Jump to ContentJump to Main Navigation
MolinismThe Contemporary Debate$

Ken Perszyk

Print publication date: 2011

Print ISBN-13: 9780199590629

Published to Oxford Scholarship Online: January 2012

DOI: 10.1093/acprof:oso/9780199590629.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 23 October 2018

An Open Theist Theodicy of Natural Evil

An Open Theist Theodicy of Natural Evil

(p.281) 17 An Open Theist Theodicy of Natural Evil

William Hasker

Oxford University Press

Abstract and Keywords

This chapter sets forth a theodicy of natural evil. General characteristics of theodicy are discussed, including the distinction between general‐policy theodicies and specific-benefit theodicies. There is also consideration of the relationship between theodicy and different theories of providence, including Augustinianism, Molinism, and open theism. It is argued that the ‘skeptical theist defense’ against the problem of evil should be rejected because it leads to an unacceptable moral skepticism. A theodicy of natural evil is presented, and it is argued that open theism makes the prospects for such a theodicy better than they are for either Augustinianism or Molinism

Keywords:   theodicy, providence, Augustinianism, Molinism, open theism, natural evil, skeptical theism

The problem of evil has received an enormous amount of attention in recent years, probably more than in any previous period in the history of religious thought. The enterprise of theodicy, however, has received only a small fraction of that attention. On the one hand, theodicy has seemed to many an unattractive proposition, unlikely of success and perhaps even morally problematic. On the other hand, a great deal of hope has been invested in the ‘skeptical theist’ defense, which argues that, because we have no reason to expect that we could know the reasons why God permits evil, it is no objection to theistic belief that we do not know such reasons.

I. Remarks about theodicy

I demur from both these judgments. As will be explained below, I believe the prospects for the skeptical theist defense are much less favorable than they are often thought to be.1 And on the other hand, I view the possibilities for theodicy as more inviting than is often supposed. But in order to set the stage for the theodicy to be proposed here, it will be helpful to include a few general remarks about the enterprise of theodicy and the ways of going about that enterprise. I take it that a theodicy is a response to an argument from evil, an argument that claims that in view of some evil that exists in this world it is incoherent or unreasonable to believe in the existence of the theistic God. A theodicy replies to such an argument by giving a justifying reason for the existence of the evil in question; a reason such that, if it indeed obtains, the permission of the evil by God is morally justifiable and does not constitute a reason to disbelieve (p.282) in God's existence or his goodness. It is unnecessary, and often unwise, for the theodicist to claim that the reason given is the actual reason God has permitted the evil; this may or may not be the case. Now, the justifying reason for God's permission of an evil consists essentially of some good that is made possible, or some evil that is averted, by God's permission of the evil in question. But these goods, or evils averted,2 may be of two kinds. Suppose the evil permitted by God is a major hurricane striking the Atlantic coastline of Florida, with both damage to property and loss of life. A theodicist might respond by pointing out that the occasional occurrence of such hurricanes is an inevitable consequence of the overall climate system of planet Earth, a climate system which much of the time, and over much of the planet's surface, produces conditions which are conducive to the flourishing of all kinds of living creatures, including human beings. Call a theodicy which relies on considerations of this sort a general‐policy theodicy; it justifies God's permission of certain evils as being the consequence of a general policy which a wise and benevolent God might well adopt.

On the other hand, the theodicist might take a different tack. In justifying God's permission of the hurricane she may appeal, not to general policies such as the maintenance of the planetary weather system, but to specific benefits that result, or harms that are averted, by the particular hurricane in question. Call a theodicy of this sort a specific‐benefit theodicy. The relationship between these two sorts of theodicy requires careful consideration.3 In one sense, to be sure, there is no inconsistency between the two: it could very well be the case both that the occurrence of the hurricane is the consequence of a wise and benevolent policy with respect to the planetary climate system, and also that there are specific benefits obtained, or harms prevented, by God's permission of the hurricane that could not have been obtained had the hurricane been prevented. However, the following question needs to be considered: In justifying God's permission of a particular evil, is it sufficient that the evil is the result of a general policy which a wise and good God might well adopt? Or is it necessary, on the contrary, that God's permission of this particular evil should have consequences which are better than any that could have been obtained had God prevented the evil? The answer to this question has momentous consequences for the enterprise of theodicy. If the answer to the first question is affirmative, it follows that, strictly speaking, theodicy has no need whatever to appeal to beneficial consequences from God's permission of a specific instance of evil. If the (p.283) evil in question is the result of a general policy that is itself wise and good, that is a sufficient answer to the problem theodicy sets out to address. To be sure, if there are specific beneficial consequences they may furnish a signal example of God's ability to bring good out of evil, and this theme is an important element in the Christian understanding of the relationship between God and evil. But theodicy need not rely on such benefits in order to accomplish its task.

Suppose, on the other hand, that a successful theodicy requires that there be a specific benefit from the particular evil under consideration. In this case, the invocation of general divine policies will never be sufficient to provide justification for God's permission of a particular evil.4 This need not mean that the discussion of such policies is entirely pointless; the specific‐benefit theodicist may welcome an argument showing that the general policies followed by God in the world's governance are on the whole wise and beneficial. But any argument of this sort necessarily falls short of rebutting the charge that a particular evil morally should not have been allowed by a good God.

