The basis for Schellenberg's atheistic arguments

In Divine Hiddenness and Human Reason, [hereafter DH] Schellenberg argues that the occurrence of reasonable (or inculpable) non-belief proves that God does not exist. For God, if God exists, would be perfectly loving, and a perfectly loving God, according to Schellenberg, would never permit reasonable non-belief. The argument is as follows:

Premise (1) is defended in both the introduction and chapter 1; premise (2) is defended in chapters 1 and 2; premise (3) is the conclusion to chapter 3. Schellenberg arrives at (4) as a result of (2) together with (3), and the conclusion (5) is arrived at from (4) together with (1). The focus of this paper will be on premise (2), since it is the crucial premise in this argument.

Schellenberg's premise (2), and thus the argument as a whole, is grounded in his reflections on what divine love would look like and what the goal of divine love would be. These reflections begin in chapter 1, which is entitled ‘Some epistemic implications of divine love’. Here Schellenberg explains what divine love would look like and what the goal of divine love would be. He concludes that divine love would seek, at all times, personal relationships with every human being who is capable of such a relationship and who does not culpably resist being in such a relationship (DH, 17–29). However, Schellenberg goes on to argue that the kind of reciprocal divine–human relationship in question entails human belief that God exists (i.e. a necessary condition of such a relationship is belief in the existence of God) (DH, 30). After all, how could an individual enter into a relationship with a being whose very existence is reasonably doubted? This line of thought leads Schellenberg to infer that God will actively seek to bring it about that no human being who is capable of relating to God is ever in a state of non-culpable non-belief.

The importance of premise (2) is highlighted in Schellenberg's 2006 preface to this work. He writes, ‘I do not start with reasonable nonbelief …, arguing that such exists but would not exist if there were a God. Rather, I begin with reflection on Divine love and allow the problematic phenomenon to receive its shape therefrom’ (DH, viii). Therefore, it is important to pay attention to what he says about divine love. Also within the 2006 preface, Schellenberg writes of premise (2) that it

Schellenberg says that his conclusion about divine love is the most important feature of his reasoning. This conclusion claims that any human being who is able to be in a relationship with God will be able to enjoy that relationship, unless that human culpably resists such a relationship. These thoughts form the basis for premise (2), which claims that God would never allow reasonable (or inculpable) non-belief to exist.

The claim that premise (2) is crucial is also found in chapter 4. After laying out the logical argument (which he calls A) for the first time, Schellenberg writes that ‘everything depends on (2) – if it is true, A is sound’ (DH, 84). He goes on to claim that argument A's status rests on whether premise (2) is plausibly denied. The last three chapters of DH bring out various arguments that attack the truth of premise (2). In each case Schellenberg concludes that premise (2) is not defeated since, according to him, neither offsetting nor outweighing goods (i.e. goods for the sake of which God would or might allow reasonable non-belief) can be shown to exist. His conclusion is that there are no available arguments that can defeat his premises, and therefore he claims that his argument from the reasonableness of non-belief to the nonexistence of God stands firm (DH, 208).

Schellenberg's most recent book is The Wisdom to Doubt: A Justification of Religious Skepticism [hereafter WD]. In this work, Schellenberg makes an extended argument for religious scepticism regarding what he calls ‘ultimism’. Religious claims involve ultimism by affirming ‘that there is an ultimate and salvific reality’ (WD, 3). Schellenberg argues that reason requires religious scepticism insofar as ‘neither religious belief nor religious disbelief … is justified’ (ibid.). The fourteen chapters of this book, arranged in three parts, all serve the argument that scepticism regarding ultimism is the only rational position.

When Schellenberg speaks of ‘theism’, he refers to a specific kind of ultimism that includes a personal God. Therefore, theism is included within ultimism, but ultimism does not necessarily entail theism. While he argues that we should be sceptics regarding whether ultimism or naturalism is true, Schellenberg uses part 3 of his book to develop various arguments against theism. His goal in these chapters is to prevent a refutation of religious scepticism on theistic grounds (WD, 1–2, 191–192). That is, even though he spends much time arguing against theism, these arguments are presented to support the main thesis of WD, namely, that reason requires religious scepticism regarding ultimism. In chapters 9, 10, and 13, Schellenberg presents arguments for atheism on the basis of divine hiddenness. The rest of this section will be limited to an investigation of those chapters.

