John Schellenberg's The Wisdom to Doubt is an impressive and important book. It is clearly written, and clearly argued. It is a valuable contribution to epistemology as well as to philosophy of religion (although Schellenberg is careful to limit the range of his epistemological points). It will have a strong effect on current discussions in philosophy of religion. In this review I can offer only the barest of bare-bones account of his rich and remarkable work; let me begin by simply recommending it highly.

‘Reason’, says Schellenberg (1), ‘requires us to be religious skeptics’. He begins by drawing our attention to the notion of ultimism, reminding us in passing that many workers in the philosophy of religion have an unacceptably narrow focus. They concentrate almost exclusively on the religions of the book, traditional theistic religions, while conscientiously avoiding any consideration of wider possibilities. Nonetheless, ‘there is unceasing disagreement and controversy over the lineaments of the Divine’ (57), and it is just such a wider perspective that any discussion in philosophy of religion must have if it is to attempt anything like a complete view of the field.

Religion, for Schellenberg, involves the claim (3) that ‘there is a reality metaphysically and axiologically ultimate (representing the deepest fact about the nature of things and also unsurpassably great), in relation to which an ultimate good can be attained’, that is, ‘that there is an ultimate salvific reality’. This very general religious view Schellenberg terms ultimism. Ultimism cannot to be equated with traditional theism, for it is consistent with non-theistic traditions. Indeed, Schellenberg's discussion of traditional theism in the third part of the book suggests that his sceptical approach is, to put it no more strongly, as likely to support atheism as it is to support current theistic positions.

Following the widening of our horizons, Schellenberg draws our attention to two aspects of scepticism. Scepticism may be categorical: it may be a scepticism about any view concerning ultimism being rationally grounded. More narrowly, it may be a scepticism based on dubiety about human abilities: a capacity scepticism which allows us to wonder whether humans can, now or ever, attain ‘basic truths about religion’ (5).

Evidence can be recognized or unrecognized. If unrecognized it can be, nonetheless, evidence we are capable of recognizing, or evidence we are, either because it is undiscovered or undiscoverable, incapable of recognizing. Unrecognized evidence within our capability can be accessible or inaccessible, and if accessible but unrecognized it may be because it is overlooked or neglected (17ff.).

Combining an extended view of religious belief with the paucity not only of our present knowledge, but even of our present ways of seeking, let alone acquiring, knowledge, we can come to see that religious scepticism is the only rationally satisfactory philosophical stance. From the believer's standpoint, we do not even have the ability to grasp what it is that might be the object of belief; from the non-believer's standpoint, we must see that it is not open to us to claim, without any doubt, that there is nothing that is religiously ultimate.

At best, the evidence we currently have, or perhaps could ever have, must represent a minute portion of the possibly relevant evidence for religious claims, and the question arises, are there ways of deciding that the evidence we do or could have is representative of the unexamined and perhaps unexaminable evidence? What would be ‘a sufficient condition for the justification of doubt with respect to the representativeness of evidence’ (34)? (We notice that such doubt will automatically lead to doubt about the propositions supposedly upheld by the evidence in question.)

These are, of course, general points, but they strike more forcefully in the religious case than in the general case. To see why, consider certain important points about the properties of propositions (35f): propositional content may be: precise: (‘definite and sharply exact’); detailed: ‘thorough and full of particulars, and thus complex or multifaceted (as opposed to simple)’; profound: ‘deep or fundamental, comprehensive in scope, and pregnant with explanatory possibilities’; attractive: ‘likely to be approved by human beings, and likely to draw forth a positive emotional response from them (as opposed to being dry, cold, uninteresting)’.

Thinking about these properties alerts us to an important result:

This is particularly true when we recognize the fact that ‘Our past reflects serious intellectual failure and deep immaturity where religious matters are concerned – a failure and immaturity that affects us all’ (88).

So, Schellenberg asks: ‘Can we – to re-emphasize what is perhaps the central question – ever fully map the alternatives there must be to any elaborated ultimism?’ (44). And, he suggests, the answer is clearly ‘No’.

So, ‘already we can see strong reasons for denying that either religious belief or disbelief is justified, and so for categorical skepticism’ (63). And, in passing, Schellenberg notices a point too often missed in discussions of the deity. ‘It is’, he remarks, ‘at least strongly plausible to suppose that whatever is true here [about ‘the Ultimate’] is necessarily true and that whatever is false is necessarily false' (67). This means that in such discussions modal operators are out of place, just as they are, for example, in mathematics. For if any given proposition about the Ultimate, u, say, must be either necessarily true or necessarily false: □u v □~u, then we have (in normal modal systems at any rate), ◊u → □ u. Which is to say: modalities collapse in such a case. What is possible is necessary, and therefore, is the case, and what is actual is necessary, and, of course, conversely.

Before turning to the application of the points made in the first part of the book to naturalism and traditional theism, Schellenberg considers the bearing of pragmatic considerations, particularly if these involve, as many (at least) of them do, simple doxastic surrender: a willingness to acquire or attempt to acquire a belief either in the absence of evidence for it, or, more strongly, in the teeth of evidence against it. Even if there are cases where ignorance is bliss, can it be right to attempt to turn ignorance into belief?

