A Holistic Response to the Problem of Evil

Faith and Philosophy: Journal of the Society of Christian Faith and Philosophy: Journal of the Society of Christian

Philosophers Philosophers

Volume 39 Issue 4 Article 5

10-1-2022

A Holistic Response to the Problem of Evil A Holistic Response to the Problem of Evil

Jonathan Fuqua

Conception Seminary College

, [email protected]

Follow this and additional works at: https://place.asburyseminary.edu/faithandphilosophy

Recommended Citation Recommended Citation

Fuqua, Jonathan (2022) "A Holistic Response to the Problem of Evil,"

Faith and Philosophy: Journal of the

Society of Christian Philosophers

: Vol. 39: Iss. 4, Article 5.

DOI: 10.37977/faithphil.2022.39.4.5

Available at: https://place.asburyseminary.edu/faithandphilosophy/vol39/iss4/5

This Article is brought to you for free and open access by the Journals at ePLACE: preserving, learning, and creative

exchange. It has been accepted for inclusion in Faith and Philosophy: Journal of the Society of Christian

Philosophers by an authorized editor of ePLACE: preserving, learning, and creative exchange.

FAITH AND PHILOSOPHY Vol. 39 No. 4 October 2022

doi: 10.37977/faithphil.2022.39.4.5

pp.593–613

A HOLISTIC RESPONSE TO THE PROBLEM OF EVIL

Jonathan Fuqua

The four standard theistic responses to the evidential problem of evil are

theodicy, Reformed Epistemology, natural theology, and skeptical theism.

It’s somewhat common for theists to combine Reformed Epistemology and

skeptical theism or natural theology and theodicy. An insufciently appre-

ciated possibility is that of combining all four of these positions into a more

holistic response to the evidential problem of evil. The chief hurdle to doing

this is that it seems that skeptical theism isn’t compatible with either natural

theology or theodicy. This rst appearance, however, is misleading. And, in-

terestingly, certain theists have implicitly put forth a holistic response to the

problem of evil. In this paper, I sketch out how one can combine all four of

the standard theistic responses to the evidential problem of evil so as to yield

a holistic response to the problem of evil. The focus will be on reconciling

skeptical theism with natural theology and theodicy.

1. Introduction

Proponents of evidential arguments from evil argue that the facts of evil

are good evidence against the existence of God.

There are four standard

theistic responses to such arguments that are prevalent in the contempo-

rary literature on the problem of evil. First, some theists give theodicies,

stories intended to show how God can be justied in allowing the evil

In my view, the most potent version of the evidential argument from evil is to be found

in Draper, “Pain and Pleasure.” It is commonly believed that Pike’s “Hume on Evil” and

Plantinga’s The Nature of Necessity have defused the logical problem of evil. Draper, himself

a non-theist, says that “logical arguments from evil . . . are rejected by the vast majority

of contemporary philosophers of religion” and that “most philosophers nowadays believe”

that the logical incompatibility of God and evil “cannot be shown” (see Draper, “The Prob-

lem of Evil,” 335). For an overview of the logical problem of evil, see Howard-Snyder, “The

Logical Problem of Evil.” It must be noted, however, that the logical problem of evil has not

completely gone away. For a new presentation and defense of the logical problem of evil, see

Schellenberg, “A New Logical Problem of Evil,” and Sterba, Is a Good God Logically Possible?

For a response to Sterba, see Feser, “The Thomistic Dissolution of the Logical Problem of

Evil.”

Faith and Philosophy594

appealed to in the evidential argument under consideration.

A second

response is Reformed Epistemology, according to which theists enjoy, or

can enjoy, non-inferential justication for belief in God such that belief in

God can be justied even if theism might be unable to gain any positive

justicatory status from its ability to explain some body of data.

A third

theistic response is that of the natural theologian, who argues that any

negative evidence against theism that is provided by the data of evil is

outweighed by the positive evidence for theism to which the premises

of natural theological arguments appeal.

A fourth response is skeptical

theism, the core claim of which is that we are not in a position to say how

likely the facts of evil are given theism.

The thesis of this paper is that these four responses are not inconsistent

with each other. In the literature on the problem of evil, it is common to

take one of these responses and develop it in some level of detail. This

focused approach has the benet of allowing us to see just how helpful

(or not) a particular response to evidential arguments from evil can be.

It has the dialectical demerit, however, of tempting us to forget that in

actual practice many theists combine two or more of the four standard re-

sponses. It also has the demerit of hindering us from seeing just how bene-

cial it can be to combine multiple responses into a more holistic response

to the problem of evil. Theists thus typically employ a subpar strategy in

their published responses to evidential arguments from evil by focusing

on just one of the four standard responses. In this paper, I sketch a holistic

response to the problem of evil.

Many theists in the past, as well as today, have employed two or more

of the four standard theistic responses. According to Rudavsky, skeptical

theism has roots in Isaiah, Job, Paul, Plotinus, John Scotus Eriugena, Mai-

monides, Aquinas, and Descartes.

But, of course, some of these also did

See Murray, “Theodicy,” for an overview of the project of theodicy. The most inuential

big picture theodicy projects in the contemporary literature are probably those of Hick’s Evil

and the God of Love, Swinburne’s Providence and the Problem of Evil, and Stump’s Wandering in

Darkness.

For an overview of Reformed Epistemology, see Bergmann, “Rational Religious Belief

Without Argument” and “Reformed Epistemology.” See also Moon, “Recent Work in Re-

formed Epistemology.” Plantinga’s Warranted Christian Belief is of course the most extensive,

inuential, and important presentation of the Reformed Epistemology project. For a recent

defense of Plantinga’s religious epistemology, see McNabb, “Warranted Religion.”

An example of the natural theology response is provided by Swinburne’s The Existence of

God. For more cutting-edge natural theology, see Craig and Moreland, The Blackwell Compan-

ion to Natural Theology; Feser, Five Proofs of the Existence of God; and Walls and Dougherty, Two

Dozen (or so) Arguments for God. For an overview of the terrain, see Baker-Hytch, “Natural

Theology and Religious Belief.”

For overviews of skeptical theism, see Bergmann, “Skeptical Theism and the Problem

of Evil,” McBrayer, “Skeptical Theism,” and Dougherty, “Skeptical Theism.” My character-

ization of skeptical theism follows Draper’s characterization of the core claim of skeptical

theism; see Draper, “The Skeptical Theist,” 176.

Rudavsky, “A Brief History of Skeptical Responses to Evil.” For an in-depth look at Des-

cartes’s skeptical theism, see Robinson, “Descartes’s Sceptical Theism.”

A HOLISTIC RESPONSE TO EVIL 595

natural theology, chief among them Aquinas and Descartes. Aquinas also

did theodicy

and, if Plantinga reads him correctly, also accepted the Re-

formed Epistemological view that it is possible to have non- inferentially

justied belief that God exists. So, it may well turn out that Aquinas him-

self utilized all four of the standard responses. And it certainly looks to me

as if Plantinga has made use of Reformed Epistemology, natural theology,

theodicy, and skeptical theism.

In short, I think some theists have been

implicitly employing a holistic or cumulative case approach; it would be

good to know if they are onto something. Of course, such a holistic re-

sponse can work only if each individual response is successful. Unfortu-

nately, I cannot defend each response here; my goal is simply to argue for

their compatibility. My task here is a second-order one: to argue for the

feasibility of a holistic response to the problem of evil.

