The nature of sceptical theism

In this paper, I understand sceptical theism as theism (there exists a unique omnipotent, omniscient, morally perfect, necessarily existent creator of the universe) together with the claim that, for all we know, there are lots of goods, evils, and connections between good and evil of which human beings are unaware.Footnote ³ We may think of the collection of facts about goods, evils, and the connections between them as an iceberg. The sceptical theist maintains that we just do not know how much of this iceberg is visible to human beings; for all we know, we see most of the berg, and for all we know we see only its tip. As Michael Bergmann points out, sceptical theists need not maintain that a significant portion of the iceberg is in fact not visible to humans, or even that there is reason to believe this; the claim is merely that we cannot tell how much of the iceberg is visible. As Bergmann puts it, what the sceptical theist maintains is simply that ‘it wouldn't be the least bit surprising if [axiological] reality far outstripped our understanding of it’.Footnote ⁴ Thus, sceptical theism as I will understand it here does not include the claim that our knowledge of good, evil, and the connections between them is massively incomplete; instead, it includes the weaker claim that our knowledge of the completeness of our knowledge of good, evil, and the connections between them is incomplete. When it comes to good and evil, we do not know how much we do not know.

With this sketch of sceptical theism in hand, let us consider how sceptical theism relates to the argument from evil. The following definitions will be useful in that task:

• D1 E is an instance of gratuitous evil=df. (1) E is an instance of evil, and (2) nothing would justify God in permitting E (if there were a God).

• D2 E is an instance of inscrutable evil=df. (1) E is an instance of evil, and (2) having thought about the matter really hard, we know of nothing that would justify God in permitting E (if there were a God).

• D3 Divine action A has beyond-our-ken justification=df. (1) there is some great good, G, such that (1) if God were to perform A, G would justify God in performing A, and (2) G or G's connection with A is beyond our ken.

It should be noted that, according to D3, that a given divine action has beyond-our-ken justification does not imply that the divine action does not also have justification that is within our ken. Additionally, D3 allows for both actual and merely possible divine actions to have beyond-our-ken justification. There may be possible divine actions that God would be justified in performing but which He does not perform.

Some versions of the problem of evil depend on the so-called ‘noseeum inference’, which can be understood thus:

• The noseeum inference E is an instance of inscrutable evil; therefore, (probably) E is an instance of gratuitous evil.Footnote ⁵

Together with the observation that inscrutable evil exists and the assumption that the existence of gratuitous evil is incompatible with the existence of God, the noseeum inference yields the conclusion that God probably does not exist.

Sceptical theists deny that the noseeum inference is a good one. One sceptical theistic critique of the noseeum inference consists of three main elements. To understand these three elements, suppose that a critic of theism points to the existence of a particular inscrutable evil, E, and offers E as evidence against God's existence. In response, sceptical theists first point to our ignorance concerning the completeness of our knowledge of good, evil, and the connections between them.

Second, sceptical theists claim that this ignorance implies that ‘we're in the dark about whether – and how likely it is’ that the particular evil E is, unknown to us, inextricably connected with some good such that God's act of permitting E has beyond-our-ken justification.Footnote ⁶ As Alvin Plantinga puts it, ‘God might have reasons we cannot so much as understand’.Footnote ⁷ The sceptical theist's claim here is that we lack justification for believing that it is false or unlikely that God's act of permitting E has beyond-our-ken justification. Bergmann puts it this way: ‘[W]e lack any good reason or justifying grounds for thinking that there isn't … a God-justifying reason [for permitting E].’Footnote ⁸

This leads to the third element of the sceptical theist's critique of the noseeum inference: because we lack justification for believing that it is false or unlikely that God's act of permitting E has beyond-our-ken justification, E's inscrutability is not a reason to think that E is, or is likely to be, gratuitous. Thus, the three main elements of the sceptical theist's critique of the noseeum inference may be (roughly) summarized as follows:

The sceptical theist's critique (STQ)

1. (1) We don't know how complete our axiological knowledge is.

