A great number of evils were and are committed in the name of supernatural beings, often with the explicit approval of religious leaders. I'll argue that this kind of evil (‘religious evil’, as I'll call it) constitutes a special class of evils which defies all theistic responses to the problem of evil currently on offer. In the last section of the article, I'll outline a tentative solution.
The concept of religious evil
A number of tragedies in recent world history have made it clear that religion is sometimes implicated in evil. Fundamentalist terrorism, sectarian violence, and religiously justified political oppression are painful reminders that the word of God can be abused in horrible ways. It seems reasonable to suggest that this phenomenon is related to the problem of evil.
Although the connection between religion and evil is of topical interest, I'd like to discuss it from a more abstract standpoint, without engaging with current issues. This strategy is justified because religious evil seems to be an old and persistent phenomenon, one that defiles large parts of the history of religion.
To simplify matters, I'll portray the problem of religious evil as a problem for traditional monotheism (specifically, as a challenge to Christian faith), but my argument can be applied, mutatis mutandis, to religion in general.
To clarify the concept of religious evil, I propose to examine some historical incidents that are remote from our day and age and hence can be considered from a certain emotional distance. And to emphasize that religious evil isn't a non-western, non-Christian phenomenon, I propose to look at religious evils perpetrated by western Christians.
Two of the most repugnant series of evils perpetrated by western Christians were the witch hunts and the crusades. The witch hunts involved the persecution, humiliation, torture, and horrendous death of thousands of innocent civilians throughout the western world from the fifteenth to the seventeenth centuries. These atrocities were usually condoned by local churches and were justified by religious ideology. From our perspective, the witch hunts appear to have been pogroms against people who had pagan or other alternative religious beliefs, or were in the way of their fellow citizens for some emotional or political reason.Footnote 1 Sometimes, witch hunts were not driven by any intelligible logic but took on a diabolical life of their own, as shown by the following report, written in 1629 by the chancellor of the Prince-Bishop of Würzburg:
The richest, most attractive, most prominent, of the clergy are already executed. A week ago a maiden of nineteen was executed, of whom it is everywhere said that she was the fairest in the whole city, and was held by everybody a girl of singular modesty and purity. She will be followed by seven or eight others of the best and most attractive persons. … And thus many are put to death for renouncing God and being at the witch-dances, against whom nobody else has ever spoken a word. … [T]here are children of three and four years, to the number of three hundred, who are said to have had intercourse with the Devil. I have seen put to death children of seven, promising students of ten, twelve, fourteen, and fifteen. (Kors & Peters (Reference Kors and Peters1972), 252)
By any reasonable moral standard, these executions were wholly gratuitous acts of horrendous evil. They also seem to be paradigm cases of religious evil, since they were apparently motivated by Christian principles and they were justified in religious language.
Let's take a brief look at the crusades before we pack the most important characteristics of these incidents into a definition. The crusades have a long and complex history which need not be rehearsed here. For our purposes, it is enough to quote two famous documents. Consider, first, a report of the sermon that Pope Urban II delivered at the general church council in Clermont in 1095, officially launching the first crusade:
I, or rather the Lord, beseech you as Christ's heralds to publish this everywhere and to persuade all people of whatever rank, footsoldiers and knights, poor and rich, … to destroy that vile race [=the Turks and Arabs] from the lands of our friends. I say this to those who are present, but it is meant also for those who are absent. Moreover, Christ commands it. All who die by the way, whether by land or by sea, or in battle against the pagans, shall have immediate remission of sins. This I grant them through the power of God with which I am invested. (Allen & Amt (Reference Allen and Amt2003), 40)
And consider the following first-hand account of the fall of Jerusalem during the first crusade, written by a French chaplain, Raymond of Aguilers, in 1099:
[N]ow that our men had possession of the walls and towers, wonderful sights were to be seen. Some of our men (and this was more merciful) cut off the heads of their enemies; others shot them with arrows, so that they fell from the towers; others tortured them longer by casting them into the flames. Piles of heads, hands, and feet were to be seen in the streets of the city. … But these were small matters compared to what happened at the Temple of Solomon … [I]n the Temple and porch of Solomon, men rode in blood up to their knees and bridle reins. Indeed, it was a just and splendid judgment of God that this place should be filled with the blood of the unbelievers … . This day, the Ides of July, shall be celebrated to the praise and glory of the name of God, who, answering the prayers of his church, gave in trust and benediction to his children the city and fatherland which he had promised to the fathers. On this day we chanted the office of the resurrection, since on that day he, who by his virtue arose from the dead, revived us through his grace. (ibid., 77–78)
Using the crusades and the witch hunts as paradigms, we can characterize religious evil as evil that is apparently motivated by beliefs about, and is explicitly justified by reference to, the purported dictates of some supernatural entity or principle. Perpetrators of religious evil, as well as their spiritual leaders, use religious language to justify the evils in question, portraying them as virtuous, beneficial, fully endorsed by supernatural powers, and perhaps even explicitly called for by revelation.
