Brief Discussion on Middle Knowledge and Divine Fatalism

In a recent Dividing Line podcast, James White was responding to a recent debate on Calvinism vs Molinism between Paul Helm and William Lane Craig. It’s a rather good podcast in general and I would recommend it. White makes many good responses worthy of consideration. In the future, I may cover some of his other points, but for now, I’m concerned with his closing remarks.

I consider myself to be broadly in the Molinist camp. There is a very common mistake I’ve seen and White makes it here as well. White made the argument that freedom requires the possibility of violating middle knowledge. He used the example that WLC used of the subjunctive conditional “If I were rich, I would buy a Ferrari”. White argues, however, that if you do become rich and are significantly free, you should be able to buy a BMW instead. But Molinism doesn’t allow for that. If God knows that you will buy a Ferrari then you couldn’t buy a BMW. The conclusion being that Molinism must be deterministic or otherwise does not truly preserve libertarian freedom.

Here’s the mistake: White, like many detractors of Molinism, conflates “will do” and “must do”. To go back to our example, if Jones were rich, he would buy a Ferrari. Let’s then say that Jones won the lottery today. Since he is now rich he will go buy a Ferrari. But, White says that since Jones is rich, he must buy a Ferrari. This is a different proposition altogether! The first case (will buy) says that Jones will freely use his financial resources to actualize a state of affairs wherein he owns a Ferrari. The second case (must buy) says that Jones does not have it within his power to refuse actualizing this state of affairs. When we say that Jones will buy a Ferrari, we are not denying the fact that he could buy a BMW instead. It is entirely within Jones’s power to buy a BMW or a yacht or donate his money to charity or anything that he wishes. The fact is that he just won’t do any of those things.

This may seem a bit difficult to understand. Let’s use another example. Assuming that you are a fairly decent human being, what would you do if someone were to hand you an infant? Would you intentionally harm this infant for the fun of it? Of course not! But, does that somehow restrict your freedom to do so? Of course not! You could harm this infant if you so chose to; it’s completely within your power but you will restrain.

This idea that we must do what God knows we will do is called divine fatalism which is dramatically different from what Molinism communicates. In this respect, I think that Dr. White has significantly misunderstood the relationship between divine foreknowledge and human freedom in the Molinist framework.

J.W. Wartick also weighed in on Dr. White’s misunderstanding of Molinism

Further Resources on Molinism

Check out the forum for this blog on Facebook for discussion of this post and many other topics!

Advertisements

Tags: , , , , ,

About caplawson

biomedical engineering // christian theism // texas a&m // molinism // coffee // ratio christi

7 responses to “Brief Discussion on Middle Knowledge and Divine Fatalism”

  1. Remington says :

    It’s been a while since I’ve listened to the Dividing Line you’re addressing, so I’m not too concerned to defend White’s original claim. Still…

    >>White made the argument that freedom requires the possibility of violating middle knowledge.

    My own criticism would be that libertarian freedom, the type of freedom Molinists typically want to affirm, makes it hard to see how there can even be middle knowledge. For one thing, there is no truth value to what an agent S would do in circumstance C, simpliciter. It’s both true and false false that S will X in C. You have to back up and talk about possible worlds. But then to hold on to PAP you either have to affirm that an agent can either actualize a different world than the one actualized or else that it simply doesn’t make sense to speak of a world being actualized until the moment of choice. Of course no Molinist wants to affirm this last option… But then again they shouldn’t want to first one either since I’m not even sure it’s coherent. I talk about this some in my podcast: in what sense has God actualized a world if S exercises his power of contrary choice to actualize a different world? And saying this is just *possible* and never actual doesn’t solve the problem and this gets back to James Anderson’s critique of a fallible God: if it’s possible there is a possible world where God fails. (And how do we know that isn’t this world? Strange thing that no one has exercised PAP in human history, no?)

    >>White, like many detractors of Molinism, conflates “will do” and “must do”. To go back to our example, if Jones were rich, he would buy a Ferrari. Let’s then say that Jones won the lottery today. Since he is now rich he will go buy a Ferrari. But, White says that since Jones is rich, he must buy a Ferrari. This is a different proposition altogether!

    I don’t see the problem, on White’s part, because I assume White is taking it for granted that God has actualized the world in which he bought the Ferrari. If that’s the case, then it’s true that White must by the Ferrari, given that God has actualized the world in which White buys a Ferrari, or you run into the problem above. He’s not saying there is no possible world in which you don’t buy the Ferrari, but that given the world God has actualized, those possibilities are no longer live options for him. It’s no longer on the table for him to not buy the Ferrari. Otherwise you run into the problem above.

    >>The second case (must buy) says that Jones does not have it within his power to refuse actualizing this state of affairs.