In chapter 9, Schellenberg makes a few modifications to his earlier argument in Divine Hiddenness and Human Reason. Instead of referring to ‘reasonable nonbelief’, he now speaks of ‘nonresistant nonbelief’.Footnote ² Additionally, his new argument is formulated with stronger, necessary premises. The basic thrust of the argument, however, remains unchanged. Schellenberg's new argument is presented in this way:

Premise (3) is a result of (1) together with (2). And when (3) is taken in conjunction with (4), Schellenberg's conclusion (5) is the result. The remaining material in chapter 9 considers and replies to possible objections to this argument.

In chapter 10, Schellenberg attempts to develop four additional new arguments against the existence of God based on divine hiddenness. He does so by examining four different kinds of non-resistant non-belief. The four types of non-resistant non-belief that are examined are former believers, lifelong seekers, converts to non-theistic religions, and isolated non-theists. For each of these types of non-resistant non-belief, Schellenberg attempts to develop a new argument for atheism.

In chapter 13, Schellenberg comes back to the divine-hiddenness arguments from chapters 9 and 10 in order to combine them, thereby creating what he calls the ‘consolidated hiddenness argument’ (WD, 291). This new argument attempts to demonstrate that the argument of chapter 9 is strengthened when the argument of chapter 10 is added to it, and also that the argument in chapter 10 is strengthened when the argument of chapter 9 is added to it.

As was the case in his previous work, these arguments in WD are all grounded in Schellenberg's reflections on divine love. In the beginning of chapter 9, he reflects on the nature that a loving God would have. He then has us think about what it would be like if God created personal finite creatures. He concludes the following:

As he reflects further on these issues, Schellenberg concludes, ‘A perfectly loving God – if those words mean anything – would ensure that some form of conscious and meaningful relationship with herself was always possible for capable and nonresistant finite persons’ (WD, 202). In other words, just as in his previous book he argues that the very nature and character of a loving God leads to the conclusion that God would never allow reasonable (i.e. inculpable) non-belief, here also he uses those same considerations to argue that God would not allow non-resistant non-belief.

In addition, just as in his previous work he bases a crucial premise for his argument on his reflections on divine love, so also here Schellenberg crafts the first (and crucial) premise of his new argument based on his reflections on divine love as found in (A), (B), and (C) above (WD, 204). The new argument, just like the previous one, concludes that, based on the presence of non-resistant non-believers, God does not exist. Also similar is Schellenberg's statement that this conclusion results from his ‘careful reasoning and consideration of the heart and soul of love’ (WD, 206).

In both of his arguments, then, Schellenberg's atheistic conclusion is grounded ultimately in his reflections about what a loving God would do, namely, that a loving God would prevent both reasonable (i.e. inculpable) and non-resistant non-belief. The following sections will call this basis into question. This paper's argument against Schellenberg is that, according to his own sceptical standards, human beings are not in a position to assert confidently that a loving God would prevent reasonable non-belief and non-resistant non-belief at all times for all people. The main source of support for this argument is found in the subject mode of religious scepticism advanced in Part 1 of Schellenberg's The Wisdom to Doubt.

The subject mode of religious scepticism

In Part 1 of WD, Schellenberg seeks ‘to motivate a transformation in our basic view of the human intellectual situation’ (WD, 12). According to his own estimation, ‘radical modesty and even pessimism … about our present capacity as individuals or in concert to establish or disestablish claims concerning the existence or nature of an ultimate and salvific reality, is called for’ (ibid.). In Part 1 of this book, Schellenberg describes seven modes and argues that each mode grants support to the sceptical position concerning all things religious. This section will look at his claims regarding what he calls the ‘subject mode’ in order to show that, once he begins arguing against theism in Part 3, Schellenberg does not exhibit ‘radical modesty’, much less ‘pessimism’, concerning his own ability to establish the nature of theistic reality, if such reality exists.

The focus of the subject mode of religious scepticism is to point out that human finitude leads to severe limitations regarding intellectual enquiry and the justification of religious belief. Schellenberg says that ‘there are, at any point in any human intellectual inquiry … various ways in which positive and negative evidence and also vindicators and defeaters can go unrecognized by us’ (WD, 17). He then proceeds to discuss, in detail, five specific ways in which evidence, vindicators, and defeaters can go unrecognized. (These five classes of unrecognized evidence are: overlooked evidence, neglected evidence, inaccessible evidence, undiscovered evidence, and undiscoverable evidence.)

Schellenberg presents a strategy for assessing if a particular belief is defeated by unrecognized evidence. Specifically, one must ask not only if one's evidence provides positive support for one's belief, but one must also ask if there is likely to be unrecognized evidence for or against this belief. According to Schellenberg, ‘If we can show as much as that there is good reason to expect much unrecognized evidence whose import could as easily turn out to be strongly negative as positive’, then the evidence one has in support of one's belief is thereby blocked (WD, 31). In order to make his case for religious scepticism more convincing, Schellenberg presents his readers with what he takes to be a sufficient condition for determining when individuals ought to be in doubt about the representativeness of their positive evidence in relation to potential unrecognized negative evidence.

