Free Will: Do You Actually Choose?

Follows: Philosophy of Mind: The Hard Problem of Consciousness and Existentialism: Freedom, Responsibility, and the Absurd

In this deep dive
1. What the question actually is
2. Hard determinism and the case against free will
3. Libertarian free will: the case for a real fork in the road
4. Compatibilism: free will without the fork
5. Libet's clock, and how a model from 2012 changed its reading
6. What's at stake: blame, punishment and believing in free will
7. Where the debate stands, and how to hold it

Raise your right hand. Or don't. Either way, you just did something it feels like you could have done differently.

Now picture the whole universe one second before you decided: every particle, every neuron, every leftover trace of your childhood, your mood, the coffee in your blood. If that exact state, run forward under the laws of nature, could only ever produce the hand going up, in what sense did you choose?

That's the free will problem. Existentialism gave you Sartre's answer: we are "condemned to be free," and pretending otherwise is bad faith. Philosophy of Mind asked what a mind even is, and how it relates to the brain. This deep dive asks whether a mind made of brain can choose at all. It's one of the few philosophical questions that also turns up in laboratories and courtrooms.

Every major position gets its strongest case here. None of them wins. They're still fighting, and some very clear-headed people sit on each side.

Part 1 — What the question actually is

Most confusion about free will comes from not saying what "free" means. Philosophers have separated two ideas that ordinary talk runs together.

1. Leeway: the ability to do otherwise. At the moment you chose coffee, could you really have chosen tea, with everything else exactly as it was?
2. Sourcehood: being the real origin of what you do. Did the choice come from you (your reasons, your character, your judgment), or did something pass through you like water through a pipe?

The usual threat to both is determinism: the thesis that the state of the world at one time, together with the laws of nature, fixes everything that happens after. Its classic picture comes from the French mathematician Pierre-Simon Laplace, who imagined an intelligence that knew every force and every position at one instant. "To it nothing would be uncertain," he wrote, "the future as well as the past would be present to its eyes."

Is the world actually like that? Physics doesn't settle it cleanly. Standard readings of quantum mechanics are indeterministic, but there are deterministic versions of quantum theory, such as Bohmian mechanics, and chaos means even a deterministic world can look random. Determinism is an open question, and as you'll see, some arguments against free will work whether or not it's true.

Why care? Mostly because of desert: the idea that people can deserve praise or blame, reward or punishment, just because of what they did, not because blaming them will do any good. The philosopher Derk Pereboom calls this "basic desert," and it's what most of the fight is really about. Nobody doubts that blame can be useful. The question is whether anyone ever truly has it coming.

That gives three main positions:

Hard determinists and libertarians agree on one thing: free will and determinism can't both be true. They're both incompatibilists. They just make opposite bets on which one is real.

Part 2 — Hard determinism and the case against free will

The case against free will is old. In the seventeenth century Spinoza was already rejecting it, and he thought that seeing through the illusion could bring a kind of calm. The modern case comes in three strong arguments.

1. The Consequence Argument. Peter van Inwagen gave it its best-known short form in An Essay on Free Will (1983):

"If determinism is true, then our acts are the consequences of the laws of nature and events in the remote past. But it is not up to us what went on before we were born, and neither is it up to us what the laws of nature are. Therefore, the consequences of these things (including our present acts) are not up to us."

Each step looks hard to deny. You can't change the Big Bang, and you can't break a law of physics. If those two together fix your choice, the choice isn't up to you. (Van Inwagen himself used this argument to defend libertarianism. A hard determinist takes the same premise and adds: and determinism, or something close enough, is true.)

2. The "luck" horn. Suppose determinism is false and some of your choices are undetermined. Does that help? Picture a choice to tell the truth or lie. If, with the past exactly the same, you might have done either, then nothing about you settles which one happens. Run the moment a hundred times and get different results. That looks less like freedom and more like a coin toss inside your skull. So the skeptic has a dilemma: if your choices are determined, they come from the past; if they're undetermined, they're partly a matter of luck. Either way you're not in control in the deep sense that desert needs. This is why Pereboom calls his view hard incompatibilism: free will of the kind desert requires is ruled out by determinism and by the most plausible kinds of indeterminism.

