Section 1 The Argument

Most of the arguments from divine hiddenness are offshoots of J.L. Schellenberg's central argument.Footnote ¹ I will be working with the following simplified reconstruction:

1. (1) If a loving God exists, then there are no nonresistant nonbelievers.

2. (2) There are nonresistant nonbelievers.

3. (3) No loving God exists.

In this context, nonresistant nonbelief is usually equated with nonculpable nonbelief and is characteristic of those who lack belief through no fault of their own. Nonculpable nonbelief is contrasted with culpable nonbelief. Culpable nonbelievers resist the evidence in some way. The existence of nonresistant nonbelievers gives rise to the problem because it is thought that a loving God would provide or make available sufficient evidence of his existence to all those who want to believe.

Philosophers have responded to this argument in a variety of ways. Some replies involve rejection of (2).Footnote ² That is, some deny that there are nonculpable nonbelievers. This is to claim that everyone who does not believe that God exists is resisting the evidence – something akin to having one's eyes closed. Other replies to the argument involve rejection of (1). This strategy generally proceeds by offering plausible reasons God might have for permitting nonresistant nonbelief.Footnote ³ For example, some appeal to a kind of benefit that accrues to individuals when God hides himself, and others appeal to an opportunity for development that the individual would not have were God to make his existence obvious to that person.

I think this work is important, but this paper will not contribute to either of these strategies. Instead, I want to advance a different route for rejecting (1). It is usually assumed that if nonresistant nonbelievers exist, it is because God has not provided or made available evidence sufficient for belief.Footnote ⁴ Schellenberg maintains that a perfectly loving God would want to be in personal relationship with his creatures and that:

The idea is that if God makes evidence of his existence available to these people and they ‘have their eyes open’, they will be in a position to believe and relate to God.Footnote ⁶ ,Footnote ⁷

In this way, the reasoning that underlies (1) can be unpacked in the following two steps:

1. (4) If a loving God exists, then he provides (or makes available) adequate evidence of his existence.

2. (5) If God provides (or makes available) adequate evidence of his existence, then there are no nonresistant nonbelievers.

I will argue that on some plausible understanding of ‘provides adequate evidence’, (5) is false. That is, I want to suggest that nonculpable nonbelief is compatible with God's providing or making available sufficient evidence for his existence. If successful, the considerations I raise undercut one route to premise (1) of Schellenberg's argument.

Two questions naturally arise when we look at the suggestion that God has not provided adequate evidence: what exactly is God obligated to do that he hasn't? and what is wrong with our evidential situation? Footnote ⁸ Discussions of the problem of hiddenness do not generally make clear how to answer these questions. In what follows, I examine several ways the argument and associated terminology could be sharpened.

Section 2 On ‘Providing’ Evidence

Consider a few statements from Schellenberg of what we would expect a loving God to do:

Without much further explanation of what this availability might look like, Schellenberg states that:

I will suggest that to determine whether adequate evidence has been provided, we need to better understand what it is to provide someone with evidence.

At first glance, this might seem like a straightforward question. In fact, given the way the notion is used in the literature, one might easily get the impression that we all have a pretty good idea what it looks like. Under scrutiny, it is not so straightforward. One difficulty is due to the numerous and complex ways evidence might be available to someone.

In The Wisdom to Doubt, Schellenberg identifies and describes various ways evidence might go unrecognized. Among these are the following categories:

Evidence is sometimes overlooked due to distractions or interruptions. It may also be overlooked because one simply has not directed attention to it yet. Since nonresistant nonbelievers are supposed to be individuals who have given sufficient attention to the evidence, this category is ruled out as a description of our evidential situation. That is, Schellenberg rules out the idea that God provides evidence but we overlook it.

