How the Free Will Debate Divides Philosophers and Neuroscientists

The Electrical Spike That Preceded a Decision

In the early 1980s, Benjamin Libet at the University of California San Francisco designed an experiment that has never quite stopped generating controversy. Participants were asked to flex their wrist whenever they felt like it, while noting on a clock the position of a revolving dot when they first felt the urge to move. Meanwhile, EEG recorded brain activity, and EMG recorded the moment muscles contracted. The results showed that a brain wave—the Bereitschaftspotential, or readiness potential—began building approximately 550 milliseconds before the muscle moved. But participants reported their conscious intention to move only about 200 milliseconds before movement. The brain was preparing to act roughly 350 milliseconds before the person was consciously aware of intending to act. If the brain initiates action before consciousness registers the decision, what becomes of free will?

Libet himself did not conclude that free will was an illusion. He noted that participants could "veto" the movement up until about 100–200 milliseconds before execution, suggesting that consciousness might exercise a "free won't" even if it does not initiate action. But the readiness potential finding was widely cited as neuroscientific evidence against voluntary control, triggering a debate that spans philosophy of mind, neuroscience, physics, and criminal law. The question is ancient. The experimental tools are new. Neither has yet produced consensus.

The Conceptual Landscape

The free will debate involves three major positions that differ in what they claim is compatible with determinism.

Compatibilism is the dominant position among professional philosophers. A large-scale survey by David Bourget and David Chalmers found that approximately 59% of philosophers identified as compatibilists, compared to 11% hard incompatibilists and 14% libertarians. The compatibilist move is definitional: freedom does not require that actions be uncaused, only that they be caused by the right kind of thing—the agent's own reasons, values, and deliberations, free from compulsion or manipulation.

Harry Frankfurt and the Second-Order Desire

Harry Frankfurt's influential 1969 paper "Alternate Possibilities and Moral Responsibility" in the Journal of Philosophy introduced "Frankfurt cases"—scenarios designed to show that moral responsibility does not require the ability to have done otherwise. If a neuroscientist could monitor your brain and intervene to ensure you made a specific choice if you were about to choose differently—but you chose that way on your own anyway—you seem responsible for your action even though you could not have done otherwise.

• Frankfurt argued that what matters for freedom is not alternate possibilities but whether the will aligns with the agent's higher-order desires—desires about what one wants to want
• An addict who is compelled to take drugs against their own first-order volition lacks freedom even if their action is self-caused
• A person who freely embraces their desires, even impulsive ones, exercises will authentically
• Peter Strawson's 1962 paper "Freedom and Resentment" argued that the question of free will should be grounded not in metaphysics but in the reactive attitudes—resentment, gratitude, blame—that constitute our moral practices

What Quantum Indeterminacy Does and Does Not Help

Some philosophers and physicists have argued that quantum indeterminacy rescues free will from deterministic physics. If at the quantum level events are genuinely undetermined—not merely unpredictable, but uncaused—then the causal closure of physics at the macro level is not as airtight as classical determinism assumed.

The libertarian free will theorist Robert Kane at the University of Texas has argued that quantum indeterminacy at the neural level can introduce genuine causal openness in human decision-making, particularly in "self-forming actions"—moments of genuine inner conflict where multiple outcomes are genuinely possible. However, most philosophers find this move problematic: a random quantum event does not obviously increase agency. If your decision is partly determined by quantum noise, it is no more "yours" than a decision determined by prior physical causes. Randomness and freedom are distinct concepts.

Revisiting Libet: What the Readiness Potential Actually Shows

Subsequent research has significantly complicated the interpretation of Libet's original findings. Aaron Schurger and colleagues, publishing in the Proceedings of the National Academy of Sciences in 2012, proposed that the readiness potential may not be a preparatory signal for a predetermined action but a slow accumulation of neural noise that crosses a threshold triggering the voluntary movement. The build-up precedes the decision not because the brain has pre-committed to acting, but because the neural dynamics that constitute decision-making take time and involve stochastic fluctuations.

John-Dylan Haynes at the Charité in Berlin extended Libet's paradigm using fMRI and found that patterns of brain activity in the prefrontal and parietal cortex could predict whether a participant would press a button with their left or right hand up to 7–10 seconds before the participant reported conscious awareness of the decision. The temporal gap between unconscious brain processes and conscious awareness was dramatically larger than Libet's 550 milliseconds. Whether this implies no free will, that free will is distributed across unconscious processes, or that conscious awareness is simply a late-arriving report on a process that is nevertheless the agent's own remains contested.

Daniel Dennett at Tufts University has argued that the framing of the free will question as a confrontation between neuroscience and agency involves a category error. The relevant question is not whether the brain initiates action before consciousness arrives, but whether the agent—a complex system comprising brain, body, history, and values—is the appropriate unit of analysis for moral responsibility. Libet's experiment does not show that the person is not in control. It shows that consciousness is not the whole of the person. That distinction changes everything.