His proposed sufficient condition hinges on the number of significant alternatives to what one believes. Hence, if one has a belief that p, then ‘the more significant alternatives to p (or evidence pertaining to such alternatives) one has not considered or thought of, the more unconsidered objections to p – and the more strongly negative evidence – may be lurking in the unrecognized evidence’ (WD, 34). In other words, when there are a great number of significant-yet-unconsidered alternatives to p, then it is likely that there is much unrecognized evidence that either directly opposes p or supports objections to p.

In order to specify when such unrecognized evidence is just as likely to be negative as positive, Schellenberg lists four general properties that increase the number of unconsidered significant alternatives. These four properties are the properties of being precise, detailed, profound, and attractive. After defining each property and giving examples, Schellenberg arrives at the following conclusion (note that E is being used for an individual's examined body of evidence):

According to Schellenberg, the more a proposition p shares in those four properties, the more significant alternatives to p exist; and the more significant alternatives to p exist, the more likely it is that unrecognized evidence opposes or contradicts p. Hence, the more p shares in those four properties, the more one should become sceptical of the representativeness of the evidence that one has in support of p. In the end, Schellenberg uses these four properties as the sufficient condition for determining when one should be sceptical about one's ability to affirm p based on one's evidence.

Undermining Schellenberg's atheistic arguments

This paper argues that Schellenberg's main arguments for atheism based on reasonable and non-resistant non-belief are not justified if his subject mode of religious scepticism is valid. As indicated above, Schellenberg's arguments in both DH and WD rest upon previous reflections regarding divine love and, more specifically, what an omnipotent and perfectly loving God would and would not do or allow. Schellenberg, then, seems to adopt the following proposition:Footnote ³

1. (1) Although I am a finite and fallible human being, I have employed my reasoning skills, prior to and independent of any divine revelation, to arrive at some conclusions regarding what an omnipotent and perfectly loving God necessarily would do (if such a God exists).

One such necessary truth, according to Schellenberg, is the following:

1. (2) An omnipotent and perfectly loving God would prevent non-resistant and reasonable (inculpable) non-belief at all times for all finite rational creatures who possess the capacity to enter into explicit, conscious relationship with God because such a relationship would be the greatest good for human beings and God, being perfectly loving, necessarily promotes the greatest good for human beings.

Further, the conjunction of (1) and (2) yields the following conclusion, showing that Schellenberg would assert:

1. (3) Although I am a finite and fallible human being, I have employed my reasoning skills, prior to and independent of any divine revelation, to arrive successfully at the conclusion that an omnipotent and perfectly loving God necessarily would prevent non-resistant and reasonable (inculpable) non-belief at all times for all finite rational creatures who possess the capacity to enter into explicit, conscious relationship with God because such a relationship would be the greatest good for human beings and God, being perfectly loving, necessarily promotes the greatest good for human beings.

It will be shown that asserting the truth of (3) stands in contrast to Part 1 of WD. Regardless of what he urges his readers to accept, Schellenberg does not adopt the same ‘radical modesty and even pessimism’ about his own ‘present capacity … to establish or disestablish claims concerning the existence or nature of an ultimate and salvific reality’ (WD, 12).

Although Schellenberg believes he has a body of evidence that he takes to provide positive support for (3), has he taken into account the potential unrecognized evidence that bears on (3)? More specifically, can one, in Schellenberg's words, ‘show as much as that there is good reason to expect much unrecognized evidence whose import could as easily turn out to be strongly negative as positive’? If this can be done, then according to Schellenberg himself, the evidence he has in support of (3) is blocked. In order to argue that there is good reason to expect much unrecognized evidence of this kind, I will focus on (3) and see how it fares when subjected to the sufficient condition elaborated in relation to Schellenberg's subject mode of religious scepticism.

In order to be justified in my scepticism regarding Schellenberg's ability to affirm (3) consistently, I will apply his own sufficient condition for determining when one ought to be in doubt about the representativeness of one's positive evidence in relation to potential unrecognized negative evidence. As stated above, this sufficient condition is determined in relation to the significant alternatives to that which is believed. By relating (3) to the four general properties Schellenberg identifies, I will show that there may be a great number of significant-yet-unconsidered alternatives to (3), and thus it may be that there is much unrecognized evidence that either directly opposes (3) or supports objections to (3).