Although the examples so far all involved horrors (torture, mass murder, execution of children, etc.), religious evil may not be confined to such atrocities but may involve all sorts of repugnant practices. Religion is often charged with such evils as hindering intellectual progress, supporting political oppression, spreading homophobia, xenophobia, and misogyny, limiting personal freedom, etc. To the extent that these charges are justified, the evils in question fall under the category of religious evils. But in the following, I won't assume that the charges in question are justified. Neither will I assume that they are unjustified. Rather, I'll restrict my attention to evils like the crusades and the witch hunts, because their historical reality is undeniable and because any discussion of religious evil is bound to address these incidents first of all.
Religious evil as a defeater of theism
Religious evil, being a form of evil, is part of the problem of evil. But it isn't just one more entry in a long list of evils that theists must reconcile with their belief in a morally perfect, omnipotent creator. Religious evil occupies a special place among evils, because it calls into question the moral authority of religion. Prima facie, religious evil is evidence that theism itself increases the amount of suffering in the world. And one might argue that this fact undermines the normative coherence of theism, along the following lines:
(1) Belief in God causes evil.
(2) God condemns evil.
(3) Whatever God condemns is objectively wrong.
(4) Therefore, by (1)–(3), belief in God causes something objectively wrong.
(5) Theists should refrain from doing anything that causes something objectively wrong.
(6) Therefore, by (4) and (5), theists should refrain from believing in God.
The argument is valid. Premise (1) is prima facie justified by religious evil. A sane theist can hardly question (2) – she must agree that God condemns war crimes, the torture and humiliation of innocents, and the murder of children, all of which have been committed in the name of God. Premise (3) is axiomatic for theists. And since (5) also seems beyond reproach, (6) appears to follow readily from theistic beliefs plus facts about religious evil.
The argument generates normative defeat for theism, portraying theistic belief as normatively self-refuting. If the argument is sound, then religious evil forces the rational theist to give up her faith.
It might be objected that the argument presupposes consequentialism about ethics. Specifically, premise (5) implies that the moral value of an action depends on its effects. If the theist goes deontological, she can avoid normative defeat by claiming that faith in God is a divinely ordained duty, regardless of its consequences, hence (6) does not follow. However, it should be obvious that this fix is both desperate and futile. For the atheologian can argue that (4) is enough to render theistic faith morally objectionable.
Alternatively, the theist can claim that religious belief causes evil only in special circumstances, for example, under unfavourable historical and economic conditions, or when the number of potential sociopaths reaches a critical level in a religious community.
But it is unclear whether this objection truly saves the theist from normative defeat. For suppose that an action A (e.g. planting landmines in a certain area) only causes evil under special circumstances (e.g. if the area in question is frequented by humans or animals). It might nonetheless be true that A is a sin, or, at any rate, that it is definitely neither virtuous nor commendable. Hence, one can coherently maintain that theistic faith is morally objectionable even if it causes evil in special circumstances only.
But suppose we disregard this point and we accept that theistic faith is not normatively undermined as long as it causes evil in special circumstances only. Then the atheologian can argue that religious evil is nonetheless evidence that theism is false, because a morally perfect God would not create a world where faith in Him causes evil under any circumstance. As long as we accept the premise that theism does cause evil, we have a defeater for the claim that theistic belief is an unambiguously positive social phenomenon. And it seems intuitively clear that a morally perfect creator would take pains to make belief in Him an unambiguously positive social phenomenon. So even if the appeal to special circumstances saves the theist from normative defeat, premise (1) can still ground an argument for epistemic defeat.
In response, the theist may try to deny (1), removing the threat of normative defeat and the threat of epistemic defeat at once. This can be done in three ways: by denying that religious evils are evils, by denying that religious evil is caused by religious beliefs, or by denying that the perpetrators of religious evil truly believe in God.
One can deny that religious evils are evils by claiming that victims of religious evil deserved to die. This is probably the line that past and present perpetrators and supporters of religious evil would take, if they cared about argumentation. Pope Urban II, Raymond of Aguilers, and the witch-hunters of Würzburg would probably say that pagans and/or witches deserved to die and their sufferings made the world a better place.
More reasonably, one can deny (1) by denying that faith in God causally contributed to religious evils. For example, one can argue that the witch hunts and the crusades were caused by psychological, social, and political factors that were unrelated to Christian faith, hence, even if references to God, Christ, and the Bible accompanied the evils in question, these speech acts did not causally contribute to the evils themselves but were mere epiphenomena. Let's call this the Epiphenomenalist Defence.
Alternatively, the theist can deny (1) on the grounds that the perpetrators of religious evil are not true believers. The theist can claim that even if the perpetrators take themselves to be theists and use sentences reminiscent of theism, they are wrong about the true content of the divine commands and they do not, in fact, serve God. The upshot of this objection is that some sort of quasi-religious beliefs are indeed involved in religious evil, but those beliefs are not genuinely theistic, because genuinely theistic beliefs cannot motivate or justify evil.Footnote 2 If this claim is true, then religion can only be said to cause evil if we use ‘religion’ in a somewhat misleading way. I'll call this the False Believers Defence.
