Free Will Defense and compatibilism
January 26, 2009 — 11:09

Author: Alexander Pruss  Category: Free Will Molinism Problem of Evil  Comments: 15

It seems very plausible that a good answer to the problem of evil will require some version of the Free Will Defense (FWD). If a FWD requires incompatibilism, then there is a very plausible argument from theism to incompatibilism.
But I think it may well be that a FWD does not require incompatibilism. First of all, a FWD does not need that freedom of will and responsibility be incompatible with determination by prior non-agential causes or by laws of nature. At most what we need for a FWD is that freedom be incompatible with total determination by prior agential causes (the case that matters is that of God’s creative act), a claim that I think some compatibilists will accept.
Second, even if freedom of will and responsibility are compatible with determination by divine agency, it does not follow that the FWD is completely out of steam. For it may be that certain kinds of good decisions depend on some of their value on something more than bare freedom of will and responsibility. For instance, for a promise to be valid, more is needed than that the object of the promise be good and that the promise be made with freedom of will and responsibility. A promise made at gunpoint is invalid, even if it is made responsibly and with freedom of will (one does, after all, have a free choice whether to utter the promise or to die, assuming one does not lose freedom and responsibility through panic, but this is not enough for validity).
Here would be one sketch of a FWD that is compatible with compatibilism (even compatiblism between freedom and responsibility, and determination by an agential cause): A love is of much greater value when the lover is not causally determined by the beloved to love the beloved. This claim is compatible with saying that the lover could freely and responsibly respond with love to the beloved even if determined to do so–for there is more that we want in a response to love than mere freedom and responsibility (e.g., someone with amazing powers of self-control could freely and responsibly respond with love to a threat, but that’s not the most valuable kind of loving response). But a failure to respond with love to God’s love is always an evil. But it might be that the only way God could ensure that there are agents all of whom respond with love to God’s love is by causally determining them to do so. (One way to argue for this is to suppose Molinism transworld unresponsiveness: In every feasible world in which agents are not determined by God to respond with love to his love, some agent fails to do so.) It might then be that God is justified in creating creatures some of whom fail to respond with love to his love.
But while this example shows that a FWD need not require the incompatibility between determination and freedom/responsibility, this FWD still requires the compatibility between freedom/responsibility and lack of determination–it requires the possibility of libertarian-type choices. (Hume thinks that freedom requires determination. Fischer, on the other hand, is an even-handed compatibilist–freedom is compatible with determination adnw ith lack thereof.)

Comments:
  • Jordan Woods

    “First of all, a FWD does not need that freedom of will and responsibility be incompatible with determination by prior non-agential causes or by laws of nature. At most what we need for a FWD is that freedom be incompatible with total determination by prior agential causes (the case that matters is that of God’s creative act), a claim that I think some compatibilists will accept.”
    I assume you have read A.A. Howsepians’, “Compatibilism, Evil, and the Free-Will Defense.” He seems to have successfully shown what you said above to be the case. Also, his account would seem enticing for theistic-compatibilists merely concerned with ‘what the Bible says’ and whom either are agnostic towards any Biblical interpretation of God’s reasons for creating our world or actually endorse the interpretation that he created out of some valuing of freedom (this intention being the intention taken up by typical FWD defenders).

    January 26, 2009 — 18:42
  • I haven’t seen the Howsepian piece. Thanks for bringing it to my attention. It’s a clever idea, but I find the middle freedom idea completely implausible. It seems clear that if someone makes me act the way I would have acted had I been acting without being causally determined, then I am no more responsible for that act, just because I would have done had I not been made to act, than I am responsible for an infinity of other acts that I would have done in other counterfactual situations.

    January 26, 2009 — 20:06
  • Jordan Woods

    I don’t presume to more ably defend the theory of Middle Freedom (MF) – which is not my brainchild – than Howsepian does in his very own article so I will leave that up to him (It cost me $30 to purchase it from Sophia! I hope being a professional philosopher saves you from such a purchase.). I only mention one thing. If you accept that there is an essence from which an agent performs libertarian free choices, then it is difficult to see why a determined world, that instantiates every choice that flows from the essence of an agent in a world God could have created, would not be sufficient to deem the agents in that world as deserving of the moral responsibility libertarians desire. I ‘feel’ your severe worry about the seeming responsibility (on the part of the agent) for all the other acts that an agent could have performed, though, this seems to require certain assumptions – about what the difference(s) in ‘being’ is/are between worlds God does and does not create and the essences of agents in those worlds – that may not be outrightly uncontroversial.
    One other thing, Howsepian does say that one “need not endorse the Theory of Middle Knowledge in order to endorse the Theory of Middle Freedom” (223). This seems to me to require some defining as to what God’s imagination exactly is. Regardless, I thought I would mention this because those who read the abstract and/or some other tid-bit about his view may think he doesn’t take the above to be the case.
    Finally, I do not hold this position though I am strongly inclined to accept it as one possible view of freedom that could be used by a Howsepian-theistic-compatibilist in an FWD.