First of all, (3) is very precise. For example, instead of saying, vaguely, that human beings in general might be able to arrive at some conclusions regarding what a loving God necessarily would do, Schellenberg states that he personally has arrived at these conclusions. Additionally, instead of saying, vaguely, that God would promote the greatest good for human beings, (3) specifies a precise condition that blocks a human being's greatest good (i.e. non-resistant or inculpable non-belief).

Second, (3) is quite detailed. That is to say, there are detailed claims made by (3), thereby creating a more complex proposition. An example here is that (3) not only speaks of a human being's greatest good but also defines it as a conscious, explicit relationship between humans and God. Granted, (3) does not try to describe everything a loving God would do, nor does it claim to specify all conditions that would block one's greatest good. There are many ways, then, in which (3) could become more detailed. Still, there are significant detailed claims made in (3).

The third property is the property of being profound. It takes very little to be convinced that (3) makes very profound claims. Schellenberg describes this property as ‘deep or fundamental, comprehensive in scope, and pregnant with explanatory possibilities’ (WD, 35). Clearly, (3) is profound in that it is making fundamental claims about what the Creator of all that exists, if such a Creator exists, would be like and would do. Moreover, it is incredibly comprehensive in scope insofar as it identifies that which is the greatest good for all human beings.

Finally, (3) is also attractive on many different levels. For example, the picture of God that it paints is an attractive one. To take a perspective that says that God, if God exists, would seek explicit, conscious relationships with human beings is very attractive. More attractive than this, perhaps, is the picture that (3) paints of the power of human reason and intellect. According to (3), human reason can be used, prior to and independently of any divine revelation, to arrive at necessary truths regarding divine love, including truths regarding relationships that God would promote and certain kinds of non-belief that God would prevent. This view of the human intellect is very attractive. It is very flattering to claim or believe that human reasoning, primarily on the basis of human analogies, can be employed to figure out what God would do and what God would prevent. On this view, human beings do not need to wait to hear from God to find out what God is like and what God seeks, if God exists. Rather, we can figure it out for ourselves.

When analysed in terms of Schellenberg's four general properties, it turns out that (3) shares in all four properties by being precise, detailed, profound, and attractive. Moreover, (3) exemplifies these overlapping properties to a great extent. That is, (3) is very precise, very profound, and very attractive. In this way it is made clear that there may be numerous significant alternatives to (3), and this makes it more likely that there may be much unrecognized evidence that opposes or contradicts (3).

Therefore, based on Schellenberg's own conclusions regarding the subject mode of religious scepticism, there is much reason ‘to suppose that the unrecognized evidence relevant to [(3)] may as well be negative as positive in its import’. Hence there is support ‘to be in doubt about the representativeness’ of Schellenberg's evidence that provides positive support for (3). If Schellenberg is correct in his use of the four properties as the sufficient condition for determining when one should be sceptical about one's ability to affirm a given proposition, then we should be sceptical about Schellenberg's ability to affirm (3) based on his evidence.

So far I have only applied Schellenberg's scepticism to the arguments in DH and the ninth chapter of WD. As noted in the first section, the tenth chapter of WD presents four additional arguments for atheism on the basis of non-resistant non-belief. To construct these additional arguments, Schellenberg identifies four specific kinds of non-resistant non-belief that a perfectly loving God would prevent: former believers, lifelong seekers, converts to non-theistic religions, and isolated non-theists. For each of these types of non-resistant non-belief, he develops what he considers a new argument for atheism. In other words, in each case Schellenberg identifies more precisely what he means by, e.g. former believers, and then argues that a loving God would prevent non-resistant non-belief in individuals who are former believers. Or again, phrased differently, Schellenberg argues that, due to the presence of former believers (or lifelong seekers, or converts to non-theistic religions, or isolated non-theists), God does not exist.

A response to these arguments can be rather brief. The same strategy used above should be applied here. Indeed, the strategy of applying the four general properties yields even more conclusive reasons for thinking Schellenberg unjustified in claiming that God would prevent these particular types of non-resistant non-belief. For his reasoning here remains based upon his foundational reflections regarding what a loving God would or would not do or allow.

Given the content of chapter 10 of WD, even more specific types of non-belief can be added into (3), thus yielding:

1. (4) Although I am a finite and fallible human being, I have employed my reasoning skills, prior to and independent of any divine revelation, to arrive successfully at the conclusion that an omnipotent and perfectly loving God would prevent reasonable non-belief and the types of non-resistant non-belief instantiated in former believers, lifelong seekers, converts to non-theistic religions, and isolated non-theists, at all times for all finite rational creatures who possess the capacity to enter into explicit, conscious relationship with God because such a relationship would be the greatest good for human beings and God, being perfectly loving, necessarily promotes the greatest good for human beings.

