The Argument from Evil, the Argument from Hiddenness, and Supernaturalistic Alternatives to Theism

Abstract

In this brief article, I consider James Sterba’s logical argument from evil, finding it to be ultimately unsuccessful. Not for the various issues Sterba raises, which do seem to be problematic if God exists, but for the logical approach itself. I encourage Sterba to shift tack, to embrace the evidential argument from evil, which is not at all concessionary, as he seems to think, and is an extremely powerful argument against the probability of theism, especially when we open the debate to the supernaturalistic alternatives to theism. I also encourage Sterba to reconsider his dismissive attitude towards the argument from hiddenness, which, in its evidential form, is also a very powerful argument against God’s existence, either employed independently or incorporated into the argument from evil.

1. Introduction

The argument from evil is a very powerful argument against God’s existence. This revolves around the notion that God, the god of classical theism, who is, among other things, omnipotent, omniscient, and omnibenevolent, finds some contradiction with the evil or suffering in the world. The logical argument from evil has long been out of favour in the philosophy of religion; however, the evidential argument from evil, tending to focus more on gratuitous evils, and shedding the unnecessary burden of deductive certainty in favour of sound probabilistic reasoning, as argued by the likes of William L. Rowe (Rowe 1979) and Paul Draper (Draper 1989), remains.

James P. Sterba (University of Notre Dame) recently published Is a Good God Logically Possible? (Sterba 2019), arguing that God’s existence is incompatible with all the evil in the world. He presents a logical argument from evil and has sportingly invited a great many talented philosophers to critique it. In this article, I consider how Sterba’s opting for the logical approach is fraught with difficulty, and how a few alterations improve his argument from evil considerably.

2. Sterba’s Logical Argument from Evil and Objections

There are different ways in which ‘logical’ and ‘evidential’ arguments from evil have been conceived (as explained in Howard-Snyder 1996). Recently I was discussing with my colleague Stephen Law, who has done his own interesting work on the argument from evil (Law 2010), and it appears that he thinks the distinction is marked effectively by the number of evils under consideration. However, when it comes to Sterba’s work and my response, logical arguments involve certainty, typically revolving around notions of impossibility, whilst evidential arguments revolve around evidences that support one hypothesis over another. The first sentence of Sterba’s introduction, which aligns well with his book’s title, makes this clear (Sterba 2019, p. 1): “The question I seek to address in this book is whether or not an all-good God who is also presumed to be all powerful is logically possible given the degree and amount of moral evil that exists in our world.” To end all doubt, Sterba, in his conclusion, states (Sterba 2019, p. 182): “All three sub-arguments conclude to the logical impossibility of God.”

It is this that makes Sterba’s efforts here so novel: he is attempting to revive an almost long-forgotten approach to the questions at hand. Sterba summarises his argument as the aforementioned three sub-arguments (Sterba 2019, pp. 185–89). The first is his Argument from the Moral Evil in the World, which includes premises such as, “there are significant and especially horrendous evil consequences of immoral actions that, if God exists, would have to have resulted from God’s widespread violation of Moral Evil Prevention Requirement I, which is logically incompatible with God’s existence, unless there is some other justification for God’s permitting those evil consequences”. The second is his Argument from the Natural Evil in the World, which finds that “the significant and especially horrendous consequences of natural evil that exists in the world would BE LOGICALLY INCOMPATIBLE WITH GOD’S EXISTENCE”. The third is his Argument from the Lack of God’s Law-like Prevention of Evil, including premises such as, “although an ideally just and powerful state would do its best to abide by Moral Evil Prevention Requirement I–III and Natural Evil Prevention Requirements I–IX, only God, if he exists, could and should insure, as needed, with law-like regularity, that there would be no significant and especially no horrendous consequences of moral and natural evil inflicted on their victims in violation of these requirements, and hence no second-order goods that would otherwise result from such evil consequences”.

