Why No Theist Should be a Compatibilist

Theologians often ask the question: if God is sovereign over all creation, is it possible for humanity to be free? And if so, how? One of the recent attempts to reconcile divine sovereignty with human freedom is called the compatibilist approach, whereby determinism and free will are considered compatible. This approach is embraced mainly by theologians in Reformed Protestant or Calvinist denominations, such as John Frame and James White. Compatibilism is not the view of Augustine and may not even be the view of John Calvin. I intend to explain why no one who believes in an omniperfect God should ever take a compatibilist view under any circumstances.

The arguments presented here are ripped from Volume 1 of the 2011 Philosophia Christi, which is the trade journal of the Evangelical Philosophical Society. This video will be about free will, which is the view that people are capable of making choices and being held responsible for those choices.

First, let's define our terms:

Determinism - Determinism is the view that all events are governed by laws, which are statements of conditions under which certain effects must occur. For a determinist, Conditions + Laws = Decision.

Simple Indeterminism - Simple Indeterminism is the view that a part of human choice is randomly determined, much like rolling dice or spinning a roulette wheel. Some interpretations of quantum physics view subatomic particles as exhibiting this type of random behavior. For a simple indeterminist, Conditions + Laws + Random Variables = Decision.

Compatibilism - is the view that all events are governed by laws, which are statements of conditions under which certain effects must occur. This is the same definition as determinism, except that the compatibilist believes that such decisions can be considered free choices of which we can be held responsible. For a compatibilist, Conditions + Laws = Decision (which we can call a free choice).

Libertarian Free Will - This is sometimes called self-determinism. Libertarian free will is the position that humans are agents who are able to self-determine some of their behaviors. The libertarian holds that for an action to be free it must be caused by the agent who performs it, and must be done in such a way that no antecedent conditions determine the act. A libertarian agent who makes a choice in a situation could have chosen otherwise, even in the same circumstances. For a libertarian, Conditions + Laws + Agency = Decision.

I will be arguing for libertarian free will and and against compatibilism on two grounds:

1. Reasoning can only be done by an agent with libertarian free will.
2. The problem of evil cannot be resolved without libertarian free will.

One series of arguments against libertarian free will that I have read over the year is the argument that libertarian free will is incoherent or unintelligible. If God exists, then there has to be at least one agent who has libertarian free will, and, presumably, is able to reason. Here's why. Theism is the belief that there is an omniperfect God who is distinct from the physical world, and from all creation. If God did not have libertarian free will, then he would not be sovereign, but would be controlled by outside forces. Hence, one must believe in a God with libertarian free will in order to be a theist. Therefore, no theist can ever argue that libertarian free will is an incoherent idea. That would be inconsistent.

First argument: Reasoning can only be done by an agent with libertarian free will. Compatibilists will argue that reason does not require free will, but can be performed by anything with rational causes. To the compatibilist, what matters is not whether or not you are being controlled (that is a given), what matters is what controls you. If you are pumped full of drugs and it causes you to have a certain set of beliefs, those beliefs are not considered rational. However, if a persuasive philosopher delivers arguments and it causes you to have a certain set of beliefs, that is considered a free and reasonable choice.

One problem with compatibilist theories of reason is that they do not tell why some reasoning belongs to a particular agent. Because compatibilists deny the existence of autonomous agents, they cannot argue that reasoning belongs to an agent if that reasoning comes from that agent's autonomous will. Instead, the compatibilist usually argues that an agent is reasoning if that agent is reasons responsive. This does not really work. A mathematician's notepad may be responsive to the reasoning of a mathematician, and is being controlled by a rational agent. Are we then to say that the notepad is reasoning? A computer is responsive to a rational programmer's instructions, but the machine is not reasoning. Using someone else's reasoning is not the same as reasoning for yourself.

Compatibilists will respond to these types of manipulation arguments by altering their definition of what it means to be rational. They say that subjects are responsible for their decisions when they issue from a mechanism that is moderately reasons-responsive and is also the agent's own mechanism. An agent is moderately reasons-responsive if that agent has two things:
1. A regular pattern of reasons receptivity. This means that the pattern of receptiveness to reasons forms a coherent and intelligible pattern.
2. The agent must at least sometimes act based on his or her best reasons.
A Mechanism is the agent's own if the agent takes responsibility for the mechanism by seeing himself as the one that made the decision.

Suppose Jill, a logic professor and also a neuroscientist who is frustrated with her student Dan, plants a logic chip in his brain. The chip causes Dan to form beliefs based on logic that he otherwise would not have formed himself. For example, if Dan comes into class very tired, and is shown that A=B and B=C, he does not reason that A=C. The chip, however, registering that Dan has A=B and B=C as beliefs, causes Dan to believe that A=C. If the chip caused the belief to form in Dan, can we say that Dan reasoned to that belief? Is Dan responsible for that belief, even though it was caused by the logic chip? Suppose that the logic chip causes Dan to think that he made the decision and that he (Dan) is responsible for praise and blame. Does that even make any difference? I think not. For suppose that Jill programmed an error into the chip, and instead of forming the true belief that A=C in Dan's head, the chip's malfunction causes Dan to believe that A≠C. Can we say that Dan is the one who committed those mistakes? Can we hold Dan responsible for that false belief? The above scenario fulfills all the compatibilist criteria for reason, yet no one would say that Dan is responsible for the beliefs formed by the chip in Dan's head.

Compatibilists may also argue that reasons are causes and that a person's behavior is determined by the best and strongest reasons. If compatibilism is true, and then the bundle of reasons that support a decision must also compel a decision in an irresistible way. However, we know that people do not always make the reasonable decision. People reject conclusions that their best arguments support.

