Home » Library » Modern Library » Nonbelief as Support for Atheism

Nonbelief as Support for Atheism


[This paper was originally presented as a talk at the “World Congress of Philosophy” at Boston University in August 1998.]


The Canadian philosopher J. L. Schellenberg has recently put forward an argument for atheism based on the idea that God is supposed to be perfectly loving and so would not permit people to be deprived of awareness of his existence. If such a deity were to exist, then, he would do something to reveal his existence clearly to people, thereby causing them to become theists. Thus, the fact that there are so many nontheists in the world becomes good reason to deny the existence of God conceived of in the given way. I first raise objections to Schellenberg’s formulation of the argument and then suggest some improvements. My main improvement is to include among the divine attributes the property of strongly desiring humanity’s love. Since to love God requires at least believing that he exists, if God were to exist, he must want widespread theistic belief. The fact that so many people lack such belief becomes a good argument for atheism with respect to God conceived of in the given way. Some objections to this line of reasoning are considered, in particular the claim that God refrains from revealing himself to people in order to avoid interfering with their free will or to avoid eliciting inappropriate responses from them or some other (unknown) purpose. An attempt is made to refute each of these objections.

Atheism of a certain sort can be supported by appeal to the existence of widespread nonbelief in God. This is shown by a Canadian philosopher, J. L. Schellenberg, in his book Divine Hiddenness and Human Reason. His argument is as follows:

  1. If there is a God, he is perfectly loving.
  2. If a perfectly loving God exists, reasonable nonbelief does not occur.
  3. [But] reasonable nonbelief occurs.
  4. [Thus] no perfectly loving God exists. [(from 2 & 3)]
  5. [Hence] there is no God. [(from 1 & 4)][1]

In this paper I shall first raise some objections to the argument and then try to show how it might be improved.


Irrelevance of the Term “Reasonable”

Schellenberg regards “reasonable” nonbelief to be that which is inculpable (i.e., for which the nonbeliever is not to blame). The distinction between culpable and inculpable nonbelief is somewhat unclear, but even if it could be sufficiently understood I am inclined to say that all nonbelief in God is inculpable. For that reason, I would accept premise 3 of the argument. Schellenberg devotes a chapter of his book to a defense of it. Although I agree with what he says there, it seems to me that even if people’s nonbelief in God were always somehow their own fault, that would be irrelevant. A perfectly loving deity would set vindictiveness aside and still want to help nonbelievers (by supplying them with evidence of his existence), despite their culpability. All it would take, for most, would be some spectacular miracle, or perhaps, as Schellenberg prefers, a religious experience. So, even if some clear sense could be attached to the distinction between culpable and inculpable nonbelief, the real force of Schellenberg’s argument would lie in the fact of nonbelief itself. The issue of whether or not the nonbelief is culpable would be irrelevant. It would therefore improve the argument if the word “reasonable” were simply omitted from it.

Premise 1

Premise 1 should be rejected because there are theists who do not view God as perfectly loving. Even some Christians think of him as an angry deity bent on punishing people for their sins. Consider for example the biblical god who ordered the total annihilation of the Amalekites (including all their children and animals), or one who would predestine some of his creatures to eternal torment in Hell (as commonly believed). Such a deity is not perfectly loving. The property of being perfectly loving is too specific, which is why lexicographers usually omit it in their definitions of “God.” To get around this, Schellenberg should put his argument forward only in relation to theists who already think of God as perfectly loving and so would accept his first premise.

Premise 2

My main objection is to premise 2. Even assuming that God is perfectly loving, why must he want people to believe in him? There are circumstances in which a man could love his children and yet not want them to be aware of his existence. Perhaps he is a moody person with periods of violent behavior. Or maybe he is hiding from bomb-throwing assassins, so anyone close to him is at risk. There is nothing in the concept of love itself that would warrant the inference drawn in premise 2. There is no contradiction in the statement “X loves Y, but X does not want Y to be aware of X’s existence.” The idea of “loving from afar” is a familiar theme in literature.

Schellenberg might object that these considerations are irrelevant because God has additional properties which forestall them. Since God is by definition all-good, it has to be in people’s best interests to be aware of such a being if indeed he exists. But what does “all-good” mean? And why must it be in people’s best interests to be aware of such a being? There are many nontheists who lack such an awareness and yet seem to be happy anyway. What is the great benefit that such people are missing? This needs clarification. Another consideration is that God may be so far beyond us that we are incapable of comprehending him. Maybe theists have such a poor understanding of God that there is no significant difference between them and nontheists, so far as God is concerned. In other words, people’s nonbelief does not bother God because the only sort of belief of which humans are capable is of no value to him. Yet, he might be perfectly loving, notwithstanding.

