Ground audit: where there is floor, where one spins, and where there is noise

This article is not about new physics. It is about ground: which numbers can serve as foundation, which cannot, and why. The thesis is that there are three categories with incompatible statutes — direct empirical datum, tautology and theory dressed as empirical — and that confusing them produces wasted work instead of knowledge. The worked case is the T2K article on the phase [1]. It was chosen for being a good piece of work: it declares every layer, reports the anomaly that bothers it, and uses could and may where the link is theoretical. Precisely for that reason it serves as a case: the problem is not in the article's honesty, it is in the reading that turns a coordinate of a formalism into an explanation of the existence of matter.