6. Conclusions
This stimulates inquiry that increases our knowledge while also raising new questions that themselves demand further answers. This useful process is undercut when observations are explained by postulating God. In this way, religious beliefs get in the way of science and the progress of knowledge.
These objections undermine Craig’s argument, even if the design hypothesis would explain fine-tuning. But it wouldn’t. No matter what happens, one can always postulate someone who designed things that way. If the universe expands, God designed it to expand.
If the universe contracts, God designed it to contract. Since God could design it either way, the hypothesis of God cannot explain why anything happens one way rather than another. Moreover, the design hypothesis works only if God can cause changes to tune the universe for life. I argued in section 4.2 that an eternal God cannot cause such changes in our temporal world, and I will develop that point in Chapter 4. The design hypothesis, thus, becomes incoherent if the designer is supposed to be an eternal God.
6. Conclusions
Craig has not given us any adequate reason to believe either in a divine source of morality or in the resurrection of Jesus or in a su-perhuman cause of religious experience or in a creator or designer of the universe. Moreover, Craig’s conclusions are bloated. Even if Craig’s arguments did establish some conclusions, they would not show the existence of God with all of His traditional features. In particular, even if some commander did dictate morality, that commander still might not be good or have the power to punish dis-
There Is No Good Reason to Believe in God 51
obedience by humans. Even if Jesus did rise from the dead, maybe he was abducted by aliens or maybe raising Jesus tired out God, so God lost His power. Even if religious experiences were some evidence for some external source, the most vivid religious experience could result from a God who is only very strong and pretty good.
Even if some creator or designer could be proven, this creator or designer might have died a long time ago. Indeed, if God designed this universe, there is much reason to doubt that He is all-good, as we will see in Chapter 4. For such reasons, nothing like the traditional God would follow even if Craig’s arguments did work part way.
Craig might respond that all of his arguments work together.
However, the combined set still would not show all of the traditional features of God, since none establishes that God is all-powerful. If there were a creator, that creator would have to be very powerful, but need not be able to control all small-scale events for all time.
Moreover, even if all of God’s features were covered, it would still not be clear that Craig’s different arguments are about the same being, since a creator might be separate from the commander of morality and also from the source of religious experience. To assume identity is just another way to bloat conclusions.
There always might be better arguments for the existence of God.
Theologians are inventive. However, until someone gives a better argument, we have no good reason to believe that a traditional God exists.
Notes
1. I refer to God as “He” because Craig and other traditional theologians use this masculine pronoun, although, as many have pointed out, it is not clear how God could have any gender.
2. Thanks to Susan Ackerman for this information.
3. David Hilbert, “Über das Unendlische,” Mathematische Annalen 95
(Berlin, 1926): 161–90. Craig cites a reprint of a translation.
4. Kurt Gödel, “Über formal unendscheidbare Sätze der Principia Mathematica und verwandter Systeme I,” Monatshefte für Mathematik und Physik 38 (1931). Thanks to Sam Levey for help on this paragraph and elsewhere.
5. Mathematical constructivists might deny this, but Craig is no constructivist, and it is hard to imagine any good reason to be a constructivist about numbers if you believe in God, since constructivism is motivated by skepticism about entities like gods.
6. For more detailed criticisms of Craig’s mathematical arguments, see
God?
Quentin Smith, “Infinity and the Past” in William Lane Craig and Quentin Smith, Theism, Atheism, and Big Bang Cosmology (Oxford: Clarendon Press, 1993).
7. Quoted in Marcelo Gleiser, The Dancing Universe: From Creation Myths to the Big Bang (New York: Penguin, 1997), 287. Thanks to Marcelo Gleiser for help at several points in sections 4.2 and 5.
8. Craig does criticize this theory: “Vacuum Fluctuation Universe theories
. . . cannot explain why, if the vacuum was eternal, we do not observe an infinitely old universe.” (8) However, the universe that we observe is (in a way) infinitely old in this view, even if its classical phase (which is the phase that we observe) is not infinitely old. So it is not clear what Craig’s objection is.
9. S. W. Hawking, “Breakdown of Predictability in Gravitational Collapse,”
Physical Review D14 (1976): 2460 (my emphasis).
10. On the reality of a singularity, see Quentin Smith in Craig and Smith, Theism, Atheism, and Big Bang Cosmology, p. 208.
11. For more detailed criticisms of Craig’s scientific arguments, see Quentin Smith, “Atheism, Theism, and Big Bang Cosmology” and “A Defense of the Cosmological Argument for God’s Non-existence” in Craig and Smith, Theism, Atheism, and Big Bang Cosmology (chaps. VII and IX).
12. On the technical problems with calculating these probabilities, see Timothy McGrew, Lydia McGrew, and Eric Vestrup, “Probabilities and the Fine-Tuning Argument: A Sceptical View,” Mind 110 (2001): 1027–1037.
13. Jeremiah Ostriker and Paul J. Steinhardt, “The Quintessential Universe,” Scientific American 284 (2001): 51–52. Tracker fields are just one example of a dynamical system in which physical observables, responding in part to feedback with the environment, evolve inevitably toward a certain final state, which is insensitive to the initial conditions. Tracker fields make fine-tuning understandable in the same way as do many other such dynamical systems throughout nature. Thanks are due to Rob Caldwell here and at other points in this chapter.
14. Ibid.; on multiple cosmoi, see Alan Guth, The Inflationary Universe (Reading, Mass.: Addison-Wesley, 1997), 245–252.
Reason Enough
William Lane Craig
Introductory Remarks
In my opening essay, I presented five reasons why I think that theism is more plausibly true than atheism. I’m grateful for Walter Sinnott-Armstrong’s stimulating and engaging interaction with those arguments. Far from being a “shotgun” approach, as he alleges, my five reasons to believe that God exists constitute a progressive, sys-tematic case for Christian theism. With each argument, we learn more about the ultimate ground of our existence: that there is a Personal Creator and Intelligent Designer of the universe, Who is the locus of moral value, Who has revealed Himself in Jesus Christ, and Who can be known in personal experience. Mine is a cumulative case for the existence of God, both in the sense that the arguments reinforce one another and in the sense that they supplement one another in order to arrive at the traditional concept of God. In assessing Sinnott-Armstrong’s criticisms, I’m therefore going to stick by the order in which I initially presented the arguments rather than follow his re-ordering.
But before I look at his criticisms in detail, I want to say a word about each of the three fallacies I allegedly commit, which Sinnott-Armstrong playfully labels the fallacies of “bloated conclusions,”
“false dichotomy,” and “excessive footnotes.”
First, do I commit the fallacy of bloated conclusions? When Sinnott-Armstrong alleges this, he isn’t referring to the specific conclusions of my individual five arguments themselves. Those conclu-53
God?
sions do follow from the premises. Rather he is claiming that even if all my arguments are sound, they do not suffice to justify the conclusion that God exists. Perhaps a Creator or a Designer of the universe exists, but such a being does not deserve to be called
“God.” I guess that depends on what you mean by “God.” Sinnott-Armstrong seems to forget that my case is a cumulative one: taken together, the arguments, if successful, do enable us to recover a striking number of the traditional divine attributes. The first argument gives us a Personal Creator of the universe who is uncaused, eternal, changeless (at least sans the universe), immaterial, and enormously powerful; the second gives us His inestimable intelligence and personal concern for intelligent creatures; the third, His absolute goodness and love as well as His metaphysical necessity; the fourth, His identity with the God of Israel revealed by Jesus of Nazareth; and the fifth, His knowability. Together the arguments do, if successful, give us a full-orbed conception of God. They give us the God of the Bible.1 It would be a strange form of atheism, indeed, which conceded that the God of the Bible exists! Clearly, if there is such a being, then atheism, in any reasonable sense of the term, is false.
Second, what about the fallacy of drawing false dichotomies?
Again, no such fallacy is committed in my first three arguments, since they are logically valid, deductive arguments, whose conclusions follow necessarily from their premises. Since these arguments are logically valid, the only question is whether their premises are more plausibly true than their negations. Of the three, only the argument for design employs disjunctive reasoning, and its first premise states, not a dichotomy, but a trichotomy, a trichotomy that Sinnott-Armstrong does not challenge, as we’ll see below. He has simply misstated my cosmological and moral arguments. As for my fourth, inductive argument concerning the historical Jesus, the form of this argument is inference to the best explanation. My contention is that given the historical facts of Jesus’ honorable burial, the discovery of his empty tomb, the post-mortem appearances of Jesus, and the origin of the disciples’ belief in his resurrection, the best explanation of these facts, when judged by such criteria as explanatory power, explanatory scope, plausibility, accord with accepted beliefs, absence of ad hoc hypotheses, and so on, is the hypothesis “God raised
55
Jesus from the dead.” I by no means exclude other explanations from the pool of live options; I merely maintain that the Resurrection Hypothesis is the best. And, of course, one may dispute the historical facts themselves, and I’m prepared to discuss the various lines of evidence supporting each one. Thus, the charge that I commit the fallacy of drawing false dichotomies is very odd and must arise from inattentiveness to my formulation of the various arguments.
Finally, the charge of excessive footnotes! This is really the charge of a fallacious appeal to authority. But as Wesley Salmon puts it,
“there are correct uses of authority and well as incorrect ones. It would be a sophomoric mistake to suppose that every appeal to authority is illegitimate, for the proper use of authority plays an indis-pensable role in the accumulation and application of knowledge.”2
In order to count as evidence, the testimony must be from an honest and reliable authority on a matter in the person’s field of expertise. In that case, “The appeal to a reliable authority is legitimate, for the testimony of a reliable authority is evidence for the conclusion.”3 Thus, while a Hollywood starlet’s endorsement of a com-mercial product does not count as evidence, the expert testimony of a DNA specialist concerning blood found at the scene of a crime does. When I quote recognized authorities like Hawking, Ruse, Kremer, Johnson, and others concerning matters in their respective fields of expertise, particularly about the state of scholarly opinion in their fields, this most certainly does count as expert testimony and, hence, evidence for the matter in question. I might note in this connection that Sinnott-Armstrong’s dismissal of scholarly opinion concerning the empty tomb of Jesus on the grounds of “bias” is quite naïve. As the New Testament scholar R. T. France points out with respect to the historical study of the four Gospels, “Ancient historians have sometimes observed that the degree of skepticism with which New Testament scholars approach their sources is far greater than would be thought justified in any other branch of ancient history. Indeed, many ancient historians would count themselves fortunate to have four such responsible accounts, written within a generation or two of the events, and preserved in such a wealth of manuscript evidence.”4 Ironically, Sinnott-Armstrong himself not in-frequently recurs to expert testimony in his criticisms, though I am struck by the weakness of that testimonial evidence (e.g., the infor-
God?
mation on Mithraism communicated to him by one Susan Ackerman). Of course, the ultimate question concerns the evidence upon which scholarly opinion is based, and I’m more than ready to discuss that evidence.
With these preliminaries out of the way, let’s now get down to a detailed examination of my arguments and Sinnott-Armstrong’s criticisms of them. I propose that we look at the arguments one premise at a time.
Cosmological Argument
My first reason, that God Makes Sense of the Origin of the Universe, is a version of the cosmological argument. Sinnott-Armstrong discusses it in section 4 of his response. Attentive readers will notice how seriously misstated his rendition of my argument in his second paragraph is. Let’s look at the argument as I actually formulated it.
