A Question about Molinism
February 20, 2006 — 9:59

Author: Ted Poston  Category: Divine Foreknowledge  Comments: 26

I’ve been thinking about Molinism of late and I’m puzzled by the claim that the counterfactuals of creaturely freedom (CCFs) are prevolitional (logically prior to God’s creative activity), contingent, and within my power. Consider the claim that the CCFs are contingent. What is the relation between the CCFs re S’s actions and S’s individual essence? Not entailment. If individual essences are necessary (and I think they are) and if S’s individual essence strictly entails certain CCFs then the CCFs are necessary. Suppose, instead, that the CCFs regarding S are made true by certain actions S performs. This conflicts with the prevolitionality condition. The CCFs are true prior to God’s creative activity; ergo prior to S being on the scene. I don’t see how S’s actions can be the truth-makers for CCFs given that they are logically prior to God’s creative activity. Truth-makers need to exist in order to make true something but I don’t exist prior to God’s creative activity; so my actions aren’t the truth-makers for CCFs about me. Another option is to view the CCFs as brute. They are true in a similar way that possibility claims are true. Problem: possibility claims are necessary, at least for an S5er like myself. So the CCFs need to be brute and contingent. Problem: the CCFs should have some relation to (i) my individual essence and (ii) my choices. If they are entirely brute then the truth of the CCFs is not explicable in terms of (i) or (ii), in which case I begin to lose the intuition that they are within my power.

Comments:
  • “The CCFs are true prior to God’s creative activity; ergo prior to S being on the scene. I don’t see how S’s actions can be the truth-makers for CCFs given that they are logically prior to God’s creative activity. Truth-makers need to exist in order to make true something but I don’t exist prior to God’s creative activity; so my actions aren’t the truth-makers for CCFs about me”.
    Let w be a world in which the following CCF is true,
    P. Had you been offered a bribe at age 15, you
    would not have taken it.
    What makes that true at w are facts in another world w’. Since you believe that there are individual essences, I don’t see offhand why you would have to exist in w for P to be true there. If you don’t exist in w then (P) says something like this: the closest world in which your individual essence is instantiated and offered a bribe at 15, is a world in which that instantiated essence does not take it.
    So you need not exist in order for your actions to make (P) true. You need only *possibly exist*.

    February 20, 2006 — 10:35
  • Kevin Timpe

    There’s lots to say here, but just a quick comment before I have to run to class.
    You write: “I don’t see how S’s actions can be the truth-makers for CCFs given that they are logically prior to God’s creative activity.” That seems exactly right. And it also is what Flint says:
    “Who or what actually caused the ones [i.e., the CCFs] that are true to be true, and the ones that are false to be false? In whose actual activity are we to find adequate metaphysical grounds for such truths?… Could God be the one who by himself causes their truth? Clearly not…. Cout it be, then, that we free creatures are the ones who, by the actions we perform in the actual world, make the relevant counterfactuals true…? Here, too, a negative answer seems to be required, and that for (at least) four reasons” (Divine Providence, 123).
    Though I don’t have time now to say much further, Flint seems to say (from what I can remember) that the truth-values of the CCFs are “basic, primitive, ungrounded facts about the way things are” (137). For me, at least, this is quite a high price to play, particularly since it seems to open Molinism up to a consequence-style argument that no agents have libertarian freedom. But off to teach about the Forms.

    February 20, 2006 — 10:55
  • Mike,
    I’m not sure I’m following you. Take
    P*: Had I been offered a bribe at age 15 I would have freely refused it.
    (Would that that were true!)
    Suppose P* is true at some non-actual world w*. So far this is just stipulation; so the question is whether the following three things are true: P* is true in w*, P* is within my control, and P* is true prior to God’s creative activity. You seem to suggest that my existence in alpha can make P* true. I agree, but in this case P* isn’t true prior to God’s creative activity. Am I missing something?
    Kevin,
    Thanks for the Flint references. Do you know if Flint addresses the issue of whether the CCFs are within my power? Flint might attempt to resist the consequence-style argument by insisting that the CCFs are contingent but when this is combined with the bruteness thesis I completely lose the intuition that these are within my power.

