Few philosophical questions intersect as intimately with everyday life as the question of free will. Every time we praise someone for their courage, blame someone for their cruelty, credit ourselves with a good decision, or hold a criminal accountable for a crime, we are making implicit assumptions about the freedom of human action. If those assumptions are wrong -- if our choices are determined by prior causes that were themselves determined, all the way back to conditions that existed before we were born -- then the entire framework of moral responsibility that organizes our social lives rests on a mistake.
The question has become more urgent, not less, as neuroscience has advanced. Benjamin Libet's experiments in the 1980s, subsequently refined and contested, appeared to show that brain activity predicting a movement precedes conscious awareness of the intention to move by hundreds of milliseconds. David Eagleman and others have documented the extent to which behavior is shaped by factors -- brain injuries, prenatal exposures, genetic variants, adverse childhood experiences -- that the person who ends up in the dock of a criminal court did not choose. The mechanistic picture of the brain as a causal system operating according to physical laws that was once confined to philosophy seminars now appears regularly in legal proceedings, psychiatric diagnoses, and public debates about punishment and rehabilitation.
Yet many philosophers argue that the threat is overstated, and that the neuroscientific arguments rest on a confusion about what free will actually is. Daniel Dennett, the most persistent and influential defender of this position, argues that no serious philosopher has ever believed in the kind of uncaused causation that hard determinists deny, and that the compatibilist conception of freedom -- freedom as responsiveness to reasons, as acting from one's own values without compulsion or manipulation -- is entirely consistent with everything neuroscience has discovered and is the only kind of freedom worth wanting. The debate between Dennett and Sam Harris, who published 'Free Will' in 2012 arguing that the whole concept is an illusion, is the clearest recent example of a genuine philosophical disagreement about matters of real practical consequence.
"We are not the authors of our thoughts and actions in the way that people generally suppose." -- Sam Harris, Free Will (2012)
Key Definitions
Libertarian free will: The view that humans have a kind of freedom that is genuinely incompatible with determinism -- the ability to have done otherwise in exactly the same circumstances, not merely in different circumstances. Not to be confused with the political philosophy of libertarianism.
Hard determinism: The position that every event, including every human thought and action, is the inevitable result of prior causes, and that this causal determination is incompatible with free will, which therefore does not exist.
Compatibilism: The view that free will and determinism are not in conflict. On compatibilist accounts, free will consists in acting from one's own reasons, values, and deliberative processes without coercion or manipulation -- a condition that can be satisfied even if those processes are themselves causally determined.
Hard incompatibilism: Derk Pereboom's term for the position that free will is incompatible with both determinism and indeterminism, and that neither obtains in a way that could ground the kind of free will required for basic desert moral responsibility.
Readiness potential (Bereitschaftspotential): The brain-wave pattern identified by Hans Helmut Kornhuber and Lüder Deecke in 1964, preceding voluntary movements by up to a second, which Libet's experiments showed begins before conscious awareness of the intention to move.
Major Positions on Free Will
| Position | Core Claim | Key Proponent | Implication for Moral Responsibility |
|---|---|---|---|
| Libertarian free will | Humans can genuinely do otherwise; freedom is incompatible with determinism | Kane, Chisholm | Full moral responsibility grounded in genuine choice |
| Hard determinism | Every event is causally determined; free will does not exist | Holbach | Moral responsibility is an illusion |
| Compatibilism | Free will and determinism are compatible; freedom = acting from one's own reasons | Dennett, Hume, Frankfurt | Moral responsibility survives determinism |
| Hard incompatibilism | Freedom is incompatible with both determinism and indeterminism | Pereboom | Retributive punishment is unjustified; forward-looking justice only |
| Free will skepticism | The folk concept of free will is incoherent; we should abandon it | Harris | Compassion replaces blame; justice becomes therapeutic |
The Classical Problem
A World of Causes
The determinism that poses the challenge to free will is the thesis that the state of the physical world at any moment, together with the laws of nature, fixes the state at every subsequent moment. In a deterministic universe, a superintelligence with complete knowledge of the current state could in principle predict every future event, including every human action, thought, and decision. The future is not open; it is fixed by the past.