II. Theodicy and models of providence

It is clear, then, that the enterprise of theodicy will be carried on in importantly different ways depending on whether we opt for a general‐policy theodicy or a specific‐benefit theodicy. And it is just at this point that the differences between models of divine providence come to be of crucial importance. The key distinction here is between what may be termed risk‐taking models of providence and risk‐free models. The difference between the two sorts of models is highlighted by the following question: Is God's decision between two or more alternative courses of action informed and guided by knowledge of the specific consequences that will follow from each course that might be chosen? If the answer to the question is Yes, then God's choice involves no risk whatever; all of the consequences that will ensue have been fully taken account of in the initial decision. If on the other hand the answer is No, then God's action does involve a greater or lesser degree of risk: how things turn out will depend on future contingencies (including and especially human free choices) that are undecided as of the logical ‘moment’ of God's decision. The risk‐free models on offer include Molinism, with its doctrine of divine middle knowledge, and (p.284) theological determinism—or as I shall generally call it, Augustinianism.5 The risk‐taking models include open theism, but also the doctrine of ‘simple foreknowledge’ as well as divine timeless knowledge, where the latter is not supplemented either with middle knowledge or with theological determinism.6

For risk‐taking models of providence, it is evident that the sort of theodicy to be pursued must be general‐policy theodicy. Since the specific consequences of the divine decision to permit a particular evil are unknowable, even to God, prior to the decision's being made,7 it would be entirely unreasonable to expect that a wise God should ensure that the consequences of permitting it are better than the consequences of any possible way of preventing the evil. Indeed, in a great many cases there may be no answer to the question as to what course of action would have had the best consequences. What the outcomes would have been for a different divine decision will often depend on what people would freely have done under those different circumstances—and, absent the counterfactuals of creaturely freedom, there is no correct answer to the question as to what they would have done. On these assumptions, the task of theodicy becomes relatively straightforward, though not by any means easy. First, one must attempt to identify the general divine policies which in their outworking result in the evils in question. Then, one must defend the claim that these are wise and good policies, policies a loving God might well adopt.

It is only slightly less evident that a risk‐free model of providence should lead to a specific‐benefit theodicy. To be sure, the believer in a risk‐free providence may well be interested to show that the general policies followed by God in his governance of the world are wise and beneficent. But the God of Molinism or Augustinianism has decided to permit the evil in question in the light of full knowledge of the evil itself and its specific consequences. It seems, then, most implausible that God would permit the evil simply on the basis of general policies, while ignoring the particulars surrounding the occurrence of the evil and its consequences. The point may be (p.285) illustrated by a criticism sometimes made of John Hick's ‘soul‐making theodicy’. It may be true, the critic admits, that in some cases severe suffering results in a dramatic change for the better in the character of the sufferer, manifesting itself in increased closeness to God and greater love and compassion towards one's fellow‐creatures. But it is also true that such suffering can have the effect of embittering the sufferer and closing her off from beneficial prospects of growth and fulfillment. When the latter outcome is the result, can we defend the divine decision to permit the suffering by pointing to the benefits of such suffering in other cases? Certainly such a defense would ring hollow, if we assume that the divine decision to permit the suffering was made in full knowledge of the results that would follow in this particular case.8 The proponent of a risk‐free providence, if she propounds a theodicy at all, must opt for a specific‐benefit theodicy.

III. The skeptical theist defense

But should she propound any sort of theodicy? I believe that the combination of a risk‐free providence and theodicy, while not inherently impossible, carries with it severe difficulties. These difficulties have led many contemporaries to abandon the task of theodicy in favor of defense9—and indeed, in favor of a specific sort of defense, the ‘skeptical theist’ defense. Let me explain why this is so. We may helpfully take a lead here from an important article by William Alston, ‘The Inductive Argument from Evil and the Human Cognitive Condition’.10 Here Alston enumerates a number of the traditional themes of theodicy: evil can serve as a deserved punishment, as a contribution towards soul‐making, as a means to humble us and turn us towards God, and so on. He then proceeds to argue that, in a great many cases, we are in no position to rule out one or another of these as the justifying reason for God's permitting this or that instance of evil. At this point, however, he confronts certain cases proposed by William Rowe for which, he is forced to admit, none of these traditional answers is at all promising. In response he writes, ‘This lacuna in the (p.286) argument is remedied by the point that we cannot be justified in supposing that there are no other reasons, thus far unenvisaged, that would fully justify God in permitting Rowe's cases…. Even if we were fully entitled to dismiss all the alleged reasons for permitting suffering that have been suggested, we would still have to consider whether there are further possibilities that are undreamt of in our theodicies’.11 This move to goods and evils that are unknown to us seems to render the skeptical theist's position impregnable: since the goods and evils in question are completely unknown to us, it is out of the question that the objector should be able to rule them out conclusively as supplying the goods made possible, or the evils averted, by this or that instance of evil. To be sure, some objectors may fail to be impressed by this move: if the goods and evils in question are entirely unknown, why should we suppose they exist at all? At best, this line of argument seems likely to result in a stand-off.

The skeptical theist's position, then, can be summarized as follows: For every instance of evil permitted by God to occur, there is indeed a greater good that is achieved, or a greater evil prevented, that justifies God's permission of the evil in question. Due to our severe epistemic limitations, there is no reason to expect that we should be able to identify these justifying benefits, and so our inability to do so in a great many cases does not constitute evidence against the truth of theism. Clearly, the ‘severe epistemic limitations’ play a crucial role in the skeptical theist's position; it is these limitations which must account for the great disparity between what seems, prima facie, to be the case and what, according to the skeptical theist, really is the case. I believe that most of us will find, if we put aside preconceptions as much as possible and judge on the basis of reason and experience, that there are an enormous number of cases in which it seems to us that there are evils which do not result either in any corresponding greater good or in the prevention of some other equal or greater evil. The skeptical theist has need of a skeptical stance in epistemology such that (i) it deprives the judgment noted above, which almost all of us are spontaneously inclined to make, of all epistemic warrant; and (ii) is not an extreme or bizarre form of skepticism which would itself invite disbelief; and (iii) does not invalidate judgments of the sort that we commonly make in everyday contexts and are such that our lives would be impaired or impoverished were we unable to make them.

The matter cannot be argued in detail in this chapter, but I am convinced that a skeptical position of the sort required is not available to the skeptical theist. In defense of my claim, I will explore briefly the skeptical (p.287) consequences of the view as recently articulated by Michael Bergmann, one of the staunchest defenders of skeptical theism.12 According to Bergmann, our epistemic limitations preclude our being able to identify any state of affairs as an all‐things‐considered good, the sort of thing a good God would wish to include in his creation. By the same token, we are unable to identify any state of affairs as an all‐things‐considered evil. But given these limitations, the very idea of our attempting to make the world a better rather than a worse place—an idea that plays an important role in both ethical theory and practice—drops out of consideration altogether. We cannot meaningfully attempt to do this, because we have no idea how to go about doing it. We may, to be sure, try to bring about the states of affairs that seem to us most likely to make the world a better place. But the attempt to do this has no real ethical significance, because we have no reason whatever to think that the states of affairs that seem to us to make the world better really do that rather than the opposite. Is this not, I ask the reader, a fairly extreme form of ethical skepticism?

