Philosophical Theology

Molinists and Calvinists agree over the soundness of the following argument, where x is a future creaturely choice.

1. Necessarily, if God foreknows x, then x will happen
2. God foreknows x
3. Therefore, x will happen

Molinists and Calvinists even agree that the following argument is not just unsound but invalid:

1. Necessarily, if God foreknows x, then x will happen
2. God foreknows x
3. Therefore, x will necessarily happen

The fallacy in view is that of transferring the necessity of the inference to the conclusion. Although necessarily, God’s foreknowledge of x is a sufficient condition for x’s eventuality, it does not follow that x itself is necessary. That’s because God’s knowledge of the future does not imply that there is only one future God could actualize. In a word, it is necessary that God know all possible contingencies and that he contingently know his creative decree.

William Lane Craig is correct that “Christian theology always maintained that God’s creation of the world is a free act, that God could have created a different world – in which x does not occur – or even no world at all. To say that God necessarily foreknows any event x implies that this is the only world God could have created and thus denies divine freedom.”

The lifting of the veil?

Let us allow x* not to be a future creaturely choice that is contingent upon God’s creative decree, but let x* be a would-counterfactual of creaturely freedom that Molinists assert is not contingent upon God’s free decree.

Molinists and Calvinists enjoy further agreement:

1. Necessarily, if God knows x*, then x* is true
2. God knows x*
3. Therefore, x* is true

Although we may not immediately transfer the necessity of the inference from the major premise to the conclusion, if the minor premise is also a necessary truth, then we may and the conclusion would not entail an illicit transfer of necessity.

Can we defend that Molinism entails this fourth and final deductive construct?

1. Necessarily, if God knows x*, then x* is true
2. Necessarily, God knows x*
3. Therefore, x* is necessarily true

From a Molinist perspective, if God knows x* is true in all possible worlds in which God could actually exist, then God cannot avoid knowing x*. Therefore, x* would appear at least to function as a necessary truth since God cannot know x* is false (or that x* doesn’t exist as true). If God cannot alter his knowledge of x*’s propositional* reality and truth value, then how is it not true that, p: necessarily, God knows x*? And if p, then how is x* not a necessary truth?

Due to the contingency of future happenings, we may not apply the fourth and final deductive construct to what will occur. However, since the Molinist middle knowledge entailed by x* is not contingent upon God’s will, it is not obvious why Molinism doesn’t reduce all would-counterfactuals of creaturely freedom to necessary truths, which is not the case for Causal Divine Determinism.

God as a necessary being exists in all possible worlds:

Molinists argue that x* is not a necessary truth because there are possible worlds in which x* is false. This is where Molnists make the bald assertion that some possible worlds are infeasible worlds. Infeasible worlds are semantic lands in which counterfactuals that would never obtain conveniently exist. (It might be useful to distinguish (a) narrow or strict logical possibility from (b) broad logical possibility or metaphysical possibility. )

However, in what functional or relevant sense is a world truly “possible” if it is not a world that God can actually inhabit or actualize? What can we make of such possibilities that God cannot experience because of their infeasibility? In other words, possible-infeasible worlds are semantic lands that God the Creator cannot actually inhabit, yet counterfactuals of creaturely freedom supposedly can! (Indeed, even God’s creatures who are alleged to be free to choose contrary to how they would cannot actually exist in a world in which they do choose as such.)

Even for the Molinist, God is a necessary being. Yet God cannot escape from counterfactual x* in any possible world in which he can inhabit. Therefore, at least on the surface, x* appears as necessary as God. Furthermore, since God cannot truly inhabit any infeasible world, then infeasible worlds would appear to be impossible worlds, not just functionally speaking but broadly-logically or metaphysically so. We will drill down further in a bit but for now, given Molinism God cannot exist within an infinite number of possible worlds (the infeasible ones).

Molinists say something like this:

“In possible worlds where x* is false, God obviously doesn’t know x* but rather knows ~x*.”

That much is true but a caveat is in order. Given the truth of x*, Molinists will not say that if a possible world with ~x* were actualized, God would know that -x* is true. Rather, Molinists acknowledge that such a possible world cannot be actualized because it’s an infeasible world virtue of ~x* alone. Accordingly, it’s not possible God exist in such a world other than theoretically.

Unpacking the distinction and locating the internal inconsistencies of Molinism:

Coming at this from another angle, the falsity of x* in any possible world is sufficient to make it an infeasible world (Iw). Yet, if Iw is also a possible world, then God as a necessary being could not have middle knowledge of would-counterfactual x* in some possible worlds in which he can supposedly exist. By applying modus tollens to the preceding if-then proposition we deduce: Iw is not also a possible world because it’s not possible that God not know all counterfactual truth. Therefore, Molinism may not park counterfactuals of creaturely freedom in possible-infeasible worlds in order to veil the necessity of their would-counterfactuals and establish their ungrounded contingency. Not being able to appeal to the infeasible land of misfit counterfactuals, all Molinist would-counterfactuals of creaturely freedom are unveiled as functionally necessary truths because they exist in every possible actualizable world, which are the worlds God can inhabit.

