APPEALS TO MYSTERY AND THE EVIDENTIAL ARGUMENT FROM EVIL

Jeffery L. Johnson

Professor of Philosophy

Eastern Oregon University

Old fashioned atheists used to argue that evil -- pain, suffering, and misery -- proved that the God of theism -- omnipotent, omniscient, morally perfect -- did not exist. In the same way that the Euclidean geometer can establish that the interior angles of a plain triangle always equal one hundred eighty degrees, the atheist sought to establish that the existence of God was logically incompatible with the obvious pain and suffering that pervades this world. It is now widely accepted, by theists and atheists alike, that the arguments actually presented to demonstrate the alleged incompatibility were unsound. They were either invalid because they ignored the logical possibility of an omnipotent and morally perfect God having a "morally sufficient reason" for allowing evil, or they smuggled in an unrealistically strong premise designed to beg the question against this possibility.

Newfangled atheists are much more likely to appeal to the "evidential argument from evil" as a way of rationally supporting their negative theological hypothesis. Although there is a great deal of consensus regarding this general strategy, there is hardly any consensus about logic of the evidential case that is being made. Some atheists utilize a probabilistic approach.(1) Others simply lay out a prose discussion and ask reflective readers to trust their sense of what is reasonable to believe.(2) I count myself as part of a growing contingent of atheists who think that the strongest appeal to evidence is to be couched in "abductive" terms, or in the contemporary jargon, as an inference to the best explanation.(3)

Two very similar take-home essay exams are turned in a couple of hours late by one of the authors. Lengthy sections are word-for-word identical. Discussions with both students reveal that one student gave her completed exam to her friend to turn in for her and left early for Spring break. The other student concedes that all of this is true, but claims that he never looked at his friend's exam, and simply turned in the two essays after he had completed writing his. I think he is not telling the truth, and that he copied his essay from his friend's. Indeed, I think I have very good evidence that he is guilty of academic dishonesty. My case, of course, is indirect and based on circumstantial evidence. That does not mean, however, that it is in any way defective or weak.

It is quite easy to make my evidence explicit.

E1. Johnson assigned a take-home essay in his Introductory Ethics course.

E2. Two of the answers were substantially the same -- they included lengthy passages that were word-for-word identical.

E3. One student completed her exam, gave it to her friend to turn in, and left town.

E4. The other student turned in the identical exams.

================================

T0. The one student copied the exam answer from his friend after she gave it to him.

I was actually once put in the position of arguing before a Student Behavior Committee that E1 through E4 provided exceedingly good evidence for T0. T0 clearly explains the substantially identical essay answers, but I claimed to my colleagues that it provided the best explanation of all the relevant data. Thus, I implicitly conceded that there were lots of other possible explanations, such as the following.

T1. The young lady broke into her friend's computer and stole his answer before giving her paper to him.

T2. Both students independently, and coincidentially, copied their answers from some published source.

T3. Both students composed their answers independently; it was a freak coincidence that the answers were substantially identical.

T4. Both students possessed extraordinary degrees of ESP. One student was unconsciously reading the mind of the other as they composed their answers.

According to inference to the best explanation data, E1 through E4, counts as evidence for some hypothesis, T0, just in case T0 does the best job of explaining the relevant data. ESP and fluke coincidences would explain the identical essays, but the cheating hypothesis is simpler, less ad hoc, and more plausible. For this reason the identical exams constitute very good evidence that there had been academic dishonesty.

The evidential argument from evil appeals to three sorts of pain and suffering in the world, and claims that the negative existential hypothesis that there is no omnipotent and morally perfect God who is in any position to do anything about it is explanatorily superior to theistic accounts that attempt to articulate a morally sufficient reason for God's allowing it.

E1. The world contains vast amounts of moral evil -- human pain, suffering, and misery caused by the actions, choices, and oversights of other human agents engaging in (free) human actions.

E2. The world contains vast amounts of natural evil -- human pain, suffering and misery caused by non-human natural occurrences like disease, genetic defects, and natural disasters.

E3. The world contains -- and has contained for millions of years before the evolution of humans -- vast amounts of animal pain.

===============================

T0. The God of western theism does not exist -- there is no omnipotent and morally perfect who bears any responsibility for the pain and suffering of this world.

Many theists challenge the quality of the atheist's evidence in the manner of a good defense attorney; they try to create reasonable doubt by carefully articulating rival explanations -- theodicies -- which they argue do a better job, or at least as good a job, of explaining pain and suffering. The following rivals have found eloquent defenders in the contemporary literature.

T1. Free will -- The pain, suffering and misery we see is the result of freely chosen actions on the part of human (and perhaps angelic) agents. God can intercede in the choices of free agents only at the expense of turning genuine freedom into pseudo-freedom.

T2. Soul-making -- The evil of this world is a necessary part of creating a challenging environment that allows for genuine spiritual growth. God desires eventual spiritual communion with fully formed human souls; the evils of this world are required for the production of these fully formed souls.

