[This article was originally published in Philo 1:1]
Arguing for Atheism: An Introduction to the Philosophy of Religion by Robin Le Poidevin (Routledge, 1996, ISBN 0-415-09338-4), 159 pp., $15.95 paper
With respect to traditional religious doctrines, philosophers tend to be a skeptical lot. It is safe to say that the proportion of atheists and agnostics among professional philosophers is far higher than in the general population (though perhaps not as high as in some other professional groups). However, when one surveys recent work in the philosophy of religion, the overwhelming preponderance of such work has been done by theists.
With certain notable exceptions, atheists have been largely quiescent. Perhaps their ennui is understandable. Many atheists seem to feel that the job of rebutting theism is done and that further such efforts would be an exercise in slaying the slain. I regard this as a most unfortunate attitude. Due to the work of a number of exceptionally qualified theistic philosophers, the defense of theism has taken a number of interesting turns in recent years. I believe that these arguments merit serious critical evaluation.
That atheist philosophers still have interesting things to say is demonstrated by Robin Le Poidevin in his recent book Arguing for Atheism. Le Poidevin is a young (born 1962) but already highly accomplished philosopher. His book is a logically deft and clearly written introduction to the philosophy of religion. It should be useful for undergraduate courses, though parts, such as the discussion of the modal ontological argument, are quite complex and certain to confuse beginners. The book is also a brief for atheism. In general, it serves both of its functions well. However, the three parts of the book are unequal in value. I found part 3, in which Le Poidevin examines the possibility of religion without God, to be of less interest than the earlier sections. Further, though I regard part 1, “The Limits of Theistic Explanation,” as a nearly complete success, I have some reservations about the treatment of the problem of evil in part 2.
In part 1 Le Poidevin examines the “big three” arguments for theism — the cosmological, ontological, and design arguments. To his credit, unlike many authors of introductory-level texts, he does not rehash hackneyed issues. His treatment is fresh and insightful and incorporates recent arguments of the most prominent current theistic philosophers such as William Lane Craig, Alvin Plantinga, and Richard Swinburne. Le Poidevin realizes that these arguments, though often very conservative in content, are defended with a rigor and sophistication that demands critical attention, not bored dismissal. Le Poidevin’s criticisms of these arguments are powerful, perhaps even decisive.
Le Poidevin’s first chapter “Must the Universe Have a Cause?” is one of the best in the book. He considers two forms of the cosmological argument, the “temporal” form, recently defended by Craig, and the “modal” form, defended by various theists back to Thomas Aquinas. The “temporal” form is given as follows:
1. Everything that begins to exist has a cause of its existence.
2. Nothing can be the cause of its own existence.
3. The universe began to exist.
Therefore: The Universe has a cause of its existence that lies outside the universe.
Le Poidevin correctly points out that the third premise is false if the universe is infinitely old or if time exists as a closed loop with each moment both preceding and following each other moment. Craig has deployed Arabic Kalam arguments to claim that an infinitely old universe is logically impossible since, he contends, no actual infinite can be traversed. For the sake of argument, Le Poidevin concedes the third premise and supposes that the universe had a beginning in time.
There are three possibilities if the universe had a beginning in time: Either time begins with the universe, or time exists for a finite period before the universe, or time exists for an infinite period before the beginning of the universe. In the first case, where time begins with the universe, Le Poidevin says it is senseless to speak of the universe as having a cause since the very notion of “cause” involves temporal priority. This might be disputed; cannot a cause be simultaneous with its effect?
I think Le Poidevin’s arguments could easily be extended to counter this objection. If God’s creative act coincided in time with the beginning of the universe, then that creative act began to exist, and according to the first premise of the temporal cosmological argument, it must have a cause. That cause must also begin to exist, and so an infinite regress threatens. Therefore, if time begins with the universe, to say that God created time is either incoherent (if causes must precede effects) or leads to an infinite regress (if causes can be simultaneous with effects). If we exempt God’s creative acts from the requirement that whatever begins to exist must have a cause, why not exempt the beginning of the space/time universe also?