    February 20, 2006 — 11:28
  • Kevin Timpe

    Ted,
    The only place that I can think of off the top of my head is the 3 chapters in “Divine Providence” that deal with the grounding objection and related issues (there may also be something related in the Thomistic objections chapter). Actually, Tom has a chapter in “Middle Knowledge: Theory and Applications”, though I can’t remember if this is a new piece, or part of his book. Regardless, there are a number of papers in that book that also address these issues as well. And I’m pretty sure that Plantinga says something about this issue in the “Profiles” volume on him.

    February 20, 2006 — 13:34
  • “You seem to suggest that my existence in alpha can make P* true. I agree, but in this case P* isn’t true prior to God’s creative activity. Am I missing something?”
    Maybe I’m missing something. How does God’s creative activity come into play relative to that particular counterfactual? You don’t exist, you possibly exist. God has not created you, he has possibly created you. So where again is God’s creative activity in this?

    February 20, 2006 — 14:19
  • Ted,
    I think you nicely summarize the difficulties Molinism faces vis-a-vis the grounding objection.
    The Molinist account of divine providence needs CFC’s to satisfy the following contraints: (1) some CFC’s are true and known by God prior to God’s creative decree, (2) CFC’s are logically contingent (b/c otherwise they’d be part of God’s natural knowledge, not his middle knowledge),j and (3) CFC’s are counterfactuals of libertarian freedom.
    Given (1)-(3) and the standard theistic view that apart from God’s creative decree only God and aspects of God exist, CFC’s must be contingent aspects of God. But then how could they be contingent?
    They can’t be contingent on de facto creaturely choices b/c then they wouldn’t be providentially available to God. They can’t be contingent on God’s volitions b/c then we would have theological determinism, not Molinism. Nor do I think they can be dependent on creaturely individual essences b/c (a) what is essential is necessary, hence individual essences could at best ground counterfactuals of compatibilistic freedom, and (b) as abstract objects, individual essences exist in all possible worlds and so cannot ground contingent CFC’s.
    I could ramble on, but suffice to say, I think the grounding problem is death to Molinism. The best the Molinist can do, I think, is to follow Flint in taking CFC’s to be brute, ungrounded, truthmakerless truths. As Kevin pointed out, that’s not a very attractive option.

    February 20, 2006 — 14:47
  • That’s right. I was taking your point in a different direction. In any case, I don’t think this helps the Molinist. If you ground the truth of the CCFs in my individual essence then it’s hard to see how they are not necessary. In which case, it’s hard to see how they are counterfactual of creaturely *freedom*, at least on a libertarian conception of freedom.

    February 20, 2006 — 15:13
  • Alan,
    Thanks for that. I was thinking this objection was slightly different than the grounding objection but if it comes to the same thing that’s fine. I was considering that this might develop into an incompatibility argument.

    February 20, 2006 — 15:24
  • Ted,
    I haven’t looked at the Flint stuff in a while, so I’m not sure I see why the grounding question is so pressing. Maybe I can be disabused of this. But here’s why I don’t find it so worrisome. Take any events e and e’ that are possible results of an indeterministic process P. Assume that the occurrence of e is much less probable than e’. It happens all of the time that P occurs and the less probable event e follows. Everyone admits that we have no contrastive explanation for e’s occurrence. That is, everyone admits that we have no explanation for why e occurred rather than the more probable event e’. Yet, of course, e occurs in the closest world at which P occurs. Since I don’t think any world is as close as our world to our world, I don’t find it so difficult to say that the counterfactual P []-> e is true here. Of course, I didn’t at all expect e on P. But counterfactuals don’t in general track conditional probablities. In any case we have, a little closer to home, a true ungrounded counterfactual here. Certainly the more grounded counterfactual seems to be P []-> e’. But then, so much for grounding.

