--- title: "The feeling of choosing" author: Dr Arman Ouveysi date: 2026-07-07 room: Horizon tags: free-will, philosophy-of-mind, compatibilism, responsibility canonical: https://drouveysi.com.au/essays/the-feeling-of-choosing/ summary: "The felt moment of authorship is the most convincing experience a person has. On what survives when you stop trusting it, why the freedom worth wanting was never freedom from causation, and what happens to blame." --- # The feeling of choosing Pay attention to the next trivial decision you make. Tea or coffee, this route or that one, reply now or later. Somewhere in the middle of it there is a felt moment, small but unmistakable, where the options hang balanced and then you, the author, tip the scales. Nothing in ordinary life is more convincing than that moment. It is the experience the whole idea of free will is built on, the thing every argument about determinism eventually has to answer to. This essay is about whether that moment is telling the truth, and about the strange, mostly reassuring discovery of what survives if it is not. Start with the trap, stated fairly, because it is one of the oldest and it has never been sprung. Every event, including every event in your skull, either follows from prior causes or it does not. Take the first horn: if your choice was fully caused, by your brain state, which was caused by your history, genes and environment, which were caused by things that existed before you did, then the choice was, in some sense, written before you arrived to make it. Take the second horn: if the choice was not fully caused, if somewhere in the process there was genuine randomness, a quantum coin flip in the neurons, that rescues nothing, because a coin flip is not authorship either. **You are no freer for being the site of noise than for being the site of clockwork.** What the traditional picture of free will requires is a third thing: a self that stands outside the causal stream and reaches in, an uncaused causer, initiating choices without itself being the product of anything. Nobody has ever located this third thing, no one can describe how it would work, and the more carefully the question is posed, the more it looks not unanswered but unaskable. Then the evidence arrives, and it aims at the moment itself. I have written elsewhere on this site about introspection's reliability record, which is poor, so I will not re-derive it, only extend it, because the sense of authorship turns out to be one of introspection's least trustworthy products. The feeling of having willed an action can be detached, experimentally, from having actually caused it. People can be handed outcomes they did not choose, and they will explain their reasons for choosing them, fluently, sincerely, in detail. The famous split-brain work showed one half of a brain confidently narrating decisions the other half had made for reasons it could not access. In certain neurological conditions a hand acts with evident purpose while its owner disowns it; under hypnosis, the reverse, actions initiated elsewhere arrive wrapped in the full felt certainty of being one's own. I will note, for honesty, that the most famous experiments in this area, the ones measuring readiness signals before conscious intention, are more contested than their popular reputation suggests, so I will not lean on them. The dissociation cases are enough, and their lesson is precise: the sense of *I did that* is not a perception of causation. It is an attribution, constructed after the fact, usually accurate, and demonstrably paintable onto the wrong canvas. **Authorship is an inference the brain draws about itself.** Mostly a good inference. Still an inference. So the ghost is in serious trouble: the extra self, outside the causes, tipping the scales. Here is where most people expect the essay to turn bleak, and where it turns the other way instead, because the surprising discovery of the last few centuries of this argument is how much of what people actually care about was never resting on the ghost at all. Consider two signatures on the same contract. One is signed after reflection, because the terms serve the signer's goals. The other is signed at gunpoint. Physics does not distinguish these cases; both are fully caused events in nervous systems. Yet everything anyone has ever cared about under the name of freedom distinguishes them absolutely, and notice what the distinction actually tracks. Not causation versus its absence. Compulsion versus its absence. The free signature flows from the signer's own values, reasons and character; the coerced one is routed around them. That observation is the seed of the position philosophers call compatibilism, and stated plainly it sounds almost deflationary: the freedom worth wanting was never freedom *from* causation. It was freedom from coercion, freedom to act from your own reasons, with the capacity to respond to better ones. Your choices are caused, and among their proximate causes are your deliberation, your values, your history of caring about things. Determined by *you* is not the opposite of freedom. It is the mechanism of it. A choice that was not determined by you, that floated free of everything you are, would not be freedom. It would be possession. Now the objections, because a position this convenient had better survive contact, and there are three that have real teeth. The first says compatibilism is a shell game: it keeps the beloved word and quietly