5626463547 | San Jose, CA 95126

Does Molinism Avoid Making God the Author of Sin? | Welty on Molinism

In “Molinist Gunslingers: God and the Authorship of Sin,” Greg Welty argues that if Calvinism has a distinct problem of making God the author of sin, then so does Molinism. Although the expression “author of sin” has different meanings, the sense at play here in the article is primarily about culpability. In this post, I will look at Welty’s argument for the claim that Molinists have a similar problem and I will suggest a couple ways a Molinist might respond, which are not considered in the paper. These might or might not work.

The Molinist’s Concern with Calvinism

The concern the Molinist brings against the Calvinist is this: if Calvinism is true, then theological determinism is true; but if theological determinism is true, then God acts as a sufficient cause for the evils that occur; but if God is a sufficient cause for the evils that occur, then God is culpable for those evils. But God is not culpable. So by a chain of modus tollens, not Calvinism.

Although Welty focuses on the work of Ken Keathley, here is William Lane Craig raising the same charge on a section of his site Reasonable Faith devoted to comparing Calvinism with Molinism:

Universal, divine, determinism makes God the author of sin and precludes human responsibility. In contrast to the Molinist view, on the deterministic view even the movement of the human will is caused by God. God moves people to choose evil, and they cannot do otherwise. God determines their choices and makes them do wrong. If it is evil to make another person do wrong, then on this view God is not only the cause of sin and evil, but becomes evil Himself, which is absurd. By the same token, all human responsibility for sin has been removed. For our choices are not really up to us: God causes us to make them. We cannot be responsible for our actions, for nothing we think or do is up to us.


Craig recognizes that a Calvinist might not embrace determinism1. But Craig’s use of the above argument is directed at Calvinists who embrace it. For those Calvinists that do embrace it, he thinks that there is that problem with which we started.

What Welty attempts to do in this article is to show that the Molinist is in no position to make this complaint, for the Molinist also has a sufficiently analogous problem if the Calvinist has a problem. That is, it would be unjustified for Craig, for instance, to single out this problem with Calvinism if Molinism also has the same or an analogous problem.2

Welty’s Argument

Welty argues that since divine providence according to Molinism is sufficiently analogous to sufficient causation, Molinism has all the same liabilities as Calvinism. There are “two movements” in Welty’s argument to show there is a sufficiently analogous problem. The first movement is a comparison of an ordinary case of causation to what we might call a “bridge case”. The second movement is a comparison of the bridge case and Molinism. What all of these cases have in common is that these are all cases of responsibility for outcomes. And that makes sense, because Molinists claim that God sans creation decided what possible world to actualize based on his knowledge of what the agents would freely do in such and such circumstances. The free choices of human agents are outcomes of God’s decision to actualize a particular world.

The ordinary case. The ordinary case is that someone has a gun, who knowingly and intentionally aims it point-blank at another person and pulls the trigger. The person could have chosen otherwise, but pulled the trigger. This person is responsible for murder.

The bridge case. The bridge case is just like the ordinary case except that it involves a different type of gun. This gun “creates ex nihilo a Bullet Bill in the chamber…. Bullet Bill is sentient, a homunculus encased in steel, possessing libertarian free will.” (63-64). Welty also notes that the person shooting the gun knows about this, and knows the counterfactuals of creaturely freedom (CFFs) about each Bullet Bill that gets fired. So he knows, for instance, that although Bullet Bill could choose to not kill Sam because there is some possible world where Bullet Bill freely twists and misses Sam, in the current circumstances Bullet Bill would hit Sam.

The Molinist case. The Molinist case is just Molinism: that God creates agents who he knows would freely do evil things in such and such circumstances.

In order to make the first movement from the ordinary case to the bridge case, Welty highlights the following features: the shooter is responsible, the thing over which the agent directly had control was not sufficient for the outcome, and there are other conditions that have to be in place over which one does not have control in order for the outcome to happen. So for example, one could pull a trigger and the gun jams. The action over which an agent has direct control (i.e. pulling a trigger) is not sufficient for resulting in a bullet killing another person. And there are conditions like what the current laws of nature are that have to obtain in order for the bullet to make it from the chamber into the heart of another person. In the case of laws of nature, we lack control over those. And furthermore, it is plausible that the laws are contingent. These contingencies and conditions beyond our control do not undermine a person’s culpability.