    February 21, 2006 — 13:35
  • Mike,
    The problem I’m getting at isn’t exactly the same as the grounding problem. The problem is that three theses Molinists are committed to appear incompatible (or at least cause me puzzlement). I’m fine with some ungrounded counterfactuals. I just don’t see how the counterfactuals of creaturely freedom can be ungrounded & yet genuine counterfactuals of libertarian freedom.
    All,
    What do you think of the general claim that brute, ungrounded truths aren’t within anyone’s control (or at least any non-divine being)? If this general claim holds then the incompatibility argument is pretty straightforward.

    February 21, 2006 — 17:42
  • “What do you think of the general claim that brute, ungrounded truths aren’t within anyone’s control (or at least any non-divine being)?”
    I would agree with that claim. If the truth of proposition P were in person S’s control, then there would have to be some action A that S could perform that would make P true. The action would presumably bring about a state of affairs that, by virtue of making P true, would be a truthmaker for P. Hence P would not be ungrounded.

    February 21, 2006 — 19:34
  • Kevin Timpe

    Can someone give me an example of a “brute, ungrounded truth”? I’m having a hard time here, so perhaps an example would help. However, it seems clear to me that if truth T1 were brute and ungrounded, then it would be outside of anyone’s control, including any potential divine being’s control. If some other truth, T2, is within a divine being’s control, than T2’s truth is grounded in or made true by that being’s volition, wouldn’t it? (Maybe these 2 questions just go to show that I’m not sure I understand brute, ungrounded truths!)

    February 21, 2006 — 23:45
  • Kevin,
    I think you have to have non-Leibnizian intuitions to think there are brute, ungrounded truths. Russell seemed to think that the existence of the universe was a brute, ungrounded fact. It has no explanation and there’s no metaphysically prior state to the existence of the universe.
    Jon Kvanvig has written on your second claim: “If some other truth, T2, is within a divine being’s control, then T2’s truth is grounded in or made true by that being’s volition.” (See his article on Maverick Molinism.) As I recall, he thinks that some CCFs are not grounded in God’s volitions but that he could act in such a way to make them false.

    February 22, 2006 — 8:26
  • Kevin Timpe

    Ted,
    Yeah, perhaps I’m just too Leibnizian (wow, never thought I’d say that!) here. Lately I’ve been reading a lot on truthmaker theory, which I’m greatly inclined toward; that probably also explains why I’m not too sure of ungrounded, contingent truths.
    It’s been a while sense I’ve read “Maverick Molinism,” so maybe I need to go back and look at it more. But initially, the claim you suggest (or suggest that Kvanvig suggests) strikes me as rather puzzling. That’s not to say it’s not true, but just that I need to go back and be convinced that it’s true.

    February 22, 2006 — 10:01
  • Blake

    I wanted to raise a related issue that has always puzzled me about Molinism. Does God know the so-called “counterfactuals of freedom” that apply to his own actions? The reason I ask is that no matter how we answer, it appears to Middle Knowledge is impossible.
    Suppose that it is simply God’s nature to know all counterfactuals as Craig and Flint seem to maintain. It seems that God would then have to know the counterfactuals that describe God’s own choices and actions. Doesn’t it then follow that God knows that he is in a possible world W, and given that he is within that world he knows also what he would do if any given cirucmstances? So God knows that in W he will freely choose to do A. But then it follows that middle knowledge collapes into simple foreknowledge because God knows what he will do in W (logically or chronologically) prior to his decision to do it. So because God knows the counterfactuals that describe his own “choices”, he already knows what he will freely decide to do given the possible world in which he is located before he can use this middle knowledge to guide his actions.
    Consider the counterfactuals that apply to God’s own actions: in W God knows that “if in possible world W, God would create possible world PW.” But God also knows already that he is in W. It follows that God knows that he will create PW. So middle knowledge collapses into simple foreknowledge.
    On the other hand, if God does not have middle knowledge of his own actions, then he has only natural knowledge of what he will do and God is in the same position as the God of open theism who knows possibilities but not what he will do given those possibilities until he decides.
    There is a very large issue here as to what God’s relation is to the various possible worlds he can survery. However, it seems fairly intuitive that only one possible world can be actual and that God must be located in that possible world. So God must be within a possible world and not transcend them all. But given that God is located within one actual world among all of the possible worlds, it seems that any view that could motivate one to beleive that God must have middle knowledge of the ations of the various creatures that God could create also motivates the view that God must have middle knowledge of his own acts. But if he does, it seems to middle knowledge collapses into simple foreknowledge.
    Am I missing something here?