My conclusion regarding Schellenberg’s argument is that, as worded, it is a failure. It should have dispensed with the unclear and irrelevant appeal to the culpability of nonbelief. To make a case for premise 1, it should have been exclusively applied to some specific deity (e.g., the God of evangelical Christianity) rather than God in general. And in place of its premise 2, it should have gone beyond the concept of divine love, perhaps making an appeal to scripture or to some additional properties of God. Overall, I would say that the basic idea of an argument from nonbelief is sound, and it does indeed provide support for a certain form of atheism, but it needs to be formulated in a different way.


We first need to recognize that there are different concepts of God and for some of them people’s nonbelief is not a serious problem. One could simply assert that God does not care whether anyone believes in him and that would end the matter right there. However, there are many theists for whom people’s nonbelief would be a serious problem. We could use survey questions in an effort to locate them. I suggest questions such as the following:

Q1: Does God have great love for humanity?

Q2: Does God strongly desire that humanity love him?

I think that more than half the theist population, at least of the United States, would answer both questions affirmatively, and for those theists there is a problem of nonbelief, which in this context is the problem of explaining why there are so many nontheists in the world. If God wants the nontheists to love him, then he must also want them to believe that he exists, since such love requires such belief. So, why, then, hasn’t he simply appeared to them or done something else that would have effectively eliminated their nonbelief? It seems that no reasonable explanation can be given here.

That gives rise to an atheological argument, similar to Schellenberg’s, which may be called “the argument from nonbelief” (ANB). Here is a formulation of it:

  1. If God were to exist, then he would have great love for humanity and a strong desire that humanity love him in return.
  2. If such a deity as described were to exist, then probably all, or almost all, present-day humans would believe that God exists.
  3. But many present-day humans do not believe that God exists.
  4. Hence, probably there does not exist a deity as described in 1. (from 2 & 3)
  5. Therefore, probably God does not exist. (from 1 & 4)

This argument is not to be applied to God in general, only to that specific concept of God with respect to which its first premise would be true. That would include the concept of all those theists who answer both survey questions Q1 and Q2 affirmatively, which I believe would be the majority of them, at least in the United States. It thus aims only at atheism with respect to that specific concept of God.

The rationale behind ANB’s premise 2 is that people cannot love God if they do not believe that he exists. Thus, if God exists then he must want people to have such belief, which implies that he would have done something to bring about universal (or near-universal) theistic belief, even if that would require giving humans some sort of “brain-boost” to help them comprehend his nature. The appeal to God’s desire for humanity’s love fills out what was missing in Schellenberg’s premise 2. In my opinion, ANB presents good evidence for the nonexistence of any deity which satisfies premise 1. However, defenses might be mounted against it.

Consider, first, the free-will defense. Might God refrain from clearly revealing himself to people in order to avoid interfering with their free will? I think not. People’s free will is not affected by them merely learning or being shown the truth about something, even by God. For one thing, people do not normally use their wills in the process of belief acquisition but rather rely on the available evidence. For another thing, they want to know the truth and therefore would not be forced against their will to acquire it if they were to be shown something. Assuming that God exists, for nontheists to become aware of that truth would actually make them more free than they were before, for it would open up options to them that were not available before. Schellenberg thinks that if God were to exist then he would reveal his existence to people by providing them with some sort of religious experience.[2] Maybe that would work, but I shall not pursue that line of thought. There are a great variety of ways by which God might impart knowledge about himself. One way would be by appearing to people and performing spectacular miracles. Another would be by inspiring humans to write scripture that possesses special properties, showing it to be divinely inspired. God could then help disseminate knowledge of such scripture worldwide. We need not pursue the details of this. It is counterintuitive to claim that God cannot do any of these things without interfering with people’s free will.

Another objection to the free-will defense makes an appeal to the idea of irrationality. There are well over a billion people on the planet earth who lack a belief in God. And with the great proliferation of conflicting belief systems even among those who are theists, presumably billions of them ascribe properties to God which he does not possess. Assuming that God wants all those people to come to believe in him and to come to have correct beliefs about him, how does he expect them to do that if he is unwilling to provide the evidence they need? Does he want them to arrive at the relevant beliefs in some irrational manner? That would certainly conflict with the usual concept of God. Theists may reply that there is evidence for the existence of God and so theistic belief is in no way irrational. But if people are aware of good evidence for God’s existence, then how is their belief in God “free”? Advocates of the free-will defense may say that there is just the right amount of evidence: enough to make their theistic belief rational but not so much as to interfere with their free will. If God were to provide still more evidence of his existence, no matter how slight, then that would cross the line and interfere with free will.

One objection here is that there really is no good evidence for God’s existence, especially when God is given the attributes mentioned in ANB’s premise 1. That is a large topic in itself, which we need not get into. Another objection is that even if there were such evidence, apparently billions of people are unaware of it and are in need of something more. Presumably for God to provide that little extra would not interfere with those people’s free will, for it would only bring their level of awareness up to that of the theists who (with no interference with their free will) already are aware of the given evidence. Thus, there would be no good reason for God to permit nonbelievers to remain that way. If there is a level of evidence sufficient for belief but less than that which would interfere with free will, then God should see to it that everyone is made aware of evidence for his existence which is at that level. There are still other objections which we need not get into. From what has already been said it is clear that the free-will defense will not work.