What I have described (following Welty) is what is true of the ordinary case. But notice that these conditions also apply to the bridge case. If someone fires a Bullet Bill knowing that in the current circumstances he would kill the person before him, the fact that Bullet Bill could have done otherwise – that there is a possible world where he chooses to not hit the person – does not absolve the shooter of culpability. Furthermore, pulling the trigger is not sufficient for the outcome that someone dies; Bullet Bill has to make the choice to hit the person. And which choice Bullet Bill would make is not something within the shooter’s control.

The second movement is a comparison of the bridge case and Molinism. Just change the initiator of the sequence of events resulting in the outcome from a human shooter to God deciding which world to actualize, and replace Bullet Bill with any libertarian agent who does something evil.

Then ask: is there any relevant difference between the nodes of the first and second movements? Welty’s view is that there is no identifiable relevant difference, and thus if God is culpable on Calvinism, so is God on Molinism.

Objections to Welty’s Argument

As far as I can see, there are two sorts of strategies the Molinist has available. She might argue that there is a relevant difference between the ordinary case and bridge case or between the bridge and Molinist case. Alternatively, she might argue that the problem is worse on Calvinism than on Molinism even if there is an analogous problem for Molinism. Welty considers objections that meet the first strategy.

Welty discusses quite a few objections. You might think that the relevant difference is that it is up to the shooter what the bullet does, but it is not up to God what a free agent does. But if it is not up to God what a free agent does because the CFFs are beyond God’s control, then neither is it up to the shooter what the bullet does, for the bullet’s properties and the laws are beyond the shooter’s control.

You might also think the relevant difference is that whereas free agents in the Molinist case could do otherwise, ordinary bullets could not do otherwise. Welty has a number of things to say about this objection. In my opinion, the most significant thing he says in response is that “given the circumstances and the truth of the relevant CFF, the agent cannot do otherwise than what he in fact does. Yes, there is a possible world in which does otherwise in the same circumstances, but that is a world in which the relevant CFF is false.”(70). And that is exactly right. If God is to know the CFFs, then then relevant CFF better be true. And holding the truth of the CFF fixed, the agent could not do otherwise.

I think Welty could continue to press his original bridge case in response to this objection as well, which he did not do. The above objection compares the Molinist case directly to the ordinary case. But recall that the bridge case is an human shooter firing a Bullet Bill knowing full-well what Bullet Bill would do in the circumstances. Even though Bullet Bill could have done otherwise, my intuition is that the shooter is culpable not just for the intention to kill, but for the murder itself. Bullet Bill’s ability to do otherwise is irrelevant to assessing whether the shooter is culpable for the murder. The shooter might be less culpable given Bullet Bill’s ability to do otherwise, but that the shooter is culpable is not in question.

The last objection Welty considers about which I will write is the following: although Welty has shown that God would be responsible for the outcome of the assassin’s sin, he has not shown that God would be responsible for the assassin’s sin itself (cf. Objection 5, p.71). In response, he writes,

Strictly speaking, this isn’t the case. In the “ordinary gun” case, I not only bring about the end result (the death of the person); I bring about the intermediate state — the bullet speeding down the barrel of the gun. As far as I can tell, I can make the same argument with respect to the intermediate state that I made with respect to the end state: both states involve my actualizing circumstances in which a substance with powers, and liabilities to exercise those powers in those circumstances, would in fact exercise those powers. And again, as far as I can tell, the “Bullet Bill gun” case is parallel in those respects. (71)

It might be that Welty is responding to a different version of the objection than what I am thinking of, for everything Welty said about intermediate state seems right to me, but I don’t think he’s got at the heart of the objection.

The point of the objection, as I read it, is that there is some sin that is not an outcome and which God would not be responsible for under Molinism, but which the shooter would be responsible for. (Welty might have something else in mind, which is why I said I might not understand the objection.) The sin – that is not an outcome – is not the bullet flying down the barrel. That is an outcome of the act of pulling the trigger. You might even think the sin isn’t the pulling of the trigger, although that might be the first sinful action. If the assassin’s sin is the choice to pull the trigger in order to kill the person – which results in the action of pulling the trigger, which results in the event of the bullet flying down the barrel, etc. – if that is the assassin’s sin being referenced in the objection, then all this talk about intermediate states is beside the point.

Even so, I don’t think the objection stands. Consider the bridge case of Bullet Bill as Welty mentions at the end. Even if Bullet Bill makes his own choice about whether to kill Anne, is there any doubt about whether the shooter is culpable? Seems not. After all, the shooter knew the CFFs about Bullet Bill. The shooter knew what Bullet Bill would choose, resulting eventually in the death of Anne. And given that there does not seem to be a relevant difference between the two cases in the second movement, it would appear that God is responsible for the shooter’s sin, contrary to the objection.

