Defenses and Theodicies

A theodicy is an explanation of the reasons that can be ascribed to an all-good, all-knowing, all-powerful God to allow evil in the world. Perhaps the most clear example of a modern theodicy was the first edition of John Hick’s Evil and the God of Love.Footnote ⁴ In the revised edition, Hick modified his preference for an Irenaean approach over an Augustinian approach to the problem of evil to try to render it as a form of “defense” rather than as a “theodicy.” This camouflage, alas, was not entirely successful. His position was a form of an “Irenaean,” “soul-making,” view that argued that suffering was or could be good for humans because it was a necessity for them to develop “higher” epistemic, moral, and theological virtues; for instance, one could not develop virtues such as forgiveness unless there were sins (evils done to one) to forgive.

As Marilyn McCord Adams successfully argued, however, horrendous evils devastate both Irenaean and Augustinian forms of theodicy.Footnote ⁵ For example, the early victims of acquired immune deficiency syndrome (as an Episcopal priest, Adams ministered to communities of AIDS victims) suffered horrendous evils. Whatever “good” might come out of that suffering and those many deaths (so far, a treatment, not a cure, for people who are HIV positive is a general benefit resulting from the plague) was not a higher good that somehow paid off those who suffered and died. Christians, McCord Adams cogently argued, could not reasonably view the horrendous evils of the AIDS epidemic as willed by God or permitted by God in a world God created for some good purpose. Horrendous evils could not be reasonably believed to be good for all the souls who suffered from soul-destroying evils (such as, for example, the Shoah, child abuse, spousal abuse). Horrendous evils overwhelm theodicies.

A defense, on the other hand, is a logical exercise that shows the compatibility of two propositions:

The primary exemplar of a defense is the Free Will Defense (FWD) formulated by Alvin Plantinga.Footnote ⁶ As far as I know, the logic of this defense has not been defeated, though challenged often.

A general, undefeated form of this defense can run roughly as follows:

1. To show the compatibility of A and B, one needs to discover possibly true (not actually true or believed to be actually true) proposition(s) that are compatible with both A and B,Footnote ⁷ such as:

Because A, C, D, entail B, and are all possibly true, then A and B are not contradictory. One can hold both A and B without contradiction.

2. The most debated particular form of defense is Plantinga’s FWD. John L. Mackie, in his argument against the rationality of believing in God as defined in A, assumed that God could actualize a possible world in which all free beings always choose to do the good and that either A or B must be false because they are not compatible.Footnote ⁸

3. Plantinga argues that the proposition that “God could make a world in which all free beings choose the good” is incoherent. His argument can be roughly summarized as follows:

4. Thus, Mackie’s argument hinges on an assumption that cannot be true and so fails. Mackie’s assumption that God could create a world in which all free beings chose the good is incoherent. God could make beings free (but then could not determine their choices of good or evil) or God could make beings good (but then they are not free to choose evil). God cannot make free beings always do the good.

In effect, this FWD is a sufficient defeater of all positions that accept even a minimal form of libertarian free will (an argument that has a compatibilist view of free will would require a different form of argument). If the FWD is correct, then the logical problem of evil is defeated on any and every non-deterministic metaphysical account. The structure and consequences of this argument have often been misconstrued.

An analogous proposition such as E, “Natural evils (e.g., earthquakes, some physical suffering of sentient beings, etc.) are brought about by the activity of free, non-human beings” can cover natural evil as well. This diabolical possibility has received far more attention than it has deserved. The defense needs this proposition to be only logically possible—not true. With A, C, and D, E entails B. So, the defense, as outré as one of its components may be, stands.

Defenses are typically supported by elaborate analysis and persuasive examples and narratives. They may be as elaborate as theodicies. The key that distinguishes defenses is that they are rebuttals of attacks, not theodicies, explanations, or justifications of a set of beliefs. Logically, one who asserts a defense does not have to believe that each proposition used in building the bulwark is true; some need only be logically possible. One who constructs a theodicy uses propositions that the author finds are as true as possible—given that human beliefs about God cannot exhaustively comprehend God.

