Monday, October 17, 2022

The Utter Failure of the Free Will Theodicy

Note: What follows is a transcript of the video above.

This paper takes aim at Free Will Theodicies in Christian apologetics by showing inconsistencies with other Chrsitian doctrines commonly held to by Free Will Theodicy defenders that would reconcile Free Will and gods omniscience and foreknowledge of our actions before we do them. In this project I will grant for the sake of argument that Libertarian Free Will both exists and is coherent, even though I do not think Libertarian Free Will is coherent. I do this because I don’t think arguing against LFW is a useful counter apologetic and I believe the Free Will Theodicy fails even granting that LFW is coherent and steelmanning the position. 

My goal is to show that Christian responses to the problem of evil while referencing Libertarian Free Will (LFW) are at best ineffectual at answering the problem of evil because LFW does not preclude god creating what I call “heaven world” where any natural number of created beings all freely choose to never sin and always freely choose to love god. Theistic and especially Chrsitain apologists want to deny that god could create such a world, despite it being a logically possible world and god supposedly being able to do anything logically possible by way of his omnipotence.

I have a particular interest in the doctrine of Molinism which itself aims at reconciling god’s omniscience with the libertarian free choices of creation. We will find that the doctrine of molinism ends up causing significant problems for the free will defense and undermines the ability to deny that god could create “heaven world”. 

Molinism states that god has full knowledge of all counterfactuals, such as “Bob will always freely choose X in situation Y”. In this schema, god looks at all logically possible worlds, and then chooses which one to instantiate along with the people who will be in it. Then creation plays out deterministically, where the sum total of situations of the universe god created plays out and each person god created goes through them. The key is that each person makes a “free decision” that god knew they’d pick for every situation they live through.

Molinism is tied very closely to the free will defense, where theistic apologists want to say that “god can not create all logically possible worlds” because “if a person Bob will always freely choose X in situation Y, then god can not create a world where Bob freely chooses ~X in situation Y”.  They will then also insist that it is still “logically possible that Bob could have chosen ~X in situation Y”, and so we are left with the conclusion that god can not instantiate all logically possible worlds due to his creation having LFW. This in turn lets them claim that it’s not possible for god to create a “heaven world”.

What strikes me is exactly how deterministic this libertarian free will ends up being. The idea is that Bob always freely chooses X in situation Y, yet the theist apologist will insist that somehow it is still logically possible that Bob chooses ~X in situation Y. I believe that this is where a hidden contradiction is being glossed over. Because the word “always” ends up creating a rigid designator for the person “Bob”.  Note that “A rigid designator designates the same object in all possible worlds in which that object exists and never designates anything else.” (Reference)

In the free will defense, the word “always” means that Bob is rigidly designated as "someone who will freely choose X in situation Y". The problem is that this means it is not logically possible for 'Bob' to choose ~X in situation Y. So it ends up that there are no logically possible worlds that god could not create. 

Consider the alternative, what if we insist that “always” is not a rigid designator for the identity of ‘Bob’ because we want to hold that it is logically possible that 'Bob' freely chooses ~X in situation Y? In this case there is a logically possible world in which ‘Bob’ freely chooses ~X in situation Y and a logically possible world where ‘Bob’ freely chooses X in situation Y. Here it is up to god to choose which world to instantiate.  If this is the case then the schema for Molinism falls apart, because then god can't have foreknowledge of which choice Bob will make in situation Y without god having to choose which world to instantiate, which in turn seems to rob 'Bob' of his supposedly libertarian free will.

What’s worse for the theist is that this alternative sets it up so that god could choose to instantiate the logically possible “heaven world”.

Perhaps an apologist will claim that the addition of the word “always” is a misrepresentation, and that god merely has knowledge of counterfactuals so that “Bob will freely choose X in situation Y” and will claim there is no rigid designator. The problem here is that we can pose exactly the same problem. We can ask if there is a possible world where “Bob will freely choose ~X in situation Y.” If the apologist says there is no world, then it is the word “will” that creates a rigid designator for ‘Bob’. If they answer yes, then the free will defense fall apart on the problem described above - because god has to choose which world to instantiate and the heaven world becomes possible for god to actualize. 

Another possible response from apologists is to acknowledge that there is a rigid designator in the fact about Bob’s free choice, and so the possible world where “Bob freely chooses ~X in situation Y” is not actually ‘Bob’ but someone else entirely. 

