem of divine hiddenness can be sufficiently supported. Perhaps we could modify it:

  • If God is perfectly loving, then for every person A who is capable of entering into a personal relationship with God, such that A has not culpably shut themselves off from relationship with God in some way, God brings it about that A believes in his existence, unless he has an overriding reason to permit A to remain in non-belief.

I can accept this premise with reasonable certainty. But then we need to modify the third premise in order for the argument to remain valid:

  • There is at least one person who has not culpably shut themselves off from relationship with God, and who has not come to believe in God’s existence; and God has no overriding reason to allow this person to remain in unbelief.

And this premise is quite plausibly false. At the very least, it seems impossible to prove from a human perspective: we cannot possibly be in a position to adequately evaluate all of God’s reasons for doing something, and we can think of a few reasons in favour of divine hiddenness that cannot be ruled out conclusively.

(Note: There is, perhaps, one very important consideration which the proponent of the hiddenness argument could raise to say that God cannot have an overriding reason to allow anyone to indefinitely remain in unbelief: the person’s eternal salvation depending on having a right relationship with God. This gets into the problems of religious pluralism and exclusivity, which I will discuss in upcoming posts.)

Thus, I think it is quite reasonable to disbelieve this modified premise (provided the theist has a response to the problems of religious pluralism and exclusivity). Which means that, while the problem of divine hiddenness introduces some tension into a theistic worldview, in the end I do not find that tension to be insurmountable. After all, there is evidence for God’s existence, so he is not and does not remain completely hidden. And it may very well be that this evidence is enough to accomplish God’s purposes in this world: enough for people to seek him and begin a loving relationship with him, if they so choose.

Obviously, I have only scratched the surface of this topic, and there is a great deal of philosophical work that has been done (and that is still being done) to explore it further. I have to admit that in this post I haven’t dealt with the problem of divine hiddenness in as much depth as I’d like. In the future, I may come back to expand on what I’ve written here. For now, I want to keep moving on to new topics (and I don’t have as much time for blogging these days), so I am content with this overview.

The Presumption of Atheism

Do we really need a reason to not believe in God? Isn’t atheism the default position?

To answer this question we need to distinguish between believing that God does not exist on one hand, and not believing that God exists on the other. The former is the belief in an absence. The latter is simply the absence of a belief.

“Atheism” is commonly defined as the belief that God does not exist, but there are a growing number of people who define it in a weaker sense as the state of not believing that God exists (beginning with Antony Flew, who first suggested to use “atheism” in this way). For clarity, I will call the former sense “positive atheism” and the latter sense “negative atheism”.

So then, is atheism the default position? The answer depends on whether you mean positive or negative atheism.

If you mean negative atheism, then yes! I am in complete agreement that negative atheism is the default epistemic state. One does not need arguments or evidence to be a negative atheist, because negative atheism isn’t a belief at all. Being merely the lack of a belief, it does not assert that anything is true, and so requires no justification. It is just a description of someone’s subjective mental state. Babies, cats, and rocks are all negative atheists, since they do not hold a belief in the existence of God. (At least, we don’t think they do.) Positive atheists are negative atheists, but the converse isn’t necessarily true – agnostics are also negative atheists. (Which is why, if you want to distinguish between atheism and agnosticism, you have to mean positive atheism.)

Negative atheism is trivial. Note that we can just as well define “negative theism” to mean not believing that God does not exist, and then that is also the default position! (Someone can lack both the belief that God exists and the belief that God does not exist, and indeed the babies, cats, and rocks of the last paragraph are in that position – along with the agnostics.)

Because “negative atheism” is so trivial, it is only a useful category in some sociological contexts – in studying the kind of things that people believe, for example. But when we are talking about more than that – when we are talking about whether or not God really exists, not just what people believe about that – what we care about is “positive atheism”.

And there, the situation is quite different. If you claim that positive (not merely negative) atheism is the default position, well, you are simply wrong. Positive atheism is a belief – it takes a stand and asserts that something is true (namely, that God does not exist). And we should have reasons for believing what we do. We should seek to have justified beliefs. Thus, positive atheism requires justification to back up the assertion that it makes, which means it is not the default position.

I think this is worth repeating. The claim that God does not exist is no less strong of an assertion about the nature of reality than the claim that he does. So atheism (that is, positive atheism) is not an epistemically neutral stance. In order to be an atheist in any philosophically significant sense of the word, you have to have reasons.