Both the Epiphenomenalist Defence and the False Believers Defence entail that religious evil is not a truly religious phenomenon. On the Epiphenomenalist Defence, religious evil is only superficially associated with religion, because religious doctrines are not relevant for its causal aetiology. And on the False Believers Defence, religious evils are perpetrated by people who are not religious in the sense in which a real theist is religious. On both views, ‘religious evil’ turns out to be a misleading label, one that wrongly suggests that (genuine) religion is seriously implicated in religious evil.
The Epiphenomenalist Defence and the False Believer Defence appear to save the theist from normative and epistemic defeat, because they entail that the candidate defeater, premise (1), is false. But the atheologian can argue that the victory is illusory. Specifically, the atheologian can claim that both the Epiphenomenalist Defence and the False Believers Defence entail alternative epistemic defeaters. Let me elaborate.
The gist of the Epiphenomenalist Defence is that religious faith is not causally relevant for the genesis of (apparently) religious evils. A proponent of this defence will claim that the people who butchered the denizens of Jerusalem in 1099 believed in God but their faith was not among the causes of their atrocious behaviour, nor was faith in God in general the cause of the crusade itself. (And similarly for other (apparently) religious evils.) Note that the Epiphenomenalist Defence, if it is to stay true to its name, must entail that the perpetrators of religious evil are religious – their beliefs can hardly be causally inert without existing.Footnote 3 Hence, the proponent of the Epiphenomenalist Defence is committed to the following two claims:
The Epiphenomenalist Defence:
(7) Agents of religious evil believe in God.
but
(8) Belief in God is not among the causes of religious evil.
To see why the Epiphenomenalist Defence leads to trouble, note that (7) entails the following:
(9) Even though agents of religious evil believe in God, their faith does not stop them from perpetrating horrible evils.
Intuitively, (9) is an epistemic defeater of theism, because it casts doubt on the power of faith to motivate sinless behaviour. If belief in God cannot stop people from sinning against humanity in the worst possible ways, as facts of religious evil suggest, then the moral efficacy of the divine message seems weaker than one would expect it to be. For one would expect an omnipotent, morally perfect creator who wants to guide his children out of sin to make it sure that belief in Him produces morally positive results. Minimally, true belief should prevent one from becoming an agent of horrendous evil. And even if it is not metaphysically impossible for God to exist in a world where His name is misused in the worst imaginable ways by agents of horrendous evil, the atheologian can argue that the amount and severity of actual religious evils make the theistic hypothesis wholly improbable.
Since (9) follows directly from the Epiphenomenalist Defence, this defence does not save the theist from epistemic defeat. It just creates a new problem, the problem of explaining why God would create a world where belief in Him does not deter his creatures from sinning in the worst possible ways.
Let's see if the False Believer Defence fares better. The False Believer Defence is simply the negation of (7):
The False Believer Defence:
(10) Agents of religious evil do not believe in God. [=∼(7)]
If (10) is true, then (1) is obviously false – if agents of religious evil are not genuinely religious, then religion can hardly cause evil. Beliefs about God couldn't have motivated witch-hunters, crusaders, and their ilk if those people did not believe in God in the first place.
To make (10) fully defensible, the theist must tell us what agents of religious evil really believe. After all, they do use ‘believing in God’ (or its counterparts in other languages) to describe their doxastic practices. The theist can claim that agents of religious evil have false beliefs about the real God or she can claim that agents of religious evil have beliefs about a fictional god who approves of horrible crimes against humanity.
The False Believers Defence avoids the pitfalls of the Epiphenomenalist Defence, because, unlike the latter, it does not entail that (true) religious faith fails to motivate sinless behaviour. But the False Believers Defence leads to epistemic defeat nonetheless, although by a different route. For it is evident that agents of religious evil, whether they are truly religious or not, appear to be religious. They take part in rituals which resemble genuinely religious rituals to a great degree, they profess to follow a god whose will is said to be revealed by the same religious texts that truly religious people hold in reverence, and they justify their acts, including their horrible crimes against humanity, by reference to purportedly divine or divinely inspired principles that are also referenced by truly religious people. For example, witch-hunters justified their crimes on the grounds that witches renounced God (something that can be seen as sinful on biblical grounds), and Raymond of Aguilers justified the atrocities of the first crusade on the grounds that God pledged the Holy Land to believers. Similar examples could no doubt be multiplied indefinitely.
If the False Believers Defence is true, then it is possible to appear religious while being, in fact, viciously irreligious. Intuitively, this is just as much a problem for theism, since it means that religious texts and practices can be twisted out of their true context and deployed for horrible purposes in an apparently religious way, creating confusion about the true nature and purpose of religion. It is unclear why a morally perfect creator who wants to promote salvation through religion would permit such practices. It seems unreasonable to allow such a confusion about religion to develop if religion is indeed the gateway to God and heaven. Hence, the atheologian can press her claim about epistemic defeat even in the face of the False Believers Defence.
