Posts Tagged ‘Molinism’

Leibnizian Supercomprehension

In a recent paper, Juan Garcia has argued that Leibniz is, in an important sense, “a friend of Molinism.”1 For those who are familiar with contemporary versions of Molinism (e.g., Flint), this suggestion is rather surprising, since Leibniz is clearly a theological determinist: he holds that God chooses every detail of the actual world. Further, a key feature of Molinism (particularly as it is understood in recent analytic philosophy) is the idea that God’s options for creation are limited by contingent but prevolitional counterfactuals of creaturely freedom. The contrary assumption, that God could have actualized any possible world, was dubbed by Plantinga, ‘Leibniz’s Lapse’. Now, Plantinga is clearly unfair in calling this a ‘lapse’: it’s a theoretical commitment Leibniz embraces clear-sightedly, with awareness that there are alternatives to it and that it has some implausible-looking consequences. Nevertheless, Leibniz does embrace this view. As a result, Leibniz surrenders the two main advantages typically assigned to Molinism in recent analytic philosophy: the ability to combine theological libertarianism with strong providence, and any Plantinga-style free will defence.

Still, Leibniz’s view does have significant similarities with Molinism and seem to employ some Molinist strategies. The main burden of Garcia’s article is to argue that Leibniz holds that counterfactuals of creaturely freedom are contingent and that God knows them prevolitionally. This seems right in a certain sense or to a certain extent, but the key issue (which Garcia does not discuss) is how Leibniz can maintain this, and here Leibniz’s notoriously weak notion of contingency is doing real work.

According to traditional Molinism, an agent’s essence or character does not include all the facts about what the agent would do in hypothetical situations, because this would render the agent’s actions in those situations necessary. Nevertheless, God ‘supercomprehends’ this essence or character in order to know facts that go beyond what is included in it. It is in this way that God knows what the agent would do.2

This doctrine is rather obscure. It leaves unexplained how or why these facts could be true of this essence. (This is, of course, a version of the ‘grounding objection’.) It also seems to have God seeing in the essence something that isn’t there.

Leibniz, however, is able to give an account of how this works. In the Arnauld correspondence, Leibniz says that, in deciding to create Adam, God employs a ‘complete individual concept’ (CIC) of Adam which includes, essentially, an entire possible world. This is because the CIC must include everything that can be truly predicated of Adam, and there are true propositions such as Adam lived before I wrote this blog post. Hence, my writing this post is included in Adam’s CIC.

How, then, can any of Adam’s actions be contingent? In a certain sense they can’t, and Leibniz makes important use of this in his discussions of the problem of evil. (See the discussion of Judas in the Discourse on Metaphysics and the discussion of Sextus at the end of Theodicy.) Arnauld, therefore, objects that Leibniz’s view falsely implies that he (Arnauld) is necessarily a celibate theologian. In fact (Arnauld says) he could instead have been a married physician. Leibniz responds by diagnosing a kind of ambiguity: this Arnauld (the one in the actual world) is necessarily a celibate theologian, but there are other Arnaulds, in other possible worlds, who are married physicians. (I’ve written about this before.) This, according to Leibniz, gives us enough contingency for moral responsibility and so forth.

This ambiguity, it seems to me, is what allows Leibniz to combine these Molinist principles with theological determinism. God knows precisely what will happen if God creates this Arnauld. At the same time, it is contingent that Arnauld, in these circumstances, becomes a celibate theologian.

On the other hand, Leibniz’s strategy does not make what is perhaps the most characteristic move of both historical and contemporary Molinism: he does not open up any middle ground between natural knowledge and free knowledge. That Arnauld, placed in these circumstances, would freely decide to become a celibate theologian is true (on the interpretation on which it is contingent) because God decided to create this Arnauld, rather than one of those other Arnaulds. Hence this is part of God’s free knowledge. That this Arnauld freely so decides is necessary, since it follows from his CIC. (Leibniz often seems like he wants to wriggle out of the conclusion that this is necessary, even when we specify this Arnauld, but the comments about Judas and Sextus and so forth seem to push the other direction.) Thus it is Leibniz’s view that counterfactuals of creaturely freedom are systematically ambiguous, and on one interpretation they are part of God’s natural knowledge while on another they are part of God’s free knowledge.

(Cross-posted at blog.kennypearce.net.)


  1. A while back, Brandon Watson also suggested, in a blog comment, that Leibniz can be understood as endorsing a kind of ‘modified Molinism’.
  2. I rely on Adams for this characterization of ‘supercomprehension’.

Read Full Post »