Free Will and Determinism

In brief

Every choice you make feels open: you could order the coffee or the tea, tell the truth or lie. Determinism is the thesis that the state of the world at any moment, plus the laws of nature, fixes exactly one next state, so that given the past, only one future is physically possible. If your decision is just one more event in that chain, caused by your brain, which was caused by your genes and history and the physics beneath them, then in what sense could you have done otherwise? The free will problem is the collision between the freedom we take ourselves to have and the causal order the sciences describe. Three families of answer divide the field. Hard determinists accept determinism and conclude that free will is an illusion. Libertarians hold that we do have free will, so determinism must be false at the point of choice. Compatibilists deny the conflict entirely, arguing that freedom, properly defined, never required exemption from causation.

The full treatment

The problem it answers

The stakes are not merely metaphysical. We praise, blame, punish, and reward people on the assumption that they could have acted differently and are the authors of what they do. If no one could ever have done otherwise, the moral and legal machinery of responsibility appears to rest on a mistake. The problem is ancient (the Stoics debated fate and assent in the third century BCE), but it sharpened once early modern physics presented nature as a closed system of prior causes producing later effects. Pierre-Simon Laplace made the threat vivid in 1814 with his image of a vast intelligence that, knowing every particle's position and velocity, could compute the entire future. If a Laplacean demon could in principle have foreseen your decision before you were born, the decision seems less like an origination and more like a link in a chain that was already drawn.

The three positions

Start with the logical map, because the terms trip people up. Two questions are in play: is determinism true, and is free will compatible with it?

Incompatibilists say free will and determinism cannot both hold. They split into two camps depending on which they keep. Hard determinists (a term from William James) keep determinism and drop free will. Libertarians (in this metaphysical sense, unrelated to the political term) keep free will and drop determinism, holding that some human choices are not settled by prior causes.

Compatibilists take the other route: free will and determinism are consistent, so we can affirm both. On this view the freedom worth wanting is not the power to have been uncaused, but the power to act according to your own desires and reasons without external compulsion. A person who chooses the tea because she prefers tea is free; a person forced at gunpoint, or acting under a compulsion she disowns, is not. Whether the universe is deterministic is beside the point.

Libertarianism and the problem of luck

Libertarian free will faces an immediate pressure: if a choice is not determined by prior causes, is it not simply random? A random twitch is no more mine, and no more free, than a determined one. The libertarian owes an account of undetermined actions that are still genuinely authored. The boldest answer is agent causation, defended by Roderick Chisholm in 1964 and developed by Timothy O'Connor. On this view the agent is a cause that is not itself an event: the person, as a substance, originates the action without being pushed to it by a prior state. Others, like Robert Kane, pursue event-causal libertarianism, locating indeterminism in the brain at moments of genuine inner conflict, so that the outcome is not fixed yet still flows from the agent's competing motives. Critics find these accounts either mysterious (agent causation) or unable to escape the luck objection (event causal).

Frankfurt cases

The decisive twentieth-century move came from Harry Frankfurt in 1969. Nearly everyone had assumed the "principle of alternate possibilities": you are morally responsible for an act only if you could have done otherwise. Frankfurt built a counterexample. Suppose Black wants Jones to do a certain thing and implants a device in Jones's brain that will intervene only if Jones shows signs of deciding otherwise. As it happens, Jones decides to do it on his own, and the device never fires. Jones could not have done otherwise, since the device would have stopped him, yet he seems fully responsible, because he acted for his own reasons and the device played no role. If the case works, it severs responsibility from the ability to do otherwise, which is exactly the ability determinism removes. This reopened the door for compatibilism: perhaps determinism threatens alternate possibilities but not responsibility, and responsibility was what we cared about.

The consequence argument

The strongest reply for the incompatibilist side is Peter van Inwagen's consequence argument, given its canonical form in An Essay on Free Will (1983). Stated plainly: if determinism is true, our present acts are the necessary consequences of the laws of nature and events in the remote past. But the remote past is not up to us, and the laws of nature are not up to us. And whatever is necessitated by things not up to us is itself not up to us. Therefore our present acts are not up to us. The argument's force lies in its "transfer of powerlessness" principle: if you are powerless over p, and powerless over the fact that p entails q, you are powerless over q. Compatibilists must either reject that principle or reinterpret "could have done otherwise" so that it survives determinism.

