Also Available in:

Feedback archiveFeedback 2010

Does God’s foreknowledge entail fatalism?


One atheopath on the CMI Facebook page claimed to have a knock-down argument against the God of the Bible: that God’s omniscience is inconsistent with free choice.

Did God create me knowing I was going to eat a bagel in the morning?

If yes, then I have no choice but to eat the bagel.

If he created me ‘knowing’ I was going to eat a bagel, but then I eat something else instead, then god is not omniscient.

They’re mutually exclusive.

First of all, this is logically back to front. We know that Napoleon lost the battle of Waterloo, but our knowledge didn’t cause Napoleon’s loss. Rather, Napoleon’s loss caused our knowledge. We can apply this to the above: God is the creator of time, thus outside time—so eating the bagel retro-actively caused God’s foreknowledge of this.

There is another fallacy in the argument, involving an ignorance of modal logic. This deals with necessity, possibility, and impossibility:

  • A proposition is necessarily true if it follows from the laws of logic, mathematics, or definition, such as “a brick, if it exists, is either red or non-red in the same time and region”, “2+2=4”, “all fathers are male”. Alternatively, it is logically impossible for a necessary proposition to be false.
  • An impossible proposition cannot be true under any possible circumstances, or logical contradictions: “It is raining and not raining in the same place at the same time”, “2+2=5”, “my father is a woman”. I.e. it is necessarily false.
  • A possible proposition is one that may be true. E.g. “it is raining today”, “x + 2 = 4”, “Bob is a father”, “the sky is blue”. Any of these can be true, but doesn’t have to be because no law of logic would be violated: i.e. it is not logically necessary that the sky is blue. These are also called contingent matters. All necessary propositions are possible propositions, but not vice versa.

Applying modal logic to the above, if eating a bagel is the proposition “p”, and “having no choice” is equivalent to logical necessity, then the argument is essentially:

Premise 1: Necessarily, God knows that I will do p.
Premise 2: Necessarily, if God knows that I will do p, then I will do p.
Conclusion: Necessarily, I will do p (i.e. there was no choice).

But this is logically invalid—the conclusion doesn’t follow. The right conclusion is:

“I will do p” rather than
“Necessarily, I will do p.”

The Fatalism Fallacy is a faulty distribution of the necessitation operator, which can be understood by bracketing the scope of “necessarily”. That is, a faulty change from:

Necessarily (if God knows that I will do p, then I will do p) to
If God knows that I will do p, then necessarily (I will do p).

Compare another similar example with the same form (based on one from philosopher Norman Swartz):

Premise 1: Necessarily, if Jenny has two cats and a dog, then she has at least one cat.
Premise 2: Jenny has two cats and a dog
Conclusion: Necessarily, Jenny has at least one cat.

Yet this is fallacious, because there was no logical necessity for her to own any cats; this is just a contingent matter that happens to be true. The correct conclusion is:

Jenny has at least one cat.

She had a choice in the matter—just like your eating a bagel.

Published: 17 August 2010