What are the limits of such defenses? First, the FWD does not defeat the empirical problem of evil for theists. As William L. Rowe has preeminently argued, the reality of genuine natural evils in the actual world is another problem.Footnote ⁹ If theists continue to believe in God in the face of genuine natural and moral (including any structural and cultural) evils, evidence for their belief is questionable, at best. The counterargument to Rowe and others is that belief in God does not have the status of a hypothesis that stands in need of evidence, but of a conviction that is accepted unless the evidence against it is overwhelming enough to defeat it for a person of faith who holds it. The believer is not arguing to support a hypothetical belief. As I put it in the original article:

The anti-theistic argument is fully successful only if it involves a person’s conversion away from faith in God. The locus of the empirical problem is not so much logic as it is the ethics of belief.Footnote ¹¹ The empirical problem of evil is compounded when one considers social evils. The evidence for horrendous evils is staggering. Rowe’s challenge has to do with natural evils. Yet natural evils may be horrendous insofar as they have social and personal components. A hurricane may well be a natural evil, but building a city in “hurricane alley” involves social and personal choices. McCord Adams’s work, discussed in the following, reflects the challenges social evils bring to Christian faith and theology and argues for placing them in a redemptive, christological framework, not in a theodicy.

Second, the FWD does not require its holder to believe that all natural evils are caused by devils or space aliens or other nonhuman entities. It does require the holder to recognize the logical possibility that devils caused nonmoral evils, but not to accept the proposition that they actually or even plausibly caused such evils. I do not know if Plantinga or other defenders believe E (I surely do not), but I do not see a way to show this proposition cannot be logically possible.Footnote ¹² Hence, it suffices to support the FWD.

Third, I would now add that a defense is a “game” played on the opponents’ “home turf.” This may not have been clear to readers of my earlier article. A defense accepts the atheists’ propositions as given only for the sake of argument. The FWD does not commit one to a position that accepts the opponents’ view of God simpliciter. The FWD is simply a defense against a specific and common modern atheistic argument against belief in God: it shows that even if one accepts the opponents’ view of God, the opponents’ attempts to defeat the reasonableness of belief in God in the face of the reality of evils fail.

One can accept the FWD as a defense against the all-too-commonly accepted argument against an assumption that God is an agent who acts directly in the actual world (like a secondary cause) without accepting the understanding of divine action that it assumes. Accepting a defense against such a view can make room for other accounts. A Thomistic account, roughly for an understanding of God as a primary agent who cannot be held responsible for evil, but who creates secondary agents who are responsible for bringing about moral and social evils, neither plays on the opponents’ turf nor endorses their view of God. Brian Davies, for example, holds a privative view of evil, that is, that evil is the absence of some good that should be there and is not, not the existence of some “thing” that is there. Hence, God cannot be considered the cause of any evil “thing.” Created beings may bring about evils, acting for something they desire, but God cannot and does not have desires. God does not create out of desire or need, but out of practically unimaginable love.Footnote ¹³

Davies explicitly rejects the FWD. He seems to assume that its rhetorical force requires the acceptance of a God who is an agent in the universe (evidently Mackie and Plantinga would accept this understanding of divine power). But such acceptance is only for the sake of argument undertaken on the opponents’ home turf. To accept it as a defense does not preclude an argument that does not accept the attackers’ and the defenders’ assumptions about God and evil. This use of a defense merely clears out the thicket of popular assumptions about divine power that Thomists do not accept. Someone like Davies could accept a defense like the FWD, but then change the subject by reconceiving how we image God.

Fourth, the FWD, more generally, does not commit one to either a particular explanation or justification of “God’s ways.” One might accept a form of “open theism”; or one of a variety of views found in “skeptical theism” in which we simply do not know what God’s reason might be for permitting evils, even horrendous evils, in the world;Footnote ¹⁴ or a “process view” in which God is not immutable; or a “Thomistic view” in which God is not a secondary agent who acts in competition with created agents; or a form of accepting the mystery of evil as a theological aporia that evokes dreadful contemplation and the exercise of a “harsh and dreadful love.”

What do defenses do? I have claimed that they defeat anti-theodicists’ logical attacks. But they also provide a way into helping someone to recognize just where the mystery lies in trying to account for God and evil. John Thiel, for example, finds that “God’s causal relation to evil is a[n] … aporia incapable of theological resolution.”Footnote ¹⁵ The possibly true claims defenses use are not explanations, but merely philosophical rebuttals. A defense makes room for a richer response: not a theodicy, but a recognition of where the mystery is or what divine love consists in. Indeed, Kilby’s refusal of an explanation of God’s relation to evil and her own explication of a profound understanding of theology’s task as gesturing to mystery is not a practical response to evil as Surin might endorse, but a profoundly insightful theological move.Footnote ¹⁶

