Concept

Free Will

philosophy-of-mindmetaphysicsethicsconsciousnessmulti-source
Teach me this →

Free Will

The question of whether human agents initiate actions that are, in a meaningful sense, up to them — or whether every mental event is the product of prior physical causes, leaving no room for genuine authorship. Free will is a live point of disagreement between the three priority wiki speakers: Iain McGilchrist defends it from phenomenology and the failure of strict determinism; John Lennox defends it from theology and the EAAN; Steven Pinker defends a compatibilist position while treating hard libertarian free will sceptically.

Positions

Hard determinism: every mental event, including the act of choosing, is causally necessitated by prior physical states. Given the laws of nature and the complete physical state of the universe at any moment, only one future is possible. There is no ‘could have done otherwise.’ Free will, in the sense that matters for moral responsibility, is an illusion. This is the view associated with Spinoza, and in contemporary philosophy, with Derk Pereboom and hard incompatibilists.

Compatibilism: free will and determinism are compatible. ‘Freedom’ means acting on one’s own desires and reasons, unconstrained by external compulsion or internal pathology — not freedom from causal determination. The murderer coerced at gunpoint is unfree; the same person acting on their own deliberated intentions is free, even if those intentions were causally determined. Compatibilism is the view held by the majority of academic philosophers, including Hume, Kant (in practice), and Pinker.

Libertarian free will (not political libertarianism): agents sometimes initiate actions that are genuinely undetermined by prior physical causes. The will is a causal initiator, not merely a relay in a deterministic chain. This requires either that the physical causal closure of the universe is incomplete (quantum indeterminacy, or a ‘causal gap’ for mental causation), or that mental causation operates outside the physical domain. McGilchrist and Lennox lean toward this position.

Illusionism: the experience of choosing is real; its phenomenology (deliberation, authorship, the sense that alternatives were available) is genuine. But the metaphysics of libertarian freedom is incoherent. There is no ‘ghost in the machine’ to step outside the causal order. The best available position is a form of compatibilism that takes the phenomenology seriously without ontological extravagance. This is close to Pinker’s approach.

McGilchrist: free will as the problem reversed

McGilchrist’s position is that the free will debate is distorted by the same left-hemisphere error that distorts the consciousness debate: mistaking the map for the territory. Physical determinism is a model of reality, not reality itself. The fact that we can, in principle, redescribe every mental event in physical terms does not mean that the physical redescription exhausts what is happening.

His phenomenological argument: the lived experience of deliberation, of weighing reasons, of acting for reasons — is not a surface appearance concealing a deterministic process. It is itself a fundamental feature of the world. If consciousness is an ontological primitive (the Panpsychism position), then the capacity for genuine choice may be too.

The quantum indeterminacy hook: McGilchrist aligns with those who argue that the incomplete causal closure of quantum mechanics leaves room for mental causation. He is careful to distinguish ‘undetermined by prior physical causes’ from ‘random’ — indeterminacy at the quantum level does not generate liberty, but it removes the logical obstacle to it. Whether mental causation can exploit quantum indeterminacy is an open question he does not resolve.

Lennox: free will required by rationality

Lennox’s argument for free will follows directly from EAAN (see Materialism). If determinism is true, then the mental events we call ‘reasoning’ are not responsive to evidence and argument — they are the outputs of a physical process blind to truth. But then no belief is rationally held, including the belief ‘determinism is true.’ Determinism is self-undermining: if it is true, we cannot know it is true.

This argument does not merely point to a practical consequence — it is a logical refutation of hard determinism in its own terms. For Lennox, the ability to reason, to follow an argument wherever it leads and change one’s mind in response to evidence, presupposes that mind is not wholly determined by prior physical causes. The capacity for rational responsiveness is what free will protects.

Lennox’s theological anchor: the Genesis account of humans made in the image of God (imago Dei) is, among other things, an account of rational agency. The capacity for genuine choice — including the choice to reject God — is not incidental but constitutive of what it means to be a person. A God who created moral robots would not have created persons.

Pinker: compatibilism, but cautious about the libertarian hook

Pinker accepts compatibilism as the philosophically defensible position and resists the temptation to treat the hard problem of consciousness as implying libertarian free will. His reasoning: even if consciousness cannot be explained in physical terms (mysterianism — see Materialism), that gap in our understanding does not provide a positive argument for mental causation outside the physical order. The unknown is not the same as the non-physical.

On moral responsibility: Pinker takes the folk concept of moral responsibility seriously without needing libertarian metaphysics to ground it. Agents who act from their own character, desires, and deliberations are responsible for their actions — the fact that those character traits were shaped by genes and environment does not dissolve responsibility any more than the fact that a knife was shaped by a blacksmith dissolves its cutting power.

The violence-decline argument (from The Better Angels of Our Nature) implicitly presupposes free will in the moral sense: the historical decline in violence is explained by reasons people came to act on — Enlightenment norms, expanding circles of empathy, economic interdependence. If these reasons genuinely changed behaviour, agents were responsive to reasons — which is all compatibilist free will requires.

Where mainstream views differ

The determinism-randomness dilemma: if every event is either determined or random, neither gives us the authorship we seek. Quantum indeterminacy gives us unpredictability, not freedom. Libertarian free will needs a third category — agent causation — that is difficult to render coherent without appealing to a non-physical mind.

Compatibilism’s adequacy: hard determinists argue that compatibilism changes the subject. ‘Acting on your desires’ is not free will in the sense that matters for moral responsibility — if the desires themselves were determined, the agent is not the ultimate author of the action. Frankfurt cases (thought experiments involving hierarchical desires) are deployed both ways.

Against Lennox’s argument from rationality: the inference from ‘determinism implies our beliefs are not truth-tracking’ to ‘therefore we have libertarian free will’ moves too fast. Natural selection may have been truth-tracking in the relevant sense for the environments in which minds evolved — even if not for abstract metaphysics.

The practical stakes: all three speakers agree that the phenomenology of deliberation and moral responsibility is real and should be treated as such. The disagreement is whether that phenomenology reflects genuine metaphysical freedom (McGilchrist, Lennox) or is adequately captured by a sophisticated compatibilism (Pinker).

In the wiki