    January 26, 2009 — 23:09
  • overseas

    >a FWD does not need that freedom of will and responsibility be incompatible with determination by prior non-agential causes or by laws of nature. At most what we need for a FWD is that freedom be incompatible with total determination by prior agential causes (the case that matters is that of God’s creative act-
    But the creative act determines the laws of nature, and the acts of non-agential causes are also determined either by the creative act or by the prior acts of other agent-causes (the bullet strikes you b/c I freely shot the gun).

    January 27, 2009 — 5:56
  • Heath White

    Alex, what significant difference do you see between determination by agential, and by non-agential, causes? I can’t see any.
    I have long thought that there could be a compatibilist FWD. The FWDer says that what’s really valuable is making morally significant free choices; the compatibilist can agree with this. The libertarian FWDer will say that choices which are determined are neither morally significant nor free, but this is just the disagreement between compatibilists and incompatibilists.
    The libertarian FWDer says that we have evil because it is not possible (for some unclear reason) that everyone always go right in a world. The compatibilist can say the same thing, appealing to some equally unclear reason. (Perhaps no finite free agent could learn to act well without the experience of evil to overcome. Or whatever.) It is tempting, but a fallacy, to infer from “All the actions of every agent are determined” to “there is some possible world in which all the actions of every agent are determined so as to go right.”
    In short, I think there could be a compatibilist FWD.

    January 28, 2009 — 7:45
  • overseas:
    That’s right. So, given theism, compatibility between FW and causal determination and compatibility between FW and causal agential determination will come down to basically the same thing. But it’s still the case that the two kinds of compatibilism are different positions, since they say different things in the absence of theism.
    Heath:
    One might have the intuition that when all of one person’s actions were knowingly determined by another person, it is only the determining person who is responsible for the actions of the determined person.

    January 28, 2009 — 9:44
  • patrick todd

    Heath,
    You say, “The libertarian FWDer says that we have evil because it is not possible (for some unclear reason) that everyone always go right in a world.” I’m not sure about this. Of course, if you’re Molinist, there’s pressure to maintain that this is not possible, since then God should have actualized that world rather than this one. But if you’re an open theist, say, I don’t see what’s (in principle) stopping you from saying that there are worlds where no one sins, and indeed that this world could have been such a world, as (perhaps) God originally intended it to be, but that this possibility clearly didn’t become actual.
    Alex,
    In response to Heath, you say, “One might have the intuition that when all of one person’s actions were knowingly determined by another person, it is only the determining person who is responsible for the actions of the determined person.” It seems to me like this intuition serves as the basis for all manipulation arguments against compatibilism. Start with this intuition, then note that there’s no principled distinction between one’s acts being determined by an agent and by mere natural causes. I hear once in a while of compatibilists wishing to make such a distinction, but I’m not sure how it could go, and it seems right to say that most compatibilits think it’s best to just bite the bullet and maintain that one’s acts can be determined by agential causes. Do you perhaps think compatibilists should resist this?

    January 28, 2009 — 13:23
  • Patrick:
    The “no principled distinction” thesis can be questioned, though I accept it. If I were a compatibilist, I might try to maintain the significance of a distinction between an action being intentionally determined by an agent, and an action being determined in some other way (either by the causality of a non-agent, or by an agent but non-intentionally).

    January 28, 2009 — 13:44
  • Heath White

    Patrick,
    Right, I had Plantinga’s FWD in mind.

    January 29, 2009 — 7:01
  • Well, I’m a compatibilist, and I agree that there is no principled distinction. But–one person’s modus ponens being another’s modus tollens–I conclude that the agent is responsible in the so-called cases of ‘manipulation’ too. But that’s because the cases of omniscient super-scientists and God are not really cases of manipulation. I think that there is a big difference between an agent causing another agent to do something and that agent manipulating the other agent to do something in a way that undercuts one’s responsibility. Whether the former counts as a case of the latter depends on the route whereby he causes the other agent to do something.