On the other hand, a theodicy is an explanation of why God allows actual evils in the real world God created. In constructing a theodicy, the believer, “tries … to tell us what God’s reason is for permitting evil. At bottom, [the believer] says, it’s that God can create a more perfect universe by permitting evil.”Footnote ¹⁷ The purpose of a theodicy is not to show that the atheologian’s attempted rebuttal is unsuccessful. Rather, the believer seeks to show that the warrants for her or his own conclusions are reliable. Here the believer must make the argument. The burden of proof is no longer on the atheologian, but on the believer. Whereas defenses show that the burden of proof is on the atheist (and is unmet), the theodicist bears the burden of proof.Footnote ¹⁸

The differences between the two forms of argument are significant. First, the theodicist must account for all of the evils, or at least the forms of evil, in the actual world. The former seems impossible as new evils happen daily. The evolution of social media has brought some new forms of good, but also new forms of evil such as the widespread, nearly unfettered, distribution of disinformation.

Second, to accept a theodicy, a believer must accept the propositions that structure the argument. If any of the propositions are false and irreparable, the argument fails. If any of the propositions are implausible or inadequate, a reader of the argument should be skeptical, at best, of its cogency. Some have tried to suggest that a theodicy is a story more than an argument. But this move, as appealing as it might be, does not make telling an incoherent story good or valid. An incoherent story is no better than a fallacious argument. The believer must accept the whole story or argument of a theodicy to be reasonable. To be felicitous, that acceptance can be properly supported only if the story or argument is actually true or a reasonable approximation of the truth. If the hypothesis is seriously flawed, then it cannot support that conviction reasonably.

Third, theodicy as a discourse practice presumes that the theodicist accepts the propositions that constitute the theodicy as true or plausible. The theodicist is arguing for a way of understanding how we can explain why God permits evils, even horrendous ones. This is a consequence of the differing burdens of proof involved in defenses as distinguished from theodicies.

Fourth, I would now add that a theodicy is constructed on the theodicist’s “home turf.” Theologians who construct theodicies presume that there is a God, whether this presumption is based on an argument such as the cosmological or ontological arguments for God’s existence, or has some other basis, or is a “properly basic belief” that is properly groundless. Anti-theistic attackers do not presume God exists. This aspect of the distinction between defenses and theodicies may not have been clear to readers of my earlier article. A theodicist does not accept an anti-theistic view for the sake of argument. The theodicist rejects the argument. The empirical requirement for a theodicy is that the evils must be explained sufficiently for the theodicy to be plausible or persuasive. The argument is not against attempts to defeat the reasonableness of belief in God, but for the reasonableness of belief in God. The burdens of proof differ.

In sum, defenses and theodicies are different sorts of arguments, undertaken in different contexts, with differing rhetorical purposes, and with differing intended main audiences. Both theodicies and defenses can, and often do, suffer from a level of abstraction that often makes them seem implausible but does not make them similar. The failure to recognize the difference plagues many more recent theological approaches to God and evil.

Kilby’s Anti-Theodicy Position

I have argued that doing theodicies creates evils, precisely in the sense noted by Karen Kilby: “Theodicies are destructive discourses which efface and perpetuate evils.”Footnote ¹⁹ Unfortunately, Kilby takes my position in The Evils of Theodicy as “most aggressive” rejecting “everything to do with theodicy.” She does not distinguish theodicies from defenses clearly, as I do, particularly in my supporting a prototype of defense found in Augustine’s authoritative instruction as a bishop to the layman Laurentius, Enchiridion.Footnote ²⁰ She is more sympathetic to the anti-theodicy position of Kenneth Surin, Theology and the Problem of Evil,Footnote ²¹ which favors “practical theodicy” (a position that can be defended if it is attacked on logical grounds, I might add). She finds me supporting abandoning “the issue as an abstract and general one, in short, and concentrate on what can be done in relation to concrete evils.”Footnote ²² I do think that theodicies should be abandoned, but not defenses, even abstract ones (given their purpose). I do think that the appropriate response to evils is to do what can be done for victims of hurricanes and earthquakes, and all forms of sin (whether individual acts, cultural patterns, or structural laws, rules, practices or presuppositions). Kilby’s position is flawed because she does not recognize that a defense is not a theodicy.Footnote ²³ To reject theodicies is not to reject defenses.