The issue here becomes “why would god create Bob?” when there seem to be infinitely many options that god has.

Lets call the base person Bob(0) who always freely chooses X in situation Y.

But what about Bob(1) who is like Bob(0) in all respects, physical attributes and prior decisions, but will freely choose ~X in situation Y?

There seems to be no contradiction inherent in the concept of Bob(1), and so the existence of Bob(1) is logically possible.

In fact there seems to be a near infinite set of possible people god could create, where for any given individual that exists there’s a version of themselves that would choose to love god in a given series of situations Y that make up the possible world god could create. So even if we grant the existence of rigid designators, the free will defense doesn’t rule out the possibility of a “heaven world” populated with the set of N number people who would always freely choose to do the good in all circumstances that occur in that world. N merely represents any natural number. 

Common Threads of the Problem

I hope it is clear that with whatever option the theist goes with in trying to keep free will in conjunction with god’s omniscience, it ends up undermining the free will defense. I believe this is because theistic apologists are mixing a necessary sort of determinism to solve the theodicy with Libertarian Free Will, which is defined by its incompatibility with determinism.

For example we can see problems previously mentioned pop up once we get rid of the sort of determinism necessary for Molinism to work: 

What about cases where “Bob sometimes freely chooses X in situation Y, but other times he chooses ~X?” There seems to be no room for this sort of libertarian freedom in these theodicies or in Molinism, because then we get into problems with god’s foreknowledge of free choices and god having to determine which logically possible world to instantiate. This makes the real choice happen in god’s hands, not the individuals. 

In fact the only way these theodicies work is if there is a strangely deterministic nature to a person's “free choice”. Notice the examples above are framed that “Bob always freely chooses X in situation Y”. In what way can the choice be said to be free for that person if situation Y is determinative of what the choice is?  If the theist says that the person’s nature determines the choice in situation Y, then it is not free because they couldn’t have done otherwise. If the theist wants to deny this common definition of Libertarian Free Will, then god could create beings who by their nature would always freely choose to do the good and love god. 

The Nature Problem

I am not an advocate of Libertarian Free Will, but my critique doesn’t depend on falsifying the concept or proving it to be incoherent. All I need to do is show conceptual problems between LFW, our notions of LFW, and Chrsitain theology and theodicies. 

One major problem is our conflicting intuitions about the nature of free choices. Consider the following three thought experiments:

1.) Imagine a young teenage boy, after running around all day on vacation, he is taken to dinner with his family at a restaurant, where in typical fashion the child is ravenous due to being active and going through a growth spurt. When offered dessert, the only option is ice cream but there are two flavors: vanilla and chocolate. The child has a strong preference for chocolate and so orders that flavor. Is the child's choice free?

It seems very much to be a case of a free choice, but then the child is bound by his nature, which in turn determines his preferences. We don’t get to pick our favorite flavors, they just are what they are, so it is not much of a surprise that in this situation the choice seems determined by factors outside of the child's control, yet we would still call this a “free choice”.

2.) Free choices do not have to be hard ones, and can be determined out to all members of our species by the nature of our species. Consider having to choose what to eat, but the only options presented are either the well rotting carcass of a dead animal or a nutritious meal. In every case for a human being capable of making an informed decision, they will choose the nutritious meal. This is because we, by our evolved human nature, are repulsed from well rotted carcasses. I hate to gross people out, but it is blessedly rare for people these days to truly encounter how terrible an experience it is to be near a rotting carcass. I didn’t experience it until I was nearly 40, so I feel compelled to describe it. 

I had a small animal die beneath my deck in the winter or early spring, only to start rotting (and smelling) in the summer. I was forced to crawl under the low deck (1.5ft clearance) and get it out of there and bury the remains. Suffice it to say the smell, sight, feel of having to move it even through gloves and garbage bags, was horrific. It was all I could do to not vomit into the layers of masks I put on to try and cover the smell.

So when it comes to the choice of what to eat, it does seem as though this choice would be a free one - after all I could choose otherwise, I just wouldn’t by my nature. So if free will is important, then god could create a world where everyone *freely* chose to always do the good and choose to love god - because it would be like deciding between eating a nutritious meal and well rotted carcass.

