Does Divine Foreknowledge Preclude Human Freedom?

I remember talking with a guy at my church a while back who expressed a similar question. Paraphrasing him, he said, “How can we act a certain way if God already knew and that’s how He created us?” What was interesting about this exchange is that even though he wasn’t articulating his question the way I have above, I could tell that’s ultimately what he was getting at. It’s a question that a lot of people have thought about even if they can’t really articulate it. Does divine foreknowledge preclude human freedom?

Knowledge and Freedom

In answering this question, I’ve found it helpful to ask if all knowledge of future events precludes free will. The answer to this question, in my view, is clearly no. I know, for instance, that I will finish writing this post. Does that mean I’m not free to stop writing it? Clearly not. I could throw it in the trash or lose interest and move on to something else–alternate possibilities abound.  Even though I know I’ll continue writing it and eventually post it, it’s possible I don’t. Here’s another example: I know that I will eat food today. Does that mean I’m determined to eat? No; I could go on a spontaneous fast; that’s clearly within the realm of possibility.

In these two cases, I know some future event will occur, but that doesn’t do away with or preclude freedom. Why? Well, notice that my knowledge isn’t actually causing anything to happen. Suppose I had no beliefs about whether I would finish writing this post. If my foreknowledge did away with my free will, then taking it away should give it back. But that’s a little weird, isn’t it? I either have free will or I don’t.

What’s going on is that my knowledge is causally unconnected from the event. What actually causes the event is my free choice as a rational agent. The knowledge isn’t doing anything. So, it’s false that knowledge precludes free will.

Knowledge and Certainty

Some might object here and say that I don’t really know I’ll continue writing this post or that I’ll eat food today. Knowledge requires certainty. If it’s possible that these statements are false, then I don’t know they are true. This objection says that genuine knowledge requires being absolutely certain.

Three problems are worth noting. First, are we certain that knowledge requires certainty? If we aren’t, then we don’t know it. Already, we’ve located an issue. Some might say that we don’t have to know that knowledge requires certainty, we just need to be justified or rational in believing it. However, the same exercise could be applied to justification/rationality. Are we certain that we are justified in believing that knowledge requires certainty? If not, we don’t know it. This is a serious worry.

Second, very few epistemologists today believe that certainty is a criterion of knowledge. In other words, most professionals believe that our knowledge can be fallible. Check out this excerpt from the Internet Encyclopedia of Philosophy article on Fallibilism:

Many epistemologists, probably the majority, wish to accept that there can be fallible knowledge (although they do not always call it this). Few of them are skeptics about knowledge: almost all epistemologists believe that everyone has much knowledge . . . In general, epistemologists also accept that . . . knowledge is rarely, if ever, based upon infallible justification: they believe that there is little, if any, infallible justification. Hence, most epistemologists, it seems, accept that when people do gain knowledge, this usually, maybe always, involves fallibility.

In other words, most experts in the field of epistemology think that knowledge doesn’t have to be certain. So I can know that I will eat food today even though I’m not absolutely certain I will.

Third, if knowledge requires certainty, we would be forced to say, actually, we don’t know much of anything. On this view, I wouldn’t know my wife loves me. Even though we’ve been married 7.5 years and she even told me this morning she loves me, it’s possible that she’s been pretending. Heck, I wouldn’t even know I have two hands! It’s possible we are living in the Matrix. The hands I think I have could be the result of a computer simulation. This view of knowledge requires giving up nearly everything we think we know; as such, it carries quite the intellectual price tag.

In my view, this objection doesn’t work. I agree with most epistemologists that I don’t have to be absolutely certain I’ll finish writing this post in order to know that I will. And if that’s the case, then foreknowledge doesn’t preclude freedom.

God and Freedom

“Alright,” the questioner says, “knowledge of future events doesn’t preclude human freedom. However, there’s a relevant difference between my knowledge and God’s knowledge. I can grant that my knowledge can be, and often is, fallible, but God’s knowledge is infallible. He can’t be wrong. So if He knows that I will do something, it has to come to pass whether I like it or not. If God knows what I’m going to do, it’s impossible that I act otherwise.”

Fair enough, there is a relevant difference here. Our knowledge is fallible in most cases. We don’t know much of anything with absolute certainty. But God, being the Maximally Great Being that He is, can’t possibly be wrong about anything He knows. Does this mean we aren’t free to do otherwise? No.

Modal Fallacy

The inference from ‘God’s knowing without error’ to ‘our not being free’ actually commits a fallacy in modal logic. To see why, let’s spell out the actual premises of the argument.

(1) Necessarily, if God foreknows that X will happen, then X will happen.
(2) God foreknows that X will happen.
(3) Necessarily, X will happen.

