3. The Argument from Ignorance
different locations) in order to cause some effects but not others in our world of time and space. Such non-homogeneous existence is incompatible with the traditional notion of God.
Many more responses are possible here, as with the problem of evil. Theists might try to develop a new notion of causation that does not depend on temporal relations. This new kind of causation could take various forms, but each form is problematic, and I don’t have time or space to go through them now. Instead, let’s see how Craig responds. Then I will assess his response.
If no response is adequate, as I think, then theists have a choice, as with the problem of evil. They can claim that God is eternal but never active in time. Or they can claim that God is active in time but not eternal. What they cannot do is hold onto both parts of the traditional conception of God. The argument from action shows that there is no God of that traditional kind.
3. The Argument from Ignorance
My next argument cannot be escaped simply by giving up traditional features of God, since it applies to many other kinds of god. I will continue to focus on the traditional God, but any kind of god that cannot be supported by evidence runs afoul of a common standard for justified belief.
One principle that is accepted throughout science and everyday life is that we should not believe in entities for which we have no evidence. No good scientist would believe in a new element or particle or force without evidence. Any scientist who did claim to have discovered a new particle but who could not produce any evidence for it would be laughed out of town.
This rule applies with special force when an entity is unusual in important respects. If someone believes that there are clothes in my closet, then he might not seem to need any evidence for this belief other than his past experiences of closets, most of which contained clothes. However, no evidence like this is available, and the need for evidence seems especially strong, when someone believes in a kind of entity that is unprecedented. Nobody should believe that there is a perfect shirt or an invisibility cloak in the closet without any evidence for the existence of such an oddity. I hope everyone agrees in that case.
God?
Religion might seem different. Religion is often said to be im-mune from the standards of science, because it lies outside the realm where evidence is possible. But this view leads to absurdity. If the fact that a religious belief cannot be subject to evidence did license one to accept that belief without evidence, then the same principle would also license many ridiculous beliefs.
An amusing example occurs in Charles Schultz’s cartoon, Peanuts.
Linus believes in the Great Pumpkin, who visits pumpkin patches all over the world every Halloween. Linus has no evidence for his belief, since he has never seen the Great Pumpkin, and Linus can explain all of his beliefs and experiences without postulating any Great Pumpkin. On the other hand, Linus also ensures that nobody can have any evidence against the Great Pumpkin, since it is visible only to those with a pure heart. Linus stays in a pumpkin patch all Halloween night (at a great cost in candy). He looks constantly and carefully for the Great Pumpkin (at a great cost in sleep). But he sees nothing. Still, this does not show that the Great Pumpkin does not exist. All it shows is that Linus’s heart was not pure, or not pure enough, at least according to Linus. This doctrine thus makes it impossible to prove that Linus is wrong. Moreover, Linus might be right. It is possible (just barely!) that there is a Great Pumpkin. However, most people agree that Linus is not justified in believing in the Great Pumpkin. More strongly, Linus should not believe in the Great Pumpkin. That is what makes the cartoon so funny. Linus is the butt of jokes in a cartoon for kids, because Linus believes something that he should not believe, given his evidence, on standards that are obvious even to children.
Linus illustrates why we should not believe in unusual entities without evidence for their existence and why this common restriction cannot be avoided by formulating one’s beliefs to make them incapable of refutation. If we weaken our epistemic standards to ac-commodate irrefutable beliefs, then we might end up believing in the Great Pumpkin or, at least, holding that many absurd beliefs like this are justified.
The same standards apply to beliefs about God. If there is no evidence for God’s existence, then we ought not to believe that God exists. This follows even if God were defined so as to remove any possibility of evidence against His existence.
Some Reasons to Believe that There Is No God 103
But is there evidence for God? Craig claims that there is. That is why he gave several arguments for the existence of God in the first part of this book. However, I refuted every one of his arguments.
Craig might have overlooked an even better argument. Or maybe there is some way to get around my criticisms. But I don’t see any.
Religious experiences are sometimes seen as evidence for God.
Many people have feelings that seem to them to come from a higher power outside of them. But many people also think that they see and hear ghosts. Their experiences are not evidence for ghosts because we and they have no independent reason to think that their experiences are accurate rather than illusions. By the same standards, when people seem to experience God, that is no evidence for God because they and we have no independent reason to believe that their religious experiences are accurate rather than illusions.
Moreover, we and they have plenty of reason to suspect that there is some better explanation of why they seem to see God. Just as people see ghosts only when they are afraid and predisposed to believe in ghosts, so people experience God only when they are predisposed to believe in God and are gripped by emotions that often distort experiences. God might appear only to those who seek or believe in Him, but then those experiences cannot count as evidence for God.
Finally, even if religious experiences were evidence for something, they could not be evidence that their source is all-good and all-powerful. The most vivid religious experience could result from a God who is very powerful and pretty good, so they cannot be evidence for a traditional God.
Miracles are also often claimed to be evidence for God. However, second-hand reports of miracles are dubious at best, for obvious reasons (which I cited in the first part of this book). On the other hand, if you think that you have witnessed a miracle first-hand, then the circumstances need to be considered carefully. Even if you and I cannot come up with any natural explanation for what you observed, that is no more miraculous than most magic tricks are. Besides, no miracle can be evidence that God is all-powerful or all-good, which is the kind of God at issue here.
I conclude that neither arguments nor experiences nor miracles provide any good evidence for the existence of God. This conclusion, together with common standards for justified belief, implies
God?
that we ought not to believe that God exists and, hence, that we ought to believe that God does not exist.
Theists often respond that it is fallacious to argue from “We have no evidence for it” to “It does not exist.” This is called an argument from ignorance. Many logic textbooks label it a fallacy.
I admit that many arguments of this form are fallacious. One cannot say that there is no gold on Uranus just because we do not have any evidence for gold on Uranus. The problem is clear: Even if there were gold on Uranus, we still would not (or, at least, might not) have evidence for it. That is why we cannot conclude that there is no gold on Uranus.
However, this point cannot be used to defend the claim that there is gold on Uranus. If there is no evidence for gold on Uranus, maybe we should not conclude that there is none, but we also should not believe that there is some. We should suspend belief until we have adequate evidence one way or the other. Consequently, this analogy cannot provide solace to religious believers. If belief in God is like belief in gold on Uranus, then the lack of evidence for God does not show that God does not exist. However, the lack of evidence also cannot be used to defend a belief in God, since it supports agnostics who hold that we should suspend belief until we have evidence (even if this is a very long time).
More importantly, arguments from ignorance are not always fallacious. Arguments of this form work fine when the phenomenon is something that we would know if it were true. If I had a pot-bellied pig on my head, I would know it, at least after I felt around up there.
Thus, my lack of evidence for a pot-bellied pig on my head is adequate reason for me to believe that there is none. Not everything is as obvious as a pot-bellied pig. But doctors can also have adequate reason to believe that a patient does not have a virus, if they look closely, they find no evidence for that virus, and the patient would have easily detectable symptoms if the patient did have that virus.
That is the situation with God. If there were an all-good and all-powerful God who could act in time, then we would have better evidence than we have. He could easily reveal himself by appearing before us. Giving us better evidence would not harm us. Why would such a God hide?
Some Reasons to Believe that There Is No God 105
Some theists answer that, if the evidence for God were stronger, believers would not need faith. However, better evidence could leave room for faith. I have faith in my wife’s love, even though I also have strong evidence of her love, namely, that she puts up with me! I would not be better off if I had to rely more on faith because I had less evidence for her love. So why is it better to have less evidence for God?
One answer might be that better evidence would take away our freedom not to believe. However, evidence does not take away any valuable kind of freedom. If it did, teachers would restrict the freedom of their students every time the teachers told the students a fact about history or performed a science experiment. If any freedom is lost in such revelations, it is a kind of freedom that is not very valuable, since it is just the freedom to believe irrationally.
Finally, even if stronger evidence did have some costs, it would also have many benefits. The new evidence would remove or reduce nagging doubts, as well as any fears that your children or friends might disavow God and end up in Hell. It would bring assurance and solace, if God is as merciful as Christians claim. Most evil people would be scared away from their horrible deeds, since few, if any, people would rape and torture if they received strong evidence that such deeds would be punished with eternal torment.
With so little to lose and so much to gain, an all-good God would reveal Himself clearly to everyone. An all-powerful God could easily reveal Himself clearly to everyone. But God has not revealed Himself clearly to everyone. Even if you have been convinced by your religious experiences, there are billions of other people on this earth now and in the past who have little or no evidence of God. Just ask yourself honestly, “Could God reveal Himself more clearly?” The answer is bound to be “Yes,” since God is supposed to be all-powerful. Now ask yourself honestly,
“What reason does God have not to reveal Himself more clearly?”
I doubt that you can come up with any reason that withstands honest and thorough reflection. Consequently, if there were a God, we would have more and better evidence than we do. That is why our lack of evidence for God supports my thesis that there is no God.
God?
4. Conclusion
Taken together, my arguments lead to the conclusion that no traditional God exists. As I said, my arguments do not pretend to be conclusive proofs, but they do provide strong evidence against any traditional God.
Even if theists accept my arguments, they can still believe in other kinds of God. Religious rituals and institutions can continue. They can still follow the central moral teachings of the great religions. My arguments show why theists should believe that no traditional God exists, but their everyday lives can still go on much as they did before. To that extent, theists have nothing important to fear from my conclusion.
Indeed, there is much to gain from my conclusion. The joys of inquiry unfettered by dogma are intense. The resulting knowledge is valuable in itself, as is the removal of any kind of ignorance and illusion. The knowledge that God does not exist also has good consequences. If people stop looking to Heaven for solutions to their problems, then they should work harder to solve the many problems in this world.
Even if this picture is overly optimistic, and life is better when people believe in a traditional God, that good life would then depend on an illusion, since that God does not exist. My main concern is not to make people comfortable but to determine whether the traditional God really does exist. He does not.
Theism Undefeated
William Lane Craig
The outspoken Madelyn Murray O’Hare once remarked, “Agnostics are just atheists without guts.” Professor Sinnott-Armstrong has guts.
He is not content to rest in an easy and undemanding agnosticism; rather he argues that we have “strong evidence,” if not conclusive proof, that God does not exist. He presents three arguments in support of his claim: the problem of evil, the problem of action, and the problem of ignorance. Since the first of these three problems is by far the most compelling (at least emotionally), whereas the latter two can be fairly easily answered, I’ll consider them in reverse order.
The Argument from Ignorance
This argument is based on the principle that “we should not believe in entities for which we have no evidence.” Since we have no evidence for God, we should not believe in Him.
The shortcoming of this argument is that even if we accept its un-derlying principle (which is not uncontroversial1), it at best supports agnosticism, not atheism. Therefore, it cannot count among Sinnott-Armstrong’s “positive arguments for atheism.”
The careful reader will note a subtle but crucial shift as the argument unfolds. Initially Sinnott-Armstrong asserts, “If there is no evidence for God’s existence, then we ought not to believe that God exists.” This is a careful application of his principle and enunciates an agnostic position. But a page later, he asserts, “This [the absence of good evidence for God], together with common standards for jus-107
God?
tified belief, implies that we ought not to believe that God exists and, hence, that we ought to believe that God does not exist.” Did you notice the shift from “we ought not to believe that God exists”
to “we ought to believe that God does not exist”? That shift, which represents the move from agnosticism to atheism, requires justification. To borrow Sinnott-Armstrong’s example, in the absence of evidence, we should not believe that there is gold on Uranus; but it does not follow that we should believe that there is no gold on Uranus. Thus, even if entirely successful, the argument so far does nothing to support atheism.
In a bizarre role reversal, Sinnott-Armstrong rejoins, “However, this point cannot be used to defend the claim that there is gold on Uranus. If there is no evidence for gold on Uranus, maybe we should not conclude that there is none, but we also should not believe that there is some.” Wait a minute! I thought this was supposed to be a positive argument for atheism. But at best this implies that we should neither believe that God exists nor believe that He does not exist.
That is classic agnosticism, not atheism. Worse, Sinnott-Armstrong goes on to say, “the lack of evidence also cannot be used to defend a belief in God.” But I don’t know of anyone who thinks that the existence of God follows from the lack of evidence (indeed, isn’t such a claim incoherent when you think about it?). Sinnott-Armstrong has shifted from building a positive case for atheism to knocking down straw men.