3. The Basic Argument. Galen Strawson's version needs no physics at all. It goes back to Nietzsche, who called the idea of a self-caused self (causa sui) "the best self-contradiction that has been conceived so far." Strawson's argument, in brief:

1. You do what you do because of the way you are, mentally speaking.
2. So to be truly responsible for what you do, you'd have to be truly responsible for the way you are.
3. But to be responsible for the way you are, you'd have to have chosen to be that way, using a self you already had. And where did that self come from?
4. The regress has no bottom. Nothing can be the cause of itself.
5. So no one is ultimately responsible.

If you didn't make yourself, and your choices flow from the self you didn't make, the skeptic asks: what exactly do you deserve?

The strongest version of the skeptic's case isn't "you're a robot, so nothing matters." Modern skeptics such as Pereboom, Gregg Caruso and Bruce Waller argue that giving up basic desert would make us better: less vengeful, more curious about causes, more humane in punishment. These "optimistic skeptics" argue that meaning, good relationships and moral judgment survive. You can still call an act wrong, still protect people from danger, still try to rehabilitate and deter. What goes is the backward-looking sense that someone deserves to suffer. The neuroscientist Robert Sapolsky's book Determined (2023) makes a version of this case aimed at general readers.

Part 3 — Libertarian free will: the case for a real fork in the road

Libertarians start from something every one of us experiences: deliberation. When you weigh a job offer, the options feel open, and it feels like it's up to you to close them. Libertarians say we should take that experience at face value unless forced not to, and they deny that we're forced.

They also accept the Consequence Argument. If determinism were true, we wouldn't be free. They conclude it isn't true, at least for some human choices. Their hard task is answering the luck problem from Part 2: showing how an undetermined choice can still be yours. There are three main answers.

Agent causation. Thomas Reid in the eighteenth century and Roderick Chisholm in the twentieth argued that a free choice is caused not by prior events but by the agent as such, a different kind of cause. Chisholm's 1964 lecture "Human Freedom and the Self" put it boldly:

"Each of us, when we act, is a prime mover unmoved. In doing what we do, we cause certain events to happen, and nothing—or no one—causes us to cause these events to happen." — Roderick Chisholm

Contemporary defenders such as Timothy O'Connor and Helen Steward have built careful versions of this view. Critics ask what an "agent cause" is, if it's not just events in the brain, and whether the evidence supports any such thing. Pereboom, for one, thinks agent causation is the only thing that could give us free will, but that the evidence is against it.

Event-causal libertarianism. Robert Kane, the best-known defender, keeps ordinary causation but adds indeterminism at a few key moments. Most of what you do flows from your character. That's fine, says Kane, as long as somewhere in your past you made self-forming actions: hard choices, often moral ones, where you were torn (say, between a duty and a temptation) and made an effort of will whose outcome was genuinely open. Either way the choice went, it went for your reasons, and you endorsed it. Those choices shape the character your later actions come from. Kane compares the indeterminacy of the effort to the indeterminacy of a particle's position in physics. Critics answer that if it's undetermined which way the effort goes, the outcome still looks lucky. Kane replies that luck isn't incompatible with responsibility when the agent is trying, succeeds and owns the result.

Non-causal views. A smaller group (Carl Ginet and others) says a free choice needs no cause at all. It's simply an action, intrinsically something you do, explained by your reasons without being caused by them. Most libertarians find that too thin.

The libertarian's bet: the felt openness of deliberation is not an illusion, and science has not shown that it is.

Is the libertarian bet empirically crazy? Not obviously. Physics doesn't establish determinism, and whether quantum indeterminism ever matters at the level of brain decisions is unknown. Libertarians need only that a few decisions, at the right moments, aren't fixed in advance. What they owe us is an account of why such an opening would amount to control rather than noise. That is still the central argument against them.

Part 4 — Compatibilism: free will without the fork

Compatibilists make a different move. They say both sides have the wrong picture of freedom. Being free isn't being uncaused. It's being caused in the right way.

The classical version. Thomas Hobbes defined liberty as "the absence of all the impediments to action that are not contained in the nature and intrinsical quality of the agent." David Hume, in his Enquiry Concerning Human Understanding (Section VIII), put it more simply:

"By liberty, then, we can only mean a power of acting or not acting, according to the determinations of the will; that is, if we choose to remain at rest, we may; if we choose to move, we also may." — David Hume

This "hypothetical liberty," Hume said, "is universally allowed to belong to every one who is not a prisoner and in chains." On this view the opposite of freedom isn't causation. It's coercion: the gun to your head, the locked door, the compulsion. A determined choice can be perfectly free, so long as it flows from what you want.