1. Whatever Begins to Exist Has a Cause Sinnott-Armstrong replies, “Craig denies that any event can be uncaused, but this is contrary to standard quantum theory.” Both clauses of this sentence are, unfortunately, misrepresentations.
First, the clause “Craig denies that any event can be uncaused”
misrepresents premise (1), which has nothing to do with events at all, but with things. It states that things do not come into existence without a cause. The premise is fully compatible with events happening without a cause and so leaves that question open. What it denies is that things pop into being without causes.
The second clause, “this is contrary to standard quantum theory,”
misrepresents quantum theory. Any physical theory is comprised of two parts: a mathematical formulation and a physical interpretation of the mathematical formulas. While the mathematical core of quantum theory has been confirmed to a fantastic degree of precision, there are at least ten different physical interpretations of the mathematics, and no one knows which of these, if any, is correct, since they are all empirically equivalent.5 Only some of these, principally the so-called Copenhagen Interpretation, are causally indeterministic. Others are fully deterministic. In order to show that quantum theory proves that events can happen without causes, Sinnott-Armstrong would have to show that these other interpretations of
57
quantum theory are not as good as the Copenhagen Interpretation, which is, of course, impossible to do.
Moreover, even in the Copenhagen Interpretation things don’t come into being without a cause. It’s true that in this interpretation so-called virtual particles can arise spontaneously out of the quantum vacuum. But Sinnott-Armstrong recognizes that the quantum vacuum is not nothing; rather it’s a sea of fluctuating energy that serves as the indeterministic cause of such virtual particles. The philosopher of science Bernulf Kanitscheider emphasizes with respect to quantum vacuum models of the origin of the universe, The violent microstructure of the vacuum has been used in attempts to explain the origin of the universe as a long-lived vacuum fluctuation. . . . From the philosophical point of view it is essential to note that the foregoing is far from being a spontaneous generation of everything from naught, but the origin of that embryonic bubble is really a causal process leading from a primordial substratum with a rich physical structure to a materialized substratum of the vacuum. Admittedly this process is not deterministic, it includes that weak kind of causal dependence peculiar to every quantum mechanical process.6
Thus, even in the disputed Copenhagen Interpretation, the quantum vacuum is a physical cause of the entities it is alleged to spawn.
The bottom line is that if we think that things cannot just pop into being out of nothing, which surely seems reasonable, then quantum theory does not constitute a defeater of that premise.
2. The Universe Began to Exist
You’ll recall that I presented both a philosophical argument as well as scientific evidence for this premise. Let’s examine each in turn.
Philosophical Argument. Sinnott-Armstrong raises three objections to my philosophical argument for the finitude of the past: 1. It would prove that there are not an infinite number of numbers.
Right! But this conclusion is problematic only if one is a Platonist, who believes that numbers are actually existing, mind-independent things. Mathematics doesn’t commit us to Platonism, and existence claims in infinite set theory don’t commit us to the mind-independent existence of such entities. As Kasner and Newman nicely put it, “the
God?
infinite certainly does not exist in the same sense that we say, ‘There are fish in the sea.’ Existence in the mathematical sense is wholly different from the existence of objects in the physical world.”7 My claim here is that although infinite set theory and transfinite arithmetic each constitutes a logically consistent universe of discourse, given their axioms and conventions, the existence of an actual infinite in the real world leads to intolerable absurdities. My argument does not deny the coherence of that universe of discourse. Thus, we needn’t fear that my argument will relieve undergraduates of learn-ing the calculus!
2. We can forbid mathematical operations on infinite quantities that lead to contradictions. That works great in the postulated mathematical realm of discourse; but in the real world you can’t stop people from taking away real objects or dividing real collections! Of course, “mathematicians today recognize that infinity can be handled without contradiction.” They do so by axiomatization and conventional rules that are laid down for their abstract universe of discourse but which make no claim for the real world.8 Sinnott-Armstrong’s assertion that zero corresponds to a really existing entity seems confused: when we say that there are zero elephants in this room, we do not mean that there are elephants in this room, and the number of them is zero! Rather we mean that there just is no such entity in the room.
3. There are counter-examples of really existing actual infinites.
Sinnott-Armstrong mentions two: (i) The real numbers between 1
and 2. To claim this as a bonafide counter-example presupposes that there are such mind-independent entities as numbers, which begs the question in favor of Platonism. And Sinnott-Armstrong says nothing to show that the intuitionists (or constructivists mentioned in his note 5), who deny even the mathematical existence of the actual infinite, are wrong. (ii) The spacetime continuum. The alleged counter-example of moving my hand across a distance presupposes that physical intervals are composed of an actually infinite number of points, which begs the question. If the interval is not a construction of points, but is logically prior to any divisions we care to make of it, then the potential infinite divisibility of the interval does not imply that it is divided into an actually infinite number of parts.
59
In presenting these putative counter-examples, I think that Sinnott-Armstrong has failed to understand the nature of what philosophers call “defeaters.” Defeaters can be of two sorts: rebutting defeaters, which try to show that some position or statement is false, and undercutting defeaters, which try to show that a position or statement has not been shown to be true. The appeal to counter-examples is an attempt to offer a rebutting defeater. The counter-example is alleged to falsify the claim against which it is directed. In order to turn back the force of his alleged counter-examples, all I have to do is offer an undercutting defeater of his examples by showing that there are other equally good ways of construing them that do not involve commitment to actual infinites. I don’t have to show his construal to be false but merely that he hasn’t shown it alone to be true. It will then be up to Sinnott-Armstrong to show us why my construal isn’t just as viable. It is he who bears the burden of proof to show that his alleged counter-examples commit us to the existence of actual, rather than merely potential, infinites; otherwise they do not defeat my argument against the actual infinite based on the absurdities that its real existence would lead to.
Hence, I really don’t find any of these objections to my philosophical argument to be persuasive.
Scientific Evidence. Sinnott-Armstrong presents two objections to my appeal to the evidence of astrophysical cosmology for the beginning of the universe:
1. The inference to God depends on a questionable interpretation of the physics. First, it’s important to realize that I make no inference to God from physics; that misimpression arises only from Sinnott-Armstrong’s misstatement of my argument. Rather the physical evidence is adduced in support of the premise The universe began to exist. This is a religiously neutral statement, which may be found in almost any textbook on astrophysical cosmology.
Second, the issue is not whether the conclusion that the universe began to exist is “essential to the physics or required by the evidence,” but whether the evidence makes that conclusion more plausibly true than its negation.9 Sinnott-Armstrong really does the reader a disservice by wrongly suggesting that recent discoveries,
God?
which are wholly consonant with the Big Bang model, somehow cast doubt upon it.10
Third, Sinnott-Armstrong confuses two questions: (i) Does the Standard Model predict a beginning to the spacetime universe? And (ii) Is the prediction of the Standard Model correct? There is simply no ambiguity about the answer to (i). As I explained in my opening statement, in the Standard Model there is just nothing prior to the initial cosmological singularity. The beginning predicted by the Standard Model involves an initial cosmological singularity, which constitutes an edge or boundary to physical space and time. Physicists John Barrow and Frank Tipler emphasize, “At this singularity, space and time came into existence; literally nothing existed before the singularity, so, if the Universe originated at such a singularity, we would truly have a creation ex nihilo.”11
As for (ii), while certainty is of course impossible here, it is note-worthy that every attempt to avert the absolute beginning predicted by the Standard Model has failed.12 Models that seek to avert the prediction of the Standard Model of an absolute beginning of the universe have time and again been shown either to be untenable or else to imply the very beginning of the universe that their proponents sought to avoid. For example, the problem with the obsolete quantum vacuum models appealed to by Sinnott-Armstrong is that they predicted that universes would form at every point in the quantum vacuum, so that given an infinite past, universes would come into being at every point in the vacuum and would eventually all collide and coalesce into one infinitely old universe, which contradicts observation. Sinnott-Armstrong’s response to this shortcoming shows that he doesn’t get it. He says in his note 8, “the universe that we observe is (in a way) infinitely old in this view, even if its classical phase (which is the phase that we observe) is not infinitely old. So it is not clear what Craig’s objection is.” The objection is that in such theories we wouldn’t be observing a young classical phase. For the various baby universes formed within the womb of the wider Mother Universe would, given infinite time, have all run into one another as they grew older and so merged into one tired, old universe, which contradicts our observation of a relatively young universe. According to Britain’s leading quantum cosmologist, Christopher Isham, this difficulty proved “fairly lethal” to vacuum fluctuation models, and so they were
61
“given up twenty years ago and nothing much has been done with them since.”13 Sinnott-Armstrong is therefore just incorrect to call such theories “as likely as other hypotheses.” As for the recent discovery al-luded to by Sinnott-Armstrong that the cosmic expansion is accelerating rather than decelerating, this startling development supports the beginning of the universe by further undermining the old oscillating model, since an accelerating expansion cannot reverse under the force of gravitation into a Big Crunch, but will proceed endlessly.
2. Theism cannot explain the origin of the universe. Notice that this objection is irrelevant to the truth of the second premise, that the universe began to exist. It is actually a criticism of my subsequent argument for the personhood of the First Cause. But let’s go ahead and deal with it here. From the two premises, it follows that the universe has a cause. But is that cause personal or impersonal? Sinnott-Armstrong agrees that an impersonal, timeless cause cannot give rise to a temporal effect like the universe. But, he insists, neither can a personal cause, because no coherent doctrine of God’s relationship to time can be formulated. I claimed, “God existing alone without the universe is either (i) before the Big Bang, not in physical time, but in an undifferentiated metaphysical time or else (ii) strictly timeless but . . . He enters into time at the moment of creation.”
Sinnott-Armstrong objects to (i) because then there was after all
“some kind of time before the Big Bang.” That’s right; but this implication is scientifically inconsequential, since such a time would be a metaphysical reality that physical cosmology does not study. Hence, its existence would contradict no scientific evidence or theory. As for (ii), he objects that a being that is “ ‘strictly timeless’ by its very nature” can never enter into time. This response illicitly presupposes that God’s temporal status is an essential, rather than contingent, property of God, which is not incumbent upon the traditional theist. If God is contingently temporal or atemporal (as I happen to think), then His entering into time is not incompatible with His nature. For more on this very important issue, see our discussion of
“The Problem of Action” in chapters 4–6.
Since my cosmological argument is a deductive argument, Sinnott-Armstrong’s final animadversions concerning inductive arguments are irrelevant.
God?
Teleological Argument
My second reason, that God Makes Sense of the Fine-Tuning of the Universe for Intelligent Life, is a form of the teleological argument.
Sinnott-Armstrong treats it in section 5 of his response. Let’s examine each of its premises.
1. The Fine-Tuning of the Universe Is Due to Either Law, Chance, or Design
This first premise presupposes that the universe is in fact fine-tuned for intelligent, carbon-based life and states the three possible explanations of this fact. Sinnott-Armstrong is skeptical that the universe is fine-tuned in the way that I describe. He is suspicious of the probabilities involved because, he claims, we have no non-arbitrary way of counting the total number of values that a constant or quantity could assume, nor the range of values that are life-permitting.
