Compatibilism Is Not What You Think It Is
Compatibilism is usually introduced as a way of reconciling free will with determinism. The standard description says that even in a deterministic universe, free will still survives in some qualified sense. We are told that intentions still matter, that deliberation still plays a role, and that responsibility can be preserved once the concepts are carefully defined.
For many readers, this framing feels unsatisfying. It sounds like an argument happening at the level of definitions, rather than an explanation of how decision-making actually works in a physical system. Worse, it often skips over the enormous amount of structure between fundamental physics and human action, as if determinism somehow operated directly on choices without passing through brains, histories, and learned patterns of behavior.
At the most fundamental level, the universe undergoes a single evolution through time. From its earliest low-entropy state, every later state follows from earlier ones according to physical laws. From this perspective, any decision I make is simply one moment in that unfolding. There is no branching of reality when decisions are made. No alternative history where something else happens.
From the inside, though, something else is going on. I experience myself as predicting possible futures, weighing outcomes, and committing to actions under uncertainty. This experience isn’t a mistake. It is what the lawful unfolding of the universe feels like when it passes through a system that has to model the future in order to act.
Seen this way, compatibilism isn’t really a compromise between freedom and determinism. It’s an acknowledgment that internal and external descriptions are answering different questions about the same process. The objective description tracks how the universe evolves. The subjective description captures what that evolution is like from within a system that participates in shaping what comes next.
This pattern is not unique to free will. The same confusion appears whenever a phenomenon has both an internal and an external description. Conscious experience is the clearest example. From the outside, perception can be described entirely in terms of neural activity and information processing. From the inside, it appears as color, sound, pain, and presence. Saying one of these descriptions is “real” and the other an “illusion” mistakes a difference in perspective for a difference in ontology.
When I deliberate about whether to take a new job, the physical description involves neural activity shaped by prior learning, emotional responses, and the current state of my brain. From that perspective, the process is entirely physical. It just follows physical laws. From the inside, however, I am weighing reasons, imagining possible futures, and committing to a choice. Both descriptions are true. The question is not which one is really happening, but which vocabulary is appropriate for understanding what is going on.
This is why debates about the so-called “hard problems” tend to stall. As David Chalmers originally framed it, the hard problem of consciousness asks why physical processes should give rise to subjective experience at all. But the difficulty often lies in the assumption that reality has to be exhausted by third-person descriptions.
Free will fits the same pattern. From the outside, behavior can be traced through brain states, prior conditions, and physical law. From the inside, it appears as deliberation, intention, and choice. Treating one as real and the other as illusory repeats the same mistake. Both are descriptions of the same unfolding process, seen from different vantage points.
There are many features of reality that exist only from a first-person perspective. Reasons don’t show up in physics, yet we act for reasons. Meaning doesn’t appear in fundamental equations, yet things matter to us. Indexicals like “I,” “here,” and “now” are nowhere in an objective description of the universe, yet every experience is structured around them. Denying the reality of these phenomena because they are perspectival doesn’t make them illusions.
Illusions are perspectives with no objective counterpart at all. Free will, like the what-it’s-like-to-be of consciousness, is a perspectival description that does have an objective counterpart. It refers to a real physical process, described from the inside rather than the outside. The question of whether free will “really exists” is therefore not well formed. It asks whether an internal description can be reduced to an external one, when the point is that they answer different kinds of questions.
This also helps clarify why the question of why there is a first-person perspective at all can feel so puzzling. The question tries to step outside that perspective and treat it as something that should appear within an objective description of the world.
But we are never outside the system we are describing. The first-person perspective is not something we encounter from the outside. It is the standpoint from which the question is being asked in the first place.
This also raises the question of what kind of system would be expected to have a first-person perspective at all. Not every information-processing system qualifies. A calculator performs computations. A language model generates responses. But neither maintains a continuous model of the world and itself over time. There’s no ongoing process guides action into the future.
A first-person perspective becomes meaningful in systems that have to continuously track their environment, maintain a persistent internal state, and use predictions to select actions under uncertainty. In those systems, the internal perspective is not optional. It is what the process of modeling and acting feels like from the inside.
The common thread here is prediction. A system embedded in a complex, evolving environment can’t wait for certainty. It has to act on partial information, using internal models to anticipate what’s happening now and what’s likely to happen next. Conscious experience is what it’s like to perform that prediction in the present moment. Free will is what it’s like to extend that same predictive machinery into the future and commit to a course of action.
Neither experience competes with an objective description of the system. Both are the inside view of what prediction feels like when it becomes essential for persistence.