Cognizant of the weakness of his argument, Sinnott-Armstrong proceeds to argue that in certain cases the absence of evidence for some thing is evidence of the absence of that thing. If we should expect to see evidence of some thing if it did exist, then the absence of any such evidence gives us grounds to think it does not exist. That seems quite right. The absence of evidence for an elephant in this room constitutes good grounds for thinking there is no elephant in this room, for if there were an elephant here, I should expect to see evidence of it. By contrast, the absence of evidence for a flea in this room provides no grounds at all for my thinking that there is no flea in this room, since if there were a flea here, I shouldn’t expect things to look any different than they do.
So the whole question devolves to Sinnott-Armstrong’s claim that
“If there were an all-good and all-powerful God, then we would have
109
better evidence than we have.” I have to say honestly that I just have no confidence at all that this is true. Indeed, this claim strikes me as enormously presumptuous.2 Who are we to say what kind of evidence God would give of His existence if He existed?
The question is not, as Sinnott-Armstrong thinks, “Could God reveal Himself more clearly?” Of course, He could: He could have inscribed the label “Made by God” on every atom or planted a neon cross in the heavens with the message “Jesus Saves.” But why should He want to do such a thing? Although I’ve found that atheists have a hard time grasping this, it is a fact that in the Christian view it is a matter of relative indifference to God whether people (merely) believe that He exists or not. For what God is interested in is building a love relationship with you, not just getting you to believe that He exists. According to the Bible, even the demons believe that God exists—and tremble, for they have no saving relationship with Him (James 2.19). Of course, in order to believe in God, you must believe that God exists. But there is no reason at all to think that if God were to make His existence more manifest, more people would come into a saving relationship with Him. Mere showmanship will not bring about a change of heart (Luke 16.30–31). It’s interesting that, as the Bible describes the history of God’s dealings with mankind, there has been a progressive “interiorization” of this interaction. In the Old Testament, God is described as revealing Himself to His people in manifest wonders: the plagues upon Egypt, the pillar of fire and smoke, the parting of the Red Sea. But did such wonders produce lasting heart-change in the people? No, Israel fell into apostasy with tiresome repetitiveness. If God were to inscribe His name on every atom or place a neon cross in the sky, people might believe that He exists, all right, but what confidence could we have that after a time they would not begin to chafe under the brazen advertisements of their Creator and even to resent such effrontery? According to the New Testament, not only is God’s existence revealed through His handiwork in nature, so that unbelievers are “without excuse” (Romans 1.20), and His moral law written on the hearts of all persons (Romans 2.14–15), rendering them morally culpable before God, but His Holy Spirit ministers to the heart of every person convicting and drawing him to God (John 16.7–11). Moreover, God in His omniscience has providentially ordered the world so that people are born
God?
at times and places conducive to their coming to faith (Acts 17.24–27).
Do we have any reason to think that if God’s existence were more manifest in the way that Sinnott-Armstrong envisions, this would be any more effective in drawing people into a saving relationship with God than the interior way that He has chosen? I think an unprejudiced answer to this question can only be that this is pure speculation on Sinnott-Armstrong’s part. He has no idea at all that in a world of free creatures in which God’s existence is as obvious as the nose on your face that more people would come to love Him and know His salvation than in the actual world. But then the claim that if God existed, He would make His existence more evident has little or no warrant, thereby undermining the further claim that the absence of such evidence is itself positive evidence that God does not exist. The whole argument is a tissue of uncertainties.
The Problem of Action
Sinnott-Armstrong argues that if God is timeless, then He cannot be causally active in the temporal world, as the Christian faith holds Him to be. Therefore, traditional theism is internally inconsistent and cannot be true.
I have to confess that I was rather pleased to find this objection in Sinnott-Armstrong’s repertoire, since it constitutes an endorsement of precisely the position that I have labored to establish in my own work on the nature of divine eternity. The Bible affirms that God is eternal in the sense that He is without beginning or end. He never came into nor will He ever go out of existence. But the Bible leaves it open as to whether we are to understand divine eternity as timelessness (atemporality) or as everlasting duration (omnitemporality). Typically, the Bible speaks of God’s being eternal in the sense of being omnitemporal, that is, enduring through every time that there is (Psalm 90.2). Indeed, Padgett in his study of divine eternity flatly concludes, “the Bible knows nothing of a timeless divine eternity in the traditional sense.”3 I think Padgett’s exegesis fails to reckon sufficiently with biblical texts that seem to contemplate time’s having had a beginning (Genesis 1.1; Proverbs 8.22–23; Jude 25; Titus 1.2–3; II Timothy 1.9). But the point remains that the Bible does not teach that God is timeless. The nature of divine eternity is a question to be answered by philosophical, not biblical, theology.
111
I have elsewhere argued that in being related to a temporal world, God, even if intrinsically changeless, nonetheless undergoes extrinsic change; that is to say, He changes in His relations with temporal things, and therefore cannot be timeless.4 Therefore, so long as a temporal world exists, God must exist in time. On my view, God is timeless without creation and temporal with creation.
Thus, ironically, this second “objection,” far from proving atheism, is actually one that I have defended even more vigorously than Sinnott-Armstrong. He does object that “nothing that exists equally, wholly, and indivisibly at all times can explain why an event occurs at one time as opposed to earlier or later.” At first we might think this assertion patently false. I, for example, exist equally, wholly, and indivisibly at every moment of my existence, and I obviously cause certain events to occur when they do rather than earlier or later. But the reason Sinnott-Armstrong makes his claim is because he is a determinist.5 Although he says he believes in free will, his understanding of freedom is a compatibilist, not a libertarian, understanding. On such a view, we are “free,” even though our decisions and actions are determined by antecedent causes, if those decisions and actions are voluntary, or in line with what we want. On a deterministic view, once all the causal factors sufficient for an effect exist, then the effect must exist, too. So if the cause of some event is equally, wholly, and indivisibly present at a time t o, then the effect cannot be delayed in its appearance until some later time t n; rather it must exist at t o. That’s why Sinnott-Armstrong says that if God is the cause of the Red Sea’s parting, then if God exists at midnight, the Red Sea must part at midnight.
The discerning reader will notice that this is basically the same reasoning I employed in defense of the personhood of the transcendent cause of the universe established by the cosmological argument.6 The origin of the universe cannot be explained as the effect of a deterministic cause, for if the cause were eternally present (in either a temporal or an atemporal sense), then the effect would be equally present as well. The only way to get an effect that begins in time from an eternal cause is by adopting a wholly different, indeterministic account of causation: agent causation.7 An agent endowed with libertarian free will can bring about new effects in time without deterministic, antecedent conditions. That is why agents
God?
(like ourselves or God) can exist equally, wholly, and indivisibly at various times and yet effect new and different events at various times.
It need only be added that Sinnott-Armstrong has said nothing to show that libertarian freedom or agent causation does not exist. So once again, the reasoning endorsed by Sinnott-Armstrong only goes to support the theistic position.
The Problem of Evil
Now we come to a really serious problem, atheism’s killer argument, the problem of evil. Since Sinnott-Armstrong makes it clear that he is not defining evil to have a moral dimension but means merely anything harmful, it will be less misleading if we characterize his argument as the problem of harm or suffering.
The problem of suffering is undoubtedly the greatest obstacle to belief in God—both for the Christian and the non-Christian. When I consider the depth and extent of suffering in the world, then I have to admit that it makes it hard to believe in God. No one who is not deranged can contemplate the many examples of innocent suffering listed by Sinnott-Armstrong and not be deeply moved and troubled by them. Their emotional impact is undeniable and makes it difficult to believe that an all-powerful and all-good God exists.
But as Sinnott-Armstrong recognizes, we as philosophers are called upon, not just to express how we feel about some issue, but to reflect rigorously and dispassionately about it. And despite the undeniable emotional impact of the problem of suffering, I am persuaded that as a strictly rational, intellectual problem, the problem of suffering does not constitute a disproof of the existence of God.
It will therefore be helpful if we distinguish between the intellectual problem of suffering and the emotional problem of suffering.
The intellectual problem concerns how to give a rational account of the co-existence of God and suffering. The emotional problem concerns how to dissolve the emotional dislike people have of a God who would permit such suffering. The intellectual problem lies in the province of the philosopher; the emotional problem lies in the province of the counselor. It’s important to keep this distinction clear because the solution to the intellectual problem is apt to appear dry and uncomforting to someone who is going through suffering, whereas the solution to the emotional problem is apt to appear su-
113
perficial and explanatorily deficient to someone contemplating the question abstractly.
Now, philosophers usually distinguish two ways in which the intellectual problem of suffering may be cast, the logical version or the evidential version. I think the traditional distinction can be more clearly and accurately made, however, if we make a distinction between the problem of suffering as an internal problem or as an external problem for Christian theism. That is to say, the problem may be presented in terms of premises to which the Christian theist is committed as a Christian, so that the Christian world view is allegedly somehow at odds with itself. On the other hand, the problem may be presented in terms of premises to which the Christian theist is not committed as a Christian but which we nonetheless have good reason to regard as true. The first approach tries to expose an inner tension within the Christian world view itself; the second approach attempts to present evidence against the truth of the Christian world view.
It’s worth emphasizing that traditionally, atheists have claimed that the problem of suffering constitutes an internal problem for Christian theism. That is, atheists have claimed that the statements A. An all-powerful, all-good God exists.
and
B. Suffering exists (in various kinds and quantities).
are either logically inconsistent or improbable with respect to each other. As a result of the work of Christian philosophers like Alvin Plantinga, it is today widely recognized that the internal problem of suffering is a failure as an argument for atheism. No one has ever been able to show that (A) and (B) are either logically incompatible with each other or improbable with respect to each other. For there is no way to prove, given the existence of God, that it is impossible or improbable that God has some morally sufficient reason for permitting any observed instance of suffering. This is a tremendous encouragement as we deal with remaining questions on which no consensus yet exists (and constitutes as well a decisive refutation of the attitude that philosophy never makes any progress!).
Having abandoned the internal problem of suffering, atheists have
God?
very recently taken to advocating the external problem, often called the evidential problem of evil. Sinnott-Armstrong’s argument represents a statement of this external problem. His four-point argument can be simplified if we take God to be essentially all-powerful and all-good and call suffering that is not necessary to achieve some adequately compensating good “gratuitous suffering.” His argument can be summarized in three steps:
1. If God exists, gratuitous suffering does not exist.
2. Gratuitous suffering exists.
3. Therefore, God does not exist.
Sinnott-Armstrong’s argument is not that suffering itself disproves God. Rather he’s talking about a very special kind of suffering, namely uncompensated, unnecessary suffering. It is this kind of suffering that is said to be incompatible with God’s existence. Sinnott-Armstrong maintains that this special kind of suffering exists, and therefore God does not exist. What makes this an external problem is that the Christian is not committed by his world view to the truth of (2). The Christian is committed to the truth that suffering exists, but not that gratuitous suffering exists. Thus the atheist claims that the apparently uncompensated and unnecessary suffering in the world constitutes evidence against God’s existence.
Now, the most controversial premise in this argument is step (2).
Everybody admits that the world is filled with apparently gratuitous suffering. We are often unable to see any reason for why harm be-falls us. But that doesn’t imply that these apparently gratuitous evils really are gratuitous. Every one of us can think back on experiences of suffering or hardship in our lives, which at the time seemed pointless and unnecessary but, when viewed in retrospect, are seen to have been ultimately to our or others’ advantage, even if we would not want to go through them again. The key move in the atheist’s argument, then, will be his inference from apparently gratuitous suffering to genuinely gratuitous suffering.
In examining this key inference, we should do well to note two general weaknesses in Sinnott-Armstrong’s style of argument. First, he typically reasons by analogy, comparing God’s relation to us with our relation to our neighbor. The problem with this method of argument in general is that any disanalogy in the comparison under-
115
mines the force of the argument. And in this case in particular God’s relationship to us is strikingly disanalogous to our relationships to our neighbors. For we and our neighbors are peers; but God is our Creator and Sovereign. Thus, I am not charged with the moral education of my neighbor; but God aims that we should learn to become mature moral agents. I am not responsible for the eternal salvation of my neighbor; but God is involved in drawing not only my neighbor but all people to a saving knowledge of Himself. I am bound by certain moral obligations and prohibitions vis à vis my neighbor (e.g., not to take his life); but God (if He has moral duties at all) is not bound by many of these (e.g., He can give and take life as He pleases). These and other disanalogies considerably weaken Sinnott-Armstrong’s argument from analogy.