The point may come home more vividly if we consider two fairly complex states of affairs—namely, the lives respectively of Mother Teresa of Calcutta, and Saddam Hussein of Iraq. We pose the question: Is one of these lives better, all things considered, than the other? We must, however, be careful to make sure we are asking the right question. If we were to ask which of the lives is morally better, the answer would be very easy. If we test the two lives against any reasonable moral standard, Teresa's life will be seen to be incomparably superior to Saddam's. But that is not the question we need to be asking. What we need to know is, which life is better all things considered, in the light of all its logical and causal connections with all sorts of other events. Which life, in short, makes a more positive contribution to the overall goodness of the world? The answer given by skeptical theism is that we simply don't know. But putting it this way tends to understate the skepticism involved. Consider, then, the following proposition: The life of Mother Teresa is better, all things considered, and makes more of a positive contribution to the goodness of the world overall, than the life of Saddam Hussein. In the light of everything we know through experience and moral reasoning, what epistemic probability should be assigned to this proposition? According to the Bergmannian skeptical theist, the answer is that the probability in question falls somewhere in the range between zero and one—and that is all that can be said. We simply have no idea whatsoever, if we reason correctly, which of these lives is better overall. And now we have reached the point where it must be left to (p.288) the reader to decide whether this is a moderate and reasonable skepticism, as Bergmann claims, or whether it is instead a bizarre, extravagant, and wholly implausible skepticism, as it seems to me to be.13

The material presented to this point is essentially preliminary; it prepares the way for the theodicy of natural evil which forms the main substance of the paper. This is an open theist theodicy, one that assumes a risk‐taking view of providence. It is, accordingly, a general‐policy theodicy rather than a specific‐benefit theodicy. The theodicy aims to provide a justifying reason for God's creation of a natural world such as the one we find ourselves in, with the various evils incident to its existence; it does not, however, address the question of God's reasons for permitting particular instances of evil. Something will be said about the reasons why God's creation contains hurricanes and earthquakes, but not about God's reasons for permitting Hurricane Katrina or the 2008 earthquakes in central China. Insofar as reasons of the latter sort are desired, they must be sought elsewhere.

IV. The structure of a world

The theodicy begins by setting out some very general, structural features of a world.14 These features are, of course, abstracted from the world in which we live, but they are general enough to apply across a wide variety of possible universes. I claim that we can see that it is good that a world with these features should exist. I will then go on to point out that the various forms of natural evil arise as a consequence of these structural features. If it is good that a world should exist with these structural features, it is also justifiable that the natural evils should be allowed to exist; they are, so to speak, the price of admission for the existence of such a world. Finally, I will consider one objection to my theodicy, an objection which puts into graphic relief the differences between Augustinianism, Molinism, and open theism.

First, it is good that there should be a world. By ‘world’ here I mean the sum total of concrete things that exist, other than God if there is a God. And to say that it is good that there should be a world, is to say that it would not be (p.289) better if, instead of any world's existing, there should be absolutely nothing at all—again, apart from God if there is a God. To say that it is good that there should be a world is an extremely minimal affirmation of the value of existence. The denial of this affirmation, while conceivable in the abstract, would be an expression of utter nihilism; for most of us, I hope, such a denial is not merely implausible but virtually inconceivable, and is not in any sense a live option. That it is good that there should be a world is of course compatible with the view that it would be far better if a world vastly different from this one should exist, so this is only a small first step towards our goal of theodicy.

Here is my second claim: It is good that there should be a complex, multi‐leveled natural world. This proposition is itself quite complex, but the various ingredients do not readily lend themselves to separate consideration. To say that a world is complex is to say that it contains many different entities and kinds of entities, interacting with each other and doing many different sorts of things. To say that it is multi‐leveled is to say that the entities exhibit different degrees of complexity, both in their internal structure and, more importantly, in their causal powers; those that are more complex in their structure and powers are thereby ‘higher’ than those that are less complex. To say that the world is natural is to say that the entities act, and interact, in accordance with their inherent causal powers, as opposed to being manipulated by some other, presumably ‘higher’, being. (Think of the difference between a puppet show and a group of human beings and animals interacting naturally. The charm of the puppet show, of course, consists in the fact that the puppets, if skillfully handled, are able to simulate many aspects of such natural interaction.)

Of the features mentioned, perhaps the idea that a world should be multi‐leveled especially invites further exploration. A world that did not exhibit this feature would consist of entities all on the same level in terms of complexity of structure and causal powers. At one extreme, this might be a world simply of ‘atoms’—the simplest, most elementary objects there can be.15 (This is not the same, by the way, as a world composed of atoms and nothing else; such a world might well contain structures of extraordinary complexity. Nothing is said here against that possibility.) I do not suppose there will be much objection to the judgment that a world consisting only of separate atoms would be uninteresting, even boring, if there were anyone around to be interested or bored. The other extreme would be a world such as Berkeley's, consisting of rational spirits and (p.290) nothing else. Notably, however, Berkeley's world contains the systematic illusion of a myriad of less complex entities, providing for those rational spirits an arena in which to act and interact with each other. In a positive way, the notion that a world should be multi‐leveled captures something of the sense of the ‘great chain of being’ that was thought to connect God and all of creation; it was important for the completeness of the whole that all the levels of metaphysical excellence—all the ‘links in the chain’—should be occupied.

Consideration of the atoms‐only and the spirits‐only worlds suggests a third desideratum: It is good that a world contain living beings that are sentient and rational. In thinking about the atoms‐only world, we were forced to imagine at least one rational being (perhaps the creator) in order to make sense of the notion that the world could be assigned any valuation. But if the world is good, then it is desirable that it be found to be so by its inhabitants, and surely their appreciation of it will require extensive sensory capacities as well as reason, which is needed to enable the evaluation. Conceivably this desideratum taken by itself could be supplied by a single sort of beings that are both sentient and rational. But a multi‐leveled world will contain beings with graduated arrays of sensory and rational capacities, allowing for a rich variety of ways in which the world can be apprehended and appreciated.