    February 25, 2006 — 12:43
  • Blake,
    How’s the argument supposed to go that middle knowledge collapses into simple knowledge? You seem to appeal to the premise that if God knows that were C actual he would do A then C is actual. Also, on the possible worlds stuff, it’s fairly standard to think that an omnicompetent being is a delimiter of metaphysical possibility. But there’s still a clear sense in which there could have been different worlds and that God actualizes one of those worlds.

    February 26, 2006 — 9:10
  • Blake

    Ted: What does it mean to be a delimiter of metaphysical possibilities when I am asking whether God has middle knowledge of his own actions? Either God does or he doesn’t have middle knowledge of his own actions. In order to have middle knowledge, God wouldn’t have to be outside any role of delimiting possibilities anymore than we have to be in order to yet have forking paths in our futures.
    Look at it this way. God isn’t actual in all possible worlds; he is actual in only one possible world though the possibility of his being actual ranges over all feasible worlds open to him to create. (If God’s existence is logically necessary then he is possibly located in all possible worlds — tho just what we do with worlds that contain overwhelming amounts of evil then becomes problematic). Nevertheless, the possibilities that are open to God are indexed according to a time or a logical standpoint of openness. So it is analogous to the way I see my own future — I am actual in only one possible world but I could have been in others had I made different choices, and there are yet possible worlds open to me to create.
    However, there is a future realm of open possibilities at any given moment in this actual world and God knows the CFFs that define all possible truths about the actual world in which God is located. So it isn’t absurd to ask what is actual prior to God’s actions — because God is actual. However, if in order to be God it follows that God must know all CFFs, it would seem that God must know the CFFs that are supposedly true statements about God’s own free acts.
    So the premise is not “if C were actual then God would do A, then C is actual” (I never said anything like that). What I suggest is that God knows the CFFs about his own actions. God knows that in the actual world where he is located (the one of the many possible worlds in which God is logically possible) that there are innumerable possible worlds that he can actualize. However, they must all be compossible with God’s (logically or chronologically) prior existence. God can actualize C or not. However, God also knows the CFFs that are feasible given the actuality of the possible world in which he is located. He is free to either strongly actualize C or not. However, God knows that given his actuality in possible world W, he will freely bring about C rather than not-C. However, what now follows is that given that God is actual in a particular possible world W, and in W there is a true CFF that says that if in W God will freely strongly actualize C, then by God’s knowing that he is in W as opposed to other logically possible worlds that he will strongly actualize C. But then God doesn’t have middle knowledge because knowing which of the possible worlds he is located in entails that God will freely choose to strongly actualize a particular possible world in virtue of his knowledge of CFFs.
    I’m exploring why this isn’t an accurate description of God’s relation to possible worlds and how, if it is, there could be middle knowledge.

    February 26, 2006 — 12:31
  • Blake

    Ted: Btw, my take on God being actual in only one of the possible worlds is Plantinga’s view as opposed to David Lewis’s theory of possible worlds. So the issue seems to be well worth exploring — especially if Middle Knowledge turns out to be impossible given Plantinga’s take on God’s relation to the universe of possible worlds.