2. (2) So, we do not know how likely it is that God's act of permitting E has beyond-our-ken justification.

3. (3) Therefore, E's inscrutability does not justify the claim that E is (or is likely to be) gratuitous (i.e. the noseeum inference fails).

It would be a mistake simply to identify sceptical theism with the rejection of the noseeum inference. That the noseeum inference fails is an implication of sceptical theism. The foundation of the sceptical theist's rejection of the inference is the claim that ‘it wouldn't be the least bit surprising if [axiological] reality far outstripped our understanding of it’.Footnote ⁹ The sceptical theist appeals to this claim to support claim (2) above, and claim (2) in turn is the basis of the rejection of the noseeum inference.

The kind of reasoning that sceptical theists employ to defuse the noseeum inference can be used in other ways as well. Consider the following example: A story in The Guardian (31 May 2006) on global warming attributes the following remark to a ‘Washington lobbyist on the religious right’: ‘Is God really going to let the earth burn up?’Footnote ¹⁰ Presumably the lobbyist was implicitly suggesting that significant global warming will not occur because a perfectly good God would not permit it to occur. But consider the following line of reasoning: (1) We don't know how complete our axiological knowledge is; (2) so, we do not know how likely it is that God's act of permitting significant global warming has beyond-our-ken justification. Furthermore, God certainly possesses enough power and knowledge to permit or even cause global warming. Thus, we lack grounds for believing that it is false or unlikely that God will permit global warming – unless, of course, we somehow have good reason to believe that even if God's act of permitting global warming has beyond-our-ken justification, God won't act on that justification. Since it appears that we do not have good reason to believe this, sceptical theism, together with some additional plausible premises, leads to the conclusion that we are not justified in believing that it is false or unlikely that God will permit significant global warming.

Please note: I do not assert that this reasoning is sound. My position instead is that this reasoning is at least as plausible as STQ. My claim is that if sceptical theism generates a successful rejoinder to the argument from evil in the way described in STQ, then sceptical theism also generates broader uncertainty about what God will or will not do, in at least some cases.

Divine lies

The Christian Bible depicts God as making certain declarations to human beings about what He will or will not do in the future. For example, the Old Testament reports that God told Noah that ‘never again shall there be a flood to destroy the earth’, and in the New Testament the writer of the Gospel of St John attributes to Jesus the assertion that ‘all who see the Son and believe in him may have eternal life; and I will raise them up on the last day’.Footnote ¹¹

It will be helpful at this point to introduce one more definition and to make two assumptions. First, the definition:

• D4 Proposition p has word-of-God justification only=df. the only reason we have for believing p is that God told us that p.

Given this definition, let us assume for the sake of argument that (1) God said the things attributed to Him in the scriptural passages quoted above, and (2) the propositions asserted by God in these statements have word-of-God justification only. The question to consider is the following: given these assumptions, can we know that these claims are true? Consider the following reasoning:

The threat of divine deception (TDD)

1. (1) IF sceptical theism is true, THEN, for any divine assertion that p, we lack justification for believing that it is false or unlikely that God's act of intentionally asserting that p when p is false has beyond-our-ken justification.

2. (2) IF, for any divine assertion that p, we lack justification for believing that it is false or unlikely that God's act of intentionally asserting that p when p is false has beyond-our-ken justification THEN we do not know p if p has word-of-God justification only (unless we have good reason for thinking that, even if God has some justification for lying about p, God doesn't act on that justification).

3. (3) So, sceptical theism implies that we do not know any proposition that has word-of-God justification only (unless we have good reason for thinking that, even if God has some justification for lying, God doesn't act on that justification).

4. (4) We do not have good reason for thinking that, even if God has some justification for lying, God doesn't act on that justification.

5. (5) Therefore, sceptical theism implies that we do not know any proposition that has word-of-God justification only.