Lineage

The problem runs through the whole Western tradition. Aristotle, in the Nicomachean Ethics, tied responsibility to actions whose origin lies "in the agent" and that are voluntary, an early compatibilist instinct. The Stoics were determinists who nonetheless located freedom in rational assent. Augustine (354 to 430) recast the question theologically as the tension between human freedom and divine foreknowledge. The modern compatibilist line was set by Thomas Hobbes (1588 to 1679) and above all David Hume, whose 1748 Enquiry defined liberty as the power to act according to the determinations of the will, and argued that far from conflicting with necessity, responsibility positively requires it, since we blame people only for acts that flow from their stable character. Immanuel Kant broke the other way, arguing that morality presupposes a freedom the natural world cannot contain, and placing the free self in the noumenal realm beyond causal law. In the twentieth century the debate professionalized around Frankfurt's cases, van Inwagen's argument, and the "reasons-responsive" compatibilism of John Martin Fischer and Mark Ravizza.

The strongest case for it

The most defensible position for most working philosophers is compatibilism, and its case is strong. It begins from ordinary usage: when we call an act free, we contrast it with coercion, addiction, manipulation, and mental illness, not with causation as such. Nobody thinks a decision becomes unfree merely because a neuroscientist could trace it. The freedom we actually value and hold each other to is the capacity to act on our own reasons and to respond to considerations, and that capacity is perfectly real in a lawful universe. Compatibilism also has the virtue of not betting the moral world on a physics question we have not settled. And Frankfurt's insight, that responsibility tracks whether an agent identifies with and acts on her own will rather than whether the universe left the future open, redirects the debate toward something we can inspect: the actual sources of a person's action.

The strongest case against it

The incompatibilists have not been answered to everyone's satisfaction, and their case comes from opposite directions.

Van Inwagen presses that compatibilist "freedom" is a consolation prize. To be free in the sense that matters for desert, and for the reactive attitudes of gratitude and resentment, an agent must be the ultimate source of her action, and the consequence argument shows a determined agent cannot be, because the chain runs back to a past she never touched. Redefining freedom so it survives determinism, he suggests, changes the subject.

Galen Strawson pushes further with his "basic argument" (1994), and it cuts against libertarians too. To be truly responsible for an act, you must be responsible for the self from which it springs, your character and values. But to be responsible for that self, you would have had to shape it, which required a prior self, which you would also have had to shape, and so on without end. Ultimate responsibility demands an infinite regress no finite being can complete. On this view free will in the deepest sense is impossible on any metaphysics, deterministic or not.

There is also an empirical flank. Benjamin Libet's experiments in the 1980s recorded a "readiness potential" in the brain fractions of a second before subjects reported a conscious decision to move, and some, like Sam Harris, read this and later work as evidence that conscious will arrives too late to be the cause. Compatibilists and careful critics reply that Libet's timing data concern trivial spontaneous flicks, not deliberated choices, and do not touch the reasons-responsive capacities at issue, but the challenge keeps the "free will is illusion" position alive in public debate.

Where it stands now

The 2020 PhilPapers survey of professional philosophers found compatibilism the most popular single view, held by a clear plurality, with libertarianism and hard determinism (or the broader "no free will") each holding a substantial minority. No position has won. The frontier has three active fronts. First, the free-will skeptics, led by Derk Pereboom and Gregg Caruso, argue that we should give up basic desert altogether and rebuild criminal justice on forward-looking, public-health lines rather than retribution. Second, work on the neuroscience and psychology of agency continues to test how much of choice is conscious and controllable. Third, the metaphysics of causation and the interpretation of quantum indeterminacy keep the question of whether the world is even deterministic genuinely open, though most agree quantum randomness, if real, would give us chance rather than control. The problem endures because it sits where physics, mind, and morality meet, and moving any one of them moves the other two.

Test yourself

Recall the last decision you agonized over. You experienced it as open: you could have gone either way. Now suppose it was fully determined by your brain state, itself the product of your history and the laws of physics. Ask yourself honestly which part of your sense of freedom that would actually take away. If the answer is "none, I still acted on my own reasons," you are leaning compatibilist. If the answer is "then it was never really mine," you feel the pull of incompatibilism, and the whole debate is contained in that difference of reaction.