    February 27, 2006 — 11:39
  • Steven Carr

    If Molinism is the doctrine that people will choose one definite way , given any logically possible scenario, isn’t it trivially true?
    Take two different sets of circumstances that I can conceive of.
    1) I am sitting down to breakfast in an hotel at 8:30 am on Wed. 2/11/2005, and a waiter is asking me ‘Tea or Coffee’, and God has infallible knowledge that I will choose tea.
    2) I am sitting down to breakfast in an hotel at 8:30 am on Wed. 2/11/2005, and a waiter is asking me ‘Tea or Coffee’, and God has infallible knowledge that I will choose coffee.
    Clearly, I can conceive of both sets of circumstances, and they are both logically possible, and they are clearly different to each other.
    It is trivially true that I will freely choose one particular way in each of those different sets of circumstances.
    But Molinism appears to be the doctrine that there there is a fact about how I will choose in logically IMpossible circumstances ie in circumstances where God does not know how I will choose.
    How can there be a fact about what will happen in circumstances which, by the definition of omniscience, are logically impossible?

    February 28, 2006 — 0:09
  • Kevin Timpe

    It isn’t trivially true that while there are possible worlds in which an agent S does action A in situation C and there are possible worlds in which agent S does ~A in situation C, the following counterfactual is true: if in C, S will freely do A.
    Why do you think that “Molinism appears to be the doctrine that there there is a fact about how I will choose in logically impossible circumstances, i.e. in circumstances where God does not know how I will choose”? It is possible that S does A in C, and possible that S does ~A in C; it just happens to be (contingently) true that S will do A if placed in C. There can’t be facts about what will happen in circumstances which are logically impossible (unless you embrace some odd view about logical impossibility)–but Molinism doesn’t require that there are (unless you think that Molinism itself is logically impossible, but that doesn’t seem to be your point here).

    February 28, 2006 — 22:50
  • Let us take a situation ‘C’ (C for coffee, perhaps)
    1) I am sitting down to breakfast in an hotel at 8:30 am on Wed. 2/11/2005, and a waiter is asking me ‘Tea or Coffee’, and God has infallible knowledge that I will freely choose coffee.
    Surely it is trivially true that in situation C, I will freely choose one particular way. Obviously I will freely choose coffee.
    All we have to do to prove Molinism is true is to check the following.
    1) Omniscience. Clearly the situation C is compatible with omniscience
    2) Libertarian free will. Clearly God’s infallible knowledge is not causing me to freely choose coffee. That would be an incoherent objection. So situation C is compatible with libertarian free will.
    So it is trivially true, that if you describe all the circumstances in which I freely choose between tea and coffee, then it is a fact that I will choose one particular way.
    Of course, the following set of circumstances is also logically possible
    2) I am sitting down to breakfast in an hotel at 8:30 am on Wed. 2/11/2005, and a waiter is asking me ‘Tea or Coffee’, and God has infallible knowledge that I will freely choose tea.
    They just happen to differ from set of circumstances 1. In set 2, I freely choose tea, while in set 1, I freely chose coffee, but I cannot see how either set or either choice contradicts any definitions of omniscience and libertarian free will.
    So it appears to me that Molinism is true.
    Of course, you cannot have the *identical* set of circumstances in two *different* possible worlds, for the obvious reason that the two supposedly identical sets of circumstances differ in at least one factor – namely God’s infallible knowledge of what is going to happen in the two worlds.

    March 1, 2006 — 0:10
  • Kevin Timpe

    Now I think I see what is going on. Molinism, as traditionally understood, involves God having knowledge of the truth-values of CCFs logically prior to His creating anything, and thus *before* you ever exist in C. It is precisely that middle knowledge is pre-volitional that allows God to use it providentially. So, on the Molinist’s view, it is middle knowledge plus God’s initial creative act that gives Him foreknowledge, and not foreknowledge that gives Him middle knowledge. If you get a chance, take a look at chapter 2 of Tom Flint’s “Divine Providence: The Molinist Account” for a very thorough presentation of Molinism.

    March 1, 2006 — 10:47
  • Heath White

    Stephen,
    Others may have different objections, but for me, the problem is “…and God has infallible knowledge that I will freely choose X.” This assumes that there is no problem of freewill and divine foreknowledge; but there is a problem, to which Molinism is one alleged solution; so we cannot assume, in defending the coherence of that solution, that there is no problem.