As long as Schellenberg remains committed to the subject mode of religious scepticism, then the content of (4) provides even more reason to be in doubt about the representativeness of one's positive evidence in relation to potential unrecognized negative evidence. For it is clear that (4) is only more precise and more detailed than was (3), thereby yielding even more significant alternatives to (4). Instead of adding more weight to his atheistic conclusion, Schellenberg has only succeeded in making us more sceptical about his ability to draw his atheistic conclusion.

Finally, in the thirteenth chapter of WD, Schellenberg combines his arguments from chapters 9 and 10 to create what he calls the ‘consolidated hiddenness argument’. Given the problems with each of the original arguments in question, combining them will do nothing to lessen the compelling doubts about Schellenberg's conclusions.

Therefore, I remain sceptical of Schellenberg's ability to remain consistent in affirming both the subject mode of religious scepticism and his divine-hiddenness arguments. That is, if Schellenberg continues to uphold his sceptical position, then his reflections about divine love will be called into question. In particular, his sceptical stance, if true, leads to the conclusion that he is not in a position to know a priori what an omnipotent and perfectly loving God necessarily would or would not allow in terms of non-resistant non-belief or reasonable non-belief. To remain consistent in his scepticism, then, Schellenberg is not able to develop the crucial premises in both of his divine-hiddenness arguments for atheism.Footnote ⁴ Hence, Schellenberg cannot claim to have conclusive proof for God's nonexistence based on non-resistant or reasonable non-belief while he also endorses the subject mode of scepticism.Footnote ⁵

Getting around Schellenberg's defences, part 1

In both of his books and in his numerous essays and articles, Schellenberg spends much time examining possible objections to his position and then offers (sometimes multiple) counter-arguments for those objections.Footnote ⁶ The argument just raised, however, is new and has not been countered by Schellenberg. In this section and the one that follows, it will be demonstrated how this paper's objection to Schellenberg's arguments from divine hiddenness is unique and is not defeated by any defence Schellenberg has used thus far.

First of all, Schellenberg seems to anticipate this paper's argument in the opening pages of chapter 14 of WD. He acknowledges that ‘a tempting response’ would be to argue that, given his capacity scepticism, ‘any claim as to the provability of atheism I might make would be self-defeating’ (WD, 297). He goes on to clarify his position by saying ‘the availability of a proof of the sort in question would not entail that we had discovered a “basic truth about the existence and nature of the Ultimate”’ (WD, 298). In other words, his capacity scepticism is properly applied only to ‘basic truths’ concerning ultimism. Hence, according to Schellenberg, an argument for atheism does not claim to discover any basic truth regarding ‘the existence and nature of the Ultimate’. Discerning the absence of a personal God, according to Schellenberg, is not the same as discerning the absence (or nature) of a different kind of ultimate salvific reality.

Schellenberg is not successful in countering the argument given above, although he has anticipated a similar argument. I can hold fast to my argument while also agreeing with Schellenberg that the availability of a proof for atheism does not necessarily entail a discovery of a basic truth about the Ultimate. That is, a defence of Schellenberg's capacity scepticism does not ipso facto rule out the possibility of atheistic arguments. However, I am not arguing that ‘any claim as to the provability of atheism’ is at odds with Schellenberg's capacity scepticism. It is not just ‘any claim’ regarding atheism that I am concerned with. Rather, I have argued that some particular arguments that Schellenberg makes (the arguments for atheism based on divine hiddenness) are defeated by his religious scepticism.

In developing his anticipatory counter-argument further, Schellenberg writes, ‘My earlier arguments never claimed that in just any instance of reasoning, problems resulting from (say) the possibility of unrecognized counterarguments should be expected to intrude’ (WD, 298). Instead, there are many beliefs that are justified based on sufficient grounds. Therefore, he believes that his arguments for atheism are not defeated by his capacity scepticism.

Schellenberg is right to insist that not all instances of reasoning will be irreparably plagued by unrecognized evidence and hence by unrecognized counter-arguments. I also agree with this and, with Schellenberg, affirm that many beliefs are and can be justified on sufficient grounds. There might be many atheistic arguments that are not hindered by unrecognized evidence. However, the argument of this paper does not claim that all instances of reasoning are defeated by unrecognized evidence. Rather, my argument is more precise. I have argued that unrecognized evidence does intrude in the specific arguments Schellenberg makes based on divine hiddenness in DH and in WD. Indeed, such is the conclusion when one adopts Schellenberg's own sufficient condition for determining when one ought to be in doubt about the representativeness of one's positive evidence in relation to potential unrecognized negative evidence.