Frankfurt's two revolutions. Classical compatibilism had a weak spot: what about a compulsive drug user who does exactly what he wants? Harry Frankfurt answered it in two papers that reshaped the field.

The first, "Alternate Possibilities and Moral Responsibility" (1969), attacked the assumption everyone shared: that you're only responsible if you could have done otherwise. Here's a standard version of a Frankfurt case, as John Martin Fischer tells it. A neurosurgeon, Black, secretly implants a device in Jones's brain. If Jones shows any sign of voting for Candidate A, the device will make him vote for Candidate B. But Jones decides, entirely on his own, to vote for B, and the device never does anything. Jones couldn't have done otherwise, because the device would have stepped in. Yet he seems fully responsible, because he chose it himself. If that's right, what matters for responsibility isn't alternative possibilities. It's what actually happened inside the agent: whether he was the source.

The second paper, "Freedom of the Will and the Concept of a Person" (1971), offered a theory of that source. People have first-order desires (to take the drug, to skip the gym) and second-order desires about which desires they want to move them. Frankfurt's unwilling addict wants the drug but, at the second level, wants that desire not to win. When it wins anyway, he isn't acting freely. His willing addict has the same craving but embraces it. The difference isn't causation, since both cravings have causes. It's whether the will is the one you identify with.

Reasons-responsiveness. Later compatibilists, notably Fischer and Mark Ravizza, focused on whether your decision-making would respond to reasons. If you'd been shown a good reason not to do it, would you have stopped? A person in the grip of compulsion wouldn't. A normal agent would. That capacity, they argue, is what responsibility tracks.

Strawson's change of subject. P. F. Strawson's essay "Freedom and Resentment" (1962) argued that the whole metaphysical debate misses how responsibility actually works. Our practices of blame rest on reactive attitudes: resentment, gratitude, indignation, forgiveness. They're responses to the "quality of others' wills" toward us. If someone injures you by accident, you don't resent it the way you would if they'd done it out of contempt. We excuse people when they turn out to have meant no ill. We exempt some people, young children for instance, or people whose illness impairs ordinary relationships, when better behavior can't reasonably be demanded of them. Strawson argued that our commitment to ordinary relationships is "too thoroughgoing and deeply rooted" for us to give these attitudes up because of a theory, and that determinism, even if true, wouldn't make everyone an exempt case.

Dennett's "worth wanting." Daniel Dennett's Elbow Room: The Varieties of Free Will Worth Wanting (1984) and Freedom Evolves (2003) pushed the practical line hardest. The kind of freedom the skeptic says we lack, an uncaused self outside nature, is something nobody needs, Dennett argued. The kinds worth wanting are real: the ability to deliberate, to foresee consequences, to respond to reasons and to be talked out of things. As the title Freedom Evolves suggests, he thought these capacities were built by evolution and culture.

The compatibilist's bet: freedom is a skill a brain can have or lack, not a hole in the laws of physics.

The strongest objections to compatibilism:

• The Consequence Argument still stands. Compatibilists have to say there's a sense in which you "could have done otherwise" even if the past and laws fixed what you did. David Lewis argued that van Inwagen's reasoning trades on two senses of "able to break a law." Incompatibilists reply that only the stronger sense matters, and they find the compatibilist sense too weak. This exchange is still going.
• Manipulation arguments. Alfred Mele imagines a goddess, Diana, who designs a zygote so that the person it grows into will commit a particular murder thirty years later, for reasons he fully endorses and in a way that responds to reasons. Many people judge that this person isn't responsible, because he was set up. But how is that different, the incompatibilist asks, from being set up by the Big Bang and the laws of nature? Pereboom's "four-case argument" makes the same point in steps.
• The Frankfurt cases may not work. David Widerker, Carl Ginet and Robert Kane pointed out a dilemma: if Jones's choice is undetermined, Black can't know in advance what Jones will choose, so the device can't guarantee the outcome; if it's determined, the case assumes the very thing incompatibilists deny. Frankfurt's defenders have built increasingly elaborate cases to escape it. Whether any succeeds is, in the Stanford Encyclopedia of Philosophy's words, "hotly contested."