The tenuousness of such analogical reasoning is particularly egre-gious in his criticism of what he calls the “Glorious Response.” A human father who thought of himself as the end-all of his children’s existence would indeed be egomaniacal. But the infinite God, who is the locus of goodness and love, is the appropriate end of all beings (even of Himself !), the summum bonum (highest good). Of course, God cares whether we are happy, and He knows that true happiness and human fulfillment are found in our knowing the highest good. Therefore, His loving purpose for humankind is to draw people freely into a saving relationship with Himself. The real question is, as Sinnott-Armstrong asserts, whether “God can surely find some more effective and efficient way to lead people to glorify Him.”
I think we have no inkling whatsoever that this is the case: it is sheer speculation. Be that as it may, the analogy of the human father is so deeply flawed as to be worthless.
Second, Sinnott-Armstrong’s method of dealing with proffered solutions to the problem of suffering is the strategy of “divide and conquer.” That is to say, he typically refutes a solution by arguing that it cannot cover all cases of suffering. What he fails to appreciate is that God’s reasons for permitting suffering are doubtlessly complex and multifarious. That implies that no one solution is intended to cover all cases of suffering. So take, for example, what Sinnott-Armstrong calls the “Virtuous Response.” Like it or not, it is an undeniable fact that moral development is impossible without hardship and suffering; without them we would be pampered and
God?
spoiled children, heedless of the consequences of our actions and indifferent toward others, rather than responsible and wise adults with a virtuous moral character. Sinnott-Armstrong responds in part by charging that such a solution is “unfair” to babies who suffer and die for the moral development of parents. But that charge falls to the ground when the Virtuous Response is conjoined with the so-called “Heavenly Response,” for the child receives a compensation so incomprehensibly great that it is incommensurable with the suffering. Sinnott-Armstrong’s complaint about the Heavenly Response is that God could send babies directly to heaven. But that forgets the Virtuous Response: so doing would short-circuit God’s work in the lives of the parents or other people. What Sinnott-Armstrong’s
“divide and conquer” strategy fails to understand is that the solutions are complementary and mutually reinforcing.
Apart from these general problems of Sinnott-Armstrong’s methodology, let me suggest three reasons why I think that the inference from apparently gratuitous suffering to genuinely gratuitous suffering is tenuous.
1. We are not in a good position to assess with confidence the probability that God lacks morally sufficient reasons for permitting the suffering in the world. Whether the suffering in the world really is gratuitous depends on whether God has morally sufficient reasons for permitting the suffering that occurs. What makes the probability here so difficult to assess is that we are not in a good position to make these kinds of probability judgments with any sort of confidence. The philosopher William Alston has surveyed six cognitive limitations under which we labor which impair such judgments, including: lack of data, complexity greater than we can handle, difficulty in knowing what is metaphysically possible, ignorance of the range of possibilities, ignorance of the range of values, and limits of our capacity to make well-considered value judgments. He concludes, “The judgments required by the inductive argument from evil are of a very special and enormously ambitious type, and our cognitive capacities are not equal to this one. . . . it is in principle impossible for us to be justified in supposing that God does not have sufficient reasons for permitting evil that are unknown to us.”8
117
Take just the first of these limits: our lack of data. As finite persons limited in space and time, we experience a brief glimpse of the history of the world. But in the Christian view, the transcendent and sovereign God sees the end of history from its beginning and providentially orders history so that His purposes are ultimately achieved through human free decisions. Only an omniscient mind could grasp the complexities of directing a world of free creatures toward one’s pre-visioned goals. One has only to think of the innumerable, incalculable contingencies involved in arriving at a single historical event, say, the Allied victory at D-day. In order to bring about this event through the actions of free creatures, God might well have to put up with myriad evils along the way. We simply have no idea of the natural and moral evils that might be involved in order for God to arrange the circumstances and the free agents required to achieve such an intended purpose. Suffering that appears pointless or unnecessary to us within our limited framework may be seen to have been justly permitted from within God’s wider framework.
To borrow an illustration from a developing field of science, Chaos theory, scientists have discovered that certain macroscopic systems—
for example, weather systems or insect populations—are extraordinarily sensitive to the tiniest perturbations. A butterfly fluttering on a branch in West Africa may set in motion forces that would eventually result in a hurricane over the Atlantic Ocean. Yet it is impossible in principle for anyone observing that butterfly palpitating on a branch to predict such an outcome.9 The brutal murder of an innocent man or a child’s dying of leukemia could send a ripple effect through history so that God’s morally sufficient reason for permitting it might not emerge until centuries later or perhaps in another country. Given our limited perspective, we cannot always discern what reasons a provident God might have in mind for permitting some harm to enter our lives. Certainly, much suffering seems pointless and unnecessary to us—but we are simply not in a position to say.
In this light, we see the shortcoming of Sinnott-Armstrong’s reply to what he calls the “Modest Response.” The problem with his appeal to human analogies is precisely that the neighbor shares the same cognitive limitations that we do, and therefore we can confidently judge that he lacks any morally sufficient reason for letting
God?
his child starve. Thus, Sinnott-Armstrong’s conclusion that the Modest Response allows anyone to “defend any action, no matter how horrible, by saying that there is some reason that we do not see,”
does not follow. He admits that in the case of a provident God some adequate compensation might arise for an apparently gratuitous evil centuries later by a causal chain we do not discern; but he insists that “we should not let uncertainty stop us from judging and acting when we have strong evidence.” But the question is, of course, whether we do or even can have such evidence in the case of God.
What, exactly, would constitute “strong evidence” that God in His providence does not have a morally sufficient reason, which will emerge in the future for permitting some harm in the present? I doubt that Sinnott-Armstrong could provide any criteria for what constitutes such strong evidence.
His realization of the shortcoming of his argument becomes clear when he tries once more to shift the burden of proof: future compensation is possible, he admits, “but we have no more reason to believe that the evil now will lead to great bliss in the long term than that this evil will lead to great misery.” Any positive argument for atheism is herewith quietly abandoned. My point is precisely that we are not in a position to make these sorts of probability judgments either way. We do not know the outcome of even seemingly trivial events, whether good or ill. Even contemporary filmmakers seem to appreciate this truth, as evidenced in the movie “Sliding Doors,” in which a young woman’s life takes two radically different courses depending upon the seemingly trivial event of whether she makes it through the sliding doors of a subway train she wants to catch. One course of her life leads to suffering and disaster, while the other leads to prosperity and success—all based upon whether a little girl play-ing with her dolly blocks the young woman’s path for a split second as she hurries to the train! Most intriguingly, however, we discover in the movie’s end that it is really the life of suffering and hardship that turns out to be the best life, while the seemingly happy life leads to her premature death. The film brilliantly illustrates the impossibility of our forming with any confidence probability judgments that some particular instance of suffering is in fact gratuitous and that God does not have a morally sufficient reason for permitting it.
119
To say this is not to appeal to mystery, but rather to point to the inherent cognitive limitations that frustrate attempts to say that it is improbable that God has a morally sufficient reason for permitting some particular harm. As Alston says, “We are simply not in a position to justifiably assert that God would have no sufficient reason for permitting evil.”10 Once we reflect on God’s providence over the whole of history and on how seemingly insignificant events can de-rail the course of history with ever-widening consequences, then it becomes evident how hopeless it is for limited observers to speculate on the probability of God’s lacking morally sufficient reasons for the suffering that we see. We are simply not in a good position to assess such probabilities with any confidence.
It’s interesting that Sinnott-Armstrong’s final attempt to deal with the Modest Response is to complain once more of the hiddenness of God: God, if He existed, would tell us what His morally sufficient reasons are for permitting suffering to overtake us. And once again I think we can repose no confidence at all in such a presumptuous assertion. Moreover, the demand is wholly unreasonable. First, since there is no one reason for God’s permitting suffering, such a revelation would have to be imparted to every person on every occasion of suffering throughout the history of the world. That would turn the universe into a haunted house. Second, such a revelation might be self-negating in many cases. If God were to reveal to us why He permitted us to suffer some particular evil, then our having that knowledge might lead to the frustration of God’s purpose. For example, the free agents might then act in ways other than they would have acted if they had lacked the knowledge, so that as a result, a future compensating event might not take place—in which case that could not have been the reason why God permitted the harm! Thus, the envisioned revelation might lead to self-stultifying situations. The demand is then logically impossible to meet.
2. Christian theism entails doctrines that increase the probability that suffering would appear to be gratuitous, even though it’s not.
This makes it all the harder for Sinnott-Armstrong to justifiably infer from the appearance of gratuitous evil to the fact of gratuitous evil. What are some of these doctrines? Let me mention three:
God?
Doctrine #1: The purpose of life is not human happiness as such, but knowing God.
One reason that the suffering in the world seems pointless is that we tend naturally to assume that if God exists, then His purpose for human life is happiness in this world. God’s role is to provide a comfortable environment for His human pets. But in the Christian view, this is false. We are not God’s pets, and the goal of human life is not happiness as such, but the knowledge of God—which in the end will bring true and everlasting human fulfillment. Many evils occur in life that are utterly gratuitous with respect to the goal of producing human happiness; but they may not be gratuitous with respect to producing a deeper knowledge of God. Innocent human suffering provides an occasion for deeper dependency and trust in God, either on the part of the sufferer or those around him. Of course, as Sinnott-Armstrong reminds us, whether God’s purpose is achieved through our suffering will depend on our response. Do we respond with anger and bitterness toward God, or do we turn to Him in faith for strength to endure? God may well be less concerned with what we go through than with our attitude while going through it.
Because God’s ultimate goal for humanity is the knowledge of Himself—which alone can bring eternal happiness to creatures—
history cannot be seen in its true perspective apart from the Kingdom of God. Martyn Lloyd-Jones has written, The key to the history of the world is the kingdom of God. . . . From the very beginning, . . . God has been at work establishing a new kingdom in the world. It is His own kingdom, and He is calling people out of the world into that kingdom: and everything that happens in the world has relevance to it. . . . Other events are of importance as they have a bearing upon that event. The problems of today are to be understood only in its light. . . .
Let us not therefore be stumbled when we see surprising things happening in the world. Rather let us ask, ‘What is the relevance of this event to the kingdom of God?’ Or, if strange things are happening to you personally, don’t complain but say, ‘What is God teaching me through this?’ . . . We need not become bewildered and doubt the love or the justice of God. . . . We should . . . judge every event in the light of God’s great, eternal and glorious purpose.11
121
It may well be the case that natural and moral evils are part of the means God uses to draw people into His Kingdom. A reading of a missions handbook, such as Patrick Johnstone’s Operation World, reveals that it is precisely in countries that have endured severe hardship that evangelical Christianity is growing at its greatest rate, while growth curves in the indulgent West are nearly flat. Consider, for example, the following reports:12
China: It is estimated that 20 million Chinese lost their lives during Mao’s Cultural Revolution. Christians stood firm in what was probably the most widespread and harsh persecution the Church has ever experienced. The persecution purified and indigenized the Church. Since 1977, the growth of the Church in China has no parallels in history. Researchers estimated there were 30–75 million Christians by 1990. Mao Zedong unwittingly became the greatest evangelist in history.
El Salvador: The 12-year civil war, earthquakes, and the collapse of the price of coffee, the nation’s main export, impoverished the nation. Over 80% live in dire poverty. An astonishing spiritual har-vest has been gathered from all strata of society in the midst of the hate and bitterness of war. In 1960, evangelicals were 2.3% of the population, but today are around 20%.
Ethiopia: Ethiopia is in a state of shock. Her population struggles with the trauma of millions of deaths through repression, famine, and war.
Two great waves of violent persecution refined and purified the Church, but there were many martyrs. There have been millions coming to Christ. Protestants were fewer than 0.8% of the population in 1960, but by 1990 this may have become 13% of the population.
Examples such as these could be multiplied. The history of mankind has been a history of suffering and war. Yet it has also been a history of the advance of the Kingdom of God. Figure 1 is a chart re-leased in 1990 by the U.S. Center for World Mission documenting the growth in evangelical Christianity over the centuries.
According to Johnstone, “We are living in the time of the largest ingathering of people into the Kingdom of God that the world has ever seen.”13 It is not at all improbable that this astonishing growth
God?
The Diminishing Task
—Across the centuries, the constantly decreasing number of non-Christians
per committed Christian
Non-Christian here means “people who do not consider themselves to be 360
Christians.”
Committed Christians here means people who read, believe, and obey the Bible.
220
The specific numbers here are correct within a small 100 AD
percentage, except for the earlier centuries.
69
1000 AD
27
21
Where do these
1500 AD
amazing numbers come
1900 AD
11
from? They were arrived at by the various contributors to the Lausanne Statistics 1950 AD
Task Force, headed by David Barrett, Ph.D., who is the 7
author of the World Christian Encyclopedia. The specific 1980 AD
figures are in the table below. The numbers in the column on the right are those used in this diagram.