While Schellenberg anticipated a generalized argument of the kind made above, he has not successfully defended against this paper's more specific argument. For I have taken Schellenberg's claims and inferences regarding divine love and demonstrated how, based on the four general properties that comprise his own sufficient condition, these specific claims (which support crucial premises in his arguments) are not warranted if we accept the scepticism in WD. That is, it is likely that there may be much unrecognized evidence that either directly opposes or supports objections to these particular atheistic arguments from divine hiddenness.

Next, someone could argue that the way in which Schellenberg develops the four general properties of his sufficient condition already contains a defence against this paper's argument. Schellenberg states, ‘I want now to develop certain points previously adumbrated into a sufficient condition for the justification of doubt with respect to the representativeness of evidence in certain nonclear cases’ (WD, 34). In other words, the four general properties establish a sufficient condition regarding when one ought to doubt the representativeness of one's evidence only in non-clear cases. Therefore, one could argue that Schellenberg's thoughts and reflections on divine love, the same thoughts that support the premises of his divine-hiddenness arguments, are in fact clear cases. If this is true, then the sufficient condition does not apply and Schellenberg can remain confident in his atheistic arguments.

This defence will not hold up upon examination of Schellenberg's remarks on what constitutes clear and non-clear cases. Schellenberg admits that there are clear cases. These are situations in which one is ‘faced with evidence for p so strong that it cannot be blocked, that there is no room for doubt as to its truth – so strong that nothing can prevent us from inferring … that the evidence in question is representative’ (WD, 31–32). As examples of clear cases, Schellenberg refers to ‘simple beliefs of reason such as the belief that if A is taller than B, then B is shorter than A or that 2+2=4’ (WD, 32). These are truths that must be true in any and every possible world. He also includes ‘certain general moral beliefs’ among clear cases, such as ‘the belief that universal justice is good’ (ibid.). Additionally, Schellenberg maintains that clear cases can include ‘the belief that my mother, for whose general virtue and particular devotion there are innumerable interwoven strands of evidence, is not out to fool anyone when she says she loves me’ (ibid.). Also included are some beliefs based on empirical research, such as ‘the belief that the earth is not flat’ (ibid.). In all of these cases, one may be secured against doubt regarding the representativeness of one's evidence based on ‘the evident truth of the belief itself’ (ibid.).

Before jumping to a comparison of these descriptions of clear cases and Schellenberg's support for his atheistic arguments, it is important to hear Schellenberg's further comments on clear and unclear cases. He emphasizes that his reader should ‘notice how narrow a class of cases has been carved out here’ (ibid.). Indeed, he says that most situations are not clear and that cases of intellectual enquiry, almost by definition, involve non-clear cases. He writes,

Anything worthy of the name enquiry, then, will be counted as a non-clear case.Footnote ⁷ Or, using a phrase cited above, if there is any ‘room for doubt’ at all, then it is not a clear case. The fact that there are clear cases is a fact that ‘has seriously limited applicability’ (ibid.), which is to say that such cases are exceedingly rare.

When Schellenberg limits the applicability of his sufficient condition to situations that are non-clear, he is not ruling out very many cases. However, it is possible that Schellenberg thinks that his propositions regarding divine love are self-evident and hence constitute a clear case. If this were so, these propositions and premises would be exempt from the sufficient condition for determining when to be in doubt regarding the representativeness of one's body of evidence. In such a case, the argument of this paper would have no force. The question at hand, then, is whether Schellenberg's claims and inferences regarding divine love count as clear cases.

The argument of this paper claims that as long as Schellenberg holds on to his subject mode of religious scepticism, he is not able to support the crucial premise in each of his main divine-hiddenness arguments. In DH, the crucial premise in his overall argument (called A) is premise (2), which states that ‘If a perfectly loving God exists, reasonable nonbelief does not occur’. Does this premise represent a clear case? Is it a clear or self-evident truth of reason? Is there any room for doubt here? I will argue that this is a non-clear case. In order to support its status as a non-clear case, I will rehearse the general moves Schellenberg has to make in order to establish premise (2).