Another defense, which is very similar, is proposed by Daniel Howard-Snyder in an exchange with J. L. Schellenberg.[3] According to it, God knows that if certain nontheists were to be presented with good evidence for his existence, then they would respond inappropriately. For example, they might come to hate God or be indifferent to him. For that reason, God simply never presents himself to them, thereby permitting them to retain their nonbelief. One objection to this defense, among others, is that a person’s immediate response to theistic belief may not remain fixed. People could come to believe in God and at first respond inappropriately, but after some time has elapsed, they may come to modify their response. Schellenberg makes this point in his reply to Howard-Snyder. He suggests that God, being all-loving, would “seek in various ways to facilitate a better disposition.”[4] I agree with that. Certainly it is hard to see how God could get any better behavior or better disposition from nontheists by having them remain ignorant of his existence than by causing them to become aware of it. I shall omit discussion of other objections. It does seem that the given defense can be refuted in somewhat like the way in which the free-will defense was refuted earlier.

Theists usually fall back on the unknown-purpose defense according to which God has some adequate though unknown purpose for permitting people to be nonbelievers. Schellenberg seems to resist this line of thought, for in his book he says: “Without strong independent evidence for the denial of my argument’s conclusion, S [a person] has no reason to appeal to the possibility of an explanation unknown to her, perhaps beyond her grasp. Without the indirect support for an appeal to the possibility of unknown explanations afforded by strong independent evidence of God’s existence, S must, if she agrees with this argument, come to believe that there is no God.”[5] This seems to be a rejection of the unknown-purpose defense out of hand without due consideration. Someone could agree with the basic points of Schellenberg’s argument and yet suspend judgment on whether or not there might be some divine purpose which would adequately explain why God remains hidden from (at least a large part of) humanity.

We are confining our investigation to theists who believe in a deity who strongly desires that (almost) everyone at least be aware of his existence. What are we to make of the claim that such a deity exists but has some unknown purpose which conflicts with his desire for (near) universal theistic belief among humans and which outweighs and overrides it, thereby falsifying premise 2 of ANB? Is there any good objection to that claim?

Theists who answer questions Q1 and Q2 affirmatively are saying that God loves humanity greatly and wants that love to be reciprocated. But if God were to want people to love him, then it would seem to be counterproductive (and perhaps even irrational) for him to stay hidden from them. Of course, it is logically possible that God desires people’s love and yet have some (unknown) purpose which overrides that desire. Nevertheless, the idea does appear to be counterintuitive.

Another consideration, the focal point of Schellenberg’s argument, is that if God has great love for people, then he presumably wants whatever would benefit them in the long run. But surely that would include a close personal relationship with himself. How in that case could God deny a large portion of humanity such a benefit by staying hidden from them? If he loves people greatly, then presumably he could not permit them to be so deprived. The appeal to the unknown-purpose defense again becomes increasingly far-fetched, though, admittedly, more is needed here to clarify the benefit that nontheists are supposedly missing out on. Perhaps what should be emphasized is the intrinsic value of wisdom, or genuine knowledge of the ultimate nature of reality. If God loves us, then he must want us to have that.


My overall assessment of ANB is as follows. I readily grant that the argument has no force whatever if the concept of God is left unrestricted (using, for example, just the dictionary definition of “God” as the all-powerful creator and ruler of the universe). However, many theists have a more definite concept of God, and ANB becomes much stronger when it is applied against that more definite concept. I think that more than half of them (in the United States at least) would answer both survey questions affirmatively. For all those people, ANB presents a formidable case for the nonexistence of their deity. On a scale of zero to 100, with fifty being the cutoff between weak and strong arguments, I would give it a score of seventy-five. (Of course, that figure would be reduced for theists who answer only one of the questions affirmatively, and it would be zero for theists in general, apart from the survey questions.) ANB may not prove conclusively that God does not exist, but it does render that result likely. It presents good support for a certain form of atheism and a serious challenge for theists which they have yet to overcome.


[1] J. L. Schellenberg, Divine Hiddenness and Human Reason (Ithaca, NY: Cornell University Press, 1993), p. 83.

[2] Schellenberg, Divine Hiddenness and Human Reason, pp. 47-57.

[3] Daniel Howard-Snyder, “The Argument from Divine Hiddenness.” Canadian Journal of Philosophy Vol. 26, No. 3 (September 1996): 433-453.

[4] J. L. Schellenberg, “Response to Howard-Snyder.” Canadian Journal of Philosophy Vol. 26, No. 3 (September 1996): 455-462, p. 460.

[5] Schellenberg, Divine Hiddenness and Human Reason, p. 211.

Copyright ©1998 Theodore M. Drange. The electronic version is copyright ©2019 by Internet Infidels, Inc. with the written permission of Theodore M. Drange. All rights reserved.

all rights reserved