X above can just be any future event you want to imagine. In his work, Dr. Craig uses the helpful biblical example of Peter denying Jesus three times (Luke 22:54-62). Since God’s knowledge is infallible, premise (1) is obviously true. Necessarily, if God knows that Peter denies Jesus three times, then Peter denies Jesus three times. However, this does not imply that Peter necessarily denies Jesus three times. That conclusion doesn’t follow. To see why this doesn’t follow logically, let’s apply the same logic in a different context. Consider the following argument:

(4) Necessarily, if I have two prime lenses and two zoom lenses, I have at least four lenses.
(5) I have two prime lenses and two zoom lenses.
(6) Necessarily, I have at least four lenses.

Does (6) follow from the previous two premises? Not at all. That I have two prime and two zoom lenses doesn’t mean that I necessarily have at least four lenses. Even though the conditional statement in premise (4) is absolutely true, it’s clearly possible I have only one prime lens. In fact, it’s possible I have no lenses at all! So it doesn’t follow as a matter of necessity that I have at least four lenses. In order for that conclusion to follow, it must be a necessary fact that I have four or more lenses. But that is not a necessary fact (e.g., there was a time in my life I didn’t own any lenses). The argument, as I say, commits a fallacy in modal logic.

Let’s go back to Peter denying Jesus. It’s clearly not necessary that Peter deny Jesus exactly three times. He could have only denied Jesus twice. He could have denied him five times. But then the conclusion this objection needs can’t be reached (without committing a fallacy). In practical terms, if Peter denied Jesus twice instead of three times, then, very simply, that’s what God would have known instead. If Peter denied Jesus once, then God would have known that. But since he denied Jesus three times, that’s what God foreknew. And again, this knowledge isn’t causing Peter to do what he does, just like my knowledge that I’ll finish writing this post isn’t causing me to finish it.

Chronological vs Logical Priority

If you’re still unsure about all this, here’s another way of looking at it. God’s foreknowledge is chronologically prior to the event (in time), but the event is logically prior to the foreknowledge. In other words, the knowledge doesn’t cause the event, the event causes the knowledge. God knew before it happened that I would eat dinner last night. But that knowledge didn’t cause me to eat. I could have just gone to bed. And if I simply went to bed, then God would have foreknown that instead.

Part of why the mix up happens is because we have a hard time wrapping our minds around what on the surface looks like backwards causation. But if we can keep this chronological/logical distinction clear in our minds, the problem disappears. The event causes the knowledge even though the knowledge is chronologically prior to the event.

Concluding Thoughts

In God, Freedom, and Evil, Alvin Plantinga makes a really interesting point:

Perhaps the following is a possible source of confusion here. If God is essentially omniscient, then He is omniscient in every possible world in which He exists. Accordingly, there is no possible world in which He holds a false belief. Now consider any belief that God does in fact hold. It might be tempting to suppose that if He is essentially omniscient, then He holds that belief in every world in which He exists. But of course this doesn’t follow. It is not essential to Him to hold the beliefs He does hold; what is essential to Him is the quite different property of holding only true beliefs (Plantinga, 72).

There’s more to be said here, but I hope this is at least a start in the right direction. Feel free to comment below with any further worries or objections.


[1] Should be noted that Open Theists reject this.

If you enjoyed reading this blog post, consider supporting us on Patreon!



Get updates on new posts, upcoming live discussions, and more.

Leave a Reply

4 Comment threads
9 Thread replies
Most reacted comment
Hottest comment thread
6 Comment authors
PeterHughBrenda von AhsenCameron BertuzziScott Christensen Recent comment authors
newest oldest most voted
Notify of

Cameron, Thanks for the post, good stuff! One possible objection that comes to mind is from Hasker’s book on Metaphysics in which he lays out a similar argument but which is slightly different. Essentially, if God knows true beliefs about what we will do, and the past is inalterable, then we cannot act otherwise as that would contradict God’s knowledge. I know this seems to falls into logical fallacy you laid out, but if one agrees that God’s foreknowledge does not causally determine us, but reveals a deterministic framework (because he knows we can only do one action in a… Read more »

Scott Christensen
Scott Christensen

If God knew infallibly that Peter would deny him 3 times, then this knowledge also indicates that God knew infallibly that Peter would not deny Christ 2 times. If Peter has agent casual libertarian free will (LFW) then he must be equally able to choose A or not choose A at the moment of choosing, and if that is the case, then God could not possibly know infallibly beforehand which choice Peter would make. However, if it is known infallibly beforehand that Peter would deny Christ only 3 times, then it is not possible he could deny him less or… Read more »


Cameron, with regards to backwards causation, do you think a special problem arises for those who endorse a presentist view of time? Take the first moment of time. God knew what I would do today because what I did today caused his knowing it back at the first moment of time. But back then, I didn’t exist; so something that didn’t exist has causal powers, which is counterintuitive.


I don’t get the Modal Fallacy part. Can’t one just take out all the modal language and make a sound argument without it? I.e.

(1) If God knows that I will perform some action X tomorrow at 5pm, then I will perform action X tomorrow at 5 pm. (infallibility of God’s knowledge)
(2) God knows that I will perform the particular action Y tomorrow at 5pm. (omniscience)
(3) Therefore, I will perform the particular action Y at 5pm tomorrow.

What’s wrong with that?