Part 5 — Libet's clock, and how a model from 2012 changed its reading

In 1965 two German researchers, Kornhuber and Deecke, reported a slow electrical buildup over the scalp before people make voluntary movements. They called it the Bereitschaftspotential, the readiness potential (RP).

In 1983 the American neuroscientist Benjamin Libet and colleagues used it in one of the most famous experiments in the history of neuroscience (Brain, vol. 106). Participants watched a spot of light circling a clock face, made a simple voluntary movement whenever they felt like it, and reported where the dot was when they first felt the urge or intention to move, a moment called W. Meanwhile EEG recorded the readiness potential.

The brain's buildup started roughly a third of a second before people said they'd decided. Libet concluded that a voluntary act "can begin unconsciously." For decades this was cited as scientific proof that conscious will is an afterthought. Libet himself didn't go that far. He proposed that consciousness could still veto a movement in the last moments, which came to be nicknamed "free won't."

Later work seemed to extend the gap. In 2008 Soon, Haynes and colleagues reported that brain activity could predict which of two buttons people would press "up to 10 s" before they were aware of deciding. In 2011 Fried and colleagues, recording 1,019 single neurons in twelve people, found that a population of 256 neurons in the supplementary motor area could predict the decision to move with better than 80% accuracy 700 ms before awareness.

Then the reading of the experiment changed. Here are the main reasons.

1. The signal may not mean "I've decided." In 2010 Trevena and Miller compared brain activity before a decision to move with activity before a decision not to move. They found no evidence of stronger signals before moving. If the buildup appears whether or not you go on to act, it can't simply be the brain committing to the act.

2. The 2012 accumulator model. Schurger, Sitt and Dehaene (PNAS, 2012) offered a new explanation of what the RP is. The brain's activity is always fluctuating at random. In a Libet task, where you're told to move whenever you like, there's only a weak, steady push toward moving. The movement happens when random noise plus that push crosses a threshold. Now here's the trick. If you line up hundreds of trials by the moment of movement and average them, the random fluctuations that happened to precede each crossing add up to a smooth ramp. The ramp is partly an artifact of how the data are averaged, not a decision building up. Their model fit the existing data and made a new prediction, which they confirmed in a second experiment: fast responses to unpredictable interruptions were preceded by the same slow buildup, even when people weren't preparing to move at that moment.

On the accumulator reading, the readiness potential is what the brain's background noise looks like when you average it backwards from a movement.

3. Arbitrary flicks aren't real decisions. In 2019 Maoz and colleagues compared arbitrary choices with deliberate ones, decisions about which nonprofit should receive a $1,000 donation. They found the usual readiness potential before arbitrary choices. Before deliberate ones it was "strikingly absent." Movements made for no reason may tell us little about decisions that matter.

4. The prediction numbers are modest. A 2021 review by Schurger, Hu, Pak and Roskies (Trends in Cognitive Sciences) points out that the fMRI predictions seconds ahead were statistically above chance but "only marginally better than a random guess." The same review notes that veto decisions also show a readiness potential, and that one study put the "point of no return" for cancelling a movement at about 200 ms before it, roughly when Libet thought the veto window opened. That makes "free won't" hard to defend as he framed it. The review's overall conclusion is that the stochastic model "strips away the conceptual basis for relating the RP to the timing of conscious awareness."

What this does and doesn't show. The reassessment undercuts the popular claim that neuroscience has proved our decisions are made unconsciously before we know it. It doesn't prove free will either. Schurger and colleagues note that stochastic models "may yet pose problems for free will, but not because of the relative timing of awareness and volition." A choice nudged over a threshold by neural noise is hardly the libertarian's ideal. And notice that none of this touches the philosophical arguments in Parts 2 to 4. The Consequence Argument and the Basic Argument were never about milliseconds.

Part 6 — What's at stake: blame, punishment and believing in free will

This is where the debate leaves the seminar room.

Punishment. Broadly, there are two kinds of justification. Retributivism says wrongdoers deserve to suffer in proportion to their wrong, whatever the consequences. Consequentialist views justify punishment by its effects: deterrence, protection, rehabilitation. Retributivism needs basic desert. If the skeptics are right, it loses its footing, and only forward-looking reasons remain.