This denial of the fact of fine-tuning is remarkable. Otherwise sober cosmologists are not beating one another out the back door to adopt metaphysically speculative scenarios of a World Ensemble for no good reason! As for Sinnott-Armstrong’s misgivings, the range of life-permitting values can in many cases be confidently established because if the values were allowed to vary beyond a certain restricted range, neither stars nor planets could exist (not to speak of biological organisms) and in some cases none but the lightest elements and in certain cases not even matter itself.14 As for estimating the range of possible values that the various constants and quantities might assume, we may take as our limit those values that are universe-permitting.15 For example, if gravity were to have a value 10100 times its present value, there wouldn’t even be a universe, but a mere singularity. Thus, we needn’t worry about the range of possible values extending to infinity. When you compare the range of possible values of the fundamental quantities permitted by the laws of nature with the range of life-permitting values, you find either that the range of life-permitting values is exceedingly small in comparison with the wider range of assumable values or else that the actual value of some constant or quantity falls improbably close to the edge of the life-permitting range. Well-established cases of this sort of fine-tuning include the strong force, the weak force, the electromagnetic force, the gravitational force, the proton-neutron mass ratio, and the cos-
63
mological constant.16 Fine-tuning is a scientifically established fact of nature that must be squarely faced.
Premise (1) states that there are only three alternatives available for explaining this remarkable fine-tuning: physical law, chance, or design. In his discussion, Sinnott-Armstrong tacitly admits that these are the only three alternatives. For of the “several competing alternatives” that he says are available to explain fine-tuning—namely, God, one Big Bang and chance alone, multiple universes each with its own Big Bang, or tracker fields—the first of these represents design, the second chance, the third chance again, and the fourth law.
So the question is: which is the best explanation?
2. It Is Not Due to Law or Chance
Sinnott-Armstrong tends to conflate these two alternatives, asserting, “when atheists ascribe fine-tuning to chance, they admit the possibility of deeper explanations.” This is incorrect; ascription to law allows deeper explanations, but chance takes the values of the vari-ables to be brute facts, just inexplicable givens. The only thing the atheist can do in order to supplement the chance hypothesis is to multiply his probabilistic resources by positing the existence of a World Ensemble in order to make the odds of the existence of a life-permitting universe more tractable. Let’s consider, then, these alternatives in order.
Law. Sinnott-Armstrong asserts that so-called tracker fields might explain fine-tuning. He refers here to the article in Scientific American on the acceleration of the cosmic expansion. The hypothetical field that drives the acceleration is called “quintessence.” It has been speculated that this quintessence field tracks the matter fields in the expanding universe in such a way as to diminish a very large initial value of the cosmological constant to nearly zero today. It’s immediately obvious that this hypothesis is, at face value, an attempt to explain merely one of the many finely tuned constants requisite for life, namely, the cosmological constant, and thus hardly constitutes an explanation of the universe’s fine-tuning. In any case, some as-trophysicists have complained that such a hypothetical field seems unacceptably ad hoc, and others have noted the severe problems facing such a proposed field.17 Finally, Collins points out that “even
God?
if such a field were discovered, it would have to have just the right (‘fine-tuned’ or ‘well-designed’) mathematical form to overcome the severe problems facing such proposals. This would reintroduce the issue of fine-tuning and design at a different level, though in a mit-igated way.”18 This has been the pattern with attempts to explain fine-tuning by physical law: Like a stubborn bump in the carpet, fine-tuning is suppressed at one point only to pop up at another.
Quoting Carr and Rees to the effect that “even if all apparently an-thropic coincidences could be explained [in terms of some grand unified theory], it would still be remarkable that the relationships dictated by physical theory happened also to be those propitious for life,” Collins concludes, “It is very unlikely, therefore, that these cases of fine-tuning and others like them will lose their significance with the further development of science.”19
Chance. Sinnott-Armstrong presents two considerations in defense of chance.
1. Improbability alone does not prove design. Of course; I made that point myself. As I presented the argument, it’s the combination of improbability with an independently given pattern that discredits chance. Imagine that Bob, who was born on August 8, 1948, receives a car for his birthday with the license plate BOB 8848. Would he be rational to shrug this off as due to chance on the ground that that combination of letters and numbers is as probable as any other? It is the combination of pattern and improbability that tips us off to design rather than chance. Given the independent pattern of vari-ables necessary for life and the enormous improbability of those values obtaining in the Big Bang, the chance hypothesis is ruled out in favor of design.
Incredibly, Sinnott-Armstrong thinks that the unimaginable improbability of your picking a white ball rather than a black ball in the lottery I envisioned (where one white ball is mixed with a billion, billion, billion black balls, and you are given one pick) should not lead you to doubt that your picking white was the result of sheer chance! I take his intransigence on this score as illustrative of the willful blindness to which determined atheism leads. For this is a classic case of a design inference: since it is overwhelmingly more probable that your pick be black rather than white, you should rea-
65
sonably conclude that your pick was not by chance. We can make this inference even more perspicuous without affecting the probabilistic features of the situation by imagining that you had to pick the one white ball several times in a row or you would be shot. Would any reasonable person who survived not think that the whole thing was rigged? Of course you would!
I presented the teleological argument using Dembski’s analysis of design inferences. But an approach using the probability calculus is feasible as well. As Collins frames the argument, one compares the conditional probability of fine-tuning relative to the hypothesis of a Cosmic Designer, on the one hand, to the conditional probability of fine-tuning relative to the atheistic single universe hypothesis on the other. What probability theorists call the “likelihood” of the design hypothesis just is the conditional probability of the fine-tuning given a Designer, and the “likelihood” of the atheistic hypothesis just is the conditional probability of the fine-tuning given a single atheistic universe. Sinnott-Armstrong admits that the probability of the fine-tuning given God’s existence seems high. But he fails to compare it with the probability of the fine-tuning given the atheistic single universe hypothesis. Fine-tuning is much more probable given God’s existence than it is given a single atheistic universe, since in the absence of a Designer there’s just no reason to expect the values of the various constants and quantities all to fall into the narrow, life-permitting range. According to a principle that probability theorists call the Likelihood Principle, the evidence of the fine-tuning of the universe therefore confirms the existence of a Cosmic Designer over the atheistic single universe hypothesis, or in other words, the hypothesis of a Cosmic Designer has a greater likelihood than the hypothesis of a single atheistic universe. Thus, fine-tuning gives us good reason to believe in the theistic hypothesis rather than in the atheistic hypothesis. Notice that the conditional probability of God’s existence given the fine-tuning of the universe, mentioned by Sinnott-Armstrong, doesn’t even come into play in the argument.
2. The Many Worlds Hypothesis makes the chance hypothesis tractable. Yes, one can so multiply one’s probabilistic resources (as by postulating billions of picks in the lottery) that the odds become acceptable. But then the question is whether it is more reasonable to postulate a World Ensemble or a Designer to explain fine-tuning.
God?
I presented four reasons why the Design Hypothesis is a better explanation than the World Ensemble Hypothesis: (1) It is simpler; (2) There is no known way for generating a randomly ordered World Ensemble; (3) There is independent evidence for the existence of God in contrast to a World Ensemble; and (4) The World Ensemble is incompatible with observations predicted by evolutionary biology.
Sinnott-Armstrong denies the greater simplicity of Design over a World Ensemble because the many postulated universes are not new types of things, as is God. But here he has confused simplicity with what we might call familiarity. The brute postulate of an indefinite collection of randomly ordered universes is definitely more complex than the postulate of a single Designer, even if universes are more familiar entities to us than are Cosmic Designers. There are various virtues of what counts as a best explanation, and I imagine familiarity is one of them.20 But that is quite a different virtue than simplicity. In any case, the proposed hypothesis is not God as such, but intelligent design; and intelligent design is not at all an unfamiliar explanation. Thus the postulate of a Cosmic Designer is another token of the same type.
Sinnott-Armstrong’s response to my remaining three objections is faltering. He tries to dismiss them as “persistent puzzles,” none of which “proves God.” But, of course, that’s not their intended purpose. They’re offered as reasons for thinking that the hypothesis of a World Ensemble is not as good an explanation as the Design Hypothesis. Nor are they mere puzzles, any more than the difficulty that undermined Boltzmann’s Many Worlds Hypothesis was a mere puzzle. Notice, too, that the inference to intelligent design, whether along the lines of Dembski’s statistical approach or along the lines of the probability calculus as advocated by Collins, is a principled inference, not “a quick fix for ignorance.” Nor is it a science-stopper, as Sinnott-Armstrong alleges, since the design theorist will formulate tests to falsify his hypothesis in hopes that it will be corroborated, thereby furthering the development of rival hypotheses.
Collins argues that the hypothesis of design in fact makes testable predictions, since it leads one to expect that cases of what he calls
“one-sided fine-tuning,” in which a constant can be shown to have an actual value very near to one edge of the life-permitting range, will turn out to involve two-sided fine-tuning, lying very near both edges of the life-permitting range. By contrast, the hypothesis of a
67
World Ensemble threatens to be a real science stopper if we have no way of knowing whether or not these other universes actually exist. Indeed, it’s debatable whether the postulation of such a World Ensemble to dismiss the appearance of design would not undermine the scientific enterprise, since any phenomenon that appears initially puzzling can be dismissed as happening by chance alone somewhere in the infinite ensemble of worlds!
With the failure of the hypotheses of law and chance, the conclusion follows that the best explanation of the fine-tuning is design.
Thus, the teleological argument nicely complements the cosmological argument.
Moral Argument
My third reason that God Makes Sense of Objective Moral Values in the World is a version of the moral argument for God. Sinnott-Armstrong tries to “get this argument out of the way” in section 1 of his response. Again, let’s look at each of the two premises.
1. If God Does Not Exist, Objective Moral Values Do Not Exist
Sinnott-Armstrong insists that “many atheists are happy to embrace objective moral values.” You bet they are! It would be hard to live without them. College sophomores may give lip service to relativism, but even they don’t really live that way. Philosophers, who are more reflective, tend to recognize the objective existence of moral values and duties. The question is whether on the atheistic world view they have any basis for doing so or whether they simply want to enjoy the benefits of theft over honest toil. Sinnott-Armstrong makes two points in defense of objective moral values within an atheistic world view.
1. The morality of an action is determined by whether it causes unjustified harm to another person. This claim is, however, very problematic for the atheist. First, given atheism, why think that it is true?
Why, given atheism, think that inflicting harm on other people would have any moral dimension at all? Sinnott-Armstrong rejects Atheistic Moral Realism. So why would it be wrong to hurt another member of our species? He answers, “It simply is. Objectively. Don’t you agree?” Of course, I agree that it is wrong, since I am a theist. But I can’t see any reason to think that it would be wrong if atheism were
God?
true. For on atheism we are just animals, and animals don’t have moral duties. Sinnott-Armstrong retorts that human beings are unlike other animals because we are “moral agents.” But this claim seems to assume precisely what needs to be proved, unless some neutral content can be given to the concept of “moral agent.”
Sinnott-Armstrong states that moral agents “make free choices” and determine their actions by their “conception of what is moral or not.”
This response is ironic precisely because Sinnott-Armstrong denies that we do make free choices in the libertarian sense21; he believes that all our actions are determined, which leaves no room for moral agency. In any case, I see no reason to think that these conditions are sufficient, rather than merely necessary, conditions of being a moral agent. Even if we freely make decisions based on our conceptions of what is moral/immoral, the question remains whether our conceptions of what is moral/immoral have any objective validity.
Second, Sinnott-Armstrong’s assertion appears to be circular. Not all harm done to other members of our species is said to be wrong, only unjustified harm. But what does that mean? It can’t mean merely that the perpetrator has a reason for acting as he does, for murderers and rapists have such reasons and may well be pursuing their self-interest. Rather the reason must be a morally sufficient reason if it is to exonerate doing harm. But then aren’t we presupposing morality in trying to ground morality? We’re saying that an action is morally unjustified if it causes harm that is morally unjustified—no duh!