Beyond this, I maintain that it is good that the creatures in the world should enjoy a considerable degree of autonomy. ‘Autonomy’ suffers from a poor reputation in some religious circles, and not without reason. To be autonomous is to be self‐ruled, and often this has been taken to imply freedom from all rule by another, even by the Creator. This is not the kind of autonomy advocated here. Each and every creature is totally dependent on the Creator for its very existence; not only for its original coming‐into‐being, but for sustaining its existence from moment to moment. And the right of the Creator to rule over his creation is not in question. Surely, however, it is conceivable that a being that depends on a superior power, and lies properly within that power's scope of control, should nevertheless in many situations be allowed the freedom to operate according to its inherent capabilities, without direct control or interference by that power. And to the extent that this is done, the intrinsic worth of the being is more clearly exhibited than it would be were this degree of independence not allowed. This is most evident, I believe, in the case of persons endowed with free will; that after all is the key insight on which the Free Will Defense relies for its credibility. But there is a good measure of plausibility in applying it to non‐personal agents as well; much of the fascination of computers and robots, for example, stems from the impression they can give of acting spontaneously without direction, and producing novel and (p.291) surprising results. This autonomy acquires an added dimension if, as seems most likely to be the case in our world, there is an element of ‘chanciness’, of indeterminacy, in the fundamental processes of the natural world. It is important, however, to stress that the autonomy praised here is a relative, not an absolute autonomy; the self‐activity of the creature is not valued so highly that any special action by the Creator, above and beyond sustaining the creatures in existence with their inherent causal powers, is ruled out as unacceptable.

Finally, I claim that it is good that there should be an evolving world, a world in which the universe as a whole as well as its component systems develop from within, utilizing their inherent powers and potentialities. This judgment is obviously dependent on contemporary science; until recently the judgment could not easily have been made, because we knew too little about the developmental history of the world to envisage an evolving world as a realistic possibility. But now we do have that history available—enough of it, at least, to perceive its inherent wonder. The majesty of the Grand Canyon, for instance, is greatly enhanced by the recognition of the hundreds of millions of years of geological history recorded in its successive strata. In this ‘golden age of cosmology’ who can help but marvel at the story of the unfolding, since the Big Bang, of the astronomical structures of which our universe is composed? And ever since Darwin there has been the story, still far from complete but continually enriched with new discoveries, of the development of life on this our earth.16 That this is so cannot be seriously contested. But is it good that it should be so? Is a world that has unfolded in this way better than if, as our fathers believed until very recently, the major features of the universe, and each separate kind of living creature, had been ‘hand‐crafted’, as it were, by the Creator? Not all will agree on this, to be sure—but I believe that it is good, and in support of this I cite earlier thinkers who reached a similar conclusion. Consider, then, these words of Henry Ward Beecher:

If single acts would evince design, how much more a vast universe, that by inherent laws gradually builded itself and then created its own plants and animals, a universe so adjusted that it left by the way the poorest things, and steadily wrought toward more complex, ingenious, and beautiful results! Who designed this mighty machine, created matter, gave it its laws, and impressed upon it that tendency (p.292) which has brought forth almost infinite results on the globe, and wrought them into a perfect system? Design by wholesale is grander than design by retail.17

Beecher, of course, wrote when Darwinism was in the ascendancy, and might be suspected of tailoring his theology to the mood of the times. But similar themes appear in much earlier Christian writers; Gregory of Nyssa, in the fourth century, wrote:

The sources, causes, and potencies of all things were collectively set forth in an instant.…Then there followed a certain necessary series, according to a certain order, as the nature of the Maker required, appearing not by chance but because the necessary arrangements of nature required succession and the things that would come to be.18

It would be difficult to find a clearer expression of the excellence of a world produced by evolutionary development, yet this was written by one wholly innocent of evolutionary science. Finally, I cite the words with which Darwin concluded The Origin of Species:

There is grandeur in this view of life, with its several powers, having been originally breathed into a few forms or into one; and that, whilst this planet has gone cycling on according to the fixed law of gravity, from so simple a beginning endless forms most beautiful and most wonderful have been, and are being evolved.19

V. The causes of natural evil

We have before us the conception of a natural world that is complex and multi‐leveled, containing creatures some of which are sentient and even intelligent. This is an evolving world, one that enjoys a considerable degree of autonomy in its operations. The task now is to show how, in our actual universe, these features lead inevitably to the sorts of occurrences we describe as ‘natural evil’. There is however an inherent limitation in this enterprise, marked in the previous sentence by the words, ‘in our actual universe’. It will not be possible, in general, to show that these same features must lead to similar results in a universe quite different from (p.293) this one, made up of different fundamental constituents and operating according to different natural laws. For the truth of the matter is that we have very little grasp of the nature, or even the real possibility, of such alternative universes. This fact, which will come to our attention repeatedly in this section, constitutes a fundamental limitation on the enterprise of theodicy: we are unable to compare the actual universe with possible universes that differ radically from it, in order to show that the actual universe is better than those others, or at least as good as any of them. However, the limitation is much more serious for the enterprise of constructing arguments from evil, in that someone who wants to claim that some very different sort of universe would be better than this one is quite unable to flesh out the alternative possibility, or even to show clearly that there are alternative possibilities. (Of course it is comparatively easy to show that there are possible universes that differ from ours in minor details, but this will turn out to be much less significant than alternatives that are radically different.)