To put a bit more rhetoric behind this statement: the atheist who says that he needs no evidence for atheism (because it is the “default”) is no better than the fideist who says that he needs no evidence for theism (because he accepts it by “faith”). Worse, even: at least the fideist usually recognizes that his position is irrational.

You’ll find atheists who say that atheism is not a claim, and therefore it has no burden of proof. But this can only be true of negative atheism: if said about positive atheism, it is blatantly false. The distinction between theism and atheism is not that one claims something and the other doesn’t. Rather, the distinction is that one is a positive existential claim (it claims that something exists) and the other is a negative existential claim (it claims that something does not exist).

And there is nothing about negative existential claims which makes them exceptions to the rule that we should have rational justification for what we believe. For comparison: philosophical idealism makes a negative existential claim, but that does not make it the default position, nor does it allow us to deny the reality of the physical world without valid reasons.

Reasons for Atheism

The saying “you can’t prove a negative” becomes relevant here. Some people assume that this is true, and that it means that justification for a negative existential claim can’t actually be given, letting atheism off the hook. But this saying is false. There are perfectly good ways to justify negative existential claims. There are, in fact, some reasons to believe atheism. Therefore, the requirement that atheists have reasons to be atheists is not unreasonable.

First, though, here are a few statements that are not reasons to be an atheist:

  • “There isn’t any evidence for God.”

If by “evidence” you mean the kind of natural, physical evidences that we study with science, then you’re looking in the wrong place. It presupposes atheism or philosophical materialism to insist that all justification for beliefs has to be evidence of this kind. Moreover, it is self-refuting, since we rely on certain non-evidential beliefs in order to acquire and use evidence at all. Construing “evidence” more broadly to mean any rational justification for a belief, this objection is false. See: all my previous posts on arguments for God’s existence.

  • “None of the arguments for God’s existence are convincing.”

This leaves you without a reason to believe that God exists, but it doesn’t give any reason to believe that he doesn’t. So at best, this leaves you with agnosticism. If you say you can be an atheist merely because you haven’t been convinced about theism, then you might as well say you can believe anything you want as long as it is not obviously false. I think rational belief requires a little bit more than that.

  • “God isn’t a valid explanation of anything.”

This objection is sometimes used against the cosmological or teleological arguments, saying that the God hypothesis is really a non-explanation. But for one, this objection is false: if God exists and created the universe, then God obviously would be an explanation for the existence of the universe. (And so the claim that God is a non-explanation just presupposes atheism, without justification.) And for two, even if true, this objection would only negate reasons for theism, rather than providing reasons for atheism.

(See also here for a further exploration of the not-a-valid-explanation objection.)

  • “Science shows that God doesn’t exist.” Or, “Science shows that we don’t need to posit God to explain anything.”

No, it doesn’t. I’ve written about this before. But by way of brief response: science allows us to explain a lot of things via natural causes, but it cannot explain everything, and it certainly cannot explain where those natural causes themselves come from. (To take up that question is to leave the realm of empirical science and go deep into the realm of philosophy.)

With those out of the way, here is one statement that is a valid reason to be an atheist (if true):

  • “There isn’t enough evidence for God’s existence, compared to the amount that there would be if he did exist.”

Absence of evidence is not evidence of absence, as the saying goes – unless we should have that evidence if the thing did in fact exist. (For a more in-depth exploration of that principle, see here.) Where this is used as an argument for atheism, it is known as the problem of divine hiddenness. More generally, we can provide justification for negative existential claims in the following way:

  • “If X exists, then Y should be true. But Y is false. Therefore, X does not exist.”

This is the logical form behind the problem of evil and other arguments against God’s existence. These are the kind of arguments that can actually justify atheism, and they are what I will be exploring in upcoming posts.

Implausibility of Theism

There is one further argument that I will address which is relevant to the question of whether or not atheism can be considered the default epistemic position. It is raised by Paul Draper in his article on atheism and agnosticism in the Stanford Encyclopedia of Philosophy. The idea behind this “low priors” argument is that theism is intrinsically improbable, so that even before taking any evidence into account, the presumption is in favour of atheism.

Draper argues for the low prior probability of omni-theism (his term for the kind of theism that I believe follows from the arguments for natural theology) by comparing it to an alternative position that he calls source physicalism:

  • Source physicalism: the physical world is ontologically ultimate, and any mental reality depends on the physical for its existence.
  • Source idealism: the mental world is ontologically ultimate, and physical reality depends on the mental for its existence.
  • Omni-theism: a subset of source idealism, which further specifies the ontological ultimate to be a unique, perfect person.