This picture has been the scientific worldview since at least Newton, though the discovery of quantum mechanics in the twentieth century introduced genuine randomness into the physical picture at the subatomic level. Whether quantum indeterminacy propagates up to the level of neural processes in ways that could matter for human decision-making is disputed. And as Sam Harris notes, even if it does, randomness is not what libertarian free will requires: a random fluctuation that determines a decision is not an exercise of the will.
The philosophical problem of free will predates modern physics. Ancient Greek debates about fate and responsibility, theological debates about divine foreknowledge and human freedom, and early modern debates in the wake of Cartesian mechanism all turned on versions of the same problem: if my actions are the product of causes, and those causes were the product of prior causes, and ultimately the causal chain runs back to circumstances before my birth, in what sense are they mine?
Compatibilism's Long History
Compatibilism -- the view that the determinism problem is a pseudo-problem arising from a confused conception of freedom -- has a long history. Thomas Hobbes argued in 'Leviathan' (1651) that liberty is simply the absence of external impediments to motion; a free action is one that is not hindered by external constraint, regardless of its causal history. David Hume argued in 'An Enquiry Concerning Human Understanding' (1748) that the very practice of holding people responsible presupposes that their actions flow from their character -- that they are causally connected to their stable dispositions -- and that this causal connection is what responsibility requires, not freedom from causation.
John Stuart Mill, in 'A System of Logic' (1843), argued that the confusion arises from conflating two senses of 'could have done otherwise.' In the metaphysical sense, nothing that has happened could have been otherwise, given the laws of nature and initial conditions. In the practical sense, a person could have done otherwise if they had chosen differently -- if they had different desires, or if circumstances had been different. It is this hypothetical sense that matters for moral and legal responsibility, and it is compatible with determinism.
The Neuroscience Challenge
Libet's Experiments
Benjamin Libet at the University of California San Francisco published his landmark experiments on voluntary action in 1983. Participants sat in front of a clock face and flexed their wrist at times of their own choosing. They were asked to report the position of the clock hand at the moment they first became aware of the intention (or urge or wish) to move. Electroencephalographic electrodes recorded brain activity. The readiness potential -- a slow negative shift in electrical potential over the motor cortex -- began approximately 550 milliseconds before the movement. Participants reported the onset of conscious intention approximately 200 milliseconds before movement. Libet concluded that the brain was initiating the action approximately 350 milliseconds before the person was consciously aware of deciding to act.
Libet himself did not conclude that free will was illusory. He proposed that conscious will might function as a veto: even if the brain initiates movements unconsciously, the person can consciously cancel a movement that has been initiated but not yet executed. Some subsequent researchers, including Chun Siong Soon and colleagues in a 2008 fMRI study published in 'Nature Neuroscience,' found that brain patterns could predict a participant's choice of whether to press a button with the left or right hand up to ten seconds before the participant reported making the decision.
Problems with the Libet Interpretation
The Libet paradigm has attracted sustained methodological and philosophical criticism. Adina Roskies and other philosophers have argued that the experiments test a very specific kind of action -- arbitrary, trivial, and unmotivated -- that may not be representative of the complex deliberate decisions that matter for questions about moral responsibility. The question of when to flex one's wrist without reason is not the kind of choice for which free will is philosophically important.
Patrick Haggard and Martin Eimer found that the late component of the readiness potential -- the specific lateralized readiness potential that predicts which hand will move -- arises much later than Libet's 550 millisecond figure, closer to 200 milliseconds before movement, reducing the gap between neural preparation and conscious experience substantially. Aaron Schurger, Jacobo Sitt, and Stanislas Dehaene published research in 2012 suggesting that the readiness potential may reflect the stochastic fluctuation of neural activity that eventually crosses a threshold, rather than a specific decision process, which would undermine the interpretation that the brain is 'deciding' before the person does.
Harris versus Dennett
Harris's Eliminativist Position
Sam Harris's 'Free Will' (2012) is a short, direct argument that free will in any meaningful sense is an illusion. Harris's core claim is that we are not the authors of our thoughts: the thoughts, desires, and intentions that arise in consciousness are not produced by a choosing self but simply arise, from neural processes that we did not cause and cannot fully observe. When a thought occurs to me, I did not choose it. When a desire arises, I did not select it. The self that appears to choose is itself a product of prior causes.