At the beginning of chapter 1 of DH, Schellenberg states that ‘if God is loving, he desires our well-being’ (DH, 17). Schellenberg then argues that the well-being of human beings would be served best through a personal relationship with God, if God exists (DH, 18–22). He makes additional arguments in order to establish that God would seek such personal relationships for all human beings and at all times (DH, 23–26). After making some qualifications regarding the role of freedom in personal relationships, Schellenberg formulates what he labels P1 (DH, 27–28).Footnote ⁸ After additional qualifications and adding another premise, Schellenberg arrives at P2 (DH, 28–31).Footnote ⁹ Eight pages later, Schellenberg settles on P2′ (DH, 38).Footnote ¹⁰ The rest of this chapter is spent clarifying P2′ and defending it against two objections (DH, 39–43). Without P2′, premise (2) of A is not established. Moreover, Schellenberg then uses the entirety of chapter 2 to argue that P2′, and hence premise (2), represents a state of affairs that is logically possible (i.e. that it is possible for God to actualize a world in which reasonable non-belief does not occur) (DH, 44–57).

After rehearsing these general moves, should one view premise (2) as representing a clear case? No, one should not. Using Schellenberg's own criteria in these matters, a premise that requires forty-one pages of clarification, inferences, arguments, and support is not a clear case. And don't forget that the entirety of Part 2 (more than half of the book) is a series of responses to potential objections to premise (2). Regardless of the truth of the matter, the need for enquiry and the existence of so many kinds of objections indicate that the case is far from clear. It is true that some of Schellenberg's beginning propositions may count as clear cases. For instance, he begins by stating that if God exists, then God is loving. And ‘if God is loving, he desires our well-being’ (DH, 17). These statements should be regarded as clear cases. What is not so clear, and what requires so much argument, is the determination of exactly what a human being's well-being includes and exactly how and when God would act in order to enhance the well-being of humans.

Similar arguments are needed to support the crucial premise of Schellenberg's divine-hiddenness argument found in WD. After beginning with ‘the traditional idea of God’, Schellenberg asks his readers to imagine God creating a concrete reality. He says that ‘it is inconceivable that a perfect personal God would exercise creativity purely for its own sake and with wild abandon, uncaring about whether the results were good or bad’ (WD, 199). Supposing further that God creates ‘finite personal creatures’, Schellenberg arrives at the following (which is also seen above):

Arguing further from (C), Schellenberg concludes that ‘God would make conscious awareness of the Divine available to every finite personal creature’ (WD, 200). He then spends four pages drawing out implications from (A) and (B) (WD, 201–204). The result of these four pages is the critical premise in his new atheistic argument from divine hiddenness. Recall that in WD, Schellenberg's premise (1) is the crucial premise. This premise states: ‘Necessarily, if God exists, anyone who is (i) not resisting God and (ii) capable of meaningful conscious relationship with God is also (iii) in a position to participate in such relationship (able to do so just by trying)’ (WD, 204). When he introduces (1) for the first time, he writes: ‘I suggest that on the basis of previous argumentation involving (A), (B), and (C), we have more than sufficient reason to affirm’ premise (1) (ibid.). In other words, Schellenberg presents premise (1) as the result of his previous inferences and arguments resulting from (A), (B), and (C).

Given all that is involved in arriving at and supporting premise (1), does it represent a clear case? No, it does not. Again, using Schellenberg's own criteria, this premise requires too much in the way of clarification, argumentation, and support to be a clear case. He even admits that there are ‘many factors that may cause us to underestimate the force of love-based arguments like the one I am developing’ (WD, 203–204). Schellenberg therefore recognizes that he is offering an argument for premise (1) just as he recognizes that there are ‘many factors’ that can influence people so as not to see the force of this argument. Recognition of these facts should lead to the verdict that premise (1) represents a non-clear case.Footnote ¹¹

To summarize, Schellenberg cannot successfully defend his divine-hiddenness arguments by claiming that his crucial premises represent clear cases. All of the work that went into constructing and defending his critical premises demonstrate that they are not ‘simple beliefs of reason’ or ‘clear truths of reason’. None of the self-evident kinds of truths he uses as examples of clear cases are analogous to his support of these premises. His premises are the result of careful enquiry and investigation into the nature of divine love. However, by his own standards for determining clear cases, anything worthy of the name enquiry ought to be counted as a non-clear case. Regardless of the degree to which one is convinced by Schellenberg's support for his premises, it is not the case that one is faced ‘with evidence for p so strong that it cannot be blocked, that there is no room for doubt as to its truth’ (WD, 31).