In 2004 the psychologists Joshua Greene and Jonathan Cohen published "For the law, neuroscience changes nothing and everything." Legal doctrine, they argued, can absorb any discovery about the brain, because the law asks about capacities, not metaphysics. But vivid neuroscience will change people's intuitions about free will, and they predicted and recommended "a shift away from punishment aimed at retribution in favour of a more progressive, consequentialist approach."

The quarantine model. Pereboom and Caruso propose an alternative they compare to public health. A person carrying a dangerous disease isn't to blame for it, yet we may quarantine them to protect others. Likewise, a dangerous offender may be incapacitated even if no one is ever basically deserving, on the ground of society's right to self-defense. They argue the analogy also limits punishment. Just as it's wrong to treat a disease carrier more harshly than needed, it would be wrong to make prisons more punishing than protection requires. Caruso extends it into a "public health" approach that puts prevention, education and mental health care ahead of punishment. The model has its critics, who press familiar worries, such as whether it treats offenders merely as tools for public safety. Defenders reply that because the justification is defense against a real threat, it can't license confining anyone who poses none.

What the law actually asks. Courts don't wait for philosophers. The classic insanity test, the M'Naghten rule from an English case of 1843, asks whether at the time of the act the defendant didn't know what he was doing, or didn't know it was wrong. That's a capacity question, and it fits compatibilism naturally: the law cares whether you could understand and respond to reasons, not whether your neurons were caused. Defenders of current practice say this is exactly right. Skeptics say it quietly assumes a kind of desert it can't justify.

Would believing in determinism make people worse? A famous 2008 study by Vohs and Schooler found that people who read a passage arguing that free will is an illusion cheated more on a later task. It was widely cited as a warning. Then came the replications. In 2020 Nadelhoffer and colleagues ran five studies, four of them pre-registered, with samples of up to 982 people, and their attempts to reproduce the effect were "largely unsuccessful." In 2023 a meta-analysis by Genschow and colleagues pooled 145 experiments, 95 of them unpublished. Anti-free-will messages did reduce people's stated belief in free will. But the authors "could not find evidence for downstream consequences" on behavior. The worry that disbelief in free will will unleash bad behavior isn't supported by the current evidence.

The live debate in one book. For a direct encounter, read Just Deserts (2021), a book-length debate between Dennett, the compatibilist, and Caruso, the skeptic, about whether anyone ever deserves blame and punishment, and what should replace desert if no one does.

Part 7 — Where the debate stands, and how to hold it

In 2020 David Bourget and David Chalmers surveyed professional philosophers worldwide about their views on dozens of questions. On free will, the roughly 1,760 who answered split like this:

Two cautions. First, a survey of opinion isn't an argument. Majorities of experts have been wrong before, and the minority positions here are held by serious people for serious reasons. Second, compatibilism winning a vote doesn't settle much, because compatibilists disagree among themselves about what free will is.

What can a careful non-specialist take away?

• Say which freedom you mean. "Free from coercion," "could have done otherwise with the same past," and "the ultimate source of yourself" are three different claims. Many arguments that look like disagreements are people talking about different ones.
• The neuroscience hasn't settled it. The Libet story is a better example of how science revises its readings than proof that you're a passenger in your own brain.
• Most of what matters in practice survives on every view. On all three positions, people can deliberate, respond to reasons and change. Protecting others from harm stays justified, and so do understanding causes and holding people to standards. The real disagreement is about one thing: whether anyone ever deserves to suffer for what they did.
• Your stance has real effects even if the metaphysics doesn't. If you lean skeptical, you may find yourself less punitive and more curious. If you lean compatibilist, you'll keep blame but tie it to capacities. If you lean libertarian, you'll take the felt openness of choice as real data. Each of these comes with a cost you should be honest about.

Whether or not your choice was settled in advance, you still have to make it. Nobody can deliberate on your behalf.

That last point is close to where Sartre would end, from the other side. You can't step outside your situation to check whether you're free. You can only choose, and see what kind of chooser you turn out to be.

