The New Atheists say belief in God is irrational.1 Many theists have countered that atheism is irrational. Others think that both sides can reasonably disagree, just as rational and well-informed scientists can disagree about string theory or the social factors contributing to secularization.
Can theists and atheists reasonably disagree about the existence of God?
Let’s begin with atheist philosopher William Rowe. After presenting his 1979 argument from evil, Rowe asked:
If one is persuaded by [this] argument for atheism, [then] what position should the atheist take concerning the rationality of the theist’s belief?
Rowe distinguishes three possible positions for the atheist. The ‘unfriendly atheist’ believes that no one is rationally justified in holding theistic belief. The ‘indifferent atheist’ has no belief about whether or not theists are rationally justified in holding theistic belief. Finally, the ‘friendly atheist’ believes that some theists are rationally justified in holding theistic belief, despite the fact that the atheist is rational to reject theism.
Rowe defends two varieties of friendly atheism, which Shane Andre calls ‘special grounds atheism’ (SGA) and ‘paradoxical atheism’ (PA):
SGA is the position (a) that the atheist is in a privileged evidential position vis-à-vis the theist, and (b) that, as a result, the atheist is justified in believing that there is no God, [even while] the theist is justified in believing the opposite. By contrast, PA is the position that, regardless of whether he has special grounds, the atheist is justified in believing that there is no God, [even while] the theist is justified in believing that there is.
Special Grounds Atheism
Andre rejects the reasonableness of SGA “unless the [atheist's] claim to special grounds can be upheld.” After all, it is not likely that the atheist is in a “privileged evidential position” to the theist with regard to evil. The theist is aware that much evil exists, and that much of it appears to be gratuitous. The atheist and the theist have access to the same evidence concerning evil, but the theist interprets it differently than the atheist does. Andre concludes:
Consequently, whereas the atheist sees suffering as evidence for his position, the theist sees it as a phenomenon whose significance is easily misunderstood and can only be fully grasped within the overall context of theistic belief.
Thus, the atheist does not have special grounds for his atheism with regard to evil. Both the atheist and the theist know that much apparently gratuitous evil exists.
Notice that Andre’s objection rests on the assumption that SGA is the position that “the atheist is in a privileged evidential position vis-à-vis the theist with regard to evil.” But these last four words do not appear in Andre’s definition of SGA. Indeed, Andre himself refers to Rowe’s argument as but one contribution among many from non-believers who have tried to show that atheists “speak from a superior evidential position to that of the theist.” And though Rowe’s original article appears to frame friendly atheism solely in the context of his argument from evil (“If one is persuaded by [this] argument for atheism, [then] what position should the atheist take concerning the rationality of the theist’s belief?”), it seems natural to interpret friendly atheism in broader terms, as a position the atheist may take whatever his justification for rejecting theism.
Rowe himself has affirmed that he intended to frame friendly atheism in response to a broader question, namely:
If the atheist’s nonbelief is justified (by whatever atheistic arguments may succeed), then what position should the atheist hold regarding the rationality of the theist?2
In this expanded context, it is easy to think of an example of reasonable SGA. Consider a teenage girl, Beth, who is raised in Alabama. She knows no unbelievers. She reads only Christian magazines, watches only Christian movies, and attends a private Christian school. She seems to have personally experienced God on many occasions, and is unaware of how the brain may produce these effects without divine intervention. Moreover, she has read some popular books arguing for theism, and is unaware of any books arguing for atheism. Surely Beth is rational to believe in God, even if today’s atheist is rational to reject theism on the basis of evidence unavailable to Beth: for example, the long history of natural explanations replacing supernatural ones. So SGA is reasonable.
Here, one might claim that we still do not avoid Andre’s objection, for Beth could see any supposed evidence for atheism as “a phenomenon whose significance is easily misunderstood and can only be fully grasped within the overall context of theistic belief.” In fact, we might expect her to say so with regard to any supposed evidence for atheism presented. If the atheist’s special grounds for nonbelief had been his knowledge of the long history of natural explanations replacing supernatural ones, then Beth, upon being made aware of this evidence, could say that even though lightning and disease and thousands of other phenomena are no longer best explained by the acts of gods and demons, nevertheless the remaining mysteries of the universe may yet be best explained by divine intentions. Or if the atheist’s special grounds for nonbelief had been his knowledge that billions of people disbelieve in the God of classical theism, forming the basis of an argument from unbelief for atheism, then Beth, upon being made aware of this evidence and argument, could say that demons afflict all those who reject theism, and that this explains unbelief without casting doubt on God’s loving nature. And so on. As Andre notes, “large-scale metaphysical theories may be so underdetermined by the facts that an ingenious thinker can, with sufficient patience and determination, weave a ‘likely account’ to preserve any of them from clear cut refutation.” But in this case, the objection to SGA rests on the uninteresting point that it may always be possible to save a theory from “clear cut refutation” if one is willing to interpret the evidence implausibly enough.
(Let us assume the reader does not accept that my examples could justify atheism. But recall, friendly atheism is proposed as a position the atheist may take if the atheist’s nonbelief is justified. It is not hard to generalize what I have said above to other potential justifications for atheism.)
More importantly, this objection ignores what Andre means when he refers to “privileged evidential position” in his definition of SGA. Andre writes:
Let us say that one person A is in a privileged evidential position vis-à-vis another person B if A, in addition to having B’s evidence for believing that P, has evidence not available to B which justifies A, as it would justify B if he had it, in believing that not-p.
Now let us put Beth’s story into Andre’s formula: “The atheist is in a privileged evidential position vis-à-vis Beth if the atheist, in addition to having Beth’s evidence for believing that God exists, has evidence not available to Beth which justifies the atheist, as it would justify Beth if she had it, in believing that God does not exist.”