To sum up: religious evil appears to be an epistemic, and perhaps also normative, defeater of theism. If the theist accepts that religion causally contributes to horrendous evils, then, arguably, theism is both normatively and epistemically undermined, because theistic faith turns out to be both inadvisable from a moral viewpoint and incompatible with any reasonable providential strategy of a morally perfect creator. In response, the theist can claim that belief in God is causally inert in cases of religious evil, or she can deny that agents of religious evil truly believe in God. The first route, the Epiphenomenalist Defence, leads to epistemic defeat because it implies that God did not make religion a strong enough incentive against sin, which seems quite improbable under the presupposition that God is an omnipotent, morally perfect being who wants to offer salvation to mankind through religion. On the other hand, the False Believers Defence also leads to epistemic defeat, because it implies that God permits agents of horrendous evils to appear religious, creating confusion about the nature and function of religion. Such a state of affairs also seems incompatible with any sensible providential strategy of a morally perfect creator who wants to promote religiosity. Or so might the atheologian argue. And it seems to me that the atheologian has a fair point. Clearly, theists must give some explanation about religious evil, an explanation that goes beyond the easy fixes discussed in this section.
The distinctiveness of religious evil
I'll now argue that the problem of religious evil is distinct from the traditional problem of evil. I'll try to show this by relating religious evil to contemporary defences and theodicies. My claim will be that the latter cannot solve the problem of religious evil even if we presuppose that they are otherwise sound.
Throughout the discussion, I assume that the context of the dialectic is the evidential argument from evil, as formulated by William Rowe (Reference Rowe1979). Unlike atheologians who argue that the existence of evil is logically incompatible with the existence of God (e.g. Mackie Reference Mackie1955), the proponent of the evidential argument (‘the evidentialist’, as I'll call her) thinks that the amount and character of actual evils make the theistic hypothesis wholly improbable. The argument from religious evil, as it was outlined in the previous section, is clearly a version of this argument. It construes religious evil as evidence that the theistic hypothesis is untenable. Hence, it makes sense to restrict the scope of our discussion to contemporary responses to the evidential argument. Since current defences and theodicies typically target the evidential argument anyway, this restriction nonetheless allows us to situate the problem of religious evil in the current analytic controversy about evil.
The free will defence
Proponents of the analytic free will defence claim that it is logically possible for God and evil to coexist. The gist of the defence is that libertarian free will makes it logically impossible for God to ensure that his creatures never commit anything evil (Plantinga (Reference Plantinga1974), ch. 9).
Most philosophers seem to agree that the free will defence does not work against the evidential argument.Footnote 4 For even if the existence of evil is logically compatible with the existence of God, the evidentialist can still maintain that the amount and character of actual evils make the theistic hypothesis rationally untenable. The evidentialist can do this in two ways.
First, she can claim that even if God cannot foresee or determine the outcome of free creaturely choices, God nonetheless must have robust probabilistic knowledge about the future, and He can make sure that horrors and atrocities do not occur on a regular basis.Footnote 5 Alternatively, the evidentialist can argue that a morally perfect creator should curtail the scope of human freedom, because the ability to rape children, perpetrate genocide, etc. add nothing of value to creation. Even if freedom is a great good, the benefits of denying those particular abilities to humans clearly outweigh the costs. Or so the evidentialist can argue.
The free will defender can reply that the abilities in question make our freedom a truly substantial phenomenon, adding a crucial existential depth to human life.Footnote 6 And, perhaps, the free will defender can also press the claim that God cannot even have probabilistic knowledge about the future course of history in a world with a large number of free creatures. Let's pretend that this reply is, in general, acceptable. I'll now argue that it still fails to defuse the problem of religious evil.
Theism is predicated on the idea that God revealed himself and the moral law through religion. The free will defence complements this with the claim that we are free in a libertarian sense and therefore all sorts of evils can occur even in a world ruled by God. These two ideas add up to the thesis that our libertarian free will must be tempered by faith in God. However, as we saw in the previous section, religious evil is easily interpreted as a sign that religion cannot perform this regulative function, either because it can motivate sinful behaviour (if premise (1) is accepted), or because it cannot prevent sinful behaviour (if the Epiphenomenalist Defence is accepted), or because there is a strange ambiguity about the meaning of ‘religion’ (if the False Believers Defence is accepted). Even if we believe that a loving God could or would or must give us significant freedom, including the freedom to rape children and perpetrate genocide, religious evil still makes the theistic hypothesis very improbable, because facts of religious evil suggest that God expressed the objective moral law in such a way that obeying his commands can motivate, or can fail to prevent, or can seemingly justify, horrible evils. It is hard to see how to square this with the hypothesis that God wants to lead us out of sin through faith.
The free will defender might reply that significant freedom involves the freedom to ignore or misinterpret anything, including divine commands.