3.) Let’s imagine a scenario where after death we are brought before god and are informed which religion is the true one. It seems odd to me that at this point it is “too late” for an individual to be given a choice to ask forgiveness for their sins and choose to love god and follow his will. This is arguably a far more “informed choice” that one could freely m ake than the choice we face now in terms of which religion we find to be true and worth pursuing. After all it is not as if people of other religions reject god or fail to love god - they all in fact believe they do and are aiming themselves at attempting to please god as they best understand his desires for them.  Even atheists presumably would want to know if they were wrong, I know I certainly do, and when given irrefutable proof of god’s existence, why would it be “too late” to ask forgiveness and freely choose to follow god? How then could this not be a free choice? If it were somehow not a free choice because we are effectively choosing between paradise and torture, then isn’t us learning about the eternal consequences of not making the choice to love god also not free? 

God’s supposed freedom

I don’t think that Christian apologists can deny that the choices laid out above are indeed free choices. Again we will come to a two pronged dilemma where neither option is good for the Christian theist.

Consider god, where most Christians consider god’s nature to be the ontological grounding of the good. Since it is given that no being can go against its own nature, it is literally logically impossible for god to do evil. As such, it seems to me and Christian apologists like Richard Swinburne, that god is not morally free when it comes to his actions. One could point out that being morally free in this scheme is not actually good, because it is not in god’s nature, but let's leave that aside.

So the dilemma is this: if god is not morally free, yet can love and be worthy of moral praise, why then do we need moral freedom to choose to love god and be saved? Or even have the capacity to sin at all? It can not be that libertarian free will is valuable, because if god is the locus of value and by nature lacks LFW with respect to morality, then it is by definition not valuable.

The alternative horn is that if god is free but it is impossible for him to do evil, then it is logically possible to create beings who would never do evil or fail to love god and yet they would still be free! As such the free will defense would fail. 

Now we can find apologists like William Lane Craig who can try to split this dilemma, by insisting that even though it is impossible for god to choose to do evil, he is still endowed with Libertarian Free Will.

“I do affirm that God’s being essentially good means that goodness is a property which God could not have lacked. Indeed, on my view God just is the paradigm of goodness in every possible world. This entails that God cannot do evil, since that would be contrary to His very nature.

First, you assume that freedom entails the ability to do the opposite of what one does. I’m persuaded that this is not true. Consider the well-known illustration of someone who, unbeknownst to him, has had his brain wired up with remote-controlled electrodes by a mad scientist who is an Obama supporter. When the man enters the voting booth, if he votes for Obama, the mad scientist will do nothing. But if he goes to vote for Romney, the mad scientist activates the electrodes, which trigger him to vote for Obama instead. Now clearly the man has no power in this situation to vote for Romney. But if he goes in and votes for Obama, doesn’t he do so freely? After all, the scientist did nothing in this case! It is just as if the man were not wired with electrodes at all. This thought experiment suggests that what is crucial to freedom of the will is not the ability to do the opposite but the absence of external causal constraints upon one’s choice: it is entirely up to you. In God’s case He is clearly free from such external causal constraints and therefore does the good freely. So He is not at all a moral automaton, but a free agent.”


There are two huge problems for Craig’s view.

First is that his thought experiment is demonstrably false. Note that in the situation where the voter “freely chooses to vote Obama” and so the mad-scientist doesn’t have to trigger the device that would force the person to vote for Obama - that doesn’t change the fact that the person could have chosen otherwise even if they could not have acted otherwise.

Note that in this scenario the only thing the mad scientist could have done was identify that the voter had chosen to vote Romney and then would trigger his electrodes to force the arm to pull the lever for Obama. However in this scenario the choice to vote Romney was still there even if the person couldn’t have acted on it. The Principle of Alternative Possibilities (PAP) is not actually falsified by this thought experiment. 

In fact in principle Libertarian Free Will entails you couldn’t falsify PAP because the choice is something of the mind, which is not deterministic. If the scientist could somehow trigger the electrodes to make the person choose to vote for Obama, then it would indicate determinism was true and Libertarian Free Will would be false! As such it seems Craig fails to do away with the intuitive and widely defended definition of LFW as “the ability to have chosen otherwise”. 

Craig’s second problem is that his apologetic fails even if we granted his definition of Libertarian Free Will. He wrongly contends that Libertarian Free Will isn’t defined as it classically is as the “freedom to do the opposite” but rather to be “free from external causal constraints”. The problem is that by god’s very nature as the supposed necessary being, god necessarily gets to decide and control what our natures are like! 

