August 30, 2009
Misconceptions about Middle Knowledge
Hi dr Craig.
Here is an argument that I wrote to a friend following a discussion of Molinism, in which I claimed that middle-knowledge and libertarian freedom are incompatible. It looks sound to me, but I would be very interested to hear where you think my reasoning fails, after all, I am not a professional philosopher. Thank you so much.
here it is:
When you consider the truth status of counterfactual statements, I believe these statements need to have in them enough information to determine them to be true or false.
Consider the following statement in order to see why:
-"If you had cut off all your hair yesterday, then you would be bald".
Two things can be said about this statement:
1) it is counterfactual
2) it is true
This statement is counterfactual, because it is contrary to the actual world: you did not actually cut off all your hair yesterday.
And this statement is true; indeed it is true that "IF you had cut all your hair yesterday, THEN you would be bald today".
But the only reason why we CAN say that this counterfactual statement is true, is because although it is counterfactual, it contains enough information IN ITSELF, to determine that it is true. Namely, the information "you cut off all your hair yesterday" is sufficient to determine your baldness of today.
Consider now the following counterfactual statement:
-"If you were to find water under the pressure of 100.000 Pa, then it would be liquid".
Let us ask: is it true?
Our answer has to be "Well I don't know!". Pressure is a necessary parameter, but in order to say that water would be liquid, I ALSO would need its temperature. Pressure AND Temperature would determine its state, and only then could I know whether the statement is true or not. If the hypothetical conditions do not determine the truth of the counterfactual, then I cannot know if it is true and neither can God, since we both are lacking information to answer that particular isolated question.
So when you apply this to middle knowledge, namely the knowledge of counterfactuals of creaturely freedom, I ask the question:
-Could God know before the creative decree, what Pilate would FREELY choose if he were placed in all the biblical conditions? (namely crucify Jesus, or not) As seen above, we are able to know this counterfactual to be true or not, ONLY IF it contains enough information to determine its truth status. Namely, we need to know that "placing Pilate in all the biblical conditions" does determine that "Pilate would choose to crucify Jesus". But this determination, by definition, is only available if Pilate does NOT have libertarian freedom. By definition, if Pilate has libertarian freedom, then regardless of ALL the right biblical conditions, he can still choose one way or the other, and we are missing information to know what he WOULD really choose. Therefore God could not have had middle knowledge of creaturely libertarian freedom prior to the creative decree.
Now again, I believe God did have middle knowledge, but the only reason why it is possible is because I believe we have a compatibility freedom, not a libertarian.
Is it clear? Feel free to challenge my thinking.
Your question is interesting, Guillaume, because you want to combine middle knowledge with a compatibilist view of freedom, according to which our actions are causally determined by the circumstances in which we find ourselves, whereas the Molinist affirms that we remain non-determined in fully-specified, freedom-permitting circumstances.
It seems to me that your view won't work because it will collapse God's middle knowledge either into His natural knowledge (God's knowledge of all necessary truths) or else into His free knowledge (God's knowledge of contingent truths posterior to His creative decree). It will be natural knowledge, if, once all the circumstances and laws of nature are specified, it is logically necessary that one choose a certain way. It will be free knowledge if it is logically contingent how one chooses once all the circumstances and laws of nature are specified. So you can ascribe to God counterfactual knowledge, all right, just as He has counterfactual knowledge of what physical events would happen under various specified circumstances, but this isn't really middle knowledge as such. To count as middle knowledge, the relevant counterfactuals have to be (i) contingently true and (ii) true prior to God's decree.
But what about your critique of the Molinist perspective? It seems to me that it is predicated on a number of misconceptions. You're right that in order to have a truth value the counterfactuals in question need to have sufficient information in their antecedent clauses specifying the circumstances. Some counterfactuals may be neither true nor false because the circumstances in their antecedent clauses are insufficiently specified. Now in practice we solve this problem by simply assuming that the actual world is held constant except for the alteration of the circumstances mentioned, and then we ask what would be true. Even in your example of cutting off all your hair, you hold constant the actual laws of nature governing the rapidity with which hair grows! But in dealing with divine middle knowledge Alfred Fredosso and Thomas Flint resolved this problem neatly by simply stipulating that what they call "counterfactuals of creaturely freedom" have fully specified circumstances in their antecedents. Such counterfactuals have the form "If agent S were in circumstances C, then S would freely do action A."
Notice that what we're talking about here is not a matter of our knowing the truth value of such a counterfactual. Rather it's a matter of its even having a truth value. The problem raised by inadequate specification is not a matter of our knowledge of such counterfactuals but more fundamentally of their being either true or false, regardless of whether we can know their truth value of not.
So in the case of what Pilate would freely do in circumstances C, we stipulate that C must be fully specified, including the whole history of the world up to the time of choosing. In virtue of the antecedent's being fully specified, the counterfactual of freedom will be true or false. But when you say,
we are able to know this counterfactual to be true or not, ONLY IF it contains enough information to determine its truth status. Namely, we need to know that "placing Pilate in all the biblical conditions" does determine that "Pilate would choose to crucify Jesus". But this determination, by definition, is only available if Pilate does NOT have libertarian freedom,
you're confusing the proposition's being sufficiently determinate to have a truth value with the choice's being causally determined. The counterfactual "If Pilate were in C, he would freely do A" can be true or false only if that statement is determinate in the sense that C is fully specified. But the fully specified circumstances are not causally determining circumstances. On the contrary, they're stipulated to be freedom-permitting circumstances. How an agent would freely choose is not determined by the circumstances he is in. It is simply up to him what he will do. So don't confuse a proposition's being determinate with an action's being determined.
So the Molinist concurs wholeheartedly with you that "if Pilate has libertarian freedom, then regardless of ALL the right biblical conditions, he can still choose one way or the other." Absolutely! He can choose one way or the other; but he will choose one way. If he were in C, he would freely choose A, though he could choose not-A instead. Don't confuse what someone would do with what he could do or think that because he would do A he could not do not-A.
There's no lack of information in such a case either. So long as the circumstances are fully specified, the counterfactual will have a truth value, and God, as an omniscient being, must know it. He knows what is the truth value of every proposition, regardless of our ignorance.