Is this plausibly correct? It is, for we may easily suppose that the atheist was familiar with Beth’s unsurprising reasons for believing that God exists, and we have already supposed that the atheist has evidence for atheism unavailable to Beth. Finally, this evidence for atheism would also justify Beth, if she had it, in believing that God does not exist. For though the long history of natural explanations replacing supernatural ones does not prove atheism, it would seem to make either the atheist or Beth (if she knew of it) at the very least reasonable to suspect that current mysteries might, like so many past mysteries, turn out to have natural explanations, and that there might therefore be little use for a supernatural explanatory hypothesis. So the story about Beth shows that SGA is, in the expanded context (beyond Rowe’s 1979 argument from evil) for friendly atheism, reasonable.
What about PA? Can the atheist who is rational to reject theism nevertheless believe some theists are rational to accept it?
Rowe illustrates something like PA by imagining two people who add up a list of numbers. The first person adds them by hand and gets sum x. The second person adds them with a calculator and gets sum y. The first person knows the second person’s calculator is broken, but the second person doesn’t know that. So the first person is rational to believe the sum is x, the second person is rational to believe the sum is not-x, and the first person is reasonable to believe the second person is rational to believe the sum is not-x.
But in this illustration, the first person is in a privileged evidential position compared to the second person. The first person knows the calculator is broken while the second person does not. So this is really a case akin to SGA, not PA. Thus, Andre writes, Rowe “fails to establish the general possibility of a position such as PA.”
While I agree that Rowe’s illustration fails, perhaps the possibility of PA can still be established.
First, let us consider whether there can be such a thing as reasonable disagreement. Then we will consider whether there can be reasonable disagreement about the existence of God.
In affirming the possibility of reasonable disagreement, I have in mind something similar to a remark made by Gideon Rosen:
It should be obvious that reasonable people can disagree, even when confronted with a single body of evidence. When a jury or a court is divided in a difficult case, the mere fact of disagreement does not mean that someone is being unreasonable. Paleontologists disagree about what killed the dinosaurs. And while it is possible that most of the parties to this dispute are irrational, this need not be the case. To the contrary, it would appear to be a fact of epistemic life that a careful review of the evidence does not guarantee consensus, even among thoughtful and otherwise rational investigators.3
Now let me be more specific. Erik Baldwin offers the following example of apparently reasonable disagreement in science:
Two physicists consider the same body of evidence and each accepts the standard criteria of scientific theory selection: explanatory power, fecundity, simplicity, scope, empirical fit, elegance or beauty, and the like. Each has a comparable grasp of the same evidential considerations and they agree about what counts as evidence. Each is equally well-trained and equally familiar with the same relevant background information. On the grounds that theory 1 is slightly simpler than theory 2, the first physicist reflectively and conscientiously concludes that the evidence best supports theory 1. The second, on the grounds that theory 2 has a slightly higher degree of empirical adequacy than theory 1, reflectively and conscientiously concludes that the evidence best supports theory 2. Both physicists are equally reasonable despite their [disagreement].4
As Baldwin observes, the history of science is littered with actual cases of apparently reasonable disagreement. Consider the disagreement between Albert Einstein and Niels Bohr about the ramifications of quantum physics. Consider the disagreement between David Sloan Wilson and Richard Dawkins over whether natural selection operates in a mostly group-centric way or a mostly gene-centric way. Consider the disagreement among sociologists of religion about the merits of competing theories of the causes of secularization.
Often, these apparent cases of reasonable disagreement arise when there is little information available with which to compare theoretical models, and when it is difficult to discern by which criteria we should choose a best explanation for observed phenomena. But this is precisely the case with the God hypothesis. God is posited to be – just like dark matter, dark energy, and multiverses – something quite unlike everything of which we have ordinary knowledge. So it may not be surprising that reasonable disagreement would arise over the question of God’s existence.
Consider the following scenario. Two philosophers, Cleanthes and Philo, consider the same body of evidence:
(i) the apparent fine-tuning of universal constants for intelligent life
(ii) the apparent origins of spacetime from a singularity at the Big Bang
(iii) the currently mainstream theories about cosmic inflation and galaxy formation
Each philosopher accepts the standard criteria of theory selection: explanatory power, fecundity, simplicity, explanatory scope, empirical fit, elegance, and the like. Each is equally well-trained in cosmology and astrophysics, and each is equally familiar with the same relevant background information.
Cleanthes proposes T1, that he evidence is best explained by the intentional action of an as yet unknown powerful being. Philo proposes T2, that the evidence is best explained by the mechanical function of an as yet unknown natural process. At this point it may well be unclear which theory has better explanatory power, fecundity, simplicity, and so on. It may also be unclear which explanatory criteria should be given most weight when choosing one theory over another. Here we seem to have an example of reasonable disagreement over the existence of God. Thus, PA can be reasonable.
Recently, this line of argument is how I defended friendly atheism. Now I’m not so sure. The whole field of epistemology of disagreement has exploded around me.
- For example, Sam Harris writes in The End of Faith, page 72, “We have names for people who have many beliefs for which there is no rational justification. When their beliefs are extremely common we call them ‘religious’; otherwise they are likely to be called ‘mad,’ ‘psychotic,’ or ‘delusional.’” [↩]
- Personal communication, March 7, 2010. [↩]
- Gideon Rosen, “Nominalism, Naturalism, Philosophical Relativism” in Philosophical Perspectives 15 (2001), pp. 71-72. [↩]
- Baldwin, Erik. Rationality Pluralism and Tradition-Based Perspectivalism (University of Notre Dame, forthcoming), chapter 1. [↩]