Suppose, then, that God starts time before the universe. Perhaps he has a countdown: “five, four, three, two, one … FIAT LUX!” In this case, the beginning of time would be an event in the mind of God — the start of the countdown. But here again, if God’s creative act in starting the countdown is the beginning of time, that act begins to exist, and the temporal cosmological argument requires that that act also have a cause. Once again the infinite regress abyss yawns.
Defenders of Kalam arguments, such as Craig, deny that an actual infinite can be crossed, so they must deny the third possibility — that time existed for an infinite period before the universe. However, if it is admitted that infinite time could have existed prior to the universe, we have to ask why the universe began when it did rather than sooner or later. The answer that God willed it to begin just before it did will not do. According to the first premise of the argument, that act of divine will must have had a cause, and, since we are supposing time to stretch back infinitely, that cause must have had a cause and so on ad infinitum. Yet it is precisely such infinite causal regresses that cosmological arguments are intended to avoid.
Le Poidevin therefore shows that if we hold that the universe had a beginning in time, as the temporal cosmological argument requires, we have excellent reasons for rejecting the first premise of that argument — that everything that begins to exist has a cause of its existence. The first two options, (a) that time begins with the universe and (b) that it exists finitely before the universe, require that time itself be caused, which is problematic whether that cause is conceived as preceding or coinciding with the beginning of time. The third and final option, that time existed infinitely before the universe, leaves us with the choice of an infinite regress of causes or the admission that something can come into existence uncaused.
A theist might respond to Le Poidevin’s trilemma by holding that God (a) creates timelessly, (b) creates in a higher time dimension, or (c) employs some form of “backwards” causation to create after the universe has begun. Le Poidevin does not consider any of these possible responses. I suspect that he feels that when we begin to speak of “timeless causation” or “backwards causation” we are in a metaphysical and semantic Twilight Zone where such notions as “causation” break down utterly. I agree, but some theists seem to think there is something to these concepts, so a more detailed consideration of their views might be in order. Perhaps Le Poidevin should note that even if it somehow makes sense to say that the universe was timelessly caused, that timeless cause might be something other than the theistic God — a Platonic Idea, perhaps.
Le Poidevin’s treatment of the other theistic arguments is equally acute, though, as noted earlier, occasionally too dense for beginners. His critique of the so-called fine tuning argument is particularly satisfying. Recently, philosophers such as John Leslie have striven to revamp the design argument by focusing on the apparent “fine tuning” of the fundamental physical constants. For instance, had the gravitational constant been ever so slightly stronger, the big bang would have been soon followed by a big crunch, producing a universe-mass black hole. If the constant had been just slightly weaker, the universe would now consist of a completely uninteresting diffuse gas. To produce an interesting universe with intelligent creatures such as ourselves, the gravitational constant had to be just right. That so many of the constants of nature are just right for the development of intelligent life-forms is for some evidence that those constants were prearranged by an intelligent designer who valued intelligent life. Of all the possible values of these constants, they ask, why should the ones just right for us to exist have emerged in the actual universe?
Le Poidevin points out that this argument rests upon a misapplication of the concept of probability. In ordinary contexts where probability calculations are useful, indeed indispensable, the assessment of probabilities depends crucially upon background information. For instance, the meteorological estimate of a 50 percent chance of rain tomorrow rests upon background knowledge of weather patterns. Against what background, Le Poidevin asks, can we judge, e.g., that it was extremely improbable that the charge on the proton would be 1.602 X 10-19 coulomb? The laws of physics cannot constitute the background since they will either be irrelevant to the charge on the proton or will entail precisely the charge it has.
Here the fine tuners reply that the issue is simply that when we imagine the whole set of possible universes, the vast majority will not contain physical laws consistent with the development of intelligent life. Surely, they say, it is much more probable that the one actual universe would fall into the vast majority of possible universes that are hostile to life.