2. Theism provides no better grounding of morality than atheism.
Sinnott-Armstrong recognizes that “Philosophers still might long for deeper explanations of why it is immoral for moral agents to cause unjustified harm.” But, he insists, this dissatisfaction would be problematic for atheism only if the theist could provide a better explanation, which Sinnott-Armstrong denies. He thinks the Euthyphro dilemma exposes the explanatory deficiency of theism by showing that God’s commands are either superfluous or arbitrary. But it seems to me that an appropriately formulated divine command morality, such as has been articulated by Adams, Quinn, Alston, and others,22 eludes the horns of the Euthyphro dilemma by forging a third alternative: our moral duties are grounded in the commands of a holy and loving God. God’s very nature is what Plato called the
“Good”: He is essentially kind, just, generous, and so forth. Just as
69
it made no sense prior to 1960 to ask whether the standard meter bar at the Bureau de Poids et Mesures in Paris was actually a meter long (since its length determined what a meter was), so it makes no sense to measure God’s moral character against some abstract standard, since His nature is determinative of what goodness is. His nature expresses itself toward us in the form of moral commands which, issuing from the Good, become moral duties for us. Thus, God’s commands are not arbitrary. If it be asked, “Why pick God’s nature as constitutive of the Good?” the answer is that God, by definition, is worthy of worship and only a being that is morally perfect is so worthy. Theism thus succeeds precisely where atheism fails, in providing a metaphysical foundation for the objective existence of moral values and duties.
2. Objective Moral Values Exist
Sinnott-Armstrong agrees. His only confusion here is to think that our ability simply to see that rape is wrong implies that no account need be given of why rape is wrong. That is to confuse moral epistemology with moral ontology.
From the two premises, the conclusion that God exists follows deductively.
Resurrection of Jesus
My fourth reason is that God Makes Sense of the Life, Death, and Resurrection of Jesus. As Sinnott-Armstrong discerns, this is properly an argument from miracles, indeed, the central miracle of the New Testament. The form of this argument is that of an inductive inference to the best explanation of a given body of historical facts.
Let’s review each premise.
1. There Are Four Established Facts Concerning the Fate of Jesus of Nazareth: His Honorable Burial by Joseph of Arimathea, the Discovery of His Empty Tomb, His Post-Mortem Appearances, and the Origin of His Disciples’ Belief in His Resurrection
Let’s take them one at a time.
The Honorable Burial. Sinnott-Armstrong doesn’t dispute this fact. Yet this fact is highly significant because it implies that the lo-
God?
cation of Jesus’ gravesite would in time have become common knowledge in Jerusalem to both Jew and Christian alike.
The Empty Tomb. Sinnott-Armstrong would rebut the evidence by asserting that o ur records come from years later, during which time distortions could arise. That’s correct; but the question remains whether or to what degree such distortions did arise. Sinnott-Armstrong’s sweeping generalization that “distortions could arise”
does nothing either to show that the accounts of the discovery of Jesus’ empty tomb are in fact unhistorical distortions or to refute the specific evidence I presented for the historical core of the narrative.
The sources for Greco-Roman history are usually biased and usually one or two generations removed from the events they record, yet historians reconstruct with confidence the course of Greco-Roman history.23 If one could really refute the specific evidence for putatively historical accounts so summarily as Sinnott-Armstrong seems to think, one might as well just close the history department at the university. In fact, most historical scholars who have examined the accounts of the empty tomb of Jesus have concluded that distorting influences have not expunged the historical core of the accounts.
The Post-Mortem Appearances. Sinnott-Armstrong merely asks, How do we know some witnesses did not hear stories about the others? He misunderstands the point. Our records of the appearances are literarily independent of one another, as a comparison of them reveals. That creates a predisposition toward historicity, since multiple, independent sources concerning an event make it less likely that the event was purely legendary or invented.24 Virtually no historian denies that the disciples had such experiences; the only question is, how are they best explained?
The Origin of the Disciples’ Belief. Sinnott-Armstrong asserts that stories about a rising Messiah were common. I’m afraid that he is just misinformed on this score. There was absolutely no anticipation that Messiah, when he came, would, instead of vanquishing Israel’s enemies and restoring the throne of David, be shamefully executed by them.25 And, of course, neither was there then any expectation that Messiah would be raised from the dead. Thus, the disciples’
suddenly and sincerely coming to believe that Jesus was risen from the dead cannot be explained as a result of Jewish influences.
71
Sinnott-Armstrong’s mention of Mithraism is baffling, since Mithraism does not even purport that Mithras was raised from the dead.26 Perhaps Sinnott-Armstrong hopes to resuscitate the hypothesis of the nineteenth-century History of Religions School that myths of dying and rising pagan deities like Osiris and Adonis led the disciples to believe that Jesus rose from the dead. That hypothesis, however, collapsed early in the twentieth century due to two factors: (1) The purported parallels were spurious. These pagan deities were merely mythological symbols of the crop cycle, as veg-etation dies during the dry season and comes back to life during the rainy season. Moreover, orientalists have questioned whether it is proper to speak of “resurrection” at all with regard to these deities.
Observing that the idea of resurrection is late and tenuous in the case of Adonis, practically non-existent in the case of Attis, and in-applicable in the case of Osiris, who merely lives on in the Under-world, M. S. Smith concludes, “it is presently impossible to accept a general category of a ‘dying and rising god’ in the ancient Mediter-ranean and Levantine world.”27 (2) The causal connection is missing. As the eminent New Testament historian Martin Hengel explains, it was precisely because of Jewish belief in the resurrection of the dead that the Hellenistic mystery religions “could gain virtually no influence” in Jewish Palestine.28 Thus, even the sceptical critic Hans Grass admits that it would have been “completely un-thinkable” for the Jewish disciples of Jesus to have sincerely come to believe that Jesus was risen from the dead because of the influence of myths of pagan deities.29
We therefore have good grounds for the four facts that serve as our inductive data base. It’s worth emphasizing that these facts, while by no means universally acknowledged, are accepted today by most New Testament historians who have written on the subject.
2. The Hypothesis “God Raised Jesus from the Dead” Is the Best Explanation of these Facts
Sinnott-Armstrong raises two objections:
1. Someone stole the body. Notice that apart from his unsubstanti-ated assertion that “many people had motive and opportunity to
God?
move the body,” Sinnott-Armstrong provides no evidence for this hypothesis. It thus has no more probability than what it acquires from our general background knowledge. On the other hand, against it stands the following specific evidence: (1) No one had a motive to steal the body. Grave robbers are after the goods interred with the body, such as jewelry, not the corpse itself. In imagining other persons with a motive, I suspect that Sinnott-Armstrong is thinking anachronistically in terms of someone’s faking the resurrection. (2) No party can be plausibly thought to have removed the body of Jesus from the tomb. Not the disciples, for their sincerity was evident in their willingness to die for their belief that Jesus had been raised; not the authorities, for they would have produced the body once the disciples began to preach the resurrection; not ordinary Jews, for there was just no reason to do such a thing. There just isn’t anybody who can be identified as the alleged perpetrators.
(3) No one apart from Joseph of Arimathea (and anyone with him) and the women even knew immediately following the crucifixion where the body of Jesus was laid. Joseph must have surprised everyone when he gave Jesus an honorable burial rather than allowing the Romans to dispatch the corpse in the customary manner. (4) The time was too short for such a plot to be hatched and executed.
Jesus was unexpectedly given an honorable burial late Friday af-ternoon; by dawn on Sunday the body was gone. The window of opportunity for coming up with such an idea, assembling the people required, and carrying out the theft is so narrow as to militate against such an explanation. (5) The theft would have eventually come to light. Such conspiracies inevitably unravel, and the Jewish authorities, who had bought off Judas to betray Jesus, would have paid handsomely for any information concerning a theft of Jesus’
corpse. Instead we find the earliest Jewish polemic lamely charging that the disciples themselves had stolen Jesus’ body (Matthew 28.13). (6) The grave clothes left in the tomb (Luke 24.12; John 20.7) show that the tomb had not been emptied by robbery. Tomb robbers would not undress and unbind the corpse and carry it off naked, with limbs dangling, leaving the wrappings behind. (7) The theft hypothesis has deficient explanatory scope. It aims to explain the empty tomb but says nothing about the resurrection appear-
73
ances or the origin of the disciples’ belief in Jesus’ resurrection (which an empty tomb alone cannot account for). Thus, the theory is multiply flawed.
2. The resurrection hypothesis contradicts all the evidence for the laws of nature violated by it. This is the real reason Sinnott-Armstrong prefers the theft hypothesis, despite its want of positive evidence and its many deficiencies! But this objection is merely a reprise of Hume’s fallacious and oft-refuted argument against the identification of miracles.30 The evidence for the laws of nature counts against the hypothesis that Jesus rose naturally from the dead (virtually any hypothesis is more plausible than that!), but it doesn’t count against the hypothesis “God raised Jesus from the dead.”
Moreover, probability theory requires that we also consider the probability of our having the evidence for the empty tomb, resurrection appearances, and the origin of the disciples’ belief if Jesus had not risen from the dead, which might strike us as very low. Sinnott-Armstrong is correct that the resurrection hypothesis is ad hoc, in the sense of postulating an additional entity, God. But his own theory is also ad hoc in postulating without evidence conspirators who stole Jesus’ body. In any case, the ad hoc-ness of the resurrection hypothesis can be outweighed by its greater explanatory power, scope, and plausibility. The religio-historical context in which the resurrection occurred (Jesus’ unparalleled life and radical claims) makes it utterly unlike Sinnott-Armstrong’s fanciful ice cream illustration. Moreover, the resurrection hypothesis is part of a cumulative case for theism, which includes several other independent reasons to think that God exists.
Apart from an unjustified prejudice against miracles, I think that it’s hard to deny that the resurrection hypothesis is the best explanation of the evidence.
Immediate Experience of God
My fifth reason is that God Can Be Immediately Known and Experienced. Sinnott-Armstrong replies to this reason in section 3 of his chapter.
God?
Sinnott-Armstrong objects that the diversity of religious claims undermines the reliability of religious experience. But he didn’t reply to my answer to this anticipated objection.31 Unless I have some reason to think that my experience of the inner witness of God’s Spirit is delusory, the mere presence of other persons claiming a similar experience in support of counter-Christian claims does not in itself serve to make my belief improper. Moreover, the objection assumes that all religious experiences “seem similar from the inside.”
That assumption is false, as an examination of the variety of religious experiences reveals. The Hindu experience of one’s subsumption in the Whole is a very different experience from Christian experience of God’s love. The veridicality of Christian experience thus need not imply that other types of religious experience are not genuine apprehensions of God in certain aspects of His being.
What makes belief in God basic for some person is the fact that this person holds it without founding it on arguments or other beliefs, and what makes it properly basic for him is that it is not held arbitrarily but is grounded in his experience. That experience need not be so overwhelming that he would be crazy not to hold the belief in question; indeed, most properly basic beliefs (like one’s memory belief of what one had for lunch yesterday) are not of such a quality. Thus, my claim is not that atheists are crazy but that one who has an immediate experience of God, such as the Bible promises to those who sincerely seek Him, may know, in the absence of any defeaters, in a properly basic way that God exists.
