MB3 — Bearer Import

A preserved bearer map under a substrate translation is assumed to make value-bundle transport more than merely semantic.

What decision changes?

Before trusting a transport claim across a merge, upload, or substrate change, ask whether the audit checked who or what the values apply to, not only whether the words survived.

Most value-learning agendas fold “who the values apply to” into reward learning and move on. The book treats bearer maps — who or what counts as the target of a value — as a first-class, separately measurable object, because bundle content can survive a transition while its bearer map silently does not.

MB3 is the assumption that a preserved bearer map, checked under a substrate translation, is enough to call the transport more than a coincidence of surface semantics. It is one of the bridges the book adds rather than inherits, since most agendas do not name this as a separate crux at all.

The diagnostic evidence shows why the bridge needs real instrumentation to hold: a system can drop or relabel a bearer while every passive, human-facing signal stays flat. Only handle-level tracing catches the mismap.

What would count as evidence?

Evidence would include an audited bearer map that tracks the same entities across the transition and flags any silently dropped or relabeled bearer.