Here is an example: Suppose James came to believe that if reason requires libertarian free will, then compatibilism is false. After reading a number of books, James now believes the antecedent of this argument. He doesn't want to conclude that compatibilism is false, so he converts the modus ponens into a modus tollens. He claims to have learned from the undeniable truth of compatibilism that reason must be redefined so as to exclude the need for libertarian free will. The fact that we can reject a conclusion that our best reasons support is powerful evidence that reasons do not compel a conclusion.

The compatibilist account of reason fails to distinguish between compulsive actions that are occurring within the brain from reasoning that the agent does. Compatibilists have tried to get around it by altering the definition of reason. These new definitions fail because they either attribute reason to inanimate objects such as a notepad, or they deny reason to human beings. Without a distinction between free choices and compulsive acts, there can be no such thing as human reasoning. Compatibilism cannot make such a distinction, and therefore cannot account for human reasoning.

Second argument: The problem of evil cannot be resolved without libertarian free will. The definition of God is "that than which nothing greater can be conceived." God is the metaphysical ultimate, and is by definition an omniperfect being. Such an omniperfect being would be omnipotent, omniscient, and omnibenevolent. Any being that lacks these attributes would be deficient, and therefore would not be God.

The classical objection to the existence of God is called the problem of evil. Evil is by definition a violation of a moral ought, and therefore a deficiency. God is perfect and can have no evil in him, yet our world is filled with evil. How can an omniperfect God co-exist with a non-omniperfect creation? The traditional free will defense of this view, supported by Augustine, has to find some way of severing God's omniperfect desires from the morally imperfect desires of created beings.

Any solution to this problem, short of denying the existence of moral evil, will require that the morally deficient wills of the creatures be autonomous from the omniperfect will of God. If it is better for some creatures with autonomous wills to enter relationship with God than it is for an entire universe to be free of evil, then we may have an explanation as to why evil exists in the first place. The very definition of autonomy restricts which possible worlds are created by God. Such is not in any way a restriction on God's omnipotence, as omnipotence is defined as maximum ability to actualize different states of affairs. To say that God could make an autonomous creature freely choose to act differently than the creature would have acted is as incoherent as saying that God should be able to make a square circle.

Compatibilism, on the other hand, faces one insurmountable obstacle which Jerry Walls calls CI.

CI: If freedom and determinism are compatible, God could have created a world in which all persons freely did only the good at all times.

This world wouldn't just be some logically possible world, uncreatable due to creaturely freedom, but a fully actualizable and creatable world. I will be arguing that any being who would not prefer to create such a world cannot be omnibenevolent. I agree with Antony Flew when he said: "Certainly it would be monstrous to suggest that anyone, however truly responsible in the eyes of men, could fairly be called to account and punished by the God who has rigged his every move. All the bitter words which have ever been written against the wickedness of the God of predestination--especially when he is also thought of as filling Hell with all but the elect--are amply justified." Jerry Walls illustrates this by asking us to imagine the following scenario:

Suppose a preschool is run by a woman who is a master of psychology, neuroscience, and human manipulation. Unbeknownst to their parents, she deliberately conditions the children who attend her preschool. Some of the children she conditions to behave virtuously and live productive lives. Some of the children she conditions to behave in a wicked manner, and sometimes even to become thieves, rapists, and murderers. Let us assume that she succeeds perfectly in her project and each child turns out exactly as she intends.

She avoids detection, and leaves the preschool to go to law school, become a lawyer, and eventually become a judge. One day, a man who once attended her preschool is brought to trial on child molestation charges. Sure enough, overwhelming evidence is brought in, including a confession by the man that he did in fact molest many children. The judge condemns him as a menace to society and sentences him to life in prison with no possibility for parole.

In prison, the man realizes that the judge was his preschool teacher and that she deliberately conditioned him to be a child molester. She could have just as easily conditioned him to be a well-adjusted person and an upstanding citizen. How could we possibly avoid concluding that this woman is evil and directly responsible for the evil acts of these former students? I do not think there is any way. If compatibilism is true, then the situation parallels our situation with God. If humans commit acts of evil, and there is no autonomous will, how can we avoid declaring God evil for making humans act in such a way?

The compatibilist might object that God needs evil to exist in order for him to be maximally great. If we do not sin, how can God show forgiveness? This view makes God dependent on creation in order to be a maximally great (omniperfect) being. If this is the case, then God is dependent on creation, and his creative decree was not a free act, but a necessary act. If God is dependent upon creation, and if his creative decree is not a libertarian free act, then classical theism is false. Such a view is not classical theism, but more of a pantheism, as it closely resembles the theology of Baruch Spinoza.

The compatibilist might fire back that evil needs to exist in order for conscious creatures to appreciate the good. This is nonsense. First, experiencing evil is not necessary to appreciate the good. Otherwise, God himself could not appreciate the good until he experiences evil.

On compatibilism, God is the great puppeteer, and there is no autonomous will. God could either instill with a sense of appreciation anyone who thinks of the good. Another option is for God to hardwire us with an appreciation of the good, the way we are hardwired to believe in other minds, to believe in the reality of the past, and to believe that objects that go out of sight continue to exist when we are not observing them. In either case, God can just make people appreciate the good. Evil is not needed to do that.

In conclusion then, we have two reasons that no one can consistently believe in God and in compatibilism. First, libertarian free will is necessary to account for human reasoning. Second, libertarian free will is necessary to reconcile God's morally perfect will from our morally imperfect wills. Moses Maimonides said that anyone who believes in God must believe in both God's sovereignty and in libertarian free will, even if that person cannot explain how both can be true. It might be explainable, or it might be a divine mystery beyond human comprehension. One thing, however, is for certain. No theist can rationally believe in compatibilism. It simply is not an option for anyone who believes in God.