From Domain Shift to Domain Rheology: Heterogeneous Evolution Laws, Open-System Exchange, and Boundary-Load Accumulation in Co-Evolving Regimes

Persistent domains can have similar current states but different transition laws, response memories, and exchange conditions. This paper asks whether those differences help predict where repeated contact leaves unresolved burden and lasting structural change. Domain rheology is a protocol-indexed response description, not an intrinsic property of a discipline. A preidentified interface connects the domains; a typed stock-flow ledger distinguishes unresolved load, realized repair, feasible export, and downstream absorption. Structural complexity, service performance, and remaining burden are measured separately. Seven elementary calibration propositions are supplemented by a conditional accommodation model: independently justified adjustment opportunities, costs, and retention can turn residual mismatch into localized structural additions. A narrow feedback closure gives stability conditions, a feasible oscillation counterexample, and limits on identifying the mechanism from equilibrium alone. Backlog alone does not imply orogeny. Twelve prospective tests require common information, service obligations, strong dynamic rivals, and held-out outcomes. External comparisons include an exactly reconstructed stateful-feedback example, reported reasoning-order and adapter tradeoffs, and field evidence on assistance, review work, and delayed data problems. They constrain missing-state, endpoint, selection, and accounting explanations; they do not calibrate the complete DRIO mechanism. Existing technical-debt, queueing, and performative-prediction work precludes broad priority claims. The remaining question is whether source-calibrated response representations improve forecasts or intervention choices at declared data and computational budgets, beyond rivals with the same observations. No new trained-model or field validation of that increment is reported. Central question: under common service obligations, do source-measured changes in laws and response properties predict where persistent contact leaves new structural accommodation and unresolved burden, beyond current mismatch, dependency history, and generic dynamics? A structural addition can be repair rather than harm; internal relief need not be net relief. Note on this release. Version 1.0 working paper; original draft 13 September 2026, evidence audit 19 September 2026. The upload contains the manuscript (about 20,200 words) and a supplement archive with two standard-library Python audit scripts and their JSON reports: check_algebra.py (40 checks in 12 groups) and check_strengthening.py (34 checks in 10 groups), both reported PASS, together with a README, MANIFEST.json, and SHA-256 checksums. These are deterministic analytical checks of the equations, boundary cases, counterexamples, feedback closure, identification examples, and measurement bounds stated in the manuscript. They are not an empirical validation: no new trained-model experiment, field study, or replication of external studies is reported here.