Le Poidevin contends that to cash out these intuitions the fine tuners must invoke some usable concept of probability, but it is very difficult to see what that concept could be. Statistical probability or chance would seem to be the most applicable, but what exactly could “chance” mean in such a context and how could it be assessed? Surely, as Hume pointed out long ago, we cannot estimate the chances for the actualization of some type of possible universe by observing how frequently certain sorts of possible universes do become actual. We cannot observe the actualization of possible universes; indeed, since possible universes are imaginary, they are not there to be observed (Le Poidevin is aware that modal realists hold that all possible universes are actual, but he argues that modal realism itself creates intractable problems for theistic arguments). Le Poidevin further argues that other construals of probability are equally inapplicable to the fine-tuning argument. Consider the view that probability is a measure of propensity. How could one argue that some kind of possible universe has a greater propensity to be actualized than another?
In part 2 Le Poidevin turns to moral arguments for atheism. Much of the discussion of the problem of evil has dealt with the question of whether it is rational for theists to believe in God given the acknowledged extent, magnitude, and apparent gratuity of so much evil in the world. Theists have traditionally responded to the problem of evil by offering a theodicy — an attempt to justify God’s permission of evil by showing how evil leads to a higher good. Perhaps the most popular of such theodicies has argued that God permits moral evil because it is better for humans to be endowed with free will, and hence be capable of moral evil, than for humans to have been made as perfect but unfree automata. Le Poidevin competently guides the reader through the complexities of the issues of freedom and determinism and argues that no conception of human free will can support a viable theodicy.
Lately some theists, such as Plantinga, have abandoned the attempt to provide theodicies and instead offer a more modest defense. Plantinga claims that it is rational for theists to believe in God and to acknowledge the full scope and depth of evil even if they cannot offer an adequate theodicy. He argues that there is no inconsistency in holding that God exists and that evil exists, not even when the evil that exists is all the evil in the actual world. Further, he claims that probabilistic versions of the problem of evil fail to show that God’s existence is unlikely given the facts of evil.
Plantinga therefore claims that theists violate no epistemic duties, and hence are not irrational, in adhering to belief in God even though they acknowledge the facts of evil and cannot explain why God permits evil. It follows that theists are rational in believing that evil is justified even if they cannot explain how it is justifted. Le Poidevin disagrees: “The question is whether we can be justified in assuming that there is such a justification. But we can only be justified in making this assumption if we are justified in our belief that there does exist a benevolent deity. But this is precisely what the problem of suffering casts into doubt. If theism is to be a rationally defensible belief, this position is simply unintelligible” (p. 104).
Le Poidevin seems to ignore the possibility that theists could have independent grounds for claiming that a good God exists. Perhaps they have no idea why God permitted Auschwitz; maybe they even admit that evil does cast doubt on God’s existence. Nevertheless, they maintain that they have other grounds sufficient to rationally convince them that God does exist and is good after all (and hence that evil must somehow be justified). Perhaps they have had overwhelming experiences in which they felt aware of God’s love and compassion. Some people have had these experiences when they themselves have suffered great evil. As an atheist I place a psychological interpretation on such phenomena, but I do not think it is irrational for people to base their belief in a good God on such experiences. I do not see that such people necessarily violate any epistemic duties in doing so. If Le Poidevin does think that they violate such duties, he needs to give more argument to that effect.
The real function of the problem of evil is not to convict theists of irrationality. The real problem of evil is that any hypothesis that seeks to give a theistic explanation of the world must face innumerable apparently disconfirming facts — the presence of a multitude of seemingly gratuitous evils. That is, any apologetic that treats theism as a well-confirmed hypothesis must deal somehow with this plethora of apparently disconfirming data. Since many of the leading theistic philosophers (e.g., Craig, Swinburne, J.P. Moreland, Hugo Meynell, George Schlesinger) do defend theism as an explanatory hypothesis, the problem of evil remains a potent challenge to such efforts. Le Poidevin endorses this hypothesis-disconfirming function of the problem of evil (pp. 102-103), but he does so almost as an afterthought.