Harris draws explicitly on Libet and on the general neuroscientific picture of the brain as a causal system. His conclusion is that the self who seems to decide is itself an emergent narrative construction, not an uncaused cause standing outside the causal order. Even our sense of deliberation and reflection -- the feeling that we are weighing options and choosing -- is itself a causal process in the brain, shaped by factors we did not choose.
Harris argues that abandoning the illusion is not merely accurate but beneficial: it naturally produces compassion for others, since their character and actions are as much products of causes they did not choose as our own are; it reduces the kind of excessive self-congratulation that comes from believing that our successes are purely our own achievements; and it clarifies the proper aims of criminal justice by removing retribution's justification.
Dennett's Compatibilist Response
Daniel Dennett has responded to Harris in several forums, arguing that Harris is attacking a conception of free will that no serious philosopher has defended and that bears no relation to the free will that actually matters for our practices of praise, blame, and responsibility. Dennett's 'Freedom Evolves' (2003) argues that free will is a natural phenomenon, a capacity that has evolved in creatures with sufficient cognitive complexity to model their own behavior, anticipate consequences, and revise their actions in light of reasons.
The relevant question, Dennett argues, is not whether our actions are caused -- of course they are -- but whether they are caused in ways that make the agent responsive to reasons. A person who acts from their own values, who can be deterred by consequences, who can reflect on and revise their desires, who was not coerced or manipulated, is free in every sense that matters. The fact that this person's values and reasoning capacities were themselves caused by prior events does not undermine their freedom; it constitutes it.
Dennett charges that Harris's argument proves too much: by the same logic, 'you' did not choose the immune response that fought off last winter's flu, so your health is illusory. The self is not a ghost in the machine that stands outside causal processes; it is a causal process, and calling it illusory because it is causal conflates two different kinds of claim.
Implications for Justice and Responsibility
Retribution and Its Foundations
The most practically consequential implication of the free will debate concerns the justification of punishment. Retributive theories of punishment hold that wrongdoers deserve to suffer in proportion to their wrongdoing, and that this desert is a reason to punish them independent of any forward-looking benefits (deterrence, rehabilitation, incapacitation). Retribution in this sense depends on libertarian free will: the wrongdoer deserves punishment only if they genuinely could have done otherwise.
Derk Pereboom's 'Living Without Free Will' (2001) and 'Free Will, Agency, and Meaning in Life' (2014) argue for hard incompatibilism -- the view that free will in the sense required for basic desert moral responsibility does not exist -- and draws the conclusion that retributive punishment is never justified. Pereboom argues that this does not require abolishing criminal justice institutions; incapacitation for public safety, treatment, and forward-looking deterrence remain legitimate. But the punitive impulse itself, the desire to make wrongdoers suffer because they deserve to, is, on Pereboom's view, a moral mistake.
The neuroscientist David Eagleman has advocated a similar position in more applied terms, arguing in 'Incognito' (2011) that criminal justice should function more like medicine -- focused on understanding and changing the biological and social determinants of offending -- and less like a moral accounting system operating on the assumption of uncaused free choice. He cites cases in which the onset of criminal behavior correlated with undetected brain tumors, traumatic brain injuries, and other neurological conditions as illustrations of how biological causation can be dramatic and undeniable.
Compatibilism and Moral Practice
Most philosophers who work on free will and moral responsibility are compatibilists, and they argue that the practices of holding people responsible, praising and blaming, rewarding and punishing, are not undermined by determinism when properly understood. P. F. Strawson's landmark 1962 paper 'Freedom and Resentment' argued that the stance of holding people responsible is not a theoretical posture that depends on a philosophical thesis about metaphysical freedom but an irreducible feature of human interpersonal life. The reactive attitudes -- resentment, indignation, gratitude, love -- are not optional extras that we could subtract from human life while leaving everything else intact; they are constitutive of what it is to relate to others as persons.
Strawson distinguished between the 'participant stance' -- relating to others as fellow agents, with the full range of reactive attitudes -- and the 'objective stance' -- treating someone as an object to be managed, appropriate when someone is severely mentally ill or too young to be held responsible. The objective stance is sometimes appropriate and sometimes required; but it is not the default for adult human interaction, and the metaphysical debate about determinism gives us no reason to adopt it as universal.
Practical Takeaways
The free will debate has implications that extend beyond philosophy seminars. The evidence is reasonably strong -- from Vohs and Schooler's research on belief effects, from clinical experience with trauma and addiction, from the moral psychology of praise and blame -- that the way people conceptualize agency and responsibility affects how they behave and how they treat others. A nuanced compatibilist position -- one that holds people responsible while remaining attentive to the biological and social causes of behavior -- seems both philosophically defensible and practically superior to both libertarian and eliminativist extremes.
In criminal justice, the shift is already underway in many jurisdictions: trauma-informed approaches, cognitive rehabilitation programs, and risk-assessment tools that identify the causes of reoffending rather than simply punishing past behavior represent practical applications of a more deterministically informed view of human action. Whether this shift should go further -- whether retributive punishment should be abandoned altogether -- remains a live and genuinely difficult question.
References
- Dennett, D. C. (1984). Elbow Room: The Varieties of Free Will Worth Wanting. MIT Press.
- Dennett, D. C. (2003). Freedom Evolves. Viking.
- Harris, S. (2012). Free Will. Free Press.
- Pereboom, D. (2001). Living Without Free Will. Cambridge University Press.
- Strawson, P. F. (1962). Freedom and resentment. Proceedings of the British Academy, 48, 1-25.
- Libet, B., Gleason, C. A., Wright, E. W., & Pearl, D. K. (1983). Time of conscious intention to act in relation to onset of cerebral activity. Brain, 106(3), 623-642.
- Soon, C. S., Brass, M., Heinze, H. J., & Haynes, J. D. (2008). Unconscious determinants of free decisions in the human brain. Nature Neuroscience, 11(5), 543-545.
- Schurger, A., Sitt, J. D., & Dehaene, S. (2012). An accumulator model for spontaneous neural activity prior to self-initiated movement. Proceedings of the National Academy of Sciences, 109(42), E2904-E2913.
- Eagleman, D. (2011). Incognito: The Secret Lives of the Brain. Pantheon.
- Kane, R. (1996). The Significance of Free Will. Oxford University Press.
- Frankfurt, H. (1971). Freedom of the will and the concept of a person. Journal of Philosophy, 68(1), 5-20.
- Vohs, K. D., & Schooler, J. W. (2008). The value of believing in free will. Psychological Science, 19(1), 49-54.
Frequently Asked Questions
What is the difference between hard determinism and compatibilism?
Hard determinism holds that every event in the universe, including every human thought, decision, and action, is the inevitable result of prior causes operating according to physical laws. Given the state of the universe at any point in time, the entire future is in principle fixed. Since free will requires that a person could have done otherwise than they did -- that their choice was genuinely open -- and since determinism rules out any genuine openness, free will and determinism are incompatible. The hard determinist concludes that free will does not exist.Compatibilism holds that free will and determinism are not in conflict because the kind of free will worth wanting does not require the ability to have done otherwise in the strong metaphysical sense. What matters for moral responsibility, praise, blame, and the practices of holding people accountable is whether a person's actions flowed from their own reasoning, values, and desires without external compulsion or internal dysfunction -- not whether the causal chain leading to those reasons could have been otherwise.Daniel Dennett, in 'Elbow Room' (1984) and 'Freedom Evolves' (2003), argues that the concept of free will that determinism threatens is a philosophical abstraction that does not capture what we actually care about in our practical lives. The kind of freedom that matters -- responsiveness to reasons, the ability to reflect on and revise one's desires, acting from one's own values -- is entirely consistent with determinism and is indeed a product of evolution.
What did the Libet experiments show and why are they controversial?
Benjamin Libet's experiments in the early 1980s measured the relationship between neural activity and the conscious experience of deciding to move. Participants were asked to flex their wrist whenever they felt like it and to note the position of a clock hand when they first became aware of the intention to move. Libet found that a characteristic brain-wave pattern called the readiness potential (Bereitschaftspotential) began approximately 550 milliseconds before the movement, while participants reported becoming aware of the intention to move approximately 200 milliseconds before movement. This meant that brain activity predicting the movement preceded conscious awareness of the intention by roughly 350 milliseconds.Libet interpreted this as evidence against traditional conceptions of conscious free will: the brain was 'deciding' to act before the person was consciously aware of deciding. He retained a role for free will through what he called the veto -- the possibility of consciously canceling an action that the brain had already initiated. Critics have raised several objections. The readiness potential may not represent a decision but a general preparation for movement; Patrick Haggard and Martin Eimer found that the specific decision to move was associated with a later, lateralized potential. The introspective reports of 'when I felt the urge' are unreliable. And the experiments involve trivial, arbitrary movements, not the complex deliberate decisions that matter to questions about moral responsibility.