swaps the product. What people believed in was the deep ability to have done otherwise, the genuine fork in the road, and that is exactly what the causal picture removes; calling the remainder "free will" is marketing. I am partially guilty as charged, and I would rather show it than hide it. Compatibilism does not preserve the metaphysics of the folk picture; it preserves the *function*. Whether that counts as rescue or rebranding depends on what you think a concept is for, and my view, unsurprising given how I spend my weeks, is that concepts earn their keep by the work they do. The work this one does, grounding responsibility, licensing praise, detecting coercion, distinguishing the willing signature from the gunpoint one, survives intact. The deep fork in the road does not survive, and I will say so without euphemism: if the fork is what you wanted, compatibilism cannot give it back, and pretending otherwise would be the actual shell game. ![Causation fork for free will, with agency-as-practice as the surviving lane](https://drouveysi.com.au/assets/articles/horizon/the-feeling-of-choosing/fig-01.svg) The second objection is the sharpest, and it is the place my position pinches. Imagine a manipulator who engineers a person's values from the ground up, so that the person later acts from those values, uncoerced, reflectively endorsing every step. Every compatibilist condition is met. The act flows from the agent's own reasons and character. And yet almost everyone's verdict is that something essential is missing, that this person is not free, which suggests the compatibilist conditions were never the whole story. My honest response is that the difference between the manipulator and an ordinary causal history of genes, parents, culture and luck is not a metaphysical difference; it is the presence of another agent's purposes in the pipeline, and our intuitions track intruding agents for excellent practical reasons: manipulators can be deterred, blamed and stopped, whereas physics cannot. I think that practical asymmetry carries most of the intuition. I do not think it carries all of it, and I would rather leave the remainder standing in the open than paper over the one spot where my confidence visibly thins. The third objection is practical: that telling people the ghost is dead makes them behave worse, and there were studies suggesting exactly that, people nudged toward disbelief cheating a little more. The replications have wobbled, so the empirical claim is softer than its fame, but suppose it holds. It would be an argument about messaging, not about truth, and more importantly it targets the wrong message. The corrosive belief is not determinism; it is *fatalism*, the idea that since everything is caused, nothing you do changes anything, and fatalism is a straightforward logical error. **Your deliberation is not a spectator to the causal chain. It is a link in it.** The outcome runs *through* your weighing of reasons, which is why refusing to deliberate "because it's all determined anyway" is like refusing to eat because your hunger was caused. The causes do not bypass you. On any accurate accounting, they largely *are* you. Which leaves responsibility, and the part of the picture that genuinely changes. What the causal view corrodes is retribution in its pure form: the idea that suffering can be intrinsically deserved, owed to the ledger of the universe, because the person could have done otherwise in the deep, forked sense. If the fork was never there, that license expires. What survives is everything forward-looking: deterrence, protection, rehabilitation, repair, the practice of holding people responsible because holding them responsible is itself a cause that shapes future behaviour. That is enough to run a justice system, and arguably a more honest one, focused on what prevents harm rather than what balances cosmic books. I want to be careful here, because both directions carry costs I do not wave away. Victims' intuitions about desert are not a bug to be engineered out of people; they are ancient, real, and any system humans must actually live inside has to leave room for them. And a purely forward-looking justice, followed to its cold conclusions, has its own monsters. This is a place for renovation, not demolition, and anyone who tells you the implications are simple has not sat with them. Back to the kitchen, then, because the essay owes the moment an answer. The feeling of choosing is real, as a feeling; nothing here denies the experience. And the choosing is real, as a process: a brain weighing reasons is a physical event that genuinely determines what happens next, which is more than most physical events can say. The only casualty is the ghost that was supposedly standing between them, the extra author outside the causes, and the honest verdict on the ghost is that it was never doing any work. Every valuable thing it seemed to provide was being provided all along by the machinery it claimed credit for. I still feel the author every time the scales tip. I have just stopped reading the feeling as testimony and started reading it as an interface, the self-model's way of rendering deliberation from the inside. The hesitation before the choice used to feel to me like evidence of freedom. It now feels like **what weighing reasons is like when you are the scales**. Same hesitation, same kitchen, same coffee. I hold it differently now, and the coffee, for the record, never stood a chance.