As I indicated earlier, the final section of the book is for me the least interesting of its three parts. The eighth chapter, “Is God a Fiction?”, examines the efforts of radical theologians to develop an instrumentalist, i.e., nonrealist view of God. According to theological instrumentalism, discourse about God, heaven, bell, etc., is purely fictional. However, such fictions are viewed as especially potent symbols. As Le Poidevin puts it: “By having an image of the goodness of God before us, we will be encouraged to lead a less selfish, and therefore more fulfilling, life. The idea of God rather than God himself, is thus an instrument through which good can be realised” (pp. 111-12).
In other words, just as reading stories of the Greek gods and heroes can be morally edifying, so the Judaeo-Christian scriptures contain a mythos that addresses perennial human concerns. Some thinkers such as Rollo May and Joseph Campbell have stressed the importance of myth for contemporary humans. Further, there is no question that many biblical stories teach great moral lessons. Christian Coalition-types might lose a bit of their self-righteousness when they recall the words of Jesus concerning the woman taken in adultery: “Let he who is sinless cast the first stone.” When Nathan confronts David over the murder of Uriah and adultery with Bathsheba, he utters an eternal rebuke to hypocrisy (2 Samuel 12: 7).
However, Le Poidevin is not merely arguing that the Bible should be read like the Iliad and Shakespeare. He defends continued religious practice by religious instrumentalists. That is, he argues that it is perfectly legitimate and reasonable for antirealists to continue to, e.g., pray, attend mass, or sing hymns. Of course, for religious antirealists these activities are a form of make-believe; by playing this game they hope to become more moral, less materialistic, and more “spiritual,” etc.
Are there any problems or dangers in thus playing a sort of spiritual Dungeons and Dragons? Le Poidevin effectively rebuts a number of objections to antirealist religiosity. I wonder, though, if such a practice does not often reflect a nostalgia for the meaty certainties of orthodox religion. Anyone guilty of such nostalgia for old-time religion should be sentenced to intensive rereading of Tom Paine, Voltaire, and Ingersoll. A cold slap from The Age of Reason would be in order: “Whenever we read the obscene stories, the voluptuous debaucheries, the cruel and torturous executions, the unrelenting vindictiveness, with which more than half the Bible is filled, it would be more consistent that we call it the word of a demon than the Word of God. It is a history of wickedness that has served to corrupt and brutalize mankind; and, for my part, I sincerely detest it as I detest everything that is cruel.”
Perhaps when we stop believing in God it is best to leave religion in the way that the advice columnists tell us to leave a soured romantic relationship: A clean break may be initially painful, but it is healthiest in the long run.
 Here I would point out that the situation is considerably more mathematically complex than these simple intuitions indicate: If there is any variability in the constants that permit life-friendly universes, there will be a nondenumerable infinity of such possible universes. For instance, if in life-friendly possible universes the value of Planck’s constant could range between 6.626001 X 10-31 joule-second and 6.626002 x 10-34 joule-second, there will be a nondenumerable infinity of such universes since there is a nondenumerable infinity of values between 6.626001 and 6.626002. Hence, if there is any variability in the possible values of constants of life-friendly universes, there will be a nondenumerable infinity of such universes. The fine tuners therefore must show that one of the nondenumerably infinite number of possible life-hostile universes is more likely to be actualized than one of the nondenumerably infinite number of possible life-friendly universes.
Keith M. Parsons is Assistant Professor of Philosophy at the University of Houston-Clear Lake.
“Lively Answers to Theists” is copyright © 1998 by Keith M. Parsons. All rights reserved.
The electronic version is copyright © 1998 by Internet Infidels with the written permission of Keith Parsons. All rights reserved.