But this reply misunderstands the problem at hand. The task is to explain why God would create a world where it appears as if religion were not an obvious antidote to sin. It is hard to see why God could not have created a world where humans cannot misuse religion in all sorts of horrible ways. In a world where religious language cannot be misused (either because of divine intervention or because of other precautions), and, as a result, religious congregations are guiltless, creaturely freedom is not seriously restricted, and the relative loss of freedom seems to be balanced by the fact that those who want to lead a virtuous life have a clear sense of what to do. If salvation is achieved through religion, as traditional theism implies, then God can only allow religion to be implicated in evil at the price of making it ambiguous what the right way to salvation is. Hence, the providential role of religious evil is not explained by libertarian freedom. Religious evil makes the theistic hypothesis improbable even if we presuppose that the free will defence is otherwise sound.
Spiritual chemotherapy
Some theist philosophers, most notably Eleanore Stump (Reference Stump1985; Reference Stump1997), argue that we all suffer from ‘a cancer of the soul’ because we are naturally predisposed to sin, and the providential function of evil is to cure this terrible disease. Just as a loving mother is not blameworthy for putting her child through the ordeal of chemotherapy if the child has leukaemia, likewise God is not blameworthy for curing our inborn sinfulness through various ordeals.
Regardless of its general merits, this theory is unable to solve the problem of religious evil. For the central message of this theory is that we must regain contact with God through atonement and faith, but religious evil calls into question the very power of religion to put us in contact with God. Indeed, religious evil seems to show that people can become the worst kinds of sinners through faith. If evil is a sort of spiritual chemotherapy necessitated by our inherent sinfulness, why does God add to our troubles by making religion implicated in the very problem that religion is meant to solve?
Punitive theodicies
Punitive theodicies portray evil as the consequence of sinful acts that separated mankind from God (see e.g. van Inwagen Reference van Inwagen1988; Reference van Inwagen2006). Proponents of this view claim that humans broke away from God at some point in the distant past, unleashing the series of catastrophes known as world history.
Unlike Stump's theodicy, which portrays evil as a bitter cure for a spiritual disease, punitive theodicies portray evil as the punishment for a collective crime. Regardless of the general merits of this idea, it fails to explain religious evil. If God punishes us with evil because we broke away from Him, then, presumably, it is an important part of the divine plan to indicate the way back. According to theists, the way back is through religion. But religious evil is prima facie evidence that religion causes even more sin. This strange feature of our world is not explained by a hypothetical original separation. The punitive theodicist owes us an explanation why a loving God who wants to steer humanity back to Himself through religion would allow the institution of religion to be compromised in this way.
Soul-building
Soul-building theodicies are predicated on the idea that disease, predation, war, and other evils contribute to the development of human virtues because they create opportunities for extraordinary bravery and sacrifice and hence for the positive exercise of our freedom. Evil, on this view, is an educational tool that a morally perfect creator is fully justified to wield.Footnote 7
The soul-building theory implies the following answer to the problem of religious evil: religious evil is yet another character-building test that God has thrown in our way to make us more virtuous. Theists must sort out the right interpretation of the divine commands from the wrong ones. The goal of this exercise is to give us a chance to become vigilant religious experts.
Regardless of the general merits of the soul-building theory, this particular explanation is unconvincing. Suppose that Kurt is a benevolent demon who wants to hone our skills in mathematical logic. Kurt whispers logical puzzles into the ears of various people, and he makes sure that those who solve the puzzles hear new ones. It seems prima facie obvious that Kurt's educational strategy is severely hampered if many people go mad because of his visitations. This would support the hypothesis that Kurt is evil or inept, or that he is merely hallucinated by mentally unstable people. But, given religious evil, God is in a similar situation if the soul-making theodicy is true.
Religious evil does not only put God's creatures through terrible ordeals, it also creates confusion about the nature and value of religion. If religious evil serves the purpose of soul-building, it can hardly serve the purpose of building religious souls. The purpose of building religious souls would be much better served by a world where religion is a safe haven of virtue, not a potential source of further corruption. Hence, even if one could show that a loving God is likely to place his creatures in a moral boot camp, the hypothesis that we are in such a boot camp fails to explain religious evil. Religious evil does not seem to be a sensible component of a divine providential strategy whose ultimate goal is to bring creatures into loving union with God through religion.
Second-person accounts
Some philosophers argue that the problem of evil cannot be solved by looking for God's instrumental reasons for permitting evil, either because there are no such reasons or because looking for them already misses the point. According to these philosophers, the proper theistic response to evil is to emphasize that God has the power to, and inevitably will, set everything right in the next life. Until then, we must accept evil as a painful but ultimately cleansing spiritual exercise, perhaps one that allows us to enter into a mystical identification with Christ. (For versions of the second-person account, see Allen (Reference Allen1980), McCord Adams (Reference McCord Adams, Adams and McCord Adams1990; Reference McCord Adams1999), and Stump (Reference Stump2001).)
Second-person accounts do not give a satisfactory answer to the problem of religious evil, even if we presuppose that they give a satisfactory answer to the problem of other types of evil. For religious evil calls into question the healing and redemptive power of religion itself, undercutting the basic message of second-person accounts. If religion itself is implicated in evil, then turning to religion to counter evil is not a prima facie reasonable strategy.
