The Continuum Shadow (CODA-CS)

The Continuum Shadow (CODA-CS) What This Paper Is About CODA-CS establishes how continuum representations descend to an exact finite calculus, and where that descent fails. When continuous processes or signals are sampled on discrete grids, fine observations are forgotten into coarser ones. A coarse observation can be honest while still losing information, like mistaking a blurry photo for the original. CODA-CS proves that refinement preserves descent: if a query cannot distinguish states that look identical after forgetting, it cannot distinguish states that look identical before forgetting. Conversely, coarse observations create fiber collisions. Projecting Boolean pairs to their first coordinate collapses four states onto two, obstructing any attempt to recover the second coordinate from the first. Crucially, writing a limit notation or a continuum formula is merely giving something a name. It does not by itself confer authority to answer a query. In CODA, operational authority requires an admitted finite shadow tower, cofinal slices, and an explicit local terminalizer. When any of these are absent, query authority is refused rather than manufactured out of continuum notation. Scope and Apparatus Mathematical Core: Refinement descent, pointwise family observation equality, Boolean fiber collisions, and the necessity of finite shadow terminalizers. Executable Verification: An exact rational Python engine and self-contained Lean 4 formalization with zero external dependencies, proving the theorems without floating-point approximations or continuum hand-waving. What Is Not Claimed: CODA-CS does not prove the gluing of overlapping shadows (which remains open as the H3 obstruction), nor does it claim that unadmitted queries lack mathematical definitions. It formalizes the refusal of operational authority without an authorized finite descent witness.