For the Molinist x* is a trans-world truth across all their feasible worlds. Accordingly, ~x* cannot actually be true and actually known by God. In that context there’s an inconsistency in their “consistent state of affairs” or possible world semantics.

Essential properties:

Assume: p is an essential property of an object o if it is necessary that o has p if o exists. In other words, if o exists, then o has p. Accordingly, o has p is a necessary condition for o to exist, and o’s existence is a sufficient condition for o having p.

Within Molinism, and not within Theological Determinism, if x* is always true of a particular person (S) whether he exists or not, then x* is an essential predicate about the person. Therefore, it is not the identical person who would choose according to ~x*, which undermines the consistency of the possible world reality of how things might have been in Iw.

A possible world (PW) is a consistent way the world could have been. Yet Molinism has many PWs in which ~x* is true. Those PWs are not just infeasible (unactualizable) worlds, they are inconsistent possible-realties because within them S is not actually S (given what is essential to S). All such PWs are narrowly logical but broadly illogical worlds. Therefore, in Molinism x* cashes out as a necessary truth because it is true in all truly possible (non-contradictory) worlds. All appeals to ~x* existing in infeasible yet possible worlds cash out as appeals to narrowly logical possibilities that don’t take into account what is metaphysically or broadly relevant. It’s akin to saying God can sin, which is narrowly logical but not broadly or metaphysically so. Given God cannot sin, a world in which God sins is not just an infeasible world; it’s an impossible world given God’s essential (and necessary) nature. Otherwise we’re not talking about the identical God. So it is for a free moral agent who’d freely choose according to counterfactual ~x*! If x* is always true about a person even without his actual existence, then x* is essential to the person whether actualized or not. Therefore, it is not the identical person who would ever choose according to ~x*, which undermines the consistency of the possible reality of how things might have been in Iw. The inconsistency is not narrowly illogical but instead broadly so. The actual existence of the person always precludes choosing ~x*. Essential to his being is x*.

In a word, if person S would never exist without x* being true about him, then x* is essential to S. Accordingly, counterfactual ~x* as it relates to S is sufficient for an impossible world. If true, then no infeasible world is a broadly-logical possible world, which leads to CCFs to function as necessary truths because their contingency is not grounded in God’s creative decree, as it is in Reformed thought.

We may not consider possible worlds in isolation from other possible worlds given that each possible world includes ideas about other possible worlds. The consistency of any given reality must coincide with other possible realities. With that in mind, Molinism is left with:

1. The actual God cannot believe something false.

2. x* is actually true.

3. The actual God cannot believe ~x*.

4. Necessarily, the actual God (as a necessary being) “exists” in all possible worlds.

5. In some possible world (PWf) x* is false (~x* is true).

6. Necessarily, the actual God believes only what is true and everything that is true.

7. In two different possible worlds the actual God believes contradictory truths x* and ~x* respectively about an identical person. (Inconsistency)

8. Some possible worlds do not entail a consistent state of affairs since they do not contemplate the actual trans-world God or identical persons as they relate to x*.

Closing out:

The possibility of actualization for God is creature-dependent within Molinism. Consequently, Molinism allows for some narrowly-logical possibilities that are purely theoretical – so much so that God cannot know them as actualized realities. These alleged possibilities can be actualized a whopping zero number of times, even though there are an “infinite number” of these possibilities. Since zero divided by infinity is zero, an event that’s supposedly possible has a probability of zero. Therefore, God cannot possibly strongly actualize the conditions for ~x* in order for the free choice of ~x* to obtain. Zero probability merely corroborates: (a) certain Molinist CCFs are not relevantly possible and (b) CCFs in the hands of Molinism function as necessary truths that are essential to respective persons.

Unlike with theological determinism, Molinism cannot reconcile possible worlds with each other, personhood or an exhaustively omniscient and necessary being. Theological Determinists have no such problems because CCFs are not trans-world truths nor artificially contingent as is the case with Molinism. At the very least, Molinism destroys the usefulness of possible world semantics. For Molinism, possible worlds are narrowly logical and not broadly so; they’re functionally meaningless.

Molinist infeasible worlds are not consistent states of affairs of how things might have been. Therefore, I don’t believe they qualify as possible worlds in any useful way.

If God is a necessary being, then he exists in all possible worlds. Yet certain features of certain possible worlds prevent God from making those possible worlds actual even though he can supposedly exist in them, supposedly without incoherence.

*If a Molinist takes a nominalist view of propositions, entailing their non-actual existence, then how might they adequately defend middle knowledge of counterfactuals of creaturely freedom? A defense for realism about propositions isn’t necessary when interacting with the realism that the concept of middle knowledge presupposes.