The takeaway

The free will debate turns on two ideas, leeway (could you have done otherwise?) and sourcehood (were you the real origin?), and on whether anyone can deserve blame. Hard determinists and skeptics argue that whether choices are determined or partly random, no one is ultimately self-made, so basic desert fails. Libertarians argue that some choices are genuinely open and still ours, and they owe an answer to the luck problem. Compatibilists argue that freedom means acting from your own reasons without coercion, which a caused brain can do. Frankfurt cases and Strawson's reactive attitudes are their strongest tools, and manipulation arguments are the strongest replies to them. The Libet experiments once seemed to show the brain decides before you know it. Later work, especially Schurger's 2012 accumulator model and Maoz's 2019 study of deliberate choices, undercut that reading without settling the philosophy. What's most at stake is punishment: retribution needs desert, and the alternatives have costs of their own.

Try this: pick one decision from this week, one you're proud of and one you regret. For each, write three lines. Leeway: could I have done otherwise, with everything exactly the same? Source: did it come from reasons I endorse, or from something I'd disown? Desert: do I (or does anyone) deserve credit or blame for it, or is it only useful to treat it that way? Notice which question your answers turn on. That's your own position, showing itself.

Sources & Further Reading

Sources
- Overviews and arguments: Stanford Encyclopedia of Philosophy entries "Free Will," "Compatibilism," "Arguments for Incompatibilism," "Incompatibilist (Nondeterministic) Theories of Free Will," "Skepticism About Moral Responsibility," "Moral Responsibility" and "Causal Determinism" (Laplace quotation; Bohmian mechanics and chaos); Internet Encyclopedia of Philosophy, "Free Will."
- Primary texts: Hume, An Enquiry Concerning Human Understanding, Section VIII; van Inwagen, An Essay on Free Will (1983), p. 16; Chisholm, "Human Freedom and the Self" (Lindley Lecture, 1964); Frankfurt, "Alternate Possibilities and Moral Responsibility," Journal of Philosophy (1969), and "Freedom of the Will and the Concept of a Person" (1971); P. F. Strawson, "Freedom and Resentment," Proceedings of the British Academy (1962); Dennett, Elbow Room (1984) and Freedom Evolves (2003); Nietzsche, Beyond Good and Evil §21; G. Strawson's Basic Argument (1994, 2011).
- Neuroscience: Libet, Gleason, Wright & Pearl, Brain 106 (1983); Soon, Brass, Heinze & Haynes, Nature Neuroscience (2008); Trevena & Miller, Consciousness and Cognition (2010); Fried, Mukamel & Kreiman, Neuron (2011); Schurger, Sitt & Dehaene, PNAS (2012); Maoz, Yaffe, Koch & Mudrik, eLife (2019); Schurger, Hu, Pak & Roskies, "What Is the Readiness Potential?", Trends in Cognitive Sciences (2021), which also cites Kornhuber & Deecke (1965).
- Responsibility and punishment: Greene & Cohen, Philosophical Transactions of the Royal Society B (2004); Pereboom and Caruso's quarantine model as summarized in the SEP's "Skepticism About Moral Responsibility"; Cornell Legal Information Institute, "M'Naghten rule."
- Belief in free will: Vohs & Schooler, Psychological Science (2008); Nadelhoffer et al., Cognition (2020); Genschow et al., Personality and Social Psychology Review (2023).
- Survey: Bourget & Chalmers, "Philosophers on Philosophy: The 2020 PhilPapers Survey," Philosophers' Imprint (2023).

Go deeper
- 🌐 Stanford Encyclopedia of Philosophy — Free Will. The best single map of the debate, with every position stated fairly.
- 🌐 Stanford Encyclopedia of Philosophy — Skepticism About Moral Responsibility. The skeptic's case at full strength, including the quarantine model and its critics.
- 🌐 Schurger et al., "What Is the Readiness Potential?" (free full text). The clearest account of why the Libet story changed, by the scientists who changed it.
- 📘 Four Views on Free Will — John Martin Fischer, Robert Kane, Derk Pereboom and Manuel Vargas. Four leading philosophers each defend their view and answer the others.
- 📘 Just Deserts — Daniel Dennett and Gregg Caruso. A compatibilist and a skeptic argue it out over punishment, in plain language.
- 📘 Elbow Room — Daniel Dennett. The classic case that the freedom worth wanting is the kind we have.
- 📘 Determined — Robert Sapolsky. A neuroscientist's full-throated case against free will; read it with the SEP entry to hear the replies.
- 📘 An Enquiry Concerning Human Understanding — David Hume, Section VIII. Free at Project Gutenberg. The founding statement of compatibilism, under twenty pages.