1989 AD
Column 1
Column 2
Column 3
Astounding
DATE
Non-Christians Com. Christians
Trend
100 AD
180
0.5
360
1000 AD
220
1
220
1500 AD
344
5
69
These are the numbers in the diagram
1900 AD
1,062
40
27
above. Despite the rapid increase of world 1950 AD
1,650
80
21
population, Christianity is simply growing 1980 AD
3,025
275
11
faster than any other global religion when 1989 AD
3,438
500
7
what is measured is its most relevant type The above numbers are published by the
(Column 2
of growth—the growth of committed
Lausanne Statistics Task Force.
divided by
adherents.
Column 3)
Note: figures in these two columns are millions.
Fig. 1. The number of evangelical Christians in the world compared to the number of non-Christians in the world over the course of human history. Neither category includes merely nominal Christians. But even in a worst-case scenario, where all the nominal Christians were included with the non-Christians as needing to be evangelized, there would still today be only about nine non-believers for every evangelical believer in the world. Undoubtedly, God’s invisible Kingdom extends far beyond the boundaries of evangelicalism.
( Mission Frontier magazine, November 1990, www.missionfrontier.org) in God’s Kingdom is due in part to the presence of natural and moral evils in the world. It may well be the case that only in a world involving natural and moral evil that is gratuitous with respect to producing worldly happiness that the maximum number of people will freely come to know God and His salvation. To carry his argument,
123
Sinnott-Armstrong would have to show that it is feasible for God to create a world in which the same amount of the knowledge of God is achieved, but with less suffering. This is sheer speculation.
Doctrine #2: Mankind is in a state of rebellion against God and His purpose.
Rather than submit to and worship God, people rebel against God and go their own way, and so find themselves alienated from God, morally guilty before Him, and groping in spiritual darkness, pursuing false gods of their own making. The terrible human evils in the world are testimony to man’s depravity in this state of spiritual alien-ation. The Christian is not surprised at the moral evil in the world; on the contrary, he expects it. The Bible says that God has given mankind over to the sin it has chosen; He does not intervene to stop it; He lets human depravity run its course. This only serves to heighten mankind’s moral responsibility before God, as well as our own wickedness and our need of forgiveness and moral cleansing.
Doctrine #3: God’s purpose spills over into eternal life.
In the Christian view, this life is not all there is. Jesus promised eternal life to all who place their trust in him as their Savior and Lord. In the afterlife, God will reward those who have borne their suffering in courage and trust with an eternal life of unspeakable joy. The apostle Paul, who wrote much of the New Testament, lived a life of incredible suffering. Yet he wrote, “We do not lose heart.
For this slight, momentary affliction is preparing us for an eternal weight of glory beyond all comparison, because we look not to the things that are seen, but to the things that are unseen, for the things that are seen are transient, but the things that are unseen are eternal” (II Corinthians 4.16–18). Paul imagines a scale, as it were, in which all the sufferings of this life are placed on one side, while on the other side is placed the glory that God will bestow on His children in heaven. The weight of glory is so great that it is literally beyond comparison with the suffering. Moreover, the longer we spend in eternity the more the sufferings of this life shrink by comparison toward an infinitesimal moment. That’s why Paul could call them a
“slight, momentary affliction”—they were simply overwhelmed by the ocean of divine eternity and joy that God lavishes on His children in heaven. Given the prospect of eternal life, we shouldn’t ex-
God?
pect to see in this life God’s reasons for permitting every evil. Some may be justified only in light of eternity.
Given these three doctrines, we should expect much of the suffering in the world to appear gratuitous. In order to show that it really is gratuitous, Sinnott-Armstrong would have to refute these three doctrines, which he hasn’t even tried to do.
3. The most important factor in weighing whether the suffering in the world is really gratuitous will be—ironically—whether God exists!
That is to say, Sinnott-Armstrong’s own argument shows that if God exists, then the suffering in the world is not gratuitous. Thus, we may argue:
1. If God exists, gratuitous suffering does not exist.
2*. God exists.
3*. Therefore, gratuitous suffering does not exist.
Thus, if God exists, then the apparently gratuitous suffering in the world is not really gratuitous.
So the issue comes down to which is true: (2) or (2*)? In order to prove that God does not exist, Sinnott-Armstrong would have to show that (2) is considerably more probable than (2*). As Daniel Howard-Snyder points out in his book The Evidential Problem of Evil, an argument from evil is a problem only for the theist “who finds all its premises and inferences compelling and who has lousy grounds for believing theism.”14 But if you have better reasons for believing that God exists, then, he says, evil “is not a problem.”15
In response to what he calls the “Overriding Response,” Sinnott-Armstrong again appeals to human analogies. He says that just because you had evidence that your neighbor was a good person, you wouldn’t believe that you were hallucinating when you saw your neighbor beat his child. This analogy is flawed in two respects. First, it is unjustified to compare God’s permitting harm and the neighbor’s perpetrating harm. It is clearly emotionally prejudicial to put God in the role of the abusive neighbor. Second, God’s having a morally sufficient reason for permitting evil is not analogous to your having a hallucination of your neighbor. (The hallucination analogy is more appropriate to a Hindu view that conceives of evil as part of maya, the realm of mere appearance.) Rather a better analogy would be if you lived next to the Sinnott-Armstrongs and one day you look
125
next door and see Walter Sinnott-Armstrong standing there while someone else is beating his daughter, and he does nothing to help.
Should you conclude, despite your previous evidence to the contrary, that Walter Sinnott-Armstrong is really an evil monster? Or wouldn’t you suspect that Walter has a morally sufficient reason for allowing this to happen (for example, that there are intruders in his house and one of them has a gun to his wife’s head and will shoot if Walter tries to stop the beating)? In the same way, if we have good evidence that God exists, then that increases the probability that He has a morally sufficient reason for permitting apparently gratuitous suffering. So you must balance the probability that apparently gratuitous suffering really is gratuitous against the probability that God exists. Even if, despite our cognitive limitations, we assign some positive probability to the former, the latter may still outweigh it.
Sinnott-Armstrong tries to block this route by saying that my arguments on behalf of theism do not prove that God is all-powerful and all-good. But my moral argument, if cogent, certainly does prove that God is all-good; and the cosmological argument certainly shows that the Creator of the universe out of nothing has sufficient power to prevent the evils that afflict our world, so that the success of my arguments does count against the gratuity of the suffering we observe. Moreover, my argument from the life, death, and resurrection of Jesus and my appeal to the proper basicality of belief in God go to support the existence of the biblical God, who is declared to be holy and almighty. In any case, Sinnott-Armstrong’s atheism wouldn’t be much of an atheism if it were consistent with the existence of a personal Creator and Designer of the universe, who is the locus of absolute value and who has revealed Himself in Jesus Christ and Christian experience! Remember, the problem of evil is supposed to be a positive argument for atheism.
In fact, far from being a positive argument for atheism, evil itself turns out, I believe, to be a positive argument for theism. For much of the suffering in the world is moral in nature; that is to say, the suffering inflicted by people on their innocent victims is genuinely evil. But then we may argue as follows:
1. If God does not exist, objective moral values do not exist.
2. Evil exists.
God?
3. Therefore, objective moral values exist—namely, some things are evil!
4. Therefore, God exists.
Thus, evil paradoxically goes to prove God’s existence, since without God things would not be good or evil as such. Notice that this argument thus shows the compatibility of God and evil without giving us a clue as to why God permits evil. We could be as ignorant of God’s reasons as was Job in the Old Testament. But even in the absence of any answer to the “why” question, the present argument proves that evil does not call into question, but actually requires, God’s existence. So although, superficially, evil seems to call into question God’s existence, at a deeper level it actually proves God’s existence, since without God, evil as such would not exist.
So, in summary, we’ve seen that Sinnott-Armstrong’s attempt to prove the second premise of his argument: 2. Gratuitous suffering exists
is by no means successful.
But it should also be noted that his first premise: 1. If God exists, gratuitous suffering does not exist itself is not obviously true. Some theists have suggested that while God could eliminate this or that specific instance of suffering without decreasing the goodness of the world, nevertheless there must exist a certain amount of gratuitous suffering in the world if the goodness of the world is not to be impaired. Thus the probability that a certain specified instance of suffering is gratuitous would not ad-versely affect theism. Indeed, it’s not at all improbable that only in a world in which gratuitous natural and moral evils exist that the optimal number of persons would freely come to salvation and the knowledge of God. Sinnott-Armstrong might say that in that case the suffering is not really gratuitous after all: it serves the greater good of securing people’s eternal salvation. But if one allows a greater good of that sort to count against the gratuity of suffering, then that makes it all the more difficult for one to prove that truly gratuitous suffering exists, for how could we possibly surmise what in God’s providential plan for history does or does not contribute to the ultimate salvation of the greatest number of people?
127
In conclusion, the intellectual problem of evil—whether in its internal or external versions—can be satisfactorily solved, and so does not constitute a proof of atheism.
To wrap up, none of Sinnott-Armstrong’s three arguments for atheism is, at the end of the day, rationally compelling or, at least, as rationally compelling as my five reasons in support of Christian theism. I, therefore, believe that the scales of the evidence tip in favor of theism rather than atheism.
Notes
1. The plausibility of the principle trades on the ambiguity of the word “evidence.” If we mean that everything we believe must be inferentially derived by sound arguments from other beliefs, then contemporary epistemologists recognize that so stringent a demand would lead to skepticism, since even such obviously true beliefs as that the external world exists or that the past is real cannot be proven in this way. Rather such beliefs are properly basic beliefs not founded on argument. On the other hand, if we take “evidence” so broadly as to include the circumstances that ground properly basic beliefs, then it is not true that belief in God is not grounded by evidence. For a good discussion, see Alvin Plantinga, “Reason and Belief in God,” in Faith and Rationality, ed. Alvin Plantinga and Nicholas Wolterstorff (Notre Dame, Ind.: University of Notre Dame Press, 1983), 16–93.
2. See the helpful booklet by Paul K. Moser, Why Isn’t God More Obvious? (Atlanta: RZIM, 2000).
3. Alan G. Padgett, God, Eternity, and the Nature of Time (New York: St.
Martin’s Press, 1992), 33.
4. See my Time and Eternity: Exploring God’s Relationship to Time (Wheaton, Ill.: Crossway, 2001), 86–109, as well as the discussion with Paul Helm, Alan Padgett, and Nicholas Wolterstorff in Gregory Ganssle, ed., God and Time (Downer’s Grove, Ill.: Inter-Varsity Press, 2001).
5. He states this during our debate “Do Suffering and Evil Disprove God?”
(April 2, 2000), Wooddale Church, Eden Prairie, Minn.
6. See pp. 5–6.
7. See further J. P. Moreland, “Libertarian Agency and the Craig/Grün-baum Debate about Theistic Explanation of the Initial Singularity,” American Catholic Philosophical Quarterly 71 (1998): 539–554.
8. William Alston, “The Inductive Problem of Evil,” Philosophical Perspectives 5 (1991): 65, 59.
9. This fact was already appreciated even in classical physics, as James Clerk Maxwell discerned:
“In all such cases there is one common circumstance—the system has a quantity of potential energy, which is capable of being transformed into
God?
motion, but which cannot begin to be so transformed till the system has reached a certain configuration, to attain which requires an expenditure of work, which in certain cases may be infinitesimally small, and in general bears no definite proportion to the energy developed in consequence thereof. For example, the rock loosed by frost and balanced on a singular point of the mountain-side, the little spark which kindles the great forest, the little word which sets the world a-fighting, the little scruple which prevents a man from doing his will, the little spore which blights all the potatoes, the little gemmule which makes us philosophers or idiots. Every existence above a certain rank has its singular points: the higher the rank, the more of them. At these points, influences whose physical magnitude is too small to be taken account of by a finite being, may produce results of the greatest importance.”
(J. C. Maxwell, “Science and Free Will,” quoted in L. Campbell and W. Garnett, The Life of James Clerk Maxwell [London: Macmillan, 1882], 443.) 10. Alston, “The Inductive Problem of Evil,” p. 61.
11. David Martyn Lloyd-Jones, From Fear to Faith; Studies in the Book of Habbakuk (London: Inter-Varsity Fellowship, 1953), 23–24.
12. Patrick Johnstone, Operation World (Grand Rapids, Mich.: Zondervan, 1993), 164, 207–208, 214.
13. Ibid., p. 25.
14. Daniel Howard-Snyder, “Introduction,” in The Evidential Argument from Evil, ed. Daniel Howard-Snyder (Bloomington: Indiana University Press, 1996), xi.
15. Ibid.
Atheism Undaunted
Walter Sinnott-Armstrong
“You just don’t get it,” say some women to men who fail to understand their point of view. I feel like saying the same thing to Craig.
He tries to respond to my three arguments, but sometimes he misses the point, and other times his reply leads him to a position that strikes me as desperate.
The Real Argument from Ignorance
Craig charges my argument from ignorance with a “subtle but crucial shift.” (107) However, this criticism misrepresents my argument.
Craig’s point is that a certain principle—“we should not believe in entities for which we have no evidence”—“at best supports agnosticism, not atheism.” (107) Of course, this principle by itself can’t yield atheism! That’s why my argument did not stop with this principle.
Perhaps I was not clear enough, but the earlier passages in my section on “The Argument from Ignorance” were merely supposed to set the stage for my argument. After all, atheists need to agree with agnostics on one thing, namely, that we should not believe in God.
My argument for atheism comes later. When Craig does discuss my later argument, he accuses me of “a bizarre role reversal” in my discussion of gold on Uranus. (108) This discussion would have been bizarre if I had claimed that God was like gold on Uranus. But I never claimed anything like that. My point was to draw a crucial contrast between the issue of gold on Uranus and the issue of God’s existence. The difference, as I tried to make clear, is that a lack of ev-129
God?
idence for gold on Uranus would not show that we should believe that there is no gold on Uranus, whereas the lack of evidence for God does show that we should believe that there is no God. That is why I said that some arguments from ignorance are fallacious, but others are not.
This crucial difference is captured by a premise in my argument that Craig ignored until later in his discussion. With this premise added, my argument takes this form:
1. If there were an all-good and all-powerful God who could act in time, then we would have strong evidence for the existence of God.
2. We do not have strong evidence for the existence of God.
3. Hence, there is no all-good and all-powerful God who can act in time.
Background premise (1), which can be called the strong evidence principle, is what enables my argument to reach the conclusion of atheism and not just agnosticism. Its basic idea is simply that, if God existed, He would make it easier for us to know Him. God would not hide his light under a bushel, for the same reasons that Jesus tells his followers not to hide their lights under bushels. (Matthew 5:15–16) The point of the argument from ignorance is that God Himself does not follow Jesus’ advice!
The strong evidence principle is, admittedly, controversial. Some theists might deny it. That is why I argued for it. I listed many benefits of God revealing Himself more clearly to us. For example, believers would suffer less from doubt, and some potential criminals would be deterred by fear of divine punishment. I also argued against many commonly claimed costs of strong evidence for God, including supposed losses of faith and freedom. My point there was that the evidence for God could be much stronger than it is without undermining freedom or faith. These arguments showed that strong evidence for God would make our lives better overall, so an all-good, all-powerful God would give us strong evidence for God, just as the strong evidence principle claims.
Once premise (1) is established, all I need is premise (2). That premise was supported by my arguments in Chapter 2, which refuted the best evidence for God that Craig could produce in Chap-
131
ter 1. As I write this chapter, I have not yet seen Craig’s response in Chapter 3 to my criticisms in Chapter 2 of his arguments in Chapter 1; so I cannot yet respond to his response. But that, of course, does not mean that Craig has succeeded in defending his arguments against my criticisms. Indeed, even if Craig (or you) did find something wrong with the particular criticisms that I gave, that does not show that Craig’s arguments do not fail in other ways or that his arguments provide strong evidence for a traditional God. If his arguments provide any evidence at all, it isn’t strong.
One major flaw, as I have said, is that Craig’s conclusions are badly bloated. Even if his arguments seem relevant to several (not all!) traditional features of God, all of his arguments together still cannot show that one single God has all of these features together.
The creator of the universe (even if there were one) need not be the ground of moral value or the source of anyone’s religious experiences today, much less the force that raised Jesus from the dead (even if this did happen). Craig cannot legitimately assume that all of his conclusions about various features apply to a single unified being. But he does assume this. This assumption is hidden when Craig formulates his arguments in terms of “God,” since to refer to God in all of his arguments is to assume that there is a single being with all of these features. If Craig’s conclusions instead referred to a creator, a ground of moral value, an external source of religious experience, and a Jesus-raiser, then it would be clear that these might be separate beings and that Craig has no reason to assume that these are all the same person (much less that the Christian Bible gives an accurate picture of that person). That conclusion goes far beyond anything that his arguments could establish even if they did work (which they do not). Moreover, this gap in his argument leads directly to the problem of ignorance: If there were an all-good and all-powerful God, He could, should, and would give us strong reason to believe that He exists as a single unified being. The fact that we have no strong evidence for a unified being with all of these features, thus shows that there is no such being.
Some readers still might be impressed by Craig’s arguments that depend on recent scientific advances or historical scholarship. These references might seem sophisticated, but they feed right into my argument from ignorance. To see why, think back 200 years to times
God?
before those scientific theories were formulated, when nobody yet had heard of a Big Bang. People at those times could not use Craig’s scientific arguments; so, even if those arguments do work today, those earlier people did not have any strong evidence for the existence of God. Thus, premise (2) was true for them. Premise (1) is also true for them, since any all-good God would care as much about them as about us, so He would reveal Himself to them. He would have no reason to let so many people in the past remain ignorant of Him for so long. Thus, premises that are only about those earlier people would be enough to reach my conclusion that there is no all-good and all-powerful God. Craig’s appeals to recent science cannot solve this problem of ignorance.
It is still possible that other arguments, different from the ones that Craig gave, provide strong evidence that God exists. However, this is merely a possibility. Until we actually see much better arguments for the existence of God, there is reason to accept premise (2). With premises (1) and (2) in place, atheism follows.
Of course, Craig rejects both premises, but that does not show that those premises are flawed. In Chapter 5, at least, Craig has not given us any good reason to doubt either premise. He does call premise (1) “enormously presumptuous,” and he asks, “why should [God]
want to do such a thing?” (109) I already answered that question when I discussed the many benefits for humans of God revealing himself more clearly. So Craig’s rhetorical question has no force at this point in our discussion.
Craig rejects my answer because, “in the Christian view, it is a matter of relative indifference to God whether people (merely) believe that He exists or not.” (109) The point of his qualification
“merely” seems to be that belief in God without love of God will not satisfy God: “what God is interested in is building a love relationship with you, not just getting you to believe that He exists.”1 (109) Fine, but God still should have some interest in mere belief, even if He prefers more than mere belief. One reason is that belief in God, even without love of God, has important benefits for humans. Imagine a contract killer who does not believe in God. Suppose that, if this contract killer did believe in God, then he would not love God, but he still would refrain from murder because he would fear punishment by God. This contract killer might be better off believing
133
in God. Even if this killer did not benefit, this killer’s potential victims would be better off if this killer believed in God. Thus, if God cares about the victims, then God has reason to make Himself more manifest to this killer and to other wrongdoers, regardless of whether they love Him.
Furthermore, even if mere belief is not sufficient to satisfy God, it is still necessary. You cannot have a love relationship with anything unless you believe that it exists. I can’t love my sister if I don’t know that I have a sister. Consequently, even if God does not want us “just” or “merely” to believe in Him, God still would want us to believe in Him, as a necessary condition for loving Him. This makes it hard to see any way around the conclusion that God cares about whether humans believe that God exists.
Here is where Craig’s second response kicks in: “there is no reason at all to think that if God were to make His existence more manifest, more people would come into a saving relationship with Him.”
(109) No reason at all? Here’s one: Those who now love and believe in God would not lose their faith or love for God if His existence were made more manifest in the proper way. Some others who do not now believe in God would come to believe in Him if His existence were more manifest. Not all of these new believers will love God, but some will, especially if God reveals His goodness. So there would be more people who believe in God and love God if His existence were more manifest.
The only way in which this argument could fail is if the new evidence made enough people lose their love for God. Craig talks about
“a neon cross in the sky” and “brazen advertisements” that make people “chafe” and “resent such effrontery.” (109) This parody misses my point. Such bungling is beneath God. Surely an all-knowing and all-powerful God could find and use some way to make Himself more manifest, at least to non-believers, without undermining believers’
love for Him. God could appear to each person with just enough evidence of the right kind to convince that individual. Or, at least, God could provide rapists and murderers with enough evidence to dis-suade them from their rampages. This evidence need not be seen by non-criminal believers, in which case the evidence for non-believers could not turn those believers against God. Moreover, even if God did give extra evidence to believers, possibly in order to relieve their
God?
doubts, this evidence could be limited to whatever is appropriate.
An all-powerful God wouldn’t need to be “brazen” in order to give people better evidence than they now have for His existence. If God gave additional evidence in some proper way, then “more people would come into a saving relationship with Him” (109), and humans would benefit in many ways at little or no cost to anyone. I might not be able to specify exactly how God would do this, but an all-knowing God could figure out how to do it, an all-powerful God could do it, and an all-good God would do it.
Thus, if there were an all-powerful and all-good God who could act in time, then we would have strong evidence for the existence of God. This is premise (1), which I called the strong evidence principle, so that premise is secure. As I said, premise (2) is supported by my series of arguments in Chapter 2. Together these premises imply atheism. Nothing that Craig has said undermines that argument from ignorance.
The Continuing Problem of Action
Craig does not question my argument that a timeless God cannot be causally active in the temporal world. Indeed, Craig supports it with citations to his own previous publications. This agreement might seem inconsequential, because Craig does not endorse the view of God as timeless. However, many traditional theologians throughout the centuries and even today hold that God is timeless. Since Craig agrees that my argument refutes that popular version of traditional theism, he must grant that my argument accomplishes a lot.
Of course, Craig does not grant this argument any force against his own view of God as existing within time. Craig sees his temporal view of God as compatible with traditional Christianity insofar as
“the Bible leaves it open as to whether we are to understand divine eternity as timelessness (atemporality) or as everlasting duration (omnitemporality).” (110) But the Bible leaves it open only insofar as it includes passages on both sides of this issue. Craig cites Psalm 90.2 in favor of the view that God is everlasting. Then he cites Jude 25, which describes God as existing “before all time” (in a common translation). This phrase (if it makes sense) suggests timelessness. If these verses are taken literally, then the Bible is not neutral in the sense of being committed to neither view. Instead, the Bible seems
135
committed to both of these incompatible views of God’s relation to time. Moreover, Jude 25 seems committed to the very view that Craig takes to be refuted by “our” argument.
Still, Bible passages can always be reinterpreted, so I do not want to make too much of these verses. Nor does it matter much to me that the view of God as timeless has been and continues to be so popular. I mention these passages only to explain why I focused on that view of God in Chapter 4. What matters now is whether my argument can be extended to refute Craig’s own view of God as everlasting or omnitemporal.
In Craig’s view, as I understand it, God exists wholly, indivisibly, and equally at all times and, presumably, also at all spatial locations.
In addition, God is intrinsically changeless, although God can change extrinsically insofar as God’s relations to other things change when those other things change. To bring this conception of God down to earth, Craig draws an analogy to humans: “I, for example, exist equally, wholly, and indivisibly at every moment of my existence.”
(111) This analogy is misleading at best. Since my life is a series of events taking place over years, the whole of me does not exist at any one moment. Even at a single moment, I am divisible, not only because my body can be cut into parts, but also because my mind can divide its attention (so I can make plans about what to do after din-ner while also feeling a headache, and my plans are separate from my headache). For such reasons, humans cannot provide a good model for Craig’s God.
But let’s imagine (if we can) for the sake of argument that humans do “exist equally, wholly, and indivisibly at every moment of
. . . existence.” There is still another crucial difference between humans and God. Humans change internally as well as externally, whereas Craig’s God is not supposed to change internally at all.
When I argued that timeless beings cannot cause changes in the temporal world, my reason was that timeless beings cannot change, and only changes can cause other changes. This argument clearly applies to any being that does not change, even if that being exists in time. So it applies to God if God does not change, whether or not God is temporal.
Of course, Craig admits that God does change externally, and Craig explains this notion by adding, “He changes in His relations with tem-
God?
poral things.” (111) It is not clear what this means. It might seem that one example is that, as my son grows stronger, he gets closer to God’s power, and God then changes by getting closer to my son’s power.