In thinking about this argument it is important to remember that merely possible divine actions can have beyond-our-ken justification. Thus, this argument does not imply that any divine assertions actually are false.

Premise (1) can be supported by reasoning that parallels the first two elements of STQ. According to sceptical theism, for all we know, there are lots of goods, evils, and connections between good and evil of which we are unaware. Thus, we lack justification for believing that it is false or unlikely that God's act of intentionally asserting p when p is false has beyond-our-ken justification.

Premises (2), (3), and (4) can be supported by reasoning that parallels the reasoning employed above in connection with the global warming example. Suppose that we lack justification for believing that it is false or unlikely that God's act of intentionally asserting p when p is false has beyond-our-ken justification. Since God has the power and knowledge to assert p when p is false, we lack grounds for believing that it is false or unlikely that God has intentionally asserted p when p is false – unless we have good reason to believe that even if God's act of lying about p has beyond-our-ken justification, God won't act on that justification. Since it appears that we do not have good reason to believe this, we lack justification for believing that it is false or unlikely that p is false, and hence we do not know that p is true.

I do not claim that TDD is sound. Instead, I claim that TDD is at least as plausible as STQ. If this claim is correct, then either (a) sceptical theism implies that we do not know any proposition that has word of God justification only, or (b) STQ fails. Thus, sceptical theists who endorse STQ face a dilemma: reject STQ, or accept that we do not know any proposition that has word-of-God justification only.

Descartes, Kant, Hobbes, and Ross

Descartes maintains that God cannot deceive on the grounds that God is perfect and ‘the light of nature teaches us that deception must always be the result of some deficiency’.Footnote ¹⁷ Let us distinguish two kinds of cases. In the first sort of case, God deceives us unintentionally. God holds a false belief that p and tells us that p is true. In this kind of case, God exhibits an intellectual or cognitive deficiency. Such a deficiency conflicts with God's perfect nature (specifically, with His omniscience) and hence is impossible.

In the second kind of case, God knows that p is false but tells us that p is true in order to secure some great good (or prevent some great evil). If there is a deficiency here, it is a moral rather than intellectual one. But in order to know that this sort of deception on the part of God would imply a moral deficiency in God, we would have to know that God could never be morally justified in deceiving us. If TDD is sound, then the sceptical theist cannot consistently maintain that God could never be morally justified in deceiving us.

Is there any way a sceptical theist could reject TDD without also giving up STQ? One strategy for resisting TDD draws on the view that one thing that we human beings do know about morality is that lying is always morally wrong, regardless of the consequences. Kant seems to have held such a view.Footnote ¹⁸ If this Kantian view is correct, then we can know that no divine lie has beyond-our-ken justification (since no lie is ever justified), and hence premise (1) of TDD is false.

The central weakness of this line of reasoning is that the claim that lying is always morally wrong regardless of the consequences, whatever its pedigree, is implausible. It is not hard to think of situations in which it is quite plausible to suppose that lying is morally permissible. Many such situations involve lying in order to achieve a great good or to prevent a great evil. In his discussion of an ideal society, Plato advocates the rulers' use of a ‘noble falsehood’, a lie told to the citizens for the good of the society as a whole.Footnote ¹⁹ John Stuart Mill advocates lying when doing so is the only way to save someone from a ‘great and unmerited evil’.Footnote ²⁰ One of the examples mentioned by Mill is that of lying in order to withhold bad news from a dangerously ill person, a scenario that is explored at great length in James Morrow's science fiction novel about a man who lies to his dying son about the seriousness of his illness.Footnote ²¹ Indeed, scenarios in which parents lie to their children for the good of the children present some of the most plausible examples of permissible lying.