But it is time to proceed with the main agenda: How is it that the realization of the features we have enumerated leads to the phenomena of natural evil? Begin with some of the largest and most impressive natural disasters: volcanic eruptions, earthquakes, and tsunamis. All of these events can have truly catastrophic consequences; they are often enormously destructive of life‐forms within the affected areas. These events result from the fact that much of the interior of the earth is in a molten or semi‐molten state; the earth's crust to which we cling is only a thin skin over the inferno beneath. It is also true, however, that the availability of the earth as a site for life depends heavily on these same facts: a planet that was geologically ‘dead’ would have little prospect of harboring life‐forms like ourselves.20 To cite one factor among many, a molten interior is a requirement for the earth to have a strong magnetic field, which shields life‐forms from the otherwise damaging effects of the ‘solar wind’ and cosmic rays. Furthermore, key transition‐points in the evolution of living creatures came about as a result of enormous volcanic eruptions, eruptions that were catastrophic in their immediate results but immensely creative in the longer term. These transition‐points are an integral part of the history that has resulted in the appearance of our own species. As a result of the earth's molten interior we have plate tectonics, resulting in mountain‐building and earthquakes, volcanism and tsunamis—and also in the presence on earth of Homo sapiens.

(p.294) Similar considerations apply to the phenomena of weather and climate, that can often have seriously harmful consequences: hurricanes, tornadoes, and drought. These phenomena result from the circulation of the atmosphere, in its interaction with earth and oceans, all governed by the fundamental laws of physics. Some of the phenomena thus produced are astonishingly complex, and even now defy accurate prediction for more than a few days in advance. They do cause harm and destruction on occasion, but they also provide good conditions for the life and growth of living creatures much of the time and over much of the earth's surface. If we are inclined to complain about this arrangement, the complaints will undoubtedly be centered on the destructive events: why can't we have just ‘normal’ weather, without the occasional catastrophe? (Never mind that a certain approximate number of hurricanes and tornadoes each year is normal; a whole different pattern of normality would be required to exclude them.) Or to put it more bluntly: why isn't the weather more responsive to human wants and needs? Raising this question leads us to an important observation: the fundamental laws of nature are impersonal: they specify the behavior of the fundamental constituents of nature in ways that may be mathematically complex but that make no reference to purposes that are meaningful in the context of human life. The stones in an avalanche do not go where they go because it would be a good thing for them to go there; the physical and chemical properties of water do not suddenly change when air‐breathing creatures are in danger of drowning. The laws of nature, it seems, are not respecters of persons.

Would it not, however, be better if things were different in this respect? Actually, whether this would in fact be better is very much open to question. But there is another, prior question to be asked: Is a situation in which the sub‐personal forces of nature adapt their operations with special regard to human concerns even possible in a natural world—one in which the entities act, and interact, in accordance with their inherent causal powers? Suppose, for instance, the desired goal is that lightning should avoid striking where it would endanger human lives. To accomplish this the lightning, or the meteorological conditions that produce the lightning, would have to be able to recognize the presence of human beings and the fact that a lightning strike in the area would endanger them, and then to select an alternative, non‐threatening target area, and finally to re‐direct the strike to that alternative area. It is wholly implausible that there is anything in the natural situation capable of this kind of discernment; what really is required is a ‘spirit of the lightning’ that deliberately (and intelligently) selects targets according to some previously determined protocol. But then, of course, we would no longer have a natural world in this respect—nor, one would think, in very many other respects if similar (p.295) ‘human‐friendly’ behavior is to be expected from other forces of nature. These scenarios simply can't be taken seriously.21

Turn now to consider living organisms, the centerpiece of the story. It is immediately evident that organisms are subject to the various kinds of natural disasters enumerated above; indeed the events would not be termed disasters at all were it not for their devastating effects on life‐forms. We must also consider the need of organisms for sustenance, the energy and nutrients required in order to live, and function, and reproduce. Plants and some micro-organisms can derive sufficient energy from sunlight and obtain needed nutrients directly from the soil. But sustaining life in this way is a slow business, and is not conducive to a way of life involving the higher functions characteristic of the more complex animals. Some highly evolved animals are vegetarian, but many are not; there are significant advantages to be derived from ingesting the highly concentrated nutrients that are available from the bodies of other animals. The general theme of evolution, of course, is that organisms evolve to occupy an available ecological niche; and some of these niches are open to predators and to parasites; disease organisms also have developed so as to exploit opportunities that are available in the ecosystem.

All this is to say that the death of living creatures is a pervasive and inescapable feature of any world remotely resembling our own. And once we have complex creatures susceptible to death and harm, pain and suffering are likewise inevitable. Many potential sources of harm can be avoided, and some harms can be alleviated, but only if the organism is sufficiently motivated. Pain is nature's way of ‘getting our attention’, and it is remarkably effective in doing so. Admittedly, great pain is sometimes suffered when the harm is already beyond the point where anything can be done about it. But to insist that pain ought to be present only when it is possible for the harm it signals to be alleviated is to make a demand that surpasses the engineering limitations of the organism.

All of this becomes especially evident, to be sure, when we consider the world in an evolutionary aspect. Whatever else may be involved, evolution as we know it depends on natural selection, and this in turn involves the winnowing process brought about by the death of less fit organisms. This is one reason some Christians object to evolution; if God has deliberately chosen to create the world through an evolutionary process, this (p.296) makes it impossible to see death and suffering as interlopers brought about as a result of human sin. But most of the reasons given why death is inevitable in a complex natural world hold with or without evolution. If anything, the idea of evolution should be the more welcome because it points to a constructive result from the vast amounts of death and suffering that are inevitable in the natural world in any case.

Is the world cruel? Under certain circumstances, it is natural for us to think so. But the world is not cruel, for it lacks the capacity to be cruel. Cruelty is defined as ‘willfully or knowingly causing pain and distress to others’ and as ‘enjoying the pain and distress of others’.22 Nature does neither of these things, for ‘nature’, conceived as a system of natural laws and forces, neither wills nor knows nor enjoys. The system of nature simply is there, and operates according to its inherent powers, with no thought for the pain or enjoyment of any creature. One might say, cruelty can exist only where there is the potentiality for kindness, and nature is neither cruel nor kind. Even to call nature ‘indifferent’ is to indulge in the pathetic fallacy, for nature is not the sort of thing to which the pleasures and pains of sentient creatures could ever make a difference. True, some animals are capable of understanding to some degree the mental state and the suffering of another creature, but this capacity—what students of animal behavior term a ‘theory of mind’—probably is less common than we are inclined to think. In all likelihood a cat, playing with a mouse before dispatching it, has no thought or awareness of the pain it is causing the mouse. Subjectively, the cat is playing with an especially amusing toy; in the broader evolutionary picture, the cat is practicing tactics to be used in hunting for prey. (It is considered quite remarkable when, among our hominid ancestors, remains are found of an individual who apparently survived for a number of years with an injury that would have prevented it from foraging effectively for food. The individual must have been maintained by support from members of its group, an early sign of altruism.)