In any case, the independent confirmation demanded by Sinnott-Armstrong is available for belief in the existence of the biblical God, as our previous four arguments show. But even in their absence, a person can be warranted in holding theistic belief in a properly basic way, even if others appeal to similar experiences to ground their incompatible beliefs in a basic way.
In summary, it seems to me that all five of my reasons to think that God exists survive Sinnott-Armstrong’s criticisms. Unless he can, then, provide sound, equally plausible arguments for atheism, the preponderance of the evidence favors theism, and the rational person will find in theism reason enough to merit his joy-ful assent.
75
Notes
1. Thus, insofar as the God of the Bible is omnipotent and omniscient, our cumulative case has justified our believing God to have even those properties.
2. Wesley C. Salmon, Logic, Foundations of Philosophy Series (Englewood Cliffs, N.J.: Prentice-Hall, 1963), 63. (How’s that for an appeal to authority!) 3. Ibid., 64.
4. R. T. France, “The Gospels at Historical Sources for Jesus, the Founder of Christianity,” Truth 1 (1985): 86. As examples, France mentions A. N. Sherwin-White, Roman Society and Roman Law in the New Testament (Oxford: Oxford University Press, 1963), 186–192, and A. T. Hanson, ed., Vindications (London: SCM, 1966), 42–43, 94–95.
5. For an elementary account of some of the options, see Nick Herbert’s delightful book, Quantum Reality: Beyond the New Physics (Garden City, N.Y.: Doubleday Anchor Press, 1985). Further interpretations have appeared since he wrote.
6. Bernulf Kanitscheider, “Does Physical Cosmology Transcend the Limits of Naturalistic Reasoning?” in Studies on Mario Bunge’s “Treatise,” ed. P.
Weingartner and G. J. W. Dorn (Amsterdam: Rodopi, 1990), 346–347.
7. Edward Kasner and James Newman, Mathematics and the Imagination (New York: Simon & Schuster, 1940), 61. For example, Alexander Abian interprets existence in set theory to mean merely that certain specified sets will be listed in an illusory table describing the theory of sets (see Alexander Abian, The Theory of Sets and Transfinite Arithmetic [Philadelphia: W. B. Saunders, 1965] 68).
8. Sinnott-Armstrong misleads in saying that Hilbert recognized that “his finitist project was undermined by Gödel’s incompleteness theorems.” What Gödel’s theorems undermined was Hilbert’s Formalist program of trying to prove the consistency of arithmetic in a finite number of steps. Nothing in Gödel’s theorems compels us to believe that actual infinites exist.
9. Actually, it is Sinnott-Armstrong’s interpretation of the physics that seems questionable. For example, the reason most physicists don’t like an initial cosmological singularity has virtually nothing to do with “doubts that gravity is always attractive,” as he claims, but because a singularity represents the breakdown of physics and so the terminus of science, an undesirable state of affairs for a scientist! It is the prospect of the initial singularity that has helped to drive the quest for alternative models. But even non-singular models like the Hartle-Hawking model still involve an absolute, if non-singular, beginning.
10. He has reference to an article by James Glanz, “Tiniest of Particles Pokes Big Hole in Physics Theory,” New York Times (Feb 8, 2001), pp. A1, A21, concerning the possible detection at Brookhaven National Laboratory of a precession in the motion of certain elementary particles. Let me explain. The precession counts in favor of a theory of elementary particle physics called supersymmetry. According to the Standard Model of particle physics, fundamen-
God?
tal particles are classed as either fermions (particles that make up matter) or bosons (particles that carry the four forces of nature). Fermions include quarks (which are the constituents of particles like protons and neutrons) and leptons (simple particles like electrons and neutrinos). Bosons include photons (which carry the electromagnetic force), gravitons (which carry the gravitational force), gluons (which carry the weak force), and intermediate vector bosons (which transmit the weak force). Many physicists find this model “ugly” because it lacks a unifying symmetry and is unwieldy. Various strategies for unifying these particles and forces are being explored. So-called Grand Unified Theories (GUTs) would unite the strong, weak, and electromagnetic forces at very high energy levels such as prevailed in the first split-second after the Big Bang into one Grand Unified force carried by hypothetical X and Y particles. Under such theories, quarks and leptons become at the same time interconvertible. Supersymmetry (SUSY) is an attempt to achieve further unification by incorporating gravitation into the GUT. In order to make fermions and bosons fully inter-changeable, they must be put into symmetrical pairings. SUSY would achieve this by postulating new particles like squarks and sleptons for fermions, and like photinos, gravitinos, gluinos, etc., for bosons. The Brookhaven results are the first empirical evidence in favor of SUSY, since the precession seems to suggest the existence of a yet undiscovered particle. Such discoveries were anticipated and fully consonant with standard Big Bang cosmology. In general we’d do well to reflect on the words of the Astronomer Royal Martin Rees, “Cosmological ideas are no longer any more fragile and evanescent than our theories about the history of our Earth. . . . The empirical support for a Big Bang ten to fifteen billion years ago is as compelling as the evidence that geologists offer on our Earth’s history” ( Just Six Numbers [New York: Basic Books, 2000], 10).
11. John Barrow and Frank Tipler, The Anthropic Cosmological Principle (Oxford: Clarendon Press, 1986), 442.
12. See my “The Ultimate Question of Origins: God and the Beginning of the Universe,” Astrophysics and Space Science 269–270 (1999): 723–740, or my
“Naturalism and Cosmology,” in Naturalism: A Critical Analysis, ed. Wm. L.
Craig and J. P. Moreland, Routledge Studies in Twentieth-Century Philosophy, (London: Routledge, 2000), 215–252. At the time of this writing all the buzz is Paul Steinhardt’s new cyclical model of the origin of the universe (see http://feynman.princeton.edu/⬃steinh/) based upon the periodic collision of three-dimensional membranes. It will be very interesting to see how well this new model fares under criticism! See Gary Felder, Andret Frolov, Lev Kofman, and Andrei Linde, “Cosmology with Negative Potentials,” http://arXiv.org/abs/
hep-th/0202017 (16 February 2002) and the therein cited literature; Arvind Borde, Alan Guth, and Alexander Vilenkin, “Inflation Is Not Past-Eternal,”
http://arXiv:gr-qc/0110012v1 (1 Oct 2001).
13. Christopher Isham, “Space, Time, and Quantum Cosmology.” Paper presented at the conference “God, Time, and Modern Physics,” March 1990; idem, “Quantum Cosmology and the Origin of the Universe.” Lecture presented at the conference “Cosmos and Creation,” Cambridge University, 14 July 1994.
77
14. Someone might say that if the fundamental quantities were radically varied perhaps forms of life as yet unimagined might have energed. It’s hard to see how any life at all could be possible in a world in which not even chemistry could exist; but let that pass. The more important point is that we need not consider such radically different worlds, but only those in our neighborhood. Imagine a fly resting in a large blank area of the wall. A single shot is fired, and the bullet pierces the fly. It is reasonable to conclude the shot was aimed, even if the wall outside the local area is covered with flies so that a randomly fired shot would strike one. Similarly, within our local area of possible universes, life-permitting universes are incomprehensibly rare, even if wildly different universes might permit some other form of life. (The fly illustration is John Leslie’s.) 15. I’m indebted to an excellent discussion of this question by Robin Collins, “Fine-Tuning and Comparison Range.” In press. Be on the lookout for his forthcoming book The Well-Tempered Universe, a tour de force on the problem of fine-tuning and design inference. Collins specifically addresses the concerns raised in the article by the McGrews and Vestrup, cited in Sinnott-Armstrong’s note 12.
16. These examples are discussed at length by Robin Collins, “Evidence for Fine-Tuning” in press.
17. Lawrence Krauss, “Cosmological Antigravity,” Scientific American (Jan-uary 1999): 59; V. Sahni and A. Starobinsky, “The Case for a Positive Cosmological Lambda-Term,” (28 April 1999); available online at http://xxx.lanl.gov/
abs/astro-ph/9904398. I’m indebted to Collins, “Evidence for Fine-Tuning,” for these references.
18. Collins, “Evidence for Fine-Tuning.”
19. Ibid. The quotation is from B. J. Carr and M. J. Rees, “The Anthropic Cosmological Principle and the Structure of the Physical World,” Nature 278
(12 April 1979): 612.
20. In their book A Subject with No Object (Oxford: Clarendon Press, 1997), 209, John P. Burgess and Gideon Rosen list seven features that tend to be used in guiding scientific theory choice. Among them are both minimality (or economy) of assumptions and coherence with familiar, established theories.
21. He makes this clear in our debate “Do Evil and Suffering Disprove God?” (April 2, 2000), Wooddale Church, Eden Prairie, Minn.
22. Philip L. Quinn, Divine Commands and Moral Requirements (Oxford: Clarendon Press, 1978); Robert Merrihew Adams, Finite and Infinite Goods (Oxford: Oxford University Press, 2000); William Alston, “What Euthyphro Should Have Said,” in Philosophy of Religion: A Reader and Guide, ed. Wm.
L. Craig (Edinburgh: Edinburgh University Press, 2001; New Brunswick, N.J.: Rutgers University Press, 2001), 283–298.
23. See A. N. Sherwin White, Roman Society and Roman Law in the New Testament (Oxford: Clarendon Press, 1963), 186–191.
24. On the importance of early, multiple attestation, see Marcus Borg,
“Seeing Jesus: Sources, Lenses, and Methods,” in The Meaning of Jesus, by Marcus Borg and N. T. Wright (San Francisco: HarperCollins, 1999), 12.
God?
25. A point emphasized by N. T. Wright, “The Crux of Faith,” in The Meaning of Jesus, by Marcus Borg and N. T. Wright (San Francisco: HarperCollins, 1999), 102, 117–118.
26. For a nice popular account see Ronald Nash, Christianity and the Hellenistic World (Grand Rapids, Mich.: Zondervan, 1984), 143–148, 169–173.
27. Mark S. Smith, The Ugaritic Baal Cycle, Vol. I. (Leiden: E. J. Brill, 1994), 70.
28. Martin Hengel, Judenthum und Hellenismus, WUNT 10 (Tübingen: J. C. B. Mohr, 1969), 368–369.
29. Hans Grass, Ostergeschehen und Osterberichte, 4th ed. (Göttingen: Vandenhoeck & Ruprecht, 1970), 133.
30. For a recent, devastating critique by a prominent philosopher of science, who is himself an agnostic, see John Earman, Hume’s Abject Failure (Oxford: Oxford University Press, 2000). See further, George Campbell, Disserta-tion on Miracles (1762; rep. ed.: London: T. Tegg & Son, 1834); Gottfried Less, Wahrheit der christlichen Religion (Göttingen: G. L. Förster, 1776); William Paley, A View of the Evidence of Christianity, 5th ed. 2 vols. (London: R. Faul-der, 1796; reprint ed.: Westmead, UK: Gregg, 1970); Richard Swinburne, The Concept of Miracle (New York: Macmillan, 1970); John Earman, “Bayes, Hume, and Miracles,” Faith and Philosophy 10 (1993): 293–310; George Mavrodes,
“Miracles and the Laws of Nature,” Faith and Philosophy 2 (1985): 333–346; William Alston, “God’s Action in the World,” in Divine Nature and Human Language (Ithaca, NY: Cornell University Press, 1989), 197–222.