The dialectic so far
I have argued that current defences and theodicies fail to solve the problem of religious evil even if we presuppose that they are otherwise successful. The reason why religious evil defies current defences and theodicies is that the latter place evil in a providential context where the ultimate goal is to make creatures more virtuous (or, at any rate, less sinful) through faith. But religious evil calls into question the moral efficacy of faith and the power of the divine message to motivate sinless behaviour. As a result, religious evil threatens the coherence of any hypothetical providential strategy the point of which is to offer salvation to mankind through religion. Generally, the problem of religious evil undermines any defence and theodicy that presupposes this principle. It seems safe to conclude, therefore, that religious evil is a distinctive type of evil, one that requires a special solution.
A note on sceptical theism
The foregoing dialectic did not address sceptical theism, which is perhaps the most discussed theistic response to the evidential argument today. According to sceptical theists, we have no reason to think that our cognitive capacities allow us to understand the providential role of evil (supposing it has one), hence we have no reason to think that evil is a defeater of theism.Footnote 8 One could argue that sceptical theism solves the problem of religious evil, contradicting my claim that the latter requires some special solution.
To the extent that sceptical theism is indeed a solution to the problem of evil in general, and not a cop-out (as some might argue), it is certainly a solution to the problem of religious evil as well. But I doubt that this makes the two problems very similar, because religious evil calls into question the providential efficacy of religion. This is clearly a special problem about the tenability of theism, over and above the problem of human afflictions in general.
Moreover, one can argue that the specificity of religious evil shows up in the context of sceptical theism as well. The practical message of sceptical theism seems to be that we should persevere in our faith instead of trying to understand God's unfathomable reasons for permitting evil. But religious evil is an impediment to persevering in one's faith, because it calls into question the moral value of faith. Many agents of religious evil persevere in their faith; indeed, it is precisely their religious fervour that fuels them with emotional energy to commit horrible sins. Hence, the practical message of sceptical theism sounds wrong as far as religious evil is concerned. And this, in turn, indicates that the problem of religious evil is different from the problem of other types of evil, even in the context of sceptical theism.
A partial solution
To round off the dialectic, I'd like to suggest a solution to the problem of religious evil. But before I launch my argument, it is important to make it clear what a solution is meant to achieve. There are different types of possible solution, some stronger than others.
Giving a solution to the problem of religious evil means explaining why religious evil is probable (or, at any rate, not improbable) on the hypothesis that God exists. Ideally, the explanation should help us understand God's reasons for permitting religious evil. But there are, in fact, two possible explanatory goals here, which must be set apart.
When we set out to explain why God permitted the witch-burnings and the crusades, we are facing two distinct explanatory tasks, corresponding to the following two questions:
(11) Why did God permit the suffering of ‘witches’ and ‘pagans’?
(12) Why did God permit ‘witches’ and ‘pagans’ to be tortured and murdered in His name?
(11) is part of the general problem of evil. It is a question about the providential function of atrocities. In contrast, (12) is a question about the providential function of a religious or quasi-religious gloss on certain atrocities. In a slogan, (11) is a question about the providential function of religious evil qua evil, whereas (12) is a question about the providential function of religious evil qua religious evil.
Accordingly, there are at least three kinds of (non-sceptical) solution to the problem of religious evil: Solutions that answer (11), explaining religious evil qua evil (type 1 solutions); solutions that answer (12), explaining religious evil qua religious evil (type 2), and solutions that answer both (11) and (12), fulfilling all relevant explanatory goals (type 3).
It seems safe to conjecture that type 1 solutions also solve the traditional problem of evil. If we manage to explain why God permitted the mass murder of innocent ‘witches’ and ‘pagans’, we are bound to have come up with good ideas about the providential role of mass murder in general, since there is nothing special about the witch hunts and the crusades qua mass murders.
In the present context, only type 2 solutions (=answers to (12) only) and type 3 solutions (=answers to both (11) and (12) ) are relevant, because these are the solutions that address the religious aspect of religious evil, which is the topic of this article. Hence, type 1 solutions can be set aside. We can then distinguish partial (=type 2) and full (=type 3) solutions to the problem of religious evil (or, more precisely, to the problem of religious evil qua religious evil, but I'll drop this qualification to reduce clutter).
My goal will be to construct a partial solution to the problem of religious evil. Giving a full solution seems to be an unduly difficult task, since, in effect, it requires us to solve the problem of (moral) evil in general. Note, however, that once we have an acceptable partial solution, we can combine it with any logically compatible defence or theodicy to get a full solution. And since existing defences and theodicies were shown to have very little to do with religious evil qua religious evil, it seems safe to speculate that an acceptable partial solution can be conjoined with any existing defence or theodicy.
Let ‘@’ denote our world, and suppose that God contemplates which of the following two worlds, if any, is worthy of being actual:
WA (‘the world of atheist sinners’)
A world where the distribution of pain, suffering, and cruelty is the same as in @, but all sinners are atheists.
WM (‘the world of mixed sinners’)
Same as WA except that the set of sinners spans the whole spectrum of actual creeds.