This theme is explored in both film and literature; examples include Morrow's novel as well as Robert Benigni's 1997 film, Life is Beautiful and Cormac McCarthy's novel, The Road.Footnote ²² These examples are particularly salient in the present context because theists often suggest that the relationship between God and His creatures is analogous to that between parents and their children.Footnote ²³ Interestingly, Hobbes appeals to just these kinds of examples in objecting to Descartes' claim that God cannot deceive us:

It might be suggested that while these kinds of examples show that it is sometimes permissible for imperfect human beings facing difficult choices to lie, they do not show that it is permissible for a perfect being to lie. After all, what reason could God have to lie to us? More to the point, what reason could God have to lie to us with respect to the claim that all who believe in Christ will have eternal life? It is hard to think of a divine justification for lying about something like that.

But of course this is precisely the sort of reasoning for which sceptical theism is supposed to make trouble. Sceptical theists maintain that the fact that we cannot think of a justification for God's permitting a certain evil does little to suggest that no such justification exists. If sceptical theism supports this claim, then it equally supports the claim that the fact that we cannot think of a justification for God's telling a certain lie does little to suggest that no such justification exists. At the very least, the sceptical theist owes us an explanation of what distinguishes the two kinds of cases. On what grounds can a sceptical theist rule out the existence of some spectacularly grand good that is connected with divine lying in such a way as to justify it?

Another reason a Christian theist in particular may find the Kant-inspired view that it is always wrong for God to lie untenable is that there are passages in scripture that appear to depict divine deception. For instance, the Gospel of St John depicts Jesus as telling his brothers that he will not attend the Festival of Booths in Judea but then attending it later in secret.Footnote ²⁵ And in one of his letters, Paul describes God as sending certain people ‘a powerful delusion, leading them to believe what is false’.Footnote ²⁶ Also relevant here is the binding of Isaac in the Old Testament.Footnote ²⁷ While God is not portrayed as explicitly lying in that episode, a natural reading of the episode has it that God intentionally creates in Abraham's mind the belief that Abraham is going to have to sacrifice Isaac, a belief that ultimately turns out to be false. The Kantian line may not be available to those who take such passages seriously.Footnote ²⁸

Such passages are also relevant to premise (4) of TDD. Here, again, is that premise:

1. (1) We do not have good reason for thinking that, even if God has some justification for lying, God doesn't act on that justification.

If the scriptural passages mentioned above do suggest divine deception, then such passages provide some indirect support for premise (4). They do this by presenting a challenge for any Christian who wants to make the case that we do have a good reason for thinking that, even if God has some justification for lying, God doesn't act on that justification. The passages discussed above may suggest that in at least some cases, God is justified in lying, and that He does indeed act on that justification.

The Kantian strategy for resisting TDD, then, is not particularly promising. However, perhaps there is a weaker view in the vicinity that will help the sceptical theist. Consider Ross's concept of prima facie duty.Footnote ²⁹ Perhaps lying is prima facie wrong; lying is actually wrong unless there is a sufficiently weighty moral reason to lie. Consider the claim that even if sceptical theism is true, we have some reason to think that it is false or unlikely that God lies. Divine lies, because they are lies, are prima facie wrong; that they are prima facie wrong constitutes a (defeasible) reason to believe that divine lies do not occur. Thus, premise (1) of TDD is false.

The problem with this line of reasoning is the following: the claim that lying is prima facie wrong is about as plausible as the claim that permitting evil which one could have prevented is prima facie wrong. According to STQ, sceptical theism implies that for each case in which God permits evil, we do not know that it is false or even unlikely that God's permission of that evil has beyond-our-ken justification. So, if permitting evil is prima facie wrong, sceptical theists must say that knowing that permitting evil is prima facie wrong does not enable us to know that it is false or unlikely that God permits evil. In light of this, it is hard to see how the sceptical theist can consistently claim that knowing that lying is prima facie wrong does enable us to know that it is false or unlikely that God lies. Thus, the Rossian strategy does not allow the sceptical theist to rebut my contention that STQ and TDD stand or fall together.

