Against the Unity of Decision: A Structural Reinterpretation of the Libet Experiment

The Libet experiment has long been interpreted as evidence that unconscious neural processes precede and potentially determine conscious decision-making. Standard debates therefore focus on the temporal ordering of neural activity, conscious awareness, and motor action, and attempt to determine what this ordering implies about free will and conscious agency. Despite extensive discussion, however, the debate has remained persistently non-convergent. This paper argues that the source of this instability is structurally mislocated. Existing interpretations share an implicit assumption: that neural activity, conscious report, and action together constitute a single unified decision event. The temporal relations observed in the experiment are then interpreted as internal relations within that unified object. The present paper challenges this presupposition. The central claim is that the Libet experiment does not, by itself, establish the structural unity of decision. To articulate this limitation, the paper distinguishes between local admissibility and global co-satisfiability. Neural activity, conscious report, and action may each be individually admissible within the experimental setting while nevertheless failing to form a jointly maintainable configuration. Temporal ordering alone does not guarantee that these components belong to a single unified decision object. On this basis, the paper reinterprets the significance of the Libet data. The observed precedence of neural activity over conscious awareness does not establish that decisions are unconsciously initiated within a unified decision structure. Rather, it reveals that the unity of decision is itself structurally unsecured. The resulting instability is therefore not primarily a disagreement about free will, but a consequence of evaluation being applied prior to the establishment of a sufficiently unified evaluative object. The paper further explains why the Libet debate persistently fails to converge. Competing interpretations stabilize different subsets of the experimental configuration while presupposing a unified decision object that is never structurally secured. The disagreement is therefore not merely empirical or theoretical, but configurational. The contribution of the paper is re-locational rather than revisionary. It does not deny the empirical findings of the Libet experiment and does not propose a new theory of free will or consciousness. Instead, it identifies a structural condition governing when distributed components can be treated as a single decision event. By locating instability at the level of configurational unity rather than evaluative interpretation, the paper clarifies both the persistence and the limits of the Libet debate.