2. Reformed Epistemology, Natural Theology, and Theodicy

In this section, I argue for the compatibility of Reformed Epistemology,

natural theology, and theodicy. Reformed Epistemology is the view that

theistic belief is or can be non-inferentially justied, that, as Bergmann

puts it, “belief in God—like the belief that I had orange juice for breakfast

or the belief that there’s a ball in front of me—can be properly basic.”

Natural theology is typically thought of as the activity of producing ar-

guments for God’s existence.

Natural theologians do assume—and seek

to show—that there is inferential justication or propositional evidence

for theism. Natural theology qua activity is compatible with Reformed

Epistemology derivatively, in the sense that the assumption which natu-

ral theologians make—that there is propositional evidence for theism—is

consistent with Reformed Epistemology.

Reformed Epistemology and natural theology are sometimes thought

of as rivals.

Reformed Epistemologists are sometimes suspicious of the

thought that theism should be understood as an explanatory hypothesis,

See Stump, “Aquinas on the Sufferings of Job.”

See the relevant chapters in Plantinga’s books The Nature of Necessity and Warranted

Christian Belief and his papers “On Being Evidentially Challenged” and “Supralapsarianism,

or ‘O Felix Culpa.’”

Thanks to a referee for some helpful comments on how to think about the nature of the

project of this paper.

Bergmann, “Rational Religious Belief Without Argument,” 538.

Thus:

Natural theology is the practice of philosophically reecting on the existence and

nature of God independent of real or apparent divine revelation or scripture. Tra-

ditionally, natural theology involves weighing arguments for and against God’s ex-

istence, and it is contrasted with revealed theology, which may be carried out within

the context of ostensible revelation or scripture. (Taliaferro, “The Project of Natural

Theology,” 1)

This is probably due, at least in part, to Plantinga’s paper, “The Reformed Objection to

Natural Theology.”

Faith and Philosophy596

a suspicion that seems uncongenial to natural theology.

When Reformed

Epistemologists voice this suspicion, they typically also argue that theistic

belief can be properly basic whether or not it enjoys the kind of proposi-

tional evidence championed by the natural theologian.

This latter notion

was, in fact, one of the main points of Plantinga’s 1967 book, God and Other

Minds, in which he argued that belief in God, like belief in other minds,

can be rational even if it can’t be established by philosophical proofs.

Additionally, a natural theologian may, if she likes, endorse the “hyperev-

identialist” view that if a theist has justied theistic belief, that is because

she has inferential support for that belief.

Such hyperevidentialism is

obviously incompatible with Reformed Epistemology.

Fortunately, Reformed Epistemology and natural theology are not, con-

trary to these initial appearances, incompatible. First, suspicion, or even

denial, that theism is an explanatory hypothesis is consistent with afrm-

ing the possibility of inferential support for theistic belief. This is an in-

stance of the more general principle that one can have non-inferential as

well as inferential support for a proposition that one does not construe

as an explanatory hypothesis. I might enjoy non-inferential justication

for the view that nonconsensual sex is wrong and also have an argument

that it is wrong without thereby understanding that proposition to be an

explanatory hypothesis. Similarly, rejection of the idea that theism is an

explanatory hypothesis does not, by itself, entail rejection of the natural

theology view that theism can enjoy inferential justicatory support.

Second, a natural theologian needn’t endorse hyperevidentialism. It’s

entirely possible for a proposition to enjoy inferential and non- inferential

support, to have both propositional and non-propositional evidence in

its favor. There is no reason a natural theologian cannot say that theism

has both kinds of justicatory support. Plantinga and Swinburne, the two

most prominent contemporary advocates of Reformed Epistemology and

natural theology, respectively, apparently concur. Plantinga, after all, is

known not only for developing a proper functionalist version of Reformed

Epistemology, but also for developing a modal version of the ontological

argument. Plantinga, additionally, is on record as saying that the ne-

tuning argument offers “mild support” for theism.

He has also, of course,

developed about two dozen arguments for theism in his “Two Dozen (or

so) Theistic Arguments.”

Plantinga has been careful to point out that the

For a critical discussion of the idea that God’s existence should be treated as an explan-

atory hypothesis, see Holley, “On Treating God’s Existence as an Explanatory Hypothesis.”

The most fully developed Reformed Epistemology currently on offer can be found in

Plantinga, Warranted Christian Belief.

See Plantinga, God and Other Minds.

This term comes from Dougherty and Tweedt, “Religious Epistemology,” 548ff.

Plantinga, Where the Conict Really Lies, 224.

See Plantinga, “Two Dozen (or so) Theistic Arguments.” For a recent collection of essays

on Plantinga’s two dozen (or so) theistic arguments, see Walls and Dougherty, Two Dozen (or

so) Arguments for God.

A HOLISTIC RESPONSE TO EVIL 597

proper basicality of theistic belief is fully compatible with its also enjoying

support from theistic arguments, and in the preface to a revised edition

of God and Other Minds, written in 1990, he takes a much more optimistic

attitude toward natural theology than he did in 1967.

Swinburne also

argues that it is possible for religious believers to enjoy non-inferential

justication for theistic belief, suggesting that natural theology is, how-

ever, needed for those believers who are aware of atheistic arguments or

who have “no experiences of the apparent presence of God.”

Like Plant-

inga and Swinburne, I conclude that there is no inconsistency between

Reformed Epistemology and natural theology.

This compatibility strengthens the theist’s response to the problem of

evil. Assume that Reformed Epistemology is correct and that belief in God

can be non-inferentially justied. Assume also that there isn’t any, or any

signicant, propositional evidence for theism. Assume further that some

evidential argument from evil really does offer good propositional evidence

against theism. The Reformed Epistemologist will likely follow Plantinga

here and argue that theism can be both improbable on our propositional

evidence and justied on our non-propositional evidence.

Now, let’s

make some contrary assumptions: that there is no good non- propositional

evidence for theism, that some evidential argument from evil really does

offer good propositional evidence against theism, but that there are some

good natural theological arguments for God’s existence. In this case, while

theism might not be probable with respect to our non-propositional evi-

dence, and improbable with respect to one part of our propositional evi-

dence, the natural theologian will say that theism is more probable than

not (or probable enough to be rationally believed) given our total proposi-

tional evidence. Now, if Reformed Epistemology and natural theology are

both correct, then the theist can say that theism has good non-inferential

and inferential support, thus responding to the problem of evil by citing

two lines of justicatory support rather than just one.

This dual response to the problem of evil would be even stronger

if the theist could add theodicy to the mix. Let’s assume that there are

some good theodicies so that the only question is whether theodicy is

Plantinga, Where the Conict Really Lies, 179.

Swinburne, Faith and Reason, 90. Swinburne’s case for the possibility of noninferential

justication for theism proceeds on the basis of his Principle of Credulity, which says that if

it seems to S that p, then probably p. Applied to theistic belief, we get the result that theistic

seemings produced by religious experiences give us good evidence for theism. See Swin-

burne, The Existence of God, 292–327. For a sustained epistemological defense of this sort of

view, using the resources of phenomenal conservatism, see Tucker, “Phenomenal Conserva-

tism and Evidentialism in Religious Epistemology,” and McCain, “Evidence and Religious

Belief.”

For more sustained arguments for this, see Sudduth’s articles “Reformed Epistemology

and Christian Apologetics” and “Revisiting the ‘Reformed Objection’ to Natural Theology.”

Evans, Natural Signs and Knowledge of God, contains a developed religious epistemology

which explicitly includes roles for both Reformed Epistemology and natural theology.

See Plantinga, “Reply to Tooley’s Opening Statement,” 154–155.

Faith and Philosophy598

compatible with Reformed Epistemology and skeptical theism. Follow-

ing Scott Coley, let’s make a distinction between weak theodicy (typically

called “defense”) and strong theodicy (typically called “theodicy”).