But that can’t be the idea, because God is infinitely powerful, so my son cannot get closer to God’s power, no matter how powerful my son becomes. Let’s try again: If I drive south from Chicago to New Orleans, I move from north of Memphis to south of Memphis. So Memphis changes from south of me to north of me, even though Memphis never changes its location. But that also can’t be the idea, because God is not supposed to have any particular spatial location as opposed to any other. One more try: If I believe that the president of the United States is in Washington when he isn’t, and I still believe it when the president gets back to Washington, then my cognitive state can change from false belief to true belief, even though my mental state did not change in itself at all. But that can’t be the idea, because God is supposed not to have any false beliefs. He is supposed to know everything in advance. There are many other models of “external change.” I can’t go through them all here. Nonetheless, my point is that, even if the notion of external change makes sense in our world, it is far from clear that it makes any sense when applied to God. External changes seem to be excluded by God’s other traditional features.
More importantly, even if God does change externally in some way, this is not the kind of change that can cause effects in our world.
It is hardly enough to part the Red Sea or cause all of the other miracles that are attributed to God. External changes surely cannot explain how God could create the world, since there was nothing external to God before the world was created, so then there could not have been any external change in God that would have caused the world to exist. Thus, even if God did change externally, such changes would not be enough to save many traditional claims about God.2
Freedom is also a red herring. Craig describes what he takes to be my view of freedom. (111) His description is inaccurate. I could explain why, but that tangent would be irrelevant,3 because I could adopt any view of freedom that I want and still give my argument that an internally changeless God cannot cause changes. To bring up freedom is to skirt the issue.
Similarly for determinism. Craig says, “the reason Sinnott-Armstrong makes his claim is because he is a determinist.” (111)
137
However, determinism has nothing to do with it. That’s why I did not mention determinism when I gave my argument. I didn’t need that assumption. My argument here would not be affected in the least if quantum mechanics convinced me that some events occur without any deterministic cause. Indeed, indeterministic quantum mechanics helps my case by providing a model of how a Big Bang could arise out of quantum soup. I discussed that issue in Chapter 2. The point here is just that, even if determinism were false, that would not begin to show that Craig’s God could cause anything.
In order to explain how his God could cause changes, Craig admits that he must depend on a “wholly different” account of causation, which he and others call “agent causation.” Craig is right that
“Sinnott-Armstrong has said nothing to show that . . . agent causation does not exist.” (112) Not yet, but now I’ll say something: Craig never explains agent causation in his essay. This lack of explanation makes it hard to criticize his view, but Craig’s idea seems to be that the cause of an act is the agent as a whole rather than any particular change in the agent.
If that’s what agent causation is, why should we admit it? Craig gives only this reason for accepting agent causation: “I obviously cause certain events to occur when they do rather than earlier or later.” (111) “Obviously”! Well, this is not obvious at all once one thinks about it a bit more deeply. Sure, in common language we say things like, “Minnesota Fats sank the eight ball.” But that is just shorthand. We all know that it was not Minnesota Fats as a whole that sank the eight ball. (He didn’t literally put everything he had into the shot. His whole body wasn’t up on the table knocking the ball into the pocket.) That common knowledge makes it unnecessary to add obvious qualifications. However, when we are being more careful and explicit, we say that what sank the eight ball was Minnesota Fats’s last shot (that is, his action of moving his arms so as to move the cue stick and knock the cue ball in a certain direction). In this and all other cases, it is not the agent as a whole who causes changes. The cause is, instead, a particular act by that agent at a particular time and place. The only reason why we do not mention this act in common language is that it is so obvious. The reason why it is obvious is that Minnesota Fats was in the pool hall all evening, but he did not sink the eight ball until midnight. His mere existence
God?
or presence cannot cause the event at midnight if he was there all night. It has to be some particular change at or before midnight that causes the event at midnight. So, if agent causation is causation by the agent as a whole, it makes no sense.4
Maybe Craig assumes a different notion of agent causation.
Maybe this other notion makes sense. It is hard to tell until Craig explains what he means. Until then, agent causation remains mysterious and dubious at best. So the problem of action still gives another reason to believe that a traditional God does not exist.
The Unsolved Problem of Evil
My final argument, which Craig aptly labels “atheism’s killer argument” (112), raises the classic problem of evil. When an argument is a “killer,” opponents often simplify it in order to hide its force. So Craig reformulates my argument with two simple premises: “If God exists, gratuitous suffering does not exist. Gratuitous suffering exists.
Therefore, God does not exist.” (114) This common reformulation obscures several important points. First, my argument is not about suffering alone. Many forms of evil or harm do not involve suffering. Death and disability are evil or harmful to some extent, even when they are painless. Second, the term “gratuitous” is misleading, because the problem of evil arises even if evil is never totally gratuitous in the sense that there is no compensation at all for the evil.
Atheists need to claim only that there is no adequate compensation for some cases of evil or harm. In response, theists cannot simply point to some small compensation. They need to argue that the compensation is adequate, which means that it must be not only important enough but also fair. Third, it is not enough that an evil is “ultimately to our or others’ advantage,” (114) as Craig says. To be justified, the evil must be necessary for that good. Otherwise, God could bring about the good without the evil. Finally, the argument targets only a traditional conception of God as all-powerful. Because God is supposed to be all-powerful, theists cannot argue simply that an evil brings about good in the world as we know it. Theists need to deny that there is any logically (or maybe metaphysically) possible world in which the good occurs without the evil or something else at least as bad. In response, atheists need to claim only that it is logically possible to gain the good without the evil, so atheists can
139
refer to worlds that are very different, even in their basic laws of nature, from the world we live in. After all, an all-powerful God could have brought about those other possible worlds instead of ours.
For these reasons, I want to return to my original formulation: 1. If there were an all-powerful and all-good God, then there would not be any evil in the world unless that evil is logically necessary for an adequately compensating good.
2. There is lots of evil in the world.
3. Much of that evil is not logically necessary for any adequately compensating good.
4. Therefore, there is no God who is all-powerful and all-good.
Craig does not deny that this argument is logically valid. How could he? Nor does he deny premise (2). That one is obvious. He does raise some questions about premise (1) later (126).5 However, Craig’s main criticisms of my argument are directed at premise (3).
I supported premise (3) by citing numerous examples of harms and then ruling out a long list of goods that are often supposed to compensate for those evils. None of these arguments is absolutely conclusive. I already admitted that. But Craig uses (or abuses) my admission in his criticisms. So I need to start with a few words about the obscure, methodological issue of burden of proof.
Craig often suggests that my argument fails unless I prove premise (3) beyond any shadow of a doubt. Craig makes this assumption, for example, when he writes,
A butterfly fluttering on a branch in West Africa may set in motion forces that would eventually result in a hurricane over the Atlantic Ocean. Yet it is impossible in principle for anyone observing that butterfly palpitating on a branch to predict such an outcome. The brutal murder of an innocent man or a child’s dying of leukemia could send a ripple effect through history so that God’s morally sufficient reason for permitting it might not emerge until centuries later or perhaps in another country. (117)
Notice the words “may,” “could,” “might,” and “perhaps.” Similar guarding terms occur again and again throughout Craig’s Chapter 5.6
Such qualifications take the sting out of Craig’s response. Sure, a butterfly might cause a hurricane. I can’t exclude that possibility.
God?
Nonetheless, when I see a butterfly fluttering, I have no reason at all to believe that this butterfly will actually cause a hurricane. Similarly, it is possible that a child’s dying of leukemia would stop all of the fighting in the Middle East. Nonetheless, when I see a child dying of leukemia, I have no reason at all to believe that this child’s death will actually prevent any war.
What does this show about my premise (3)? In any example that I give to support premise (3), I must admit the possibility that the cited evil is necessary for some adequately compensating good. (It’s not clear how this necessity could be logical, but I will let that go for now.) So it is possible that this example does not really support premise (3). Still, all that I have to admit is a possibility. Despite this possibility, in each of my examples, we have no reason to believe that the cited evil is actually necessary for some adequately compensating good.
Now apply this point to the range of examples. There are many cases of many kinds where we have no reason, even after careful reflection, to believe that an evil is necessary for any adequately compensating good. It is possible that all of these evils are justified in ways that we cannot see, just as it is possible that every fluttering butterfly causes a hurricane. But to admit that this is possible is not to admit that it is true or even plausible. Surely we have plenty of reason to believe that much evil out of this vast set is not logically necessary for any adequately compensating good. That is all that premise (3) claims.
So we have plenty of reason to believe premise (3).
Is that enough for my argument? Not if each premise must be proven beyond a shadow of a doubt, as Craig seems to assume. However, that burden of proof is too heavy to be met on any important and controversial matter. We could not live our lives well if we never took any risk of being wrong. We need to rely on fallible reasons. So it is enough that we do have plenty of reasons to accept premise (3).
The same point applies to another of Craig’s complaints. He sometimes adds that we cannot “assess with confidence the probability that” (116) my premises are true. I admit that I cannot assign any definite probability between 0 and 1 to my premises. However, I also cannot assign a definite probability to many things I believe, such as that the butterfly fluttering in front of me now will not cause a hurricane. This limitation does not make my beliefs unreasonable.
141
If we had to assign definite probabilities to every belief, then we could not believe much, if anything, of importance in everyday life.
As before, Craig’s criticism depends on a very high standard that my belief does fail, but which need not be met in order for beliefs to be justified.
In other places, Craig suggests a different issue regarding burden of proof. He assumes that I need to show that Christians as such have reasons to accept my premises, so it is not enough to show that other people who are not Christians have such reasons. This assumption comes out, for example, when Craig replies, “the Christian is not committed by his world view to the truth of ” (114) premise (3). The suggestion seems to be that my argument begs the question by invoking premises to which Christians are not committed.
Well, that depends on my goal and my audience. If I were trying to persuade Christians who immediately reject anything that is incompatible with their prior religious beliefs, then my task would be hopeless. So I am not trying to persuade such dogmatic Christians.
This is not to admit that my arguments have no force against them.
My arguments show that all traditional theists have adequate reasons to renounce their belief in God, so they should renounce that belief, even if they are not persuaded by those reasons, because they do not listen to my arguments. Religious beliefs lead many (not all!) people to refuse to listen to reasons, but that does not mean that those reasons lack force as reasons. All it means is that those people cannot be persuaded. But, as I said, my goal is not to persuade them. Nor is my argument aimed at atheists who immediately reject anything that is religious without caring about whether they have adequate reasons to do so. They don’t need any argument from me or anyone else. Instead, my intended audience includes only those whose minds are open, that is, who are willing to consider evidence against the existence of God as well as for the existence of God and to form their beliefs on the basis of that evidence.
Most Christians whom I know fall into this category. However, Craig seems to assume that a Christian will reject any premise to which he is “not committed by his world view.” What is this “world view?” If their world view as Christians includes only the essential doctrines of Christianity and the logical implications of those doctrines, then Christians are not committed to my premise (3) on that
God?
basis. However, Christians are not just Christians. They are also humans with common sense, so they hold many beliefs, as well as standards of adequate compensation, that are not essential doctrines of Christianity. These other views do commit most (even if not all) Christians to my premises.
Craig mentions one of these commonsense beliefs: “Everybody admits that the world is filled with apparently gratuitous suffering.”
(114) Craig emphasizes, “that doesn’t imply that these apparently gratuitous evils really are gratuitous.” (114) This statement is accurate if all it denies is a formal logical implication. Nonetheless, appearances still do affect the burden of proof. If a stick looks bent, then it is reasonable to think that it is bent, as long as there is no reason that overrides or undermines the visual appearance. If the stick also feels straight, or if a magician is waving it, then I should not trust my eyes. But if I feel it, look at it closely from all angles in various circumstances, and find no reason to doubt its bent appearance, then I have reason to believe that it is bent. Its bent appearance still doesn’t “imply” that it is really bent, but no mere experience could logically imply that. We need to draw conclusions from how things appear to us, at least after close inspection.
Sticks are concrete (in one sense!), but the same method applies to less tangible topics. Would it benefit my family in the long run if I were to burn down our house tonight? Possibly. Does that possibility make it reasonable to believe that we would benefit? No. Why not? Because we do know one thing: Burning down my house would cause serious problems for my family in the short term. Those known costs set up a presumption or burden of proof that needs to be overcome before contrary beliefs can be reasonable. Given those known costs, if I have no reason to believe that burning down my house will bring adequate benefits, then it is not reasonable for me to believe that burning down my house will benefit my family overall in the long term. Moreover, it is reasonable for me to believe that burning down our house will not benefit us overall in the long term. Don’t you agree? If so, we both seem committed to some standard like this: When something causes serious short-term costs, and we have no reason to believe that it is necessary for any adequate compensation in either the short term or the long term, then it is not reasonable to believe that it is good overall in the long term, and it is
143
reasonable to believe that it is not good overall in the long term.