31. See pp. 27–28.
This page intentionally left blank
Some Reasons to Believe that
There Is No God
Walter Sinnott-Armstrong
In the first part of this book, I criticized Craig’s arguments for the existence of God, but criticism goes only so far. Even if there is no decent argument for the existence of God, there also might not be any decent argument against the existence of God. Consequently, my criticisms are compatible with agnosticism, which is the view that we cannot know whether or not God exists. In this section, I will go further and present positive arguments for atheism, which is the view that God does not exist.
Why would anyone argue against the existence of God? If you feel sure that God does not exist, why come out of the closet and announce your reasons publicly? You are bound to make enemies.
You won’t convince anybody. Why bother? Many friends have asked me these questions. They deserve an answer.
My answer is that I am a teacher, so my job is to educate. I am also a philosopher. Philosophers question common assumptions and inspect the reasons for and against those assumptions. That is why I want to help readers get clear about the evidence for and against the existence of God, so that they can decide for themselves. If you look at the evidence with an open mind, then you will conclude that God does not exist, or so I believe. But even if you do not reach this conclusion, if you do achieve a better understanding of the evidence on both sides, then I will have achieved my main purpose.
81
God?
My goal is not to undermine religion. Some atheists want to sub-vert religion, because they see religion as the source of many social and personal problems. They have a point. Many wars in the past and today result (in part) from religious differences. Witness the Crusades long ago and the current conflicts in Ireland and the Middle East.
Many fine people have been executed, tortured, tormented, and os-tracized in the name of religion. Witness not only the Inquisition but also Matthew Shepard, who was tortured and killed because of his harmless sexual proclivities. Many more people suffer because religious institutions resist certain advances in education, culture, and medicine. Witness stem cell research, which promises to cure juve-nile diabetes along with other horrors; its funding is restricted by the United States government because of opposition by religious leaders (and almost nobody else). So religious belief is costly. Those costs are what motivate some atheists to argue against the existence of God.
I share many of these concerns, but they are not my reasons for arguing that there is no God. In my view, religion is not all bad.
Faith leads many people to perform wonderful acts of charity that they might not perform if they did not have faith. Faith can bind families and communities together in mutually beneficial ways. Some people need faith to face difficulties and find meaning and hope in their lives. Like law, science, art, and guns, religion is a powerful tool that can be used for great good as well as for great evil. I have no desire to obstruct the benefits of religion.
Luckily, the truth will not hurt us. Religion can play its positive roles without all of the fancy doctrines that theologians construct.
Most religious believers love the songs, stories, rituals, communities, and moral teachings of their religions. They do not understand or care much about obscure details of theology. Indeed, many people would come to doubt their religion if they thought through the extreme theories of theologians. Dubious doctrines, no matter how popular, weaken religion by making it more questionable. So the faithful should not worry about reconsidering dubious doctrines.
They should welcome this enterprise with an open mind.
One particular collection of doctrines will be my target here.
These doctrines are not common to all religions. They are not accepted by Hindus or Buddhists, or even by all Jews or Christians.
According to many historians, this collection of doctrines was not
Some Reasons to Believe that There Is No God 83
put together until the early Middle Ages, so it is not part of early Judaism or early Christianity, including the New Testament. This historical claim might be questioned, and some Christians might insist that these doctrines are essential to true Christianity. Still, many other people who call themselves Christians reject these doctrines, at least in the versions that cause trouble. Anyway, I take no stand on what counts as true Christianity. Whenever they got started, these doctrines became central to later Judaism and to Islam, as well as to many currently popular versions of Christianity. That is what makes it worthwhile to question them here.
What are these dubious doctrines? According to traditional (even if not original) Christianity, God is defined to be:
• All-good (⫽ God always does the best that He can)
• All-powerful (⫽ God can do anything that is logically possible)
• All-knowing (⫽ God knows everything that is true)
• Eternal (⫽ God exists outside of time)
• Effective (⫽ God causes changes in time)
• Personal (⫽ God has a will and makes choices) From now on, when I refer to God, I will mean a being with these defining features. When I argue against the existence of God, I will be arguing that nothing has all of these features.
My arguments will not be proofs in the mathematical sense. I don’t see how there could be any absolutely conclusive proof of any view about God, either that He does exist or that He does not exist. There is not much, if anything, of importance that one can prove with certainty beyond any doubt whatsoever. So I won’t even try to give a conclusive proof. What I will give is adequate evidence against the existence of God. That is enough to justify the belief that God does not exist.
This evidence comes in the form of three arguments (although more could be given). None of these arguments is original. Still, they are worth repeating, because many people do not fully appreciate their force.
1. The Problem of Evil
My first argument focuses on the first three defining features of God.
It is very common and simple: There is lots of evil in the world.
God?
There would not be so much evil if there were an all-good and all-powerful God. Therefore, there is no such God.
In its most powerful form, this argument focuses on examples like this: Many babies each year are born with Down’s syndrome. Most of these babies, with normal pediatric care, will grow up healthy. A significant number, however, have intestinal obstructions that will kill them if they do not receive an operation. Without the operation, dehydration and infection will cause these babies to wither and die over a period of hours and days. Today this operation is relatively simple, but not long ago these babies could not be saved. So just think about a baby born with Down’s syndrome and an intestinal blockage in 1900. This baby suffers for days, then dies. If the details of this example are doubted, or if one example does not seem to be enough, then I could describe many more birth defects and diseases that cause babies excruciating pain and death after only very short lives. There is no shortage of horror.
To complete the example, suppose a doctor could have cured the child and enabled it to live a happy life with a simple operation that would not cost much to anybody. Nonetheless, the doctor chose not to save the child. Why not? Imagine that the doctor just did not feel like operating that day. I would consider this doctor to be a moral monster. I assume that most of you would agree. If the doctor can save the child with little cost to anyone, but the doctor chooses to let it die, then the doctor cannot be a good person, to say the least.
Now apply these standards to God. In the traditional conception, God is all-powerful, so He could have saved the child easily. All He had to do was remove the intestinal obstruction. Doctors can do that, so God can, too. Better yet, God could help the baby secretly and painlessly by changing the child’s genes so that this baby would never have had Down’s syndrome or, at least, an intestinal blockage. God’s cure need not harm anyone else. The doctor might need the operating table for another patient, but God’s mercy would not use up any medical resources. The doctor might be too busy, but not God.
A doctor might worry that the child or the child’s parents will suffer more later, but God could prevent that future suffering as well.
All of this might take a miracle, but there is nothing logically impossible about God helping this baby with no harm to anyone, so God can do it.
Some Reasons to Believe that There Is No God 85
Given these alternatives, did God have an adequate reason to let this baby suffer and die? I cannot see any. I bet you can’t either. If God did have some reason to let this baby suffer and die, He could tell us His reason, and He would have no adequate reason not to tell us. Thus, since we see no adequate reason for it, this baby’s suffering is evidence that there is no God who has all of the features in the traditional conception.
Notice that each bit of unjustified evil is evidence against God, according to this argument. If I claimed to be all-good, one bad act would refute me. If I claimed to be all-powerful, you could disprove this claim by finding one thing that I cannot lift. Similarly, even one bit of unjustified evil disproves the existence of God. Still, it is worth adding that there is lots of evil in the world. Many children through the centuries have been born with intestinal blockages. There are many more kinds of birth defects that lead to short lives filled with intense suffering. Every day many children die from disease or mal-nutrition. The world includes floods, hurricanes, tornadoes, volcanoes, earthquakes, lightning, and many other natural disasters. Children and adults get trapped under rubble and then die after hours of horrible suffering. Take any individual natural disaster. An all-powerful God could prevent this natural disaster from occurring. He could do so without harming people as much as they are harmed by the natural disaster. So there seems to be no adequate reason why He should let these natural disasters occur. That means that God would not let them occur if He were all-good. But they do occur. So there is ample evidence against the existence of any all-powerful, all-good God.
That’s the basic idea, but my argument can be presented more formally and generally like this:
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.
Arguments like this are sometimes said to show that evil is logically inconsistent with an all-good and all-powerful God. I do not claim that
God?
much here. The point of my argument is, instead, that evil is evidence against the existence of an all-good and all-powerful God, at least when that evil is not logically necessary for any adequately compensating good. Thus, my argument is a version of what is called “the evidential argument from evil” rather than “the logical argument from evil.”
This argument depends on the notion of evil, which needs to be explained. When you call people evil, you probably think that their motives and characters are morally bad. In contrast, I will use the term “evil” more broadly to include anything that is harmful or bad, even if no moral agent causes it or could prevent it. In this wide sense, there are many disagreements about what is evil, but I need not settle them here. The only examples that I need are obvious.
Under “evil,” I include intense pain (or suffering), serious disabilities (mental and physical), and death. There are probably also other evils. I do not deny, for example, that unfairness and injustice (both distributive and retributive) are evils. Thus, I do not assume any crude form of utilitarianism. The reason why I focus on pain, disability, and death is only that there are plenty of these evils for my argument, and there is less question that they are evils. The other evils are harder to identify and more questionable, so I will not use them in my argument. In fact, I don’t even need all of the items on my list of evils. Even if one item on my list is questioned, the others should be enough. But it does strike me and almost everyone else as obvious that pain, disabilities, and death are evils.
What makes these things evil? That is a tough question that I do not need to answer here. It is enough for my argument that these things are evil, even if it is not clear what makes them evil or what it means to call them evil. Still, if you demand a general account of evil, my colleague Bernard Gert plausibly identifies an evil as anything that all rational people avoid for themselves unless they have an adequate reason to want it. Pain, disability, and death are evils because anyone who fails to avoid them without an adequate reason is, to that extent, irrational.
Of course, we often do have an adequate reason to accept pain, disability, and even death. I pay my dentist to drill my teeth in very painful ways. Is this crazy? It would be, if the drilling did not serve any purpose; but a trip to the dentist is not irrational because the dentist prevents more pain in the future.
Some Reasons to Believe that There Is No God 87
When is an evil justified? Only when it is necessary to gain some good or to prevent some evil. If I could prevent tooth decay with an innocuous pill rather than by paying a dentist to drill my teeth, it would normally be stupid to choose the drill over the pill. Moreover, not every reason is adequate. Even though leg cramps are painful, and even if there is no other way to prevent periodic leg cramps except by cutting off someone’s legs, the pain of leg cramps is still not an adequate reason to cut off that person’s legs. There are disagreements about which reasons are adequate, but some cases are clear, and everyone recognizes some reasons as inadequate. In any case, evils can be justified only by benefits that are adequate to compensate for those evils. This is just common sense about what is rational.
Morality enters the story when evil is caused not just to oneself but to other people. (Morality might also restrict some harms to oneself and to lower animals, but we can avoid those controversial issues here.) It is morally wrong to cause evils to other people, although again we have to add “unless one has an adequate reason.”
Morally, we also should not fail to prevent important harms to others when we can easily prevent those harms at little or no cost to anyone, including ourselves. A common example is a baby who crawls into a pool and starts to drown. If I can save this baby’s life at no cost other than my getting wet, then it would be immoral for me to walk away and let it drown. Maybe I shouldn’t save it when somehow I know that it will grow up to become a mass murderer.
Maybe I don’t have to save it when other people can and will save it if I don’t. Maybe I have an excuse when saving it would be dangerous or when there are too many drowning babies for me to save them all. Moralists disagree about how far anyone needs to go to prevent harm to others. It is not controversial, however, that we should prevent serious harm when nobody else can or will prevent it and we can easily and safely prevent it and similar harms with no significant cost to anyone.