A weak theodicy, or defense, is an explanation for some evil E which appeals

to some God-justifying reason R and says that, for all we know, R holds

and would justify God (if God exists) in permitting E. A strong theodicy

says that R does hold and that it justies God (if God exists) in permit-

ting E. A strong theodicist assumes that we have an ability to sometimes

identify those goods which would justify God in allowing certain evils.

Obviously, this is not incompatible with either the claim that (i) we have

non-inferential justication for belief in God or (ii) we have inferential jus-

tication for belief in God. The weak theodicist makes a weaker assump-

tion, namely that we have an ability to identify those goods which, if they

obtain, would justify God in permitting certain evils. Clearly, this claim

doesn’t conict with Reformed Epistemology or natural theology either.

If there is a good strong theodicy out there, then the theist can argue that

theism is non-inferentially justied and enjoys propositional support from

both natural theology and theodicy. And if there is a good weak theodicy

out there, the theist can say that theism enjoys non-inferential justication

and inferential justication (from natural theology), and that, owing to

weak theodicy, we can’t say whether E is propositional evidence against

theism or not. So, combining Reformed Epistemology with natural theol-

ogy and theodicy gives the theist a stronger response to the problem of

evil than she would have with any one or two of these alone. Now, I’m not

saying that there are any good theodicies. All I’m saying is that these three

options are compatible, and that if they all work, the theist who combines

them has a stronger response to the problem of evil than the theist who

only uses one or two of these strategies.

3. Skeptical Theism Plus Natural Theology Plus Theodicy

You can endorse everything I’ve said so far without thinking that the the-

ist’s overall epistemic situation is a very good one. For you might think

that theism gets some support from non-inferential sources, some from

natural theology, and some from theodicy, but that the total evidence still

favors some non-theistic hypothesis over theism. For instance, it might

be that there are some evils for which we have no theodicy and that the

negative evidence provided by these evils swamp the positive epistemic

support provided to theism by Reformed Epistemology, natural theology,

See Coley, “Skeptical Theism is Incompatible with Theodicy.”

It is interesting to note that Reformed Epistemologists and natural theologians have

both combined their favored approach to the justication of theistic belief with theodicy.

Plantinga, for instance, has defended a theodicy in his 2004 paper “Supralapsarianism, or

‘O Felix Culpa’”; Swinburne has done the same in his 1998 book Providence and the Problem

of Evil. Plantinga and Swinburne, then, seem to be committed to the compossibility of Re-

formed Epistemology, natural theology, and theodicy.

A HOLISTIC RESPONSE TO EVIL 599

and theodicy. Skeptical theism, if applicable to the arguments from evil

that appeal to these evils, could help with this problem.

One potential problem for skeptical theism, however, is that it looks to

be incompatible with certain natural theological arguments. Wilks argues,

for instance, that skeptical theism assumes that we are unable to say that

any apparent evils are all-things-considered evil and hence not justied

by some greater good.

But if we cannot say this about evils, then we

cannot say this about goods. And if we cannot say that certain goods are

all-things-considered good, then we cannot say whether God would bring

about such goods. Wilks points out that this kind of skepticism would un-

dermine the design argument, for on this kind of skepticism we wouldn’t

be able to say whether the goods produced by an orderly universe would

be the sorts of things God would want to bring about.

A second potential problem for skeptical theism is that it looks to be

incompatible with theodicy. Coley explains why: any possible good G that

a theodicist might appeal to as an explanation for why God allows some

evil E will fall prey to the following skeptical theist claim: there might be

a good G* that is much greater than G and which entails the prevention of

E. If a theodicist is giving a strong theodicy, and she’s saying she knows

of a good G that really would justify God’s permission of E, then skepti-

cal theism will tell her that G might be outweighed by G*, which entails

the prevention of E. And if this theodicist is giving a weak theodicy (or a

defense), and she’s saying she knows of a possible good G which would

justify God’s permission of E, then—says Coley—she’s just doing what

the skeptical theist does, and her position reduces to that of the skeptical

theist. In short, weak theodicy, on Coley’s telling, is just a version of skep-

tical theism, which happens to be incompatible with strong theodicy.

So far, then, it looks like skeptical theism is incompatible with both nat-

ural theology and theodicy. The basic reason for this seeming incompati-

bility is that natural theology and theodicy appear to require the following

assumption: we are in a position to know (or at least justiedly believe),

at least in part, which things God would likely allow or produce in the

world. Call this the “likelihood assumption.”

Skeptical theism, however,

Wilks, “The Structure of the Contemporary Debate on the Problem of Evil.”

See Wilks, “The Structure of the Contemporary Debate on the Problem of Evil.” Wilks

doesn’t make his case on this point in terms of goods and evils being all-things-considered

good or evil. This is how Bergmann reconstructs Wilks’s argument, and in my presentation

of Wilks I follow Bergmann’s reconstruction. See Bergmann, “Skeptical Theism and the Prob-

lem of Evil,” 398.

The natural theologian, making the likelihood assumption, argues that certain goods

are quite likely on theism and hence that their existence is evidence for theism. The theo-

dicist, making the likelihood assumption, argues that certain goods which God is likely to

allow or produce logically require certain evils or their permission, and hence that the exis-

tence of these evils is not unlikely on theism (and may in fact even be likely on theism). See

Anderson, “Skeptical Theism and Value Judgments,” for a good argument that the distinc-

tion between an evil and its permission is (sometimes, at least) important. The basic point is

that it may be the case that an occurrence of an evil is all-things-considered bad, but that its

Faith and Philosophy600

rejects the likelihood assumption and tells us that we are not in a position

to know (or justiedly believe) what God would allow or bring about in

the world, thus undermining natural theology and theodicy. As Cullison

puts it, skeptical theism seems to imply that “any attempt to reason about

what God would do is in jeopardy.”

Following Cullison, let’s call this the

“Reasoning about God Problem,” or the “RAG problem” for short. Given

the RAG problem, it looks like the theist will have to choose between two

incompatible responses to the problem of evil, one which includes natural

theology and theodicy, and has room for Reformed Epistemology, and the

skeptical theist response, which also has room for Reformed Epistemol-

ogy but not for theodicy and natural theology. In the remainder of this sec-

tion, I will outline two possible responses to this dilemma, one concessive

and one non-concessive.

3.1. The Concessive Response

The concessive response concedes that skeptical theism and theodicy

are incompatible, but denies that skeptical theism is incompatible with

natural theology. A theist who adopts this response will be able to utilize

the resources of Reformed Epistemology, skeptical theism, and natural

theology, but not the resources of theodicy. The compatibility of skep-

tical theism, natural theology, and Reformed Epistemology falls short

of the claim I am ultimately aiming for here, but it’s an option worth

pausing to take note of.

As Wilks himself points out, skeptical theism does not make trouble

for all natural theological arguments.

The reason is that not all natural

theological arguments involve predicting that there is some good that God

would likely bring about and thus that theism is more likely given this

good. Some versions of some natural theological arguments, such as the

standard approach to the ne-tuning argument, do indeed involve making

likelihood assignments, i.e., saying that we can know or justiably believe

propositions about how likely some good or evil would be given theism.

Perhaps such arguments are out-of-bounds, given skeptical theism. Many

other natural theological arguments, however, don’t involve any likeli-

hood assignment and thus don’t require the likelihood assumption. Many

of Plantinga’s two dozen or so theistic arguments, for example, do not in-

volve any likelihood assignment at all. The ontological and cosmological

arguments don’t require any likelihood assignment, nor does the moral

argument. In general, the traditional, classical versions of the arguments

of natural theology are best thought of as metaphysical demonstrations of

the existence of a necessary being who serves as the ground of existence

rather than as abductive-style arguments to the effect that theism better

permission is not. It may be the case that God’s permission of an evil is necessary for some

outweighing good even if the occurrence of that evil is not.

Cullison, “Two New Versions of Skeptical Theism,” 250.

Wilks, “The Structure of the Contemporary Debate,” 317–318.

A HOLISTIC RESPONSE TO EVIL 601

predicts certain facts than rival hypotheses.

Even the contemporary

version of the ne-tuning argument can be articulated without presup-

posing the likelihood assumption. William Lane Craig’s version of that

argument, for example, involves positing three possible explanations for

the ne-tuning—chance, necessity, or design—and then ruling out chance

and necessity, leaving design as the only plausible, or even possible, ex-

planation standing.

I’m not here endorsing (or even articulating) any

such arguments; I’m simply noting that many of the traditional arguments

of natural theology look to be fully compatible with skeptical theism in

that they don’t require making predictions about God’s behavior and thus

don’t require the likelihood assumption.

This is a good place to point out that it’s not clear that natural theolog-

ical arguments which say that some good is more likely on theism than

on some nontheistic hypothesis, such as naturalism, really do require the

likelihood assumption. If they don’t, then probabilistic natural theology

may well be out of the woods. Poston argues that our inability to say

whether some evil is gratuitous or not in no way imperils our knowledge

that certain things have value.

Using Swinburne’s examples, Poston argues that we know, for in-

stance, that consciousness and freedom are valuable. Following Poston,

let H stand for “humanly free creatures exist,” T for theism, and N for nat-

uralism. Given the value of consciousness and freedom, we can say that

the likelihood of H given T is not too low. We can also say that the likeli-

hood of H given N is low.

Thus, we can say that H is evidence for theism

over and against naturalism. Poston does concede—to Beaudoin

—that

the goods of consciousness and freedom “might also realize a disastrous

consequence that lies beyond our comprehension.” However, it could also

be that these goods lead to “unconceived felicitous consequences.” In the

end, then, these “unconceived values wash out,” leaving us with our orig-

inal judgment that H is more likely on theism than on naturalism.

Poston’s claim that the “unconceived values wash out” is not special

pleading for the theist. Poston is making essentially the same move that

There are also moral arguments which don’t involve likelihood assignments. Wolter-

storff, for example, argues that there is no adequate secular grounding of human rights and

then suggests that, given the existence of human rights and the existence of a plausible the-

istic grounding of such rights, we have a reason to believe theism. See Wolterstorff, Justice,

361ff.

See Craig, “Design and the Anthropic Fine-Tuning of the Universe.”

See Poston, “Skeptical Theism within Reason.”

According to Poston:

Naturalism doesn’t predict the existence of humanly free creatures. Why? There is noth-

ing about the content of the naturalist hypothesis that predicts H. Value considerations per-

taining to H play no role whatsoever in naturalism’s ability to account for H. According

to naturalism, H is the result of blind processes working over millions of years. It’s very

surprising that H would be true given naturalism.” (“Skeptical Theism within Reason,” 319)

See Beaudoin, “Evil, the Human Cognitive Condition, and Natural Theology.”

Poston, “Skeptical Theism within Reason,” 321.

Faith and Philosophy602

Draper makes in his Bayesian argument from evil. One of the premises in

this argument is that naturalism predicts “the data of good and evil” better

than theism does.

Draper concedes that “we should not be able to expect

to discern potentially God-justifying reasons for allowing the evils we nd

in the world.”

However, he does not think that this undermines his ability

to say that the data of good and evil are more likely on naturalism than

on theism. Our situation is this: we know that (i) the world’s evils are bad

and thus that God would have a reason to prevent them, (ii) there might be

God-justifying reasons for the world’s evils unknown to us, and (iii) there

could be reasons unknown to us for God to not create a world containing

the data of good and evil. Draper argues that, in this situation, “it is the

known reasons that must break the tie”—i.e., that our knowledge of the

badness of the world’s evils still gives us a reason to think that those evils

are less likely on theism than on naturalism.

In other words, when I know

that there might be unknown God-justifying reasons for preventing some

evil E, and also that there might be unknown God-justifying reasons for

allowing E, I can still say that E is less likely on theism than naturalism—for

“it is the known reasons that break the tie.” Poston makes this move, essen-

tially, when he argues that our unconceived reasons against bringing about

humanly free creatures and our unconceived reasons for bringing about

humanly free creatures “wash out,” leaving us with our original judgment

that God has a good reason to bring about humanly free creatures.

3.2. Non-Concessive Responses

I want to now argue that there are ways of combining skeptical theism

with both natural theology and theodicy. The core claim

of skeptical the-

ism is the following:

Skeptical Theism: We are not in a position to say how likely the facts of evil

are given theism.

One possible reason for thinking that skeptical theism is true is simply

that we are beings with limited cognitive ability.

Given our cognitive

Draper, “Explanation and the Problem of Evil,” 72.

Draper, “Explanation and the Problem of Evil,” 79.

Draper, “Explanation and the Problem of Evil,” 77.

I’ve discussed Poston’s views in a section on concessive responses because it suited

my dialectical purposes to do so, but I should note that this does not mean that I think that

Poston himself would endorse the concessive response over the non-concessive response. As

a referee points out, Poston’s views in fact seem to fall into the non-concessive camp. I agree

and thank the referee for helping to clear up any confusion on this point.

As Bergmann notes, skeptical theism has a theistic component and a skeptical compo-

nent. A nontheist could embrace the skeptical component without embracing the theistic

component. So, weirdly, a nontheist can be a skeptical theist, of sorts. See Bergman, “Skepti-

cal Theism and the Problem of Evil,” 375.

For more on our cognitive limitations, and how those prevent us from being in a posi-

tion to see why God would allow the particular evils he does (if he exists), see Alston, “The

A HOLISTIC RESPONSE TO EVIL 603

limitations we have good reasons to be very humble when thinking about

the likelihood of some instance of evil given theism. Consider, in this vein,

Bergmann’s skeptical thesis, ST4:

ST4: We have no good reason for thinking that the total moral value or dis-

value we perceive in certain complex states of affairs accurately reects the

total moral value or disvalue they really have.

Call this “state of affairs skepticism.” Anyone who endorses state of affairs

skepticism has a good reason to think that she is in no position to afrm

the reliability of her seeming that some instance of evil E is not part of a

larger complex state of affairs which is overall good and which defeats E.

State of affairs skepticism seems quite plausible, at least in this respect:

our cognitive limitations are such that it is very difcult, perhaps even

impossible, for us say whether the value we seem to perceive in a complex

state of affairs accurately reects the total value in that state of affairs, or

in some larger, even more complex state of affairs in which the state of

affairs under consideration is a part. So, for example, I can see that S’s

having cancer is bad, but I can’t see that S’s having cancer isn’t part of a

larger complex state of affairs that is overall good and defeats the badness

of S’s having cancer. If I am aware of my own cognitive limitations on the

matter, then I have a defeater for any seeming to the effect that God’s al-

lowance of S’s having cancer is unlikely: if S’s having cancer is in fact part

of a larger complex state of affairs in which it is defeated by some good G,

and if God strongly desires that G be brought about, then it may be quite

likely that God would allow S’s having cancer.