Perhaps Craig might say that on his scheme of LFW god makes it so that we may not be free in preferring a nutritious meal to a rotting carcass, or in our favorite flavor of iced cream, but we are free in choosing to love god or not. But then he would also have to say that we are not really free in loving our children, because by our nature we have an immense attachment to them and a desire to love and protect them. Similarly children by nature trust and form bonds with their parents so long as they are not mistreated.  This is not to mention the mountains of apologetics about how humans are designed to have a “longing for god” or the “god shaped hole in our hearts” which contradicts the idea that god created our natures so that we could freely choose to love god or not. 

But the problem goes much deeper. In fact Craig’s entire endorsement of Molinism and the Free Will Defense itself rests on the nature of an individual person being what determines that they will “always freely choose X in situation Y”. This gets down to the very core of the rigid designator of any given person on Molinism, and Craig can’t avoid this without tearing down the entire molinistic framework. By choosing to create Bob(0) or Bob(1), god is deciding the nature of the individual he is creating which in turn determines what he will supposedly “freely choose” in a specific situation Y. The key here is that we are not just talking about random choice X in situation Y, but this is directly tied to “will Bob freely choose to love and follow god given the sum total of circumstances in his life?”.  In such a molinist schema that Craig endorses, this freedom is entirely determined by the nature of the being created and put into the world. If it was not, then god couldn’t have foreknowledge of what each person would choose in any given situation. 

The fact is that Molinism as a doctrine entails that the nature of the creating beings is chosen by god, because that’s how god is able to have foreknowledge of their supposedly free choices in any given situation.

A Better Situation

One theodicy I’ve found was put forward by Dr. Joshua Rasmussen in a paper where he attempts to deal with the problem of god not having moral free will undermines the idea of it having any value.

To quote the abstract:

“Theists typically think the freedom to choose between right and wrong is a great good . Yet, they also typically think that the very best being—God—and inhabitants of the very best place—heaven—lack this kind of freedom. The question arises: if freedom to choose evil is so good, then why is it absent from the best being and the best place? I discuss articulations of this question in the literature and point out drawbacks of answers that have been proposed. I then propose a new answer by showing how freedom to do evil could result in certain good situations even if it does not contribute to the intrinsic greatness of a certain being or place.”

Dr. Rasmussen claims that in a situation where a wife wants her husband to surprise her with a vase of her favorite flowers, the situation where the husband does that of his own accord rather than being causally determined to do so by his wife just seems prima facie to be a “better situation”.

I think the primary problem with this is that for Dr. Rasmussen to say the situation is “better” there must be some normative ground for assessing situations beyond simply appealing to our intuitions.  After all, theists are quite fond of demanding some sort of grounding for normativity from atheists, so it seems fair to ask what grounds this evaluation. 

We know the grounding can’t be in god’s nature, because god simply does not get to choose whether or not he loves us. In fact we could ask Dr. Rasmussen if it would be a “better situation” if god were not perfectly loving but could in fact choose whether or not to love us, being free to hate some of his creation - and so perhaps Dr. Rasmussen’s Christian universalist outlook would be fatally flawed and he should instead adopt some sort of Calvinist framework. 

Conversely if we say there is no normative ground for assessing what is and is not a better situation, then why is it not the case for other normative properties, or all of morality? 

Unfortunately Dr. Rasmussen’s theodicy just reduces down to “it seems as if free will with respect to morality seems valuable to us, even if god doesn’t have it” and I don’t see how the theist can hold to that and say that god is the greatest possible being if he lacks something that we are to find so valuable. 

The Logical Problem of Evil, Resurgent

Because of these reasons, it remains logically possible for god to create a “heaven world”, where N number of people all freely choose to always do good in the sum total of situations Y that will occur in that world. 

If we say that a person can make “free choices” even if their nature determines what they will freely do in any given situation in a deterministic way so that god can know it, then this is a logically possible world.

If we deny this, then god can’t be omniscient and foreknow what any given individual will do in the future given a specific situation, contradicting not only millenia of theology, but also a number of stories in the Christian bible as well.

The theist must pick which way they want to go: Either the problem of evil is not defeated by the Free Will Defense, or their deity cannot be omniscient and so cannot be god by definition.


No comments:

Post a Comment