In WA, all sinners are atheists, but in WM, some sinners claim to be theists, and otherwise everything is the same as in WA. For simplicity, we may assume that both WA and WM are very similar to @. (Perhaps WM is @.)
To stay neutral on problems about free will, let's say that God is trying to decide which one of WA or WM should be in the set S of worlds that are candidate-actual in the sense that once God's providential decisions are settled, any further facts about which member of S will turn out to be actual will depend on free creaturely acts. Since we are looking for a partial solution to the problem of religious evil, we may take God to have decided that S includes worlds with much suffering. God is now contemplating whether He should allow any of the candidate-actual evils to have a religious gloss on them.
For the sake of illustration and without loss of generality, suppose that (a counterpart of) Raymond of Aguilers exists in WA. (Raymond of Aguilers, remember, is the French chaplain who justified the first crusade by reference to Christ.) And suppose that (a different counterpart of) Raymond of Aguilers exists in WM. And let's say that the only difference between WA and WM, as far as Raymond is concerned, is the following:
(13) In WA, Raymond of Aguilers is an atheist who justifies the (relevant counterpart of) the siege of Jerusalem by saying that might is right.
(14) In WM, Raymond of Aguilers professes to be a Christian and justifies the (relevant counterpart of) the siege of Jerusalem by saying that Christ was on their side.
Let's compare the moral scorecard of Raymond of Aguilers in WA and WM. It seems clear, first of all, that Raymond of Aguilers is wrong to try to justify the siege of Jerusalem in any way, since the siege is an evil. This fault appears on his scorecard in both worlds. Moreover, by (13), Raymond makes a mistake about the moral laws in WA, because he claims that might is right. And by (14), Raymond makes a mistake about religious principles in WM, because he claims that Christ assists mass murderers. Since we're assuming that these are all the morally relevant facts, it follows that WM is not worse than WA in terms of Raymond's moral scorecard iff making a mistake about the secular moral laws (=moral laws that are not explicitly related to God or to religious practice) is at least as bad, objectively, as making a mistake about religious principles.
Let's look at another example. Suppose that Günther is a witch hunter of Würzburg. In WA, Günther is an atheist who kills people on the grounds that they are genetically inferior, while in WM, Günther claims to be a Christian and kills people on the grounds that they refuse the sacrament. Günther commits the same basic sin in both worlds: he kills innocent people and he tries to portray this evil as morally right. But he makes different mistakes about moral justification in the two worlds. In WA, he claims that people with a different genetic make-up deserve death, and in WM, he claims that people who refuse the sacrament deserve death. Günther makes a mistake about the secular moral laws in WA, and he makes a mistake about religious principles in WM. Suppose, again, that this is the only relevant difference between WA and WM, as far as Günther is concerned. It follows that WM is not worse than WA, in terms of Günther's moral scorecard, iff making a mistake about the secular moral laws is at least as bad as making a mistake about religious principles.
These examples can be generalized to all evils in WA that have a religious gloss in WM, yielding the following preliminary conclusion: a world with a religious gloss on some evils is not worse, from a divine standpoint, than an otherwise similar world without such a gloss iff making a mistake about the secular moral laws is at least as bad, objectively, as making a mistake about religious principles. And a world with a religious gloss on some evils is better than an otherwise similar world without such a gloss iff making a mistake about the secular moral laws is worse than making a mistake about religious principles. This might be the case if building and maintaining a secular moral order is a more important human duty than mastering the interpretation of religious tenets. Let me sum up this last idea in the slogan that Secular Morality Trumps Religiosity (the SMTR thesis). Secular morality trumps religiosity iff it is more important for us to master the secular moral laws (=moral laws that are not explicitly related to God or to religious practice) than to master religious principles. If this thesis is true, then God might have sufficient reason to allow WM instead of WA to be candidate-actual, and, generally, to prefer worlds with religious evil to otherwise similar worlds without religious evil.
One can object to this defence along the following lines: suppose we accept that, by the SMTR thesis, perpetrators of religious evil are less blameworthy in WM than in WA as far as mistakes in moral justification are concerned. We might nonetheless claim that WM is much worse, overall, than WA. This might be the case, for example, if evils like the witch hunts and the crusades are caused by religious beliefs in WM. Such adverse effects of religion might make WM much worse from a divine standpoint.
I propose to meet this objection by combining the SMTR thesis with the False Believers Defence, i.e. with the claim that agents of religious evil are not truly religious. This means that religion does not cause evil in worlds like WM.
However, as we saw, the False Believers Defence has problems of its own. Specifically, it entails that God allowed the nature and function of religion to seem ambiguous because of the presence of apparently religious evil practices. So the combination of the SMTR thesis and the False Believers Defence is still vulnerable to the evidentialist challenge.
Further, it might be argued that religious evils can have spill-over effects that make a world like WM worse than WA. For even if genuinely religious beliefs do not cause any evil, it might be the case that the religious gloss on certain evils deters some people from becoming religious. And if religion is the way to salvation, then this might entail that fewer people are saved. Religious evil may have ramifications that lower the number of people who go to heaven. If this is true, then, again, worlds like WM might come out as worse than worlds like WA from a providential standpoint.