    January 30, 2009 — 10:29
  • Seth

    I am relatively new to this world but I will do my best to contribute…thank you in advance for your tolerance.
    Patrick: I agree with your compatibilist’s attempted distinction between an action being intentionally determined by an agent, and an action being determined in some other way. If you take that further back in time: were the intentionally determined actions of the manipulator also intentionally determined by another agent, or were they determined another way? Then, were those actions which determined the manipulator’s actions intentionally determined or not? And so on…
    I have struggled with compatibilism’s use of FW and it’s distinction between “forced” decisions and ones made in other ways i.e. our character and psychological make up. It would seem impossible for there not to be a forced or manipulated action within a person’s life, which would then serve as causes for other actions. And, if inevitably our actions are predetermined by a mixture of intentionally and unintentionally determined actions, how can FW exist? Or, do the actions which fall 2nd and further behind in the chain have no bearing on the new action?

    January 30, 2009 — 10:36
  • Tim:
    In any case, to get the discussion back on track, do you have the same intuitions about love and punishment as you do about responsibility? In other words, is x’s loving y just as valuable when x is manipulated by y to love y, and would y be justified in punishing x for a crime that y manipulated x into doing? (This being the superscientist-type manipulation.) It seems quite coherent to say that there is responsibility despite manipulation, but to answer these two questions in the negative.

    January 30, 2009 — 11:20
  • Hi Alex. I’m not sure how much stock you should put in my intuitions in this sort of case–I’m probably pretty aberrant here…
    But in case, interesting question. A few random thoughts.
    (1) In some sense, I *am* trying to cause my children to grow up into certain sorts of people, and one of things I’m hoping to accomplish to that they’ll love me (and that I’m worthy of that love). Of course, in my case, my actions are far from the causal factor at work, and I’m far from perfect in my understanding (not to mention my virtue). But I don’t see this is ‘manipulating’ my children or making their love for me less valuable.
    (2) Two things that may be going on with the ‘superscientist’ cases that may fuel your sorts of intuitions. One is that (normally) ‘manipulation’ *does* involve some sort of interference with the person that undercuts freedom, and we subconsciously import this sort of attitude even into the superscientist case where it shouldn’t apply. But the second, more interesting reason, is that the scientist seems to be in a far different position vis-a-vis the ‘manipulated’ person than I am with my child. To put it in Peter Strawson’s terms–and I’m *really* sympathetic to Strawson–I’m still taking the ‘participant attitude’ with my child as I interact with him as he’s growing up. Whereas the super-scientist seems to be taking the ‘objective attitude.’ The POV of the superscientist (or of God) when it comes to resentment, blame, and punishment is so different from ours–from ordinary participants in the moral community–that I’m not sure what to say regarding it.
    (3) If you’re compatibilist about responsibility, there’s a good chance that you think that compatibilist freedom is the only sort of freedom available and/or worth having. So it’s not as though having agent causation (if it were possible), or occasional quantum freak events working in some self-forming actions, a la Kane, would make us ‘more free.’ There aren’t two types of freedom–mere compatibilist freedom that’s good enough for responsibility and also a more robust, spiffier libertarian type.

    January 31, 2009 — 13:40
  • No one should have to pay $30 for a PDF:
    http://www.andrewmbailey.com/papers/Howsepian07.pdf

    January 31, 2009 — 16:28
  • Tim:
    1. The child case is interesting. I think what I could say is this. A loving response not determined by the parents’ endeavors would be more valuable. But that does not mean that the parents’ endeavors are mistaken. For it could be that both (a) a loving responsible not determined by the parents’ endeavors would be worth having, but (b) it is better to go for, say, a 70% chance of a loving response determined by the parents’ endeavors over a 40% chance of a loving responsible not determined by the parents’ endeavors (just as it is better to go for a 70% chance of getting $100 than for a 40% chance of getting $120).
    2. “If you’re compatibilist about responsibility, there’s a good chance that you think that compatibilist freedom is the only sort of freedom available and/or worth having.”
    I don’t know that that is true. Compare two cases:
    A. George, who is quite a normal person, has a gun held to his head, and after a rational calculation, concludes that it is better to give the gunman the $100 being asked of him.
    B. Sally, who is quite a normal person, is offered the opportunity of buying a sufficiently functional car for $100, and after a rational calculation, concludes that it is better to give the seller the $100 being asked of her.
    I think a libertarian should say that in both cases there is freedom (unless there are further facts to the contrary, such as that Sally is controlled by a neuroscientist or George panics and loses rationality just before he gives the money). I suspect that a compatibilist ought to say the very same thing. After all, in both cases there is responsibility. (Suppose that George knows that the $100 will be used to buy a bomb that will kill hundreds. Then he has done wrong, and bears some responsibility for these deaths. But even without such an additional posit, he is responsible.)
    Thus, in both cases there libertarian freedom if libertarianism is true, and compatibilist freedom if compatibilism is true. But at the same there is a further kind of freedom which is worth having, and which is present in case (2) but not in case (1).
    Moreover, there may be other kinds of freedom that are worth having.

    January 31, 2009 — 18:43