T3. Causal and epistemological regularity -- Pain, suffering, and misery is the result of the world's operation under consistent and understandable causal laws. God could intercede, but this would require interfering with the causal regularity (this would be a kind of aesthetic flaw). More importantly, without causal regularity human agents are incapable of understanding, and to a limited extent, controlling their world. Thus, divine interference with the causal regularity of the world would implicate genuine free choice.

I have undoubtedly left off this list a theistic account of evil that you, dear reader, think is at least deserving of discussion. The remedy for this oversight is quite easy.

T4. Reader supplied morally sufficient reason -- God's reason for allowing pain, suffering, and misery is . . .

I argue that the atheist's hypothesis, T0, does a better job of explaining the tremendous amounts of evil than any of these theodicies by themselves, or in combination. This latter point is crucial, since I agree with Peter Van Inwagen about the following.

It seems to me that the theist should not assume that there is a single reason or a tightly interrelated set of reasons for the sufferings of all sentient creatures. In particular, the theist should not assume that God's reasons for decreeing, or allowing, the sufferings of nonrational creatures have much in common with His reasons for decreeing, or allowing the sufferings of human beings.(4)

Many theists are willing to simply register their disagreement with my plausibility ranking. They argue that some appropriate combination of the standard theistic theodicies does a perfectly adequate, some would even say superior, job of accounting for all the pain and suffering. I disagree, of course, but those colleagues and I may well have reached some termination point in rational dialogue. We may simply have to agree to disagree. At the very least, it will require a much more leisurely context than the present to even begin the job of cataloguing the explanatory deficiencies in the traditional theodicies.

I am concerned in the present context with a very different sort of theistic response to the above evidential argument. Some theists implicitly concede that they do not have a better explanation of evil -- moral, natural, and animal pain. They counter, however, with an argument that the sort of cosmic understanding implicit in the business of theodicy is actually quite beyond human moral and intellectual capabilities. Theists who adopt this strategy often seem to sympathetically feel the force of the evidential argument from evil. They admit that evil is troubling problem for theism at a number of levels -- ethical, theological, and epistemological. Basically, they admit that evil must remain a kind of theological mystery. But they claim that this an inevitable feature of the human intellectual condition, and that it in no way counts against the rationality of theistic belief.

What I will be calling the mystery theodicy begins with a reminder of the obvious limitations of the human cognitive condition with respect to the metaphysical, moral, theological issues that are central in the evidential argument from evil.

[O]ur epistemic situation is such that we are unable to make a sufficiently well grounded determination that [gratuitous evil exists]. . . . [T]he magnitude or complexity of the question is such that our powers, access to data, and so on are radically insufficient to provide sufficient warrant for accepting [its existence].(5)

It is hard to deny that all of natural theology is risky epistemological business. There is a kind of intellectual arrogance in committing oneself to the explanatory superiority of the atheist's, or the theist's, account of pain and suffering. But, although modesty and caution are theoretical virtues, cynicism and intellectual paralysis are hardly admirable. Purely theoretical work in mathematics or the humanities, to say nothing of the natural sciences, always runs the risk that the theorist has overlooked something, is not privy to all the data, or is simply trying to talk about something that exceeds current, or future, human intellectual competence. That is hardly an argument for not pursuing these intellectual enterprises, nor for trying to support our best theories with good evidence. Obviously we may need to revise our theories on the basis of new evidence, some superior explanatory candidate, or because the whole theoretical structure is leading nowhere. I would be happy to revise my atheism on the basis of new evidence, or some combination of superior theodicies, or even the perception that the secular naturalistic hypothesis left too many questions unanswered. My problem is that the theory explains all too many facts about this world. And it does so in a clear, concise, straightforward manner.

We typically appeal to appeal to mystery in non-theological contexts when one or more of four conditions apply.

The atheist's hypothesis can hardly be challenged on (1) or (3). Atheism is a relevant and direct response to a theological question that tries to smuggle in a theistic response -- Why does an omnipotent and morally perfect God allow all the suffering? Because there is no such being. Furthermore, we understand the existence, pervasiveness, and long biological history of pain and suffering all too well. Some theist may argue that atheism is flawed on grounds like (2). This is an interesting and challenging response that cannot begin to be adequately answered in the present context. I take it to be obvious that the problem of evil ceases to be much less serious if conclusive independent evidence for the existence of God is forthcoming. But, of course, whether we have reached this ancient goal of natural theology remains, at best, highly controversial. The heart of the mystery theodicy, therefore, is based on the epistemological and methodological concerns in (4). Some direct response to this worry is required by atheists.

I argue that appeals to fundamental human ignorance as a way of deflecting the evidential argument from evil are both theologically dangerous, and grossly overestimate the problem. We can see this by probing to very interesting thought experiments proposed by contemporary theists who advocate the mystery strategy.

Wykstra articulates the question at the heart of the mystery theodicy. "[I]f there were an outweighing good of the sort at issue, connected in the requisite way to instances of suffering like this, how likely is it that this should be apparent to us?"(6) His answer is "not very." A detailed and highly technical discussion is offered to support this response. But, it is vividly summarized with a catchy, but I believe misleading, metaphor.