It is worth noting that it is not obvious what constitutes sufficient attention. How easy does it need to be to take account of or appreciate the evidence? There is considerable variation on the circumstances one might be in, many of which do not fall neatly into categories. Consider a few examples. Suppose I write you a note and hide it in your house. There is a sense in which the note is available to you, though it is not easy for you to find. After looking for 20 minutes, you may get distracted. Or, you look in the wrong place. Does this count as overlooked evidence? Suppose further that I write you a note but in a language you do not understand. You have the means to hire a translator (and I know this), but you have not found the time to do so yet. Or perhaps the letter is hundreds of pages long and you have not set aside the time to read it yet. Does this count as evidence provided? We should avoid ruling out too quickly the possibility that there are some people to whom God seems hidden but who fall into this category.

Schellenberg delineates other categories, as follows:

Neglect of the evidence may be due to laziness or involve culpable self-deception. The explanation for why I do not have the evidence involves some failure on my part—perhaps I should have been more attentive. In this kind of situation, the evidence is available and the agent is culpable for not observing it. By contrast, when evidence is inaccessible one is in no way responsible for failing to possess it. For example, the evidence might be in a cave and no one has found it. It is possible to uncover, and we would recognize it as evidence were we to see it, but we have not discovered it.Footnote ¹³

Although this discussion takes us in the right direction, these options fail to fully describe our evidential situation. In particular, the evidential situation of the group that Schellenberg is especially interested in discussing – that of former believers – does not fall into one of these categories.

Former believers, for Schellenberg, display the most disturbing type of nonresistant nonbelief: this is the nonbelief of those who regret the loss of belief and wish to regain it but are unable to do so. Schellenberg asks us to imagine individuals who:

I suggest we need to consider another category: Defeated Evidence. The problem is that it is often assumed that if adequate evidence is available and an agent is not in willful resistance to the evidence, the agent will believe that God exists. But on a plausible understanding of ‘available evidence’ this is not the case. God might provide evidence – evidence which on its own is sufficient for belief – but one might also possess defeaters, or misleading evidence, that makes agnosticism rational despite the availability of evidence. So the dichotomy of either ‘culpable resistance’ or ‘not enough evidence’ strikes me as an incomplete description of the types of evidential situations one could be in. The epistemic situation of many people is more complicated than this.

Consider the following case where under normal circumstances we might take the evidence provided as a paradigm example of ‘adequate evidence’. Suppose I have a red cup, and I want you to know that the cup I have is red. One thing I might do is tell you I have a red cup, or I might take the red cup and set it on the table in front of us, in clear view. Placing the red cup on the table seems like a paradigm example of providing you with sufficient evidence for believing there is a red cup in front of you. But consider this case where we introduce what is often called ‘defeating’ evidence:

Here we have the presence of evidence normally sufficient to make belief rational, plus misleading evidence. It is a common judgment among contemporary epistemologists that in cases of this sort the testimony defeats the evidence you have that the cup is red. Thus, should you continue to believe that the cup is red, your belief would fail to be rational. (Many also agree that this is a case of knowledge-defeat. That is, on the assumption that you knew the cup is red prior to hearing about the trick lighting, your knowledge is defeated after hearing the testimony.Footnote ¹⁵ )

One point I want to draw attention to is that the situation in Red Lighting clearly differs from a context where there is a mere absence of evidence – that is, where I provide you with no reason at all to think that the cup is red. (I neither tell you about it nor show it to you, nor give you any indication that it is a red cup.) In Red Lighting, there is some clear sense in which I have provided evidence that the cup is red. Here are a few questions we might ask about this case:

It is difficult to answer these questions when your evidential situation involves defeaters. It is not immediately obvious that you are not in a position to believe the cup is red or that I have not provided sufficient evidence. We certainly cannot determine this just by looking around with our eyes open. But it seems equally clear that should you fail to believe that the cup is red, your lack of belief is not due to willful resistance of the evidence. The problem is not that your eyes are closed.

The difficulties that arise in Red Lighting also arise for some alleged cases of divine hiddenness. Clearly there are situations where agents have evidence that God exists but also have evidence that suggests God does not exist. Plausibly, God provides evidence—evidence normally adequate for belief—and agents fail to believe without willful resistance of the evidence. If such cases are possible, (5) is false.

