Continuum electromechanical models require bulk forces and surface tractions at the resolution of the target mechanical prediction. Agreement in macroscopic electromagnetic fields, or even in total force, does not by itself establish that these local loads are determined. Starting from microscopic carrier balances, we formulate an output-first minimum-closure diagnostic that separates two ordered questions: transmission closure (C_T) specifies how Lorentz forces on carriers reach the selected mechanical host, whereas projection closure (C_P) asks whether the retained electromagnetic state determines the required load once transmission is fixed. Two controlled equal-state comparisons isolate these noninterchangeable failure modes. In a uniform prescribed external field, bound dipoles and individually anchored charges have the same retained electromagnetic state and retained projection of the applied-field carrier loading, yet produce different host loads because their support topologies transmit endpoint forces differently. With transmission fixed, aligned and staggered line-dipole arrangements have the same retained polarization and continuum field, yet produce different boundary tractions because registry-dependent near-field correlations omitted by retention survive as an interfacial load. These witnesses establish structural nonuniqueness within the declared model classes, rather than universal material magnitudes. The diagnostic stops when global balance fixes the requested total force or torque on the receiving system; otherwise it establishes transmission before testing whether a projection residual affects the selected output above tolerance. Missing closure may be supplied by a support or carrier-kinetic model, targeted reconstruction, or an equivalent validated reduced law. A macroscopic load law is therefore sufficient whenever transmission is fixed or irrelevant and all output-relevant residuals are negligible or represented.