Controversy arises when there are significant costs in preventing an evil or gaining a benefit. Still, a few general points seem clear.
As in the case of dentists, to be adequate as compensation for an evil, the compensating benefit must be important enough, and there must be no better way to prevent the evil. When other people are
God?
affected, it also must not be unfair to cause or allow the evil in order to gain the benefit. For example, a doctor would not be justified in taking one person’s liver without consent to transplant into another person, even if that other person would die without the liver.
The death of the unwilling organ donor is an evil, and the other person’s life is an important compensation. Still, this compensation is not adequate morally, because it is unfair to sacrifice one person to save another without consent (other things being equal). This restriction on fairness in compensation shows, again, that I do not assume any crude utilitarianism.
This little bit of moral theory is common sense that can be accepted by those who believe in God, as well as those who do not.
This moral common ground is worth highlighting here, because atheists are often accused of being amoral or immoral. This accusation has no basis. Atheists can, and often do, accept very stringent moral principles that restrict not only causing harm but also failing to prevent harm. The moral principles of atheists need not be any different from the moral principles of theists (except on religious issues, such as blasphemy).
The problem of evil arises when we apply these common moral principles to God. God is supposed to be more skilled than any human. God can save drowning babies without even getting wet. God can prevent tooth decay with no pain at all. It might seem that even God cannot do this, because tooth decay results from laws of nature.
But God is supposed to be able to perform miracles. God can intervene in nature and even change the laws of nature. This follows from the definition of God as all-powerful in the sense that He can do anything that is logically possible. The only evils that such a God cannot avoid without loss are those that it is logically impossible to avoid without loss. Consequently, the only evils that God is justified in allowing are those that are logically necessary to promote some adequately compensating good. That explains why I used the term
“logically necessary” in my premises.
As I said, my argument is very old. Over the centuries, many Christians have responded to it, but no response is adequate, or so I will try to show. I cannot consider every possible response, but I will quickly survey the most prominent responses. The first two responses question premise (2). The next six responses deny premise
Some Reasons to Believe that There Is No God 89
(3). The last three responses raise broader issues about method.
Some of these responses are rarely used today, but they are still worth running through to cover the territory.
The Unconscious Response: Maybe there is no evil in the world, since, for example, the babies who seem to suffer might not be conscious, in which case they never really feel any pain.
Criticism: Nobody could believe this while they watch these poor babies squirm. It is possible that God anesthetizes such babies and then makes them squirm as a spectacle for us. But why would God do that? Moreover, anesthesia cannot remove the evils of disability and death. It is still logically possible that no evils occur, since even disability and death might be illusions, but that abstract possibility cannot keep our experiences from being strong evidence that there is much pain and disability and death in this world.
The Privation Response: Although pain, disability, and death occur, evil does not exist in reality, because evil is merely the privation of good, just as darkness is the absence of light.
Criticism: Anybody who has ever stubbed a toe or given birth knows that pain hurts. It is a positive sensation and not just the absence of pleasure. People who are numb or asleep do not feel pain, even though they experience the absence of pleasure and any other feeling. So pain is not a privation. More importantly, even if pain, disability, and death were all just privations, that would not make them any less evil or any less real. It would be a sick joke to try to comfort parents whose child just died by saying that their child is not really dead but merely deprived of life. Whether or not evils are privations, they should be avoided when there is no compensating good, so no metaphysical doctrine about privation can explain why God would let them occur.
The Contrast Response: If the whole world were black, we would not be aware that it is black. For that awareness, we need some other color to contrast with the black. Analogously, if there were no evil in the world, we would not be aware of the good. God then allows evil to make us aware of goodness, since this awareness is itself a good.
God?
Criticism: To make us aware of blackness, all we need is to see a speck of white. To make us aware of goodness, all we need is to see a speck of evil. One stubbed toe is enough, or maybe one stubbed toe per person. There is far too much evil in this world for it all to be necessary to enable us to conceive of goodness. Some theists might claim that we need more evil around us in order to appreciate how good our lives are. Even if so, this is hardly laudable, and it is not logically necessary that our minds work this way, so God could make us better.
The Punishment Response: God lets people suffer because of sins for which their suffering is just retribution.
Criticism: Even babies suffer, but babies have not sinned, since they are not morally responsible yet, so there is nothing to punish them for. That is why I used babies in my examples.
Some theologians respond that God punishes babies for original sin. Adam and Eve sinned in the Garden of Eden. Babies pay now.
But that would be grossly unfair. Long ago, governments did punish whole families and other groups for crimes by one member, so then the punishment response might have seemed plausible by analogy. Today, however, almost everyone agrees that group punishment is barbaric. You should not be punished for what your father or your state’s governor did, especially if they did it before you were born.
Why not? Because their acts were beyond your control, and it is unfair to punish people for what they cannot control. This widely accepted principle of justice would be violated by punishing babies for original sin.
God is supposed to be unlike humans who punish, since God knows the future. Maybe God punishes babies in advance for sins that they will commit if they live. But that still seems grossly unfair.
If people have free will, as most traditional theists assume, that makes it unfair to punish them in advance, while they still have a chance to choose not to sin. Besides, babies with Down’s syndrome are not likely to become such bad sinners that they deserve to die painfully before they sin. It is logically possible that every baby who suffers and dies would have become a mass murderer, but we have no evidence at all for such speculation. It is just as likely that these babies would have become heroes. Our evidence, thus, suggests that
Some Reasons to Believe that There Is No God 91
the evils of this world are not always distributed according to past or future sins, so these evils cannot be justified as punishment.
The Heavenly Response: Even if sinless babies suffer, they are compensated many times over by being given eternal life.
Criticism: God could send the babies straight up to heaven before they suffer. Because this option is available, their suffering is not logically necessary for them to get to heaven, so heaven cannot justify their suffering. To see why, compare a father who beats his daughter and then gives her a gift to make up for it. The beating is still immoral even if the gift is so good that she would rather have both the beating and the gift than neither. Why? Because he could have given her the gift without the beating. That would be better for her than both together and also better than neither. Similarly, if God lets babies suffer and then sends them to heaven for eternity, they are better off than if they had neither the suffering nor heaven.
But that cannot make it right to let them suffer, since God could have sent them straight to heaven without the suffering. He could have caused the mother of the Down’s syndrome child to miscarry, so that the baby would never have been born, would never have suffered, and would have lived only in heaven. Why didn’t God choose that option? Heaven cannot answer this question.
The Virtuous Response: Evil builds good character. The child’s suffering enables the parents to become more courageous and observers to become more compassionate. These and other moral virtues compensate for the evil.
Criticism: Education should be effective, efficient, and fair. Letting babies suffer and die is often a very ineffective way to teach courage and compassion. For one thing, suffering and death sometimes occur in secret. If nobody lives to tell the story about a suffering baby, nobody can learn from it. Even when evils are known, many observers become hardened and cynical, or despondent and isolated, rather than morally virtuous. And even when evil does teach some courage and compassion, greater losses often occur. The child’s parents break up, neglect their other children, and turn away from God. If a schoolteacher used techniques this harmful, you would immediately fire that teacher. And God would know exactly which peo-
God?
ple would react well to evils. So it is hard to imagine any reason why God would adopt such a crude teaching technique in the cases where He knows that it won’t be effective.
Moreover, when a lesson is expensive, it is inefficient to repeat it more often than is necessary. Surely God can create moral virtues with less harms than we experience. One starving or deformed child should be enough to create compassion, if the news spreads. We don’t need millions.
Most importantly, even if the parents or others did become better off as a result of the baby’s suffering, it would still be unfair to let the baby suffer in order to help those separate individuals. Compare a neighbor whose baby is suffering horribly from disease. The neighbor can help easily, but the suffering of this baby will teach a valuable lesson to his other children and to himself, so he lets the suffering continue unabated. This neighbor is a moral monster. Don’t you agree? But this neighbor is analogous to God if God lets babies suffer just to build moral virtues in other people, especially since God has more and better options than any human neighbor has.
The Glorious Response: Craig sometimes says that God uses evil to maximize the number of people who know, glorify, and gain salvation from God. God’s purpose is not to make humans happy but, rather, to make them glorify Him.
Criticism: This would be horribly unfair to those babies. If a father told us that he doesn’t care whether his children are happy as long as they know how powerful he is and glorify him, we would think that this father is an egomaniac. God might have purposes other than making us happy, and one of his purposes might be to make us glorify Him, but our happiness can still be among His purposes. If God doesn’t care at all about whether we are happy, then He is not good by our standards or in any sense that matters to us.
Moreover, an all-powerful God can surely find some more effective and efficient way to lead people to glorify Him, since evil so often turns people away from God.
The Free Will Response: You wouldn’t want a robot for a child, even if the robot always obeyed you, right? So you must admit that freedom is worth the price of much misbehavior. This value also jus-
Some Reasons to Believe that There Is No God 93
tifies God in giving us free will, even though He knew that we would often misuse our free will and choose or cause evil.
Criticism: I do not deny that people have free will, although it is surprisingly hard to say what free will is. I also do not deny that free will is very valuable, although it is again hard to say why. Of course, I would never trade my children for robots, even though they do (rarely!) misbehave. What I do deny is that all evil can be justified by the value of freedom or free will.
Although I want my son to have free will, I would stop him from killing his sister. If I didn’t, then I would be guilty of a serious crime.
So why doesn’t God stop people from killing each other? At least, why doesn’t he stop most of them? There is way too much killing in the world to think that an all-good God could stop it but chooses to sit idly by just in order to protect the free will of murderers.
Moreover, many murderers can be stopped without robbing them of their free will at all. Consider a pedophile who contacts young children on the Internet, talks them into meeting him, then kidnaps them, rapes them, and kills them. Let’s call one victim Mary. What would God have to do to save Mary? Just make this pedophile’s com-puter crash right before he contacts Mary. The pedophile would still have free will, but he would not cause as much pain and suffering.
So why doesn’t God foil evil plans or turn evil people into klutzes?
Finally, even if the value of free will did explain why God allows evil that is caused by humans, it still would not explain why God allows natural evil. Natural evil is evil that arises independently of human actions. It includes most of the pain, anguish, and loss of life that is caused by lightning and earthquakes and diseases and so on.
The free will defense does not apply to natural evil. Diseases, birth defects, tornadoes, and earthquakes do not occur because of free will, so God could prevent them without limiting our free will at all.
That is why I used natural evils as my examples. The value of free will cannot compensate for the suffering and death of babies with intestinal blockage, since their birth defects have nothing to do with free will.
Free will does show something about the problem of evil. It is logically possible that every bit of evil in the world results from an act of free will, so there is no truly natural evil. Even lightning and birth defects could be caused by free choices of invisible spirits. This
God?
possibility is supposed to show that any evil is logically consistent with an all-good and all-powerful God. I am not sure whether these propositions really are consistent, since an all-powerful God might be able to stop such invisible spirits without depriving them of free will (as in the case of the pedophile above). Luckily, I need not settle this issue. My argument is not about logical consistency. It is about evidence. We have no reason whatsoever to believe in invisible spirits that cause evils that appear to be natural. In the absence of any such reason, we have evidence that these evils are natural, at least after we look hard but find no connection to anyone’s free will.
This evidence then gives us adequate reason to believe that the value of free will cannot justify God in allowing these evils.