I do not take these considerations to refute the argument from hiddenness but, rather to undercut one motivation for the first premise and to invite clarification of the argument. A few points to draw: first, we need a better understanding of the notion of providing (or making available) evidence. The literature is suffering as a result of a vague understanding of the locution. Second, we cannot reason from ‘nonculpable nonbelievers’ to the absence of adequate evidence. At least, this inference is not warranted without further argumentation.Footnote ¹⁶

Third, there is a significant difference between an evidential situation where one fails to have evidence which, taken on its own, makes p probable and an evidential situation that involves evidence for p combined with defeaters of that evidence. The latter makes something along the lines of a free will defense plausible. Current resolutions of the hiddenness problem place the blame either on God or the nonbeliever. Either God has not provided adequate evidence, in which case God is at fault for failing to do something he ought to do, or there is culpable resistance of the evidence, which instead attributes fault to the nonbeliever. Shifting the focus to defeated evidence introduces a third option: a third party has misled the nonbeliever. (For example, a child might believe the testimony of a well-intentioned atheist family member, or a student might trust an authority figure who convinces the student that her religious experience is not to be taken seriously.)

In this way, the free will defense is relevant to the problem of hiddenness. Just as the free will of agents contributes to an explanation for why God allows (at least some) evil in the world, the free will of agents can explain the introduction of certain defeaters and thus contribute to an explanation of divine hiddenness. Of course, the freewill of agents will not explain every case of defeated evidence, since not every instance of nonbelief is due to defeaters that are introduced by agents. But insofar as the presence of evil in the world is a central source of counter-evidence against theism, and since a great deal of the evil in the world is due to the free will of agents, the relevance of free will to the hiddenness argument is significant.

Section 3 Is God Hidden? Fixing the Argument

There are various ways to understand the claim that God is hidden. If ‘hidden’ is taken as synonymous with ‘God has not provided adequate evidence of his existence’, then the reasoning above shows that the existence of nonculpable nonbelievers is not enough to conclude that God is hidden. Given this gloss on the notion of ‘hidden’, the following is false:

One way to fix the argument is to suggest that despite the availability of evidence for belief, there is still something inadequate about our evidential situation: it is still impoverished in a way a loving God would not permit. We can ask again: what's wrong with our evidential situation?

One option is to argue that a loving God would provide sufficient evidence, where we understand ‘sufficient’ as providing adequate evidence for belief plus defeater defeaters. This initially seems to be a plausible fix. But note that this makes the argument more difficult to defend – it makes what is required of God more demanding. It might be easy to get on board with the idea that God ought to provide adequate evidence when doing so is minimal or easy. Once we introduce a requirement that God provide defeater defeaters, the demand for evidence is raised significantly.

One question that naturally arises concerns whether God is required to defeat every defeater anyone has. Is he obligated to make everyone's total evidence favor his existence (perhaps in a way that is obvious to each agent)? We might wonder whether this statement better captures what advocates of the argument expect of a loving God:

I have two worries about revising the argument in this way. One worry is that this version of the argument will strike some as less intuitive than the original argument (for the reason given above). Another worry is that there will be no limit to the amount of evidence God must provide. Notice that it is not clear what minimally I can do to get you to believe the cup is red once you believe the room has trick lighting.

Schellenberg at times uses language which suggests that he thinks a loving God would do whatever he can to ensure the belief of those seeking belief. So perhaps he would be happy with this reconstruction of the argument. But at the same time, Schellenberg maintains that he is not asking for much—nothing extreme, no compelling proofs or wondrous signs. He states that:

It is plausible that if providing sufficient evidence requires that God defeat all defeaters, this may result in a requirement for fireworks or something similarly overwhelming.Footnote ¹⁸ Yet advocates suggest that they have something less than this in mind when they look for a stronger evidential position. So there is at least some cost to advancing the argument in this way, and some evidence that advocates of the argument would resist this strengthening but may be forced into it.