Both of these objections can be deflected by claiming that (i) God wants to guide us out of sin primarily through morality, not through religion, and (ii) salvation depends on secular moral virtues most of all, not on one's degree of religiosity. I'll call the conjunction of (i) and (ii) ‘the sola virtute principle’. The gist of this principle is that salvation depends primarily on morality, not on religion; the soteriological role of religion is to foster secular morality. (This principle is not meant to entail that humans can save themselves by being moral. The sola virtute principle is consistent with the idea that salvation only happens through grace. The principle only claims that secular moral behaviour trumps facts about religiosity in terms of God's decisions about salvation.)
The sola virtute principle defuses both of the last two objections. The first objection was that a morally perfect creator who wants to guide us out of sin through religion is unlikely to allow religion to be compromised by religious evil. The sola virtute principle defuses this objection, since the principle entails that God wants to guide us out of sin through morality, not through religion, so religious sinners do not compromise God's strategy any more than atheist sinners do.
The second objection was that religious evil might deter some people from becoming religious, lowering the number of people who are saved. The sola virtute principle defuses this objection as well, because the principle entails that failing to become religious does not, in itself, disqualify anyone from heaven, so the fact that religious evil might deter some people from becoming religious does not threaten God's efforts, ceteris paribus, to save as many as possible. The objection would only work if one could show that the religious gloss on religious evils makes people who would otherwise be moral immoral. But it is hard to see how the atheologian could justify this claim.
The atheologian can retort by saying that God is unlikely to permit religion to be implicated in evil, no matter what. Religious evil defiles God's name and commands. Even if this does not directly hurt the divine plans, it is a useless and unbecoming feature of creation.
In response, a proponent of the present defence can say that God might nonetheless have sufficient reason to actualize a world like WM instead of a world like WA if God wants to make it clear that praising Him is less important than building and maintaining a strong secular moral order. A world like WA fails to underscore this principle, since all sinners are atheists in WA, which lends inductive support to the hypothesis that being a good person is the same as being a religious person. In contrast, religiosity is not coextensive with morality in WM. So the history of WA, unlike the history of WM, is inductively deceptive if secular morality is above religiosity on the list of human duties.
To sum up, my proposed partial solution to the problem of religious evil consists of three components: the thesis that secular morality trumps religiosity, the False Believers Defence, and the sola virtute principle.
It might be objected that a sane theist cannot accept this package, for two reasons. First, these ideas are objectionable on scriptural grounds. The Bible clearly advocates faith in God as one of the foremost duties of those who want salvation. Second, accepting the theory in question may seem like ideological suicide for theists. For the theory may seem to imply that religious belief has very little to do with salvation, undercutting one's reasons for being religious. Let me tackle these complaints in reverse order.
The partial solution I'm proposing indeed entails that religious belief is not immediately but only mediately relevant to salvation, through its effects on secular morality. But this does not undercut theists’ reasons for being religious, because it does not follow that theists do not need religion to be moral. Neither does it follow that religion is not needed to foster secular morality in general. The solution I'm proposing can only be objectionable to theists if they think that atheists cannot be saved and/or that secular morality is of secondary or no importance from a truly religious viewpoint. I find it unreasonable for theists to believe such things.
As for the claim that faith is indispensable for salvation on biblical grounds, I reply that this principle must be restricted to those who need religion to be moral. A proponent of the present partial solution will say that such commands are only valid in communities where the Bible is treated as a sacred text. This idea can be upheld without vicious relativism, for one can consistently maintain that belief in God is not an objective duty for all yet it is an objective duty for some, namely for those who need religion to be moral. Moreover, the theist who accepts my proposed solution can add that even though faith in God is supererogatory for some people, it is nonetheless a great virtue which is recommended for everyone.
It might be argued that a sane theist cannot relativize her beliefs in this way. And perhaps my proposed solution is objectionable for further reasons as well. If the objections are decisive, then the proposed partial solution is unsuccessful, and theists must look for some other defence or accept defeat.
Summary
I have argued that evils committed in the name of God (‘religious evils’) generate a special version of the problem of evil. I drew on historical examples to clarify the concept of religious evil, then I argued that religious evil is an epistemic and/or normative defeater of theism. I tried to show that the problem of religious evil cannot be solved by any defence or theodicy currently on offer, and, finally, I constructed a partial solution. My partial solution combined three ideas: (i) secular morality trumps religiosity, (ii) perpetrators of religious evil are not genuinely religious, and (iii) salvation depends primarily on (secular) morality, not on religiosity.
Although I assumed a Christian context throughout the article, my claims are not meant to be restricted to Christianity or monotheism. Any religious system that does not portray its ruling deities or principles as advocating cruelty and violence is faced with the problem of religious evil if the system in question has been used to justify evil acts. And similar partial defences are presumably constructible, mutatis mutandis, for such systems as well.Footnote 9