It should be obvious, though still important to take note of, that state

of affairs skepticism is not incompatible with Reformed Epistemology’s

claim that we can have non-inferential justication for theistic belief, nor

with the natural theologian’s claim that we can have inferential justica-

tion for God’s existence—the ontological and cosmological arguments, for

example, are unaffected by the sort of skepticism at play in Bergmann’s

ST4. ST4 does, promisingly, seem to undercut arguments from evil that

appeal to tokens or instances of evil, for we cannot say whether any given

token of evil is or is not part of a larger complex state of affairs which is

overall very good and which defeats the token evil in question. But does

ST4 allow for theodicy?

If we make a distinction between tokens of evil and types of evil, we

might open a path to combining skeptical theism and theodicy by arguing

that we can be skeptical theists about arguments from evil that appeal

to tokens of evil while offering theodicies in response to arguments from

evil that appeal to types of evil. The basic idea is that while we may be

in a position to say which types of evil-requiring goods God would allow

Inductive Argument from Evil and the Human Cognitive Condition”; Howard-Snyder, “The

Argument from Inscrutable Evil”; and Poston, “Skeptical Theism Within Reason.”

Bergmann, “Skeptical Theism and the Problem of Evil,” 379.

Faith and Philosophy604

or produce in a world, we cannot say, of any token instance of evil, how

likely that particular token is given theism. So, we can make rough likeli-

hood assignments about types of evil, but not about tokens of evil. On this

approach, I may be able to offer a theodicy for animal suffering in general

without being able to explain why God allowed Rowe’s fawn to suffer

and die in that forest re. This approach opens the door to combining

skeptical theism with full-bore probabilistic natural theology: skepticism

about tokens can be combined with the non-skeptical thought that God

would likely allow or bring about certain types of goods in a world, such

as a world governed according to laws of nature, or a world with rational

creatures who can appreciate beauty, and so on.

This approach to combining skeptical theism with theodicy and natural

theology makes use of the following idea:

Token Skepticism: Having an explanation for a type of evil does not entail

having an explanation for a token of that type.

By combining token skepticism and skeptical theism we get a view we

can call “token skeptical theism.” As I read van Inwagen’s Gifford Lec-

tures on the problem of evil, he endorses token skepticism.

Someone

who embraces token skepticism is free to endorse or reject the idea that

God would allow gratuitous evil.

If a token skeptical theist embraces the

compatibility of theism and gratuitous evil, then she can say, in response

to an argument from evil that appeals to a particular instance of evil E,

that E may well, for all we know, be gratuitous. So, if E is Rowe’s fawn suf-

fering and dying in a forest re, this token skeptical theist can reply that

E’s possible gratuitousness is not incompatible with theism. And on the

assumption that gratuitous evil is compatible with theism, it will be very

difcult to say how likely E is on theism, thereby providing a response to

a probabilistic argument from evil that says that E, while perhaps com-

patible with theism, is very unlikely on theism. Of course, it may initially

seem to us that E isn’t very likely on theism. But once we realize that we

are in no position to say how likely particular tokens of evil are on theism,

because having an explanation for E-type evils doesn’t entail having an

explanation for any particular E-type token of evil, we gain a defeater: our

seeming that E isn’t likely on theism loses its justicatory juice (and this

holds even if that seeming persists).

I should note what is perhaps obvious: namely, that the view that gratu-

itous evil is compatible with theism in no way implies the much stronger

and implausible view that theism is compatible with all evil being gratu-

itous. Rather, the idea would be that some evil is permitted for the sake

of a greater good and yet that God is morally justied in permitting other

See van Inwagen, The Problem of Evil.

For defenses of the compatibility of theism and gratuitous evil, see Hasker, “The Neces-

sity of Gratuitous Evil,” and Howard-Snyder and Howard-Snyder, “Is Theism Compatible

with Gratuitous Evil?”

A HOLISTIC RESPONSE TO EVIL 605

evils even though no greater good necessitates them (nor are such evils

necessary for the prevention of an even worse evil). If something like this

were true, then it would open the door to the following sort of view. If we

can’t think of a God-justifying reason for some evil, then maybe that evil

is gratuitous. But we can think of a God-justifying reason for some evils,

hence some evils can be explained and some cannot. You might think that

this is, in fact, a decent approximation of our actual cognitive situation on

these matters: we can explain some but not all evils, and our inability to

explain all evils is due in part to the fact that our world contains gratu-

itous evils. The possibility that some evils may be gratuitous while others

are not helps to explain why the compossibility of theism and gratuitous

evil does not render skeptical theism superuous. Evils that we cannot

explain may be gratuitous or they may have explanations that lie beyond

our ken. Gratuitous evils are not thereby unjustied, we should note; so, a

skeptical theist retort is not superuous here: “I don’t know of a theodicy

that covers evil E,” a theist might say. “It may be that E is explained by a

good that lies beyond my ken, or it may be that E is justied even though

it is gratuitous. Either way, my inability to explain E does not imply that

it is unjustied.”

But, of course, a token skeptical theist doesn’t have to endorse the com-

patibility of theism and gratuitous evil. A token skeptical theist who re-

jects gratuitous evil can say that any seeming that God would not allow

E because E is apparently gratuitous is defeated by the realization that

we aren’t in a position to say that E is gratuitous. I might have a good ex-

planation for E-type evils without having a good explanation for E itself,

assuming token skepticism. E isn’t explainable by my theodicy for E-type

evils, but it might be explainable by some other theodicy, maybe one I’m

currently not aware of. I can’t say that E is gratuitous simply because I’m

not aware of any good, God-justifying explanation for it. In light of the

foregoing, I am likely to be in the dark about the likelihood of E given the-

ism. If I have a good theodicy for E-type evils, then I won’t be in the dark

about them even if I’m in the dark about E itself: skeptical theism about

tokens seems to t comfortably with theodicy about types.

What if the atheologian in question, instead of claiming that E is gratu-

itous, argues in Draper-style that E is unlikely on theism but not on some

alternative hypothesis? Here, a token skeptical theist, who rejects gratu-

itous evil, can simply say that we aren’t in a position to say how likely

tokens of evil like E are on theism due to the fact that explaining a type of

evil doesn’t imply explaining all instances of that type—and that, again,

this holds even if we have a good theodicy for E-type evils in general. Of

course, it may well be that having a theodicy for E-type evils raises the

probability of E itself. But that, of course, doesn’t in turn imply that we can

say how likely E is on theism. Perhaps the best we can say here is that, in

light of our theodicy for E-type evils, E itself isn’t too unlikely on theism.

Given token skepticism, it may well be that E is explained by some theo-

dicy other than the one we have for E-type evils in general, one which is

Faith and Philosophy606

beyond our ken. In any case, the general idea here is that, since a theodicy

for a type of evil doesn’t necessarily account for every token of that type,

arguments from evil that appeal to a particular instance of evil will run

headlong into the token skeptical theist’s claim that we are not very good

at judging the likelihood of tokens of evil.

One crucial question here is whether we should embrace token

skeptical theism in the rst place. Maybe it’s not a very good view. As

I indicated above, a full articulation and defense of the view isn’t possi-

ble here, but it’s worth noting that there may well be an asymmetry here

between explanations of non-agential happenings and explanations of

agential happenings. If I’ve got an explanation for why, in general, water

freezes when it gets below a certain temperature, then it looks like I’ve got

an explanation of any particular instance of water freezing. Here, in the

non- agential realm, it looks like having an explanation for a type may well

entail having an explanation for every token of that type. The same may

not be said, however, for happenings in the agential realm. I may correctly

surmise that you let your daughter stay up late on Friday nights because

she doesn’t have school the next day. However, that explanation may not

hold for every single Friday night. Perhaps this Friday you planned on

putting her to bed early because she has somewhere to be early the next

morning, but due to some medicine she took, which makes it difcult to

sleep, you’re letting her stay up late this Friday night.