Draper’s claim is that 1) source physicalism and source idealism have equal intrinsic probabilities, principally because of the symmetry between the two positions, and 2) omni-theism is a very specific version of source idealism and therefore it must be much less intrinsically probable than source idealism in general. From this it follows that 3) omni-theism is much less intrinsically probable than source physicalism.

These claims are, of course, contestable. Despite Draper’s assertion, there is an asymmetry between source physicalism and source idealism, both in regards to our epistemic position (we can doubt the existence of the physical world, but we cannot rationally doubt the existence of our own minds), and in regards to their ontological status (there are mental phenomena that do not seem possible to derive from the physical world even in principle). It seems to me that these are a priori considerations that could tip the balance in favour of source idealism in terms of its intrinsic probability, over source physicalism.

Second, it seems to me that Draper’s claim about the specificity of omni-theism, compared to source idealism, is greatly exaggerated. Over and above source idealism, omni-theism simply claims that the ontological ultimate is personal instead of impersonal, one instead of many, and perfect instead of imperfect. The first of these additional claims arguably has the advantage due to similar considerations as those in the last paragraph. And the latter two claims are in some sense simpler than their counterparts, and could claim greater intrinsic probability on that basis.

Finally, omni-theism is actually a very simple hypothesis: it postulates the existence of a single entity (an immaterial person) who can be described by perhaps just three properties (knowledge, power, and goodness) and who possesses those properties in the simplest possible way (to the maximum degree, with no limitations). Compare this to Draper’s source physicalism, which has to postulate the brute existence of a number of physical entities, laws, constants, and initial conditions (or some kind of mechanism for generating such). So I would say that considerations of simplicity actually tip the balance of intrinsic probability in favour of theism!

So I don’t believe that Draper is successful in his argument that omni-theism is many times less probable, intrinsically, than source physicalism. Which means that is not a good reason to take atheism as the “default position”.


An aside on Draper’s arguments in the article referenced above:

I might add that Draper’s complete “low priors” argument also includes the premise that the total evidence does not favour omni-theism over source physicalism. (Where “evidence” in this context means any rational source of justification for a belief – experience, arguments, etc.) Combining this with the premise that omni-theism is intrinsically much less probable than source physicalism, he concludes that omni-theism is very probably false, and thus, atheism (understood just as the denial of omni-theism) is very probably true.

As I said, I think there is reason to doubt the premise about the intrinsic probabilities of omni-theism and source physicalism. But even if that were established, I absolutely disagree with this claim about the balance of the evidence. Justifying it requires one to undercut or defeat all of the arguments for omni-theism from natural theology, or at least to balance them with equally strong arguments for source physicalism. And that, from my perspective, is a very difficult task.

(Draper’s appeal to the “fallacy of understated evidence” is a good starting point for discussion about potential weaknesses or balancing considerations of certain theistic arguments, but it in no way shows that those considerations actually succeed in reducing the force of those arguments by a significant degree.)

Interestingly, the failure of this premise is related in a way to the potential success of the other one. One of the ways that Draper tries to justify his premise about the intrinsic improbability of omni-theism is by saying that omni-theism makes very specific claims, each of which must reduce its probability. But if we’re being rational, we theists don’t just make those claims out of the blue. We make specific claims about God and reality because of reasons that we see for them (for example, from natural theology). So in the end, theism’s level of intrinsic probability doesn’t really matter. The important thing is its probability when the evidence is taken into account.

The put it another way, omni-theism is not postulated a priori as an arbitrary refinement of source idealism, any more than the Standard Model of particle physics is postulated a priori as a refinement of source physicalism. These things are postulated as hypotheses to account for facts that need to be explained. So the fact that omni-theism is more specific than source idealism is no more of a strike against theism, to my mind, than is the fact that the Standard Model is more specific than source physicalism a strike against atheism.

One more note: Draper has a second argument for atheism in that article, his “decisive evidence” argument. This one isn’t really a distinct argument, since its premises basically assert that 1) none of the argument for theism are successful, and 2) there is an argument for atheism that is successful. (There’s more nuance to it than that, but when you really break it down the premises are justified only if those two things are true.)

End aside.


So far, I’ve explored two reasons that are often put forward for rejecting theism in favour of atheism: arguments that theism is impossible, and the presumption of atheism as the default epistemic position. And in my honest evaluation, neither of these carry any rational weight whatsoever.

In the next few weeks I will explore some atheistic arguments that I think do have force.