[T]he outweighing good is at issue is of a special sort: one purposed by the Creator of all that is, whose vision and wisdom are therefore somewhat greater than ours. How much greater? A modest proposal might be that his wisdom is to ours, roughly as an adult human's is to a one-month old infant's. (You may adjust the ages and the species to fit your own estimate of how close our knowledge is to omniscience.)(7)

It may seem like caviling to press so hard an analogy that is clearly designed to make the uncontroversial point that human knowledge is a far cry from infinite knowledge. But, I believe that other work is being done here, and it deserves analysis. One-month old infants are in capable of any relevant reasoning. They don't "understand" adults; they don't form moral or theological judgments, and the don't evaluate evidence or judge explanatory plausibility. They are at this juncture in their careers intellectually cut off from all of these rational pursuits. A certain strand in natural theology has always been attracted to the idea that the human intellect is similarly cut off from rational understanding or evaluation of the divine. Consider Hume's character, Demea.

When I read a volume, I enter into the mind and intention of the author; I become him, in a manner for an instant, and have an immediate feeling and conception of those ideas which resolved in his imagination while employed in that composition. But so near an approach we never surely can make to the Diety. His ways are not our ways. His attributes are perfect but incomprehensible.(8)

One month old infants cannot "enter into the mind and intention of" adults. They cannot make "so near an approach" to the concerns of adults. Adults' way are not their ways. Adults' attributes are incomprehensible. Such an intellectual break comes at a high price. One-month old infants do not love adults, nor respect them, nor worship them. They don't even fear them. They simply interact with them in a relationship of instinctual biological dependence.

For natural theology to be at all interesting -- and I would go farther and say for religion to be interesting -- we must suppose some understanding of the nature of the deity that is the object of religious attention. Theists believe in an omnipotent, omniscient, and morally perfect God. Certainly they must have some better understanding of infinite power, knowledge, and goodness than that of a one-month old infant. If they didn't, why even claim to believe in God, rather some other metaphysical presence?

Wykstra invited us, in his original metaphor, to adjust the ages according to our own estimates. Notice what happens when we make the child a bright seven year old, and in order to keep the range of intellects comparable, we make the adult a genius and a saint. There are, of course, many respects in which the seven year old cannot understand the adult. Wykstra's metaphor is very useful in reminding all of us interested in the evidential argument from evil that these sorts of intellectual limitations are central to the discussion. But I would argue that they in no way preclude us from judgments of explanatory plausibility, any more than they would preclude the seven year old from claiming to have evidence about the adult. Suppose for example, the child claims that some action on the part of the adult is unfair. She may have relevant evidence in support of this judgment.(9) We may smile from our own adult perspectives at all she overlooks or misunderstands, but we shouldn't accuse her of improper reasoning. She is stuck, as are we all, with making the best of what you've got. She's bright, reflective, and inquisitive. Her judgment is that some action is unfair. It is not simply that she doesn't like the action, but on the basis of thought and comparison this is a reasoned judgment. Notice, further, that her reasoning looks even better if she asks others what they think, and receives no better theory. And finally, her reasoning begins to look compelling -- from our adult perspective -- if our imagined genius/saint is fully aware of her puzzlement and intellectual discomfort, and chooses to remain silent.

Thus, I am claiming that the atheist can fully concede that there is an inherent limitation of the human cognitive condition that makes all of natural theology -- including consideration of the evidential argument from evil -- very difficult. The atheist, and of course the theist as well, should be modest about the strength of any of the evidence. Reasonable thinkers may disagree about what data is relevant, and about rankings of explanatory plausibility. None of this counts against the quality of the atheist's argument. We, like my imagined seven year old, have no choice but to exercise our own intellects. From our perspective, pain and suffering is relevant data that cries out for explanation. Atheism, however psychologically sad, does a pretty good job of accounting for the misery. No better explanation -- so claims the atheist -- is forthcoming. Consequently, according to the model of inference to the best explanation, there is evidence, perhaps good evidence, in support of atheism.

There remains one consoling feature of the mystery theodicy, for both theists and atheists. Maybe the obvious intellectual modesty required by this line of reasoning allows us to hold out some hope that relevant new data may be forthcoming. Perhaps some wise theist will articulate a theodicy, or combination of theodicies, that will do a better job of explaining evil. Perhaps other evidence will become apparent -- divine revelation, or our continued existence in some afterlife state. The evidence is against this now, but perhaps.

ENDNOTES

1.    Salmon.

2.    Rowe.

3.    Johnson, Russell, etc.

4.    Van Inwagen, p. 157.

5.    Alston, p. 98.

6.    Wykstra, 155.

7.    Ibid.

8.    Hume (Pike), p. 38.

9.    I should note here that inference to the best explanation is much more straightforward as a model of good evidence in cases where the theory being defended is causal or existential, rather than normative. The model can be expanded to cover axiological theories, however. See, Wright.