But if nature can be neither kind nor cruel, should we conclude that cruelty is rather to be attributed to the one who planned and created such a system? Not unless, reversing the judgments we have made up to this point, we are prepared to say that the existence of the world is a bad thing overall—that it would be better that nothing at all should exist (or nothing other than God) than that such a world as this one should be. Such a nihilistic denial of the goodness of reality, if consistently maintained, (p.297) cannot be answered by argument; fortunately, however, this is very rare. (More often, it is inconsistently maintained; the person who makes such a claim shows through other judgments and through actions that he or she really does value life in spite of its pains and hardships.)

At this point a natural response for some is to say, ‘No I don't say that a good God should not have created any system of nature at all; what I do say is that such a God would have created a better system than this one, producing a far more favorable balance of pleasure over pain, of happiness over misery’. We need to see, then, why such a ‘better system of nature’ is not a possibility of which we can have any real grasp. Our failure to grasp such a thing is not a matter of mere ignorance, comparable to our lack of information about some as‐yet‐undiscovered species of insect. This is a fundamental ignorance, and one of the reasons it is so can be found in the phenomenon known as ‘fine tuning’. It is a remarkable fact, widely acknowledged in the scientific community, that many of the fundamental constants of nature are balanced as if on a knife's edge, within an extremely narrow range that is essential for the existence of life as we know it.23

That this is so is not seriously controversial. What is controversial is what should be made of this fact. Many have argued that such a remarkable combination of improbabilities points to the existence and activity of Something or Someone that has deliberately selected, from all the possible values of these physical constants, the precise combination that makes possible the existence of sentient and rational beings. Needless to say, there is great resistance to this suggestion, but for our present purpose we have no need to take sides in this controversy. What is important is that, so far as we can tell, there very likely could not be a universe with significantly different natural constants than this, in which intelligent observers could live and raise philosophical questions about its existence. True, we cannot absolutely rule out that some combination of values very different from those in our own universe would make possible some form of life which we are at present unable to imagine. Nor can we rule out absolutely the possibility of a universe even more radically different from our own, in which the fundamental constituents and forces, and the laws of nature, are completely unlike anything presently within our ken. But anyone who would base an argument on such possibilities as these is whistling in a darkness so profound that no light is likely ever to shine into it. (p.298)

VI. A natural order theodicy

It is time to draw the threads together and to present formally the theodicy of natural evil that emerges from our deliberations. This theodicy will be termed the Natural Order Theodicy in virtue of its central theme; it can be stated in four propositions.

  1. 1. The actual universe is a complex, multi‐leveled natural world, containing creatures that are sentient as well as some that are intelligent. The world has developed to its present state through a complex evolutionary process, and enjoys a considerable amount of autonomy in its functioning.

  2. 2. The universe so constituted makes possible a large amount of good, both in the order and beauty of the physical universe and in the development and flourishing of a myriad of living creatures. It also unavoidably contains a great deal of suffering and death.

  3. 3. There is no good reason for us to suppose that some alternative order of nature, capable of being created by an all‐powerful God, would surpass the present universe in its potentiality for good or in its balance of good versus evil.

  4. 4. In virtue of 1–3, it is good that God has created this universe; there is no basis for holding God morally at fault for doing so, or for supposing that a perfectly good Creator would have acted differently.

It is hardly to be expected, of course, that the bare statement of these propositions should by itself be sufficient to produce conviction. But taken together with the considerations in the preceding pages, along with additional supporting information drawn from the reader's background knowledge, I submit that this constitutes a plausible and satisfying theodicy with respect to God's permission of the kinds of natural evil that are inherent in the existence of a universe such as this one.

Clearly many things could be said by way of objection to this theodicy, but space does not allow for most of them to be addressed here. I close by considering one protest that seems, on the face of it, to constitute an emphatic rejection of this and of any other attempt to justify the existence of the world's natural evils. The protest comes from Annie Dillard, author of Pilgrim at Tinker Creek. She writes, ‘Any three‐year‐old can see how unsatisfactory and clumsy is this whole business of reproducing and dying by the billions. We have not yet encountered any god who is as merciful as a man who flicks a beetle over on its feet. There is not a people in the world who behaves as badly as praying mantises….The universe that (p.299) suckled us is a monster that does not care if we live or die’.24 When I first read this, I wanted to say, ‘Do you really hate your life that much?’ Dillard's words seem to be saying that it is a bad thing that a world such as this should exist; it would have been far better that the creator (if any) should have had second thoughts, and refrained from creating it. But assuredly Dillard herself, and the other human beings she cares about, and for that matter all of us, are the products of precisely the world‐system and the evolutionary process she deplores. To wish that none of this had existed, is to wish that neither you yourself nor other people whom you love had ever existed—and that is not an easy thing to wish for, unless you are already desperately unhappy. Rejecting the world in this way may not be impossible, but there is a steep price to be paid for the rejection, a price few of us (I hope) will be willing to pay.

Taken in context, though, this is not really what Dillard means to be saying. In the words quoted she is expressing her shock that the world of nature, produced by an evolutionary process, is so often sharply at odds with human values and emotions. But this is not, in the end, a reason to reject such a world; more likely it is the emotions that are misdirected, when applied to natural occurrences such as predation. She writes, ‘Although it is true that we are moral creatures in an amoral world, the world's amorality does not make it a monster. Rather, I am the freak. Perhaps I don't need a lobotomy, but I could use some calming down, and the creek is just the place for it. I must go down to the creek again’ (1975, 182).

The non‐moral character of the evolutionary process, so vividly portrayed by Annie Dillard, poses an important question for theodicy. The point is well captured in a poem by Robert Frost. He begins by describing a grisly little scene in which a white spider sits on a white flower holding up a white moth it has trapped and killed. He then asks,

  • What brought the kindred spider to that height,
  • Then steered the white moth thither in the night?