Skeptical theisms based on token skepticism sit well with the idea that

probabilistic natural theology arguments are, properly understood, really

about types of goods rather than tokens of those types. Recall Poston’s

H—humanly free creatures exist. Theists (or some of them, anyway) think

that H is not too low given theism due to the intrinsic value possessed

by consciousness and freedom. What such theists think, properly under-

stood, is that God would likely bring about a world containing things of a

certain type; theists don’t think that God would bring about any one par-

ticular humanly free creature, or homo sapiens specically. So, for instance,

my existence is not something anyone would predict as likely given the-

ism, but the existence of beings like me does seem—or so the argument

could go—likely given theism. The same line of thinking, i.e., of thinking

I intend what I am saying to be compatible with the claim that, in some cases, having

an explanation for a certain type of evil might entail having an explanation for every token

of that type. Suppose, for instance, that animals do not feel morally relevant pain and also

that animal pain is in general quite biologically useful; in such a case, God would have good

reasons to allow animal pain and no good reason not to. It might then follow, at least on the

assumption that there is no other morally sufcient reason to eliminate or minimize animal

pain, that we have an explanation for animal pain in general that covers every instance of

animal pain. Though I think this view is plausible, I am not endorsing it, but am merely call-

ing it to our attention as a potential example of the possibility of having an explanation for a

type of evil that covers every token of that type. For a recent argument against the idea that

animals do indeed feel morally relevant pain, see Miller, “Do Animals Feel Pain in a Morally

Relevant Sense?”

A HOLISTIC RESPONSE TO EVIL 607

about what God would allow or bring about in terms of types rather than

tokens, seems to apply to how theists are thinking (or should be thinking)

when they do theodicy. In offering the free will theodicy as an explanation

for moral evil, for example, the theist isn’t saying that any one particular

token of morally bad action is necessary to secure the good of free will

in general. Rather, theists are arguing, or should understand themselves

to be arguing, that a good of a certain type necessitates God’s allowance

of evils of a certain type. In sum, when giving probabilistic natural theo-

logical arguments, the theist is best understood as arguing that it’s likely

that God would actualize a world with certain types of goods; and when

giving a theodicy, the theist is best understood as arguing that goods of

a certain type require the permission of evils of a certain type. This more

holistic approach seems to offer an answer to Cullison’s RAG problem: we

can reason fairly well about the types of things God would allow or bring

about, but not very well about which particular tokens of those types God

would allow or bring about.

Another worry about a version of the holistic response based on token

skepticism is that it is based on a misunderstanding of theodicy, for no

theodicist—someone might argue—claims to have good explanations for

all the particular tokens of evil we are confronted with.

Whether this is

true for any given theodicist will depend in part on whether that theod-

icist thinks that a theodicy for every type of evil explains every token of

that type. The worrier here seems to assume that no theodicist has ever

made that assumption, but I doubt whether that is true or not. In personal

conversation with me, both Plantinga and Swinburne argued that having

a theodicy for a type does entail having an explanation for every token of

that type. Also in personal conversation, Draper suggested the opposite,

arguing that having a theodicy for a type does not entail having a theodicy

for every token of that type. Given this, as well as the paucity of explicit

discussion of the matter, it is likely that the worrier’s assumption on this

score may well be false. Critics of theodicies sometimes respond to theod-

icies by pointing out that a proposed theodicy doesn’t explain all the facts

of evil. Laura Ekstrom, for example, says the following:

[O]ur experiences of, and awareness of, facts about evil in the world provide

awfully good reason to doubt that a hybrid case will succeed in covering

the full range, distribution, and intensity of evils. It seems especially unrea-

sonable to think that the pain endured by all sentient non-human animals

by way of neglect, abuse, torture, disease, and natural disaster can be justi-

ed by a hybrid theodicy that draws on the values of free will, punishment,

character- building, and connection with God.

Ekstrom’s complaint seems to presuppose or at least be consistent with

the idea that a theodicy for a type doesn’t cover every token of that type.

Thanks to a referee for helping me see the need to address this issue.

Ekstrom, God, Suffering and the Value of Free Will, 94–95.

Faith and Philosophy608

I think, then, that the matter is a contested one and thus that the objector

here is not justied in claiming that no theodicist claims to have good

explanations for all the particular tokens of evil we are confronted with.

But regardless of what theodicists do or do not assume about the scope

of their theodicy projects, it remains open to the atheologian to make ar-

guments from evil that appeal to types or tokens. And of course it remains

open to a critic of theodicy, someone such as Ekstrom, to complain that

extant theodicies don’t explain all the facts about evil. So, in response to

atheological arguments from evil in general, it is good for the theist to

have at her disposal responses to both types of argument, and thus it is

important to distinguish them and to note the possibility of offering a

skeptical theist reply to arguments from evil that appeal to tokens and to

offer a theodicy reply to arguments from evil that appeal to types. And in

response to complaints about the explanatory inadequacy of a theodicy,

it can be very useful to make use of the holistic response described here

by replying with skeptical theism about tokens and theodicy about types.

Other sorts of non-concessive holistic responses seem possible, at least

in principle. Let me briey outline, without developing, three of them.

For starters, you might simply think that we can explain some but not all

evils. For example, perhaps you are persuaded by Stump’s theodicy and

thus think we can explain the suffering of fully functional adult human

beings but also that we currently have no good explanation for the suffer-

ing of sentient beings, human or otherwise, that fall outside the scope of

this theodicy. Or perhaps you propagate what Ekstrom refers to as a “hy-

brid theodicy,” in which you offer different theodicies for different evils;

even with a hybrid theodicy on hand, however, you might still think that

you cannot explain every type of evil, or at least every token of every

type.

You might also think that, in virtue of having good inferential and/

or noninferential reasons to believe that God exists, and that God would

not allow an evil without having a justifying reason for doing so, that

God does have such reasons and thus that your inability to think of them

is not evidence that they aren’t there. This move would be a version of

what Rowe refers to as the “G. E. Moore shift.” The skeptical theism here

would be based on the fact that you have good reasons to think that God

exists and would not allow pointless evil. It could of course be helpfully

Ekstrom:

This raises the question of the plausibility of a hybrid theodicy or a bundle case.

Perhaps a theist could argue that, for some cases of evil, the God-justifying reason

for causing or allowing them is punishment, whereas with other cases of evil, the

God-justifying reason for allowing them is the preservation of created beings’ power

of libertarian free will, whereas in other cases of evil, the God-justifying reason for

causing or allowing them is temporally extended character development, and in

other cases of evil, the God-justifying reason for causing or allowing them is that

they provide avenues to knowledge of, and intimacy with, God. (God, Suffering, and

the Value of Free Will, 91)

I should note that Ekstrom does not here endorse this approach herself.

A HOLISTIC RESPONSE TO EVIL 609

buttressed by reections on the limitations of human cognition in general

but also on the fact that you do have some explanations for some evils.

The skeptical theist bit in all this is, of course, the thought that an inability

to think of a God-justifying reason for evil E is no reason to think that there

is no such reason. Is it even reasonable to demand that theists be able to

produce explanations for every evil, or the exact amount of evil, and so

on? In how many domains of inquiry, for example, can we explain every

factum that is, in principle, explainable? To demand universal explana-

tions in this case seems like overkill, given our cognitive limitations; more

strongly, it seems like special pleading on behalf of religious skepticism.