I find this argument quite compelling, though not enough to declare that now we certainly have a way to conclusively prove that God does not exist. While theists cannot really move the goalposts much on God’s omnipotence and omniscience, there seems to me to always be sufficient wriggle room to raise objections about what God’s omnibenevolence really entails; what ‘good’ actually is; what ‘evil’ actually is; Reichenbach seems to do just this (see Reichenbach 2021, pp. 4–5); and what can be considered reasonable prevention requirements, trade-offs, rights, justifications, and rationale, even if that results in significant concessions or explanations that need not be probable, merely possible. Perhaps like Michael Tooley or Laura Ekstrom, I would try to convince Sterba to go down the more fruitful evidential/probabilistic route.

To assert that there is finally a sound logical argument from evil is very bold, so, naturally, there have been many replies to Sterba’s work. One of Elizabeth Burns’ objections is that “it is not necessary to define divinity in this way, and that this is the third and fatal flaw in Sterba’s argument” (Burns 2021, p. 8), going on to speculate on different ways God could be conceived, stating that “the God in question is not a God of the kind that Sterba describes” (ibid. p. 11). This is in alignment with my concern that the critics merely need to disagree on definitions, though this particular attempt seems far too costly. By making big alterations to what God is, we are no longer discussing God, the god of classical theism. Indeed, such theodicies actually reinforce the power of the argument, strengthening the view that the being in question does indeed not exist, and perhaps it is another god that exists (a notion we shall revisit).

William Hasker demonstrates another major approach to critiquing such arguments, questioning what the relevant terms, such as ‘prevention requirement’, actually mean and entail (Hasker 2021), whilst also cleverly charging that, if Sterba insists on such requirements being placed on God, then it is he, Sterba, who is guilty of crafting a straw person argument, of effectively creating his own version of God, who “was devised precisely in order to show that he does not exist”, and one who “has little or nothing to do with the existence of the God in whom Christians believe—Yahweh, the God of Israel, the Father of Jesus Christ” (p. 7). Even if Sterba is correct, it takes little effort for the believer to quickly alter, in a relatively minor way, their view of God, not massively, just enough to avoid the consequences of Sterba’s premises. This is one of the primary reasons I find the logical approach untenable. Aside from the fact that absolute certainty is required, which is itself a likely insurmountable obstacle, if only for our cognitive limitations, the theist need only assert that the sort of god Sterba or others is arguing against is not exactly their own god. It is all too simple for the theistic philosopher to add more and more on to the definition of their god, par for the course, unless critics opt for the evidential/probabilistic route, where this would immediately be exposed as doing nothing for (perhaps even harming) the prospects of showing theism to be probable.

Sterba acknowledges that his argument could contain a fatal flaw and invites criticism (p. 191). For me, the ‘fatal flaw’ is, of course, the appeal to logical argument in the first place. While theists and non-theists alike find it troubling that there would be evil in a world created and effectively run by a god who could easily prevent it, the theists’ old appeal to God’s inscrutability is successful, particularly if they are open to (slightly) shifting the goalposts. God could always have some absolutely befuddling—to us—reason for allowing things we consider evil, and this is made worse by the fact that maybe those things are not evil after all. Just as with most things, we cannot be certain. That is why I suspect that all the logical arguments against God’s existence must ultimately fail. Despite the logical argument from evil being very powerful, including Sterba’s form, bringing theists and atheists together in acknowledging that there is something quite odd about the perfect God allowing so much torment and injustice, logical arguments will always be subject to our epistemological gap. We simply cannot be sure about, well, anything (except our own existence, with some disputing even that, and perhaps, by extension, the universe’s existence).

Sterba has done his homework and knows the typical objections that would arise when crafting a logical argument from evil. There is no need to repeat it all here, as appeals to nebulous concepts of God’s intentions and free will are well known. Sterba, as is to be expected, finds these objections wanting, and so he should. However, it is not enough, when positing a logical argument, an argument from certainty, that the theistic hypotheses are wanting. It is not enough for Sterba to rely on words peppered throughout his book such as ‘widespread’ and ‘significant’. When crafting a logical argument, we need to be sure, and we simply cannot be. We must accept to some extent, as Socrates allegedly did, that ‘we know nothing’.