Most theists are committed to that commonsense standard, even though it does not follow from any essential Christian doctrine. They have reason to accept it, because without it they would be lost in their everyday lives. They might end up burning down their houses!
Craig seems to think that this standard is undermined by my claim that “we have no more reason to believe that the evil now will lead to great bliss in the long term than that this evil will lead to great misery.” (118) However, my point was that, without any reason to believe that either long-term effect is more likely than the other, possible long-term benefits are canceled by possible long-term costs.
But short-term consequences are another matter. We can still know a lot about the evil now and its short-term costs, at least in the cases that I was discussing. What is known about these evils then sets up a presumption that is not rebutted by any other reasons. That is why my admission of ignorance about long-term effects supports my argument rather than undermines it in the way that Craig suggests.
This commonsense standard leads right into my argument. We know that evil occurs, as premise (2) says. The existence of that evil sets up a presumption. This presumption is never overcome, because we have no reason to believe that this evil is logically necessary for any adequate compensation in either the short term or the long term.
I supported this premise by ruling out the many proposed justifications of evil. The presumption plus the absence of any rebuttal makes it reasonable to believe that the evil is not good overall. That, in turn, makes it reasonable to believe that there is no all-good and all-powerful God.
So far I have focused on standards for when beliefs are reasonable, but my arguments against proposed justifications of evil also employ a different kind of standard, namely, standards for when compensation is adequate. Here again I claim that my standards are commonsense principles that most Christians accept, along with almost everyone else.
For example, Craig suggests that God’s reason for allowing babies to suffer and die might be a combination of heaven and virtue: The babies go to heaven, and other people develop moral virtue. In Chapter 4, I pointed out that an all-powerful God could send the babies straight to heaven without the evil, but Craig responds, “that
God?
forgets the Virtuous Response: so doing would short-circuit God’s work in the lives of the parents or other people.” (116) I didn’t forget that. Rather, I assumed that gains to other people would not be adequate compensation for harm to the baby, for reasons that I gave in my very next section on the Virtuous Response. One reason is that an all-powerful God would have at His disposal less harmful ways to teach lessons. Another reason is that it would be unfair to let the baby suffer just to help separate individuals. Here I was appealing to a commonsense standard of adequate compensation, namely, that it is morally wrong to use some people merely as means to benefit other people, at least when one could benefit those other people in a less harmful way. This standard would be accepted by conse-quentialists, by Kantians, and also by defenders of the natural law doctrine of double effect (since, in the relevant case, God would intend harm as a means, thereby violating this doctrine). This standard is not completely clear, but surely one example is letting a baby die painfully merely to teach a moral lesson to other people, when one has less harmful ways to teach those other people. Any human who did this would be considered a moral monster, so the same compensation also should not count as adequate for God.
Craig complains about my analogies, so I should explain why I use them. I admit that analogies are dangerous. One reason is that opponents can always find some disanalogy. No analogy is perfect.
But, also, not every difference matters. The analogy still works if the differences that do exist do not matter. So analogies can be useful when theists would simply deny any direct claim about God. If theists admit the analogous case, their admission reveals their commitment to a general principle, and then the burden is on them to explain why that general principle does not apply to God. That is the strategy behind my analogies. I am trying to show that common sense commits you to the premises of my argument. If you agree with me about my everyday examples, that admission reveals your own standards for reasonable belief or adequate compensation. My argument simply applies those same standards to belief in God.
Craig suggests that Christians might deny that such common standards apply to God: “I am bound by certain moral obligations and prohibitions vis à vis my neighbor (e.g., not to take his life); but God (if He has moral duties at all) is not bound by many of these (e.g.
145
He can give and take life as He pleases).” (115) Of course, if God is all-powerful, then God is able (in the sense of having the power) to “take life as He pleases.” However, that ability shows nothing about “moral obligations and prohibitions.” So I take Craig’s final parenthetical claim to be that there is nothing morally wrong with God killing anyone at any time for any reason, no matter how trivial the reason is. In this view, people have no more rights than mos-quitoes with respect to God. Why not? Maybe because God is so powerful. But might does not make right. Maybe because God created us. But parents are prohibited from killing the children whom they create and nurture. Why? Because the children are separate people with rights. This reason applies to God as well. If God gave us free will, then He made us separate people, so why wouldn’t we have rights not to be killed by Him needlessly? I see no good reason to exempt God from moral standards, so I assume that He would not be justified in letting babies suffer and die merely as a means to benefit other people when he has better alternatives.
If Craig continues to insist that God is morally permitted to “take life as He pleases,” then his obstinacy will strike many people as desperate. Most Christians will hope for a better response. After all, God is supposed to be all-good. Craig’s response admits that God violates the only moral standards that we know. If God’s goodness is compatible with letting babies die painfully merely as a means to help other people when God could use less harmful means to help those other people, then this is a kind of goodness that we do not understand or have any reason to care about. Why should we love or worship a God who does such things? It might sound neat to say that God is not subject to our standards, but this ploy leaves it un-clear what it is that makes God good. In the end, we need to use our own standards, because we cannot understand any others.
Luckily, Craig gives other responses. Sometimes he argues that all evil can be justified in terms of common moral standards. What a theist needs is for each bit of evil to be justified by something.
Theists do not need a single justification to cover all bits of evil. They can use different justifications for different bits of evil, as long as every bit ends up being justified. Craig accuses me of forgetting this logical point by adopting a “divide and conquer” strategy. I plead innocent. I did not “typically refute . . . a solution by arguing that it
God?
cannot cover all cases of suffering,” (115) as Craig charges. Instead, I gave several examples, such as birth defects, and argued that no proposed compensation, either alone or in combination with others, is adequate to justify the evil in those specific cases. Even if “the solutions are complementary and mutually reinforcing” (116) when they apply, that does not show that they do apply to the cases in my argument. Besides, I argued (four paragraphs back) that the solutions do not mutually reinforce each other in the way that Craig claims, at least in the examples that I gave. So, far from “divide and conquer,” my strategy was to focus my forces in a few examples that undermine all proposed justifications together.
Later in Chapter 5, Craig does propose one justification that might seem to solve the problem of evil, if it worked. He claims that “the purpose of life is . . . knowing God,” (120) and “it is precisely in countries that have endured severe hardship that evangelical Christianity is growing at its greatest rates.” (121) This is supposed to show that severe hardship serves the purpose of life.
The first problem here is that Craig’s factual claim is dubious at best. He cites Johnstone on China as one example. (121) Reportedly, harsh persecution of Christians during Mao’s Cultural Revolution was followed by unparalleled growth in the number of Christians.
This report should come as no surprise, because it must be based on what people communicated about their beliefs. The number of people who are willing to publicly proclaim their Christianity is likely to go down while Christians are persecuted harshly, even if the number of believers does not go down. Anyway, let’s grant for now that the number of Christians did go way down under Mao, because of fear and inability to obtain information about Christianity. That low point would explain why many people in China did become Christians when information became available and fear was removed. But the number of Christians is still probably less than what it would have been if there had been no persecution in the first place. That comparison is the relevant measure of whether the evil of persecution led to more knowledge of God. So neither Craig nor Johnstone is justified in calling Mao “the greatest evangelist in history.” (121) Most importantly, even if Mao’s policies did increase the number of Christians, Craig himself says, “It is estimated that 20 million Chinese lost their lives during Mao’s Cultural Revolution.” (121) To glo-
147
rify such a tragedy dishonors the memory of the victims. Surely an omnipotent God could find a less costly way to produce Christians in China.
Craig also mentions El Salvador and Ethiopia. I will not discuss those cases here because similar points apply. Even if the number of Christians does go up after tragedies, there is no reason to think that an all-powerful God could not find a less harmful way to help people come to know and love Him.
Besides, if every such tragedy really were necessary for a higher purpose, then we ought to welcome tragedies and not fight them.
We should let Mao persecute the Chinese, and let earthquakes de-stroy lives in El Salvador, and let famines ravage Ethiopia, and so on. If all such tragedies were God’s plan, to interfere in any of them would be to interfere in God’s plan and to prevent the greater good of more people becoming Christian. Luckily, not all Christians hold such beliefs. Many Christians work hard against evils in this world.
But Craig suggests a different view of tragedies when he says that so many deaths and so much suffering are a “slight, momentary affliction” that “only serves to heighten mankind’s moral responsibility before God.” (123) Such a view takes away our incentive to help each other in this world. Beliefs with this result should be avoided, at least in the absence of any reason to accept them. And nothing that Craig says gives us any reason to accept such a view or to believe that tragedies like those in China, El Salvador, and Ethiopia really do cause growth in evangelical Christianity or serve any worthwhile purpose.
Although Craig fails to specify why God allows evil, he has one move left. He claims that his arguments for the existence of God show that God must have some reason to allow evil, even if we cannot figure out what it is. The conclusion that God has some unknown reason is supposed to follow from Craig’s arguments for the existence of God.
Unfortunately, this move is no better than his arguments for the existence of God. I already argued at length in Chapter 2 that Craig’s arguments for God fail miserably. I have seen no reason to take any of that back. Later, in Chapter 5, Craig merely repeats his claims that “my moral argument, if cogent, certainly does prove that God is all-good; and the cosmological argument certainly shows that the
God?
Creator of the universe out of nothing has sufficient power to prevent the evils that afflict our world.” (125) Again, Craig merely assumes without any argument that the “God” who is all-good is the same person as the creator. That assumption begs the question here, since the existence of evil is compatible with one being who is all-good and another separate being who is all-powerful. Besides, even if the moral argument were cogent evidence for some personal source of objective value (which it is not), it still would not “certainly
. . . prove” that this source is all-good, since this source might fail its own objective standards of goodness. Regarding his claims for his cosmological argument, notice that Craig’s use of the assuring term
“certainly” is again misleading, since it is not at all certain that something powerful enough to create the whole universe around fifteen billion years ago was ever all-powerful, is still alive today, or is able to prevent every specific occurrence of evil. Many things that are powerful on a large scale (such as a supernova or the Big Bang) do not last forever and are as clumsy as a bull in a china shop when it comes to small details. Similarly, it might not be within a creator’s abilities or job description to fix genetic defects in suffering children.
Of course, if a traditional God did exist, then, by definition, He would have “sufficient power to prevent the evils that afflict our world.” But that would hardly show that God does exist. Indeed, God’s supposedly unlimited power is exactly what creates the problem of evil and, thereby, demonstrates that God does not exist.
Craig’s arguments for the existence of God need to be much better than these in order to give us adequate reason to believe that all evil is justified in ways that we cannot see.
Craig does formulate one new argument when he claims that “evil
. . . requires God’s existence” because “If God does not exist, objective moral values do not exist.” (125–126) This premise is, however, a strange way to begin an argument at this point in our book, because I already showed in Chapter 2 why that premise is dubious at best. Rape is morally wrong because it harms the victim, even if there is no God; so objective moral values can and do exist without God. Moreover, atheists can consistently and justifiably believe in moral values as much as theists can. Just consider all of the non-religious moral theories that philosophers develop.
Where does all of this leave us? Craig is not convinced by my ar-
149
guments. Nor am I convinced by his. That impasse should not surprise anyone. The more important question is whether our arguments provide reasons for open-minded inquirers who are not previously committed either for or against the existence of God.
Craig’s arguments constantly depend on a prior belief in God.
How else could he hold that a child’s dying of leukemia and Mao’s Cultural Revolution were good? How else could he start arguments with phrases like “Once we reflect on God’s providence . . . ”? (119) How else could he think that the problem of evil could be solved by Christian doctrines like “The purpose of life is . . . knowing God”
(120) and “God’s purpose spills over into eternal life”? (123) None of these claims has any force for anyone who is not already committed to the existence of God.
In contrast, I tried to base my arguments on commonsense standards of reasonable belief and adequate compensation. These principles are not peculiar to atheists. Most Christians also use the same standards in their everyday lives. More importantly, these principles are accepted by almost everyone who is not committed in advance either for or against the existence of God. That makes them neutral starting points.
Of course, when you put these principles together with obvious facts, they yield the conclusion that God does not exist. Otherwise my argument could not work. But that is no problem for my argument. Every argument has to start from premises that lead to its conclusion. The strategy in a good argument is to start from premises that are independent of the conclusion in the sense that the premises would and should be accepted by people who are not already committed one way or the other about the conclusion. My premises pass this test. Craig’s do not. That is why I conclude that my arguments provide better reasons to believe that God does not exist.