In the next two sections, I will consider two other objections to Welty’s argument

Strategy 1 Objection

I said above that there were two strategies for the Molinist. The first strategy was to find some way of stopping the slide between the ordinary case to the bridge case to the Molinist case. I’ll consider one way to adopt that strategy in response to Welty’s argument.

In Welty’s argument, the bridge case involving Bullet Bill is important because it introduces the idea of an agent who brings into existence some other free agent, and yet the first agent is culpable for the outcome of the second agent’s free choice. Now if there a way to argue that in the bridge case, Bullet Bill isn’t an agent with free choices, then it is not possible to move from the ordinary gun case to Molinism. How might that go?

In the introduction to the bridge case, I quoted Welty’s description of the gun that kills: the gun “creates ex nihilo a Bullet Bill in the chamber…. Bullet Bill is sentient, a homunculus encased in steel, possessing libertarian free will.” You might even think that Bullet Bill comes out with beliefs and a full blown character.

But now consider the following quote from one libertarian:

According to libertarians, moral virtues cannot be imposed upon one person by another and cannot be instilled, produced, or brought about by a sufficient cause external to the agent.3

The same would be true of moral vices according to libertarians.4 Now if this is the libertarian view, notice that Bullet Bill comes into existence ex nihilo. He didn’t develop his character through being ultimately responsible for it (e.g., satisfying Kane’s UR condition). The gun made Bullet Bill that way. And how Bill is when he comes into existence makes a big difference in what he is inclined to do, even if it does not determine what he would do.

This is quite unlike what happens with humans who develop their characters over time, who gradually become moral agents starting with less free choices to eventually fully free choices. And that’s the crucial difference between the second case and the third. Bullet Bill, like the ordinary bullet, is not responsible.

There are two things I think Welty can say in response to this objection.

First, it is not necessary to the bridge case that the bullet come into existence ex nihilo. Bullet Bills can be come into existence when they are manufactured in the bullet factory, and while they sit in their cases, they be thinking about all the horrible things they might or might not do. By the time Bullet Bill gets fired down the barrel, he would have developed his character. Tweaks like this are easy.

Second, you might wonder whether it is possible for God to create a free agent with a fully developed character. In fact, you might think Adam and Eve are two such cases. Even if you thought they were not real persons in actual history, you might still think that they were beings God could have made. As far as I am aware, there isn’t any case in scripture where God is depicted doing something that he could not do. But as I pointed out here, the problem of just creation is that God creates agents who are morally good, and that is a widely accepted view in ecclesiastical history among both libertarians and compatibilists. So it seems to me that the motivation with denying Bullet Bill could be responsible conflicts with the idea that God created a morally good human person, or that God could do that.5

Strategy 2 Objection

The second strategy is to concede to Welty that there is an analogous problem for Molinists, but the problem is nevertheless worse on Calvinism. How might that argument go?

After noting that the ordinary bullet is not responsible but Bullet Bill is responsible, Welty writes:

So there is a clear difference here in the two cases because of a difference in the bullets. But does Bullet Bill’s individual responsibility somehow lessen my responsibility, the one who aims and shoots the Bullet Bill gun and thereby kills three people in rapid succession? I don’t see how. (65)

In Thomas Nagel’s article, Moral Luck, he points out that luck challenges our attributions of responsibility. As luck increases, the less plausible it is that one is responsible. It is actually not a simple matter to say what luck is. But it is often thought that indeterminism introduces luck. From the shooter’s perspective, he does not control what Bullet Bill does, and it might be thought that because Bill’s choice is not in the shooter’s control, the shooter is to some extent lucky that Bullet Bill did what he wanted him to do. By contrast an analogous case for Calvinism would involve the shooter causing and determining Bullet Bill to do what he choose to do. That’s supposed to seem worse.

I actually don’t think this strategy is very good. For one thing, it requires the Molinist to say that God is lucky each time a creaturely agent chooses to do God wanted the agent to do. Molinism is actually a theory about providence, and saying this much makes Molinism look like a much weaker version of providence than Molina and his followers intended.

