Schellenberg's Free Will Offense
J.L. Schellenberg presented an argument in his 2004 article “The Atheist’s Free Will Offense” against what he claims are two fundamental tenets of theism: (1) that a perfectly loving God would give His creatures libertarian free will; (2) that these finite beings would make progress towards perfection through their own actions because they have free will. According to many present-day philosophers of religion, if a being is perfectly loving then that being will allow finite beings to have free will so that those finite beings can make decisions about how they move towards perfection.
Schellenberg believes that their belief in free will, as well as their belief that finite beings move towards perfection as a result of their own actions, stems from some form of cultural or religious bias rather than actually knowing something about the nature of God. Schellenberg also believes that God can still promote the good of individuals if they do not have free will. If, however, finite beings did have free will then Schellenberg claims that this freedom to choose would be positive evidence for atheism.
Dismantling the puppet objection
John Hick argued that without free will, human responses to God would be fixed and mechanical. According to Hick, humans would lack the ability to make choices on their own and thus would essentially be nothing more than puppets. Schellenberg disputes this argument stating that humans would be just as mentally and emotionally complex in a world in which free will did not exist. Humans would continue to exhibit insight, engage in inter-personal communication, develop intellectually and morally, etc., despite the absence of free will.
Although humans may be determined in a world in which God exists to act based upon their true nature, they would still not be acting against their desires or will; whereas individuals who are placed under hypnosis by a hypnotist do not truly desire to perform the actions they are doing, they are simply forced to do so. Therefore, the type of determinism that exists in a world with a god is different than the type of manipulative control that a hypnotist has over an individual.
Rejoinder: You might find that denying free will removes the mechanism of perfection. In some theistic traditions, humans achieve their highest state when their free will is infused with divine grace. Grace is a gift from God that, when willingly accepted, empowers and elevates human nature, but it requires a free response from the recipient to be effective. If your actions are determined or ordered toward The Good (which is God), you cannot freely participate in this infusion. Love is not a reaction to a stimulus but a self-giving movement toward the good of another. This means the movement of your will toward another person is not forced by your design or your surroundings. If you cannot choose to withhold love, your expression of it lacks the personal quality required for a union with God. A world with free will produces better results because it allows for the highest possible good: a voluntary union of wills between creator and creature.
The sharing model
Rather than viewing the divine-human relationship as analogous to relationships between two adult equals, Schellenberg offers the sharing model as a way to understand the divine-human relationship. He states that the relationship between a parent and a child is a better analogy for understanding the divine-human relationship than that of two adult equals. As parents generally know how their young children will react to comfort or hunger, etc., this knowledge does not preclude an authentic and personal relationship between the parent and the child. While the relationship between a finite creator and an infinite creature is inherently asymmetrical, Schellenberg argues that the advantage of experiencing an eternal closeness with an infinitely rich reality, outweighs the difference between whether this closeness is experienced as libertarian or not.
To justify the sharing model, Schellenberg uses a thought experiment involving a parent and a child to demonstrate how a meaningful relationship exists without free will.
The adult-child analogy: You should consider the relationship adults have with their children, where the parent is responsible for the existence and initial nature of the child.
Predictability and authenticity: Parents know how children will respond to comfort, fear, or hunger, yet this knowledge does not prevent an authentic personal relationship appropriate to their roles.
The genetic engineering scenario: Imagine a child created through sophisticated genetic engineering where the parent possesses deep knowledge of everything the child will do.
Persistent value: Schellenberg argues that even in this deterministic scenario, a meaningful relationship remains possible because it is structured by sharing experiences, such as a parent helping a child understand the universe.
Application to God: If a human parent can have an authentic relationship with a child whose nature they designed, then a Divine creator can have an authentic relationship with persons whose lives are determined.
Rejoinder: A theist might argue that the parent-child analogy fails to account for the goal of a divine relationship. While a parent nurtures a child, the final aim of human parenthood is to raise an independent adult who relates to the parent through choice. If God maintains a relationship where you are permanently in the position of a determined child, the relationship never reaches the maturity of a friendship or a marriage of wills. Theists suggest that God desires a union with persons who possess the dignity of self-possession. Without the ability to reject God, your presence in the relationship is an extension of God’s will rather than the presence of a distinct other. Love is self-giving. To give yourself, you must first possess yourself through the power of choice. If you cannot withhold your will, you are not a partner in a sharing model but an object of divine experience.
The impact on the free will defense
The free will defense claims that for persons to achieve their deepest good, they must have the power to choose for good or ill, making horrific suffering a possibility. Schellenberg argues this is flawed because an everlasting relationship with God mediated in other ways, including the sharing model, is not inferior.
If an option exists that allows God to realize equal value without the risk of evil, a loving creator will choose it. Schellenberg formalizes this logic, and I present it in a simplified form below:
If God exists and free will poses a serious risk of evil, there must be no countering option available to God.
The sharing model is a countering option that realizes equal value without evil.
Therefore, God and free will cannot both exist.
Rejoinder: The acceptance of the first premise depends on how you define divine love and the limits of divine power. Schellenberg operates on the principle that a perfectly loving being will always minimize risk and maximize safety for its beloved. If God is like a parent, and a parent has two ways to provide a child with a great good, the parent should choose the way that involves no risk of horrific suffering. Under this view, if the sharing model provides an equal level of intimacy without the risk of evil, God is morally bound to select it.
Theists often reject this premise by arguing that the options are not actually equal in value, all else considered. You can see this in the selected text. If free will is a necessary condition for a specific type of highest good, such as a voluntary union of wills, then a world without free will is missing that value. From this perspective, the sharing model is not a countering option because it provides a different, lesser category of relationship. The premise fails if the risk of evil is the only way to achieve the specific good of a free response to divine grace.
A central tension lies in whether the sharing model truly realizes equal value. Schellenberg assumes that divine love is primarily concerned with providing a safe and rich experience for creatures. If you define the highest good as the absence of suffering and the presence of shared knowledge, then God must choose the sharing model. However, theists argue that the premise is false because the sharing model lacks the unique value of libertarian freedom. If a free choice to love God is infinitely more valuable than a determined experience of God, then no countering option exists. The risk of horrific evil is a cost God might pay to achieve the specific good of a free creaturely response. Accepting the premise requires you to believe that a determined relationship is just as good as a free one.
Free will as atheistic evidence?
If Schellenberg’s arguments hold, the consensus that God would give creatures free will is false. Most people believe that free will exists in our world. However, if free will exists and is incompatible with the existence of a perfectly loving God who had safer options to provide equal value, then God does not exist. If Schellenberg’s inference right, then the seeming reality of free will provides a reason for denying the truth of theism rather than defending it.
Moving forward requires both sides to agree on the criteria for divine love. You can build consensus by focusing on the shared desire for the flourishing of persons. Atheists and theists can find common ground by examining the limits of parental responsibility. A parent who protects a child from every possible harm while removing their agency is often seen as overbearing. A parent who allows a child to suffer horrific harm for the sake of autonomy is often seen as negligent. By identifying the point where protection becomes control and autonomy becomes abandonment, both sides can clarify what a perfectly loving creator should prioritize. Consensus relies on defining whether the value of a free response outweighs the value of a world without horrific suffering. If you can agree on the weight of these values, the logical gap between the free will defense and the atheist offense begins to close.