Notes
1. Craig’s guarding term “just” is misleading, because neither I nor anyone else believes that all God cares about is “getting you to believe that He exists.”
Any all-good God would care about many things other than belief in Him. But that does not mean that belief in God is not one of the things that God would care about, at least instrumentally as a means to other goods.
God?
2. For more arguments that external (or Cambridge) changes cannot cause effects, see Sections I–II of Sydney Shoemaker, “Causality and Properties,” in his Identity, Cause, and Mind (Cambridge: Cambridge University Press, 1984), 206–209.
3. At several points Craig says that I adopt or assume a view that I do not hold and that my argument does not depend on. When the ascribed view is implausible or absurd, this common rhetorical trick might lead some readers to distrust anyone who would hold such a view. That danger makes it tempting to respond to each charge, but then I would not have space for more important issues. For that reason, I will restrain myself and not bother to respond to each accusation. My silence should not be misconstrued as an admission of Craig’s claims.
4. My brief argument against agent causation derives from C. D. Broad,
“Determinism, Indeterminism, and Libertarianism,” in his Ethics and the History of Philosophy (London: Routledge, 1952), 195–217. For more arguments, see Alvin Goldman, A Theory of Human Action (Princeton, N.J.: Princeton University Press, 1970), 80–85; and Hilary Bok, Freedom and Responsibility (Princeton, N.J.: Princeton University Press, 1988), 44 ff.
5. There, Craig says that premise (1) “is not obviously true” and has been questioned by “[s]ome theists.” (126) Since he uses the third person, he does not seem to share these theists’ views. In any case, Craig says how I would respond to those theists. His only criticism of my response brings the issue back to premise (3). That is why I focus only on premise (3) in my text.
6. Here are a few examples: “the envisioned revelation might lead to self-stultifying situations.” (119) Maybe, but couldn’t an all-powerful God avoid this problem? “Many evils occur in life that are utterly gratuitous with respect to the goal of producing human happiness; but they may not be gratuitous with respect to producing a deeper knowledge of God.” (120) Or they may be gratuitous in that respect as well. The question is not what “might” or “may” be, but what we have adequate reason to believe
A
Big Bang, 4–5, 9, 32, 43–45,
Ackerman, Susan, 51 n. 2, 56
49, 60–61, 64, 132, 148
action, problem of, 61, 98–101,
Big Crunch, 61
110–112, 134–138
bloated conclusions, fallacy of,
Acts, 110
32, 50, 53–54, 131
actual infinite. See infinite,
Bohm, David, 6
actual vs. potential
Boltzmann, Ludwig, 14–15, 66
Adam, 90
Brandenburger, Robert, 13
Adams, Robert M., 68, 77 n. 22
Buddhists, 82
advertisements for God, 109,
burden of proof, 139–145
133
agnosticism, 81, 107–108, 129
C
Alston, William, 27, 68, 77 n. 22,
Carr, B. J., 64
116, 119
causation, 5, 8, 45, 100–101
analogies, 12, 16, 38, 39, 44,
agent causation, 111–112,
46, 47, 48, 89, 90, 92, 96,
137–138, 150 n. 4
98, 100, 104, 114–115,
whatever begins to exist has
117, 124–125, 135, 144
a cause, 8–9, 56
argument from ignorance. See
chance, 10–11, 49, 64–67
ignorance, argument from
change, external or extrinsic,
atheism, 3, 81, 107–108, 129
111, 135–136, 150 n. 2
atheistic moral realism, 19–20,
chaos theory, 117, 139
34, 67
Chaotic Inflationary Universe
theory, 8, 13, 117
B
Collins, Robin, 63–66
Barrow, John, 15, 60
common sense, 142–144, 149
beg(ging) the question, 7,
Copenhagen Interpretation, 6,
48–49, 58, 96, 141, 148
56–57
Ben and Jerry, 38
Corinthians, 123
151
Index
cosmological argument, 41, 56,
apparently gratuitous
125, 147
suffering, 114, 116–120,
creator of universe, 31, 50–51,
124–126, 138, 142
53–54, 125, 131, 148
contrast response, 89
Crusades, 82
emotional problem, 112
Cyclical Model, 76 n. 12
free will response, 92
glorious response, 92, 115
D
heavenly response, 91, 116,
Davies, P. C. W., 9, 10
123, 143–144
debating God’s existence, ix-xi
intellectual problem, 112,
defeaters, 59
127
Deltete, Robert, 6
internal vs. external version,
Dembski, William, 65–66
113–114, 127
design(er), 10, 13–14, 16–17,
logical vs. evidential version,
31, 47, 50–51, 53, 65–66,
86, 113–114, 124
125
modest response, 95–97,
determinism, 68, 111, 136–137
117–119
divide and conquer strategy,
overriding response, 97,
115–116, 146
124
divine commands. See morality,
permitting evil, 116–119,
divine command theories
123–124, 126, 145
of
privation response, 89
Down’s syndrome, 84, 90–91
punishment response, 90
standard response, 94–95
E
unconscious response, 89
Eddington, Sir Arthur, 5
virtuous response, 91,
equivocation, fallacy of, 33
115–116, 143–144
Euthyphro dilemma, 35–36,
evolution, 15, 17–20, 33, 49
68–69
excessive footnotes, fallacy of,
Eve, 90
32, 55–56
evidence, 36, 39, 48, 83, 85,
experience, religious, 25–28,
94, 96, 103–106, 127 n. 1,
31, 38–40, 51, 53–54,
130–131, 140–142
73–74, 103, 125, 131
evil, 86, 138
expert testimony, appeals to,
justified, 87–88
55–56
natural, 93, 96–97, 117, 121
explanation, 16, 45, 50, 66,
evil, problem of, 83–98, 101,
100
112–127, 138–149
simplicity vs. familiarity, 66
153
F
eternal and timeless, 5, 8,
fallacies. See begging the
31, 46, 83, 98–101,
question; bloated
110–111, 134–138
conclusions; equivocation;
gender, 51n. 1
excessive footnotes; false
immaterial, 5, 54, 98
dichotomy; genetic;
personal, 5–6, 31, 45–46, 54,
ignorance, argument
61, 83, 111
from; inverse gambler’s;
temporal, 61, 99–100,
straw man
110–111, 134–138
false dichotomy, fallacy of, 32,
transcendent, 8
54–55
uncaused, 5, 54
familiarity, 66
unity, 131, 148
fine-tuning argument, 9–17,
Gödel, Kurt, 42
46–50, 52 n. 13, 62–67
Grand Unified Theories,
France, R. T., 55
76 n. 10
free will (and freedom), 34, 68,
Grass, Hans, 71
90, 92–93, 111, 117,
Great Pumpkin, 102
136–137, 145
H
G
Hawking, Stephen, 8, 9, 32,
Genesis, 110
43–44, 55, 75 n. 9
genetic fallacy, 20
Healey, John, 18, 32
Gert, Bernard, 86
Hengel, Martin, 71
ghosts, 103
Hick, John, 26, 32
God
Hilbert, David, 4, 32, 42,
all-good, 31, 83–86, 97–98,
75 n. 8
105, 112–114, 125, 134,
Hindus, 74, 82, 124
138–139, 145, 147–148
holocaust, 17
all-knowing (omniscient), 31,
Howard-Snyder, Daniel, 124
83, 117, 133–134
Hoyle, Fred, 4, 32
all-powerful (omnipotent),
Hume, David, 5, 32, 73
31, 51, 83–85, 97–98, 105,
112–114, 117, 119, 125,
I
133–134, 138–139,
ignorance, argument from, 96,
147–148
101–105, 107–110,
changeless, 5, 54, 99, 135–136
129–134
effective (active in time), 31,
as sometimes fallacious, 104,
83, 98, 101
130
Index
infinite, actual vs. potential, 4, M
7, 41–43, 57–59, 75 n. 8
Many Worlds Hypothesis,
Inquisition, 21, 80
12–15, 49–50, 62, 63,
inverse gambler’s fallacy, 14
65–67
Isham, Christopher, 60
Mao, 121, 146–147, 149
Islam, 83
Mark, 22–23, 37
Mathematics. See also infinite
J
calculus, 41, 58
James, 28, 109
constructivism in, 51n. 5, 58
Jesus, burial and resurrection
numbers, 57–58, 98–99
of, 21–25, 32, 36–38, 51,
platonism in, 57–58
55, 69–73, 125, 131
set theory, 6, 57–58, 75 n. 7
Jews, 23–24, 71, 82–83
transfinite arithmetic, 7,
John, 23, 72, 109
58
Johnson, Luke, 24
Matthew, 23, 72
Johnstone, Patrick, 121, 146
McCullagh, C. B., 24
Joseph of Arimathea, 22,
Messianic expectations, 24, 37,
24–25, 72
71
Jude, 110, 134–135
mind, 16, 135
miracle, 36–38, 69, 73, 78 n. 30,
K
84, 88, 103
Kanitscheider, Bernulf, 57
Mithras, 37, 56, 71
Kasner, Edward, 57–58
moral agency, 68
Kenny, Anthony, 4
moral argument, 125, 147
kingdom of God, 120–123
moral epistemology vs. moral
Kremer, Jacob, 23, 55
ontology, 69
morality, 17–21, 32–36, 67–69,
L
87–88, 117, 121, 131, 145,
law (of nature), 10, 36, 44,
148
63–64, 67, 73, 100
divine command theories of,
Lemaitre, 43
35–36, 50, 68–69
Leslie, John, 10
evolution and, 17, 20
life, intelligent, 9, 46–50, 62
objectivity of, 17, 21, 69
Linde, Andrei, 13–14
Multiple Cosmoi Hypothesis.
Linus, 102
See Many Worlds
Lloyd-Jones, Martyn, 120
Hypothesis
Lüdemann, Gerd, 23–24
Murray O’Hare, Madelyn,
Luke, 72, 109
107
155
N
Q
natural disasters, 85, 93
quantum epoch, 43–45
natural law. See law of nature
Quantum Gravity Universe
Newman, James, 57–58
theory, 8
Nielsen, Kai, 5, 28 n. 5
Quinn, Philip, 68, 77 n. 22
Nietzsche, Friedrich, 17–18,
quintessence, 63
32–33
nothing, cannot get something
R
from, 6, 9, 44, 57
Rees, M. J., 64
religion, 82
O
religious schools, history of, 71
Ockham’s Razor, 13
Robinson, John A. T., 22
origin of universe. See
Romans, 109
universe, origin of
Ruse, Michael, 17, 18, 32–33,
Oscillating Universe theory, 8, 61
55
Russell, Bertrand, 17, 18,
P
32–33
Padgett, Alan, 110
Paul, 22–23, 123
S
pedophile, 93–94
Salmon, Wesley, 55
Penrose, Roger, 9
Schultz, Charles, 102
Plantinga, Alvin, 28, 113
Shepard, Matthew, 82
Plato, 69
shotgun strategy, 31, 53
Polkinghorne, John, 13
simplicity, 49, 56, 66
possibility, logical, 36, 94,
singularity, initial cosmological,
139–140, 150 n. 6
8, 60, 75 n. 9
probability, 47, 62, 140–141
Smith, Mark S., 71
likelihood principle, 48, 65
space, 4–5, 43, 98–101
specified, 11–12, 141
stem cell research, 82
problem of action. See action,
straw man, fallacy of attacking
problem of
a, 34, 108
problem of evil. See evil,
strong evidence principle, 130
problem of
supersymmetry, 75–76 n. 10
properly basic beliefs, 26–28,
40, 125, 127 n. 1
T
Proverbs, 110
Taoism, 27
Psalms, 110, 134
Taylor, Richard, 19–20
purpose of life, 120, 146, 149
teleological argument, 41
Index
theism, 3
origin of, 3–9, 41–46, 56–61
time, 98–101
sub-atomic events and origin
classical, 43–44
of, 6, 44–45
metaphysical, 8, 46, 61
thermodynamic equilibrium,
physical, 8, 43
14
Timothy, 110
utilitarianism, 86, 88
Tipler, Frank, 15, 60
Titus, 110
V
tracker fields, 49, 52 n. 13, 63
Vacuum Fluctuation Universe
theory, 8, 52 n. 8, 57, 60
U
universe. See also Big Bang;
W
fine-tuning
World Ensemble. See Many
astronomy and astrophysical
Worlds Hypothesis
discoveries, 4, 59, 61
Wright, N. T., 24
cosmic acceleration, 61, 63
finite or infinite, 3–4, 6,
Z
41–42, 52 n. 8, 57–58
zero, 41–42, 58