However, if we drop the requirement of certainty, and embrace probabilistic reasoning, namely ‘refining’ Sterba’s argument into an evidential form, theism faces—at present—an insurmountable challenge. Yes, it may be possible that God has some odd reason for allowing these evils, but is it probable? Are there not alternatives to theism where the presence of these evils, amongst other things, makes more sense? Suddenly the burden is thrust right back at the proponents of theism, who can no longer appeal to mere epistemic possibilities, (not even confirmed as ontological possibilities), but must now show why their innumerable excuses and theodicies, are probable, and this they generally cannot do, especially when God’s inscrutability is widely considered a feature and not a bug. However, shift tack to the evidential approach, and now the appeals to God’s inscrutability fall apart, to such an extent that the case for God’s existence is, in general, already over, as every excuse lumped in together with the God hypothesis drags its probability further and further downwards (see Lataster 2018).

I sympathise with Sterba, as he has taken on an impossible task. Furthermore, it is unnecessary. While it is extraordinarily difficult to produce a sound logical argument from evil, impossible even, it is just as extraordinarily easy to produce a compelling evidential argument from evil. In a flash, all the typical objections fall apart. It is not enough to simply say ‘God might this’ and ‘God might that’, and ‘but will somebody please think of the free will?!’. The theist can have no compelling retort. They are in a worse position than the atheistic proponent of the logical argument. They rely on notions that are unproven and seem very improbable. ‘Free will’ suddenly looks to be a hollow claim when we realise that we can’t be sure it exists (interestingly, the Bible hints that it effectively does not in many passages, such as 2 Thessalonians 2:13), and that we might exercise our hypothesised free will more ideally if we didn’t have to face all these ostensibly unnecessarily evils that block our spiritual progress. I leave it to my fellow purveyors of knowledge to determine if the knowledge of one of the most important things one could know, that God exists, might have some impact on us properly exercising our free will to make the crucially important decision to follow God or not, and to obey his commands, moral and otherwise.

However, it is not even necessary, once we embrace the evidential/probabilistic approach, that we thoroughly scrutinise and analyse, revealing just how improbable God’s existence seems to be, compared with a naturalistic world order.

3. An Evidential Revision

Logical arguments from evil suffer, and Sterba’s is unfortunately no exception, beholden to the fact that theists can always appeal to mere possibilities, no matter how extraordinarily improbable (such as the supernatural afterlife), as well as unknowns (God’s supposed inscrutability, being particularly key). This is still possible with the evidential arguments from evil, however, they do not help, and arguably even make the situation more dire for the theist.

As outlined succinctly in The Case Against Theism (Lataster 2018), all the proponents of the evidential argument from evil need do is point to the apparent relative improbability of such a good, knowledgeable, and powerful god allowing seemingly unnecessary instances of suffering. It, at least, looks odd. While, on the hypothesis of naturalism, there is nothing odd about it. While on the hypothesis of naturalism, there is nothing odd about it. Nor is it odd on alternatives to both naturalism and classical theism. It would not be odd on theories involving several gods, who might plot to foil each other’s plans, or a hypothesis cen-tred around a morally indifferent god. It is not necessary for the atheist to prove that there is some incompatibility between God’s existence and the presence of unnecessary suffering. So long as all else is held equal (particularly easy to do when crafting alternative supernaturalisms) or, even further, adding to the case against theism, the critics of the case for theism only need to show that the presence of unnecessary suffering is more expected in some alternative to God’s existence, whether that be naturalism or other forms of supernaturalism. For example, imagine a god identical to God, except that it is not all-good or all-powerful. The presence of unnecessary suffering is more expected in this god’s existence, making its existence more probable than God’s. Since there are several such hypotheses, it follows then that God’s existence is very improbable.

