Section 6: Free Will

This section addresses what free will actually is within a relational, open territory — not as a metaphysical abstraction but as a specific structural capacity of agents, with a precise mechanism and a bounded taxonomy of what can be willed.

Section 6A: Free Will and the Individual

The case for Frame A on free will, argued from within — stated honestly.

The Honest Conclusion

Frame A’s honest conclusion about free will is not compatibilism. It is that free will does not exist.

If reality consists of individual substances governed by fixed laws, and if the physical world is causally closed — every neural event determined by prior neural events, which are determined by prior physical events, all the way back to initial conditions — then the felt sense of choosing freely is an illusion. The decision has already been made by the time consciousness registers it. The experience of deliberation is a post-hoc narrative the brain constructs to make sense of outputs it has already generated. There is nothing to discuss. The outcome was fixed before the agent arrived.

This is not a fringe position. It is where Frame A leads when followed without flinching. Daniel Wegner’s The Illusion of Conscious Will (2002) makes the case with neuroscientific precision: the feeling of willing is a causal inference the brain makes about its own outputs, not a genuine causal force. Derk Pereboom’s hard incompatibilism reaches the same conclusion from the philosophical direction: no coherent account of agent causation survives the causal closure of the physical world, and the honest response is to abandon the concept rather than redefine it into something unrecognisable. (See Pereboom, Living Without Free Will, 2001.)

The Calvinist Convergence

What is striking is that this conclusion is not unique to secular materialism. Calvinist theology arrives at the same place from a completely different direction.

On the Calvinist account, God’s sovereignty is total and unconditional. Election is not foreknowledge of what agents will freely choose — it is the prior determination of what they will do, before and independent of anything in them. The agent does not choose God. God chooses the agent, and the choice is made before the foundation of the world. Human action follows from the divine decree as necessarily as physical events follow from prior causes. The agent is not a genuine originator of its own destiny. It is an instrument through which a predetermined outcome is expressed.

The structural identity between hard determinism and Calvinist predestination is not a coincidence. Both are Frame A pursued to its conclusion in domains that take the question seriously — one in physics, one in theology. Both say: the outcome was fixed before the agent arrived. Both say: what looks like genuine choice is, at the level of ultimate explanation, not that. The convergence across secular and religious Frame A is evidence of the position’s genuine depth. It is not a mistake. It is the view from inside a consistent individualist ontology applied without rescue operations.

Compatibilism as Concession

Compatibilism — the attempt to show that free will is consistent with determinism by redefining what free will means — is not Frame A’s position. It is Frame A’s attempt to recover something after the honest conclusion has already been reached.

The compatibilist move is to argue that the freedom relevant to moral responsibility is not the freedom to have done otherwise given identical prior causes, but the freedom to act from one’s own reasons and values without external coercion. This is a real distinction and it matters practically. But it concedes the central point: there is no genuine ontological openness, no real bifurcation, no moment at which the outcome was genuinely undetermined. The compatibilist agent acts from its own character — but that character was itself determined, and so were the actions that flow from it. The freedom is linguistic, not ontological. It describes a kind of action, not a kind of causation.

Hume saw this clearly. His compatibilism was not a defence of free will as ordinarily understood. It was a recommendation to stop caring about what cannot be had and redirect attention to what can be preserved: the practices of praise, blame, and moral formation that remain useful even in a fully determined world. This is honest. But it is a counsel of resignation, not a vindication of free will.

What Frame A Leaves

Frame A’s position on free will is coherent, consistent, and — stated honestly — bleak. The will is not a cause. It is an experience of causation. The agent is not an originator. It is a conduit. Responsibility, on this account, is a social practice rather than a metaphysical fact — useful for coordinating behaviour, allocating consequences, and shaping future actions through incentive, but not grounded in any genuine capacity to have done otherwise.

This is the position. It does not need to be rescued. It needs to be confronted — which is what Frame B does.

Section 6B: Free Will and Structure

The case for Frame B on free will, argued from within.

Dissolving the Old Framing

The compatibilism/determinism debate inherits Frame A’s assumption that the physical world is causally closed. Once that assumption is in place, free will either has to be fitted into the gaps (compatibilism) or abandoned (eliminativism). The debate is real, but it is conducted within a frame that may be wrong.

Frame B dissolves this framing not by choosing a side but by challenging the assumption. The territory is not causally closed. It is genuinely open — structurally, at every level from quantum mechanics to cognition, as established in Sections 3 and 4. The question is not whether free will is consistent with determinism. The question is what free will actually is in a territory where genuine openness is a structural feature and agents are its most developed instantiation.

The answer is already partially in view. Section 4 established that agents are selection nodes — the instantiation of the bifurcation process that has become reflexive. Section 5 established that the strange loop of self-reference constitutes a genuine interiority within which multiple outcomes are simultaneously held. What remains is to characterise the will precisely: what it does, at what levels, and what its structure implies.

Libet Revisited

The Libet result, read carefully, supports the Frame B picture rather than undermining it.

Libet’s own interpretation of his results has been systematically misread in the popular debate. He found not only the readiness potential preceding conscious intention, but also that subjects retained the ability to veto the action after becoming aware of the intention and before the movement occurred. The will, on Libet’s own account, operates primarily as inhibition — not as the initiator of neural preparation but as the gatekeeper that permits or arrests what the neural system has prepared. (See Libet, Mind Time, 2004.)

Subsequent reanalysis has strengthened this reading further. Schurger and colleagues showed that the readiness potential is better understood as stochastic neural noise reaching a threshold than as a deterministic preparatory signal — which means the neural system is not committing to an action before consciousness arrives, but running a noisy process that consciousness can gate. (Schurger et al., PNAS, 2012.)