The Standard Response: God’s goodness is different from human goodness. He is good by His own standards, even though He allows evil with no compensating good.
Criticism: This response admits that God is not good by our common moral standards. This is hardly welcome news for anyone who wants a model to guide his life. It is hardly a God worth worship-ping, except out of fear. It is not a God we could love or admire.
We don’t have any idea what to expect from such a God. In the future, will He allow or cause even more evil without compensation?
Will this show that He is even better than before? Nobody can tell, if we cannot understand the standards by which God is deemed good.
Moreover, this response could justify atrocities. Imagine a human tyrant who kills and rapes or, at least, lets babies die painfully when he could save them easily. His followers claim that he is still a good person, because he has a divine right to rule without being subject to common moral standards. Whatever they say, however, we have plenty of reason to believe that this tyrant is not a good person. We should oppose him. God is no better if He lets babies die painfully for no adequate reason.
Some religious believers counter that, unlike a tyrant, God is not subject to our moral standards because He created us, so we owe Him everything. But that does not give God the right to abuse or neglect us any more than it gives human parents the right to abuse and neglect their children. Others say that our moral standards do not apply to God because He is so powerful. But a tyrant who is so
Some Reasons to Believe that There Is No God 95
powerful that nobody can stop him can still be judged to be morally bad. Might does not make right. This leaves no good reason to exempt God from moral judgment by our own standards.
The Modest Response: God has adequate reason to allow each bit of evil, but we cannot always see His reason, for we are just measly little humans.
Criticism: Anyone can try to defend any action, no matter how horrible, by saying that it is justified by some reason that we do not see. Suppose your neighbor lets his child suffer and starve. He has plenty of food and money, and his child seems normal, so you do not see anything to justify his neglect. Then his friend says, “Maybe he is a good person, and he has an adequate reason, but we just don’t see it. Maybe this child’s death will turn out for the best.” You wouldn’t buy this for a minute. You might give your neighbor a chance to defend his actions. However, if he gives no adequate reason when he could and would give one if he had one, then you would be justified in concluding that he has no adequate reason. Your conclusion would become even more secure if your neighbor let several of his children die in various horrible ways for no apparent reason. Similarly, if we look long and hard at a natural evil, such as an intestinal blockage, and we find nothing to suggest any adequate compensation, then we are justified in believing that there is no adequate compensation for that evil. This conclusion becomes more secure when the same pattern gets repeated in case after case after case.
We cannot be sure. It is always possible that some adequate compensation will arise centuries later by some obscure causal process of which we are now ignorant. However, we should not let uncertainty stop us from judging and acting when we have strong evidence. It is possible that our neighbor has his reasons for letting his children starve and that the children’s deaths will work out for the best in the end, but we still may and should criticize and stop him.
If we were not allowed to reach any conclusion without being completely sure, then we would never be allowed to reach any conclusion on any important matter, since we can never be completely sure about anything important (at least if it is controversial). The demand for certainty leads to ignorance and inaction.
God?
Admittedly, we often lack strong evidence. Before we ask our neighbor what is going on, we might not have enough evidence to reach any conclusion, since most people know their own children and their own situation better than their nosy neighbors do. That is why we need to investigate carefully before we judge. However, after thorough investigation to the best of our abilities, if our neighbor reveals nothing and we see nothing that comes close to an adequate reason or suggests any promising avenue for further investigation, then we are justified in concluding that our neighbor has no adequate reason to let his children suffer and die. Analogously, if we look long and hard at a particular case of evil, but God reveals nothing and we see nothing that could compensate for that evil, and nothing that tells us where to look further, then it is reasonable to conclude that there actually will be no adequate compensation for this evil. Our justification becomes even stronger as we accumulate more and more evidence from more and more cases of natural evil.
This conclusion might seem rash if we would not be able to fore-see an adequate compensation even if there were one. But why think that? The idea might be that compensation could come far in the future. That is possible, 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. These speculations cancel each other, if there is any force to cancel in the first place. The idea might be that heaven awaits the suffering child, but I already criticized that response on the grounds that God could send the child straight to heaven without suffering. Add my criticisms of other responses, and then there is no plausible story about any adequate compensation for which we would not have evidence even if it were forthcoming.
Besides, why wouldn’t God give us more evidence than we have on this matter? Why would He hide His light under a bushel? If He exists, He could easily show us His reasons for allowing evil. Our ignorance about His reasons is another form of evil, since it is a disability and causes distress to many people. So God would give us better evidence of adequate compensation for evil, if there were any.
It is still possible that God has a hidden reason. You might assume this if you already believe in God, but that assumption begs the question here. Just try to look at all of the evils in the world as you would if you were not already committed for or against the ex-
Some Reasons to Believe that There Is No God 97
istence of God. Any neutral survey of the evidence is then bound to point away from the existence of God.
The Overriding Response: If there is a God, there must always be an adequate reason for evil, even if we cannot see it. Therefore, every argument for the existence of God is an argument that all evil is somehow compensated.
Criticism: Suppose you see your neighbor beat his child, then I claim that you must be hallucinating, since I have arguments to show that your neighbor is a good person who wouldn’t beat his child.
That would not convince you for a second unless my arguments were very strong. Similarly, the overriding response works only if there are very strong arguments for the existence of a traditional God. But there are none. This claim was supported by my criticisms of Craig’s arguments in the first part of this book. It is also recognized by many Christians who espouse justification by faith rather than by reason or by acts.
Notice how much an argument would need to show to refute my argument from evil. It would not be enough to prove that the universe was created by a very powerful being. A refutation of my argument would have to show the existence of a God who is both all-good and all-powerful, since a limited creator is compatible with my argument. None of Craig’s arguments comes close to showing that anything is both all-powerful and all-good. So none of Craig’s arguments for the existence of God could undermine my argument against the existence of God.
Other arguments might be constructed, so I cannot be sure that no future argument will ever support the overriding response. All I (or anyone) can do is inspect each new argument when it is presented. Still, it is hard to imagine that any new argument for the existence of God will come close to proving the existence of an all-good and all-powerful God. Any argument that has eluded so many smart people for so long will not be very obvious, so it probably won’t provide anything like the overwhelming amount of evidence that we have for the existence of so much natural evil with no compensating good. If no new argument is stronger than the evidence of evil, then no new argument for the existence of God can undermine my argument from evil against the existence of God.
God?
Conclusion: Theists have a choice. They can say that God is not all-powerful, or they can say that God is not all-good. What they cannot do is face the evidence of evil in this world and still believe in the traditional Christian God who is both all-powerful and all-good.
That is the only kind of God that we are arguing about here. So my argument shows that there is no God of the disputed kind.
2. The Problem of Action
In addition to being all-good and all-powerful, the traditional Christian God is also defined as eternal and active in time. These additional properties lead to new problems.
To say that God is eternal is to say that He exists outside of time.
This kind of existence is mysterious, but it can be explained by analogy with numbers. The number three is what the numeral “3” refers to. A numeral is a symbol that can be written down at a particular time and place. In contrast, a number is not the kind of thing that can ever be written down.
To see how peculiar numbers are, consider how you would react if anyone said that the number three exists in North America but not in Antarctica. Antarctica lacks many things that we take for granted in North America, but not numbers. You know this without looking around Antarctica. What would you look for? This point about space applies also to time. If anyone said that, although the number three exists today, it did not exist in the days of dinosaurs, then you would sneer without checking the fossil record. Again, what would you check for? This shows that, if numbers exist, they do not exist at some times and places as opposed to others. Moreover, numbers exist or fail to exist equally at all times and places. A number cannot get bigger or smaller or change in any way.
These points about numbers are not controversial, but they are important here in two ways. First, atheists are sometimes accused of denying the existence of anything beyond the purely physical world. Some atheists do go that far, but other atheists admit the existence of non-physical minds and even abstract objects, including numbers. All that atheists have to deny is the existence of one kind of non-physical object, God. The more important lesson is that, if numbers exist outside of time, then they are eternal in the same way
Some Reasons to Believe that There Is No God 99
as God. This makes it fair to use numbers to determine what follows from the doctrine that God is eternal.
If the number three exists outside of time, this keeps it from causing any changes inside time. If you have triplets, that might make you very busy, but the number three itself cannot be what makes you busy. Why not? Because the number three is exactly the same when you are busy as when you are not. The number three was no different before, after, or as your triplets were born, so the number itself cannot explain why you became busy at the time when they were born.
More generally, the cause of any event must bear the right relation in time to that event. In a baseball game, if I swing at a pitch too soon, I will miss. If I swing too late, I will miss. The only way to hit a ball is to swing at just the right time (and in just the right place).
If the number three exists outside of time, it cannot hit a ball or cause any event within time. Slogan: If you are not in time, you are inert.
The same restriction applies for the same reasons to any God who is outside of time. Such a God cannot change or exert force at just the right time, so He cannot cause changes within time. This result has been overlooked by many traditional Christian theologians, but it does follow from their doctrine of eternity.
The problem is that the traditional Christian God is also supposed to be active in time. He is supposed to have created the world, to cause miracles, to answer prayers, to be incarnated, and so on. None of this makes sense if God really is eternal and, hence, outside of time. If God has one of these traditional features, then He cannot have the other. Consequently, there cannot be a God who is both eternal and also active in time.
Theists sometimes respond by giving up the traditional doctrine that God exists outside of time. They claim instead that God is present equally, wholly, and indivisibly in each moment of time. This omnitemporal kind of God is not my target here, but it is worth showing why this move does not help anyway. If God makes the Red Sea part at noon, and if God is equally, wholly, and indivisibly present at midnight, then God would also have caused the Red Sea to part at midnight. Otherwise, His presence could not be sufficient to
God?
cause the Red Sea to part at noon. More generally, 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. That is why God still cannot cause changes in time if He is equally, wholly, and indivisibly present at each moment of time.
Another possible response is that laws of nature can be causes.
The law of gravity, for example, exists either outside time or at all times. Yet this law might seem to cause balls to fall. This is inaccurate. A ball falls at a particular time and place because the earth attracts it at that particular time and place. The ball would not move in the same way if the earth were not there then, even though the law of gravity would still hold without the earth. Thus, it is not the law of gravity that causes the ball to fall. It is, instead, a particular gravitational attraction in particular circumstances that causes the ball to fall when, where, and how it does. Diffuse gravitational fields and waves cause other effects, but all of these causes exist in time and have properties that vary from one place to another, unlike God.
Gravity in general still might seem to be a causal factor instead of a cause. Just as a match will not light without (some) oxygen, so the ball will not fall without (some) gravity. Such background conditions might seem analogous to an eternal or omnitemporal God, if events in this world would not occur without God. This move, however, takes us far away from any traditional notion of God. Neither gravity in general nor the law of gravity can create worlds, cause miracles, answer prayers, make choices, lay down laws, or send sons to save us.
It would make no sense to worship gravity or the law of gravity. So this analogy cannot salvage anything like the traditional notion of God.
In a final analogy, some theists say that God exists in a dimension outside of time and space, but He can still have effects inside time and space just as a three-dimensional pencil can cause marks on two-dimensional paper. (Of course, paper has more than two dimensions, but let’s play along.) Still, the pencil must touch the paper at one point but not another to make a mark at that one point but not the other, so the pencil must be in some parts but not others of the paper’s two dimensions, even if it exists in a third dimension as well.
Similarly, even if God did exist in some dimension other than time and space, He would also have to exist in some parts of time and space but not others (or He would have to exist in different ways in