It is obvious to most that the presence of unnecessary suffering is surprising, if God exists, even to theists, who grapple with the problem of evil and try to contrive ingenious solutions. None of this would be necessary if it were so obviously expected, as it would be in naturalism or other alternatives. Theists come up with notions about the afterlife, free will, God’s inscrutability, the possibility that God has perfectly reasonable but currently unknown reasons for allowing such, etc. All are things that adversely affect the probabilistic case for theism, while doing no harm at all to the cases for naturalism, certain polytheisms, certain forms of alternative monotheisms, and certain pantheisms. By broadening our scope to consider not only theism and naturalism, to consider numerous other divine models, it becomes clear that the presence of unnecessary suffering in the world is damning evidence against theism indeed, rendering theism extraordinarily improbable, even if some sort of god, such as a pantheistic one, actually exists. Hence, even if the theist refines their view of their god, helpful when swatting away the logical argument from evil, their efforts are in vain. The theist cannot refine their divine model so much that they become a pantheist, for example. It would then be they, and not Sterba, who would be describing a god quite different from what they began with.

Laura Ekstrom agrees that, while the focus on evil will be fruitful, the evidential approach is optimal, with Sterba objecting: “In his debates with atheists, Craig is especially good at getting his opponents to admit that given their arguments, God is still logically possible.” That concession, at least since Plantinga’s exchange with Mackie, is taken by the theists to be quite significant. So, it is worth noting that it is a concession my argument does not make to the theists, but Ekstrom’s does (Sterba 2021, p. 3).” [Revert all this, there is no justification for editor to alter quoted passages.] The focus ought to remain on what we can demonstrate to be probably the case, rather than what our intellectual opponents might or might not say or do. Furthermore, the possibility of God’s existence is not much of a concession, since so many things are possible in an epistemic sense, and, if it is Plantinga’s ontological argument that Sterba hints at, that is discredited for this very reason (confusion over epistemic and ontological possibilities), which Plantinga himself admits (Plantinga and Sennett 1998, pp. 65–71). Theistic philosophers apparently do not tend to lose sleep over the possibility that God does not exist, and I advise Sterba to be unconcerned that God’s existence can be said to be possible. In any case, the evidential approach is not at all concessionary; it is the method we all use in attempting to get closer and closer to objective truths, as in the natural sciences, and, used correctly, renders theism a particularly improbable hypothesis. The case for theism is done yet more damage in expanding the evidential approach to consider all the alternative hypotheses. The only true concession is that we are intellectually humble in accepting that we can know almost nothing with certainty, but this should be considered a fundamental requirement for objective intellectuals, to avoid being lumped in with those who cling to unproven beliefs.

Sterba could say that the outs I grant the theist when it comes to a logical problem of evil are too generous, and that they would be improbable. In that, he would be correct, though this further reinforces that the more fruitful approach is evidential; the evidential approach is transparent and objective. Discussions about what is improbable or not would be futile in arguing with a theist over a logical argument; over an evidential/probabilistic argument, however, swatting away improbable theodicies is precisely the point. It is then a very easy task to show that these ‘excuses’ do not raise the probability of God’s existence, and may even lower it, with the improbability merely shifted from one side of the equation (for instance, the consequent probabilities’ or likelihoods’ side) to the other (for instance, the prior probabilities’ side). With the probabilistic approach, there is no reprieve to be had via improbable excuses, mere possibilities, and sooner or later the bill comes due (for more on the probabilistic approach, and how it renders theism very improbable, see Philipse 2012; Lataster 2018). Of course, I do not claim that evidential arguments will be more compelling to the populace than logical arguments. The point here is simply that, as with democracy, being a system of governance with many drawbacks, we must make do with the best we have. Tangentially, I suspect that certain actors in the field would prefer the discussion revolved around logical arguments, realising their futility, and shifting the focus away from the evidential arguments, whose conclusions are damning and indisputable.