In carving out such a narrow class of clear cases, Schellenberg ensures that the premises for his divine-hiddenness arguments are regarded as non-clear cases. These premises are therefore not exempt from the sufficient condition elaborated under the subject mode of religious scepticism since the sufficient condition applies to all non-clear cases. As a result, this paper's argument is justified in subjecting Schellenberg's crucial premises to the four properties that comprise the sufficient condition for determining when to be in doubt regarding the representativeness of one's body of evidence. Therefore, my argument stands firm: as long as Schellenberg holds on to his subject mode of religious scepticism, he is not able to support the crucial premise in each of his main divine-hiddenness arguments.

Getting around Schellenberg's defences, part 2

Schellenberg has also supported and defended his divine-hiddenness arguments in many other ways. This section examines the most prominent of these. It will become evident that these defences do not speak against or hinder this paper's argument in the least.

In the argument formulated in DH, the crucial second premise states, ‘If a perfectly loving God exists, reasonable nonbelief does not occur’. Schellenberg believes that his prior reasoning regarding the nature of divine love has established at least prima facie evidence that (2) is true. He then goes on to establish what must happen in order to plausibly deny (2). Plausibly denying (2) relies on the conditions of (2)'s falsity. According to Schellenberg,

After establishing these two conditions of (2)'s falsity, the conclusion is that, in order to plausibly deny (2), one must show that both conditions for (2)'s falsity are plausible. As he fleshes out what this means, he concludes that (2) can be denied (and hence his argument can be rebutted) ‘only if (i) it is shown conclusively that an offsetting good necessitating the permission of reasonable nonbelief exists, and/or (ii) it is shown to be plausible that an outweighing good requiring the permission of reasonable nonbelief exists’ (DH, 87–88). Therefore, according to Schellenberg, one must deny (2) only by arguing that (2) is (plausibly) false by appealing to either an offsetting or outweighing good in virtue of which reasonable non-belief will be instantiated.

All the chapters in Part 2 of DH present possible offsetting or outweighing goods. For example, Schellenberg examines the position that says that God would allow reasonable non-belief for the sake of instantiating the good of moral freedom, or for the good of religious diversity, and so on. In each case, he judges that the goods in question do not meet the criteria for plausibly denying (2), and so he concludes that his argument is sound.

Similarly, in WD the crucial premise is defended in the same way. After laying out his newly formulated argument from divine hiddenness, Schellenberg again looks at numerous possible objections to his argument. However he believes that serious objections to his argument must all have a similar form. Regarding this form, he says:

Similarly to his previous work, Schellenberg is focusing on objections that attempt to show the reasons why God would allow non-resistant non-belief (or objections that attempt to show that it is plausible that God has such reasons but that they are unknown to us). Such reasons involve ‘some great good’ that can be realized only in the event that non-resistant non-belief also exists.

Schellenberg spends much time in WD presenting possible objections along these lines. But again, in each case he believes he demonstrates how the objections to his argument fail. He has yet to find some ‘great good’ or any offsetting or outweighing good that would warrant the existence of non-resistant non-belief. Therefore, he concludes that his argument is undefeated.

The argument presented against Schellenberg above, however, is not one of these kinds of arguments. It does not seek to present any kind of actually existing good, either outweighing or offsetting. It does not seek to plausibly deny the truth of (2) in a direct manner. It does not seek to argue directly for the falsity of his crucial premises. Therefore, his numerous responses to the numerous objections that he presents in his works do not contain a response to my argument.

Instead of attempting to deny his crucial premises directly, I have argued that, if one accepts Schellenberg's subject mode of religious scepticism, one should doubt Schellenberg's ability to develop his crucial premises in the first place. That is, if one is satisfied with this kind of scepticism, then the resulting view of what our limited human reason can accomplish prohibits Schellenberg from concluding, a priori, what an omnipotent and perfectly loving God would do in regard to the existence of non-resistant non-belief or reasonable non-belief. The crucial premises of his divine-hiddenness arguments, and hence those arguments as a whole, cannot succeed if he is correct regarding scepticism. Therefore, this paper avoids playing by his rules and thereby avoids the back-and-forth game of proving and disproving what would count as (possibly or actually existing) offsetting and outweighing goods.

Schellenberg believes that his personal reflections regarding the nature of divine love establish a presumption in favour of his crucial premises (DH, 91). Hence, he continually operates under the assumption that a successful objection against his arguments must provide positive evidence regarding the (at least plausible) falsity of his premise(s). This paper's objection to his arguments, however, points out that his own scepticism, if true, undercuts the supposed presumption in favour of his crucial premises. His arguments, therefore, never get off the ground if his sceptical case in WD is sound; and in such a scenario one is not under any obligation to provide positive evidence in the form of actually existing offsetting or outweighing goods. As a result, this paper's argument is untouched by the many defences he has given for his arguments.