Schellenberg's answer (and here, as throughout this review, it is not possible to discuss his stance in detail) is that, even if there are ‘non-truth-oriented benefits’ associated with belief, ‘religious skepticism has access to [benefits] at least as great as those alleged … to be associated with belief’ (123). And, he argues, that while ‘there is an undeniable attractiveness in the disbeliever's perspective, charitably and sympathetically construed’ (125), the pragmatic benefits of the disbeliever's stance are also outweighed by the benefits accruing from scepticism in religious matters.

Now, how do these considerations operate when we consider the more common types of belief and disbelief? To take disbelief in its most common form, naturalism, we should notice at once that ‘naturalists tend to be overly influenced by the least mature forms of religion, which in their doctrinal manifestations can be spiritually unimaginative and psychologically naïve’ (148).

Parts 1 and 2 provide a sufficient justification for religious scepticism, but ‘the present climate in philosophy of religion is one in which such skeptical voices may still be hard to hear’ (191), and, with this in mind, Schellenberg considers in Part 3 some typical positions in contemporary Western philosophy of religion. In particular, he looks, after providing ‘an answer to naturalism’ (chapter 7), at arguments from religious experience, at a variety of subtle and interesting considerations based on the hiddenness of God, at various versions of the argument from evil, and at the free-will defence, concluding that ‘free will, if it exists, is a problem for the theist rather than a solution’ (193), a point argued for in detail in chapter 12, ‘The free-will offence’.

In his discussion of divine hiddenness (198ff) Schellenberg makes use throughout of the notion of ‘non-resistant non-believers’. Consider the following three propositions (204):

1. (1) 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.

2. (2) Necessarily, one is at a time in a position to participate in meaningful conscious relationship with God only if at that time one believes that God exists.

Hence,

1. (3) Necessarily, if God exists, anyone who is (i) not resisting God, and (ii) capable of meaningful conscious relationship with God, also (iii) believes that God exists.

‘We can see that this state of affairs does not actually obtain’, which is to say ‘God is hidden’ (205). Moreover, this leads inexorably to a further conclusion (206). Since,

1. (4) There are (and often have been) people who are (i) not resisting God, and (ii) capable of meaningful conscious relationship with God without also (iii) believing that God exists,

it follows, from the conjunction of (3) and (4), that

1. (5) God does not exist.

Importantly, we are here dealing with non-resistant non-believers, not with ‘people walking around demanding that God “show himself”. Some philosophers may do this, but they are usually individuals who have long since concluded that God does not exist and think the world is better off that way’ (231–232).

And so, finally on this point, ‘God is, if God exists, neither unjust nor (to reintroduce the other relevant attributes) ungenerous, uncaring, or unloving. Thus the existence of lifelong seekers suggests the non-existence of God’ (234).

What about the ‘perennially interesting [but] flawed’ (257) argument from free will? Once again I shall simply report a small portion of Schellenberg's subtle argument, altering Schellenberg's symbols slightly. (The argument, which I here compress, occurs on 283–284.)

Suppose finite persons possess and exercise free will. Let F be the claim that this state of affairs obtains. Let G be the claim that God exists. Let E be the claim that the possession and exercise of free will by finite persons poses a serious risk of evil, and let O be the claim that there is an option available to God that prevents finite persons possessing and exercising free will.

But now, we have three straightforward premises:

1. (1) (G v F v E)→~O);

2. (2) O;

3. (3) E;

with three straightforward justifications.

For the first consider the

But if there is a countering option, then God will not countenance the combination of what we might call finite free will and its possibly attendant evil. Premise (1) follows by contraposition.

Premise (2) is argued for at some length in chapter 12, and I will not here recapitulate Schellenberg's argument, concentrating instead on what follows if he is correct in supposing that he has sketched a model of personal relationship between God and finite persons which justifies the acceptance of premise (2) (270–282). And finally, premise (3) we may take as obvious, ‘giving finite persons free will must always carry with it a serious risk’, a point not only accepted but emphasized by philosophers who want to utilize a free-will defence to the problem of evil. However, we see at once that our three premises yield the conclusion,

1. (4) ~G v~F,

that is, either God does not exist, or there is no free will, or, of course, both: God does not exist and there is no free will. Clearly, for anyone who goes along with the almost universal acceptance of free will, it follows, finally, that,

1. (5) ~G.

that is, God does not exist.

This review has merely scratched the surface of this remarkable work. It may be that some will treat the first section, in which the importance of considering the relevance of ultimism is stressed, as being less relevant to our immediate concerns, and treat the third section of the book as being: (a) more directly relevant, and (b) lending strong support to atheism. This, however, would be to misread both the book and Schellenberg's stance and intent. Although his argument leads us into ‘the dark valley of religious skepticism’, it may well be that ‘there are important intellectual vistas we can see only from there’ (311).

Let me finish as I began: this is an extremely important book: philosophers in all areas will benefit from reading it, and thinking about its arguments, and it should be required reading for philosophers of religion.