blog.kennypearce.net

Molinism and the Logic of Subjunctive Conditionals

The following is a plausible principle of the logic of subjunctive conditionals:

◊(p□→q), ◊p ⊨ ◊q

This is to say that if a subjunctive conditional is possible true, and its antecedent is possibly true, then its consequent is also possibly true. This principle is validated by most accounts of subjunctive conditionals, including those that allow for non-trivial counterpossibles.

If Molinism is true, then this principle is very likely false.

According to Molinism (as I use the term here), God exists necessarily and essentially possesses the attributes of omnipotence, omniscience, moral perfection, etc. God possesses two types of knowledge logically prior to God's creative decision: natural knowledge of necessary truths and middle knowledge of counterfactuals of creaturely freedom (CCFs). The CCFs are contingent truths about what possible free creatures would do if faced with various choices. The core of the Molinist theory of providence is the idea that God can employ these CCFs in deciding what to create. (The account in this paragraph is based on contemporary analytic Molinists like Plantinga and Flint, not on a reading of Molina himself.)

Because the Molinist accepts divine necessity, the Molinist (like other Anselmians) is committed to the view that God is the delimiter of possibilities. (This lovely phrase is due to Tom Morris, p. 48.) According to this view, there are some consistently describable outcomes that God necessarily prevents, because to do otherwise would be inconsistent with God's character.

However, if Molinism is true, then God can give a free creature a choice about whether q and still ensure that q does not eventuate. For instance, Tom Flint thinks that Molinism allows for the reconciliation of Papal infallibility with Papal free will. (If Flint had not written on this question, it would never have occurred to me that Catholics might have a problem of Papal free will!) According to Flint, if any person is such that if he were Pope he would make false statements ex cathedra, God can prevent that person from becoming Pope. But this, according to Flint, does not take away the freedom of actual Popes to say whatever they want in whatever circumstances they want.

Now suppose for the sake of argument that a Pope making a false statement ex cathedra is such a bad outcome that it is inconsistent with God's goodness to permit it.* It doesn't seem, on the Molinist picture, that this should change anything at all. Thus, if Flint is correct, it would appear that ordinary humans can be free to do the impossible.

This also leads to the invalidation of the (purported) logical principle with which we began. The key point here is that the Molinist takes the CCFs to be contingent. (The principle can be maintained if one adopts Leibniz's view that if the antecedent of a CCF precisely identifies one possible creature then the CCF is necessary.) Hence, for some person S, there will be some possible worlds at which if S were Pope, S would be Pope and make false statements ex cathedra and some other possible worlds at which if S were Pope, S would be Pope and not make false statements ex cathedra. Because the latter worlds exist, it is possible that S be Pope (it is consistent with the divine goodness to permit S to be Pope). This means, however, that the former conditional has a possibly true antecedent and a necessarily false consequent. Yet, according to Molinism, the conditional is possibly true.

---

Note

* I of course don't actually believe this, since I'm Protestant. Further, even orthodox Catholics needn't (and perhaps shouldn't) endorse the strong claim that it would be inconsistent with the divine goodness to permit this. They can (and perhaps should) go with the weaker claim that it is inconsistent with God's providential plan for the actual world to permit this. However, it's a convenient example requiring only a very small modification to a relatively simple example defended by an actual Molinist. In a number of papers Flint has employed a similar strategy in defense of the consistency of the Incarnation with the possession of free will by Christ's human nature. In this case, the necessity claim is much more plausible: it is inconsistent with the divine goodness for one of the divine Persons to be hypostatically united with a human nature that sins. However, this example is much more complicated than the Papal infallibility one.
Posted by Kenny at February 14, 2019 11:52 AM