    But the Molinist wants to say *God* has already actualized a world. God chose which world would be actual, not James White. So I’ll ask you the question I asked in my podcast: does Jones have the power to actualize a world different than the one God has actualized? If so, what does it even mean to say God has actualized a world (since he failed to, given Jones’ counter-actualization)? If not, in what significant sense does Jones have it within his power to refuse the actualization God has given him, so to speak?

    >>This may seem a bit difficult to understand. Let’s use another example. Assuming that you are a fairly decent human being, what would you do if someone were to hand you an infant? Would you intentionally harm this infant for the fun of it? Of course not! But, does that somehow restrict your freedom to do so? Of course not! You could harm this infant if you so chose to; it’s completely within your power but you will restrain.

    I don’t think that’s a good analogy. Take for instance Robert Kane’s (he’s a philosopher who affirms LFW) Luther example. (Caveat: I haven’t read Kane’s book in years so I apologize if I get something wrong here). Kane says it’s plausible that Luther was telling the truth when he said “Here I stand, I can do no other.” That’s because given Luther’s history (e.g., character formation, past experiences, etc.) it’s plausible that Luther really couldn’t do otherwise than what he did. Well it seems to me that for most of us we are in just such a situation when holding an infant. So I don’t affirm that I have the (Libertarian) freedom to do otherwise in such a scenario as you present.

    >>This idea that we must do what God knows we will do is called divine fatalism which is dramatically different from what Molinism communicates.

    It’s also a perennial problem in philosophy how God foreknow what will do and yet we’re not in a situation of “must do”. Molinism doesn’t give us any solution to that problem, which is why it’s legitimate for people like White to point to this issue as a difficulty for Molinism.

    • caplawson says :

      Hi, Remington! Thanks for this great comment. I’ve been mulling it over between schoolwork these few days.

      “For one thing, there is no truth value to what an agent S would do in circumstance C, simpliciter. It’s both true and false false that S will X in C.”

      This seems to me to simply be an assertion that counter-factual statements have no truth value. I’m not convinced that this is the case. First, it seems rather intuitive considering my daily conversation and planning is riddled with subjunctive statements I take for granted. I would wager this is true of pretty much everybody. I don’t have a good defeater for this intuition so it seems to me to be perfectly rational to accept that counterfactuals have truth value. Secondarily, there is a whole system of specifically counterfactual logic which strongly suggests to me that at bare minimum, it is internally consistent to have counterfactual truth values. However, if I were presented with a good argument that counterfactuals have no truth value, I would change my mind.

      It’s true that an agent cannot actualize a world other than the world actualized as this would entail a logical contradiction. On the other horn of the dilemma presented is that it’s nonsensical to speak of a world being actualized until the moment of choice. I’m not sure what you mean by ‘moment’ here. In a temporal sense, yes, that would be nonsense as the world was actualized by the creative decree of God. The remainder of the critique is dependent on the earlier assertion that counterfactuals have no truth value. Assuming there are no counterfactual truths, then there are two types of worlds: the possible and the actual. In this respect, God would be fallible in the way that Dr. Anderson postulates. But, let’s assume counterfactuals do have truth values. S finds herself in C with choices X, Y, and Z. So there is a possible world wherein she does X, another where she does Y, etc. However, there is no world where S does X, Y, and Z simultaneously. These decisions are mutually exclusive. Therefore, it is necessary that she makes one decision. The content of that one decision is what populates the truth value of our counterfactual. So, when saying “possible but never actual”, it’s important to ensure that the logical (not temporal) moments are in the correct order which is a follows
      1. In C, it is possible for S to X, Y, or Z
      2. S freely chooses Z in C
      3. God knows (2) is true
      4. Creative decree
      There is nothing about 2-4 that negates the truth of (1), so, S genuinely does have the freedom to actualize a world that entails ~Z. However, doing so would change the truth value of (2). There is a possible world where S does ~Z in C but there is not a world where “God knows S would do Z in C” and “S does ~Z in C” are simultaneously true. To clarify, I’m not saying that God gets his knowledge from things; He knows all things in Himself and from Himself. He knows what every one of an infinite possible individuals would do in any conceivable circumstance. Also, I’m interested in this podcast you alluded to. If you were to link to it, I would be glad to listen to it.