Another issue concerns whether there is some amount or kind of evidence God could provide that would defeat all defeaters, make belief that he exists rational, but do so without placing each agent in a position to know that he exists. One wonders, thus, whether one can avoid construing the central premise of the problem of hiddenness as follows:

The line of thought in this paper leads the argument in this direction.Footnote ¹⁹ But I suspect that advocates of the hiddenness argument may not be happy with this construal of the argument.Footnote ²⁰ The challenge then is to set up the argument such that it does not require God to place each of us in a position to know he exists, but nevertheless requires a great deal of evidence.

Regardless of whether a loving God must provide everyone with defeater defeaters—we might agree that doing so seems like a good idea—that it would be better if everyone had defeater defeaters, and thus it still makes sense to look for an explanation for why our evidential situation is not stronger than it is. Of course, as the literature attests, the reason might not be forthcoming. In the final section, I will discuss one way in which thinking of one's evidential situation as involving defeaters may impact how we illustrate the problem of hiddenness.

Section 4 Glimpses

There are a number of analogies to divine hiddenness on offer in the literature. These analogies attempt to portray what God and nonbelievers are like. Schellenberg, the primary advocate of the hiddenness argument, compares God to a bad parent – a mother who has abandoned her child. In one picture he offers, God is compared to a parent silently watching from a distance as her child suffers and cries out, earnestly seeking the parent he believes loves him.Footnote ²¹ Those on the opposing side have portrayed culpable nonbelievers as analogous to people stubbornly clenching their eyes shut so as to resist the evidence that would otherwise make it obvious to them that God exists.

The analogies we use to discuss this issue are important. They prime us to see the argument in a certain light and serve as aids in making the argument more or less persuasive. When we focus on the bad parent analogy, we are more likely to be inclined to affirm premise (1). By contrast, if we think that nonbelievers are all clenching their ‘eyes shut’ and that getting evidence that God exists is as easy as opening one's eyes, it is much easier to deny premise (2), that there are nonresistant nonbelievers. But each of these depictions neglects an important aspect of the debate—namely, that our evidence is more complicated than this. (Note that the situation of defeated evidence is not captured by either of these images particularly well.) If in fact the evidential situation of many is mixed, our analogies ought to reflect that fact. In this way, some of the analogies in the literature fall short, and I want to suggest that we focus on a different set of analogies. By way of conclusion, I here put forward an image-type that strikes me as more adequate to the task of representing our evidential situation. (This kind of image seems to at least better depict the situation of former believers or those who have had some kind of religious experience.)

Consider glimpses. They provide the viewer with a partial vision, a glance. Glimpses are often momentary, and they are usually not available on demand or at all times.Footnote ²² One feature of this kind of evidence is that in many cases it is easy to defeat. That is, while sometimes a glimpse is sufficient, given one's background knowledge, to draw a conclusion, it can also be outweighed by counter-evidence—sometimes very easily. Glimpses seem better fit to represent our evidential situation for two reasons: first, they concede that the evidence is not so clear as to make it obvious that God exists; second, unlike the image of a silent God, they represent God as having made some effort to reveal himself—while leaving it open whether he is obligated to give us more than a glimpse.Footnote ²³ Analogies of this kind allow us to depict the evidential situation of the nonbeliever while maintaining neutrality with respect to the premises of the argument.

I do not mean to suggest that the analogy of glimpses will resolve the hiddenness debate. There is still ample room for disagreement regarding when glimpses constitute sufficient evidence and under what circumstances glimpses are defeated.Footnote ²⁴ Although representing our evidential situation as similar to glimpses will not in itself resolve the issue, the image offers a more accurate representation of the complexity of our evidential situation than many of the analogies in the literature. A potential result of refocusing the central analogy for the problem of hiddenness to that of glimpses is that theists can agree that our evidential situation could be stronger than it is, without losing sight of the important point that it is not as though we have no evidence that points to God's existence. There is evidence for God's existence, even if the evidence is not as strong as we might have expected to receive, if there is a loving God. The question remains whether such glimpses are enough.Footnote ²⁵ ,Footnote ²⁶