4. Hiddenness

Like so many philosophers and scholars of religion, I consider the problem of evil, or of gratuitous suffering, to be a major issue for theistic philosophers attempting to demonstrate that God, the god of classical theism, exists. Even more impressive to me, however, is the problem of divine hiddenness. It was disappointing then, to see Sterba very casually dismiss the ‘need’ for the problem of hiddenness in his book, presumably because the problem of evil is so powerful. Sterba even says that, if his argument works, “it would no longer make sense to go on to raise a problem of divine hiddenness”.

Like the problem of evil, the problem of hiddenness is so powerful because it points to a piece of evidence in the world that seems quite odd, or is even outright unexpected, if God exists. That God would remain hidden from the people he desires a relationship with, and apparently makes it so much harder for those he gifted with more intelligence and more knowledge to come to believe in him, seems utterly preposterous. To paraphrase J.L. Schellenberg, while remaining relatively reserved, if God exists, it is at least quite odd that there also exists non-resistant non-belief (Schellenberg 1993). At best, this great non-resistant non-belief in God is not exactly expected in theism, if not outright unexpected. However, this is not the case with naturalism. In naturalism, the evidence of divine hiddenness is 100% expected. If God does not exist, we certainly would not expect him to show up. The same can be said for supernaturalistic alternatives to theism, such as the deisms. If there were a god who cared not for human interactions, and even wished to remain undiscovered, it would make sense that so many of us honest seekers after truth do not encounter him. This makes the argument from divine hiddenness very powerful, on its own, in demonstrating theism’s improbability. This evidence is more expected in naturalism, and more expected in other alternatives. It is less expected in theism. This is simply indisputable. It is not out of the realm of possibility that God, who by definition and according to theistic religious traditions, wants a relationship with us, would again have some absolutely befuddling—to us—reason to remain hidden, but it is certainly not 100% expected, as it would be for alternatives such as naturalism or certain deisms. I see no reason to overlook this, whether or not Sterba’s logical problem of evil is considered successful. His position is especially troublesome, as it could be used against him. For example, if we have another good argument against God’s existence—and we do (several, in fact)—we could say that we have no need for Sterba’s work on the argument from evil, particularly if I am correct in supposing that the logical approach is fruitless.

Furthermore, the hiddenness of God can itself be considered a great evil. If God is so good, one would expect that God would make more of an effort to grace us with his presence, which presumably would improve all of our lives, increase our chances of ending up in Heaven rather than Hell, make life easier for theistic scholars like Craig, Plantinga, and Swinburne to convince those pesky atheists that God really does exist, help convince those who commit evil acts to reconsider, inspire those who commit good acts to continue, etc. Here, we have one more totally unnecessary, and arguably totally unexpected, evil that God is ultimately responsible for. I implore Sterba, then, to not only embrace the power of the hiddenness argument, but even to incorporate it into his work on the problem of evil, creating a ‘superargument’, as it were.

5. Conclusions

I do not wish to be overly cautious in declaring Sterba’s argument a failure. As a logical argument, it just is, though that is true of nearly all logical arguments. The problem is not with Sterba’s conceptualisations, which can and ought to be yet utilised, but the logical approach itself. It allows theistic critics too much wriggle room, both with a god that can be ever-so-narrowly defined, and with mere possibilities as outs. When seen from an evidential standpoint, however, Sterba’s argument is a very good one, making a worthwhile contribution to the philosophy of religion.

By shifting focus to the evidential arguments against God, atheistic claims are far more reasonable and justifiable, with theistic retorts being increasingly impotent. This is amplified further when we consider the argument from divine hiddenness, alluding to yet another evil in the world, and accept that God is a very specific type of god, opening the analysis up to supernaturalistic alternatives to theism, revealing that, since many gods are very compatible with the evils in the world, including divine hiddenness, that God’s existence is incredibly improbable indeed. I implore Sterba to embrace the argument from hiddenness, even incorporating it into his work on evil, and to fully embrace the probabilistic approach. I further implore the philosophers of religion, in general, to pay more heed to the supernaturalistic alternatives, so common outside of the Western world, and who must play a key role in determining the probability of classical theism being true.