Further objections and replies

I have argued that sceptical theists who endorse STQ face a dilemma: reject STQ, or accept that we do not know any proposition that has word-of-God justification only. In this section I consider and respond to various objections to my argument.

One strategy for responding to my argument involves attempting to de-horn the dilemma by suggesting that religious believers should not be particularly troubled by the conclusion of TDD. According to this strategy, sceptical theists should respond to my argument simply by accepting the soundness of TDD. One version of this strategy rests on the claim that if God tells us something, we can be confident that He wants us to believe it, even if we cannot know that it is true. Moreover, the fact that God wants us to believe something is reason enough for us to believe it. In this way, it might be argued that even if TDD is sound, religious believers still have good reason to believe what (they think) God has told them.

The first thing to notice about this response is its unusual implications concerning the nature of divine revelation. I take it that the traditional understanding of divine revelation has it that if God tells us that p, then He is revealing the truth of p to us. On the alternative understanding under consideration, if God tells us that p, then He is not revealing the truth of p at all. Rather, He is revealing something about Himself, namely that He wants us to believe p. Furthermore, the kind of reason we have to believe p is not one that provides warrant for our belief that p (I use the term ‘warrant’ to indicate whatever factor(s) make the difference between mere true belief and knowledge). Instead, it is a reason that provides prudential and/or moral justification for our ‘act’ of believing p. The response in question construes divine revelations that p as a lot like divine commands to believe p.

A deeper problem with this response is that, given the soundness of TDD, it is hard to see how sceptical theists can be confident that when God tells us that p, He wants us to believe p. In order to know that, we would have to know what God's ultimate goal in telling us p is. But sceptical theism appears to make it impossible for us to know this. Given sceptical theism and the soundness of TDD, we lack justification for believing that it is false or unlikely that God's act of telling us that p has beyond-our-ken justification. And this implies that, for all we know, God's goal in telling us that p is not that we come to believe p but rather that we react in some other fashion. If we have no idea what God's ultimate goals may be, then we have no idea what His more immediate goals may be, and hence we have no idea how He wants us to respond when He tells us something – or, indeed, does anything else. The basic problem here is that if we have no idea what God is up to, then we have no idea how He wants us to respond to His interactions with us.

There is another response that also involves accepting TDD but seeking to defuse its importance. This response begins with the observation that nothing in TDD implies that God is not morally perfect. Thus, even if we cannot know any propositions that have word-of-God justification only, we can at least be sure that, in the end, everything will turn out for the best.

It is important to recognize just how little this tells us in the context of sceptical theism. Given sceptical theism, it may be that the ultimate goods that God is pursuing are entirely beyond our understanding. So, even if we can be confident that everything will turn out for the best, we have no grounds for accepting any particular specification of just what form this great final outcome will take. For example, for all we know, everything turning out for the best includes all those who believe in Christ being annihilated at the moment of death. This seems a far cry from any recognizably Christian understanding of the fate of believers after death. To know that all will turn out for the best is, in the context of sceptical theism, to know very little. It might be more accurate to say that what can be known, despite the soundness of TDD, is simply that everything will turn out for the best in the end – whatever that means.Footnote ³⁰

Thus, it appears that the sceptical theist who wants to endorse STQ faces the task of making the case that TDD and STQ do not stand or fall together. This would involve making the case that although our lack of knowledge regarding the completeness of our axiological knowledge prevents us from knowing, with respect to any particular evil, that it is false or unlikely that God's act of permitting that evil has beyond-our-ken justification, our lack of knowledge regarding the completeness of our axiological knowledge does not prevent us from knowing, with respect to any particular proposition that has word-of-God justification only, that it is false or unlikely that God's act of lying in asserting that proposition has beyond-our-ken justification. I cannot see how such an argument would go; of course, it does not follow that no such argument can be made. Perhaps this paper will prompt some sceptical theist or other to make this kind of argument.