This is precisely the symmetry-breaking picture. The neural system generates a space of prepared possibilities — a stochastic process operating near the edge of chaos, where small perturbations can propagate into committed action. The will operates by tipping this system — not by originating action from nowhere, but by selecting among genuinely available trajectories at the moment of bifurcation. The energy cost is minimal. The cascade provides the amplification. The selection is real.

The Taxonomy of Will

Free will is not a single capacity. It operates at four levels, each building on the one below, each with different scope and different implications for ethics.

Inhibition. The most fundamental and most reliably documented form of will. The neural system prepares options through a stochastic process. The will operates primarily by permitting some preparations to proceed and arresting others — veto power, not origination. It requires minimal energy, operates at the last moment before commitment, and is the will’s most primitive and most certain expression. It is also, as Libet showed, available even when conscious initiation is not: you may not have initiated the cascade, but you can stop it.

Attentional selection. The agent directs the representational system — choosing what to think about, which of the internally available possibilities to bring into focus, which aspects of the relational field to illuminate. This shapes the space within which subsequent bifurcations are resolved. Over time, habitual attentional patterns constitute character: the agent becomes the kind of entity that tends to attend to certain possibilities and leave others in shadow. Moral formation operates here — not in single dramatic choices but in the accumulated pattern of what is attended to, what is held in view, what is allowed to become vivid.

Metacognition. Available to agents of sufficient complexity: thinking about thinking, representing one’s own representational process, choosing to revise the framework within which selections are being made. Metacognition recursively expands the space of possible selections — an agent that can examine its own attentional habits can choose to change them; an agent that can examine its own values can choose to revise them. This is the highest expression of individual will within the framework, and the level most closely connected to the capacity for good faith: the willingness to examine one’s own frame and remain open to revising it.

Relational will. This level is irreducibly social and cannot be reduced to the three above. It is the will as exercised in and through relation — the choice of how to be with another, how to inhabit the between, how to hold the relational field. Relational will is not the application of individual will to a social context. It is a qualitatively different kind of exercise: the selection of a relational stance, a way of being in contact, that constitutes rather than merely expresses the agent’s identity in that moment. Choosing to forgive is not merely an inhibitory write against the resentment attractor. It is a relational act that alters the fabric between persons — and in doing so, alters both persons. Choosing to be genuinely present to another, rather than managing them from behind a self-model, is a relational will act with ontological consequences that extend beyond the agent making it.

This fourth level is where the will taxonomy connects directly to the ethics and love sections. The first three levels describe what an individual agent can do within its own representational system. The fourth describes what agents do together — the will as it operates at the level of the relational field rather than the individual node.

The Texture of Constraint

One further point requires stating here because it completes the picture established in Section 5. The will does not operate in open space. It operates within the specific attractor landscape the agent has accumulated — the history of prior selections that have shaped the manifold, the relational field that has constituted the self, the particular configuration of constraints that makes this agent this one and not another.

This is not a limitation on genuine will. It is its medium. A forgiveness made against the grain of a powerful resentment attractor — made by someone for whom it costs something real — is a more significant exercise of will than one that costs nothing. The constraint gives the selection its weight. The texture of the attractor landscape is not the obstacle to freedom. It is what makes freedom meaningful rather than merely random. It is what makes it a life rather than a series of undifferentiated events.

The will, on this account, is always embedded. It is always operating within constraints it did not fully choose. And the constraints are not incidental — they are partly constitutive of who is doing the choosing. This is Frame B’s account of agency: not the isolated individual exercising unconstrained choice, but the relational self navigating genuine openness within a structured field that it has partly inherited, partly built, and partly become.

Section 6C: Free Will as Witness

What the free will argument establishes — and what it hands to the ethics sections.

The two accounts do not have equal resources at this point in the argument.

Frame A’s compatibilism is coherent and practically adequate. It preserves moral responsibility, connects freedom to character and deliberation, and does not require anything the neuroscience cannot support. Its limitation is that it purchases this coherence by setting aside the question of genuine ontological openness — by treating it as either unnecessary or unanswerable. The Libet result, on Frame A’s reading, is a challenge to naive libertarianism, not to compatibilism. Frame A can live with Libet.

Frame B’s account goes further. It takes the Libet result as positive evidence for the symmetry-breaking picture — the stochastic neural process, the veto capacity, the tipping of a system poised at the edge of chaos — and integrates it into a taxonomy that runs from inhibition through relational will. The openness is not a gap in the causal fabric. It is a structural feature of a relational territory, operating at every level from quantum mechanics to cognition, and the will is the agent’s capacity to participate in its resolution.

The asymmetry that matters here is this: Frame A’s compatibilism can ground ethics in character and responsiveness to reasons, but it cannot ground the claim that selections are ontologically significant — that the fabric is genuinely different depending on how the resolution goes, and that the branch not taken is not recoverable. For Frame A, the weight of moral responsibility is a social and psychological fact. For Frame B, it is a structural feature of what agents are and what their selections do to the relational fabric.

That difference is what the ethics section turns on. If selections are genuinely ontologically significant — if agents are real contributors to an open relational territory whose structure they are continuously shaping — then the ethics is not a set of constraints imposed on agents from outside. It is a description of what it means to act in accordance with what agents actually are: nodes in a relational fabric whose selections matter, permanently and non-invertibly, to the structure of the whole.

What the will taxonomy hands to the ethics section specifically: relational will is where ethics lives. The first three levels of will are preconditions — they describe the agent’s capacity to navigate its own representational system with integrity. Relational will is where that integrity, or its absence, enters the shared fabric. The ethics of Frame B is built at this level: not the ethics of an individual protecting its interests or following rules, but the ethics of a relational node choosing how to inhabit the field that constitutes it.