Could Atheism Prove that God does not Exist?

March 20, 2009


The skeptical atheist – the original and genuine atheist – has competition even among atheists. Distinctions between ‘positive’ and ‘negative’ atheism, and between ‘strong’ and ‘weak’ atheism, have appeared in recent literature. Definitions of these types of atheism vary across atheists. What ‘positive’ and ‘strong’ atheists have in common is their view that one is justified in believing that no god exists, and they regard ‘weak’ or ‘negative’ atheists as only holding the lesser view that one is only justified in not believing that any god exists.

What exactly would be the difference between concluding that one should believe that X does not exist, and concluding that one should not believe that X exists? Let’s try it on Santa Claus. Could I say, “I should not believe that Santa exists, but I should not believe that Santa does not exist.” If I can’t bring myself to believe that Santa doesn’t exist, I am admitting that for all I know, Santa might exist. This “weak anti-Claus” stance seems too weak to someone convinced that Santa does not exist, for lots of reasons involving the extreme implausibility that Santa does exist. No one sees Santa, his North Pole hideout hasn’t been discovered, his Christmas eve schedule would violate natural laws, etc. These facts about Santa encourage the “strong anti-Claus” stance, but they can simultaneously encourage the weak anti-Claus stance, too. After all, if Santa is admittedly so mysterious and so unnatural, it is very hard to imagine how to show that he doesn’t exist! In the absence of a definitive proof that this amazing Santa doesn’t exist, the weak anti-Claus stance is the more reasonable alternative. The burden of proof is definitively shifted to the strong anti-Claus position. Returning to god, can the positive/strong atheist conclusively eliminate the possibility that no god exists?

It is, contrary to legend, quite possible to prove a negative. But not all negatives. It all depends on the kind of negative, the non-existence of something, that you aim to prove. When talking about alleged supernatural beings, there aren’t many successful options.

Many atheists believe that there are rational proofs that god does not exist. For example, some atheists are so impressed by the argument from the existence of evil that they conclude that this argument proves that god cannot be omnipotent, omniscient, and benevolent. There are many ways for Christian theology to reply to this argument, and we will cover the ensuing debate in a later blog entry. But suppose, just for a minute, that there really is a perfectly valid argument for that negative conclusion. Well, what does that argument exactly prove? Only one thing: that one specific kind of god cannot exist: a god having omnipotence, omniscience, and benevolence.

Two lessons are learned here. First, the atheist is reminded that there might be other kinds of gods. Second, the theologian is reminded that it is possible, in theory, to prove that some specific gods do not actually exist.

There are two basic ways to design non-existence proofs. The “dialectical non-existence proof” argues that two or more characteristics of a specific god are logically incompatible. On the reasonable assumption that a definition of something having logically incompatible characteristics can only be the definition of a necessarily non-existent entity, dialectical non-existence proofs can prove that specific kinds of gods cannot actually exist. For example, many Christians believe both that god is perfect and that god can suffer along with us. Figuring out how a perfect being can suffer requires some fancy refinements to god to avoid the harsh verdict of a dialectical non-existence proof. And even if these refinements go badly and one characteristic of god must go, theology is often flexibly accommodating to such modifications to its conception of god. Avoiding dialectical non-existence proofs is, from theology’s point of view, just another way for humanity to learn more about god. It also keeps theologians very busy.

The other kind of proof confronts a specific kind of god with the actual existence of something else, where it is necessarily impossible that both can exist together. This “evidential non-existence proof” attempts to demonstrate that some specific god cannot exist if something else (the “disprover”) actually does exist. Of course, this sort of proof works well only if there is conclusive evidence of the actual existence of the disprover. Theologians are attracted towards investigating the validity of a proof’s logical steps, but ordinary believers have a notoriously expeditious way of disposing of the problem, by stubbornly denying the existence of the disprover. Consider the example from the previous paragraph. What sort of evil could disprove the existence of god? There just couldn’t be any! Or, expressed from the theologian’s perspective, what sort of god would permit getting disproven by any actual turn of affairs? Not surprisingly, Christian theology has already carefully insulated god and god’s plan for the universe from any and all possible evidence. What appears to be evil really isn’t; what we must nevertheless declare to be evil (such as the Holocaust) still has some inscrutably divine sanction, for all we know. A debate over god and evil soon sidetracks into a debate over the extent of our knowledge of god. Revising god (well, our conception of god) is endlessly productive and profitable for theology. Here’s another example. Does natural evolution prove that god did not specially create humanity? Well then, god must have designed the natural laws responsible for humanity’s origins. Keeping god out of harm’s way from actual evidence has also helped to keep theologians employed and busy.

The atheist can offer impressive proofs that specific and inflexible gods do not exist. Logic, obvious evidence, and scientific knowledge can rule out a wide variety of gods. Unfortunately, the number of potentially conceivable supernatural entities (some have already been thought of, but most have not) far outruns the number of disprovable gods. But perhaps the intellectual’s gods don’t really count. An atheist could still feel proud that many the gods which have been worshipped by the great mass of humanity have suffered disproof. Nevertheless, that accomplishment, though nobly executed, is hardly the same thing as successfully proving that no god could possibly exist. The human imagination will, in all likelihood, forever outrun reason’s logic or science’s facts.

When an atheist proudly claims that god can be disproven, he overstates the actual achievement, ignores imaginative theology, and encourages religious believers to suppose that the only reasonable atheist is the one who can prove that their god does not exist. This bold tactic unfortunately sets off a philosophical vs. theological arms race which no one can win. Indeed, this strategic race has already begun. The ordinary believer cheers on theologians protecting god from refutation, but the needed theological refinements to god in turn make god more and more mysterious, which in turn forces atheists to design ever-more intricate arguments against god, and when these arguments fall short, the believers rejoice at the atheists’ dismay and congratulate themselves for their blind faith in incomprehensible mystery. Atheism has a poor strategy if it mainly results in the spread of fideism.

To further appreciate the magnitude of the task of proving that no god exists, compare it to the task of proving that no extraterrestial life exists. Would any scientist, no matter how skeptical about alien life she might be, eagerly undertake such a demonstration? With what degree of confidence could a scientist, us
ing only current scientific knowledge, assert that no alien life exists anywhere in the universe? Now, keep in mind that today’s scientists are rightly skeptical about alien life, in the sense that we do not yet have good evidence of alien life. Scientists cannot reasonably assert that alien life exists, even if they suppose that such life has a fair probability of existing somewhere else out in the vast universe. Nor can scientists reasonably assert that alien life does not exist. And scientists cannot even affirm not believing that alien life does not exist. There simply isn’t enough evidence at present for either the ‘weak’ or ‘strong’ position about alien life. The entire dichotomy between ‘weak’ and ‘strong’ stances breaks down completely here, and the same situation holds for the existence of god.

The only useful category remaining is skepticism, pure and simple: all scientists should be skeptical about alien life, and everyone should be skeptical about god. What kind of atheist are you? I recommend answering, "a skeptical atheist."