5 / 06
Image of birds flying. Image of birds flying.

#22 Must the Atheist Be Omniscient?

September 17, 2007

I’ve heard some Christian philosophers explain that atheism is untenable because philosophically/logically one can’t posit an absolute negative. Also, absolute certainity that there is no God means that one would have to be omniscient. How do we respond when someone says the same about the theist? Wouldn’t a theist have to be omniscient to be certain that God exists? Thank you!


Flag of United States. United States

Photo of Dr. Craig.

Dr. craig’s response


I’ve also heard both of those claims batted about in popular apologetical discussions, Annissa, and I find neither of them to be persuasive.

The first claim is, ironically, usually found on the lips of atheists, who thereby seek to excuse themselves from bearing any share of the burden of proof in the discussion. Usually, the claim is that a universal negative cannot be proved, and therefore the claim that “There is no God” is unprovable. The second claim is typically given as the reason why a universal negative cannot be proved: no matter how much knowledge you have acquired, there will always be more facts that you do not yet know, and perhaps the exception is among them. So one can never prove that there is no God. Perversely, this is somehow interpreted, not as an admission that atheism is indefensible, but as a demonstration that it is in no need of defense!

Unfortunately, the argument is misconceived on a couple of counts.

First, negative, universally quantified statements can be proved. We do this all the time. When we make statements about “all” or “none,” we are speaking about what is the case with respect to a certain domain. We are saying that all or none of the members of that domain have or has a certain property. If the domain is not too large, I can confidently make universally quantified affirmative and negative statements. For example, I am quite confident that “No U.S. Senator is a Muslim.” Or again, if I have a typical sample of the domain, I can make inductive inferences on the basis of the evidence from the sample to the whole, even if the whole domain is too large for me to canvass; for example, taking as my domain all the microbes on Earth, I can confidently assert, “No microbes have brains.”

Now someone might say that while it is admittedly true that negative, universal statements can sometimes be proven, still the point remains that in the case of God, the domain is too large and our sample too small to come to any negative conclusion. But those who propound this argument seem to think that the way one determines whether God exists is by taking a sort of universal survey to see if anything answering to the description of God exists somewhere out there. There are, however, other ways of coming to a knowledge of negative, universally quantified statements than doing an inductive survey.

For example, we can have knowledge of negative, universally quantified statements on the basis of things’ essential properties; for example, “No water molecules are composed of CO2.” (Even if something looked and behaved just like water but was made of CO2 , it still would not be water but just a look-alike substance.) Or if we could show that a notion is logically impossible, we would know that it does not exist; for example, “There are no married bachelors.” Significantly, many atheists have tried just this route to proving that God does not exist, arguing that the idea of a being which is all-powerful or all-knowing is logically incoherent.

Atheists will also typically present deductive arguments against God’s existence, which will rule out God’s existence without an inductive survey. For example, the atheist might argue:

1. If God exists, gratuitous (unnecessary, pointless) evil does not exist.
2. Gratuitous evil exists.
3. Therefore, God does not exist.

If the premisses of this argument are true, then it follows that God does not exist.

The point is that it is not at all impossible to prove negative, universally quantified statements, and atheists have historically tried to present non-inductive arguments to show that there is no God.

Second, the statement that “God does not exist” is not a universally quantified statement. When the theist asserts that “God exists,” the word “God” is being used as a proper name, not as a common noun. It is not a statement like “Dogs exist” but rather like “Lassie exists.” In order to prove that God does not exist, one need not prove that there are no gods whatsoever. Our interest is in one specific being, not in all the other beings which may have been imagined or worshipped throughout the world. So the claim that “God does not exist” is really a singular claim, like “Sherlock Holmes does not exist” or “Harry Potter does not exist.” No one thinks that negative, singular claims cannot be proven.

So whether this claim is being made by the atheist trying to shirk the burden of proof or by the Christian apologist intent on showing that atheism is inherently unprovable, the claim is, I think, false. Of course, absolute certainty is not available, but that’s really a red herring, since we have absolute certainty about almost nothing. Demanding absolute certainty will only lead to an unlivable scepticism.

Now as to your question whether this argument backfires on the theist, it seems to me that it does. There might at first seem to be an asymmetry between the theist and the atheist. If one has knowledge that God does exist, then it doesn’t matter what facts lie outside the domain of one’s knowledge—you still know the fact that God exists. But if one has no knowledge of the fact that God exists, then maybe that fact lies outside the domain of one’s knowledge. To illustrate, if you found a gold marble, then you know that gold marbles exist, regardless of what facts lie outside your limited field of knowledge; but if you haven’t found a gold marble, that’s no good reason to think that gold marbles do not exist, for perhaps the evidence for their existence lies outside your limited field of knowledge.

The weakness is this reasoning is that the atheist could have, not merely no knowledge of God’s existence, but positive knowledge of the fact that God does not exist, and then he, too, would know this fact regardless of what other facts lie outside the realm of his knowledge. So neither the theist nor the atheist need be omniscient to know what he claims to know.

Now this response might seem to miss the point. The point, one might say, is that the theist could be justified on the basis of his limited knowledge in believing that God exists, whereas the atheist can never be justified in believing on the basis of the limited evidence that God does not exist.

But, as we have seen, there’s no good reason to think that that is the case. The argument assumes that the way one comes to a knowledge of God is by doing a sort of inductive survey and seeing if God comes up in your dragnet. In fact, the evidential situations of the theist and atheist seem symmetrical. Outside the atheist’s field of knowledge could lie the evidence that God exists; but in the same way outside the theist’s field of knowledge might lie the decisive disproof of God’s existence.

The bottom line is that we have no choice but to go on the basis of the knowledge and evidence that we do have—just we do in all other affairs of life.

- William Lane Craig