In addition, there’s still a problem that the Molinst would be conceding and for which an answer would be required. That is, a Molinist would still have to explain to an open theist how God is not culpable. For whatever answer the Molinist gives, it is not clear that the Calvinist could not also provide that same answer.6

Finally, this approach reminds me of two pundits I saw on television today, where one pundit argued that his candidate was preferable to the other candidate because although his candidate has said some racist things, his candidate was not as racist as the other candidate. I have no sympathy for such a strategy as this, and I do not welcome it in the area of philosophical theology.

Concluding Comment

I think Welty is right. The above objections are not successful and I think the analogy is pretty tight. Nice job, Greg. And cool title.



  1. See Craig’s comments on Turretin.
  2. Obviously, how Welty would have to respond to an open theist raising this criticism of Calvinism would be different. He knows that.
  3. Kevin Timpe, Free Will in Philosophical Theology, p.108. Timpe is quoting Thomas Talbott.
  4. I don’t think it is necessary to give other citations that this is a view libertarians hold since I don’t believe it is something that would be challenged.
  5. If you object that God made Adam out of the ground and Eve from Adam’s rib, is that really an important difference from being created ex nihilo? I don’t see any reason to think so.
  6. Welty mentions three different approaches on pp. 73-74. I don’t consider apophaticism among the things the Molinist can say since the Molinist, like the determinist Calvinist, is providing a positive model.
2018-06-07T21:29:00+00:00 September 13th, 2016|5 Comments


  1. Devan September 13, 2016 at 11:00 am

    How is it that God is supposed to be analogous to someone firing a gun at someone? Is the shooter omnipresent, omnipotent, omniscient? The shooter seems like a limited, faillible, fallen person. One is God, the other isn’t, so I don’t see how you can apply this situation to God.

    What about the concept of evil or sin being an absence of good? Like rust in a car, or coldness being the absence of heat? Good being turned towards God, evil turned from him.

    With God, he has an ultimately good purpose in bringing about his Kingdom, but he can utilize the sinful actions of people, like a Judo master, to bring about a greater good?

  2. Jonathan September 13, 2016 at 12:04 pm

    How is it that God is supposed to be analogous to a Judo master?

  3. James A. Gibson September 13, 2016 at 12:05 pm

    The relevant concept in play here that raises the challenge against Calvinism is causal sufficiency. What Welty rightly points out is that we can take away causal sufficiency is an ordinary case and you will have the intuition that the shooter is still culpable. Then apply the bridge case. Same intuition. And there isn’t anything we can see that stops the second movement. If you think it isn’t causal sufficiency that’s doing the work, then discussed objection that the Molinist typically raises against the Calvinist goes away. And then Welty would still be right: that there is a problem for Calvinism only if there is an analogous problem for Molinism.

  4. Sam Garcia September 18, 2016 at 11:10 am

    Ye do err, not knowing the Scriptures.

    I think the Old Testament Law is worthy to be looked at in this case, where a man who looses an ox gores a man.

    Exodus 21
    28 If an ox gore a man or a woman, that they die: then the ox shall be surely stoned, and his flesh shall not be eaten; but the owner of the ox shall be quit.

    29 But if the ox were wont to push with his horn in time past, and it hath been testified to his owner, and he hath not kept him in, but that he hath killed a man or a woman; the ox shall be stoned, and his owner also shall be put to death.

    The ox is not a deterministic bullet, but a creature. A deterministic bullet or weapon would kill. The owner is only responsible if he knew for sure the ox had a past tendency to do so.

    Since in Calvinism, a man can only sin and not do anything else but sin, God has the responsibility. Since man in Molinism can do something against his circumstances (or why else did Jesus wonder at few times when He found faith or unbelief?), God is not responsible.

  5. James A. Gibson September 18, 2016 at 12:27 pm

    Thanks for the comment, Sam. I guess I have two thoughts about that: one about the Scriptures, the other about the argument under discussion. Regarding whether I don’t know the Scriptures: yeah, I probably have some incorrect interpretations about some of them; probably about eschatology, if I had to guess which ones. But I don’t think your use of Exodus 21 is all that compelling. You’re taking a passage that focuses on the epistemic state of an ox owner, which is relevant to holding the ox owner responsible or not, and then inferring something about the metaphysics of responsibility. So much for that argument. The second thought I have is that you don’t understand how analogies work. Everyone knows an ox isn’t a bullet but a creature. A bullet might kill whether it were deterministic or not. It’s not like if I shoot a bullet at someone where the chance that it kills is, say, 50%, makes me less responsible. Again, I appreciate the reminder to read my Bible.

Comments are closed.