It may occur to us—and I feel certain it occurred to Frost—that there are other sorts of events in nature that are far more disturbing than the capture of a moth by a spider. Consider an episode from a television program about the wolves in Yellowstone National Park.25 An invading troupe of wolves had moved into the Lamar Valley and were contesting it with the Slough Creek pack that was in occupancy. After an initial skirmish, the invaders coolly and deliberately set up a blockade around the dens in which the Slough Creek pack was raising its cubs. The (p.300) resident wolves were too few and too weak to break the blockade, and were unable to reach the cubs in order to feed them. After two weeks every one of the cubs had perished of starvation, whereupon the invaders withdrew from the scene.

Now return to Frost's question: What—or Who—was responsible for this gruesome set of circumstances? For both Augustinians and Molinists, the answer is entirely clear: this particular sequence of events, like every other, is the consequence of God's intentional, specific, and minutely detailed choice of a world to make actual. As Alfred J. Freddoso so eloquently states the matter,

God, the divine artisan, freely and knowingly plans, orders, and provides for all the effects that constitute His artifact, the created universe with its entire history, and executes His chosen plan by playing an active causal role sufficient to ensure its exact realization. Since God is the perfect artisan, not even the most trivial details escape His providential decree.26

God, then, specifically planned and intended27 the spider's capture of the moth, the blockade and starvation of the wolf cubs, and many, many other such ‘trivial details’. Nothing could be clearer than this, however chilling the consequences as we contemplate the fate of the wolf cubs. But now let's look at Frost's answer to his own question, when he asks for the explanation of these disturbing events:

  • What but design of darkness to appall?—

And we may well be appalled; I think we will be unwilling to accept, but also will find it difficult to contest, Frost's description of divine providence in such a case as a ‘design of darkness’. But Frost's conclusion is qualified: his closing couplet in full goes like this:

  • What but design of darkness to appall?—
  • If design govern in a thing so small. (1969, 302)

If design govern—but suppose it doesn't? Suppose, that is to say, that the endless contest of spider and moth, of predator and prey, is the product of an evolutionary process which is unintelligent and without intrinsic (p.301) purpose, even though purposed by divine wisdom to perform the function of engendering a rich array of biological life? That of course is what open theism takes to be the case: God has planned and brought into being the universe with all of its inherent laws and structure, and has permitted the natural course of events to proceed, but God has not specifically ‘planned, ordered, provided for, and ensured’ each and every event that takes place, including the capture of the moth, the starvation of the wolf cubs, and a great many even more unsavory events that disfigure human history. God, indeed, is deeply grieved at many of these events, yet in his gracious wisdom he grants to the creation the degree of autonomy needed to act according to its own nature, and refrains from frequent, intrusive intervention into the course of earthly events. And isn't this a better way for us to understand the situation? Not better for the wolf cubs, to be sure; nothing we can say is going to help them in the slightest. But the more directly God is said to be involved in planning the tragic events of this world and ensuring their occurrence, the more strongly we are led to question whether such activities are indeed compatible with the wisdom and love of God. It is just for this reason that many of us have concluded that Augustinianism has no hope of a tenable solution for the problem of evil. For Molinism God's control is less complete, but only slightly so—and the advantage becomes miniscule in dealing with natural evil, where presumably the moral freedom of creatures is not a significant consideration. For Molinism as for Augustinianism it is still very much the case, as Freddoso rightly says, that God plans, orders, provides for, and ensures the occurrence of each instance of evil and suffering that takes place. Open theism, on the other hand, denies this, and that is an important advantage for this view over its rivals. We may all the same be appalled at some of these events, even as we rejoice at the wonder and beauty of the creation overall. But there is no ‘design of darkness’, and no reason in all this to question the wisdom or the goodness of the Creator. But he is not a tame God, and he has not given us a tame world.28


Bibliography references:

Alston, William (1991) ‘The Inductive Argument from Evil and the Human Cognitive Condition’, Philosophical Perspectives, 5, 29–67. Reprinted in Daniel Howard‐Snyder (ed.), The Evidential Argument from Evil (Bloomington: Indiana University Press, 1996), 97–125. (p.302)

Basinger, David (1986) ‘Middle Knowledge and Classical Christian Thought’, Religious Studies, 22, 407–22.

Beecher, Henry Ward (1885) Evolution and Religion (New York: Fords, Howard and Hulbert).

Bergmann, Michael (2009) ‘Skeptical Theism and the Problem of Evil’, in Thomas P. Flint and Michael C. Rea (eds), The Oxford Handbook of Philosophical Theology (Oxford: Oxford University Press), 374–99.

Darwin, Charles, The Origin of Species, A Facsimile of the First Edition (Cambridge: Harvard University Press, 1954; originally published 1859).

Dillard, Annie (1975) Pilgrim at Tinker Creek (New York: Bantam Books).

Farrer, Austin (1966) Love Almighty and Ills Unlimited (London: Collins).

Freddoso, Alfred J. (1988) ‘Introduction’, to Luis de Molina, On Divine Foreknowledge (Part IV of the Concordia), tr. Alfred J. Freddoso (Ithaca, NY: Cornell University Press), 1–81.

Frost, Robert (1969) ‘Design’, in Edward Connery Lathem (ed.), The Poetry of Robert Frost (New York: Holt, Rinehart, and Winston).

Gonzalez, Guillermo and Jay W. Richards (2004) The Privileged Planet: How Our Place in the Cosmos is Designed for Discovery (Washington, DC: Regnery Publishing).

Gregory of Nyssa, Apologetic Treatise on the Hexaemeron, in J. P. Migne (ed.), Patrologia Graeca, vol. 44, col. 72.

Hasker, William (1989) God, Time, and Knowledge (Ithaca, NY: Cornell University Press).

—— (2004) ‘The Sceptical Solution to the Problem of Evil’, ch. 3 of William Hasker, Providence, Evil and the Openness of God (London: Routledge).

—— (2008) The Triumph of God Over Evil: Theodicy for a World of Suffering (Downers Grove, IL: InterVarsity Press).

—— (2009) ‘Why Simple Foreknowledge is Still Useless (in spite of David Hunt and Alex Pruss)’, Journal of the Evangelical Theological Society, 59, 537–44.

—— (2010) ‘All Too Skeptical Theism’, International Journal for the Philosophy of Religion, 68, 15–29.

Keller, James (2007) Problems of Evil and the Power of God (Aldershot, UK and Burlington, VT: Ashgate).

Public Broadcasting System documentary, ‘In the Valley of the Wolves’, first broadcast on November 4, 2007.

Rees, Martin (2000) Just Six Numbers: The Deep Forces That Shape the Universe (New York: Basic Books).

Rogers, Katherin (2004) ‘Augustine's Compatibilism’, Religious Studies, 40, 415–35.

Sanders, John (1997) ‘Why Simple Foreknowledge Offers No More Providential Control than the Openness of God’, Faith and Philosophy, 14, 26–40.


(1) See William Hasker (2004 and 2008, 180–7).

(2) From now on I will sometimes omit the reference to evils averted, but this should always be understood as a relevant possibility.

(3) The ideas developed here were suggested by some remarks of James Keller (2007, ch. 5).

(4) There will be an exception to this in those cases, if any, in which it can be shown that the divine policy, to be effective, must admit of no exceptions. This however will have to be argued for each such policy, and in many cases this will prove difficult or impossible to do.

(5) ‘Augustinianism’ serves as a non‐denominational label to refer to all varieties of theological determinism, including Calvinism and Thomism. Katherin Rogers (2004) has argued persuasively that Augustine was a compatibilist in all of his written work, including the early On Free Will.

(6) Only recently has it come to be realized that these two models offer no advantage whatsoever for divine providence, as compared with open theism which rejects comprehensive divine knowledge of the actual future. For argument, see David Basinger (1986); Hasker (1989, ch. 3); and John Sanders (1997). For comment on two recent attempts to circumvent these arguments, see Hasker (2009).

(7) For some risk‐taking models, such as simple foreknowledge, ‘prior’ here must be understood as logical or explanatory priority rather than temporal priority.

(8) I do not in fact believe that this criticism applies to Hick; he does not, as I understand him, postulate a specific divine decision to permit, or to prevent, suffering in individual cases. Hick is a general‐policy theodicist, not a specific‐benefit theodicist.

(9) As understood here, a defense, unlike a theodicy, does not meet the argument from evil by proposing goods made possible, or evils averted, by God's permitting some instance of evil. Rather the defense attempts to block the argument from evil without making any such proposal. (See for example Plantinga's ‘Free Will Defense’ against the logical problem of moral evil.)

(10) Alston (1991); reprinted in Daniel Howard‐Snyder (1996).

(11) Alston (1991, 58); in Howard‐Synder (1996, 119).

(12) Michael Bergmann (2009). For my discussion of Bergmann's views, see Hasker (2010).

(13) It is fair to say that Alston, and some other advocates of skeptical theism, probably did not have in mind such an extreme skeptical view, and may not have accepted it had it been proposed to them. For the reason why this strong form of skepticism seems inevitable, see Bergmann's and my articles.

(14) Much of the material in the remainder of this chapter is taken from Hasker (2008, ch. 5). It is used by permission of InterVarsity Press PO Box 1400 Downers Grove, IL 60515. www.ivpress.com.

(15) The atoms of modern science, of course, are not elementary objects but have a complex internal structure. The word ‘atoms’ is used here simply to represent the simplest elementary components of physical reality, whatever those components may in fact be.

(16) To avoid possible misunderstanding, let me state emphatically that I am not endorsing a conception of ‘moral and spiritual evolution’ in which the need for redemption is satisfied by the evolutionary progress of humankind. Human sin and degradation are not part of the divine intent in creation, and they cannot and will not be remedied by the self‐perfecting of that creation.

(17) Beecher (1885, 114). I am indebted to Michael Murray for this quotation.

(18) Apologetic Treatise on the Hexaemeron, in Patrologia Graeca, edited by J. P. Migne, vol. 44, col. 72. My thanks to Ernan McMullin for supplying this reference.

(19) Charles Darwin, The Origin of Species, 490. I believe the point made by Beecher, Gregory, and Darwin can stand on its own merits, but there is an interesting ad hominem aspect to this situation. Theistic believers, whatever their initial predispositions, will find it hard to contest the point, once they are convinced by the evidence that God has in fact created the world through an evolutionary process. And non‐theists are likely to find it awkward to insist that it would be much better for the world to have been created by a complex series of specific divine actions, rather than allowed to evolve naturally!

(20) For further elaboration, see Gonzalez and Richards (2004, ch. 3).

(21) Austin Farrer states it well: ‘The atomic world is run by the atomic energies. And it is a manifest absurdity to suppose that they can consider the whole, or even consider one another. They cannot consider anything; the principle of their action is simply that it should go on discharging itself’ (1966, 56).

(22) From the Random House Dictionary of the English Language.

(23) For a readable and authoritative presentation of the data supporting this claim, see Martin Rees (2000, 22).

(24) Annie Dillard (1975, 180). I am indebted to Michael Murray for this quotation.

(25) ‘In the Valley of the Wolves’, first broadcast on PBS on November 4, 2007.

(26) Alfred J. Freddoso (1988, 3). Freddoso gives this as a statement of the traditional view of divine providence, a view which is held in common by Thomism, Calvinism, and Molinism.

(27) Both Augustinians and Molinists tend to deny this, claiming that the evil events are ‘permitted’ by God but not ‘intended’. In the light of Freddoso's statement of the traditional view of providence, this claim appears to be incoherent. What artisan creates an artifact ‘unintentionally’? To be sure, with any human artisan there will be some features of the artifact that escape the artisan's attention and are not intended. But this is specifically denied to be the case for the perfect divine artisan.

(28) I want to express my thanks to John Martin Fischer and to the members of his seminar on God and Free Will at the University of California at Riverside for their stimulating discussion of an earlier version of this material.