Why does the free will debate matter for criminal justice?
The free will debate has direct implications for how we justify criminal punishment. Most philosophical theories of punishment invoke either deterrence (punishment prevents future crimes by making them costly), rehabilitation (punishment aims to change the offender), or retribution (punishment is deserved because the offender freely chose to act wrongly). Retributive justifications depend directly on libertarian free will: punishment is deserved only if the offender could genuinely have done otherwise.If hard determinism is true -- if every action is the inevitable product of prior causes including genetics, upbringing, and circumstance -- then retributive punishment loses its foundation. The murderer could not have done otherwise, any more than a rock can be blamed for falling. The philosopher Derk Pereboom, who defends what he calls 'hard incompatibilism,' argues that this does not mean we should abandon all criminal justice institutions, since detention for public safety and rehabilitation remain justified -- but retributive punishment inflicted simply because someone 'deserves' to suffer does not.Neuroscientist David Eagleman, in 'Incognito' (2011), argues that neuroscience increasingly reveals the biological and developmental determinants of criminal behavior -- brain lesions, prenatal exposures, adverse childhood experiences -- and that this should shift criminal justice toward a more forward-looking, medical model focused on what interventions actually reduce reoffending, rather than backward-looking moral accounting.
What is Sam Harris's argument against free will?
Sam Harris argues in 'Free Will' (2012) that free will is a thoroughgoing illusion. His argument has two main prongs. First, drawing on Libet and subsequent neuroscience, he argues that conscious decisions arise from neural events that are not themselves the product of conscious decision -- we did not choose the desires, beliefs, and dispositions that generate our choices. Second, he argues that even if determinism is false and quantum indeterminacy introduces genuine randomness into brain processes, this randomness is not what free will requires. A random neural fluctuation that alters a decision is not an exercise of free will; it is a form of noise.Harris argues that abandoning the illusion of free will has beneficial consequences: it increases compassion (seeing others as products of causes they did not choose), reduces excessive self-congratulation for good outcomes, and clarifies what actually works in criminal justice by removing the illusion of purely deserved punishment.Daniel Dennett responded in 'Intuition Pumps and Other Tools for Thinking' (2013) and elsewhere that Harris is attacking a straw man -- a libertarian conception of free will that no serious philosopher defends. The question is not whether our choices are caused (of course they are) but whether they are caused in the right way -- by our own reasoning, values, and deliberative processes rather than by coercion, manipulation, or pathology. Harris, Dennett argues, has used a philosopher's trick of focusing on an extreme scenario (total randomness vs total determinism) to dismiss the actual phenomenon of interest.
Does believing in free will affect how people behave?
Research by social psychologists Kathleen Vohs and Jonathan Schooler, published in 2008 in 'Psychological Science,' found that participants who read passages arguing that free will is an illusion subsequently cheated more on a test than a control group, and a subsequent study found that they were also less helpful and more aggressive. The studies suggested that belief in free will might function as a psychological resource for self-regulation: people who believe their choices are genuinely up to them may exert more effort to align those choices with their values.The Vohs and Schooler findings proved difficult to replicate reliably, and subsequent meta-analyses found smaller and less consistent effects. The theoretical interpretation is also disputed: it is unclear whether the studies demonstrate something about free will belief specifically or about a broader orientation toward agency, meaning, and personal responsibility.Whatever the empirical picture, the question of whether belief in free will is beneficial even if it is false is itself philosophically interesting. It is one version of the question William James raised about pragmatic justification for beliefs: is a belief that is psychologically useful but metaphysically false nonetheless worth holding? Most philosophers who deny libertarian free will -- including Dennett, Pereboom, and Eddy Nahmias -- argue that a correct understanding of compatibilist free will provides all the psychological and practical resources that people actually need without requiring false metaphysical beliefs.