    March 1, 2006 — 11:11
  • Heath White

    This thread seems to have turned into a general discussion of Molinism, so I have an argument I’d like to try out on everyone. Here’s the gist.
    Molinism includes the view that God infallibly knows counterfactuals of the form, “If it were the case that P, it would be the case that S freely chooses A.”
    The freedom involved is supposed to be libertarian freedom, which is indeterministic. So any claim of the form “S freely chooses A” entails “It indeterministically occurs that S chooses A.”
    It is a piece of standard logic that “P []-> Q” and “[](Q->R)” entails “P []-> R”, where the []-> is counterfactual implication. So the two premises above entail that God infallibly knows counterfactuals of the form, “If P were the case, it would indeterministically occur that E”, where E is that S chooses A. And if God knows such counterfactuals, then they are true.
    But it seems to me that they are not true. Suppose that flipping coins and rolling dice are genuinely indeterministic events. Then there is no truth to either “if you flip that coin, it will come up heads” or “if you flip that coin, it will come up tails.” Even if the probabilities lean heavily one way, there is no truth to either “If you roll that die, it will come up 6” or “If you roll that die, it will not come up 6.” The reason there is no truth to either counterfactual is simply that the worlds in which the coin comes up heads/tails, or in which the die comes up 1/2/3/4/5/6, are equally close. In general, there are no true counterfactuals with indeterministic events in the consequent.
    Stop here for a minute. If that argument works, then there are no true counterfactuals of indeterministic events, hence no true counterfactuals of libertarian free actions, hence no knowledge of such counterfactuals. Hence Molinism is false.
    Objection: Yes there are true counterfactuals of indeterministic events. They are the ones whose consequents would be true in the would-be actual world. Thus, if you *do* flip the coin, and it *does* come up heads, then it was true that “if you were to flip that coin, it would come up heads”.
    Reply: I am not sure the needed notion of would-be actual worlds makes any sense (what’s the truth value if you don’t flip the coin?), but let’s run with it. There are multiple worlds in which P (flipping the coin, or rolling the die, plus the whole history of the world up til that point) obtains. Some of those worlds have one outcome and some have another, by the definition of indeterminism. God knows a counterfactual of the form, “If P were the case, then it would indeterministically occur that E” by stipulation from the objection.
    Thus, he must know which world is the would-be actual world, that is, he knows which P-world he is referring to when he thinks of this counterfactual. Thus the true form of the counterfactual is something like, “If P (where w1 is actual) were the case, then it would indeterministically occur that E.”
    But w1 essentially includes E. So the counterfactual is entailed by, “Necessarily, if w1 is actual, then E occurs.” But this is a piece of God’s natural knowledge, and so God’s knowledge of counterfactuals of indeterministic events is natural knowledge, not middle knowledge. This goes for those indeterministic events which are libertarian free actions as well. Thus, even granting the coherence of the objection, Molinism is still false.
    ****
    OK, there’s my argument. I’d appreciate any thoughts from commenters.

    March 1, 2006 — 11:38
  • Steven Carr

    Indeterministic events.
    It might be the case that 2 worlds are identical up to the toss of coin and then branch therefater (assuming coin tosses are indeterministic).
    It might be that the 2 worlds are identical, except that they differ in one respect. God knows what the coin will do. (His knowledge does not cause the coin to do anything)
    Heath writes ‘There are multiple worlds in which P (flipping the coin, or rolling the die, plus the whole history of the world up til that point) obtains.’
    So, if God is omniscient, the whole history of the world up till that point is different in each world.
    The argument would work if any being, not just God, was omniscient. The present knowledge of what will happen would be enough to distinguish between worlds.

    March 1, 2006 — 12:14
  • Statistics for February

    Total Pages:   31,327 Total Visits:   16,631 Avg Pages per Day:   1118 Avg Visits per Day:   593   Most Popular Pages 1   The Problem of the Trinity   530 2   Four Versions of Open Theism   329 3   Chr…

    March 1, 2006 — 15:34