Another possibility is that of being a skeptical theist qua generic theism

but a theodicist qua religious theism. So, suppose you endorse the view

that we have good inferential and/or noninferential reasons to think that

God exists. Qua generic theist, you might say that we aren’t able to explain

why God allows evil but that this is no reason to think that there aren’t any

such explanations at all. One reason for denying that we have the ability

to predict God’s behavior, qua generic theists, is that you might not think

of God as a well-behaved moral agent, that is, as someone who is subject

to and perfectly follows the same moral law that govern us. Without de-

nying divine goodness, you might take the line that that goodness does

not consist in being a well-behaved moral agent and thus that your ability

to predict divine behavior is not particularly strong. In their own ways,

Brian Davies, Mark Murphy, and Edward Feser have all recently given ar-

guments against the view that we should think of God as a well-behaved

moral agent.

But you might not be just a generic theist, for you—like

Davies, Murphy, and Feser—might also be a religious theist, that is, an

adherent to a theistic religion that posits divine revelation. You might then

say that, given the additional resources provided by divine revelation, we

can offer good theodicies for evil. So, if the atheologian wants to attack ge-

neric theism and you are content to operate just with the tools provided by

natural reason, you might respond by combining Reformed Epistemology,

natural theology, and skeptical theism. On the other hand, if the atheo-

logian wants to attack a specic version of religious theism, the one you

advance, you might then reply by combining Reformed Epistemology,

natural theology, and theodicy, bringing in the additional resources made

available to you by your faith tradition.

Yet another option involves making a distinction between non-

comparative probability judgments (“X is unlikely [likely] given the-

ism”) and comparative ones (“X is more likely [unlikely] on naturalism

than on theism”).

Skeptical theism might rule out non-comparative

probability judgments, meaning we can’t say how likely (full stop) ne-

tuning is, given theism, nor can we say how likely (full stop) some evil

See Davies, The Reality of God and the Problem of Evil, 84–105; Murphy, God’s Own Ethics,

106–116; Feser, “The Thomistic Dissolution of the Logical Problem of Evil.”

Thanks to a referee for pointing out the possibility of this version of the holistic response.

Faith and Philosophy610

E is given theism. But this seems compatible with being able to say that

ne- tuning, say, is more likely on theism than naturalism. I may not be

able to say how likely it is that the U.S. Congress will declare war on X,

but I know that it is more likely that they will declare war on X than it

is that Vice-President Harris will declare war on X (for the simple rea-

son that vice-presidents don’t have the authority to declare war). The

ruling out of non-comparative probability judgments also seems com-

patible with saying, in Draper-style, that some evil E is more likely on,

say, the hypothesis of indifference than it is on theism. So, this version of

the non- concessive holistic response allows the theist to wield skeptical

theism against non-comparative likelihood judgments of the sort, “evil E

is evidence against theism because it is unlikely given theism”; in reply,

the skeptical theist will argue we are in the dark about the likelihood

of E given theism. Draper-style comparative arguments from evil would

have to be met with another response, but there’s no in-principle reason

to think that one or more of the other elements of the holistic response

couldn’t be brought in for that job.

4. Conclusion

The holistic response to evil would not be a silver bullet. There is no

guarantee that theists will be able to come up with good theodicies; no

guarantee that the arguments of natural theology work; no guarantee that

skeptical theism can be defended from the serious objections which have

been leveled against it, such as that it implies too much skepticism; and no

guarantee that we really can have noninferential justication for theistic

belief. Theists will still have to do—or, rather, keep doing—the hard work

of defending each of the four standard responses to the problem of evil.

However, the result reached here is not insignicant. If the four standard

theistic responses to the problem of evil are in fact compatible, then, so

long as each one is defensible in its own right, it suggests that the theist

has at her disposal a powerful, holistic response to what is perhaps the

most serious objection to God’s existence.

Conception Seminary College

References

Alston, William. 1991. “The Inductive Argument from Evil and the Human Cog-

nitive Condition.” Philosophical Perspectives 5: 29–67. https://doi.org/10.2307

/2214090.

Anderson, David. 2012. “Skeptical Theism and Value Judgments.” International

Journal for Philosophy of Religion 72: 27–39. https://doi.org/10.1007/s11153

-012-9358-1.

Baker-Hytch, Max. 2023. “Natural Theology and Religious Belief.” In The Cambridge

Handbook of Religious Epistemology, edited by Jonathan Fuqua, John Greco, and

Tyler McNabb (Cambridge University Press), 13–28.

A HOLISTIC RESPONSE TO EVIL 611

Beaudoin, John. 1998. “Evil, the Human Cognitive Condition, and Natural Theol-

ogy.” Religious Studies 34: 403–418. https://doi.org/10.1017/s0034412598004557.

Bergmann, Michael. 2001. “Skeptical Theism and Rowe’s New Evidential Argu-

ment from Evil.” Nous 35: 278–296. https://doi.org/10.1111/0029-4624.00297.

Bergmann, Michael. 2009. “Skeptical Theism and the Problem of Evil.” In The

Oxford Handbook of Philosophical Theology, edited by Michael Rea and Thomas

Flint (Oxford University Press), 374–399.

Bergmann, Michael. 2012. “Rational Religious Belief Without Argument.” In

Philosophy of Religion: An Anthology, 6th ed., edited by Michael Rea and Louis

Pojman (Cengage Learning), 534–549.

Bergmann, Michael. 2023. “Reformed Epistemology.” In The Cambridge Handbook

of Religious Epistemology, edited by Jonathan Fuqua, John Greco, and Tyler

McNabb (Cambridge University Press), 41–55.

Coley, Scott. 2015. “Skeptical Theism is Incompatible with Theodicy.” International

Journal for Philosophy of Religion 77: 53–63. https://doi.org/10.1007/s11153

-014-9494-x.

Craig, William Lane. 2003. “Design and the Anthropic Fine-Tuning of the Uni-

verse.” In God and Design: The Teleological Argument and Modern Science, edited

by Neil Manson (Routledge), 155–177.

Craig, William Lane, and J. P. Moreland, eds. 2009. The Blackwell Companion to

Natural Theology (Wiley-Blackwell).

Cullison, Andrew. 2014. “Two New Versions of Skeptical Theism.” In Skeptical

Theism: New Essays, edited by Trent Dougherty and Justin McBrayer (Oxford

University Press), 250–263.

Davies, Brian. 2006. The Reality of God and the Problem of Evil (Continuum).

Dougherty, Trent. 2014. “Skeptical Theism.” In The Stanford Encyclopedia of Philoso-

phy (Spring 2014 Edition), edited by Edward N. Zalta. http://plato.stanford.edu

/archives/spr2014/entries/skeptical-theism.

Dougherty, Trent, and Chris Tweedt. 2015. “Religious Epistemology.” Philosophy

Compass 10: 547–559. https://doi.org/10.1111/phc3.12185.

Draper, Paul. 1989. “Pain and Pleasure: An Evidential Problem for Theists.” Nous

23: 331–350. https://doi.org/10.2307/2215486.

Draper, Paul. 1996. “The Skeptical Theist.” In The Evidential Argument from Evil,

edited by Daniel Howard-Snyder (Indiana University Press), 175–192.

Draper, Paul. 2009. “The Problem of Evil.” In The Oxford Handbook of Philosophical

Theology, edited by Thomas Flint and Michael Rea (Oxford University Press),

332–351.

Draper, Paul. 2013. “Explanation and the Problem of Evil.” In The Blackwell

Companion to the Problem of Evil, edited by Daniel Howard-Snyder and Justin

McBrayer (Wiley-Blackwell), 71–79.

Ekstrom, Laura. 2021. God, Suffering, and the Value of Free Will (Oxford University

Press).

Evans, C. Stephen. 2010. Natural Signs and Knowledge of God: A New Look at Theistic

Arguments (Oxford University Press).

Feser, Edward. 2017. Five Proofs of the Existence of God (Ignatius Press).

Feser, Edward. 2021. “The Thomistic Dissolution of the Logical Problem of Evil.”

Religions 12: 268. https://doi.org/10.3390/rel12040268.

Hasker, William. 1992. “The Necessity of Gratuitous Evil.” Faith and Philosophy

9: 23–44. https://doi.org/10.5840/faithphil1992911.

Hick, John. 2010. Evil and the God of Love, 3

ed. (Palgrave Macmillan).

Faith and Philosophy612

Holley, David. 2010. “Treating God’s Existence as an Explanatory Hypothesis.”

American Philosophical Quarterly 47: 377–388. https://www.jstor.org/stable

/25734163.

Howard-Snyder, Daniel. 1996. “The Argument from Inscrutable Evil.” In The

Evidential Argument from Evil, edited by Daniel Howard-Snyder (Indiana

University Press), 286–310.

Howard-Snyder, Daniel. 2013. “The Logical Problem of Evil: Mackie and Plant-

inga.” In The Blackwell Companion to the Problem of Evil, edited by Daniel

Howard-Snyder and Justin McBrayer (Wiley-Blackwell), 19–33.

Howard-Snyder, Daniel, and Frances Howard-Snyder. 1999. “Is Theism Compati-

ble with Gratuitous Evil?” American Philosophical Quarterly 36: 115–130. https://

www.jstor.org/stable/20009958.

McBrayer, Justin. 2010. “Skeptical Theism.” Philosophy Compass 5: 611–623. https://

doi.org/10.1111/j.1747-9991.2010.00306.x.

McCain, Kevin. 2023. “Evidence and Religious Belief.” In The Cambridge Hand-

book of Religious Epistemology, edited by Jonathan Fuqua, John Greco, and Tyler

McNabb (Cambridge University Press), 29–40.

McNabb, Tyler. 2015. “Warranted Religion: Answering Objections to Alvin Plant-

inga’s Epistemology.” Religious Studies 51: 477–495. https://doi.org/10.1017

/s003441251400033x.

Miller, Calum. 2021. “Do Animals Feel Pain in a Morally Relevant Sense?” Philoso-

phia 49: 373–392. https://doi.org/10.1007/s11406-020-00254-x.

Moon, Andrew. 2016. “Recent Work in Reformed Epistemology.” Philosophy Com-

pass 11: 879–891. https://doi.org/10.1111/phc3.12361.

Murphy, Mark. 2017. God’s Own Ethics: Norms of Divine Agency and the Argument

from Evil (Oxford University Press).

Murray, Michael. 2009. “Theodicy.” In The Oxford Handbook of Philosophical Theology,

edited by Michael Rea and Thomas Flint (Oxford University Press), 352–373.

Pike, Nelson. 1963. “Hume on Evil.” The Philosophical Review 72: 180–197. https://

doi.org/10.2307/2183103.

Plantinga, Alvin. 1967. God and Other Minds (Cornell University Press).

Plantinga, Alvin. 1974. The Nature of Necessity (Clarendon Press).

Plantinga, Alvin. 1981. “The Reformed Objection to Natural Theology.” Proceed-

ings of the American Catholic Philosophical Association 15: 49–62. https://doi

.org/10.5840/acpaproc1980547.

Plantinga, Alvin. 1990. God and Other Minds, rev. ed. (Cornell University Press).

Plantinga, Alvin. 1996. “On Being Evidentially Challenged.” In The Evidential Argument

from Evil, edited by Daniel-Howard Snyder (Indiana University Press), 244–261.

Plantinga, Alvin. 2000. Warranted Christian Belief (Oxford University Press).

Plantinga, Alvin. 2004. “Supralapsarianism, or ‘O Felix Culpa.’” In Christian Faith

and the Problem of Evil, edited by Peter van Inwagen (Eerdmans), 1–25.

Plantinga, Alvin. 2007. “Two Dozen (or so) Theistic Arguments.” In Alvin Plant-

inga, edited by Deane Peter-Baker (Cambridge University Press), 203–227.

Plantinga, Alvin. 2008. “Reply to Tooley’s Opening Statement.” In Knowledge of

God, by Alvin Plantinga and Michael Tooley (Wiley-Blackwell), 151–183.

Plantinga, Alvin. 2011. Where the Conict Really Lies (Oxford University Press).

Plantinga, Alvin, and Michael Tooley. 2008. Knowledge of God (Wiley-Blackwell).

Poston, Ted. 2014. “Skeptical Theism within Reason.” In Skeptical Theism: New

Essays, edited by Trent Dougherty and Justin McBrayer (Oxford University

Press), 307–322.

A HOLISTIC RESPONSE TO EVIL 613

Robinson, Thaddeus. 2013. “Descartes’s Sceptical Theism.” Religious Studies 49:

515–527. https://doi.org/10.1017/s0034412512000418.

Rowe, William. 1979. “The Problem of Evil and Some Varieties of Atheism.” American

Philosophical Quarterly 16: 335–341. https://www.jstor.org/stable/20009775.

Rudavsky, T. M. 2013. “A Brief History of Skeptical Responses to Evil.” In The

Blackwell Companion to the Problem of Evil, edited by Daniel Howard-Snyder and

Justin McBrayer (Wiley-Blackwell), 379–395.

Schellenberg, John. 2013. “A New Logical Problem of Evil.” In The Blackwell

Companion to the Problem of Evil, edited by Daniel Howard-Snyder and Justin

McBrayer (Wiley-Blackwell), 34–48.

Sterba, James P. 2019. Is a Good God Logically Possible? (Palgrave Macmillan).

Stump, Eleonore. 1996. “Aquinas on the Sufferings of Job.” In The Evidential

Argument from Evil, edited by Daniel Howard-Snyder (Indiana University

Press), 49–68.

Stump, Eleonore. 2010. Wandering in Darkness: Narrative and the Problem of Suffering

(Oxford University Press).

Sudduth, Michael. 2003. “Reformed Epistemology and Christian Apologetics.”

Religious Studies 39: 299–321. https://doi.org/10.1017/S0034412503006553.

Sudduth, Michael. 2009. “Revisiting the ‘Reformed Objection’ to Natural

Theology.” European Journal for Philosophy of Religion 2: 37–62. https://doi.org

/10.24204/ejpr.v1i2.340.

Swinburne, Richard. 1998. Providence and the Problem of Evil (Oxford University

Press).

Swinburne, Richard. 2004. The Existence of God, 2nd ed. (Oxford University Press).

Swinburne, Richard. 2005. Faith and Reason, 2nd ed. (Oxford University Press).

Tucker, Chris. 2011. “Phenomenal Conservatism and Evidentialism in Religious

Epistemology.” In Evidence and Religious Belief, edited by Kelly James Clark and

Raymond J. VanArragon (Oxford University Press), 52–73.

van Inwagen, Peter. 2006. The Problem of Evil (Oxford University Press).

Walls, Jerry, and Trent Dougherty. 2018. Two Dozen (or So) Arguments for God: The

Plantinga Project (Oxford University Press).

Wilks, Ian. 2004. “The Structure of the Contemporary Debate on the Problem of Evil.”

Religious Studies 40: 307–321. https://doi.org/10.1017/S0034412504006900.

Wolterstorff, Nicholas. 2008. Justice: Rights and Wrongs (Princeton University Press.)