      Dr. White’s error is that he is conflating the necessity of the inference as the necessity of the action. We know that
      1. If Jones were rich, he would buy a Ferrari
      2. Jones is rich
      The rule of modus ponens holds in counterfactual logic, so, we can correctly infer
      3. Jones will buy a Ferrari
      This is a necessary inference given the validity of logic for ((P→Q)^P)→Q is a necessary truth. It is absolutely certain that Jones will buy a Ferrari. But the action isn’t necessary! We can’t conclude that “It is necessary that Jones will buy a Ferrari” from (1) and (2). That would be like saying
      ((P→Q)^P)→◻︎Q
      Consider another famous argument
      1. If it is raining, then it is wet outside.
      2. It is not wet outside
      From this, we can only conclude
      3. It is not raining
      It’s appropriate to say “It is not raining” necessarily follows. However, to say “It is necessary that it is not raining” would imply some kind of trans-world drought.

      Is it still on the table for Jones to not buy the Ferrari? Look back at my earlier structure of logical moments but replace S with Jones and Z with “buys a Ferrari”. Of course Jones could do something else with his money, but, doing so would change the truth value of (2). How do we know that this won’t occur in the actual world? Because (2) is logically prior to the creative decree.

      Does Jones have the power to actualize a world different than the one God has actualized? No, that would entail a logical contradiction. Prior to the actualization of this world, God knew what the possible individual Jones would do in every conceivable circumstance. In what significant sense does Jones have it within his power to refuse the actualization God has given him, so to speak? It would entail a logical contradiction to have a world that was both actualized and not actualized. But, why is it necessary for Jones to ‘buck the system’ in order to have LFW? I would posit an extremely simplified definition of LFW as ‘the ability to genuinely decide between alternatives in any given circumstance’. It seems to me that God’s knowledge [of what an individual’s decision would be in any given circumstance] is causally inert.

      I did cheat a bit on the baby example. What I was attempting to illustrate was an instance where we have practical certainty about a free action. I doubt you would say that there is no possible world wherein you harm an infant. Because harming an infant does not entail a logical contradiction, we can conclude that there is indeed one possible world where you harm an infant. However, you know that in the actual world, if you were to hold an infant, you would not use your free will to intentionally harm it. You have certainty about a future free decision. I cheated on this one because it gets to be really messy when talking about the malleability of the will, soft libertarianism, etc. Given the background circumstances, experiences, regeneration, and the like, your will has been shifted such that certain decisions, like harming a baby, become immensely difficult to undertake. I’d rather not get sucked into this rabbit hole, but, I will acknowledge its existence.

      Finally, I agree, the interrelation of free acts and foreknowledge is quite an issue. One large error, as I mentioned earlier in this reply, is that the inference’s necessity is being imported to the actions necessity. Secondarily, I think that the presupposition that subjunctive conditionals have no truth value is a false one. I do think Molinism’s answer, albeit incomplete, is the best in my estimation. However, if I were presented with a more convincing alternative I would change my mind.

      • Remington says :

        Hi Capt. Thanks for the response. I like the idea of taking time to think things over before responding. In fact as I was listening to James White’s Dividing Line yesterday he was talking about this very issue: the urge to quickly respond and our need for patience.

        I would follow that advice myself, except for the fact that once I let something go it seems to run away from me, never to return. Cases in point can be found on my blog 🙂

        Nevertheless, in an attempt to take baby steps towards the more responsible thing, let me throw out a few quick thoughts and then I’ll think it over some more and hopefully give a more thoughtful response later.

        1. I don’t think I’m denying that counterfactuals have truth values. I affirm that there can be truth values to counterfactuals, such as what Chorazin and Bethsaida would have done (Luke 10:13). But I think I have a reason that can account for counterfactuals having truth values that the Molinist doesn’t want to take: counterfactuals are ways God could have ordered the world. In other words, they are grounded in God.

        *Maybe* I would suggest it’s like the grounding problem for moral facts in atheism. The atheist is correct to say “Look, it’s just obvious to me that moral claims have a truth value” the problem is that we don’t see how to account for that on atheism, but we do on theism. Just a thought.

        2. I don’t think I’m simply denying that counterfactuals have truth values. That is, I’m not just asserting a grounding problem in the section you quote. As I said, there is no truth value to “S will X in C *simpliciter*” Rather, I was trying to derive something from Molinist premises.

        The Molinist wants to say that it’s possible that Sally X in C and that Sally ~X in C. Right? Well if that doesn’t mean that there is a possible world where Sally would X in C and that Sally would ~X in C then I just don’t know what the Molinist means when he says it’s possible that S X/~X in C. If he just means it’s contingent in the sense of it not being logically necessary or something like that, fine. I think the Calvinist affirms that events are contingent in that sense. But I don’t think that gets the Molinist PAP. It’s sort of like saying “If Remington had wings and feathers he could fly.” Great, that’s not a very robust alternative for me is it?

        Well, those are my baby steps. Steve Hays has had a few interesting musings on this issue of counterfactuals. I’ve linked to some of it on my blog.

        Thanks again.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: