Compatibilism and the Free Will Defense
November 10, 2007 — 8:00

Author: Jeremy Pierce  Category: Divine Providence Free Will Problem of Evil  Comments: 23

I’ve long suspected that the basic structure of Plantinga’s free will defense doesn’t require a libertarian view of free will, but I’ve never gotten around to trying to figure out in detail why that might be so. Well, Andrew Fulford has a proposal. Relying on the notion of creaturely integrity, Andrew offers an account of why God’s options might be limited by how God himself may have intended a person’s compatibilist freedom to work itself out, and for all we know this may be true for every actual person. In other words, it may well be that transworld depravity of a very particular sort may be true. It’s possible for all we know that, for each actual person, there is no possible world in which that person does no wrong. There is the problem of dealing with non-actual people, but that’s where God’s choice to actualize people with a certain kind of creaturely integrity comes in. Perhaps it’s true that anyone with the right sort of creaturely integrity, that God would have good moral reasons for wanting to bestow on people, will be transworld depraved in the way Andrew imagines.

What’s interesting about this proposal is that objections to it seem to be the same sort that people might raise against Plantinga’s own libertarian version of transworld depravity or his use of it. If that’s right, then he’s used the basic structure of the free will defense without relying on libertarian freedom.

Comments:
  • Christian Lee

    J-
    “It’s possible for all we know that, for each actual person, there is no possible world in which that person does no wrong.”
    Would you accept the following interpretation of the above quote: for all we know, every person must have performed some act that was wrong, at some time in their life?
    If so, that seems false. We know that children die at all different ages and that its extremely likely that some child has died, or will die, just after acquiring personhood and before they have done wrong. Moreover, it is surely possible for their to be such a case, even if, as is very unlikley, there is no actual case.

    November 10, 2007 — 15:50
  • Christian, would you say the same thing to Plantinga? If so, then I rest my case (or rather Andrew’s case). If not, why not? The argument is that Plantinga’s argument could work just as easily with this view as it does with his libertarian view. I’m not sure how the issue you’re pointing out shows anything particularly problematic with this account that isn’t also an issue for Plantinga.

    November 10, 2007 — 15:55
  • “It’s possible for all we know that, for each actual person, there is no possible world in which that person does no wrong.”
    Plantinga does not say that there is no possible world in which each actualizable agent does something wrong. On the contrary, he affirms that there is a possible world in which no agent does anything wrong. He does say that, possibly, there is no feasible world in which this is so (where a feasible world is one that God could weakly actualize).
    One other contrast. Plantinga’s argument does not have the sorts of the burdens that Andrew’s argument seems to have. One burden on Andrew’s argument is–as Christian points up–there have been lots of actual beings that have died long before they could have done anything wrong. Andrew has to show that even those beings that seem to have died before doing something wrong, have in fact done something wrong. Plantinga does not have this burden. He can argue that it is at least possible that no created essence dies before he does something wrong. I don’t think there’s much chance of showing that every actual being is twd. It is comparatively more plausible to claim that, in some world or other, every creaturely essence is twd.

    November 10, 2007 — 17:09
  • Forgive me if I betray my ignorance in responding; I’m nowhere near as educated as the people who contribute to this blog.
    Mike:
    As I read Plantinga, he seems to be saying that it is possible that there is no possible world where, if God were to actualize it, every person would not do something wrong. I’m not sure how my defense is any different: it could be the case that, there is no possible world such that, if God gave all actual creatures integrity, God could prevent all evil from occuring if he actualized those creatures. I’m not sure what difference libertarian freedom makes in this hypothetical.
    I still fail to see how the second problem is any more burdensome for me than for Plantinga. I’m not sure why I could not formulate the (creaturely integrity defense) CID such that it is possible no creature with integrity would die before committing evil. Again, I don’t see what libertarian free will adds to this.
    I think the ultimate point behind my defense (besides defeating the problem of evil argument from a Calvinist point of view), is to show that all the defender really needs to show there is a limitation on God’s omnipotence in the extent to which he could prevent evil. Calvinists often talk as if God has the power to stop everything, but I think that is a bit unqualified, and thus I think Calvinists could still use the CID along the same lines libertarians used the FWD.

    November 10, 2007 — 17:40
  • Perhaps, to continue the discussion a bit, I could make another comment.
    I don’t think the argument that there could be some agents who exist without committing evil is necessarily a defeater (though, if it was, I don’t see how the CID would be any more open to the problem than the FWD). One could reformulate the idea to be: it might be the case that any agent who is morally responsible, if actualized and given integrity/LFW, will commit some evil that God cannot prevent. This could avoid the apparent problem with infants, etc., at least as far as I can see. There might be other possibilities as well. Then one could reformulate the defense: if God wanted to give creatures both moral responsibility and integrity, perhaps he could not do so without either allowing evil or violating integrity/LFW.

    November 10, 2007 — 19:02
  • As I read Plantinga, he seems to be saying that it is possible that there is no possible world where, if God were to actualize it, every person would not do something wrong.
    No, he is definitely not saying that. He’s explicit that there is a world W in which creatures do no wrong at all. The FWD is designed to show that there is a world W’ from which W is not weakly actualizable. The actualization of W from W’ depends on the cooperation of liberatarian free agents. But since they are all twd in W’, W is not feasible from W’. That’s the basic story.
    I’m sure I’m don’t get entirely your CID defense. I went to your blog, but I doubt I’m reading it right. So, let me ask a question. Suppose, first, there is the world W in which no agent goes wrong. Suppose, second, that God’s causing agents to act morally is consistent with their freedom and integrity. In that case we can have integrity + no wrongdoing in the actualization of W. Why not actualize W? If the answer is that “actualizing W is inconsistent with God’s intentions”, the natural question is why God would intend anything other than actualizing W? . But I’m probably not getting entirely your CID.

    November 10, 2007 — 21:45
  • The actualization of W from W’ depends on the cooperation of liberatarian free agents. But since they are all twd in W’, W is not feasible from W’.
    Which, I think, is basically what I’m saying: it is possible that God cannot actualize W because all agents are twd in W’, and (possibly) they are twd because to make them otherwise would be to violate their creaturely integrity (i.e., God would have to contradict what he already planned about their history in order to prevent them from doing evil).
    If the answer is that “actualizing W is inconsistent with God’s intentions”, the natural question is why God would intend anything other than actualizing W?
    Perhaps because actualizing W would mean creatures have no integrity: they do opposite and contradictory (in the strict sense of doing different things in the same sense at the same time) things, they have no stable historical identity. This is perhaps to qualify your second supposition: it is perhaps the case that even within a compatibilist framework, God could not cause some people to do right at some times in a manner consistent with their integrity. (Just as Plantinga would modify it to say that within a libertarian framework there are some cases where God could not prevent people from doing evil without violating their free will.)
    If you further ask “Why is it that their integrity is not consistent with W being actualized?”, I think that would be the same as asking Plantinga “Why is it that actualizing W is not consistent with libertarian free will?” It is not necessary that libertarian free will result in evil choices, so the only response a Plantingan would have would be to say “perhaps it just happens to be the case.” I’m not really claiming anything more than that.

    November 10, 2007 — 23:44
  • Perhaps it would help for me to state my argument a different way.
    As I understand it, the FWD defense is set up to respond to the classic problem of evil argument against God’s existence.
    This argument is roughly as follows:
    1. God could do all things.
    2. God would want to prevent all evil.
    3. There is evil in the world.
    4. Either God can’t do all things or does not want to prevent all evil; in either case, 1 or 2 is false (and Christian theism is false with it).
    Plantinga’s argument seems to be trying to modify 1. He does this by pointing out that:
    1a. God cannot causally prevent someone who is libertarianly free from doing evil.
    He takes this limitation on God’s omnipotence and then comes up with a hypothetical possibility, which is essentially his defense (his possible reason why God allowed evil):
    5. Perhaps it is the case that God could not prevent evil from occurring in the actual world without at some point violating an agent’s libertarian free will. (Or, in other words, perhaps it is the case that transworld depravity exists.)
    Thus he comes up with a scenario in which, while God would prevent any evil if he could, he cannot actually do so. Hence the problem of evil fails because it wrongly defines 1.
    My argument basically switches 1a and 5 with the following:
    1a’. God cannot causally prevent someone from doing evil whom he also planned would do evil.
    5′. Perhaps it is the case that God could not prevent evil from occurring in the actual world without at some point violating an agent’s historical integrity. (Or, in other words, perhaps it is the case that transworld depravity exists.)
    It seems to me this is also a possible scenario in which, while God may want to prevent all evil, he cannot actually do so. Thus, once again, the problem of evil also fails because it falsely defines 1.
    The substantive difference between Plantinga’s argument and mine is that in Plantinga’s argument the source of God’s restriction is something built into the structure of the agent (their libertarian freedom), whereas in my argument the restriction is entirely in historical possibilities (their inability to do the opposite thing at the same time). But in both cases it is not necessary that the feature of reality that provides the restriction (LFW or CI) actually do so. It just happens to be the case. For Plantinga this is because LFW choices, qua LFW choices, cannot necessarily do anything; necessity is the opposite of libertarian freedom. For me it is because it is not clear that creaturely integrity must mean that God could not prevent them all from doing evil. But, because of the nature of a defense vs. a theodicy, it is not necessary for either of us to prove that this aspect of reality (LFW/CI) must restrict God’s omnipotence. It is enough to show that it could do so, and that this might be the reason why God could not prevent evil.
    Hopefully I have cleared something up here. I’m pretty sure the reason people are having difficulty understanding my position is because I’m not as careful with my words as an analytic philosopher should be; but this is just because I’m not educated as much in that field. I apologize for the confusion this generates.

    November 11, 2007 — 1:39
  • As far as I can see, your view is this: It’s possible that, for each creatable agent A and each strongly actualizable states of affairs T in which A can be created, God plans for A to go wrong in T and arranges the causal order so that A is caused to go wrong in T. But this is just to say that it is possible for God to plan and strongly actualize the wrongdoing of agents.
    There are at least two problems.
    1. God’s plans are freely chosen, so IF He planned such a world, He could, before any agent is created, change His plans for agents and see to it that they never go wrong. After all, God’s plans are up to God. If He were to change them and see to it that no agent ever goes wrong, there would be no violation of integrity. The plans are changed long before there are any existing agents.
    2. There is an obvious and powerful atheistic response: God simply could not plan for agents to go wrong any more than He could himself cause evil. So, even if (1a’) were true, it is impossible for God to freely plan that any agent would do evil. There is no such world in which, as a matter of contingent fact, God freely plans for agents to do evil (and then, on top of that, strongly actualizes the wrongdoing!)

    November 11, 2007 — 11:31
  • As far as I can see, your view is this:
    That is just part of my view; in this paragraph you have just summarized compatibilism, which I hold to. But that is not all I’m saying
    1. God’s plans are freely chosen, so IF He planned such a world, He could, before any agent is created, change His plans for agents and see to it that they never go wrong.
    This is an assertion I’m trying to question; perhaps there are some worlds (ones where creatures have integrity) where God could not strongly actualize them and the creatures would not choose to do evil. Perhaps God’s plans are limited in the sense that God can only create a world where agents will go wrong, and his commitment to their creaturely integrity means he cannot prevent them from doing evil. God’s plans are up to God, but that doesn’t mean he can plan just anything. There are logical limits (furnished by his nature, his other plans, etc.) to what he can do.
    2. There is an obvious and powerful atheistic response: God simply could not plan for agents to go wrong any more than He could himself cause evil.
    Why should I believe this is “simply” not possible? Perhaps it is. Even given compatibilism, there are other possible reasons why God may not be able to do this. Maybe out of all the best possible worlds (there may well be many equally good best possible worlds), none of them involves a plan where no evil occurs (this seems to be Plantinga’s response anyway).
    And, as a final note, I’m pretty sure the same problems could be raised against Plantinga’s FWD. The only difference in those scenarios between myself and Plantinga is strong vs. weak actualization. But your objections don’t require either version to do what they’re meant to do.

    November 11, 2007 — 12:55
  • Mike:
    For what it’s worth, too, it seems that the Bible explicitly contradicts the atheist’s maxim in 2.
    “Joseph said to them, ‘Do not be afraid, for am I in the place of God? But as for you, you meant evil against me; but God meant it for good, in order to bring it about as it is this day, to save many people alive.” (Gen 50:19-20)
    So the biblical theist is committed by divine revelation to contradict the atheist’s supposition. I also should point out that not even an open theist can avoid this problem, if he or she wishes to be biblical.

    November 11, 2007 — 13:34
  • Sorry to multiply posts needlessly, but I realized my post at 12:55 was wrongly worded.
    In the second last paragraph I said “Even given compatibilism, there are other possible reasons why God may not be able to do this.” I should have said “there might be reasons why God would have to do this.” Compatibilism isn’t even really an issue in that objection. It’s whether a greater good can justify God’s intending evil actions would occur (in any sense of the word “intend”).

    November 11, 2007 — 13:38
  • I thought you said that the loss of integrity arose in any case where God PLANNED for agents to behave one way, and thereafter acted contrary to that plan. Isn’t that the source of the loss of integrity, viz., God’s interrupting his own plan for agents and causing them to act in an unplanned way? My (initial) objection was that this plan was freely chosen by God. And since that is so, God could have changed that plan long before there were any agents in existence and avoided any integrity violation. Certainly he could have done that, since his plans are up to him. I don’t see how you’ve responded to this objection.

    November 11, 2007 — 13:49
  • Andrew,
    If you want to argue that, possibly, God’s plans are not freely chosen, then we need some explanation why. The reason cannot be that there would be some loss in integrity, since choosing a plan on which no one goes wrong does not entail a loss of integrity (recall he could choose such a plan before any agents exist). So there has to be some other explanation. This is where all of the work comes in, it seems to me.

    November 11, 2007 — 14:03
  • And since that is so, God could have changed that plan long before there were any agents in existence and avoided any integrity violation.
    But it’s not obvious that God could have planned a world in which twd doesn’t exist. It’s possible that the only plans God could have made would have included twd. The integrity issue comes up after the fact, when the world with twd exists, and explains why God could not have changed it even given compatibilism. (This is what libertarianism does for Plantinga: twd gives God no option but to make a world where people choose to do evil, and once it exists God cannot prevent evil because to do so would be to violate LFW.)
    The reason cannot be that there would be some loss in integrity, since choosing a plan on which no one goes wrong does not entail a loss of integrity (recall he could choose such a plan before any agents exist).
    But why should I accept that such a possible world would be actualizable? Perhaps it is the case that there is no plan where no one goes wrong and they also do not lose integrity. Transworld depravity would explain why there is no plan where no one goes wrong, and the goodness of creaturely integrity would explain why God cannot prevent evil once the world with transworld depravity has been actualized.
    Further, I don’t think it’s fair to raise this objection against me, and not against Plantinga. There is no obvious reason why weakly actualizing a plan where no one goes wrong should HAVE TO violate libertarian freedom. Why couldn’t God actualize such a world? There’s nothing that makes it impossible; it’s just the nature of libertarian freedom that there could be no answer to this question.

    November 11, 2007 — 14:25
  • Andrew, my guess is that we are simply at an impasse. I re-read your post where you say,
    there could some scenario in which, because of something God has ordained about the course of A’s life, God could not also ordain that A be protected from sin.
    But what God ordains in the first place is up to God. How could it not be up to God? And if it is up to God–if God freely ordains–then he could have ordained instead that no one goes wrong. In doing so there would be no loss of integrity; or not as far as I can see, anyway. In any event, I think we are both repeating things we have said already. I just don’t see us making much progress on this particular issue.

    November 11, 2007 — 17:51
  • “It’s possible for all we know that, for each actual person, there is no possible world in which that person does no wrong.”
    What about these two objections (neither of which is fully convincing to me)?
    1. There is a possible world where I do no wrong because God gives me a blissful life where I can make no morally significant choices.
    2. Suppose we escape (1) by qualifying the claim as follows: “It’s possible for all we know that, for each actual person, there is no possible world in which that person makes a significant choice but does no wrong.” This seems to imply that everybody’s first significant choice was wrong. For suppose that Bob’s first significant choice was right. Then God could have prevented Bob from making any further wrong choices by giving him a blissful life where he could do no wrong. Is it plausible to suppose that everybody’s first significant choice was wrong? Moreover, it implies each of us has the essential property of being such that her first significant choice was wrong. But surely we could have been snatched at conception, transported somewhere else, etc., in a way that ensured our first choice would not be wrong. Would that harm our historical integrity? I doubt it–there is not much historical integrity there to be harmed at an age before we make our first significant choice.

    November 13, 2007 — 15:44
  • Alex, we might need to distinguish between pre-fall and post-fall. I think it’s a plausible view that post-all everyone’s significant choices are tainted by sin such that no one’s choices are not on some level sinful. But what matter is whether this is true pre-fall, and obviously it can’t be true of anyone who didn’t sin before the fall. So that does seem to be a problem.

    November 13, 2007 — 15:52
  • Hmm… I guess I don’t see the problem.
    As long as one of our significant choices would be evil, does it make a difference if it was the first one, in terms of providence? I mean, if everyone at some point in their life would choose to do evil, would that not be twd, regardless of whether it was their first choice?

    November 16, 2007 — 0:44
  • I think it’s worth remembering one thing in this discussion (at least for my benefit, to clarify my argument), since we are now moving on to issues of what seems “plausible” or “likely” in possible worlds.
    When I read Plantinga’s arguments, I see lots of things like the following (bold is my emphasis, italics is Plantinga’s):
    “Now suppose we assume that Curley was free with respect to the action of taking the bribe-free to take it and free to refuse. And suppose, further, that he would have taken it.”
    “Of course, there are possible worlds in which he is significantly free (i.e., free with respect to a morally significant action) and never does what is wrong. But the sad truth about Curley may be this.”
    “Obviously it is possible that there be persons who suffer from transworld depravity. More generally, it is possible that everybody suffers from it.”
    The common theme here, to my understanding, is that Plantinga has not demonstrated any of these things is necessary. It is speculation piled on top of speculation.
    But that is fine within the parameters of what he has set out to do. Plantinga is only trying to come up with a defeater to an argument that depends on the premise that it would be impossible for God not to be able to stop all evil, since he is “almighty”. The atheist argument, ultimately, has to demonstrate a universal negative, and thus Plantinga’s speculations serve as a helpful response: “but look, I can even imagine some logically possible scenarios in which that is not true. So your universal negative is unprovable.”
    Most of the responses to my defense here, so far, seem to have been along the lines of attacking it’s plausibility or necessity. But: a) Plantinga’s defense does not require that it necessarily be the case (it might not be the reason God allowed evil, Plantinga himself makes this statement), b)plasubility is a relative, continuum like thing. Perhaps my argument is less plausible than Plantinga’s, or even not as plausible (to some people) as atheism. But that’s really irrelevant to the original purpose of the argument, at least as I understood it: it is possible, and thus the atheist can’t assume or demonstrate a universal negative.
    Correct me if I’m way off here…

    November 16, 2007 — 13:45
  • Andrew, the reason the first act matters is because (presumably) there’s a world in which the person just does that first act and doesn’t live long enough to commit the second one. So if it’s possible for the first act to be good, then there’s a possible world where the person does nothing but good. Since the claim you’re making is that maybe for each actual person there’s no world in which the person does only good things, it follows that even the first act is not good or that the first act can’t take place without some evil act taking place immediately afterward of necessity.
    If you had something different in mind, I’d be curious to hear it, but that’s how I’m reading the claim you’re making.

    November 16, 2007 — 14:02
  • Since the claim you’re making is that maybe for each actual person there’s no world in which the person does only good things, it follows that even the first act is not good or that the first act can’t take place without some evil act taking place immediately afterward of necessity.
    Perhaps the latter is the case: perhaps there is no world in which someone will make one good choice and then never make an evil choice later. That doesn’t seem to me to be an impossible scenario.
    And just so I’m clear, is there any reason this criticism wouldn’t also apply to Plantinga’s argument? I don’t see how lib. freedom would change anything about this objection… in which case we’re just objecting to any defense…

    November 16, 2007 — 15:29
  • Another thought occured to me that is relevant to defenses from twd:
    All Christian theists (including libertarians) are required to posit not just the possibility, but the actuality of one person living a morally significant life who never does evil: Jesus.
    I think what is really important about Plantinga’s defense is not that everyone have twd, but that if God were to create all actual people (with lib. freedom/creaturely integrity), at least one person would eventually do evil. That’s all that needs to be supposed to counter the problem of evil, and certainly I think it is not only possible, but even plausible (not that the degree of plausibility makes a difference when the argument is about what is possible, as far as I understand).

    November 16, 2007 — 16:03