Theodicies, I have argued in a number of places,Footnote ²⁴ also tend to fail to discriminate between personal and social (whether cultural or structural) evils. In effect, they render social evils as practically natural, as if all structural and some cultural arrangements are not to be understood in moral terms. Kilby also seems to think this imputes a moral failure to theodicists. She writes:

Later, Kilby is explicit: “What critics like Surin and above all Tilley do is to push the idea that such intellectual failing has a moral dimension… . [Theodicies] are bad for us and a bad thing altogether.”Footnote ²⁶

This point is also not quite accurate. It confuses the communicative action of writing a theodicy with the readers’ acting in response to reading them. First, those theodicists, who simply do not recognize cultural and structural evils as types of evils distinct from personal sin or a nexus of sins, cannot “try to reconcile us to evils” because they do not see them. You cannot try to reconcile me to a phenomenon you do not see! Perhaps some so-called “natural evils,” like Rowe’s fawn dying miserably alone in the forest, are just natural evils, but most “natural” evils have social and personal components. Hurricanes are natural. Building cities in their paths is not. Second, if there is a moral failure, it is the readers’: those who accept a theodicy blindly and act without consideration of the practical need to work against evils perhaps “sin” by omission.

Theodicists are not immoral or depraved but deprived of (for various reasons) the ability to recognize social evils. If writing theodicies produces evil, it is primarily a cognitive evil. Readers may or may not be blinded to recognizing evils or may not be led to people being “reconciled” to evils or to a caricature of Julian of Norwich’s famous line, “All shall be well, and all manner of thing shall be well.” But these are readers’ responses to an intellectually problematic discourse. In sum, Kilby’s invocation of the immorality of theodicy seems too broad: it seems to conflate theory and praxis. She thus misplaces the locus of any moral evil of theodicy.Footnote ²⁷ Theodicies may be a bad thing, but they may not be altogether bad.

Nonetheless, Kilby’s own constructive work is truly impressive in that she relocates the theological aporia. She uses an analogy with mathematics: “Mathematics can provide us, in other words, with an example of clarity and mystery varying in direct, rather than inverse proportions.”Footnote ²⁸ As our understanding of the mathematics of astrophysics deepens, the sense of awe does not diminish, but the mystery understood in and through the aporias that point to that mystery intensifies. The same is true of theological exploration: our sense of mystery deepens.

Reflecting on the work of Denys Turner, Katherine Tanner, and St. Thomas Aquinas, she distances her work from theodicies or defenses that construe divine and human agency as competitive. She concludes: “To say that divine and created agency cannot be contrasted [and thus not competitive] is not to offer an explanation of how they go together, but a rule which defeats all possible explanations.”Footnote ²⁹ Such a rule is a “stop card” marking the danger of theodicy. She finds that theodicies seek to resolve the insoluble, effectively blocking the path to contemplating the beauty of divine simplicity.

Kilby appreciates the focus and aim (though not necessarily all the content) of Karl Rahner’s approach: What Rahner “is aiming to do is not to clear up mysteries that are around in Catholic theology … but to intensify the sense of the mysterious.”Footnote ³⁰ She seems to accept the purposes of philosophical defenses and to reject the purposes of theodicies. She writes:

Yes, but defenses do not diminish the mystery of evil as they are fundamentally dismissals of relatively superficial understandings of the mystery.

Kilby recognizes the challenges posed by social evils.Footnote ³² She also finds that personal and structural evil cannot be distinguished clearly. Hence, she remains cautious about exploring the problem. She focuses on structural evils that entangle our actions with evil. In capitalism, for example, we seem to be unable to avoid such entanglement—unless, of course, we give up any entanglement with the capitalist system. She does not, however, discuss cultural evils such as the pervasive attitudes of racism, sexism, and others that damage people regardless of the systems in which they are entangled. More work remains to be done in this area.

One way forward would be to show that there are approaches to social evils that can and should be deconstructed. Construing social evils, for example, as simply an agglomeration of personal sins without considering the cultural and social contexts that incline people to sin or entangle them in sin seems a missed opportunity, at least so far. It seems to me that an ideology of individualism lies behind many of the refusals to name social evils. Such reduction of social evils to personal sins or natural evil neglects the power of culture and system. It also seems that corporate actions (for example, the civil rights movements and the US farm workers movement, which are not merely negative protests, but movements for ameliorating or extirpating extant social evils in favor of a “better way”) may be possibly important in imagining and understanding which cultural components and social structures we ought prefer. The issue is not a general one, but particular challenges about particular, concrete cultural phenomena and social structures: Is the world better or worse if we support free trade, for example? Is it an unsupportable social structure because it is thoroughly evil or is it a structure productive of some evils, or is it an optimal structure, given the circumstances? The real issues are practical and retail, not intellectual and wholesale.

Nonetheless, I find her constructive work illuminating even as I am not convinced about the moral issue. But while my descriptive analysis distinguishing theodicies and defenses may not line up perfectly with her distinction between writings of God and evils that minimize versus those that aim to intensify (and rightly place) the sense of mystery that our desire to understand generates, I hope our approaches yield different, but accurate and compatible insights.Footnote ³³

McCord Adams and McCullough: Explanation or Justification?

Theodicy is a modern discourse project. The first essay in this project was written by the future Anglican Archbishop of Dublin, William King, in 1702, De Origine Mali. An English translation by Edmund Law was published in 1731 as An Essay on the Origin of Evil. The neologism “theodicy” followed as the title of Leibniz’s response to the issues, including lengthy discussion of King’s work, in his Théodicée in 1710. It is an anachronism to call premodern reflection on God and evil a theodicy or a defense, unless one is clear about why one is using an admittedly anachronistic notion.

A key characteristic of King’s work—and most modern theodicies and defenses—was its ignoring of actual evils, such as the killings at the Battle of Boyne in 1690, which settled the Williamite War in the favor of the Anglicans. King was then Bishop of Derry in the north of Ireland. This characteristic of not mentioning significant real evils is a central characteristic of the modern project of theodicy, in contrast to Milton’s Paradise Lost (1667) and other works on God and evil previous to the modern project. Projects like those of Kilby, Ross McCullough, and McCord Adams try to retrieve the insights of premodern views about God and to avoid the problems of modern accounts of God’s relationship to evil.

To call a theodicy an attempt “to justify God’s ways to man”Footnote ³⁴ is misleading. Despite her insights, Marilyn McCord Adams claims that her project in Christ and Horrors is not really a theodicy because it is an explanation, not a justification. Her reasoning presumes a de re (talk about things) approach in theodicies (and defenses).Footnote ³⁵ But even if a theodicy is an explanation, it is not an explanation of God’s ways. It functions not to justify God, but to show our set of theological beliefs reasonable. Defenses are a de dicto (talk about facts and reasons, for example) approach that defends the compatibility of A and B. “I have argued that horrors are so bad that no candidate reason-why … is remotely sufficient to exhibit the compatibility of horrors with Divine goodness-to created horror-participants, and that any attempt to construe one or more of them as sufficient underestimates how bad horrors are and caricatures God into something monstrous.”Footnote ³⁶ I entirely agree: modern theodicists caricature our belief in God.

But the author of such a caricature, as well of any theodicy or defense, is a human, writing de dicto, about what humans say about God, whose ways are not our ways. Explanations are justifications, not of God, but of our believing in the infinite goodness of God in the face of evils, even horrendous evils, a point she concedes when she claims that she is describing a “cosmos held together by Christ and to exhibit the coherence of Christology.”Footnote ³⁷ McCord Adams’s hypothesis is a justification of holding not merely theistic belief, but Christian belief. It is an explanation that attempts to justify our beliefs as reasonable. And it is finally in the genre of Christology, so McCord Adams has, in effect, changed the subject: she seeks to show how we can believe in true salvation in Christ in the face of the horrendous evils we know of. The subject of her work is christological, not theodical.

McCullough follows McCord Adams down the path of separating justification from explanation. His theodicy is an explanation of how God does not create evil, “so there is no way of linking evil to that Intelligibility that underlies creation and makes possible our explanations.”Footnote ³⁸ Again, this explanation is a de dicto justification of our beliefs about God and evil. McCullough writes in a de re mode, but his propositions and arguments function de dicto. They support a set of beliefs some people hold about God: he is “not asking what it is that mitigates the evils of history, what justifies them by overcoming them or integrating them into some greater good.”Footnote ³⁹ He seeks to show how we account for the origin of evil. His strategy is like that of Archbishop King’s: to explain what “we” believe about the origin of evil.

McCullough seeks to develop a theodicy that holds together God’s opposition to evil and God’s noncompetition with creatures.Footnote ⁴⁰ His key move is to attribute evil not to result from misuse of free will, but of human deficiency in acts that God wills effectively that we do. God wants the best of us. We resist the best for us and in so doing fall short of what God wants. God is not responsible for that falling short, but we are responsible for missing the mark, especially in our creaturely creativity.Footnote ⁴¹ McCullough also ignores social evils, thus replicating in his theodicy one of the main evils of the modern discourse about God and evil.